EUROMEDLAB 2013 Posters

EuroMEdLab 2013 - postErs
T203 MEASUREMENT OF BASAL ANTI-MULLERIAN HORMONE LEVEL OF HEALTHY FERTILE WOMEN IN MACEDONI M. Boshkovska(1), I. Gjorgoski(2), A. Momirovska(1)
1
T204 ASSOCIATION OF VITAMIN D AND SOLUBLE FMS-LIKE TYROSINE KINASE 1/PLACENTAL GROWTH FACTOR RATIO WITH PREECLAMPSIA IN THE SECOND TRIMESTER Z. Boundi, C.M. Puche Morenilla, F. Caizares Hernndez, J. Calle Luna, E. Martnez Snchez, M. Martnez Villanueva Hospital Virgen De La Arrixaca, Murcia Background: Preeclampsia (PE) is a disorder emerging from placental dysfunction in the latter half of pregnancy. The condition is diagnosed following new-onset hypertension and proteinuria after 20 weeks of gestation and affects up to 2-5% of pregnancies worldwide. PE is one of the foremost complications of pregnancy and a major cause of maternal and fetal mortality. The angiogenic factors (Placental Growth Factor (PlGF), Soluble fms-like tyrosine kinase 1 (sFLT-1)) and 25-hydroxivitamin D(25-OH-D) may play an important role in the pathogenesis of angiogenesis. 25-OH-D is a risk factor for preeclampsia, its active form, 1,25-dihydroxyvitamin D3 stimulates VEGF vascular endothelial growth factor) expression in smooth muscle cells. PlGF is a pro-angiogenic factor that may play a role in placental development during pregnancy. sFLT-1 binds PIGF and prevents its interaction with its endogenous receptors. We sought to determine whether there is an association between midgestation serum 25-OH-D levels and angiogenic factor activity for development of preeclampsia. Methods: Healthy women with term deliveries were used as controls (n=24). We identified 25 patients with risk of PE who met one or more of the following criteria (IMC >35, arterial pressure >140/80 mmHg without proteinuria, age >40 years, multiparious, pregestational hypertension or diabetes mellitus. From 25 women with risk of PE, we identified 5 cases with PE and 20 women without PE. We measured midgestation maternal serum of 25-OH-D and the angiogenic factors using chemiluminescent immunoassay.The ratio of sFLT-1/PlGF was determined to have more predictive value than either variable alone. Results were analyzed by SPSS15. Results: There were significant differences in sFLT-1/PlGF ratio between controls, patients with risk and patients with preeclampsia (means SD:4.42.3, 7.13.7, 11.15.6 respectively, P:0.001). None of the angiogenic factors were significantly correlated with 25-OH-D, just found correlation between PlGF and sFLT-1 (correlation coefficient:0.446 P=0.001). The sFLT-1/PlGF ratio was the most predictive marker for preeclampsia (area under curve:0.820). Conclusions: Our findings confirm that sFlt-1/PlGF ratio is better marker of development of preeclampsia than a single angiogenic factor.
Private Health Institution Adrialab, Biochemistry and microbiology diagnostic laboratory 2 Department of physiology and biochemistry, Institute of Biology, Faculty of Natural science and mathematics, Skopje
Backgroud: With aging in women the number of primordial follicles is reduces, which is accompanied by a decrease in serum concentrations of anti Mulerian hormone. Several studies have shown that serum concentrations of anti mulerian hormone strongly correlate with the number of antral follicles and much more specific marker for predicting ovarial reserve and induction induction than age or other conventional serum tumor markers (follicle stimulating hormone, estradiol, or inhibin B). Methods: The method of work is a classic sandwich ELISA (Anti Mullerian Hormone (Mullerian Inhibitoring substance), Immunotech, Beckam Coulter, Webster, TX). Experimental group consisted of randomly selected 126 healthy girls and women aged between 22 to 50 years from Skopje, Republic of Macedonia. Were divided into five groups according to age to see the distribution of AMH by age. Results: Analyzed mean values of serum concentrations of AMH, median age and the standard deviation as a measure of individual deviation. The mean decrease in relation to increasing age is 1, 824SD 0,7999 ng/mL. And that decline serum concentration is harder to 35th year (approximately 2 ng/mL), after the 40th year decreasing intensity decreases and is approximately 1 ng/mL. Median age and median concentrations are strongly positioned in linear digression R> 0,5167, which confirms that there is a positive correlation between age and serum concentrations of AMH. Discussion: We confirmed that AMH strongly correlates with oocyte number. The results from our study are highly correlated with other similar studies of AMH. The serum level of AMH, was decreasing over aging, which is also correlated with decreasing of the number of antral follicles over aging. Thous makes AMH serum concentration a better marker marker for prediction the ovarial reseve in women then other markers.
biochimica clinica, 2013, vol. 37, SS
S393

T205 LOCAL INTERLEUKIN-8 AS BIOMARKER FOR ADVERSE PREGNANCY OUTCOMES IN PREGNANT WOMEN WITH BACTERIAL VAGINOSIS S. Cauci(1), M. Di Santolo(1), J.F. Culhane(2)
1 Department of Medical and Biological Sciences, School of Medicine, University of Udine, Italy 2 Department of Obstetrics and Gynecology, Drexel University College of Medicine, and Department of Pediatrics, University of Pennsylvania, School of Medicine, Childrens Hospital of Philadelphia, USA
T206 LEVELS OF PAPP-A IN THE 1ST TRIMESTER OF PREGNANCY IN RUSSIAN POPULATION WITH TWOSITE MONOCLONAL ELISA N. Chepurchenko(1), A. Burkov(1), O. Udalova(2), A. Obriadina(1)
1
RPC Diagnostic Systems, Nizhniy Novgorod, Russia Regional Clinical Diagnostic Center, Nizhniy Novgorod, Russia
2
Background: Preterm birth (PTB) is an increasingly prevalent, complex condition associated with a high risk of infant mortality and morbidity, including cerebral palsy, blindness, hearing loss, and also hidden disabilities such as school difficulties and behavioural problems that become apparent and persist into adolescence. PTB has a recognized complex multifactorial etiology. Numerous clinical studies have shown a direct relationship between reproductive tract infection/inflammation and PTB. Among microbial alterations, bacterial vaginosis (BV) has received particular attention, because it is a highly prevalent condition among childbearing and pregnant women potentially treatable by antibiotics. Non-invasive biomarkers to assess which pregnant women with BV are more likely to have an adverse pregnancy outcome are highly warranted. Our aim was to assess if vaginal interleukin-(IL)-8 measured in early pregnancy is associated with adverse outcome among BVpositive women. Methods: A total of 1,806 women were enrolled at <20 weeks gestation, in Philadelphia, Pennsylvania, USA. 800 women were BV-positive (Nugent 7-10), 707 of them had birth outcome data. Vaginal IL-8 concentrations was measured in 105 BVpositive women who had an adverse preterm outcome, including 66 preterm births (20-<37 weeks, of which 52 were spontaneous) and 14 late miscarriages (12-<20 weeks), and in 295 BV controls (term normal birthweight infants). To asses whether IL-8 associated risk had a U-profile, the upper (>66th percentile) and lower (<33rd percentile) tertiles of IL-8 were compared with the middle tertile (33rd to 66th percentile). Results: The 33rd percentile (first tertile) of vaginal IL-8 concentrations corresponded to 1140 pg/mL, and the 66rd percentile (second tertile) corresponded to 4830 pg/mL. None of the IL-8 tertiles was associated with increased risk for any adverse preterm outcome, nor preterm birth and miscarriage with or without exclusion of women with concurrent STDs. Conclusions: IL-8 is not a risk biomarker for preterm birth among BV-positive women in early gestation at average 12 weeks gestation.
Background. Pregnancy-associated plasma protein-A (PAPPA) is an established biochemical marker that is using in the prenatal screening for chromosomal abnormalities in the first trimester of pregnancy. Methods. Serum concentration of PAPP-A was measured using a monoclonal two-site ELISA (DS-EIA-PAPP-A). Results. Maternal serum PAPP-A values from 3579 singleton white pregnant women with no Down in pregnancy outcome and 16 women with Down syndrome fetuses (Central Russia, mean age 27 years, range from 16 to 43 years) have been measured. Results of specimens testing were used to calculate median (normal range, 5th and 95th percentile) for each gestational week: 8 week of gestation median 0.60 mIU/ml (from 0.19 to 1.82 mIU/m); 9 week of gestation median 0.88 mIU/ml (from 0.3 to 2.03 mIU/m); 10 week of gestation median 1.31 mIU/ml (from 0.28 to 2.84 mIU/m); 11 week of gestation median 2.02 mIU/ml (from 0.25 to 4.65 mIU/m); 12 week of gestation median 3.39 mIU/ml (from 0.22 to 6.46 mIU/m); 13 week of gestation median 4.27 mIU/ml (from 1.49 to 9.11 mIU/m). A statistically significant correlation between PAPP-A medians and maternal body weight was found (p <0,05). Medians were ranged according forth body weights (45 kg, 60 kg, 75 kg, 90 kg) for each gestational week: 8 week of gestation 0.73; 0.64; 0.51; 0.44 mIU/ml respectively, 9 week of gestation 1.26; 0.97; 0.84; 0.73 mIU/ml, 10 week of gestation 1.52; 1.34; 1.25; 1.13 mIU/ml, 11 week of gestation 2.32; 2.13; 1.84; 1.48 mIU/ml, 12 week of gestation 3.92; 3.35; 3.19; 2.41 mIU/ml, 13 week of gestation 5.57; 4.86; 3.68; 3.32 mIU/ml. A downward trend in the medians PAPP-A with an increase age of pregnant women (P >0,05) was observed. For risk calculation in prenatal screening PAPP-A concentrations are indicated as MOM (multiple of medians). The median of MOMs for women with no Down in pregnancy for our population was observed as 1.07 and for Down syndrome pregnancies the medians of MOMs for PAPP-A are increasing during the first trimester: 11 week of gestation 0.45; 12 week of gestation 0.47; 13 week of gestation 0.5. Conclusions. We defined normal limits of serum PAPP-A in Central Russia population with test DS-EIA-PAPP-A. It is necessary for estimation of abnormal pregnancy risk.
S394

T207 CORRELATION BETWEEN LIPID AND HORMONAL PROFILE AND INTIMA MEDIA THICKNESS AT DELIVERY IN PREGNANCIES COMPLICATED BY INTRAUTERINE GROWTH RESTRICTION C. Cosma(1), D. Faggian(1), M. Zaninotto(1), S. Visentin(2), E. Cosmi(2), A. Lapolla(3), M. Plebani(1) Department of Laboratory Medicine, University-Hospital, Padova, Italy 2 Department of Health of the Woman and Child; University of Padua, Italy 3 Department of Medical and Surgical Sciences, University of Padua, Italy Background: this study has examined the associations aIMT measured in utero during the early third trimester of pregnancy and neonatal and maternal lipid and hormone profile at delivery. Methods:AGA=97, SGA=57 IUGR=11 were enrolled. The lipid profile (LP), Adiponectin (A), and insulin (I) and leptin (L) were measured in mothers (M) and fetuses (F) at delivery. Data were analyzed by Analyset.it,considering significant p <0.05. Results: aIMT was significantly higher in IUGR than SGA and AGA (P <0.05).The LP maternal was significant higher compared to neonatal. The fetal concentration of L is lower than maternal but this difference is only significantly in IUGR (P=0.0343).Fetal L was positively correlated with birthweight (p <0.05).The fetal concentration of LDLox was significantly higher in IUGR than SGA and AGA (P <0.05).There was a hypertrigliceridaemia in SGA and IUGR groups than AGA (P= 0.05) and among SGA fetuses there is a significant positive correlation between fetal aIMT and tryglicerids concentration (P <0.05).There is a negative correlation between oxidized LDL and aIMT that among IUGR and SGA was not significantly positive.A and I have a significant positive correlation between maternal and fetal values.The concentration of A was significantly lower in the M than in the F (P <0.05) while NEFA were higher. Maternal A and NEFA in SGA group were significantly higher than in AGA ones (P <0.05). in SGA we had a significant negative correlation of increased fetal I with low aIMT, while in AGA we had no correlation and in IUGR a non significant positive correlation.There was in a non-significant way a correlation between high aIMT and maternal high I (SGA and IUGR). In AGA there is a significant negative correlation between mother A and fetal aIMT, while in the SGA group the correlation was significantly positive (P <0.05). Conclusion: metabolic syndrome of the mother might reflect on the fetal metabolism and probably influence the health of fetal vessels. LP of the mother is correlated with LP of the F and newborns with low birthweight had higher chances to present higher colesterol and LDL. Low birthweight was associated with dislipidemic state as well as high tryglicerids were associated with increase aIMT in IUGR.
1
T208 MATERNAL SERUM SCREENING BESPOKE CALIBRATION, DOES IT MAKE A DIFFERENCE? K. Donalson(1), S. Turner(1), N. Mumtaz(1), B. Nix(2) Biochemistry Department, Royal Victoria Infirmary, Newcastle, UK 2 Medical School, Heath Park, Cardiff, UK Background: Four biochemical markers, alpha-fetoprotein (AFP), free or total human chorionic gonadotrophin (hCG), unconjugated estriol (uE3) and Inhibin A are used in the risk calculation for Trisomy 21 in the second trimester of pregnancy (14+2 20 weeks). All 4 markers were not originally designed for this purpose, consequently the standard concentrations supplied by the manufacturers are inappropriate for screening. It was considered if a bespoke set of standards was used this would improve the specificity and sensitivity of each assay. Methods: Beckman Coulter Access2 AFP, Total hCG, uE3 and Inhibin A kit methodology was used in all assays. Appropriate in house calibrator levels were calculated by determining the 5th and 95th centiles of each of the four markers from samples previously run in the laboratory. The manufacturers standards were run to establish their calibration curve. This was followed by the in house standards and a set of six low, medium and high quality controls (QCs). Results: A linear curve fit was the most appropriate for AFP, total hCG and Inhibin A in house calibration. A log 10, log 10 linear transformation model was needed for uE3. Each QC level was read against both calibrations for each marker. Standard deviations (SD) were calculated for each of the controls for all four markers. AFP showed a mean SD for the low medium and high controls of 0.32525, 0.89245, 2.0532, total hCG 3.7071, 5.1379, 9.8352, uE3 0.01259, 0.01763, 0.04234 and Inhibin A 2.7796, 4.0602, 9.3653 for the manufacturers calibration curve. For the in house calibration curve the SDs were AFP 0.28366, 0.7454, 1.58916, total hCG 3.13719, 4.3398, 8.0061, uE3 0.01032, 0.01783, 0.0521 and Inhibin A 2.7357, 3.8884, 8.8638. Conclusion: It appears that the in house calibration curves provide a more then adequate fit over the range of standards used. The standard deviations associated with the back calculated QC values are, in the majority of cases smaller using the local in house calibration curve than those provided using the manufacturers software thus improving specificity and sensitivity.
1
S395

T209 RESULTS SCREENING TEST FIRST TRIMESTER OF PREGNANCY AND PRENATAL DIAGNOSTICS KARYOTYPE IN BANJA LUKA CLINICAL CENTER A. orevi, Mirjani-Azari, Maleevi, Milju Department of laboratory diagnostics, Clinical Center Banja Luka, Bosnia and Herzegowina Introduction: Downs syndrome (DS) is the most common chromosomal disorder in live births with a prevalence of 1: 650 800. Combined screening in the first trimester of pregnancy (10 to 14 weeks gestation) is used for the statistical evaluation of risk for Downs (DS), Edwards (ES) and Pataus syndrome (PS). Trisomy 21 is associated with biochemical ( pregnancyassociated plasma protein-A, PAPP-A and free-beta hCG, -hCG ) and ultrasound markers ( nuchal translucency, NT and crown rump length, CRL ). Objective: Our aim is to present the positive results of the screening test of the first trimester and results of karyotyping amniotic fluid samples obtained by amniocentesis in the second trimester of pregnancy. Materials and Methods: The study population includes 1437 pregnant women between 10 and 14 gestational weeks. Biochemical markers, free -hCG and PAPP-A were performed by electrochemiluminescence imunoassay on Roche COBAS E601 analyzer. Risk calculations were performed using antenatal screening software SsdwLab v5.0 (Roche). Karyotype analysis of fetus was performed on 16 metaphases. At 8 metaphases were done using GTG bands, while the other 8 metaphases determination was performed using classical colored slides. Results: From 1437 pregnant women tested, high risk (cut-off >1: 250) was found in 171 (12.00%). Of the 171 high risk pregnancies, 69 (40.35%) had an increased risk for DS based on biochemical and ultrasound parameters, while ES and PS increased risk was found in 5.20% of patients. In 171 pregnant women with an increased risk for DS, amniocentesis was performed on 111, of which 108 (97.29%) had normal findings. Pathological karyotype was found in 3 cases (2.71%). Conclusion: Screening plays an important role in the early detection of risk for DS and it is recommended for pregnant women, regardless of age. Early diagnosis can reduce the incidence of late stage termination, which would otherwise be routinely applied at the request of the patient. Screening tests do not give a definitive prenatal diagnosis. The final diagnosis should be made only by applying invasive methods and karyotyping. T210 A NON-INVASIVE PRENATAL DIAGNOSIS EXPERIENCE IN SOUTHERN PART OF TURKEY: HIGH RESOLUTION MELTING ANALYSIS TO DETECT THE PATERNAL MUTATIONS OF CELL-FREE FETAL DNA IN MATERNAL PLASMA E. Dundar Yenilmez(1), A. Tuli(1), C. Evruke(2) University of ukurova, Faculty of Medicine, Department of Medical Biochemistry 2 University of ukurova, Faculty of Medicine, Department of Obstetrics and Gynecology Backround: We examined paternal mutations in cell-free fetal DNA (cff-DNA) by high-resolution melting analysis (HRM) and investigated differences in the levels of fetal and total cff-DNA in mothers with -thalassemia or sickle-cell traits. Methods: We examined cff-DNA from 32 pregnant women with -thalassemia, 57 pregnant women with sickle cell anemia (HbS), and 15 healthy pregnant women as control subjects. The plasma was separated from maternal peripheral blood within 1 h of sample collection, and the DNA was extracted from 1-mL plasma. DYS14 and -globin genes were analyzed to detect fetal and total DNA by quantitative real time PCR. Paternal mutations were examined in cff-DNA by HRM analysis. Results: We observed increased multiples of the median (MoM) levels of DYS14 in fetuses affected with HbS and decreased DYS14 MoM levels in -thalassemia-affected fetuses. We successfully detected 22 paternal -thalassemia mutations using HRM analysis in cff-DNA when the fetus had different mutation as the mother in -thalassemias.. However, we could not distinguish cff-DNA from HbS-affected individuals. Conclusion: Fetal DNA quantification and MoM levels were not informative for detecting HbS and -thalassemias in early pregnancy. HRM is a useful method for non-invasive prenatal diagnosis for detecting paternally inherited mutations in the maternal plasma.
1
S396

T211 NONINVASIVE PRENATAL DIAGNOSIS OF CYSTIC FIBROSIS S. Galbiati(1), A. Monguzzi(1), N. Soriani(1), F. Lalatta(2), M. Seia(3), F. Damin(4), M. Cretich(4), M. Chiari(4), M. Ferrari(5)
1
T212 RELATIONSHIP BETWEEN INSULIN SENSITIVITY AND LEPTIN IN PREGNANT WOMEN WITH GESTATIONAL DIABETES MELLITUS M. Genova(1), K. Todorova(2), K. Tzatchev(1) 1Chair of Clinical laboratory and Clinical Immunology, Medical Faculty, Medical University, Sofia, Bulgaria 2University Clinic of Endocrinology, Medical University, Pleven, Bulgaria Background: Gestational diabetes mellitus (GDM) individuals have higher body fat than healthy. The obesity in GDM patients results in elevation of leptin concentrations. The aim of this study was to determine leptin influence on insulin sensitivity during the late pregnancy. The index QUICKI-IS is used to assess the insulin sensitivity. Methods: Pregnant women (n=102, gestational weeks 24 vs 254) were included in the study . Based on 75 g 2-h OGTT, the participants were stratified into the following groups (IADPSG criteria): I-st group: healthy pregnant women with normal glucose tolerance (NGT), n=49; II-nd group: pregnant women with GDM, n=53. All women were tested for venous plasma glucose (by measuring of oxygen consumption ), serum insulin ( by ECLIA) and serum leptin (sandvich ELISA). The index QUICKI is defined as follows: QUICKI=1/[(log(Ins0)+log(Glu0)], where Ins0 is the fasting plasma insulin level (U/mL) and Glu0 is the fasting blood glucose level (mmol/L). Results: Both groups were compared for gestational age (non significant difference) and BMI (significal difference, P=0.011). Women with GDM had significantly higher levels of serum insulin (13.848.43 vs.11.357.38 U/mL, P=0.02), higher levels of venous plasma glucose (5.931.04 vs. 4.630.28, P <0.0001 ) and serum leptin levels (16.9611.89 vs.8.495.38 ng/mL, P=0.002) in comparision to NGT group. In contrast, insulin sensitivity calculated from QUICKI-IS was significantly lower for GDM group (0.560.11 vs. 0.650.1, P=0.001). Spearman correlation analysis revealed significant correlation between leptin and QUICKI index ( r= -0.740 for NTG and r= 0.728 for women with GDM). Conclusions: The women with GDM had significantly higher levels of serum insulin and leptin concentrations in comparision to the NGT group. A signicant negative correlation between leptin levels and insulin sensitivity is established by us. Our findings suggest that leptin might contribute to development of GDM by decreasing insulin sensitivity.
San Raffaele Scientific Institute, Genomic Unit for the Diagnosis of Human Pathologies, Center for Translational Genomics and Bioinformatics, Milan, Italy. 2 UOSD di genetica Medica, Fondazione IRCCS C Granda Ospedale Maggiore Policlinico, Milano. 3 Laboratorio di Genetica Molecolare, Fondazione IRCCS Ca Granda Ospedale Maggiore Policlinico. 4 Istituto di Chimica del Riconoscimento Molecolare, C. N. R.,Milano, Italy. 5 Universit Vita-Salute San Raffaele, Milan, Italy. 6 Diagnostica e Ricerca San Raffaele SpA, Milan, Italy. Background: The possibility to retrieve fetal DNA from maternal plasma has made available a new source of fetal genetic material for noninvasive analysis of numerous fetal pathological conditions. Due to the scarcity of fetal DNA in maternal plasma and the difficulty in detecting fetal mutated alleles, noninvasive prenatal diagnosis (NIPD) has not yet attained a widespread clinical application. Our goal is to develop and validate accurate tests for NIPD which combine high sensitivity and ease of use during the first trimester of pregnancy. The project is focused on the set up of methodologies for the identification of fetal paternally inherited mutations/polymorphisms in CFTR gene. Methods: Two research lines are been investigated: the development of amplification protocols based on COamplification at Lower Denaturation temperature-PCR (COLD-PCR) combined with Sanger sequencing and highly sensitive microarray substrates which could allow the detection of fetal minority sequences without any enrichment strategy. Results: Full COLD-PCR: We have developed assays for the identification of fetal paternally inherited F508del and G542X mutations in maternal plasma. In total, 5 diagnoses were performed including 4 cases where the father carried the F508del mutation (in 2 of these the fetus inherited the paternal mutation) and 1 case where the father carried the G542X mutation and the fetus inherited the wild-type allele. In all cases the fetal paternal mutated allele was not detectable with conventional PCR while it became evident at different proportions after enrichment with full COLD-PCR. The results obtained were in complete concordance with those obtained on fetal DNA extracted from chorionic villi. Microarray: We have developed assays for the identification of fetal paternally inherited F508del and G542X mutations in maternal plasma in all the couples previously tested. The results obtained were in complete concordance with those obtained with COLD-PCR. Conclusions: The application of the full COLD-PCR protocol and the microarray might be extended to any other CF gene mutation for noninvasive prenatal diagnosis of cystic fibrosis and other common genetic diseases and has the potential to be easily transferable to clinical diagnostic laboratories.
S397

T213 BIOCHEMICAL CHARACTERIZATION OF HEALTHY NEWBORNS IN THE PROVINCE OF VILLA CLARA, CUBA T. Gmez Hernndez, L.C. Bequer Mendoza, L. Salazar Torres, L. Prez de Alejo Rodrguez, Y. Romero Garca, O.R. Molina Hernndez, M. Villar Valds, O.L. Gonzlez Gonzlez, V. Hernndez Moreno, A. Mollineda Trujillo, A.D. Alfonso Pestano, M. Rodrguez Prez Unidad de Investigaciones Biomdicas. Universidad de Ciencias Mdicas de Villa Clara, Cuba. Background: Laboratory studies in newborns constitute an important tool in the diagnosis and treatment of neonatal conditions. In our region are used as the reference limits reported in the literature, which generally come from abroad, which complicates the evaluation and interpretation of normal ranges in the neonate and, undoubtedly, play an important role in behavior and neonatal clinical management. In order to characterize biochemically healthy newborns Villa Clara was quantified various indicators in cord blood. Methods: We studied 80 infants (40 females and 40 males) who met the inclusion criteria, and 22 biochemical parameters were measured in umbilical cord blood. Were established reference values for each parameter, taking the interval corresponding to the central interval interpencentile (95%) delimited by the 2.5 and 97.5 percentiles. Results: The reference intervals determined for blood chemistry parameters were: urea: 2.01-5.45 mmol/L, creatinine: 24.6595.75 mol/L, SGPT: 0.47-28.70 U/L, SGOT: 3.35-89.20 U/L, total bilirubin: 20.86-148.37 mol/L, alkaline phosphatase: 97.37-421.87 U/L, GGT: 12.13-176.12 U/L, cholesterol: 0.592.96 mmol/L, triglycerides: 0.07-0.97 mmol/L and VLDL: 0.310.45 mmol/L. Regarding immunological parameters the results were: IgM: 0.04-0.34 g/L, IgG: 7.26-13.43 g/L, IgA: 0.00-0.08 g/L, C3: 0.49-1.50 g/L and C4: 0.03-0.22 g/L. About micro minerals the results were: iron: 10.21-36.13 mol/L, copper: 6.18-15.82 mol/L and zinc: 11.15-24.09 mol/L; and macro minerals: sodium: 102.60-150.40 mmol/L, potassium: 3.249.69 mmol/L, calcium: 1.08-2.72 mmol/L and magnesium: 0.451.15 mmol/L. Conclusions: Knowledge of reference intervals presented in this study is an important social, scientific and economic, offering mothers and their children the opportunity to receive a quality service that ensures future welfare, and gives professionals health a new tool for the early diagnosis and successful of different complications in this age group, thus avoiding unnecessary medication and decreased hospital stay. T214 HORMONE LEVELS IN EXTREME PREMATURITY: ESTABLISHMENT OF GESTATIONAL APPROPRIATE REFERENCE INTERVALS FOR FSH, LH AND PROLACTIN R. Greaves(1), J. Pitkin(2), C.S. Ho(3), J. Baglin(1), M. Zacharin(4)
1 2
RMIT University, Victoria, Australia Murdoch Childrens Research Institute, Victoria, Australia. 3 Prince of Wales Hospital, Shatin, Hong Kong SAR 4 Department of Endocrinology and Diabetes, The Royal Childrens Hospital, Victoria, Australia Background: Premature neonates, in particular extremely premature neonates, suffer from a number of common morbidities and prophylactic maternal glucocorticoid administration prior to delivery is now almost universal. Immaturity of the endocrine system and its potential impact on morbidity is the subject of numerous studies. Reports suggest significant differences in serum levels of pituitary hormones in extremely preterm neonates compared to older preterm and full term neonates. However, whilst serum hormones are often measured there is little normative data available for this cohort. The aim of this study is to develop reference intervals for three pituitary hormones measured in babies born between 25 and 32 weeks gestation. Methods: Blood was collected from 249 (129 male and 120 female) extremely premature infants on successive occasions whilst inpatients in neonatal wards. All infants in this cohort did not have any evidence of ambiguous genitalia or other endocrine related abnormalities Samples (median of three per neonate) were collected in conjunction with routine capillary blood collection. The serum was analysed for prolactin, follicle stimulating hormone (FSH) and luteinizing hormone (LH) by automated electrochemiluminescence immunoassay (Roche Cobas 8000 - E602 module). Infants that did not survive beyond the equivalent of term were excluded from the statistical analysis. Results: Reference intervals were established with samples collected from 230 of the 249 extremely preterm infants; representing 521 (267 male and 254 female) samples analysed throughout the first six weeks of life. The distribution was nonGaussian and initial assessment established the 95% central range for each pituitary hormone. For male extremely preterm infants the ranges are: prolactin 605 - 4798 mIU/L; FSH 0.2 4.7 IU/L; and LH 0.3 8.3 IU/L. The female extremely preterm infant ranges are: prolactin 666 5854 mIU/L; FSH 5.7 >174 IU/L; and LH 0.4 167 IU/L. Conclusions: We describe gestation appropriate reference intervals for three pituitary hormones measured in the first six weeks of life for babies born <32 weeks gestation. Utilisation of these reference intervals will permit the correct and timely interpretation of results for this preterm population.
S398

T215 ANTIOXIDANT DEFENSE IN PREGNANT WOMEN U. Grudzien, K. Gawlik, D. Pawlica, B. Solnica Department of Diagnostics, Chair of Clinical Biochemistry Jagiellonian University Medical College, Krakow, Poland Background: Oxidative stress is described as an imbalance between the production of reactive oxygen species (ROS) and the level of antioxidant activity. Pregnancy is a state of oxidative stress arising from increased placental mitochondrial activity. Aim of the study was to assess differences in the levels of antioxidants in healthy pregnant women compared to nonpregnant women. Methods: Antioxidants measurements were performed in 29 healthy pregnant (HP) and 22 healthy non-pregnant (HNP) women. Activity of glutathione reductase (GR), glutathione peroxidase (GPx) and FRAP (Ferric Reducing Ability of Plasma) were measured in plasma, CRP and uric acid (UA) levels were measured in serum. Results: In plasma GPx activity a trend towards lower values in pregnant women (560,69 U/L for HP, 651,12 U/L for HNP; P=0,62), and towards higher values in FRAP in pregnant group (1,115 mmol/L for HP, 1,067 mmol/L for HNP, P=0,36) was found. No difference in the GR activity and in uric acid levels (58,04 U/L for HP and 59,44 U/L for HNP; 2,02 mol/L for HP and 2,07 mol/L for HNP) was observed. Significant increase in serum CRP was observed in pregnant compared to nonpregnant group (1,78 mg/L for HP, 0,44 mg/L for HNP, P < 0,01). Significant correlations between GPx activity and uric acid levels (P=0,026 r= 0,33) in both groups, between uric acid and FRAP in the HNP group (r=0,74, P <0,001) and between FRAP and CRP in the HP group (r=0,4340, P=0,039) were found. Conclusions: Our results indicate that the antioxidant defense assessed based on GR, GPx, FRAP and uric acid measurements does not differ significantly in pregnant compared to non-pregnant women. Uric acid appears to be an important low molecular weight antioxidant. Our findings suggest that oxidative stress in pregnancy is the result of increased production of ROS, which is not balanced by unchanged antioxidant defense. Significantly higher serum CRP levels in pregnant women indicate mild inflammatory response in this group. T216 COPEPTIN: A NOVEL EARLY BIOMARKER FOR FOETAL GROWTH RESTRICTION? A.T. Hansen(1), P. Sandager(2), N. Uldbjerg(2), A.M. Hvas(1) Department of Clinical Biochemistry, Aarhus University Hospital, Denmark. 2 Department of Obstetrics and Gynaecology, Aarhus University Hospital, Denmark Background: Pregnancies complicated by foetal growth restriction (FGR) is a stressful maternal state with increased neonatal mortality and morbidity. FGR is usually diagnosed by ultrasound in the late second trimester. Early biochemical prediction of FGR is of great interest since early treatment has been shown to reduce risk of developing FGR in high risk pregnancies. An increased level of copeptin is found associated to various stress conditions, such as ischaemic stroke and sepsis. Copeptin remains stable ex vivo and is easy to measure in contrast to vasopressin and cortisol. The aims our study were 1) To establish first and second trimester reference ranges for copeptin, and 2) To investigate if first and second trimester levels of copeptin are elevated in pregnancies later complicated by FGR. Methods: In this case-control study, we measured copeptin levels in maternal serum at gestational week 12 and 19. The blood samples were collected at Aarhus University Hospital, Denmark from 2002 to 2004. Pregnant women who subsequently developed FGR were defined as cases (N=39) and each case was matched on maternal age with 3 pregnancies not complicated by FGR (controls, N=119). Levels of Copeptin were analysed using a commercial kit from BRAHMS (Copeptin ultrasensitive Kryptor); an automatic immune fluorescent analysis. Reference ranges were calculated as 95% prediction intervals and presented as anti logarithms with 90% confidence intervals. Paired and unpaired t-test was performed testing the null-hypothesis of no difference within and between groups. Results: No significant changes were found in Copeptin levels from week 12 to 19 in controls (P=0.61). Copeptin levels declined significantly from week 12 to 19 cases (P=0.02). No significant difference in copeptin levels was found between cases and controls in week 12 (P=0.10) and week 19 (P=0.81). Conclusions: The decline in copeptin in FGR pregnancies was not clinically important, since it does not add any predictive information on FGR. An unspecific marker as copeptin is not a suitable single biomarker for prediction of FGR.
1
S399

T217 sFlt-1/PlGF RATIO AS BIOMARKER IN THE DIAGNOSIS OF PREECLAMPSIA IN SINGLETON AND MULTIPLE PREGNANCIES. I. lvarez-Fernndez(1), B. Prieto(1), Y. Ruano(2), A. Escudero(2), F.V. lvarez(1)
1
T218 TRIMESTER-SPECIFIC REFERENCE LIMITS FOR THYROID HORMONES IN RUSSIAN POPULATION WITH THE MONOCLONAL ELISA E. Kochina, E. Matveeva, A. Burkov, A. Obriadina Laboratory of New Technology Department, RPC Diagnostic Systems, Nizhniy Novgorod, Russia Background: Maternal thyroid dysfunction may be associated with increased risks of adverse pregnancy and perinatal outcomes. Therefore, a local reference range for thyroid hormones in pregnant women is required. Methods: Samples were tested for thyroid stimulating hormone (TSH), free thyroxine (FT4), total thyroxine (TT4), free triiodothyronine (FT3), total triiodothyronine (TT3), antibodies to thyroid peroxidise (TPO-Ab), human chorionic gonadotropin (HCG). All testing was performed using ELISA kits (RPC Diagnostic Systems, Russia). Results: 360 serum samples from white women (Central Russia and Volgo-Viatsky Region, mean age 29) with normal pregnancy were examined. Reference limits (2.5th and 97.5th percentile) were calculated using TPO-Ab negative specimens with normal levels of HCG. Trimester specific median, and 2.5th 97.5th percentile ranges for each analyte are reported below: TSH 1st trimester - median 1.26 IU/mL, normal interval from 0.10 to 2.84 IU/mL, 2nd trimester - median 1.39 IU/mL, normal interval from 0.22 to 3.25 IU/mL, 3rd trimester median 1.61 IU/mL, normal interval from 0.47 to 3.32 IU/mL; FT4 1st trimester - median 13.7 pmol/L, normal interval from 10.9 to 22.9 pmol/L, 2nd trimester - median 13.0 pmol/l, normal interval from 9.9 to 16.9 pmol/L, 3rd trimester - median 10.7 pmol/L, normal interval from 8.3 to 14.9 pmol/L; TT4 1st trimester - median 123 nmol/L, normal interval from 84 to 172 nmol/L, 2nd trimester - median 133 nmol/l, normal interval from 107 to 165 nmol/L, 3rd trimester median 127 nmol/L, normal interval from 97 to 158 nmol/L; FT3 1st trimester - median 2.97 pg/mL, normal interval from 1.96 to 5.0 pg/mL, 2nd trimester median 2.99 pg/mL, normal interval from 2.25 to 4.54 pg/mL, 3rd trimester - median 2.72 pg/mL, normal interval from 2.11 to 3.97 pg/mL; TT3 1st trimester - median 1.4 ng/mL, normal interval from 0.86 to 2.15 ng/mL, 2nd trimester - median 1.61ng/mL, normal interval from 1.12 to 2.33 ng/mL, 3rd trimester - median 1.73 ng/mL, normal interval from 0.94 to 2.42 ng/mL. Conclusions: We determined trimester-specific normal limits for serum thyroid hormones in Central Russia population with the ELISA (RPC Diagnostic Systems, Russia). These intervals may reduce the risk of mis-interpretation of thyroid function during pregnancy.
Servicio de Bioqumica Clnica. Hospital Universitario Central de Asturias, Oviedo, Spain 2 Servicio de Obstetricia y Ginecologa, Hospital Universitario Central de Asturias, Oviedo, Spain Background: Preeclampsia (PE) is a disorder characterized by the onset of hypertension and proteinuria after 20 weeks gestation. Angiogenic/anti-angiogenic factors, as placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFlt-1), have been purposed as biomarkers in the monitoring of women at high PE risk. The aim of the study is to assess the utility of sFlt-1/PlGF ratio at the obstetric triage area to distinguish those women who will develop PE from those who will not. It has been described that sFlt-1 and PlGF show different concentrations in singleton and twin pregnancies, so we want to demonstrate if these differences are clinically relevant to diagnose PE when sFlt-1/PlGF ratio is used. Methods: Serum samples from 214 pregnant women between 20 and 40 weeks gestation were collected; 192 were singleton and 22 multiple pregnancies. sFlt-1 and PlGF concentrations were determined by electrochemoluminescence immunoassay (Cobas 6000, Roche Diagnostics). Performance of sFlt-1/PlGF ratio was analyzed using receiver operating characteristic (ROC) curve. Differences between groups were compared using Mann-Whitney U test. SPSS software was used for data analysis. Results: Subjects were classified in two groups according to gestational age at sampling time: 34 (n=58) and >34 weeks gestation (n=156). Women who developed PE (31 and 36 in each group, respectively) showed higher values of sFlt-1/PlGF ratio than normal pregnancies. Areas under ROC curves for the ratio were 0,857 (95% confidence interval: 0,751-0,962) and 0,797 (95% confidence interval: 0,711-0,884), for 34 and >34 weeks gestation, respectively. Cutoff values for the ratio with best performance were 22 for 34 and 54 for >34 weeks gestation. Ratios were significantly different when singleton vs multiple non-PE pregnancies were compared at 34 weeks gestation, but similar diagnostic power was obtained either including or not multiple pregnancies. Conclusions: Increased values of sFlt-1/PlGF ratio may be useful to identify women who will develop PE at the obstetric triage area in the third trimester. In this study so far, differences observed in sFlt-1/PlGF ratio between singleton and multiple gestations do not seem to have clinical relevance to diagnose PE in the third trimester.
S400

T219 THYROID DISORDERS IN WOMEN OF REPRODUCTIVE AGE IN KOSOVO A. Kotori(1), V. Haxhibeqiri(2), M. Zhuri(1), V. Mulliqi-Kotori(3), S. Haxhibeqiri(1)
1
T220 SELECTIVE TRANSFERT OF A SINGLE BLASTOCYST STAGE EMBRYO: A COMPARISON VITRIFICATION / SLOW FREEZING M. Kuentz(1), L. Janny(1), H. Pons-Rejraji(1), A.S. Gremeau(2), L. Dejou-Bouillet(2), J. Pouly(2), F. Brugnon(1)
1 CHU Clermont Ferrand, CHU EStaing, Ple Gyncologie Obsttrisque Reproduction Humaine, AMP, CECOS, 1 Place Lucie Aubrac, Ferrand 2 CHU Clermont Ferrand, CHU EStaing, Ple Gyncologie Obsttrisque Reproduction Humaine, unit mdecine de la reproduction, 1 Place Lucie Aubrac, Ferrand
Medical Institution for Laboratory Diagnostics Li-ori Prishtina Kosovo 2 University Clinical Center, Department of Clinical Biochemistry, Prishtina, Kosovo 3 University Clinical Center, Pediatric Clinic, Department of Endocrinology Prishtina, Kosovo Background: Thyroid diseases are very common in women of reproductive age. Normal thyroid function is essential for maintaining normal reproductive capacity. Any type of thyroid dysfunction undiagnosed and untreated may be the cause of infertility. Goal of the investigation was to evaluate thyroid disorders in women of reproductive age in Kosovo. In this study were included women of reproductive age 18 to 40 years, frequented in our clinic, from January 2006 up to December 2010, who are trying to conceive and women that have undergone methods of assisted reproduction technologies. Testing of Thyroid-stimulating Hormone (TSH), triiodothyronine (T3), thyroxine(T4), Thyroid antibodies (aTPO-Ab, aTg-Ab) and prolactin was performed in all women. A total of 860 infertile women underwent thyroid screening test and 176 females in a control group. Of these, 16.1% (106 of 860) were with hypothyroidism (subclinical and overt Hypothyroidism). Hypothyroxinemia was tested in women with elevated TSH and women with TSH within range. Hypothyroxinemia was seen in 0.39% (4 of 754) of the tested women with TSH within range and was seen in 3% (4of 106) of the tested women having elevated TSH. Women with elevated TSH received aTPO Ab test and of these, 14.9% (16 of 106 ) tested positive. Females with thyroid disorders and ovarian dysfunction have increased period of infertility in comparison with control group. Thyroid disorders are one of the causes of infertility. Women that have undergone methods of assisted reproduction technologies and have thyroid disorders with positive TPO-Ab have increased risk of miscarriage. Based on many scientific facts we therefore propose that a systematic screening of TSH, freeT4 and TPOAb could be considered in all women with infertility
Introduction: Data from the literature tend to show the benefit of the blastocyst stage embryo vitrification compared to slow freezing. However, no study has examined the contribution of vitrification under a policy of selective transfer of a single embryo. The aim of our study was to compare the outcome of the first and second attempts at IVF/ICSI couples who benefited from a transfer of a single embryo at the blastocyst stage with slow freezing or vitrification of embryos. Materials and methods: This retrospective study compares two groups of pairs of balanced numbers (n=35) for whom cryopreservation of surplus blastocysts was performed by slow freezing (n=128) or vitrification (n=130). Recruitment periods protocols for slow freezing and vitrification respectively extend from January 2007 to December 2011 and from November 2010 to March 2012. Couples enrolled in their first or second attempt at IVF/ICSI, including the womans age is less than or equal to 36 years with at least one supernumerary blastocyst quality for cryopreservation (Gardner classification: B3/or B4 AA, AB, BA or BB) are included. Methods of embryo culture (G series), slow freezing/thawing (G-Blast Kit Freeze/Thaw Kit G-Blast) and vitrification/warming (Rapid Lives Blast, Rapid Warm Blast) are performed according to the recommendations provider (Vitrolife, Sweden). Results: The two groups of couples have considered similar demographic parameters. After freeze/thaw slowly, the survival rate of blastocysts was lower (68%, n=63 blastocysts thawed) compared with blastocysts vitrified/warmed (86%, n=35 blastocysts warmed). The number of embryos thawed/warmed and transferred is similar for both groups (1.090.06 vs. 1.110.07). After thawed embryo transfers or reheated, a significant difference was observed concerning the implantation rate (slow freezing: 9% vitrification: 36%) and pregnancy rate (slow freezing: 4%; vitrification: 32%) between the two groups (P <0.05). Conclusion: Culture until the fifth day of embryonic development allows the application of an effective policy of selective transfer of a single blastocyst. However, slow freezing (or) blastocyst (s) supernumerary (s) appear to lose this benefit. The preliminary results obtained by vitrification/warming are very encouraging
S401

T221 PLACENTAL GROWTH FACTOR (PLGF) AS MARKER FOR PREDICTING PRE-ECLAMPSIA. V. Manolov(1), B. Marinov(2), V. Vasilev(1) Medical University, Alexandrov hospital - Central Clinical Laboratory 2 Medical University, Maichin dom hospital - General Gynecological Ward Maternity Risk Clinic Background. Placenta growth factor (PlGF) is a homodimeric glycoprotein, 46-50 kDa in size, belonging to the vascular endothelial growth factor (VEGF) sub-family. We perform this study to evaluate the ability of placental growth factor (PlGF) measurements in gestational weeks 24-28 to predict preeclampsia. Methods. In this study we include 30 pre-eclamptic and 40 healthy pregnant women. Levels of PlGF were measured in the stored serum. Pre-eclamptic subjects were also divided into women with early-onset (<37 weeks) and women with lateonset pre-eclampsia (> or =37 weeks). Results. Levels of PlGF were found to be higher in the preeclamptic group in both trimesters. No differences were found between early- and late-onset pre-eclamptic. This finding was confirmed by high specificity (97.06%) and sensitivity (88.89%), a positive predictive value (94.12%) and a negative predictive value of (94.29%) for PlGF. Conclusions. PlGF might be used as markers for predicting pre-eclampsia.
1
T222 ANGIOTENSIN-CONVERTING ENZYME (ACE) AS A PREDICTIVE MARKER OF PREECLAMPSIA AT 19-21 WEEKS OF GESTATION A. Martnez-Ruiz, J. Prez-Fornieles, J.A. Vlchez-Aguilera, N. Sancho-Rodrguez, I. Cebreiros-Lpez, I. De MiguelElzaga, M. Martnez-Villanueva, M.D. Sarabia-Meseguer, P. Martnez-Hernndez Clinical Analysis Service of University Hospital Virgen de la Arrixaca, Murcia, Spain Background: Preeclampsia (PE) is a syndrome of unknown etiopathogenesis. Previous studies about preeclampsia have focused on the increase in free radicals in the feto-placental unit with poor perfusion. It is believed that the reninangiotensin-aldosterone system (RAAS) has a role in the poor perfusion of the placenta. It is uncertain whether there is a preexisting impairment in RAS in preeclamptic pregnant women or not. The aim of this study was to investigate the role of angiotensin-converting enzyme (ACE) serum levels as a useful marker in the prediction of PE at 19-21 weeks of gestation. Methods: We included 101 pregnant women with a priori risk of developing PE (inclusion criteria: preexisting renal disease, chronic hypertension without proteinuria, history of PE in a previous pregnancy, etc) in a prospective study of one year and 80 controls. PE was diagnosed if normotensive woman had two systolic blood pressure measurements of 140 mmHg repeated in a 6 hour interval and/or a diastolic blood pressure measurements of 90 mmHg after the 20th gestation week; together with proteinuria of more than 300 mg/24 hours specimen (6 h interval). ACE levels were measured by spectrophotometry assay (Biosysten, ATOM). Students t test was used to compare ACE serum levels in both groups: Group 1 (pregnant with PE) (G1), Group 2 (control group of normotensive pregnant women) (G2). We used ROC curve analysis to calculate the area under the curve (AUC). Results: Of the 101 pregnant women included, 9 developed PE. When we compared G1 and 2, we obtained significant differences for ACE serum levels (53.719.7 vs 30.412 U/L) (P <0.001). AUC for ACE was 0.882 and the cutoff point with better sensitivity and lower false positive rate was 40.4 U/L (sensitivity=66.6%, specificity=85%, positive predictive value=33.3%, negative predictive value=95.7% and false positive rate=15%). Conclusions: ACE is useful marker to predict the subsequent development of PE at 19-21 weeks of gestation.
S402

T223 SOLUBLE FMS-LIKE TYROSINE KINASE 1: A PROMISING DIAGNOSTIC BIOMARKER FOR ECTOPIC PREGNANCY A. Martnez-Ruiz, J. Prez-Fornieles, J.A. Vlchez-Aguilera, N. Sancho-Rodrguez, I. Cebreiros-Lpez, I. De MiguelElzaga, M. Martnez-Villanueva, M. D. Sarabia-Meseguer, P. Martnez-Hernndez Clinical Analysis Service of University Hospital Virgen de la Arrixaca, Murcia, Spain Background: Ectopic pregnancy (EP) remains a considerable cause of maternal morbidity and mortality worldwide. It is diagnosed using a combination of transvaginal ultrasound and serial b-human chorionic gonadotrophin (bhCG) serum levels. Therefore is important to develop a biomarker that can differentiate between an EP or an intrauterine implantation. The aim of this study was to investigate if placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFlt1) serum levels were useful markers to differentiate an EP of a missed abortion (MA). Methods: We included 70 pregnant women (35 EP and 35 MA) at 6-9 weeks of gestation with hCG serum levels between 800-3500 UI/L. PlGF and sFlt-1 levels were measured by electrochemiluminescence assay (Roche Diagnostics, Manheim, Alemania). U Mann-Whitney test was used to compare the two markers in both groups and we used ROC curve analysis to calculate the area under the curve (AUC). Results: When we compared EP group with MA group, we didnt find significant differences for bhCG (18541070/20891859 UI/L) (P= 0.780) because selected pregnant women presented bhCG levels between 800-3500 IU/L as an inclusion criteria. PlGF levels were similar in both groups (16.13.6/15.23 pg/mL) (P =0.277). We only obtained significant differences for sFlt-1 (8020/170125 pg/mL) (P <0.001). The AUC for sFlt-1 was 0.829 and the cutoff point with better sensitivity and lower false positive rate was 85 pg/mL (sensitivity=77%, specificity=94%, positive predictive value=93%, negative predictive value=80% and false positive rate=5%). Conclusions: sFlt-1 is a useful marker to differentiate between an EP or a MA when bhCG levels are similar in both groups. T224 PLACENTAL GROWTH FACTOR AND SOLUBLE FMSLIKE TYROSINE KINASE 1 AS MAIN BIOMARKERS OF PREECLAMPSIA AND INTRAUTERINE GROWTH RESTRICTION A. Martnez-Ruiz, J. Prez-Fornieles, N. Sancho-Rodrguez, J.A. Vlchez-Aguilera, I. Cebreiros-Lpez, I. De MiguelElzaga, M. Martnez-Villanueva, M.D. Sarabia-Meseguer, I. Tovar-Zapata Clinical Analysis Service of University Hospital Virgen de la Arrixaca, Murcia, Spain Background: Placental growth factor (PlGF) and soluble fmslike tyrosine kinase-1 (sFlt-1) are altered in preeclampsia (PE) and intrauterine growth restriction (IUGR). The aim of this study was to investigate if PlGF and sFlt1 serum levels are useful markers in the prediction of PE and/or IUGR at 11-13 weeks gestation. Methods: We included 101 pregnant women with a priori risk of developing PE (inclusion criteria: preexisting renal disease, chronic hypertension without proteinuria, history of PE in a previous pregnancy, etc) in a prospective study of one year. PE was diagnosed if normotensive woman had two systolic blood pressure measurements of 140 mmHg repeated in a 6 hour interval and/or a diastolic blood pressure measurements of 90 mmHg after the 20th gestation week; together with proteinuria of more than 300 mg/24 hours specimen (6 hour interval). IUGR was dened as a birth weight below the 10th percentile for gestational age. PlGF and sFlt-1 levels were measured throught electrochemiluminescence (Roche, Manheim, Alemania). U Mann-Whitney test and students t test were used to compare the two markers in different groups: Group1 (pregnant with PE) (G1), Group2 (pregnant women with IUGR) (G2) and Group3 (pregnant women at risk for PE but ultimately not develop PE or IUGR) (G3). Results: Of the pregnant women included, 9 developed PE and 16 developed IUGR. When we compared G1 and 3, obtained significant differences for PlGF levels (35.411.1/60.726.3 pg/mL) (P=0.004) and the sFlt-1/PlGF ratio (43 [26-52] / 24 [1735] (P =0.033). When we compared the IUGR cases, G2 with G3, no significant differences were found for any marker studied. Finally we compared G1 and G2 and we observed significant differences only for PlGF levels (35.411.1/69.4 24.4 pg/mL) (P=0.009) Conclusions: PlGF and SFlt-1/PlGF ratio are useful markers to predict the subsequent development of PE but not IUGR, in the first trimester of pregnancy. Conclusions: PlGF and SFlt-1/PlGF ratio are useful markers to predict the subsequent development of PE but not IUGR, in the first trimester of pregnancy. Of the two markers studied PlGF has a higher AUC for the diagnosis of PE.
S403

T225 EVALUATION OF THE DYNAMICS OF THYROID HORMONES LEVELS DURING NORMAL PREGNANCY BY ELISA E. Matveeva, E. Kochina, A. Burkov, A. Obriadina RPC Diagnostic Systems, Nizhniy Novgorod, Russia. Background: Pregnancy is associated with changes in thyroid function, which are result of a normal physiologic state. Methods. Levels of thyroid stimulating hormone (TSH), free thyroxine (FT4), free triiodothyronine (FT3), total thyroxine (TT4), total triiodothyronine (TT3) during normal pregnancy were evaluated using ELISA kits (RPC Diagnostic Systems, Russia). Results: Two groups of samples were investigated. The first group consisted of 360 serum samples from white women with normal pregnancy (Central Russia and Volgo-Viatsky Region, from 4 to 39 gestational weeks). The second group included 103 serum samples from white non-pregnant women (the same region). The Mann-Whitney U test (in case of non-normal distribution) and t-statistic (in case of normal distribution) were used to compare differences between two independent groups. TSH level was lower in pregnancy state (1 trimester - median 1.26, P <0.001; 2 trimester - median 1.39, P=0.003; 3 trimester - median 1.61, P >0.05) than in normal state (median 1.83 IU/mL). FT4 level was lower in pregnancy state (1 trimester median 13.7, P >0.05; 2 trimester - median 13.0, P <0.001; 3 trimester - median 10.7, P <0.001) than in normal state (median 14.1 pmol/L). FT3 level was the same in pregnancy state (1 trimester - median 2.97, P >0.05; 2 trimester - median 2.99, p >0.05; 3 trimester - median 2.72, P >0.05) as in normal state (median 2.91 pg/mL). TT4 level was greater in pregnancy state (1 trimester - median 123, P <0.0001; 2 trimester - median 133, p <0.0001; 3 trimester - median 127, P <0.0001) than in normal state (median 96 nmol/L). TT3 level was greater in pregnancy state (1 trimester - median 1.40, P <0.0001; 2 trimester median 1.61, P <0.0001; 3 trimester - median 1.73, P <0.0001) than in normal state (median 1.19 ng/mL). Conclusions. We observed a statistically significant suppression serum TSH in the 1 trimester of pregnancy when the human chorionic gonadotropin levels reach their peak. Concentrations of FT4 decrease during pregnancy in contrast to FT3 due to the high affinity of thyroxine-binding globulin for T4. Total T4 and T3 levels increase from 1 to 2 trimesters and they reach a plateau from 2 trimester to term. We defined normal trimester specific limits for these hormones in Central Russia population. T226 CORRELATION BETWEEN SERUM HOMOCYSTEINE AND CERULOPLASMIN LEVEL IN THE LAST PART OF PHYSIOLOGICAL PREGNANCY M.E. Muresan(1), O. Micle(1), L. Antal(2), S. Vicas(3), L. Micle(4), I. Ignat-Romanul(5), A. Popa(6)
1 Preclinical Disciplines Department, University of Oradea, Medicine and Pharmacy Faculty, Oradea, Romania, 2 Surgery Department, University of Oradea, Medicine and Pharmacy Faculty, Oradea, Romania, 3 University of Oradea, Faculty of Environmental Protection, Oradea, Romania 4 Clinical County Hospital of Oradea, Romania 5 Dentistry Department, University of Oradea, Medicine and Pharmacy Faculty, Oradea, Romania 6 S.C.Bioservice SRL, Romania
Background: Pregnancy in humans is a physiological state in which there are majors anatomic and metabolic changes in order to support and provide the needs of the growing conceptus. Homocysteine is an amino acid which contains sulfur and derivates from demethylation of methionine. Hyperhomocysteinemia in pregnancy it has been associated preeclampsia, spontaneous abortion, and premature delivery. The objective of our study was to investigate the correlation between homocysteine concentration and ceruloplasmin in the last part of physiological pregnancy. Methods: We studied 45 women enrolled in the following groups: (1) (n =15) non-pregnant women (control group), (2) (n=15) second trimester of pregnancy (group 1), (3) (n=15) third trimester of pregnancy (group 2). We measured serum homocysteine concentrations using an enzymatic method, C Reactive Protein (CRP) were determined by immunoturbidimetry. As the most important plasmatic antioxidant factor, the level of ceruloplasmin was determined using the Ravin method. Results: Compared with the non-pregnant group (7.830.98 mol/L), serum concentrations of homocysteine were significantly low in pregnant women groups (P <0.001). No significant difference was found between pregnant women in the second trimester (4.250.69 mol/L) and pregnant women in the third trimester (4.080.89 mol/L) (P >0.05). The serum levels of ceruloplasmin in pregnant women in the second (43.752.50 mg/dL) and the last trimester of pregnancy (46.606.20 mg/dL) were higher than in the reference group (32.111.62 mg/dL) (P <0.001). There was no significant modification of the serum ceruloplasmin concentration between group 1 and group 2 (P >0.05). No significant correlation was observed between homocysteine and ceruloplasmin. Conclusions: The concentrations of serum homocysteine were significantly lower in pregnant women groups compared with non-pregnant controls, but between the pregnant women groups no difference was found. The main plasma antioxidant factor, ceruloplasmin was increased in pregnant women groups in comparison with the reference group. No correlation was found between homocysteine and ceruloplasmin.
S404

T227 OXIDATIVE STRESS MARKERS IN PREGNANCY WITH PATHOLOGICAL CARYOTIPE A. Nikolic(1), M. Milosevic Tosic(1), J. Bosic(2), J. Oros(1)
1
T228 ANALYSIS OF SERUM PROTEIN FRACTIONS FROM WOMEN WITH RECURRENT PREGNANCY LOSS L. Nowak-os(1), G. Odrowaz-Sypniewska(1), J. Zegarska(2), M. Zalewska-Zacharek(2), J. Kyszejko-Molska(2)
1 Department of Laboratory Medicine, Nicolaus Copernicus University in Torun, Collegium Medicum in Bydgoszcz, Poland 2 Department of Obstetrics, Gynecology and Gynecological Oncology, Nicolaus Copernicus University in Torun, Collegium Medicum in Bydgoszcz, Poland
Clinical center of Vojvodina Urgent Center, Department of urgent laboratory 2 Institute of Cardiology Sremska Kamenica Fetal trisomy 21 is the most frequent form of aberrant karyotype in pregnancy and screening methods for detection of fetal trisomy 21 are urgently needed. Cu, Zn- superoxide dismutase is an enzyme with important roles in oxidative stress equilibrium and the physical location of the gene encoding Zn, Cusuperoxide dismutase on chromosome 21 makes it a likely candidate for the Down syndrome screening. The aim of this study was to investigate the role of the fetal trisomy 21 on the redox system status in mothers and to investigate the possibility of using anti- oxidative enzymatic activities in the prenatal screening for the Down syndrome fetuses in the first trimester of pregnancy. Erythrocyte hemolysates from the study group (30 pregnancies with the fetal trisomy 21) were obtained and compared with the control group erythrocyte hemolysates (50 healthy pregnancies with the normal karyotype) for the following pro-oxidative (lipid peroxidation) and anti-oxidative parameters (SOD, catalase, gluthatione peroxidase activities and acidum uricum). The standard biochemical markers of the aberrant karyotype in the first trimester (feta-HCG, PAPP-A) were also compared between the control and the study group. Our data indicate that there is a significant increase in the lipid peroxidation and superoxide dismutase activity in the study group compared to the control group and no change in the catalase activity between the two groups. Additionally, we found a significant decrease in the gluthatione peroxidase activity in the study group compared to the control group, and an increased level of acidum uricum and of the index of oxidative stress in the study group. The analysis of our data suggests that there may be a change in the oxidative stress balance in the study group which may lead to the excessive hydrogen peroxide production. The analysis of the oxidative stress parameters in the control group suggests that there is a balance between the pro- and anti-oxidant parameters. The test for the biochemical markers of the aberrant karyotype (free beta HCG and PAP-AP) was positive in 93.75% of the pregnancies with the fetal trisomy 21. Of all parameters investigated, the values for the SOD activity in the 13th and 14th gestation week showed the best prognostic value.
Recurrent pregnancy loss occurs in 1 5% of women at reproductive age. Miscarriage incidence correlates with, among other factors, gestational age, womens age and the number of previous abortions. In most women, no cause of recurrent pregnancy loss is usually found. Therefore it seems important to study all factors possibly inducing pregnancy disorders. Objective: We assessed the fractions of serum proteins from those women with recurrent pregnancy loss, who gave vaginal birth once prior to having at least 3 miscarriages, and those who had never given birth. Methods: The study group consisted of 52 women (aged 36.04.9) with recurrent pregnancy loss. Nine of them (17%) reported one earlier regular pregnancy ending with childbirth without complications. Control group comprised 30 nonpregnant women (aged 36.13.6), who had given vaginal birth to healthy children at least twice. Serum protein fractions were separated by electrophoresis in the SDS PAGE buffer system using a Mini PROTEAN 3 cell device. BioRad SDS PAGE Molecular Weight Standards covering mass range of 6.5-200 kDa were used as a reference. Gels were stained with Coomassie Blue R 250 solution. BioRad QuantityOne software was used for the assessment of molecular weight of each protein fraction. Results: Electrophoretic separation revealed 39 protein fractions of 10 243 kDa. Particularly interesting was a 38 kDa fraction present exclusively in serum of women with recurrent pregnancy, who had never given birth. Another fraction (74 kDa), not detected in the control group, was found in all women with recurrent pregnancy loss. Protein fractions of 76 and 151 kDa were present only in the control group. Conclusions: The presence of atypical, low-or mid-weight proteins, including a 38 kDa fraction, in women with recurrent pregnancy loss may potentially play a role in the pathomechanism of this disorder.
S405

T229 LIPID PEROXIDATION IN OVERWEIGHT AND OBESE PREGNANCY B. Rejc(1), J. Osredkar(2), K. Gerak(1) University Medical Centre Ljubljana, Department of Obstetrics and Gynaecology, Zaloska cesta 2, Slovenia 2 University Medical Centre Ljubljana, Clinical Institute of Clinical Chemistry and Biochemistry, Zaloka cesta 2, 1000 Ljubljana, Slovenia Background: Overweight and obesity is a growing problem all over the world also for women of reproductive age. The study was undertaken to evaluate the role of oxidative stress (OS) in overweight and obese pregnant women. 8-isoprostanes (8-IP) were estimated as a marker of lipid peroxidation in maternal urine and amniotic fluid. Methods: We prospectively recruited Slovenian pregnant women undergoing amniocentesis at at Department of Obstetrics and Gynecology at University Medical Center Ljubljana between 15 and 27 weeks of gestation. Eighty nonsmoking women with healthy, singletone pregnancy and normal pregnancy outcome were included. The groups were formed according to WHO body mass index (BMI) classification, as lean and normal weight pregnant women (BMI <25) and overweight and obese women (BMI 25). Urine and amniotic fluid samples were collected at recruitment. 8-isoprostanes were assayed by 8-isoprostane ACETM Competitive Enzyme Immunoassay Kit (Cayman Chemical, Ann Arbor, MI, USA) following manufacturer instructions. Normally distribution of data was checked with KolmogorovSmirnov test. Non-parametric Mann Whitney U-test and Spearman`s correlation were used for statistical analysis. A p value <0.05 was considered to indicate statistical significance. Results: We found significant difference in amniotic fluid 8-IP concentration between groups (P <0,05). The 8-IP level in amniotic fluid was lower in the group with BMI 25 (19.87 ng/L) than those in the group with BMI <25 (24.66 ng/L). There was no difference in urine 8-IP between groups. The 8-IP in amniotic fluid and urine did not correlate with each other. Conclusions: We successfully detected 8-IP in urine and amniotic fluid. Absence of correlation between maternal and fetal compartment indicate that there might be different level of oxidation products excreted to maternal and fetal side. Our results of comparison between defined groups were not consistent with previous findings and expectations.
1
T230 BIOCHEMICAL PROFILING STUDY IN UMBILICAL CORD BLOOD AS PREDICTORS OF NEONATAL DAMAGE. S. Gruccio(1), M. Pandolfo(1), M.B. Di Carlo(1), G. Santa Cruz(1), R. Giuliano(2), G. Negri(1), H. Ruda Vega(2), M. Vazquez Blanco(3), B. Perazzi(1) Department of Clinical Biochemistry-Faculty of Pharmacy and Biochemistry, University of Buenos Aires, Argentina 2 Obstetrics Division-Hospital de Clnicas, University of Buenos Aires, Argentina 3 Cardiology Division-Hospital de Clnicas-University of Buenos Aires, Argentina Background: During pregnancy inflammatory,metabolic and immunologic disorders affecting fetus,are known,such as abortion,intrauterine growth retardation (IGR), low birth weight and neonatal death. The objective was to analyze different biochemical parameters (BP) in maternal venous blood (MVB) and umbilical cord blood (UCB) of newborn from healthy mothers and mothers with basal pathologies and associated to gestation that allow the early detection of perinatal complications. Methods: Samples from MVB (283) and UCB (283) were analyzed. Delivery was by cesarean. Mothers and newborn were classified controls (C-n=99), pathological (P-n=184). Maternal pathologies: diabetes, hypertension, anti-phospholipid syndrome, hyper/hypotiroidism, intrahepatic-cholestasis and genital infections. Pathological newborn: IGR and/or fetal distress. BP: Glucose, urea, creatinine, uric acid (UA), bilirrubin (B), proteins (P), albumin (A), transaminases (ALT/AST), alkaline-phosphatase, gammaglutamyltranspeptidase (GGT), creatinkinase (CK), lactatedehydrogenase, iron, calcium, phosphorus, magnesium, sodium, potassium (K), cholesterol (CHO), triglycerides (TG), hsCRP were determined by recommended methods-Roche autoanalyzer. Student/Mann Witney tests were applied, P <0.05. Results: -P newborn from P mothers showed significant decrease:in gestation weeks (GW) and newborn weight (NW) with respect to C newborn from C mothers (P <0.05); significant increases in CHO, TG, UA, K, B, AST, GGT (P <0.05) and significant decreases in CK, P, A (P <0.05). -P mothers related to C mothers showed significant increase in UA, ALT, AST, GGT (P <0,05). Conclusions: In P newborn from P mothers with respect to C, the decrease in GW/NW would be related to IGR that accompany these pathologies; increases in CHO, TG, UA, K, B, AST, GGT to cellular destruction associated to maternal pathologies and to deficit in pulmonary development by IGR; decreases in CK, P, A to IGR. The increase of UA, ALT, AST, GGT from P mothers with respect to C mothers would be associated to inflammatory process. A future study with greater number of samples by maternal pathology, is proposed.
1
S406

T231 BLOOD LIPID PROFILE IN SECOND AND THIRD TRIMESTER OF PREGNANCY J. Petrovic(1), V. Stankic(1), M. Zamurovic(1), S. Popovac(1), L. Radakovic(2) Gynecologycal and obstetrical clinic Narodni front, Belgrade, Serbia 2 Medical center Vukovar, Vukovar, Croatia Pregnancy induces significant metabolic changes. The concentration of plasma lipids increase appreciably during pregnancy. The lipid levels are affected during maternal hormonal changes(rise in insulin, progesterone, 17 betaestradiol, Human placental Lactogen). Other maternal factors such as BMI (Body Mass Index), maternal weight gain, nutrition, pre-pregnancy lipid levels and various medical complications of pregnancy may also have significant effects on lipid metabolism and plasma levels. To test any changes in lipid profile, we determinated cholesterol, tryglycerides, HDLand LDL-cholesterol in 33 pregnant women three times. I second trimester (24 +/- 3 gestation weeks), II - third trimester (36 +/- 1Gestation week) and III - after delivery (30 +/- 10 days). Method: Enzimatic colour tests for quntitative determination of cholesterol, truglycerides, HDL- and LDL-cholesterol in human plasma. The measurement were classified into the following categories: second trimester (24 +/- 3 GW), third trimester (36 +/- 1 GW) and after delivery (30 +/- 10 days). Results: I group - cholesterol mean 6,77 +/- 1,097 mmol/L; tryglycerides mean 2,476 +/- 0,719 mmol/L; HDL- mean 1,87 +/- 0,496 mmol/L; LDL- mean 3,763 +/- 0,894 mmol/L; II group cholesterol mean 7,15 +/- 1,279 mmol/L; tryglycerides mean 3,366 +/- 1,495 mmol/L; HDL- mean 1,728 +/- 0,395 mmol/L; LDL- mean 3,94 +/- 0,912 mmol/L; III group cholesterol mean 5,72 +/- 1,032 mmol/L; tryglycerides mean 1,48 +/- 1,073 mmol/L; HDL- mean 1,624 +/- 0,504 mmol/L; LDL- mean 3,23 +/- 0,758 mmol/L. Conclusion: cholesterol, trygycerides, HDLand LDl-cholesterol increase to third trimester and decrease after delivery.
1
T232 DETERMINATION OF MEDIAN LEVELS OF THE FREE BETA SUBUNIT OF HUMAN CHORIONIC GONADOTROPIN IN WOMEN FROM RUSSIAN FEDERATION USING A TWOSITE MONOCLONAL ELISA Y. Ptitsyna(1), O. Udalova(2), A. Burkov(1), A. Obriadina(1) RPC Diagnostic Systems, Nizhniy Novgorod, Russia Regional Clinical Diagnostic Centre, Nizhniy Novgorod, Russia
2 1
Background. The level of free subunit of the human chorionic gonadotropin (free -HCG) is an important serum marker for biochemical screening in the first trimester of pregnancy. Methods. Serum concentration of free -HCG was measured using a two-site monoclonal ELISA (DS-EIAGONADOTROPIN-free HCG). Results. We evaluated serum samples from pregnant white woman from Central Russia and Volgo-Viatsky Region, Russian Federation, mean age 27 years, range from 16 to 40 years. The study population consisted of 3345 singleton, nondiabetic pregnancies at 8-13 weeks of gestation resulting in the delivery of phenotypically normal neonates and 16 pregnancies with trisomy 21. Median values and 95 % confidence intervals were calculated for 8-13 gestational weeks. 8 weeks of gestation: median 46.5 ng/mL (12.1 to 169 ng/mL), 9 weeks of gestation: median 44.7 ng/mL (12.7 to 134.3 ng/mL), 10 weeks of gestation: median 43.8 ng/mL (6.2 to 135.8 ng/mL), 11 weeks of gestation: median 38.9 ng/mL (10.6 to 133.0 ng/mL), 12 weeks of gestation: median 35.9 ng/mL (8.2 to 135.6 ng/mL), 13 weeks of gestation: median 17.3 ng/mL (4.75 to 115.6 ng/mL). MoM values were calculated by dividing an individuals marker level by the median level. In cases of Down syndrome (trisomy 21) the median MoM values of free -HCG were significantly higher than in control groups: 11 weeks of gestation - 2.4 MoM, 12 weeks of gestation 1.8 MoM, 13 weeks of gestation 7.8 MoM. In control groups Median Mom consisted 1.00,004. The study population was divided according to maternal weight. Medians for groups with different weight (45, 60, 75, 90 kg) were calculated for every week of gestation. 8 weeks of gestation (54.1, 46.5, 44.3, 32.8 ng/mL), 9 weeks of gestation (56.8, 52.5, 46.1, 30 ng/mL), 10 weeks of gestation (55.0, 43.9, 39.4, 37.3 ng/mL), 11 weeks of gestation (48.4, 38.9, 34.5, 32.8 ng/mL), 12 weeks of gestation (42.5, 35.9, 34.8, 27.4 ng/mL), 13 weeks of gestation (22.0, 18.2, 14.3, 5.5 ng/mL). Conclusions. We defined normal limits of maternal serum free -HCG in Central Russia population with DS-EIAGONADOTROPIN-free HCG. It is necessary for estimation of the risk of abnormal pregnancy. A significant relationship between free -HCG values and maternal weight was obtained.
S407

T233 EVALUATION OF 2ND TRIMESTER BIOCHEMICAL AND ULTRASONOGRAPHIC MARKERS FOR THE PREDICTION OF PREECLAMPSIA D. Rizos(4), G. Karampas(2), A. Haliassos(1), D. Hassiakos(3), K. Panoulis(3), G. Mastorakos(3), N. Vitoratos(3), M. Rizou(1)
1
T234 FETAL URINE ANALYSIS TO PREDICT POOR POSTNATAL FUNCTION IN TWO CASES OF CONGENITAL URINARY OBSTRUCTION C. Rossi(1), M.R. Dess(2), M. Pellegrino(3), G. Noia(3), L. Masini(3), C. Manzoni(4), A. Giona(3), A. Facente(3), A. Caruso(3), C. Zuppi(1), C. Call(1)
1 Dipartimento di Medicina di Laboratorio,Policlinico A.Gemelli, Roma, Italia 2 Dipartimento di Medicina di Laboratorio, Policlinico Universitario Tor Vergata, Roma, Italia 3 Dipartimento di Ostetricia e Ginecologia, Policlinico Universitario A.Gemelli, Roma, Italia 4 Divisione di Chirurgia Pediatrica, Policlinico Universitario A.Gemelli, Roma, Italia
Diamedica, Clinical Chemistry Lab, Athens, Greece Obstetrics & Gynecology Department, Konstantopoulio General Hospital, N. Ionia, Attica, Greece 3 2nd Dept of Obstetrics & Gynecology, Medical School, University of Athens Aretaieion Hospital, Greece 4 2nd Dept of Obstetrics & Gynecology, Hormone Laboratory, Medical School, University of Athens, Aretaieion Hospital, Greece
2
Background: Recent studies have examined the combination of biochemical and ultrasonographic markers for the prediction of preeclampsia and found that the addition of uterine artery Doppler data to the biochemical markers data improved the predictive performance of biochemical markers in both 1st and 2nd trimester. We aimed to evaluate the performance of the biochemical markers:Placental growth factor (PlGF), soluble fms-like tyrosine kinase-1 (sFlt-1), Asymmetric dimethylarginine (ADMA) and human Neutrophil Gelatinase-Associated Lipocalin (NGAL) in the 2nd trimester of pregnancy for the prediction of preeclampsia. Methods:The data used in this study are part of a wider ongoing investigation project on biochemical and ultrasound markers for the prognosis of adverse pregnancy outcomes.In this nested case-control study we prospectively measured PlGF, sFlt-1,ADMA,NGAL and Doppler uterine artery pulsatility index (PI) in 12 pregnancies that developed preeclampsia and 41 uncomplicated pregnancies,in the 2nd trimester (20th 26th week) of pregnancy. Results: ADMA (12327.9 mol/L) and NGAL (51.828.2 ng/mL) concentrations were significantly higher while PlGF (225152 pg/mL) was significantly lower in preeclamptic pregnancies compared to normal pregnancies (10425.3; 30.420.2; 330157 respectively) in the 2nd trimester. Also uterine artery PI was significantly higher in preeclamptic pregnancies (1.240.42) compare to normal pregnancies (0.950.30). The concentrations of sFlt-1 didnt differ significantly between the two groups. We evaluated a logistic regression model to predict the probability of preeclampsia using as predictors the concentrations of biochemical markers, the PI and the maternal body mass index (BMI). Using the forward stepwise selection method we found that BMI, sFlt-1 to PlGF ratio and NGAL were significantly independent predictors. The model with these predictors had sensitivity 67% and specificity 96% for the prediction of preeclampsia. The addition of PI in the model didnt improve the sensitivity and specificity. Conclusions: During 2nd trimester of pregnancy, ADMA and NGAL were significantly increased while PlGF significantly decreased in preeclamptic pregnancies. sFlt-1 to PlGF ratio and NGAL together with maternal BMI are effective predictors.
Background: Untreated fetal lower urinary tract obstructions have a mortality rate of up to 45%. We evaluated the clinical usefulness of fetal urine analysis for the prediction of poor postnatal renal function in two cases. Methods: Patient 1: 26th week, diagnosis of left fetal abdominal mass (suspected multicystic-dysplastic kidney). Patient 2: 14th week, diagnosis of fetal megabladder. In both cases, 7 echoguided cystocentesis were performed. Urine fetal samples were assayed for electrolytes, creatinine, albumin, osmolality, 2-Microglobulin and CystatinC on Cobas311-Roche, BNII nephelometer Siemens,Osmometer AI. Results: In patient 1, the parameters of renal function were elevated, supporting an impaired renal function (Na:130.13.1 mEq/L; K: 4.140.38 mEq/L; Cl:109.32.1 mEq/L; Osmolality: 258.17.4 mOsm/Kg; Creatinine: 8.81.4 mg/dl; 2Microglobulin: 0.470.11 mg/dL; CystatinC: 0.070.01 mg/dl). At day 4 after birth, the child underwent left nephroureterectomy, with a definitive diagnosis of segmental form of renal dysplasia. The cystourethrography, performed at 3 months of age, showed a right healthy renal parenchyma, with a normal renal function. In patient 2, the fetal urine biochemical tests were normal (Na:6116.6 mEq/L; K:2.570.5 mEq/L; Cl:5018.4 mEq/L; Osmolality: 134.330.14 mOsm/Kg; Creatinine: 13.144.4 mg/dL; 2 Microglobulin: 0.400.23 mg/dL; CystatinC: 0.030.02 mg/dL) suggesting a normal renal function,even in the presence of megabladder, suggesting Prune-Belly syndrome. At birth, general conditions were fair with normal biochemical parameters and urinary tract ultrasound. The statistical analysis of biochemical tests of the two patients showed significant differences for Na, K, Cl, Osmolality and CystatinC (P <0.05), but not for creatinine and for 2-Microglobulin (Students test). Discussion: The use of echo guided invasive techniques has allowed to consider the fetus as a little patient and invasive approaches can be made with a very low risk-benefit ratio.The biochemical evaluations on fetal urines, have made possible the monitoring of the disease,leading to pregnancies up to 37th and 38th weeks,respectively, with appropriate approach and without aggressive treatment. In conclusion, the biochemical tests have been of great support to the formulation of the diagnosis and to reduce the intra-bladder pressure.
S408

T235 COST/EFFECTIVENESS OF RH NEGATIVE PREGNANT WOMEN MANAGEMENT J. Duplantie(1), O. Martinez(1), A. Bois(1), L. Nshimyumukiza(1), J. Gekas(5), E. Bujold(6), V. Morin(6), M. Valle(2), Y. Gigure(3), C. Gagn(4), F. Rousseau(3), D. Reinharz(1)
1 Dpartement de mdecine sociale et prventive, Facult de Mdecine, Universit Laval 2 Centre de recherche, CHU de Qubec 3 Dpartement de biologie molculaire, biochimie mdicale et pathologie, Facult de Mdecine, Universit Laval 4 Dpartement de gnie lectrique et de gnie informatique, Facult de Sciences et Gnie 5 Dpartement de pdiatrie, Facult de Mdecine, Universit Laval 6 Dpartement dobsttrique/gyncologie, Facult de Mdecine, Universit Laval, Canada
T236 VOLATILE ORGANIC COMPOUNDS IN AMNIOTIC FLUID DURING NORMAL HUMAN PREGNANCY R. Minet-Quinard(2), S. Ughetto(3), D. Gallot(1), D. Bouvier(1), D. Lmery(4), C. Thonat(2), L. Blanchon(1), V. Sapin(1)
1
EA 7281 Retinoids, Reproduction and Developmental Diseases / School of Medicine, Auvergne University, Clermont-Ferrand, France 2 Biochemistry and Molecular Biochemistry Department / Universitary Hospital, Clermont-Ferrand, France 3 Medical Information Department, Universitary Hospital, Clermont-Ferrand, France 4 Obstetrics and Gynecology Department, Universitary Hospital, Clermont-Ferrand, France Background: Amniotic fluid is essential for fetal development and maturation. Biochemical composition changes during pregnancy and reflects the metabolic status of the fetus and the mother. Volatile organic compounds (VOC) are present in various biologic fluids of the organism (urine, blood, saliva) and many are in the expired air. The endogenous or exogenous biochemical and physiological functions are not yet clearly understood. However, their presence may indicate a pathological process. The purpose of this study was to identify the VOCs present in amniotic fluid during physiological pregnancy and to analyze the relationship between these compounds and the lifestyle of the mother. Methods: The amniotic fluids were collected during the second trimester of pregnancy in 84 women. VOC composition was analyzed with a mass spectrometer (MS5973, Agilent) coupled with gas chromatography (GC6890, Agilent). Results: One hundred and twenty-three VOCs were detected in the amniotic fluid with a relative abundance between 0.001 and 63%. These compounds belong to 13 chemical families (alkanes, Maillard compounds, halogens, terpenes...) and the majority of them (90/123) had an exogenous origin. Acetone was the most abundant compound (relative abundance 63%) and most common (present in 100% of amniotic fluid). A statistical association was found between tobacco consumption by the mother and some VOC presence [benzene (P=0.0098), styrene (P=0.0007), 2.5 dimethylfurane (P=4.91x10-8), but-3enenitrile (P= 6.234x10-6)]. Conclusion: During normal pregnancy, the fetus is exposed to many VOC with mainly exogenous origins and with some carcinogenic properties. This compounds presence results in part of lifestyle of the mother and in particular tobacco addiction. Consequences on the childs long term become are to be determined.
Background: Currently, to prevent alloimmunization to Rhesus D antigen (Rh) during pregnancy, all Rh-negative women receive prophylaxis (anti-D antibodies). Measurement is therefore not targeted to women who have a specific need. Knowing fetal RhD would allow identifying women who are actually at risk of alloimmunization. There are several options to identify these women, but their cost/effectiveness (C/E) is still unknown. Methods: A virtual population of 10000 Rh negative pregnant women was built to simulate the C/E of preventing alloimmunization. The model considers four options: 1) the systematic use of anti-D immunoglobulin, 2) fetal RhD genotyping, 3) immunological determination of the father Rh factor, 4) mixed screening: immunological determination of the father Rh factor, followed, if the result is positive, by fetal RhD genotyping. The outcomes were measured for the first and an eventual second pregnancy (in 54.93%). Outcomes considered were: 1) the total direct costs under the perspective of the public health care system; 2) the number of fetus with hemolytic disease (HDF), 3) infant mortality. Results: Regarding the first pregnancy, two options emerged as the most C/E options: systematic prophylaxis and immunological Rh typing of the father with overlapping confidence intervals between both of them. Regarding the second pregnancy, immunological typing of the father emerged as the most C/E option when the outcome considered was the number of fetus without HDF. When, as an outcome babys survival at 28 days was considered, the results were similar to those of a first pregnancy. In all cases (first or second pregnancies and the combination of the two) fetal genotyping option doesnt appear to be C/E. In sensitivity analysis, fetal RhD genotyping appears as the most C/E option if the cost of fetal RhD genotyping drops below $CAD 140. Conclusion: Immunological Rh typing of the father and routine prophylaxis are the most C/E options. Whereas the immunological typing of the father option probably could not be adopted by the majority of clinicians, routine prophylaxis remains the preferred option. However, this could change if the cost of RhD fetal genotyping goes below $CAD 140.
S409

T237 FIRST TRIMESTER COMBINED PRE-NATAL SCREENING: EVALUATION OF EIGHT YEARS AS FMF CERTIFIED LABORATORY M.J. Sousa, R. Ribeiro, A.M. Albuquerque, J.G. Sousa, G. Sousa Centro Medicina Laboratorial Dr. Germano de Sousa, Lisbon After eight years as Fetal Medicine Foundation (FMF) Certified Laboratory, the authors want to demonstrate the importance of using correct technology and correct procedures that allow detecting the highest number of Positive Cases of Trissomy 21, 18 and 13, and lowering the unnecessary amniocentesis. Methods: Screening analysis was performed between June 2005-2012. The FMF-CL uses a immunoassay analyser with Time-Resolved-Amplified-Cryptate-Emission (TRACE technology) KRYPTOR (BRAHMS). The biochemical markers are Free hCG and PAPP-A for First Trimester Screening, from BRAHMS. The three level Internal Quality Control are analysed on a daily basis, with a Coeficcient Variation less than 3%. The Certified Laboratory participates in UK-NEQAS for First Trimester Downs Syndrome Screening. For statistics evaluations the laboratory uses a ANOVA with 95% of Confidence Interval, and descriptive statistics permormed on Minitab Software. Results: All the procedures recommended by FMF, including quality control, technology used for biochemistry markers and the ultrasonographic data that the laboratory is working with (an essential component of biochemical screening is accurate dating of the pregnancy by ultrasound, otherwise the detection rate is reduced by about 10%), allow us to give very accurate risks, and decrease the number of amniocentesis in women aged 35 or more. We evaluate in our population the correlations between Free Beta-HCG/ PAPP-A MoMs and several maternal characteristics, including racial origin, weight, smoking and method of conception, being evident an increase in PAPP-A MoMs in smokers, and decrease in diabetic patients. We analyze the cross-effect of the smoke and diabetes in both biochemical variables. In both we verify the effect potentiating of factors that are not as evident when only one is present. We interpreting the results between PAPP-AMoMs and birth weight, and compare with other bibliography. Summary of effects of various factors on the MoMs studied, Discussion: An alternative strategy for first-trimester combined screening is for biochemical testing and ultrasound scanning to be carried out in two separate visits, with the first done at 9 to 10 weeks and the second at 12 weeks, allowing earlier assessment. Based on bibliography, this method would improve the detection rate from 90% to 93%. In this period 20486 patients were examined. The median maternal age at term was 33 years, and there were 23% aged 35 or more. The risk estimated was 1:300 or higher in 3,35% fetuses. Techinal variations with more than 4% of Coeficient Variation in the determination of Free HCG and PAPP-A, can give variations in the calculation of the risk over than 25% [Spencer, 2003, DS News]. Laboratories that do no follow the FMF criteria can present False Positive Rates (FPR) of 30% and over. Our FRP is 3,14%. Regarding FMF the FPR maximun expected is 5%. The smoker variable and diabetes are effectively important parameters when combined risk is performed, and the decision action is taken based on FMF criteria and there accurated risks. T238 USING SOFTWARE SSDWLAB 5 IN FIRST AND SECOND TRIMESTER FOR SCREENING DOWNS, EDWARDS AND PATAUS SYN G. Spasic-Obradovic, K. Ille Laboratory of Biochemistry, "Zvezdara University Medical Center", Belgrade, Serbia Background: First and second trimester screening are indispensable in early identification of Downs, Edwards and Pataus syndrom in fetuses. We performed combined protocol which included maternal biochemistry(free beta-hCG, PAPP-A ,AFP and beta hCG) and sonografic determination(CRL, NT,BPD). In first trimester screening was performed between 11-14 weeks SsdwLab version 5.0.This software makes use of an algorithm described by Palomaki and is based on mathematical calculation using Gaussian multivariate distribution. Risk analysis is based on maternal age, NT,CRL as well as on the results of biochemical parameters free beta hCG and PAPA-A, corrected by different factors like e.g. maternal weight, smoking and ethnic background of the pregnant woman.The risk for a trisomy 21affected pregnancy in second trimester,14-18 weeks , can be calculated using the algorithm as described by Wald and the respective assay-specific prametars. Methods: Free beta-hCG, PAPP-A, AFP, beta hCG were performed by electrochemiluminescence immunoassay on COBAS E 411 immunoassay analyzer. Results: Total of 1806 samples from clinical routine with known outcome were examined for one year. 42 were from pregnancies with confirmed Downs syndrome in first and 38 in second trimester. 8 out of the 1806 samples were positive for Edwards or Pataus syndrome in first trimester. Cut-offs for Trisomy 21 and 18-13 were 1:250. Conclusion: Using software SsdwLab version 5.0 we were screening 2,7% pregnancies with risik in first and 2,1% in second trimester.
S410

T239 THE USE OF NEW LABORATORY MARKERS IN THE DIAGNOSIS OF ENDOMETRIAL CARCINOMAS - THE PILOT STUDY D. Stejskal(1), M. Svestak(2), D. Ondrova(3), R. Pilka(3)
1
T240 SPERM FATTY ACID COMPOSITION AND THE MALONDIALDEHYDE LEVELS IN HUMAN SEMINAL PLASMA X. tramov(1), K. Vorkov(1), R. Hampl(2), R. Kanr(1)
1 Department of Biological and Biochemical Sciences, Faculty of Chemical Technology, University of Pardubice, Czech Republic 2 Sanus, In Vitro Fertilization Clinic, Pardubice, Czech Republic
Department of laboratory medicine Institute od medical chemistry and biochemistry, faculty medical, Palacky university, Olomouc 3 Department of obsteric and gynecology, University hospital, Faculty medical and dentistry, Palacky University Olomouc
2
Background: Abnormal or pathological findings on the endometrium is a frequent finding in women of childbearing age, as well as in postmenopausal women. The most important finding is endometrial cancer, which is currently one of the most common gynecological malignant tumors. Diagnosis is usually set up on the basis of histopathological examination of biopsy material. There is currently no reliable biomarker that could diagnose suspected or established on the basis of biochemical tests. Calgizzarin (S 100A-11) and TFF are proteins, which in addition to physiological functions in the body and interfere significantly in carcinogenesis. Their abnormal values in serum could therefore may indicate pathological changes in the endometrium. Interesting for the diagnosis of endometrial carcinoma is the AIF-1. Their concentrations were till now used as a diagnostic indicator of Ca endometrium. Aim: to verify the effectiveness of diagnostic determination of AIF-1 TFF, TFF-2, TFF-3 and S100A-11 in the diagnosis of endometrial carcinoma. Methods: 44 women were middle age tested. S100A-11 (BioVendor, DSX) TFF-1 (BioVendor, DSX), TFF-2 (BioVendor, DSX), TFF-3 (BioVendor, DSX), AIF-1 (BioVendor, DSX) and Ca-125 (Siemenes, Centaur XP) were analysed in sera. Results: 30 women without endometrial carcinoma and 14 women with carcinoma were tested. Individuals with carcinoma had significantly higher values S100 A-11 and TFF-3 in sera (S100 A-11 7.2 vs. 4.2 ng / ml, P <0.01; TFF-3 1.0 vs. 2.7 ng / ml, P <0.05). Subgroup with with carcinomas did not significantly differ in values of Ca 125, TFF-1, TFF-2 and AIF with the healthy subgroup. Linear regression model with stepwise regression included only as a diagnostic marker of choice S-100 A11. Conclusions: S100 A-11 appears as a potential indicator of the presence of endometrial cancer. Similarly interesting appears TFF-3. In later stages of the project, we will investigate these parameters in subjects repeatedly endometriosis and monitor their prognostic significance. In Conclusion, firts results reveal a real possibility to use these biomarkers in diseases with endometrial cancer.
Background: Infertility appears to be very common problem of reproductive-age couples nowadays. The semen quality decreases for example because of lifestyle, infections, autoimmune or chronic diseases. The aetiology and pathogenesis are still not exactly understood, accordingly majority is considered idiopathic. Sperm lipid membrane, mainly consisted of polyunsaturated fatty acids (PUFAs), is particularly susceptible to oxidative damage. Good sperm qualities are important to suitable sperm motility and successful fertilization. Lipid peroxidation occurs when the production of potentially destructive reactive oxygen species (ROS) exceeds natural antioxidant capacity of the cell. PUFAs in the phospholipid membrane are the most attacked substrate. Many markers of oxidative stress are described; malondialdehyde (MDA) is one of the most determined. It is mutagenic, carcinogenic and atherogenic. Methods: MDA levels were measured using HPLC with fluorescence detection as a MDA(TBA)2 adduct. To seminal plasma sample EDTA solution and 2-thiobarbituric acid were added. The mixture was stirred and incubated. After derivatization, cold n-buthanol was added and the mixture was vortexed and centrifuged. The buthanol layer was filtered through nylon filter and transferred into cramped vial. Phospholipid PUFAs were determined by GC-FID. The raw semen specimens were centrifuged, supernatant was carefully removed and the pellet was used further. After protein precipitation, lipids were extracted and separated by preparative TLC into 5 classes. Phospholipid fraction was isolated and after the addition of an internal standard it was hydrolyzed and converted to methyl esters of fatty acids. This way prepared samples were analyzed. Results: MDA levels in a group of all patients ranged between 1.12-1.84 mol/L, in smokers (S) between 1.03-1.91 mol/L and in nonsmokers (NS) between 1.15-1.81 mol/L. Total sperm membrane phospholipid fatty acids were profiled into 4 groups - saturated acids (S 60.59%, NS 61.07%), PUFA 3 (S 14.52%, NS 14.93%), PUFA 6 (S 10.06%, NS 10.07%) and other acids (S 14.84%, NS 13.94%). Conclusions: High levels of PUFAs are proper for sperm membrane motility and flexibility. Despite this, PUFAs are susceptible to lipoperoxidation.
S411

T241 FREE BETA HUMAN CHORIONIC GONADOTROPIN AND PREGNANCY-ASSOCIATED PLASMA PROTEIN A IN THE FIRST TRIMESTER SCREENING E. Trujillo Arribas, P. Camacho Martinez, M. Garcia Pinilla, H. Macher Manzano, J.M. Guerrero Montvez H. U. Virgen del Rocio, Sevilla, Spain Background: The aim of this research is to study the influence of the lapsed time between the extraction and the determination procedure in the case of free beta human Chorionic Gonadotropin (free beta hCG) and Pregnancyassociated plasma protein A (PAPP-A), and its effect on the calculation of risk of chromosomal anormalities. Free beta hCG and PAPP-A increases over time, but it is to be determined how long it needs to cause a significant difference on the values and on the associated risk for the prenatal screening. Materials and methods: A group of 38 women in the first trimester of pregnancy were included in the study, blood samples were taken during the routine check-up. Blood samples were stored and transported to Virgen del Roco Laboratories in the same conditions (uncentrifuged and in isothermal iceboxes). Once in the laboratory the samples were centrifuged and analyzed using the electrochemiluminescence immunoassay ECLIA on cobas 8000 (Roche Diagnostics). Results: Both beta hCG and PAPPA increase over time. There were significant changes among the determinations corresponding to the samples extracted four, seven, and nine hours before the analysis (P=0.007). Both beta hCG and PAPPA have a linear ascending trend as the statistics and the profile graphs indicate (P=0.000). Comparing two to two (the different levels of the factor -time lapsed since the extraction), the critical levels associated to every comparison show that there are relevant differences among all the combinations of the different levels for both variables (P=0.000). The associated risk has notable difference among the determinations carried out at different times from the extraction (P=0.000). Such a difference, however, has no influence on the screening results. Conclusions: Normal handling of samples is likely to increase the results of the determination of free beta hCG y PAPP-A as time till their processing increases. However, in most cases, these variations do not affect the screening results (positive or negative screening). T242 THE IMPACT OF BILIRUBIN PHOTOTHERAPY ON NEUROBLASTOMA CELL VIABILITY J. Vanikova(1), A. Cepa(2), M. Sticha(2), M. Cerna(3), L. Muchova(1), L. Vitek(1)
1 Institute of Medical Biochemistry and Laboratory Diagnostics, 1st Faculty of Medicine, Charles University in Prague 2 Department of Organic and Nuclear Chemistry, Faculty of Science, Charles University in Prague 3 Institute of care for mother and child, Prague
Background: Pathological newborn jaundice is a severe condition which may lead to irreversible damage of the neonatal brain. To prevent this potentially fatal disease, endangered neonates are routinely treated by phototherapy, in which lipid-soluble bilirubin is transformed onto more polar bilirubin photoisomers (PIs). However, only scarce data exists on the potential biological effects of these PIs, which might account for certain side effects accompanying neonatal phototherapy. Thus, the aim of our study was to assess whether PIs may affect viability of neuronal cells. Methods: The effect of bilirubin PIs was studied on neuroblastoma cell line SH-SY5Y incubated with bilirubin (10100 umol/L) for 24 hrs and exposed to photo-irradiation using clinical phototherapeutic lamp (Drger Photo-Therapy 4000). Bilirubin PIs were determined by HPLC (C18 column, isocratic elution with 92% 0.1 mol/L di-n-octylamine acetate/8% water). Cell viability was tested by staining with crystalline violet, cell apoptosis was measured by MTT test. Results: Phototherapy of the neuroblastoma cells exposed to bilirubin lead to significant decrease of the bilirubin concentration (by 15-25%) due to its photoisomerization. Surprisingly, significant decrease in viability and increased apoptosis rate were observed in neuroblastoma cells exposed to various levels of bilirubin PIs (P <0.05 for both parameters, and most of bilirubin concentrations). Possibility, that our observations on cell viability might be due to the effect of photoirradiation per se, was excluded in separate experiments, in which no effect of phototherapy on neuronal cell proliferation was observed. Conclusion: Here we report that bilirubin PIs may be even more harmful for neuroblastoma cells compared to non-irradiated unconjugated bilirubin, and this observation might account for some known side effects of bilirubin phototherapy, especially under conditions where bilirubin PIs clearance from the body is impaired.
S412

T243 SERUM NEUTROPHIL GELATINASE-ASSOCIATED LIPOCALIN AND PLASMA NITRIC OXIDE LEVELS IN PREECLAMPTIC AND HEALTHY PREGNANTS N. Dogan(1), S. Yildirmak(1), V. Mihmanli(2), M. Vardar(1), Y.G. Ozbanazi(1), M. Cakmak(3), F. Sezgin(4)
1
T244 INTERPRETATION OF APTT PROLONGATION A POSTANALYTICAL EXTERNAL QUALITY ASSESSMENT SURVEY OF 16 COUNTRIES E. Ajzner, D. Rogic, P. Meijer, A.H. Kristoffersen, E. Somen, A.P. Faria, P. Carraro, J. Watine, S. Sandberg EFLM/EQALM Working group for Postanalytical External Quality Assessment Background: Activated partial thromboplastin time (APTT) is the second most common coagulation test performed in clinical laboratories worldwide. It is sensitive to quantitative and qualitative defects of intrinsic coagulation factors and it also might detect presence of factors specific and nonspecific inhibitors. The aim of our study was to assess the extent of laboratory practice variation in Europe when dealing with an unexpected prolonged APTT result. Methods:The survey was designed as a case report with multiple-choice questions. It was distributed electronically through EQALM and ECAT representatives to laboratories in 35 countries, with an invitation letter aimed at persons responsible for coagulation. Results:16 countries provided 95% out of 902 answers received, with an average response rate of 20% (range 2-77%). The results show considerable variation in handling an unexpected prolonged APTT, in terms of preanalytical errors exclusion approach and techniques, subsequent mixing studies and results reporting. When considering preanalytical factors, only 37% of laboratories use a laboratory method for excluding heparin presence in the sample. In spite of the fact that mixing studies represent a straightforward way to gather initial orientation on possible causes of truly prolonged APTT (presence or lack of inhibitors), 27% of participating laboratories do not perform them at all, while further 12% do it only upon physicians specific request. Among those who perform mixing studies, there is great variation in the sources of normal plasma, which is also non-buffered in at least half of the laboratories, while the possible effect of mixture preincubation is not known in 45% of cases. The interpretation of three possible mixing studies scenarios has also shown substantial variation. It is worth mentioning that almost half of the responders were not sure about the characteristics of their APTT reagent, ie. its relative sensitivity to factor deficiencies or LA presence. Conclusion:There is considerable variation in laboratory practice concerning evaluation and interpretation of prolonged APTT results. This might be an indication towards a possibility of false or mis-interpretations during investigations of the reason behind a prolonged APTT result.
Okmeydani Educational and Research Hospital, Department of Biochemistry, Istanbul, Turkey 2 Okmeydani Educational and Research Hospital, Department of Obstetric and Gynecology, Istanbul, Turkey 3 Gulkent State Hospital, Department of Biochemistry, Isparta, Turkey 4 Istanbul University, Department of Industrial Engeneering, Istanbul, Turkey Background: We aimed to evaluate serum neutrophil gelatinase associated lipocalin (NGAL) and plasma nitric oxide (NO) levels in preeclamptic and healthy pregnant women above twentyfour gestation weeks. Materials and Methods: 49 healthy and 21 preeclamptic, totally 70, pregnant women participated voluntarily in the study. Presence of 140 mmHg and above systolic and 90 mmHg and above diastolic blood presure which emerges after 20th gestation week, proteinuria more than 300 mg/24 hour and edema has been used as diagnostic criterion for preeclamptic pregnant women. Measurements of serum NGAL and plasma NO were performed with ELISA and photometric method, respectively. Results: Serum NGAL (ng/mL) and plasma NO (M) levels of healthy and preeclamptic groups did not show a statistical difference [124.68 (72.42-218.82) (medyan (min-max), (39,8315,84); 120.4450.88, 37,0714,48 (MeanSD), respectively]. In preeclamptic group, a statistically meaningful correlation was found between level of NGAL and body mass index of sampling time, creatinin and NGAL, total protein and NO and albumin and NO. Conclusion Serum NGAL levels, correlated with serum creatinin levels in our study, may be the early marker of renal damage which may develope mainly due to inflammation and endothel damage. We could not find a statistical difference for serum NGAL and plasma NO levels between healthy pregnant and preeclamptic groups. Varieties peculiar to humans in preeclampsia, impossibility of obtaining first trimester tissue material as an evidence of inadequate trophoblast invasion, different appearance of maternal reaction to underlying main pathology in every case may restrict clarification of etiopathogenesis.
S413

T245 COMPARISON OF KRAS GENOTYPING METHODOLOGIES: DOKUZ EYLUL UNIVERSITY MOLECULAR ONCOLOGY LABORATORY EXPERIENCE Y. Baskin(1), G. Calibasi(1), A. Amirfallah(1), S. Sarioglu(2), H. Ellidokuz(3)
1 Dokuz Eylul University, Institute of Oncology, Department of Basic Oncology, 35350, Izmir, Turkey 2 Dokuz Eylul University, Faculty of Medicine, Department Of Pathology, 35350, Izmir, Turkey 3 Dokuz Eylul University, Institute of Oncology, Department of Preventive Oncology, 35350, Izmir, Turkey
T246 PREANALYTICAL RATING PHASE OF BLOOD SAMPLING: SURVEY S. Avram, A. Stojni Clinical Center Banja Luka, Institute of Laboratory Diagnostic, Banja Luka, Bosnia and Hercegovina Introduction: Preanalytical errors are counting up to 70% of the total number of errors in laboratory diagnostics. According to the ISO 15189:2007 preanalytical phase can be defined in chronological order, which is based on the requirements of clinicians, preparation of patients, taking the primary samples, transport to the laboratory, and it ends when the analytical process begins. The study was conducted at the CCBL clinics and it was based on the process of blood sampling among nurses with secondary education and nurses with college degree, in order to access the actual assessment. Materials and methods:The anonymous survey which included 7 questions of closed type (Likert scale; never=1, rarely=2, often=3 and always=4) had a purpose to examine the blood extraction procedure as preanalytical phase of laboratory practice among nurses with secondary education and nurses with college degree at Clinical Center Banja Luka (N=882). Results: Of the total number of participants in the survey, 53% gave responds. Of that percentage (53% answered question), 85% were nurses with secondary education, and 15% of nurses with college degree.The mean value and standard deviation of all procedures was determined to be 3,27 0,60. The rating time record of blood sampling was 2.42, and this value represents the lowest score of all the survey questions. The best score obtained in the survey was based on the question refereeing to the mixing test tubes, and it was determined to be mixing blue and purple caps immediately after extracting blood and its value was 3.95. Conclusion: Our results highlight the need to educate the majority of staff responsible for blood sampling , in order to prevent the source of errors and quality assurance of clinical laboratory testing. For the achievement of high quality standards, it is necessary to continuously educate the personnel directly involved in the blood sampling.
Background: KRAS is a predictive marker for the response to EGFR based targeted therapies in metastatic colorectal patients. KRAS mutation testing for codons 12 and 13 is recommended by the American Society for Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) guidelines for planning CRC chemotherapy. DNA sequencing, PCR or microchip based methods for determine the KRAS gene mutations are used in many routine laboratories. No specific methodology is recommended for this test. The aim of this study was to compare 5 methods of KRAS mutation analysis tests and provide practical and affordable method. Methods: Five different methods for the analysis of KRAS codon 12, 13 from formalin fixed paraffin embedded tissues were evaluated. ARMS-PCR (DxS, Therascreen KRAS Mutation Kit, Manchester, UK), PCR/reverse-hybridization test strips (ViennaLab K-Ras StripAssay, Austria), classical DNA sequencing (Beckmancoulter, GenomeLabTM DTCS-Quick Start Kit, USA), pyrosequencing (Qiagen, PyroMark Kras Kit), microarray based gebotyping technology (Autogenomics Inc., Infinity Biofilm Chip Microarray, KRAS Assay, USA). Agreement was assessed by (Kappa) statistics using SPSS (Version 19.0, SPSS Inc., Chicago, USA). Results: In this study, each methods were compared with each other. The results of DNA sequencing by ARMS-PCR are 1=0.529 (95% CI, 0.180.87), by pyrosequencing 2=1; (95% CI, 1.001.00) and by microarray 3=0.89 (95% CI, 0.681.09). The results of ARMS-PCR by by pyrosequencing are 4=0.78 (95% CI, 0.381.17), microarray 5=0.65 (95% CI, 0.360.94) and by PCR/reverse-hybridization test strips 6=0.61 (95% CI, -0.041.237). The results of pyrosequencing by microarray 7=0.76 (95% CI, 0.351.18). The results of microarray by PCR/reverse-hybridization test strips are 8=0.60 (95% CI, 0.051.27). Discussion: This method comparison study provides data on KRAS mutation analyses. The results suggest that commercial services may provide different results. Pyrosequencing reliability is outstanding according to DNA sequencing. The lack of concordance among results of some methods, suggest that quality assurance programs will be necessary to ensure consistent and accurate results.
S414

T247 BIOLOGICAL VARIATION REFERENCE DATA: TIME FOR DEVELOPMENT OF STANDARDS? W.A. Bartlett Biological Variation Working Group, European Federation of Clinical Chemistry and Laboratory Medicine, NHS Tayside Blood sciences Department Background: Characterization and understanding of the biological components of variation in laboratory data is a fundamental requirement for laboratory medicine specialists. The data are used in the setting of quality standards and in the determination of significance of change in two consecutive results (reference change values (RCV). The characteristics of the data define their utility to providers and users of services, and so methods of data production and other data attributes are of great importance. There are parallels with production and use of reference values as identified by the IFCC and more recently in CLSI guidance. That approach identifies need for characterization of populations studied, methods for production of data, and the statistical treatment of data. The need for this degree of definition is accepted in the context of reference values, but not so in the context of biological variation data (BV). Indiscriminate application of poorly characterized, or produced, BV data leads to delivery of erroneous quality standards and RCVs, and compromises other applications of BV data. It follows that there is a need for standards for production of BV data, standards for reporting of them, and standards for their transmission, since these all impact on their utility and commutability. Methods: A working group has been constituted by the EFCCLM to consider the issues. As a first step the group has undertaken work to deliver a checklist to help users and publishers of biological variation data to appraise existing and future studies of biological variation data in a structured way. Results: An outline critical appraisal checklist has been developed to be further validated, and is available for wider consideration, discussion and testing (www.biologicalvariation. com/Tools.html). The checklist identifies critical factors to be considered, ranging from the need to adequately describe populations studied and the analytical methods used, to the importance of detecting outlying data and appropriate use of statistics. Conclusions: There is a need for standardization in the context of BV. An outline critical appraisal checklist has been developed to enable assessment of existing data sets and promote good practice in delivery of new data. T248 EXTERNAL QUALITY ASSURANCE FOR 25-OH VITAMIN D (RESULTS OF 2010-2012 CYCLES) S. Bianchi(1), S. Giovannini(2), M. Pierini(2), R. Ndreu(1), A. Vannucci(1), D. Battaglia(1), C. Vassalle(1)
1 2
Fondazione G.Monasterio CNR-Regione Toscana QualiMedLab-CNR, Pisa, Italy
We reported the results of the Immunocheck External Quality Assessment (EQA) scheme for 25-OH vitamin D (25OHD)2010-2012 cycles- conducted by the QualiMedLab-CNR (Pisa, Italy) in co-operation with ProBioQual (Lyon, France), including 250 among Italian (25%) and French (75%) laboratories (2012). Methods: Each participant, identified by a code number, indicated on a datasheet the method of analysis used, and the results obtained assaying the control samples. An on-line access permits to visualize analyzed and archived data, including distribution of all results and those of the participants own method group. All control materials (12 samples/year) are delivered at ambient temperature as lyophilized samples at the beginning of the annual cycle. Six rounds are expected, each one including two different test samples. The 2010-11 cycle included four samples with deficient (10 ng/mL), six and five samples with insufficient (11-29 ng/mL), and two and three with sufficient concentration (30 ng/mL). The 2012 cycle, actually not completed, included three with deficient and seven samples with insufficient concentration. Results: Number of EQA participants increase during the 201012 period (from 110 to 169 and 259). In parallel, there has been a significantly reduction in inter-laboratory imprecision, from 29% in 2010 to 26% and 23% in 2011 and 2012, respectively. Majority of laboratories utilized automated immunoassay (90%), while 10% of the overall participants had manual assays. Liaison DiaSorin was the method most utilized (40.3%), then Roche (23.5%), IDS (10.9%), ABBOTT Architect (10.9%) DiaSorin RIA (4.1%), others (10.5%). The withinmethod variability between laboratories (CV %) resulted 12.5, 18.5, 12, 14, and 17.6% for Liaison, Roche, IDS, Abbott, Diasorin-RIA, respectively. The majority of methods are positively biased with respect to the Consensus Mean (21.9, 11.5, 15.2 and 0.6% for total Liaison, IDS, Sorin-RIA and Abbott, respectively), the exception being the Roche which had a mean bias of 23.5%. Conclusions: Results of the present study indicate that running EQA schemes are needed to allow the comparison of results from different laboratories and to reliably evaluate their performance.
S415

T249 THE CDC HORMONE STANDARDIZATION PROGRAM J.C. Botelho, H.W. Vesper Centers for Disease Control and Prevention, Atlanta Steroid hormone measurements are increasingly used in patient care and public health. However, the accuracy and reliability of hormone tests prevent appropriate detection, treatment and prevention of diseases. The aim of this effort is to standardize steroid hormone measurements such that accurate and comparable measurements are obtained regardless of the measurement procedure. The CDC has established higher order reference measurement procedures for total testosterone and total estradiol in serum using LCMS/MS to assign target values to matrix-based materials. These measurement procedures are traceable to primary reference materials and to JCTLM certified reference measurement procedures. As part of the CDC Reference Laboratory Services non-pooled sera from single donors obtained following the protocol from CLSI C37-A with target values assigned are being provided for calibration. As part of the Standardization Services the CDC is operating a standardization certification program, HoSt. In this program, 4 blinded challenges are sent over the period a year. Values obtained are used for bias assessment as described in CLSI EP9-A2 and final assessment criterion is based on biological variability data. At present, 22 participants are enrolled in the HoSt Testosterone Program. Participants include clinical, academic, and pharmaceutical laboratories as well as immunoassay manufactures. 82% of participants have been able to meet the established criterion and laboratories that are successful are published on the CDC website (http:// www.cdc.gov/labstandards/hs.html). Estradiol has been added to the HoSt program and is executed in the same manner. The CDC is also working with external quality assessment providers on accuracy-based surveys as well as monitoring services for large clinical trials. In addition the CDC is working to establish reference ranges using the NHANES data set that can serve as common reference ranges for laboratories that are standardized to the CDC. The CDC Hormone Standardization Program collaborates closely with its partners such as the Partnership for Accuracy in Hormone Testing (PATH) to assure clinical and public health needs are met. Initial data show improvements in testosterone measurements performed in patient care. T250 THE ROLE OF THE SERBIAN CHAMBER OF BIOCHEMISTS IN THE QUALITY IMPROVING OF WORK AT THE CLINICAL BIOCHEMICAL LABORATORIES V. Canic Serbian Chamber of Biochemists, Belgrade Background: Serbian Chamber of Biochemists was founded as an independent, professional organization whose purpose is to promote the conditions necessary for performing duties required by the professions of medical/clinical biochemists. The aim of the work is the implementation of testing in the process of human resource evaluation, laboratory evaluation and implementation of internal and external control of work with recommendations for their improvement. Method: represents an independent research which is conducted using of Questionnaire on the organization of clinical biochemical laboratories (the study sample included 10% of all registered laboratories). The data were analyzed by the descriptive statistics method. Results: represent the present structure of employees with university degrees. The results show that in primary and secondary activity are working most of physicians-specialists in biochemistry, while pharmacy graduates medical biochemistry specialists are most represented at the tertiary level. By the form of ownership, 84,4% laboratories belonged to the public and 15,6% to the private sector. Pharmacy graduates M. Sc. in medical biochemistry are most represented in the private sector. By the educational level of lab managers, specialist in biochemistry was present in 90% of private laboratories and 85,2% of public laboratories. Biochemists participation on the work of professional collegiums at the primary health care level, is 68,8%, at the secondary level 77% and at the tertiary level 85,7%. Implementation of internal work control, in 90% of laboratories, is carried out daily with commercial control material, with commercial control material periodically in 18,8%, frozen pool serum daily in 15,6%, and with frozen pool serum periodically in 3,1%. System control yesterday-today is done in 28% of laboratories. 88% of laboratories are involved in external monitoring. Here is a significant statistical difference between the private sector-20%, compared to public sector-100% of laboratories. Research has shown that only one laboratory is accredited according to ISO 17025, while the ISO 9001 certification was performed in 4 laboratories. There is a significant difference between private and public sectors, by the opportunities for professional training (20% of the private sector and 78% of public sector institutions). Conclusion: condition of the human resources does not satisfy the standards. The relation, between basic profile and specialists in medical and clinical biochemistry, is uneven. Status of health workers is not regulated within the health system. Accreditation is at the very beginning. It is necessary to legislate the implementation of external and internal work control.
S416

T251 QUALITY SPECIFICATIONS BASED ON BIOLOGICAL VARIATION WHERE ARE WE? I. Celap(1), J. Juricek(3), A. Perovic(2), N. Puc(4), V. Supak Smolcic(5), I. Vukasovic(1) Clinical Institute of Chemistry, University Hospital Center Sestre milosrdnice, Zagreb, Croatia 2 Department of Clinical Chemistry, General Hospital Dubrovnik, Dubrovnik, Croatia 3 Clinical Department for Laboratory Diagnostics, Dubrava University Hospital, Zagreb, Croatia 4 Department of Medical Biochemistry, General Hospital Tomislav Bardek, Koprivnica, Croatia 5 Clinical Department of Biochemistry and Hematology Rijeka, Clinical Institute of Laboratory Diagnostics, Clinical Hospital Center Rijeka, Rijeka, Croatia Background: Consensus recommendations on the quality targets in laboratory medicine supports hierarchy quality model. In this concept, a model higher in the hierarchy should be preferred over lower one. The analytical performance in the majority of laboratories in our country is evaluated according to manufacturer recommendations, which is the lowest model in the hierarchy. Thus, we aimed to compare analytical coefficients of variation (CVa) against criteria for imprecision based on biological variation from five different laboratories. Materials and methods: Imprecision of 17 biochemistry parameters (amylase, AST, ALT, bilirubin total and conjugated, Ca, Cl, CK, CK-MB, creatinine, CRP, glucose, K, LD, Na, urea and total protein) was monitored by analyzing two levels of commercial quality control material, three times per day, over one year. CVa was monitored on four different platforms (AU2700, AU480, Architect ci 4100 and Cobas c501). Obtained CVa for each platform was evaluated against criteria for imprecision based on biological variation. Results: Comparison against biological variation was as follows: all monitored parameters met the optimum quality specifications only on AU480, while CVa monitored on the rest of the platforms (AU2700, Architect ci 4100 and Cobas c501) did not meet quality specifications based on biological variation for four parameters (sodium, chloride, calcium and total protein). Conclusions: According to our results sodium, chloride, calcium and total protein did not meet quality requirements on three different platforms. One of the possible reasons is low withinsubject biological variation which is difficult to achieve with available analytical methods.
1
T252 EVALUATION OF THE VERIGENE RESPIRATORY VIRUS PLUS NUCLEIC ACID TEST (RV+) FOR RAPID DETECTION OF INFLUENZA VIRUSES C. Gae Ryung, L. Chae Seung Department of Laboratory Medicine, Brain Korea 21 Graduate School of Medicine, College of Medicine, Korea University, Seoul, Republic of Korea Background: The Verigene Respiratory Virus Plus Nucleic Acid Test (RV+) (Nanosphere, USA) is a fully automated sample-to result platform capable of detection of respiratory virusesspecific nucleic acids directly from clinical specimens. This qualitative nucleic acid multiplex test based on nucleic acid amplification (NAT) followed by hybridization to capture gold nanoparticle-conjugated probes which are utilized to detect the presence of captured target DNA. This system differentiates influenza A and its subtypes (H1, H3, and 2009 H1N1), influenza B, and RSV subtypes (A and B). We evaluated the ability of the RV+ assay to detect influenza A and B virus from clinical specimens. Methods: Nasopharyngeal swabs, taken from 210 patientsinfluenza A (n=80), B (n=80), A and B mixed (n=1), and influenza-negative(n=102) were analyzed using Verigene Respiratory Virus Plus Nucleic Acid Test (RV+) assay, the Seeplex RV15 ACE Detection (Seegene, Seoul, South Korea), and the Anyplex FluA/B Typing Real-Time Detection (Seegene, Seoul, South Korea). Results: Compared to the Seeplex RV15 ACE Detection, Verigene Respiratory Virus Plus Nucleic Acid Test (RV+) assay had sensitivities and specificities of 100% and 100% for influenza A, and 100% and 99.4% for influenza B. Compared to the Anyplex FluA/B Typing Real-Time Detection, Verigene Respiratory Virus Plus Nucleic Acid Test (RV+) assay had sensitivities and specificities of 95.8% and 100% with RNA samples, and 87.5% and 100% with unprocessed specimens. Conclusions: The RV+ assay for detection of influenza A and B was simple and relatively rapid with performances comparable to that of the conventional molecular methods- approximately 2.5 hour with 15-minute hands-on time. However, a low throughput is a drawback of this assay as only one sample can be tested at a time.
S417

T253 USING BD LABORATORY CONSULTING SERVICES TO UNDERSTANDING THE IMPACT OF THE PREANALYTICAL PHASE ON SAMPLE QUALITY AND SAFETY, A MULTI COUNTRY PERSPECTIVE K. Schlueter(2), S. Church(1) BD Diagnostics Preanalytical Systems, Oxford, UK BD Diagnostics Preanalytical Systems, Heidelberg, Germany
2 1
T254 VALUATION OF NON CONFORMITIES IN LABORATORY: HOW TO IMPROVE THE QUALITY PROCESS FOR PATIENT SAFETY L. Papay Ramrez(3), J. Alonso Garca(3), M. Robles Garca(3), L. Bolaos Bernal(3), A. Rubio Ruz de la Oliva(3), S. Garca Linares(3), T. De Haro Muoz(1), R. Snchez Navarro(2), N. Coronado Alvarez(3) Laboratory Director. San Cecilio University Hospital, Granada, Spain 2 Medical Analyst. San Cecilio University Hospital, Granada, Spain 3 Resident Specialist in Medical Laboratory, San Cecilio University Hospital, Granada, Spain Objective: Evaluate the incidents that negatively influence in the safety of the patient, registered during the period January -September 2012, which were action derived necessary improvement. Method: We have a record system of Non-Conformities (NC) which the whole personnel has access by login and password. The detected NC and the corrective actions are selected. Statistical methods be used for describing our collection of data: a) failures identification; b) frequency; c) realized action; d) actions done: corrective immediate action (CI), preventive action (PA), critical incident; e) percentage of not solved/solved NC.The quality manager compiles every term these entries and proceeds to its evaluation and reporting that is sent to the Direction. The report includes: a) percentage of registered NC; b) causes; c) effects; d) corrective/preventive measures; e) graphics. This report is object of Annual Review for the Direction. Results: There are 9.929 (3.16%) NC from 314.219 related to errors in the patients identification, 23,9% are not repairable mistakes. Effect: the samples cannot be processed. Corrective immediate action: rejection of requests. Preventive action: information from the Unit of Communication of the Laboratory to the Center/Service petitioner. 38.9% are errors of codification in the steering wheels of request (repeated and/or incorrect codes), CI: recoding by the Laboratory, research and communication causes to the Center/Service petitioner. 23,1% incorrect requests (no origin, no destination, no age, no sex). 14,1% are errors detected in samples. Cause: urgent tests in non-urgent requests that overcome the maximum time among extraction/receipt and the use of inadequate containers, Effect: rejection of samples. Conclusions: The NC valuation provides a starting for his management and control. Is proposed to increase the formation/information in the whole implied personnel actions, in extraction points in order to adapt them to the quality procedures pertinent.
1
Background: The complexity of the preanalytical (PA) phase has precluded standardisation of PA processes, despite its impact on sample quality, laboratory efficiency, or patient & healthcare worker safety. The BD Laboratory Consulting Service Preanalytical Review provides a consistent method, based on a standard data collection form, to audit PA procedures and practices in hospitals in different countries. Blood collection processes were assessed from storage of blood collection materials through specimen collection, transportation, processing of the samples and the resulting sample quality. By following the samples through the complete process, it was possible to link specific PA attributes to sample quality deficiencies. Methods: A consistent method and data collection form were used for audits (N=48) of all blood collection systems. Data were collected by observation of PA phase procedures and practices and resultant sample quality. Results: The PA phase was observed for 3597 blood collection tubes over 1350 collections. Sample quality was assessed for 8016 chemistry and 3532 coagulation tubes. For collections that resulted in hemolysed samples, 48% had prolonged use of tourniquet, 31% used catheters and for 38% the disinfectant was not allowed to dry. For serum samples with fibrin where the PA process had been observed, 26% had less than 30 minutes between collection and centrifugation and 81% had not been mixed. The following list gives the percentage of collections where a particular behaviour was observed, incorrect patient identification procedure, 56%; tubes labelled prior to collection 61%; coagulation tubes filled to less than 90% of tube volume 7%; gloves not worn 37%; incorrect activation of needle safety device 19%. Conclusions: The BD Preanalytical Review standardised methodology allows comparison of results between departments and between institutions and countries. The prospective nature of the reviews permits identification of issues within an institution based on more data than that from rejected samples alone. It makes the link between collection procedures and the consequences for the sample quality & efficiency of the institution, providing evidence for all those involved that improved compliance has an impact.
S418

T255 ONE YEAR EXPERIENCE OF A QUALITATIVE SCORING SCHEME FOR EQA BLOOD SMEAR INTERPRETATION A. Fanelli(1), M. Borsotti(2), R. Caporale(1), A.M.G. Gelli(1), B. Peruzzi(1), M. Quercioli(2)
1
T256 INTERLABORATORY COMPARISON OF METHODS IN LABORATORY MEDICINE IN PREPARATION FOR ACCREDIATION OF THE LABORORATORY R. Findeisen, B. Straube, K. Weber, B. Schmoz, E. Guhr Oberlausitz-Kliniken gGmbH, Institute for Clinical Diagnostric and Transfusion Medicine, Bautzen 2 ELBLAB GmbH Centre for LaborMedizin, Riesa 3 Medizinisches Labor Dresden - Elsterwerda, Dresden, Germany Background: The College of American Pathologists (CAP) has been involved in interlaboratory comparison surveys since 1946. The European Community Confederation of Clinical Chemistry (EC4) has established a working group on laboratory accreditation. The aim of this group is to explore the possibilities for harmonisation of accreditation and quality systems in clinical laboratories in the European Community (EC) (2). Part 1 of ISO/IEC Guide 43 provides guidance on the development and operation of laboratory comparisons for use in proficiency testing schemes [see also DIN EN ISO 15819:2007; points 5.6.6. and 5.6.7.]. Methods: The Department of Laboratory Medicine of the hospital consists of two laboratories in different towns. The distance between the laboratories is 18 kilometres. The Department of Laboratory Medicine will be accreditated by Deutsche Akkreditierungsstelle GmbH [DAkkS] according to the Guidelines 93/42/EWG; 90/385/EWG and DIN EN ISO 15189 in the future. Results: Passing-Bablok-Regression and Pearsons Correlation Coefficient (R) were calculated for the following metohods: Creatinine (Y= 0,949x+4,21; R=0,9992), Bilirubine total (y=1,02x-0,39, R=0,9947), sodium (y=x-1, R=0,839), potassium (Y=x, R=0,987), chloride (Y=x, R=0,947); aspartate aminotransferase (Y=x-0,01, R=0,985), alanine aminotransferase (Y=0,923x-0,0004, R=0,988), gammaglutamyltransferase (y=0,980+0,059, R=0,999), C reactive protein (Y=1,019x-0,48, R=0,999), TSH (Y=1,004x, R=0,985), BUN (Y=1,006x-0,02, R=0,998), uric acid (Y=0,984x-1,3, R=0,999), cholesterol total (Y=0,967x+0.07, R=0,995), HDL cholesterol (Y=0,993x-0,031, R=0,994), LDL cholesterol (Y=0,938x+0,109, R=0,995), triglyceride (Y=1,017x-0,044, R=0,999), total protein (Y=0,977x+0,608, R=0,986), albumine (Y=x, R=0,994), phosphorus (Y=0,941x+0,049, R=0,992), creatin kinase (Y=1,008x+0,0042, R=0,999), alkalic phosphatase (Y=1,017x, R=0,985), alpha-amylase (Y=1,014x0,002, R=0,999), lipase (Y=1,013x-0,07, R=0,988), lactate dehydrogenase (Y=0,938x+0,109, R=0,995), calcium (Y=1,09x-0,21, R=0,95); cholinesterase (Y=1,025x+5,18, R=0,989), magnesium (Y=x, R=0,954) and iron (Y=0,987x0,205, R=1,00). Conclusions: Since two years an interlaboratory competition system of the most tests in clinical chemistry, haematology and haemostaseology has been installed. The competition system works IT-based. The competition system shows that the routine measurement procedures have an acceptable traceability to the reference systems.
1
SOD Laboratorio Generale, AOUCareggi, Firenze SOD Controllo di Qualit in Laboratorio, AOU Careggi, Firenze
2
Aim: In 2012 the regional EQAS organizer has set up a peripheral blood smear qualitative scoring scheme, following the EQALM guidelines. Methods: Registered laboratories in the smear scheme were 179. Samples were selected by the scientific committee of the haematology unit of the AOUC laboratory. One sample of 4 different blood smears stained with MGG was sent to the participants for morphological evaluation: a chronic lymphatic leukaemia (LLC), an haemolytic anemia (HA), an acute lymphoblastic leukaemia (LLA).and a primary myelofibrosis (PMF) Relevant patients details and complete blood count were included. The participants were asked to select up to three significant morphological abnormalities using a coding list provided by the EQAS organizer. A diagnostic suggestion (free text) was well accepted. For each survey, individual results were assessed against the consensus answers given by the referees. The assessment of individual results was calculated using a score system. At the end of each exercise each laboratory received a report with the morphological alterations identified by the participants and by the referees, the reached score, a clinical background of the patient including relevant immunophenotype or cytogenetic data, the diagnosis. Results: Overall participation in this morphology scheme was high (70%). Different performance levels were registered relative to pathological cells: the mean score of exercise n1 (LLC) was 0.40, of exercise n 2 (HA) was 1.70, of exercise n 3 (LLA) was 0.70 and of exercise n 4 (PMF) was 1.3, indicating that pathological lymphoid cells were the most difficult to identify by the participants. Different performance levels were detected relative to the laboratory category: higher score were shown by great laboratories, specialized laboratories and hospital laboratories. No significant differences were found between private and public laboratories. Conclusions: The results revealed substantial concordance with data of literature. We are going to improve the EQA morphology scheme by the introduction of a list of suggested diagnosis and the delivery of a report, at the end of each survey, with a short discussion of the case, some comments and bibliography.
S419

T257 COMMUTABILITY OF TWO REFERENCE MATERIALS FOR TWO COMMERCIAL LITHIUM ASSAYS E. Frusciante(1), I. Infusino(1), C.A. Ferrero(2), M. Panteghini(1)
1
T258 INTEGRATION OF CONTINUOUS QUALITY IMPROVEMENT FOR BIOSAFETY WITH THE CONTINUOUS QUALITY IMPROVEMENT FOR THE ENTIRE LABORATORY: OUR EXPERIENCE G. Giuliani, G. Candelieri, S. Lino, S. Pastori, R. Ottaviano Servizio di Medicina di Laboratorio - Azienda Ospedaliera Guido Salvini Garbagnate Milanese, Milan, Italy Background: Biosafety laboratories must ensure adequate safety conditions to avoid potential hazards associated with the handling of biologic materials, the manipulation of genomes, the creation of synthetic organisms, and the spread of multidrug-resistant bacteria. The protection of laboratory workers from disease agents transmitted by aerosols, droplets, blood and body fluids is a must in different sections of a laboratory such as chemistry, hematology, virology, etc. CDC and NIH addressed the topic in their publication BMBL. Recently, CDC published the Guidelines for Biosafety Laboratory Competencies. This guideline outline the importance of the competencies of the laboratory workers in terms of skills, knowledge, and abilities required for working with biologic agents at the three highest biosafety levels (BSLs 2-4). More recently, the publication of the Guidelines for Safe Work Practices in Human and Animal Medical Diagnostic Laboratories outline the culture of safety, the biological risk assessment and the fundamental safety practices in diagnostic laboratories. Moreover, it is crucial the integration of continuous quality improvement for biosafety with the continuous quality improvement for the entire laboratory. Methods: Of the four domains that BLC follows, we started to introduce the Domain 1 Potential hazards which is comprehensive of the following sub-domains: Biologic materials and Chemical materials. We classified our laboratory personnel (technicians, laboratory specialists and manager) in three different level (entry-, mid-, and senior level) on the basis of the skills and competencies. Results: on a total of 16 and 51 specialists and technicians respectively we trained 8 and 37 at midlevel, 6 and 10 at senior level, 2 and 4 were entry level. Conclusions: the implementation of these guidelines among laboratory workers increased the awareness of biorisk, including willingness to report concerns, response to incidents, and communication of risk. Moreover, it developed an enhanced culture of safety that is open and nonpunitive, encourage questions, and is willing to be self-critical.
Centre for Metrological Traceability in Laboratory Medicine (CIRME), University of Milan, Milan, Italy 2 Sentinel Diagnostics, Milan, Italy Background: The traceability of lithium measurements in serum is ensured by the availability in the JCTLM database of two secondary reference materials (RM), i.e. the NIST SRM 956c (electrolytes in frozen human serum) and the IRMM BCR-304 (lyophilised human serum). As these RM are intended for use in the calibration and traceability validation of commercial systems, information on their commutability is central. Here we tested the commutability of these RM using two commercial methods measuring serum lithium, based on direct potentiometry (Ion-Selective Electrodes Direct, Roche Cobas Integra 400) and on a colorimetric approach (Multigent Lithium, Abbott Architect c16000), respectively. Methods: A total of 27 leftover human serum samples were collected, aliquoted and stored at -80 C until their use. We measured lithium concentrations with the two systems in each biological sample, in SRM 956c (3 levels) and in BCR-304 in duplicate in two different runs on the same day. Manufacturers control materials were used to validate analytical runs. The commutability of RM was estimated from Deming regression analysis of the measured results in native samples using the 95% prediction interval (95PI) and multiples of the standard error of regression (Sy-x), in accordance with the CLSI C53-A standard. Results: The SRM 956c results did not fall inside the 95PI based on the results for the native clinical samples. In addition, using an acceptance criterion for commutability of 3 times the experimental Sy-x (0.066), the relative residuals (rr) for SRM 956c (2.956 level 1, 4.044 level 2, 3.209 level 3) were all outside the acceptable range. On the other hand, BCR304 results fall inside the 95PI, but its rr (0.197) was not within the acceptable range. Conclusions: Our results show that SRM 956c was not commutable between the methods evaluated. BCR-304 showed better, although not perfect, commutability and should be preferred to align lithium assays to higher-order references. The uncertainty of BCR-304 (2.9%) is however relatively high and this may become an issue for fulfilling the goal of acceptable uncertainty of lithium measurement for clinical laboratories (4.3%), as calculated assuming a time interval between doses of 12h and a drug average half-life of 24h.
S420

T259 CALCULATION REFERENCE VALUE CHANGE IN BIOCHEMISTRY LABORATORY M. Morito Aguilar, F. Cava Valenciano, S. Gutierrez Moreno, F.J. Garca igo, A. de Lzar de la Villa, A. Torregrosa Benavent, M.G. Serrano Olmedo Soraya Gutirrez Moreno, Laboratory Medicine Department Background: As one aspect of quality control in the laboratory, we have to provide objective meanings for changes in consecutive results. If we want to establish with confidence a change between successive results, the difference between both has to be higher than the variation of three factors: preanalytical variation (CVp), analytical variation (CVa) and intra-individual biological variation (CVi). If we standardize preanalytical procedures, the CVp is considered minimal and the total variation is reduced to CVi and CVa components. Two consecutive results are significantly different if numerical variation between both is greater than the combined variation inherent to the two results. This value is called Reference Value Change (RVC). Methods: Analytical variation coefficients were obtained from internal quality control measured by an Advia 2400 Autoanalyzer (Siemens Healthcare Diagnostics) during 2011. To calculate the RCV the following equation was used: RCV = 21/2 *Z*(CVa2 +CVi2)1/2. Results: We show the results of the calculation of CVa and RCV for a 95% confidence for each analyte, respectively: Glucose: 1.90, 16.65; urea: 2.65, 34.87; uric acid: 1.48, 25.28; cholesterol: 3.12, 17.28; creatinine: 3.88, 18.20; triglycerides: 5.28, 59.74, total bilirubin: 2.43, 66.30; direct bilirubin: 9.06, 105.03 , gamma-glutamyltransferase: 4.01, 39.83; alanine aminotransferase: 3.63, 68.09; aspartate aminotransferase: 3.16, 34.12; creatine kinase: 3.12, 63.78; amylase: 3.56, 5.26; alkaline phosphatase: 3.35, 20.02; iron: 4.60, 74.54; albumin: 2.73 , 11.45; calcium: 1.94, 7.53, total protein: 2.24, 9.72; Sodium: 1.42, 4.38, potassium 0.98, 13.58; chlorine: 1.78, 5.96. Conclusions: The RCV is easy to calculate and is the most appropriate tool to evaluate changes in successive results. Biological quantities with both high imprecision and an elevated CVi, results in high RCV. The incorporation of the VRC to lab reports is very useful to interpret changes in two consecutive measurements of the same analyte. T260 EVALUATION OF THE IMPACT OF STANDARDIZATION PROCESS ON THE QUALITY OF SERUM CREATININE DETERMINATION IN ITALIAN LABORATORIES I. Infusino(1), A. Carobene(2), F. Ceriotti(2), E. Frusciante(1), M. Panteghini(1)
1 Centre for Metrological Traceability in Laboratory Medicine (CIRME), University of Milan, Milan, Italy 2 Diagnostica e Ricerca San Raffaele, Istituto Scientifico Ospedale San Raffaele, Milan, Italy
Background: Creatinine determination in serum is the key indicator of kidney glomerular function. A reference measurement system for standardization of creatinine measurements is available and virtually all IVD manufacturers have recently aligned their creatinine assays to this system. The aim of this work was to verify if and how these standardization efforts have improved the state of the art of creatinine determination in Italy. Methods: An analysis of Prolarit EQAS results using control materials with target values assigned by a traceable method (enzymatic assay calibrated against the NIST SRM 967) was carried out. Results: Results obtained during 2006, 2010, and 2011 schemes by participating laboratories showed a general good alignment at creatinine concentrations ~2.00 mg/dL, with 2011 results except for one method group well inside the desirable bias (4%). At higher concentrations, whereas the overall bias was small in 2010, for some groups using alkaline picrate (AP) methods it became significantly negative in 2011. The performance markedly worsens at creatinine physiologic concentrations, where a significant positive bias (up to ~20%) is still present for most of the AP-based analytical systems. Unexpectedly, with few exceptions, no evident improvement in individual assay bias was noted from pre- (2006) to poststandardization (2011) periods. The enzymatic method groups were the only always presenting an acceptable bias for all concentration levels, in addition to showing the lowest between-laboratory variability. The number of laboratories using enzymatic methods, however, still remains only 7% of the total. Conclusions: Our EQAS performance data indicate that most of the current standardized creatinine methods based on AP reaction do not perform well, mainly at the lower creatinine concentrations. This inaccuracy of creatinine measurements can adversely impact the estimation of glomerular filtration rate by equations and the evaluation of kidney function in pediatrics.
S421

T261 QUALITY AND PATIENT SAFETY IN HEALTHCARE: ASSESSMENT OF PERFORMANCE OF PREANALYTIC PHASE IN CLINICAL LABORATORIES BY SIX SIGMA OR QUALITY INDICATORS D. ren Emekli, D. Aslan, N. Zorbozan Pamukkale University School of Medicine Depatment of Medical Biochemistry Denizli/Turkey Background: The preanalytic (Pre-A) phase is responsible 46%-77% of total errors in the total testing process. Every laboratory should have a policy for detection and prevention of errors. The frequency of errors should be determined systematically in a standardized manner. In this study, we aimed to assess the performance of Pre-A phase by two methods: the Six Sigma approach and Quality Indicators (QIs) Model developed by the IFCC Working Group on Laboratory Errors and Patient Safety (IFCC WG-LEPS). We compared them for effectiveness. Methods: We selected the most possible Pre-A errors (sampleimproperly labeled; sample-hemolyzed; sample-improper transport; sample-insufficient; sample-damaged; sampleclotted; information (clinical)-insufficient; order-incorrect; order-duplicate; container-inappropriate). We first standardized the terminology and structured the reporting system in our Hospital Information System (HIS) and Laboratory Information System (LIS). Data were collected monthly for the period of January 2012June 2012; analyzed, and the QIs and sigma metrics were calculated. QIs were assessed according to the quality specifications (QSs) proposed by the IFCC WG-LEPS. We have chosen the quality performance as 4.6 sigma which shows 10% waste with 1 000 DPM. The QI value and sigma metrics for each error was evaluated. Results: The QIs for sample-hemolyzed and sample-clotted didnt match the QS for desirable performance. The sigma metrics which are found below 4.6 are as follows: 3.2 (samplehemolyzed), 4.2 (sample-clotted), 4.4 (sample-insufficient) and 4.3 (sample-improperly labeled). The results obtained with both methods are comparable. Conclusions: Our results showed that both QIs Model and Six Sigma approach may be useful for assessment of performance of preanalytic phase. But, the Six Sigma approach may be more promising because the QSs for QIs are not defined and standardized yet. Every laboratory can choose the most applicable assessment method in its own situation, but the HIS and LIS should be structured for error collection and data management, and the skills of laboratory staff also should be improved on data management and the usage of quality tools in order to identify and eliminate the error sources. T262 COMPARING, BY USING SIX SIGMA LEVEL OF THE ANALYTICAL PERFORMANCES OF DIFFERENT LABORATORIES WHICH USE VARIOUS ANALYZERS OF BIOCHEMISTRY O. Gulbahar(1), M. Kocabiyik(1), M.Z. Ciraci(1), M. Cingirt(1), C. Demirtas(1), F. Ucar(2), N. Bayraktar(3), F. Akbiyik(4), M. Yavuz Taslipinar(2) Medical Faculty of Gazi University Diskapi Yildirim Beyazit Treatment and Reserch Hospital 3 Medical Faculty of Baskent University 4 Medical Faculty of Hacettepe University
2 1
Background: The analytical performances of medical laboratories, in terms of patient safety, should be satisfactory. Six sigma level is the most expressive measurement of quality in the evaluation and comparing of the performances of laboratories. In this work, it is purposed to compare the six sigma levels of biochemistry parameters in different laboratories using various analyzing systems.Methods: In the work, it is defined the six sigma levels belonging to the total biochemistry parameters of the Central Biochemisty Laboratory at Gazi University Hospital that have the Beckman Olympus AU2700 devices and the different three laboratories. The six sigma levels are calculated by using internal QC data.Results: While the parameters having sigma levels 6 in our Biochemisty Laboratory using the Beckman Coulter systems are glucose, creatinine, uric acid, total bilirubin, AST, ALT, ALP, HDL, Ca, direct bilirubin for level 1; those having sigma levels <3 are urea, Na, Cl for level 1. While the parameters having sigma levels 6 in the laboratories using the Abbott systems are glucose, creatinine, cholesterol, TG, HDL, total protein, Ca, K, GGT, direct bilirubin for level 1; those having sigma levels <3 are urea, total bilirubin, AST, ALT, LDH, Cl for level 1. While the parameters having sigma levels 6 in the laboratories using the Siemens systems are glucose, TG, HDL, LDH, K for level 1; those having sigma levels <3 are urea, ALT, Ca, Na, Cl, GGT, direct bilirubin for level 1. While the parameters having sigma levels 6 in the laboratories using the Roche systems are glucose, uric acid, AST, ALT, ALP, TG, HDL, total protein, LDH, Ca, K, GGT for level 1; those having sigma levels <3 are Na for level 1. Conclusions: It is examined that the six sigma levels as an indicator of analytical performances in different laboratories are similar in some parameters but different in others. These differences can result from not only the performances of analyzers but also the factors such as the criteria belonging to the QC procedures that are used by respected specialist. As a result, in terms of patient safety, the performances of laboratories can be evaluated by this easy and valid method and if they are under expected performance, it could be made improvement works.
S422

T263 EXTERNAL QUALITY ASSESSMENT (EQA) FOR QUANTITATIVE FECAL BLOOD IN STOOL (FIT) P. Kocna(1), T. Zima(1), M. Budina(2), T. Ichiyanagi(3) Institute of Medical Biochemistry and Laboratory Diagnostics, 1st Medical Faculty Charles University, Prague, Czech Republic 2 SEKK, Pardubice, Czech Republic 3 Eiken Chemical Co. Ltd., Tokyo, Japan Background. Colorectal cancer (CRCA) is the second most frequent malignant disease in Europe. CRCA screening we began in the Czech Republic in 1994, and population-based national screening with FOBT was started in 2002, since 2009 with immunochemical test (FIT). External quality assessment (EQA) of haemoglobin determination in the stool has been started in January 2012 as a part of the national EQA. The aim was to improve the analytical quality of new, quantitative FIT for colorectal cancer screening. This EQA programme is provided by SEKK (member of EQALM) which is accredited according to ISO/IEC 17043:2010 and provides EQA programmes for Czech and Slovak republics. This study will compare our new Czech FOB EQA with 15 years running ECQS annual Eiken program for users of OC-Sensors. Methods. In the Czech Republic the Faecal Occult Blood (FOB) EQA programme is organised 2-times per year, in Japan EQCS once per year, with 2 liquid samples produced by Eiken. The results are collected using web application and evaluated according to the ISO13528. We use robust means of all results as the assigned value for each sample. Also group based statistics (based on manufacturers of kits) are evaluated. Results: There ran 2 rounds of the new FOB EQA programme up to now. Participants in two rounds - May (n=34) and October (n=32) - analysed samples with OC-Sensor analyser (mean CV% were 10 and 8.7), Sentinel-Gold methods (mean CV% were 16 and 18) and Smart Eurolyser POCT (mean CV% were 52 and 76). Eiken ECQS data from 880 facilities indicating smaller CV% with OC-Sensor Diana (3.5 and 3.2), OC-Sensor Micro (5.2 and 3.7), others analyzers (5.8 and 12). Conclusions. The new FOB EQA programme that started in January 2012 was joined by 33 clinical laboratories analysing quantitative FIT samples. Our results confirm the importance of the newly implemented FOB EQA for the quantitative haemoglobin in the stool samples and support further activities aimed to improve quality and optimization of the colorectal cancer screening programme. The results obtained in the Czech EQA rounds show higher CV in comparison to the Eiken EQAS. It is probable that this behaviour is also influenced by the number of participants which is uncomparably larger in the Eikens EQAS.
1
T264 BIOLOGICAL VARIATION AND REFERENCE CHANGE VALUE OF PROCALCITONIN IN HEALTHY INDIVIDUALS D. Pineda-Tenor, E.J. Laserna-Mendieta, J. Timn-Zapata, . Cabezas-Martnez, A. Julin-Jimnez, M. GmezSerranillos-Reus Laboratory of Clinical Chemistry, Virgen de la Salud Hospital, Complejo Hospitalario de Toledo Background: The measurement of procalcitonin (PCT) in patients with systemic inflammation, infection and sepsis has considerable utility, from both a diagnostic and a prognostic point of view. PCT has demonstrated excellent correlation with the severity of infection, and its monitorization has been employed for antibiotic management of patients with systemic bacterial infections. Our objetive was to establish the biological variation (BV), index of individuality (II) and reference change values (RCV) of PCT in healthy individuals. Methods: Samples were collected from 45 subjects (PCT <0.5 ug/L) at 24 hours intervals over 5 days (PCT half-life=24-30 h). PCT was assayed using a Cobas e411 (Roche Diagnostics). The analytical coefficient of variation (CVA) was calculated from the between-run data quality control (N=111). BV data were estimated using nested analysis of variance, according to the method published by Fraser and Harris. The II was calculated as the ratio CVw/CVg (within- and between-subject biological coefficients of variation respectively). The RCV was estimated using the formula RCV=Z(2)1/2 (CVA2+CVw2)1/2 , where Z denotes the level of statistical significance (Z=1.96 for P <0.05). Results: The PCT values ranged from 0.02 to 0.46 ug/L (225 points). The analytical variability estimated by CVA was 7.11%. The BV coefficients, CVw and CVg, were 29.94% and 60.75%, respectively. The index of individuality was 0.49 (II <0.6 suggest marked individuality). The clinically significant difference between two consecutive results estimated by RCV was 85.23% (P >0.05). Conclusions: The low index of individuality shows that the use of population-based reference limits is inadequate for interpretation. In this case, the use of the RCV has been proposed to be the best reading strategy. The inclusion of this data in diagnostic algorithms for sepsis may increases the usefulness of PCT, mainly for monitoring the effectiveness of therapy.
S423

T265 A TWO-YEAR EXPERIENCE OF IMPLEMENTING GENETICAL EXTERNAL QUALITY CONTROL FOR RHD FETAL GENOTYPING BY CNRHP N.R. Da Silva(1), S. Huguet-Jacquot(1), C. Toly-Ndour(1), A. Cortey(2), B. Carbonne(2), A. Mailloux(1)
1 Centre National de rfrence en hmobiologie prinatale, ple Biologie mdicale et pathologie, GHU Est Parisien, APHP, Paris 2 Centre National de rfrence en hmobiologie prinatale, ple prinatalit, GHU Est Parisien, APHP, Paris
T266 IMPLEMENTATION OF A PROFICIENCY TESTING FOR THE ASSESSMENT OF THE PREANALYTICAL PHASE OF BLOOD RNA F. Malentacchi(1), K. Guenter(2), P. Verderio(3), S. Pizzamiglio(3), C. Ciniselli(3), A. Tichopad(4), M. Kubista(5), R. Wyrich(2), M. Pazzagli(1), S. Gelmini(1)
1 University of Florence (UNIFI), Clinical Biochemistry Unit, Department of Clinical Physiopathology, Italy 2 QIAGEN GmbH, Hilden, Germany 3 Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy 4 Institute of Biotechnology AS CR, Prague, Czech Republic, 5 Charles University, Faculty of Medicine in Pilsen, Pilsen, Czech Republic 5 TATAA Biocenter AB, Gothenburg, Sweden, and Institute of Biotechnology, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Background: A positive RHD fetal genotyping diagnoses the RH1 fetomaternal incompatibility for the anti-RH1 alloimmunized pregnant women. For the non-immunized ones, a negative test will avoid injection of IgRH. Since the RHD fetal genotyping became a key to the monitoring of RH1 negative pregnant women, an increasing number of laboratories performed such test. It appeared essential for the CNRHP and part of its missions, to offer a quality assessment program based on an external quality control (EQC). The CNRHP can rely on more than ten year experience in the fetal RHD genotyping by PCR from maternal blood and its EN ISO 15189 acreditation to establish such control. The aim of this presentation is to review the EQC program two years after its launch. Methods: Positive control specimen were prepared from RH1 negative plasma donors spiked with various concentrations of RH1 positive plasma in order to reflect RH1 positive fetuses of different gestational ages. Negative control specimen, made from RH1 negative plasma donors, remained unspiked. Once tested, the samples were conveyed to the laboratories with a feedback form where they had to state the material and methods used, the results and the clinical biological interpretation. The control samples were sent twice a year. Results: Over these two years, 9 series of samples were prepared and sent to 6 or 7 laboratories (3 in 2010, 4 in 2011, 2 in 2012) reaching each year a 100% response rate. In 2010, the EQC results were consistent with those expected although the laboratories use different extraction and amplification protocols. In 2011, two laboratories made erroneous clinical interpretations despite right analytical results. In 2012, the EQC concluded to both analytical or/and interpretation errors. Only a single laboratory returned the right analysis with the good clinical interpretation. Conclusion: The presented EQC responds to the criteria required to evaluate the practices of laboratories performing fetal RHD genotyping. The ideal EQC should be prepared from maternal plasma from a single pregnant woman containing a predetermined quantity of fetal DNA but the collection is impossible in practice. The next step is the transfer of EQC program conducted by the CNRHP to an EN ISO/IEC 17043 certified entity.
Background: Molecular in vitro diagnostics will play an important role in future health care practice and gene expression profiling promises to provide insight into normal biological and pathological processes with the hope of predicting disease outcome and indicating individualized courses of therapy. In this field, significant improvements of downstream assays and data analysis (analytical process) have been made during the last years. In contrast, the influence of the pre-analytical steps, such as sample collection and stabilization, have been highly underestimated. SPIDIA (Standardisation and improvement of pre-analytical procedures for in vitro diagnostics, www.spidia.eu) is a four-year largescale integrated project funded by the European Commission that works on the standardization and improvement of preanalytical procedures for in vitro diagnostics in order to close the gap between the more elaborated analytical procedures and the less standardized pre-analytical processes. Methods: We performed a multicenter study within the EUgranted SPIDIA project to investigate blood collection and shipping influence on some RNA quality parameters: yield, purity, integrity, RT-qPCR interference and IL1B, IL8, c-Fos and GAPDH gene expression. Two models were designed: ExpA Ten laboratories collected blood from an own donor into two different tubes (with or without stabilizer) and extracted RNA at two different times; Exp B - Blood was drawn from a single donor and shipped to ten laboratories in two different tubes (with or without stabilizer) for RNA extraction. Results: In both models and collection tubes, reliable results were obtained for purity, yield, GAPDH expression, and interferences. For blood collected in unstabilized tubes we observed a substantial variation in RIN (ExpA) and in transcription levels of IL1B, IL8 and c-Fos (ExpB). Overall the variability was higher among data obtained from unstabilized blood samples. Conclusions: We defined the experimental setup for a larger ring trial throughout Europe. The chosen downstream analyses verified their potential, serving as adequate markers to test quality of blood RNA.
S424

T267 HARMONIZATION OF AN LC-MS/MS ASSAY FOR THERAPEUTIC DRUG MONITORING OF THE IMMUNOSUPPRESSANT DRUG TACROLIMUS D. Mason(1), T. Annesley(2), E. Champarnaud(3), C. Mussell(3), L. Harter(1), L. Calton(1), D. McKeown(4) Waters Corporation, Milford, USA The University of Michigan 3 LGC Limited 4 Analytical Services International Ltd
2 1
T268 ACCREDITATION OF CLINICAL BIOCHEMISTRY LABORATORY DZ SAVSKI VENAC BY THE STANDARDS OF THE AGENCY FOR ACCREDITATION OF HEALTH CARE INSTITUTION OF SERBIA (AZUS) V. Milatovic Jezdic1, J. Mitrovic2, S. Jankovic1
2 1
Primary Health Center Savski venac, Belgrade University Childrens Hospital, Belgrade
Background: Monitoring blood levels of the immunosuppressant drug tacrolimus in solid organ transplant recipients is considered the standard of care. Trough concentrations are commonly monitored and are generally regarded as a good surrogate for tacrolimus exposure. Thus, dose adjustments that are critical to regulating the appropriate level of immunosuppression are made in part based on laboratory results. While liquid chromatography-tandem mass spectrometry (LC-MS/MS) is often considered the goldstandard, tacrolimus test results are not yet harmonized across laboratories because of the wide variety of calibrators, sample pre-treatment protocols and instrumentation used today. Here we demonstrate the successful harmonization of tacrolimus measurements using a commercially available test kit, through a proficiency testing survey. Further, we compare these results to a reference measurement procedure for tacrolimus. Methods: A 40 member whole blood panel (spanning the range of approximately 2.0 25 ng/mL tacrolimus) consisting of twenty pooled patient samples and twenty tacrolimussupplemented samples was prepared and distributed to seven laboratories in the United States and Europe. All testing was performed on a common LC-MS/MS system and as specified in the kits directions for use. Four patient pools were prepared in sufficient volume to allow for value assignment by an exactmatch isotope dilution mass spectrometry method (EM-IDMS), i.e., a reference measurement procedure (RMP). Results: Good agreement between all laboratories was observed for the proficiency testing panel (2.0 5.4% CV for patient pools and 3.7 12.2% CV for supplemented samples). Further, for the four sample pools having value assignment by EM-IDMS (4.58, 7.66, 11.90 and 19.82 ng/mL) the difference between this RMP and the mean measurements from the participating laboratories ranged from 0.6 4.4%, demonstrating excellent accuracy of this routine assay across the currently accepted therapeutic range for tacrolimus. Conclusions: The authors believe this is the first study to demonstrate harmonization of an LC-MS/MS assay for tacrolimus across multiple laboratories and should lay the foundation for similar surveys involving other measurands.
Background: Agency for Accreditation of Health Care Institutions in Serbia was established on the basis of the Law on Health Care to perform professional, regulatory and development activities. This law bestows her setting standards for the accreditation of health institutions. Accreditation is a process for evaluating the quality of health care institutions by applying the optimal level of acceptable standards. Standards have been developed and recommended by health experts and used the guidelines of the International Association for quality in health care. Methods: Qualitative research methods: a retrospective analysis and database searching. Results: Accreditation Standards for Primary Care Centres may be divided into three categories: 1. Patient care standards: These standards are structured to follow the process of care for the patient, from the time of intake into a primary care service, through planning and delivery of car to completion of care or referral to another form of care. 2. Support service standards: Support service sections have been developed: environment, human resources and information management. 3. Leadership: These standards cover the leadership in process accreditation goes through the following phases: Application for accreditation: Primary Care Centre institutions obtain the necessary documents from the AZUS. Self-assessment: performed by the staff. They assess the quality of work and evaluate compliance with the standards. In this way they are looking at areas that could be improved. Assessment by surveyors: is an independent evaluation by a trained assessor hired by AZUS. Continuous audit: contribute to maintaining the level of quality that is achieved. Laboratory accreditation in Primary Care Centres is based on a comparison of laboratory performance with the given standards. AZUS has developed 8 standards. Conclusions: Accredited facility has proved to have: detailed procedures for all activities that are conducted, comprehensive quality system that actively searches for problems in providing services and tries to solve them. The advantages gained through accreditation are as follows: improving the quality of health services, reducing risk and increasing safety for patients and employees and reducing costs.
S425

T269 ESTABLISHING QUALITY ON PROCESS IMPROVEMENT AT PRE-EXAMINATION PHASE - THE WAY FORWARD L. Lam, S.K. Ong Department of Laboratory Medicine, Alexandra Hospital, Singapore Introduction: Our laboratory that serves 300-bed General Hospital started operation on August 2010 and we achieved College of American Pathologists (CAP) accreditation 8 months later. Quality indicators on specimen unacceptability are measured, collected and data-transformed knowledge served as continuous process improvement. We report our laboratorys journey into operational delivery performance at the pre-examination bench. Material and methods: At first contact with specimens on preexamination bench, staff at Core Laboratory use predefined quality indicators to accept specimen. Quantitative analysis data on rejected specimens are tracked weekly during our Quality Management Meeting. Specific actions are acted upon and data are being interpreted. Interventions taken and process monitored. Data on unacceptability (August 2010-December 2012) were calculated for trends and drifts. Results: Longitudinal data showed weekly unacceptability decreased from 2% to 0.4% (lowest) after several Plan-DoCheck-Action cycles. Interventions such as introduction of laboratory phlebotomist service, lectures on specimen collection are conducted during the Resident and Nursing Orientation Days, reporting of unacceptable specimens to Incident Hospital Reporting System managed by Medical Affair Department and returning of memo received from doctors/nurses for minor amendment on forms/tubes back to wards, out-patient clinics and Emergency Department(ED). ED had the highest % of specimens being rejected because of anticoagulated tubes clotted (50%), under-filled citrated tubes (20%), labeling errors (15%) and insufficiency (10%). Conclusion: Specimen rejection has remained at mean 0.7% (1.1% to 0.4%) since March 2011. Similar data from Microbiology Division shows mean rejection at 9% with poor specimen collection as the main specimen rejection criterion. Plan to improve specimen quality at ED is being discussed. In summary, monitoring of quality indicators during regular meetings, follow-up actions and interdepartmental cooperation ensured continuous process improvement in attaining stable quality service and patient safety. T270 MEASUREMENT UNCERTAINTY IN THE MEDICAL LABORATORY - IMPLEMENTATION AND EVALUATION OF TWO DIFFERENT FORMULAS IN CLINICAL CHEMISTRY PARAMETERS: TOTAL CHOLESTEROL, CREATININE AND GLUCOSE MEASUREMENTS. R. Plcido Raposo(1), A.M.M.S. Freitas(2), M.A. Gomes(3), S.L. Morgan(4), M. Fasca(5) School of Health, University of Algarve Portuguese Institute for Accreditation 3 School of Health, University of Algarve 4 Cranfield Health, Cranfield University 5 Gnstica, Clinical Laboratory
2 1
Background: Laboratory Accreditation by ISO:15189 requires the application of uncertainty calculation in medical laboratories. Specifically it requires laboratories to calculate this data, and to provide patient results along with their associated uncertainty values. The Guide to the expression of uncertainty in measurement describes in general terms the calculation of measurement uncertainties (MU); however, the specific aspects of clinical laboratory testing are not addressed. Several models of uncertainty have been studied, but there is no consensus on implementation. Methods: The aim of the study was to investigate two potential MU formulas (MU-A and MU-B) within a Portuguese ISO:15189 Accredited Clinical Laboratory. Results were compared with Total Error, currently accepted by the Portuguese Institute for Accreditation as MU value. Three accredited analyte measurements were considered: Total Cholesterol, Creatinine and Glucose, measured in human serum samples using two Cobas 6000c501(Roche) analysers. Results: Cholesterol - Normal: TE=4.7; 5.4%; MU-A=4.2; 4.8%; MU-B=4.6; 5.1%; Pathological: TE=4.2; 4.5%; MU-A=3.7; 4.1%; MU-B=4.1; 4.4%; Creatinine - Normal: TE=12.5; 10.9%; MU-A=11.6; 10.4%; MU-B=9.9; 8.5%; Pathological: TE=10.5; 9.9%; MU-A=10.2; 9.8%; MU-B=8.2; 7.7%; Glucose - Normal: TE=3.9; 4.4%; MU-A=3.6; 4.0%; MU-B=7.0; 7.2%; Pathological: TE=3.9; 4.6%; MU-A=3.6; 4.0%; MU-B=7.0; 7.3% Conclusions: MU-B formula was capable to provide reliable values, allowing definition of procedures variability, well representing the dispersion of values reasonably attributable to the measurand final result. Became clear the necessary Investment from manufacturers, Reference Laboratories and International Organisations to promote and produce certified reference material with high metrological traceability, focusing values at levels of critical decision, with uncertainties associated to the assigned values. General laboratory investment is also needed to improve practice in the preanalytical phase, but also to assess and evaluate their own specific pre-analytical uncertainty. In addition, guidelines and tables with new goals/limits, defined according to the evaluation methodologies and tools being introduced in the clinical laboratory, must be developed.
S426

T271 EVALUATION OF FLUORIDE-CITRATE MIXTURE IN PRESERVING GLYCEMIA AFTER COLLECTION. L.F. Senz Mateos, T.J. Palomino Muoz, A.U. Muoz Colmenero, C.M. Cabrera Morales, E. Fernndez Grande, A. Sastre Gmez, P. Garca-Chico Seplveda Clinical Laboratory Department, University General Hospital of Ciudad Real, Spain Introduction: The preanalytical loss of glucose during the first 1-2 h after collection constitutes a large source of error that interferes with both diabetes diagnosis and the process of clinical decision-making regarding management. The standardized pre-analytical processing (i.e.: centrifuging samples immediately in a refrigerated centrifuge and placing the removed plasma promptly into an ice slurry) is not a practical solution in our outpatient blood collection area composed of 51 blood collection facilities. As it has been reported that acidification of blood drawn into tubes containing sodium fluoride inhibits glycolysis quickly, our aim is to study in our setting whether glucose level in blood collected into tubes containing fluoride-citrate mixture is effectively preserved within two hours after collection. Material and methods: The study was carried out from 25 selected outpatients coming to our blood collection unit. Venous blood was drawn into both two tubes prepared with lithium heparin and two tubes containing fluoride-citrate mixture. Pairs of tubes were stored under transport conditions for one or two hours and then centrifuged and analyzed for glucose. Paired Students t-test was used for glucose levels randomized comparison. Results: Significance paired Students t-test revealed that glucose levels in tubes prepared with lithium heparin and stored one hour after collection were higher than those stored two hours after collection by approximately 8 mg/dL on average (P <0.05). There was not statistically significant difference of glucose levels between both times as far as tubes prepared with fluoride-citrate mixture were concerned. The closeness between both DIF1FLU and DIF2FLU curves shows that there is no change in blood glucose level at any patient over two hours as far as citratefluoride mixture is concerned. On the contrary, the gap between DIF1HEP and DIF2HEP curves reveals that glucose level falls in the course of two hours at every patient as far as tubes prepared with lithium heparin are concerned. Conclusions: We conclude that drawing blood into tubes containing fluoride-citrate mixture to preserve glucose within two hours after collection is an excellent alternative for the standardized preanalytical processing. T272 SELECTION OF REFERENCE GENES FOR EXPRESSION ANALYSES IN LIVER OF RATS WITH IMPAIRED GLUCOSE METABOLISM A. Hernndez1, R. Curi2, L. Salazar3
1 Centro de Biologa Molecular & Farmacogentica, Universidad de La Frontera, Temuco, Chile 2 Dept. of Physiology and Biophysics, Institute of Biomedical Sciences, University of So Paulo, So Paulo, Brazil 3 Escuela de Tecnologa Mdica, Universidad Catlica de Temuco, Chile
Background: Hepatic gene expression studies are vital for identification of molecular factors involved in insulin resistance. However, the need of normalized gene expression data leads to the search of stable genes useful as reference in specific experimental conditions. To evaluate expression stability of potential reference genes for real-time PCR gene expression studies, in rats with insulin resistance, early programmed in intrauterine environment for maternal insulin resistance and triggered by exposure to a high sucrose and fat diet in adult life. Methods: Male rats coming from insulin resistant (F1IR) or normal (F1N) mothers were fed a standard rodent diet from postnatal day 21 to 56, and then divided in two groups each. One group from normal and one from insulin resistant mothers were fed a high sucrose and fat diet (groups F1lR + HSFD and F1N + HSFD respectively) and the rest were fed a control diet (groups F1lR + CD and F1N + CD) for 14 days. Afterwards, glucose metabolism related tests were performed. After liver extraction, RNA was isolated and gene expression analyses of seven potential reference genes (Actb, Gapdh, Gusb, Hprt1, Ldha, Rpl13a and Rplp1) were performed. LinRegPCR software was used to analyze raw data and determinate baseline corrections, threshold lines, efficiency of PCR reactions and corrected Cq values. Evaluations of gene expression stabilities and of the number of necessary genes for normalization were assessed with geNorm tool. Results: All samples from all groups showed acceptable PCR amplification efficiencies. The most stable genes were Rplp1, Ldha, Hprt1 and Rpl13a and the less stable was Gapdh. For all groups, just 2 to 3 of the most stable genes were necessary to optimal gene expression data normalization in rat liver. Conclusion: Genes encoding ribosomal proteins are the most appropriated for normalization of expression data in the presented animal model. By contrast, Gapdh, one of the more used genes in normalization, is not recommendable due to its high intergroup variation.
S427

T273 EXTERNAL QUALITY ASSESSMENT SCHEME FOR FAECAL OCCULT BLOOD TESTING (FOBT): AN ESSENTIAL TOOL TO ASSURE THE RELIABILITY OF RESULTS L. Sciacovelli, S. Secchiero, A. Faggian, L. Zardo, M. Seguso, M. Plebani Centre of Biomedical Research, University-Hospital of Padova Introduction: The colorectal cancer (CRC) is the most common newly-diagnosed cancer and the second most common cause of cancer deaths in Europe. The objective of a CRC screening programme is to discover disease in its early latent stages and, consequently, reduce mortality. To achieve the potential benefit of CRC screening, quality must therefore be optimal at each step in the process. The reliability of FOBT results and their decisional levels are fundamental to assure the correct identification of clinical condition. International guidelines recommend the adoption and use of an Internal Quality Control procedure and the satisfactory participation in External Quality Assessment Scheme (EQAS) to ensure standardized and reliable FOBT results. In this context the Centre of Biomedical Research implemented a specific EQAS to monitor and assess the analytical performances of laboratories. The aim of this work is to describe the results obtained during these years. Methods: The results of participants to EQAS, from 2006 to 3rd survey of 2012 (46 samples, about 90 participants), have been analysed and grouped on the basis of concentrations. In particular the inter-laboratory variability (CV%) of diagnostic systems used by participants have been evaluated and results have been compared with each decisional level used. Results: The concentrations range, the percentage of agreement among laboratories about results interpretation and the CV% average related to quantitative methods are reported. <15ug/L: negative = 98.3%, CV% = 67.5%; 30-50ug/L: negative = 74.7%, positive = 24,7%, doubtful = 0.6%, CV%=30.6%; 55-79ug/L: negative = 64.9%, positive = 33,3%, doubtful = 1,8.0%, CV% = 14.6%; 80-96 ug/L: negative = 48.4%, positive = 45.8%; doubtful = 5.8%, CV% = 12.8%; 100182 ug/L: positive = 89.7 %, negative = 9.5%, doubtful = 0.8%, CV% = 9.3%; 220 -1381ug/L: positive = 98.1%, negative = 1.9%, CV% = 10.1%. Moreover, data demonstrate that only the 77% of laboratories use the cut-off advised by international recommendations (100 ug/L). Conclusion: The participation in EQAS is a useful tool for clinical laboratories to monitor their performance, to know the improvement needs and analytical performance of diagnostic system commercially available. T274 EXTERNAL QUALITY ASSESSMENT SCHEME FOR BLOOD GAS ANALYSIS: AN ESSENTIAL TOOL TO CONTROL THE RELIABILITY OF RESULTS OF POCT S. Secchiero(1), A. Faggian(1), E. Babetto(2), C. Marcato(2), P. Carraro(2), M. Plebani(2)
1 Centre of Biomedical Research, University-Hospital of Padova, Italy 2 Department of Laboratory Medicine, University-Hospital of Padova, Italy
Background: Many regulatory bodies have developed standards for the performance and governance of Point-ofCare Testing (POCT). In accordance with the risk management perspective, all users of POCT are required to undergo specific training managed and coordinated by the laboratory which must assume the responsibility of POCT. Moreover, the adoption of an Internal Quality Control procedure and the satisfactory participation in External Quality Assessment Scheme (EQAS) to ensure standardized and reliable POCT results, are recommended. The Department of Laboratory Medicine of Padova consists of 3 laboratories and a specific laboratory staff has the role of supporting all POCT and blood gas analyzers in the 3 hospital sites. Since 2011, more than the 4 blood gas analyzers in the laboratories and 20 POCT decentralized in care units participated in the EQAS of the Centre of Biomedical Research. The aim was to investigate the performances of the 24 blood gas analyzers over 20 months for ten parameters of the EQA scheme: pH, pO2, pCO2, tCO2, Na+, K, Cl-, Ca++, Glucose and Lactate. Methods: Data relating the performances of the 24 blood gas analyzers (10 Siemens RapidLab 1265, 13 Siemens RapidPoint 405 and 1 IL Gem Premier 3500) were extracted from the overall data of EQAS participants (145), for a total of 2234 results. In the EQAS the analytical performance is calculated as Index Score (IS) with the following acceptability limits (Total error, %), based on biological variation: pH = 0.5, pO2, pCO2, tCO2 = 8.6, Na+ = 2.0, K = 4.4, Cl- = 3.3, Ca++ = 6.1, Glucose = 5.2, Lactate = 15.2. Results: Total number of unacceptable performances (PNA), for parameter: pH=0, pO2=29, pCO2 =8, tCO2=9, Na+= 2, K = 2, Cl- = 1 Ca++=15, Glucose=16, Lactate=0. For blood gas and glucose the higher number of PNA was associated to samples with lower concentrations. For pO2 and Ca++ the higher number of PNA was observed for the 3 POCT of a single hospital site. These results are also affected by pre-analytical problems. Conclusions: EQA Program for blood gas analysis appears an essential tool to monitor the analytical quality of POCT. It allows to identify the POCT with unacceptable performances that need corrective actions coordinated by the laboratory.
S428

T275 BLOOD GAS ANALYSIS IN ITALY: STATE OF THE ART RESULTING FROM THE EXTERNAL QUALITY ASSESSMENT PROGRAM. A. Faggian, S. Secchiero, L. Sciacovelli, M. Plebani Centre of Biomedical Research, University-Hospital of Padova, Italy Background: External Quality Assessment (EQA) Program for blood gas in Italy was implemented for the first time in 2009 by the Centre of Biomedical Research. At present 80 laboratories and 65 POCT (Point of Care Testing) for a total of 145 are participating. The scheme consists of four surveys of two liquid control samples each. Aim: the evaluation of the state-of-the-art of the parameters analyzing the inter-laboratory variability and the degree of harmonization among diagnostic systems. Methods: Data from 20 control samples distributed from 2010 to 2012 EQA cycles, grouped for diagnostics systems (IL GP3000 = 27, IL GP4000 = 18, Radiometer ABL = 20, Roche Omins/Cobas = 7, Siemens RapidLab = 16, Siemens RapidPoint = 57) were analyzed. Results: Inter-laboratory variability (in the concentration range studied), calculated as mean CV% and (range): pH (7.10-7.64) = 0.11 (0.07-0.14), pO2 (67.5-149.3 mmHg) = 4.98 (3.55-6.58), pCO2 (17.0-79.7 mmHg) = 3.74 (2.48-5.46), tCO2 (17.6-26.6 mmol/L) = 3.10 (2.45-3.95), Na+ (112.6-165.2 mmol/L) = 0.67 (0.46-0.95), K+ (2.74-6.61 mmol/L) = 1.62 (0.81-2.69), Cl(74.0-119.0 mmol/L) = 1.25 (0.87-1.88), Ca++ (0.54-1.89 mmol/L) = 2.20 (1.42-3.72), Glucose (1.90-20.1 mmol/L) = 2.95 (2.38-4.06), Lactate (0.88-7.90 mmol/L) = 4.33 (3.75-5.26). The analytical variability results sufficiently low for all parameters with the exception of pO2 and lactate whose CVs% mean are between 4 and 5. RapidLab shows the higher variability for pO2, Ca++ and Glucose, RapidPoint for pCO2 and tCO2, GP3000 for Na+ and Lactate, GP 4000 for K+ and ABL for Cl.Harmonization among diagnostic systems is very high for pH and Na+, while some critical situations appear for pO2, tCO2, Ca++, Cl- and Lactate. E.g.: for pO2, Omnis/CobasB221 and RapidLab show a bias (positive and negative, respectively) depending on concentration; for Ca++ Radiometer ABL shows a positive bias depending on concentration; for Cl- GP4000 shows a mean bias of +3.6%. Conclusions: EQA Program for blood gas analysis is useful to define the state of the art of analytical performance of diagnostics system commercially available and appears an essential tool for clinical laboratories to monitor the analytical quality of POCT. T276 ACCURACY ASSESSMENT OF SERUM CREATININE MEASUREMENT ON ROCHE MODULAR DP AND VITROS ANALYZERS S. Shiesh(1), W. Yu(1), C. Chang(1), W. Tsai(2)
1 Department of Medical Laboratory Science and Biotechnology, National Cheng Kung University, Tainan, Taiwan 2 Department of Clinical Pathology, National Cheng Kung University Hospital, Tainan, Taiwan
Background: Serum creatinine concentration is critical for the assessment of renal function. We aimed to develop an isotope dilution liquid chromatography tandem mass spectrometry (LCIDMS/MS) method for measuring creatinine, and to assess the accuracy of serum creatinine measurement on Modular DP (Roche) and Vitros analyzers. Methods: Serum samples were deproteinized by methanol containing the internal standard (IS) D3-creatinine. Supernatants were analyzed using a LC-MS/MS (API 5000) system in positive ionization mode. The mobile phase consisted of 95% of 0.025% formic acid and 5% of methanol. Creatinine and IS were quantified using ion transitions of m/z 114 to 44 and 117 to 47, respectively. The method was calibrated with the purified standard reference material SRM 914a from NIST and assessed the accuracy by analyzing SRM 967. Serum creatinine concentrations in patient serum samples (n=40) determined by kinetic Jaffe method (Roche), enzymatic method (Vitros), and LC-MS/MS were compared. Results: Intra-assay imprecision of the LC-MS/MS method was 1.5% and 0.4% at 73.0 and 536 mol/L creatinine, respectively, while inter-assay imprecision was 4.3% and 1.0%, respectively. The method was accurate indicated by the recovery in SRM 967 (99.81.0% and 99.60.6% of target value for levels 1 and 2, respectively). A good correlation was obtained between the LC-MS/MS method and Jaffe method (Y=0.984X-0.006, r=0.999), and LC-MS/MS and enzymatic method (Y=1.028X0.125, r=0.999). Hemolysis (up to 20 g/L) caused falsely high results in Jaffe method, but tended to have lower creatinine results in enzymatic method. Conclusions: We have developed a fast and simple method for the quantification of serum creatinine by isotope dilution LCMS/MS and used this method as a reference method to validate serum creatinine automated assays.
S429

T277 COMMON URINALYSIS CURRENT PRACTICES AND WAYS TO IMPROVE THEIR RELIABILITY M. Shishenkov(1), S. Lukova(1), K. Nancheva(2), V. KolevaTopova(3)
1 2
T278 ENZIMATIC CREATININE: LIMIT OF BLANK (LOB) AND VERIFICATION OF DETECTION LIMIT (LOD). USE OF ACTIVATED CHARCOAL AS BLANK SAMPLE L. Contardi(1), M. Torres(2), N. Dolcemascolo(1), S. Quiroga(1), V. Correa(1)
1 2
Dept. Clin. Lab. and Immunol. Military Medical Academy Clin. Lab. Multiprofile hospital Stara Zagora 3 Clin. Lab. Tokuda hospital Sofia Introduction: Urinalysis is one of the first tests used in medicine from the uroscopy to qualitative and quantitative automated analyses today. It makes up to 12% of the workload in the routine Bulgarian laboratory. It consists of 10 13 dry chemistry parameters and microscopic investigation, performed by qualified medical personnel. Urinalysis is used for screening, diagnostic and monitoring purposes. The laboratory is responsible for reliable test results. The goal is to present our experience in applying automated analyzers (DIRUI H800/FUS100) in the routine urine lab, the dry chemistry and microscopy quality control, and explain a surprising source of interlaboratory variation in the microscopy. Material and methods: Two years long we are using DIRUI H800/FUS100, >35000 urines are tested, all with dry chemistry and microscopic investigations. Productivity is presented as time/sample for 1, 10 and 230 samples. The quality is controlled with materials, purchased by the producer and through manual microscopic confirmation of selected samples - with pathology. Control cards can be plotted manually or with DIRUI software. Regression comparisons are calculated between manual (x) and automated (y) counts of RBC and WBC in 292 urines, analyzed by the workstation and manual microscopic. Results: The productivity of the workstation is 1 min/sample (230 sample series), 1.5 min/sample for 10. Single sample takes 2.5 minutes (all with 10 dry chemistries + microscopy). Comparison of WBC counts n = 162; y=1.189.x+1.3 r =+0.913; RBC counts n = 130 y=0.847.x + 7.3; r = +0.870. We were surprised to see, that the high power field (HPF) of objective 40 is not a reliable measuring unit to present the microscopic findings. We measured the diameters of HPF 40x of 12 microscopes and calculated a CV of 31%; when the HPF surface is calculated the CV rises up to 73%! Discussion: Optimal productivity and quality of results is achieved through automation and standardization of the urinalyses. The technician is responsible for the quality and is an expert evaluates the findings on the PC screen, verifies them and selects samples for manual microscopic confirmation
CEMIC Departamento de Anlisis Clnicos ProgBA - Programa Buenos Aires
Background: LoD is the lowest amount of analyte in a sample that can be detected regardless of error. LoB is the highest measurement result that is likely to be observed for a blank sample (without analyte). In a Clinical Chemistry lab it is necessary to know analytical performance at some analytes low concentrations; for example, enzymatic creatinine, used in pediatrics replacing Jaffe reagent, whose Limit of Quantification is high. The difficulty lies in getting a sample whose creatinine concentration is zero. So, we verified Creatinine Enzimatic reagent`s LoD manufacturers claim. Methods: reagent: Creatinine Plus, Roche Diagnostics. Analyzer: Cobas 6000. To set the blank limit, we use an activated charcoal adsorbed serum, whose protein concentration was 6,7 g/dL and albumin 4,6 g/dL. This ensures a suitable matrix, similar to patients samples. In this serum, the creatinine concentration reported by the instrument was 0,00 mg/dL. The adsorbed sample was processed to get 20 replicates, in order to calculate mean and standard deviation (SD) of absorbance measurements. To calculate LoB we used LoB = meanblk + 1.65 SDblk. In order to express LoB in mg/dL, we used a serum of known concentration. We plotted a straight line with these two mg/dL points, to interpolate LoB absorbance. To verify LoD, we measured 35 replicates (at different days) babys serum dilutions (initial concentration: 0.14 mg/dL) and calculated SD. LoD=LoB+1.645 SD; LoD manufactures claims = 0.056 mg/dL . Results: LoB = 0,048118 (expressed as absorbance); LoB = 0,009 mg/dL (< 0,01 mg/dL); LoD=0,046 mg/dL. Conclusions: We obtained LoD=0.046 mg/dL, lower than manufacturer claims (0.056 mg/dL). So, LoDs Creatinine Plus Reagent is therefore verified and may be used to measure creatinine in pediatrics and in adults patients with decreased muscle mass, such as, burned, mutilated patients and pregnant women.
S430

T279 EMPOWER YOUR LAB - NEW INTEGRATED EQA DESIGN H. Stepman(1), U. Tiikkainen(2), L. De Grande(1), J. Pelanti(2), D. Stckl(3), H. Laitinen(2), L. Thienpont(1)
1
T280 IMPROVING THE COMPARABILITY OF PARATHYROID HORMONE MEASUREMENTS A REPORT FROM THE IFCC WORKING GROUP FOR PTH C. Sturgeon, W.D. Fraser, R. Singh, J. Souberbielle, S. Sprague, H. Vesper Dept of Clinical Biochemistry, Royal Infirmary of Edinburgh, UK Background: Renal physicians strive to maintain PTH concentrations for patients with chronic kidney disease (CKD) within guideline limits, but poor method comparability means there is currently serious risk of clinical misinterpretation. The potential for under-or over-treatment is significant, representing a major challenge to patient safety. Method: At a meeting convened in September 2010 and attended by representatives of relevant clinical and scientific professional organisations and manufacturers of most PTH methods, the current status of PTH measurement was reviewed and priorities for improvement identified. The IFCC Scientific Division subsequently established a Working Group for PTH with the aim of undertaking this work. Results: The Working Group is actively raising awareness of clinical implications of method-related differences in PTH among both clinicians and laboratorians. Establishing preanalytical requirements for PTH is also a priority with a systematic review of stability and specimen requirements currently in preparation. In the longer term, re-standardization of PTH methods in terms of an appropriate International Standard (IS) is required. Provided its commutability can be assured, the recently established IS 95/646 for PTH1-84 is a suitable candidate. Establishment of a well-characterized panel of samples of defined clinical provenance to enable manufacturers to determine appropriate reference intervals and clinical decision points is also being undertaken by the Working Group and will provide an invaluable clinical resource. Recent developments in mass spectrometry mean that a candidate reference measurement procedure for PTH is now achievable and will represent major progress. Concurrently, evidencebased recommendations on clinical requirements and performance goals for measurement of PTH are being developed. Conclusions: Improving the comparability of PTH results requires support from many stakeholders but is achievable.
Laboratory for Analytical Chemistry, Faculty of Pharmaceutical Sciences, Ghent University, Ghent, Belgium 2 Labquality, Helsinki, Finland 3 STT-Consulting, Horebeke, Belgium Background: Labquality has experience in performing external quality assessment (EQA) services for more than 40 years. Recently it has become clear that there is a need for new kinds of modern and even integrated EQA-services in particularly with the aim to empower medical laboratories for their future tasks, such as contribution to the development and implementation, as well as participation in global health-care policies. Methods: Labquality recently developed the Empower product line comprising the following 4 pillars: (i) master comparisons with panels of single donation sera (ii) virtual EQA-1 and (iii) virtual EQA-2 based on monitoring of patient percentiles and internal quality control (IQC) data across laboratories, and (iv) conceptual/statistical education to share a common vision on analytical quality. Results: Master comparisons give laboratories a calibration fixpoint and information on basic quality of their assays and own performance; patient percentile monitoring serves as a realtime quality indicator for their daily performance; patient- and IQC-monitoring establish evidence about mid- to long-term variation of the instrument-calibrator-reagent combination, backed-up by information from other laboratories of the peer group; in case of unacceptable lot-to-lot variation, laboratories have a basis for factorizing; the link patient/IQC data strengthens the quality management/assurance system of laboratories. Conclusions: Labqualitys new Empower product line adds on to the knowledge of the reasons for assay variation, strengthens the laboratories position in claims versus manufacturers and creates a tool for developing realistic quality goals and for strengthening the physician/laboratory interface by transparent communication on performance. It delivers data on the performance of assays from other manufacturers that may help laboratories in decisions on the acquisition of new instruments. Last but not least, the education by EQAorganizers on analytical quality, backed-up by evidence created from the empower project, will allow a common understanding between manufacturers and laboratories about realistic performance specifications and the needed quality management/assurance activities.
S431

281 UK NEQAS PILOT SCHEME FOR ANTI-MLLERIAN HORMONE EARLY EXPERIENCE N. Syme, R. Al-Sadie, C. Sturgeon UK NEQAS, Department of Laboratory Medicine, Royal Infirmary of Edinburgh, United Kingdom Background: Anti-Mllerian hormone (AMH) is increasingly used to assess ovarian reserve in women undergoing investigation for infertility and/or assisted conception. A Pilot United Kingdom National External Quality Assessment Service (UK NEQAS) scheme for AMH has been operating since July 2010. Participants receive monthly specimens to assess analytical performance. Occasional surveys of practice are also undertaken. There are currently 93 participants, a substantial proportion of whom are outside the UK. Methods: Participants receive five liquid serum specimens monthly and are requested to return their results to the EQA centre within three weeks. A personalised report summarising their results is returned with the next set of specimens a week later. In 2011 surveys were sent to 75 laboratories participating at the time; 70 % of these laboratories are located outside the UK. The survey focussed on clinical applications, reference intervals and interpretation. Results: Most participants (63%) report results in pmol/L but the remainder use mass units. Five methods are currently represented in the scheme. Between-laboratory agreement is poor, both overall (the coefficient of variation was 24% in 2011) and within-method, particularly at low concentrations. The absence of an International Standard probably contributes to the differences observed. Surveys of practice show wide variation in reference intervals quoted, even by users of the same method; relatively few centres use age-related reference intervals, despite good evidence that AMH concentrations change with age. Decision limits for assessing ovarian reserve also vary widely, so that the same AMH concentration might be interpreted differently depending on the centre in which it was obtained. For example decision levels used to assess risk of ovarian hyperstimulation syndrome ranged from >28.6 to >70.0 pmol/L. Most laboratories use reference intervals provided by the manufacturer, 31% have determined their own reference intervals and 24% use values from the literature. Conclusions: There is a need to improve between-method agreement of AMH assays and to achieve consensus about units for reporting, appropriate reference intervals and reliable decision levels for assessing ovarian reserve. T282 SUITABILITY OF THE MEASUREMENT IMPRECISION OF COMMON URINARY BIOCHEMICAL ANALYTES D. Szke, F. Braga, C. Valente, M. Panteghini Clinical Biochemistry Laboratory, Luigi Sacco University Hospital, Milan, Italy Background: Imprecision monitoring is central in the performance evaluation of an analytical system. To be suitable in clinical setting, the measurement imprecision should fulfil goals derived from biological variation data of the corresponding analyte, when available. Here we aimed to evaluate the long-term imprecision in our laboratory of measurements of 11 urinary analytes measured on the Roche Cobas 6000 system. Methods: Data were obtained for evaluated analytes by drawing up an internal quality control (QC) programme based on daily measurements of a lyophilized control material (Assayed Urine Control Level 3, Randox Laboratories, lot no. 580UC). CVs were calculated along the whole examined period (February 2012 to September 2012) (n=165) and compared with corresponding goals for imprecision derived from biological variability data (analyte concentration in 24 h urine sample), with the exception of urinary chloride and glucose for which biological variability data are missing. Results: Analyte concentration means and CVs (optimum goals in parentheses) were as follows: albumin, 182 mg/L, 3.0% (15.3%); calcium, 194 mg/L, 2.9% (6.9%); chloride, 258 mmol/L, 2.2% (not available); creatinine, 1.91 g/L, 2.1% (6.0%); glucose, 2.78 g/L, 1.9% (not available); magnesium, 323 mg/L, 3.5% (11.4%); phosphate, 852 mg/L, 2.0% (6.6%); potassium, 105 mmol/L, 2.1% (6.8%); sodium, 193 mmol/L, 1.9% (6.0%); urea, 14.7 g/L, 2.5% (5.7%); and urate, 214 mg/L, 2.7% (6.2%). Conclusions: Our study shows that in routine laboratory practice and over a clinically and analytically relevant timespan, the imprecision of the common urinary analyte measurements on the Cobas 6000 system fulfils optimum goals of analytical performance. Although intra-individual biological variability of urinary analytes is usually relatively high, resulting in less stringent analytical goals when compared to those of serum measurements, the excellent CVs of urinary analyte measurements qualify the suitability for clinical application of these tests.
S432

T283 REGULATORY SCIENCE FOR LABORATORIES IN GLOBAL AND ETHIOPIAN SITUATIONS A. Taklewold Laboratory management and Quality assurance in Addis Ababa University, College of Health sciences, School of Medical Laboratory Sciences. Objective: To explore the regulatory science and describe its application for medical Laboratories comparing Global and African orientation with Ethiopian harmoniztion. Method: Seminar Review was conducted on electronically available documents on Regulatory Science, Health and Laboratories regulations in the Global and Sub-Sahara African perspectives. Result: A total of about 480 documents were searched based on relevance of their topics. 60 documents including 22 web pages have been upraised for this seminar. Documents from regulatory review meetings reports, books, scientific reviews, policy documents and guidelines have been given priority during appraisal of the documents. The outcome of the study has been organized in to four categories: the regulatory science1, the FDA2 and CLIA882 experience, International Health Regulation3, harmonization in Sub-Sahara Africa3 and Experience of Ethiopia4. Discussion: Regulatory science is the science of developing new tools, standards, and protocols to assess the safety, efficacy, and quality of medical devices. The regulations of medical devices in developed countries are guided by national agencies and overseen for standardization and collaboration by ICH. Having committed to share responsibility in regulation and maintenance of global health, ICH is working to develop and standardize global prospects to ensure safety, effectiveness and quality of human use of medical devices. The World Health Assembly passed IHR-2005 enabled the WHO and member states to improve local and global capacity for detection, intervention, and response to diseases of international and public concern. Both the ICH and WHO are working in collaboration to enable countries to manage their population health and communicate their findings for global concern. Ethiopia has established Food, Medicine and Health care Administration and Control Agency by proclamation 661/2009 17 years after National Health Policy had been documented. Tools, standards and protocols to measure substantial equivalence of devices and practices is not yet developed. T284 ANALYSIS OF INTESTINAL PARASITES IDENTIFYING CLINICAL LABORATORIES PARTICIPATING IN A PROGRAM OF EXTERNAL QUALITY ASSESSMENT J.M. Vargas-Morales, L.E. Fragoso-Morales, S. SalazarGarca, M.T. Valle-Garca, M.N. Robledo-Aguilar, A. Rodrguez-Rodrguez, A.G. Colorado-Castillo Universidad Autnoma de San Luis Potos, Facultad de Ciencias Qumicas, San Luis Potos, S. L. P., Mxico. Background: External Quality Control (EQC) is a procedure in which an organization outside clinical laboratories for assessment methods of analysis, including testing coproparasitologic useful in the diagnosis and treatment of intestinal parasites infections, however, there are problems diagnosis, mainly of overdiagnosis and also the lack of experience of laboratory personnel in identifying parasites, which promotes the indiscriminate use of antiparasitic drugs. The objective is to evaluate the ability of clinical laboratories linked to a program of EQC in the correct identification of the parasites during the period of 2007-2010. Methods: We developed an analytical study of the results obtained by clinical laboratories linked to the EQC program of the Facultad de Ciencias Qumicas, Universidad Autnoma de San Luis Potos, Mxico (PEEC), in the period of 2007-2010, where on average per year involved 71 laboratories. The identification was assessed by sending photomicrographs of the following parasites: Blastocystis hominis, Giardia lamblia, Entamoeba histolytica/dispar, Entamoeba coli, Trichuris trichiura, Enterobius vermicularis and Ascaris lumbricoides. For statistical analysis were used Microsoft Office Excel and Epi InfoTM programs. Results: 867 results were analyzed, 579 of protozoa (479 successful) and 288 of helminths (287 successful). T. trichiura and A. lumbricoides scored 100% acceptable results (142 and 78) responsible for helminth infections more common in Mexico. E. vermicularis scored 98,5% correct results. For protozoa, E. coli was the most successful identified 92% (70), G. lamblia followed by 83% (123) and B. hominis with 81% (175) and E. histolytica/dispar with 79% (111). The ability to identify protozoa against helminths, was 59,9 times to correctly identify helminths (P <0,001). The identification analysis protozoa, is 2.5 times the ability to identify E. coli to B. hominis (P <0,05) and 4,4 times the ability to identify E. coli vs. E. histolytica/dispar (P <0,001). No differences between helminths. Conclusions: Clearly there are problems in identifying protozoa; laboratories with unsatisfactory results should implement training programs to improve identification skills.
S433

T285 APPLICATION OF GUIDELINES AS A REQUIREMENT FOR ACCREDITATION OF CLINICAL CHEMISTRY LABORATORIES: THE DUTCH EXPERIENCE. W. Verboeket-van de Venne(1), H. Vermeer(2), Y. Kluiters-de Hingh(3), M. Thelen(4), H. Kleinveld(5), W. Oosterhuis(6)
1 Atrium Medical Centre Heerlen; Netherlands Society for Clinical Chemistry and Laboratory Medicine (NVKC), Working Group on Guidelines 2 Albert Schweitzer Hospital Dordrecht; NVKC, Quality Committee 3 Sint Elisabeth Hospital Tilburg; NVKC, Working Group on Guidelines, Quality Committee 4 Amphia Hospital Breda; NVKC, Quality Committee 5 Atrium Medical Centre Heerlen 6 Atrium Medical Centre Heerlen; NVKC, Working Group on Guidelines, Quality Committee
T286 A TRACEABILITY CHAIN FOR THE STANDARDIZATION OF DIRECT BILIRUBIN TESTS F. Weber, R. Thiele, H.J. Kytzia Roche Diagnostics GmbH, Penzberg, Germany Background: Beyond the determination of Total Bilirubin, differential diagnosis of an icterus requires discrimination between the water-soluble post-hepatic Direct Bilirubin fraction (consisting of glucuronidated - and -bilirubin, covalently protein-bound -bilirubin and photo-bilirubin, a water-soluble cis-trans-isomer of -bilirubin) and the Indirect unconjugated -bilirubin. The Doumas reference method for the measurement of Total Bilirubin is in place, while for Direct Bilirubin neither a specific method nor a standard material is available yet. In the following we suggest a traceability chain for the standardization of Direct Bilirubin. Methods: A fast HPLC method (less than 15 minutes) was able to resolve the -, -, -, and - species and photo-bilirubin into clearly separated fractions and to quantify those using NIST reference material Unconjugated Bilirubin as a standard. The Doumas reference method for Total Bilirubin and the Roche methods BILTX and BILD2 were used for bilirubin determinations. Results: Comparison of the HPLC with the routine and reference methods results indicated that -, -, and -bilirubin were correctly recovered by the routine methods BILTX and BILD2. This was also true for photo-bilirubin, a very polar fraction that formed upon exposure of -bilirubin samples to blue light, and for ditaurobilirubin, a non-physiological conjugated bilirubin that gave a separate HPLC peak when dissolved in water, but was transformed rapidly and quantitatively into -bilirubin when added to serum. Therefore, human sera with low total bilirubin (5 M) were supplemented with different concentrations of ditaurobilirubin (up to 170 M) and used as physiological direct bilirubin standards. They gave an excellent correlation when measured with the Doumas Total Bilirubin reference method compared to the routine method BILD2, showing their suitability for the standardization of Direct Bilirubin routine tests. Conclusion: A traceability chain for the standardization of Direct Bilirubin assays has been established from the Doumas Total Bilirubin reference method via a commercially available bilirubin derivative to the Direct Bilirubin concentration in human sera.
Background: Quality and quality management are concepts within the clinical chemistry laboratory that always have received considerable attention. An important tool to demonstrate the quality of the products and services of the laboratory is accreditation. In the Netherlands, medical laboratories are accredited according to the CCKL-mark, based on the ISO 15189 standard. It is acknowledged in this standard that a country can have its own specific regulations or requirements. Within the Netherlands Society for Clinical Chemistry and Laboratory Medicine (NVKC) there is a growing need for guidelines with recommendations on laboratory procedures. Methods and results: According to a newly adopted procedure within the NVKC, three guidelines were recently developed with financial support from the Dutch Quality Fund for Medical Specialists (SKMS): Anemia protocol in primary care, Consultation by laboratory specialists and Point-of-care testing in primary care. All members of the NVKC could comment on the concept guidelines. Comments and suggestions for improvement were processed in a final version. After approval by the Quality Committee of the NVKC, the guidelines were put to vote during a plenary membership meeting and accepted. Each of the guidelines contains minimum standards as well as target standards. Minimum standards provide a lower limit of appropriate care that laboratories must comply to. In individual cases one can by exception deviate from this, if substantiated. Target standards provide optimal care. Ideally, these standards should be pursued by laboratory protocols. Conclusions: 1) The main goal for the development of NVKC guidelines was to harmonize processes between clinical chemistry laboratories, thereby improving quality. 2) The Dutch procedure of authorizing guidelines implies an active demand for input of the NVKC members and a transparent feedback regarding the processing of this input. A formal status of the guidelines was obtained by approval during a plenary membership meeting. 3) Application of the guidelines is greatly improved by subjecting them to audits. Specifically for the Netherlands, implementation of these professional guidelines is a requirement for accreditation of clinical chemistry laboratories.
S434

T287 PERFORMANCES OF THIRD PART QUALITY CONTROL IN A COAGULATION SYSTEM L. Zardo, L. De Valentin, R. Turrini, G. Piaserico Analysis Laboratory of Civil Hospital, Castelfranco Veneto (TV), Italy The guidelines for the running of the Internal Quality Assessment (IQA), conceived by the working group SIBIoC in 2008, report both the characteristics the materials must have and the formalities of execution and validation of IQA. Among the characteristics, internal control materials must have, the guidelines report they have better to be independent (of a third part) in case of validation of the analytical series, that is to say different in comparison to those either manufactured or supplied together with the reagents. The use of third part materials allows to highlight problems of degradation and/or shift of the calibrator that could be hidden if calibrators and control materials with the same matrix are used. The aim of this work was to evaluate the performances of third part control materials on a coagulation system. We tested two Biorad control levels and, for each one, we had 30 determinations for PT, APTT, Antitrombin (AT) and Fibrinogen (FGB). The Biorad controls were tested on ACL TOP 700 instrument using dedicated reagents (RecombiplstaTin, SynthAsil, Liquid AT, QFA). The coefficients of variation (CV%) relating to the values obtained from each test, were for the normal and pathological level as follows: PT=4.0% and 4.6%, APTT=3.5% and 3.0%, AT=3.0% and 4.0%, FBG= 6.0% and 6.0%. All the CV% resulted to be lower than the target of total error defined on the basis of biological variability (PT=5.3, aPTT=4.5, AT=8.3, FBG=13.6). They also resulted to be lower than the optimal acceptable limits proposed by the accredited EQA Program managed by the Centre of Biomedical Research of Padua and defined on the basis of the biological variability but modified on those of the state-of-the-art (PT=5.3, aPTT=6.7, AT=6.3, FBG=6.8). The results of this work demonstrate that the tested third part materials have characteristics of precision corresponding to the quality objectives defined both on the basis of the biological variability and of the state-of-the-art. Moreover, having the tests been performed up to 48 hours from their reconstitution, they show a good stability and it is obvious to think that processing them within 24 hours from reconstitution, as happens in a routine process, can further ameliorate the performance above all for PT and APTT. T288 CELIAC DISEASE DIAGNOSIS AND MONITORING: COMPARISON BETWEEN CHEMILUMINESCENCE AND ELISA ANTITRANSGLUTAMINASE AND ANTIDEAMIDATED GLIADIN PEPTIDES MEASUREMENTS A. Aita(1), D. Basso(1), E. Rossi(2), M. Pelloso(2), G. Guariso(3), C.F. Zambon(2), F. Navaglia(1), E. Greco(2), D. Bozzato(2), A. Padoan(2), P. Fogar(1), M. Plebani(2)
1 Department of Laboratory Medicine, University-Hospital of Padova, Italy 2 Department of Medicine - DIMED, University of Padova, Italy 3 Department of Womens and Childrens Health, University of Padova, Italy
Background: Celiac disease (CD) diagnosis is based on antitissue transglutaminase (tTG) and anti-deamidated gliadin peptides(AGA) determination. The majority of commercially available assays for tTG and AGA are ELISAs with sensitivity and specificity above 95%, never reaching however 100%. The aim of this study was to compare the analytical performancesand the clinical utility for CD diagnosis and monitoring of tTG and AGA of the IgA and IgG classes measured by new chemiluminescent(QUANTA Flash, Inova) and established ELISA (QUANTA Lite, Inova) methods. The one run measurementsof IgA and IgG tTG and AGA (htTG/DGP screen) by QUANTA Flash and QUANTA Lite were also evaluated. Methods: We studied 321 children (155 CD; 166 controls). tTG IgA and IgG, DGP IgA and IgG, h-tTG/DGP screen were measured by QUANTA Flash and QUANTA Lite. Results: Intra- and inter-assay coefficients of variations of QUANTA Flash were: 1.6% and 3.0% for tTG IgA, 3.7% and 4.0% for tTG IgG, 8.6% and 6.8% for DGP IgA, 8.6% and 3.2% for DGP IgG, 5.1% and 2.4% for h-tTG/DGP screen. The most sensitive (96.1%) and specific (97.0%) test for CD diagnosis was QUANTA Flash tTG IgA (cut-off:16 U), while the less sensitive (82.1%) and specific (78.6%) was QUANTA Lite tTG IgG (cut-off:4 U). We performed binary logistic regression analyses considering CD diagnosis as dependent, and including as predictors QUANTA Flash or QUANTA Lite tTG IgA and IgG, DGP IgA and IgG, and h-tTG/DGP screen. QUANTA Flash allowed to obtain a better correct overall classification of patients (96.60%) with respect to QUANTA Lite (95.14%). Among QUANTA Flash predictors, tTG IgA only was selected as significantly correlated with CD (r=0.263, P <0.0001, Exp(B)=1.0506, 95% CI=1,0286-1,0731). 18 CD children on GFD had a complete follow-up with measurements at 4, 12 and 24 months. QUANTA Flash tTG IgA was the most reliable index (Repeated measures analysis of variance, between subject effect F=5.03, P <0.05) in monitoring GFD. Conclusion: QUANTA Flash tTG IgA measurement is an extremely sensitive and specific index not only for CD diagnosis, but also for the monitoring of CD patient on GFD.
S435

T289 EVALUATION OF A PARTICLE-ENHANCED TURBIDIMETRIC CYSTATIN C ASSAY ON THE ROCHE COBAS 8000 ANALYZER AND ASSESSMENT OF CYSTATIN C-BASED EQUATIONS FOR ESTIMATION OF THE GLOMERULAR FILTRATION RATE H. Akbas, F. Davran, G. Yucel Department Of Biochemistry, Akdeniz University, Faculty Of Medicine, Antalya, Turkey Backgrounds: Estimation of the glomerular filtration rate (GFR) is essential for the evaluation of patients with chronic kidney disease (CKD). Cystatin C is a low molecular protein that freely filtered through the glomerulus and reabsorbed and catabolized by tubular cells. Recently, serum cystatin C-based equations were proposed as markers for estimation of GFR. The aim of this study was to evaluate the analytical performance of new cystatin C particle-enhanced turbidimetric assay (PETIA) on the COBAS 8000 analyzer and compared it particle-enhanced nephelometric assay (PENIA). Estimated GFR, which was generated from cystatin C-based equations, was compared by using two cystatin C assays. Methods: PETIA cystatin C measurements (n=185) were performed on the COBAS 8000 analyzer (Roche Diagnostics,Switzerland). PENIA cystatin C measurements (n=185) were done by using BN II nephelometer (Siemens NLatex Cystatin C, Germany). . Serum creatinine levels were analysed by rate-blanked and compensated Jaffe method in Cobas 8000 analyser. Estimated GFR (eGFR) for these cystatin C assays was calculated using the equations recommended by the manufacturers: Grubb et al. and Hoeks et al. Imprecision, limit of detection, quantification and linearity data were determined. Bland-Altman analysis was used to examine the differences between the assays. Results: The measurement range of cystatin C PETIA was between 0.4 mg/L and 8.0 mg/L. Within-run and between-run CVs were less than 5% at two quality control levels. PETIA cystatin C values were higher than PENIA results (mean of differences: -0.14, SD: 0.16, %95 confidence limits: -0.45-0.17). Both assays are linear with a good correlation coefficient (r=0.989). The BlandAltman analysis for the eGFR calculated with the two assays showed a systematic bias. Conclusion: PETIA cystatin C method is a reliable alternative to determinate of the cystatin C with good linearity, precision and accuracy on the Cobas 8000 analyzer. It provides a valid quantitative measurement of cystatin C with comparable % CVs in quality-control as well as patient samples. Additionally, PETIA provides a rapid analysis of a large number of samples with a low turnaround time and is well suited to detect renal dysfunction with a 24 h availability. T290 SIMULTANEOUS DETERMINATION OF SIROLIMUS AND EVEROLIMUS BY LC-MS/MS AUTOMATED PROCEDURE: A FOUR-YEAR EXPERIENCE IN KIDNEY TRANSPLANTATION RECIPIENTS H. Akbas(1), A. Dinckan(2), H. Kocak(2), G. Suleymanlar(2), G. Yucel(1)
1
Department Of Medical Biochemistry, Akdeniz University, Faculty Of Medicine, Antalya, Turkey 2 Tuncer Karpuzoglu Transplantation Center, Akdeniz University Hospital, Antalya, Turkey Background: Therapeutic drug monitoring (TDM) of immunosuppressive drugs as well as sirolimus and everolimus is frequently used in organ transplantation for dosage adjustment. High-performance liquid chromatography with tandem mass spectrometry (LC-MS/MS) is a reliable technique for TDM with many advantages. The main objective of this study was to evaluate on-line solid-phase extraction method in combination with LC-MS/MS for the simultaneous determination of sirolimus and everolimus in whole blood. Methods: The whole blood sirolimus and everolimus levels (n=7720) were determined in kidney transplantation recipients between December 2008 and November 2012. The current method uses protein precipitation for sample preparation. Analyses were performed using a triple quadrupole LC-MS/MS with a C18 column (Tandem Gold, Zivak Company, Turkey). Blood samples (100 L) were prepared by protein precipitation. Ascomycine was used as the internal standard (ZinMass Immunosuppressants LC-MS/MS Analysis Set (Whole Blood). Mass spectrometric analysis was performed by selective ion monitoring with an electrospray ionization in positive mode. Four-level blood calibrators and three internal quality control materials at different concentrations were used for assay. External quality control assessments were done by using the international proficiency testing scheme (www.bioanalytics. co.uk, England). Results: The analytical time was 4 min and the assay was linear from 1.5 to 50 g/L for everolimus and sirolimus. Calibration curves were linear between the ranges. Within-run and between-run CVs were less than 10% for three quality control levels. The limits of detection and quantification were determined (sirolimus: 0.66 g/L, 1.98 g/L ; everolimus: 0.48 g/L, 1.60 g/L, respectively). The external quality control results showed acceptable agreement with other methods. Conclusions: This method allows for the simultaneous determination of sirolimus and everolimus in whole blood in a short time with high linearity, precision and accuracy. LCMS/MS method provides a reliable and rapid automated procedure that can be preferable for therapeutic drug monitoring of immunosuppressive drugs in clinical laboratories.
S436

T291 RAPID AND SIMPLE SANDWICH IMMUNOASSAY FOR DETECTION OF TOTAL CYANOBACTERIAL HEPATOTOXINS (MICROCYSTINS AND NODULARINS) S. Akter, M. Vehniinen, U. Lamminmki Division of Biotechnology, Department of Biochemistry and Food Chemistry, University of Turku, Finland Background: Cyanobacterial blooms cause local and global problems with their potent toxins and pose serious health risks for human and animals. Microcystins (MCs) and nodularins (Nods) are structurally related cyclic peptide cyanobacterial toxins having more than 100 variants. Most MCs and Nods are potent hepatotoxins, tumor promoters and possible carcinogens. The WHO guideline value for cyanobacterial toxin MC-LR in drinking water is 1 g/L. Commercially available immunoassays are based on indirect competitive method and require hours to perform. Simple and rapid, yet sensitive methods for first line screening of total MCs and Nods are in high demand for assessment of water quality and safety. Methods: Antibody fragment (scFv) capable of recognizing immunocomplexes (IC) consisting of a capture antibody (Ab) bound to MCs or Nods were selected from a synthetic phage displayed Ab library. The scFv was expressed in E. coli as fusion with Alkaline Phosphatase (AP), purified and used in sandwich assay for detection of total MCs and Nods. In a onestep assay reagents and toxin standards in water were added together, incubated for 10 min followed by one washing step and signal development. Signal of Eu labelled anti-AP Ab was detected by time-resolved immunofluorometry. Total assay time was 15 min. Results: It is very rare and difficult to establish generic sandwich format assay for small molecules (~1000 dal) where both capture and secondary Ab must have generic binding capabilities. Using synthetic phage display library, we have successfully isolated high affinity anti IC scFv with generic specificities towards MCs and Nods and developed a 15 min rapid and simple single-step sandwich format assay. The sensitivity (blank+2SD) of the assay for major hepatotoxins (MC-LR, -dmLR, -RR, -dmRR, -YR, -LY, -LF -LW, and Nod-R) varies from 0.02 to 0.1 g/L (4-20 pg/well). Conclusions: The assay was capable of detecting all the tested nine major hepatotoxins at levels ten times below the WHO guideline limit (1 g/L). The assay is very rapid and simple having only one washing step. It offers robustness for costeffective automation and high throughput possibilities for fast screening of large number of samples for the detection of total MCs and Nods. T292 NEW INSTRUMENT FOR EASY DETERMINATION OF RHEOLOGICAL PARAMETERS OF ERYTHROCYTES B. Albea(1), A. Marenzana(1), H. Castellini(3), B. Riquelme(2) Grupo ptica Aplicada a la Biologa, Instituto de Fsica Rosario (CONICET-UNR). Rosario, Argentina 2 rea Fsica, Facultad de Cs Bioqumicas y Farmacuticas (UNR). Rosario, Argentina 3 Dpto. de Fsica. Facultad de Cs. Exactas, Ingeniera y Agrimensura. Universidad Nacional de Rosario, Argentina The evaluation of the rheological parameters of human red blood cells allows diagnose possible alterations that can induce serious complications in microcirculatory disorders and diseases such as diabetes, hypertension, anemia, etc. The red blood cell (RBC) must be able to deform without breaking or obstruct microcapillaries but there are certain diseases in which the membrane is altered and cannot normally be distort or break. In this context, we developed a new computerized rheometer to determine the stationary and dynamic rheological parameters of RBCs for economical and easy use in conventional and specialized biochemical laboratories, which is in patenting step. This new instrument is based on laser diffractometric technique and has a GUI (Graphic User Interface) for easy use to an unskilled operator. As in the Erythrodeformeter developed previously by Rasia et al, this instrument permits to evaluate the viscoelasticity of RBC by applying the laser diffractometry technique. In order to carry out these measures, a thin layer of RBC is suspended in a viscose media and placed between two parallel concentric disks. In the stationary mode, the lower disk rotates at constant speed and in the dynamic mode, it moves at sinusoidal oscillating speed at frequencies in physiological range. Photometric signals from diffraction pattern are used to determine all RBC rheological parameters, which are averaged over several millions of cells. The parameters obtained with the new rheometer correlate with the erythrocyte rheological parameters obtained with other similar instruments but they are more complete and accurately. The new erythrocyte rheometer can determine all the rheological parameters of erythrocytes in a simple and easy form to help in the diagnosis of vascular and haematological diseases and allow analyze their evolution in terms of treatment and medication that the patient receives. Also, the easy determination of the dynamic viscoelasticity of RBCs permit to know the part of the membrane that is altered in order to design a medication according to the disease.
1
S437

T293 COMPARISON BETWEEN ANALYTIC RESULTS OBTAINED BY TWO VERY COMMON HPLC SYSTEMS IN HAEMOGLOBIN DIAGNOSTICS: EVALUATION ISSUE AT CSMR A. Amato, D. Ponzini, P. Di Biagio, D. Zei, D. Gianni, L. Masi, R. Piscitelli ANMI Onlus - Centro Studi Microcitemie Roma, Rome, Italy Background: The CSMR (Centro Studi Microcitemie di Roma) deals from more than 50 years with prevention and diagnosis of Thalassaemia syndromes and haemoglobinopathies. The determination of the haemoglobin fractions HbA2 and HbF, very important parameters to reach diagnostic results, is currently performed using the cation exchange HPLC method. In this study we compared the results obtained for the haemoglobin fractions HbA2 and HbF by the Variant II system (BIO-RAD Laboratories, Hercules, CA, USA) and the G8 system (TOSOH HLC-723 G8). Furthermore, when possible, we compared the analytic results of abnormal haemoglobin fractions according to identifying and quantitative capacity. Methods: 156 samples coming from our ambulatory were studied. These samples were analyzed as routine, by the HPLC Variant II (BIO-RAD Laboratories, Hercules, CA, USA) and successively, they were examined by the G8 system (TOSOH HLC-723 G8). The analyzed samples were divided into three subgroups: a. Hb A2 <2.0% (n=15) b. 2.1%< Hb A2 <3.2% (n=111) c. Hb A2> 3.3%. (n=30). In all subgroups the median calculation was carried out for the HbA2 and HbF. We used the Pearson correlation coefficient to compare the HbA2 and HbF obtained values. Some samples already selected by the Variant II for the presence of rare haemoglobin variants, were tested by the G8 as well. Results: The analysis of the results obtained by the two methods pointed out a correlation coefficient of 0.99 for both HbA2 and HbF. This result indicates a very strong correlation. As for the HPLC separation of the examined haemoglobin variants, we obtained excellent results by the G8 system even if the retention times are different respect with Variant IIS. We highlighted the following haemoglobin variants: Hb G Copenhagen, Hb O Arab, Hb San Diego, Hb Koln, Hb Toulon and one more Delta variant still under identification. Conclusions: Both systems provided equivalent results that can be interpreted in the same way in diagnostic terms. The excellent overlap obtained by two different systems, leads us to consider also the G8 HPLC as a reliable support for first diagnosis level in thalassaemia and haemoglobinopathies. T294 STUDY OF REPRODUCIBILITY OF A NEW CAPILLARY ELECTROPHORESIS INSTRUMENT (MINICAP FLEX PIERCING) AT CENTRO STUDI MICROCITEMIE OF ROME A. Amato, D. Zei, P. Di Biagio, D. Gianni, E. Melis, D. Ponzini, O. Sarra ANMI Onlus - Centro Studi Microcitemie Roma, Rome, Italy Background: Centro Studi Microcitemie of Rome has been working for years in prevention and diagnosis of hemoglobinopathies. For the detection of these pathologies the identification and the precision in the quantification of the hemoglobin fractions HbA2 and HbF are crucial. Aim of the study: In this study we tested a new Capillary electrophoresis instrument, the MINICAP Flex Piercing (SEBIA, Parc Tecnologique Leonard de Vinci, Evry Cedex, France) to evaluate the precision performance of quantitative measurements. For this purpose we performed intra and inter assay study on HbA, HbA2 and HbF. Assessment was also extended to hemoglobin variant HbS as it is widely present in our laboratory. Materials and methods: For intra- and inter-assay reproducibility, 5 samples were selected as follow: normal whole blood from patient of our ambulatory; whole blood with elevated HbA2 from patient of our ambulatory; heterozygous A/S from patient of our ambulatory; SEBIA Pathological HbA2 Control; SEBIA Normal HbA2 Control. All assay were performed on samples stored at -80 C. Reproducibility withinrun: each sample was run 5 times on both capillaries, using the Minicap Hemoglobin procedure. The same protocol was then repeated the same day using a different reagent lot number. The mean, Standard Deviation (SD) and Coefficient of Variation (CV; n=20) were calculated for each sample and each hemoglobin component. Reproducibility between-runs: the 5 samples were run within the same series for 10 consecutive days, using Minicap Hemoglobin procedure. The same protocol was repeated the same day using a different reagent lot number. The mean, SD and CV (n=20) were calculated for each sample and each hemoglobin component. Results: In the within run study we have evaluated a good reproducibility for HbA2 (1.26<CV%<1.70), HbS (CV= 0.49) and HbA0 (0.05<CV<0.31). As expected, since there is evidence in literature, the CV of HbF is major (CV=5.66). The assay between runs shows a good reproducibility too. Infact for HbA2 we have calculated 1.12<CV%<1.77, for HbS the CV is 0.75 and HbA0 shows a CV from 0.05 to 0.47. As expected the CV of HbF is major (CV=8.1). Conclusions: The results show a good performance of the Minicap flex pearcing about its precision in the reproducibility of the results.
S438

T295 ANALYTICAL PERFORMANCE OF COBAS C311 IN THE PEDIATRIC LABORATORY M. Aralica, J. Matica Rijeka Clinical Hospital Center, Clinical Institute of Laboratory Diagnostics, Rijeka, Croatia Background: The Cobas c311 analyzer (Roche, Mainheim; Germany) is routine clinical chemistry analyzer suitable for laboratory with workloads of 50-200 samples per day. The aim of our study was to evaluate analytical performance of Cobas c311 in the pediatric laboratory. Methods: The validation was performed for 30 analytes: glucose, urea, creatinine, uric acid, total bilirubin, direct bilirubin, ALP, AST, ALT, GGT, LDH, CK, AMY, total protein, albumin, CRP, IgA, IgG, IgM, ferritin, triglyceride, cholesterol, HDL, iron, calcium, magnesium, inorganic phosphorous, sodium, chloride and potassium. The imprecision study consisted of within-run imprecision (N=15) and between-run imprecision (N=15). Method comparison study was done with routine analyzer AU400 (Beckman Coulter, USA; N=20). Results were judged according to quality specification given in Biological variation database by Ricos and colleagues. Results: For all tested analytes coefficients of variation (CV) were below 5% for within-run imprecision. Of 30 tested analytes, 26 of them fulfilled quality specification for total imprecision except: creatinine, sodium, chloride and ALP. The Passing and Bablok regression analysis was used for method comparison study. Linear equations, as well as 95% confidence intervals for intercept and slope were calculated. The proportional error was found for: total bilirubin, direct bilirubin, CK, AST, LDH, CRP and magnesium. The constant error was revealed for: GGT, IgG and chloride. Both constant and proportional errors were detected for urea and IgM. Other tested analytes were in agreement with routine method. Conclusions: Overall analytical performance of Cobas c311 was satisfactory. Imprecision study for majority of tested analytes was acceptable. However, method comparison study showed a need for regression line modifications in almost half of tested methods. Both imprecision and inaccuracy requirements were not satisfied only in case of chloride measurement. Despite of imperfect analytical performance Cobas c311 analyzer can be implemented in routine pediatric laboratory after certain adjustments. T296 THE CONVENIENT PURIFICATION OF BIOCONJUGATED UPCONVERTING NANOPARTICLES BY HIGH GRADIENT MAGNETIC SEPARATION R. Arppe, O. Salovaara, L. Mattsson, S. Lahtinen, T. Soukka Division of Biotechnology, University of Turku, Finland Background: Upconverting nanoparticles (UCNPs) have the unique luminescent property of converting low-energy infrared light into visible emission which can be utilized in reporter and imaging technologies. However, as the particle size of the UCNPs approaches the size of biomolecules, the handling of reporters becomes cumbersome with traditional purification methods such as centrifugation. While the lanthanide-doped crystal structure of UCNPs is responsible for the upconversion luminescence, the lanthanide dopants also bring paramagnetic properties to the UCNPs. This enables the use of high gradient magnetic separation (HGMS) as a method to purify and separate them and their conjugates. HGMS produces high magnetic gradients which can capture even weakly paramagnetic materials. Methods: UCNPs were bioconjugated with an excess amount of streptavidin or monoclonal antibodies and thereafter purified from unbound biomolecules by HGMS or by conventional method (ultrafiltration). The purification efficiency of the two methods was compared by analyzing e.g. the yield and by using the bioconjugated UCNPs as reporters in a heterogeneous bioaffinity assay. Results: The HGMS-purification of the bioconjugated UCNPs resulted in higher yield compared to ultrafiltration, as it is difficult to detach the UCNPs from the filter membrane and redisperse them into solution. Ultrafiltration also tends to aggregate the UCNPs which was observed as higher non-specific binding and standard deviation when they were used as reporters in a heterogeneous assay. A more sensitive assay was performed by using the HGMS-purified UCNPs. Conclusions: HGMS is a fast, convenient and highly selective purification method for UCNPs and it can be used to separate bioconjugated UCNPs from unbound biomolecules or other reagents used for their surface modification process producing highly sensitive reporters with high yield and purity. It can also be used for buffer exchange and for concentrating the UCNPs which are difficult with other purification methods. The separation is solely based on the intrinsic paramagnetism of luminescent lanthanide dopants without the need to embed separate, optically inactive magnetic materials within the UCNPs.
S439

T297 HIGH-THROUGHPUT ANALYSIS OF AN IMMUNOSUPPRESSANT DRUG PANEL IN WHOLE BLOOD USING ULTRAFAST SPE/MS/MS K. Schlicht(3), P. Wallemacq(2), M. Barbieri(1), V. Miller(3), W. LaMarr(3)
1 2
T298 QUANTITATIVE ANALYSIS OF FREE AND TOTAL THYROID HORMONES IN SERUM USING LC/MS/MS R. Doyle, K. McCann, M. Barbieri Agilent Technologies, Inc. Liquid chromatography triple quadrupole mass spectrometry (LC/MS/MS) has become an essential tool for small molecule quantitation due to its high sensitivity and specificity, excellent reproducibility and the ability to perform simultaneous analysis of multiple analytes. Thyroid hormones can be challenging compounds to analyze due to the low levels in biological matrices such as plasma relevant to clinical research. In order to address this challenge, a sensitive liquid chromatographytandem mass spectrometry method for the simultaneous analysis of Thyroxine (T4), 3,3,5-Triiodothyronine (T3) and 3,3,5-Triiodothyronine (rT3) in serum samples has been developed. Two separate sample preparation techniques were used for the determination of the thyroid hormones protein precipitation for the total concentration and ultracentrifugation for the free concentration. A labeled internal standard was included for each of the analytes to ensure accurate quantitation. Quantitative analysis was performed using multiple reaction monitoring (MRM) transition pairs for each analyte and internal standard. Standard liquid chromatography (LC) was used with a reverse-phase C18 analytical column. Final concentrations were calculated by comparing the response of the analyte to a known concentration of internal standard and plotting the result on a calibration curve developed using stripped human serum spiked with standards. The LC-MS/MS parameters were optimized and calibrated over the range of 1 pg/mL to 1000 pg/mL for free T4, T3 and rT3 and 1 pg/mL to 1000 pg/mL for total T3 and rT3 and 1 ng/mL to 1000 ng/mL for total T4 hormone concentrations. The calibration curves show excellent linearity and reproducibility across the entire range of analysis. Accuracy of the methodology was verified using NIST Standard Reference Material (SRM 971 Hormones). A sensitive and specific LC/MS method has been developed the simultaneous analysis of Thyroxine (T4), 3,3,5-Triiodothyronine (T3) and 3,3,5Triiodothyronine (rT3) in serum. A simple filtration sample preparation for free thyroid hormones and a liquid-liquid extraction sample preparation for total thyroid hormones allows for determination down to low pg/ml levels.
Agilent Technologies Inc., Cheadle, UK Cliniques Universitaires St-Luc, UCL, Brussels, Belgium 3 Agilent Technologies Inc., Wakefield, MA, USA In many clinical research laboratories, liquid chromatographymass spectrometry (LC/MS) methods of analysis of immunosuppressant drugs have proven superior because of their increased sensitivity and selectivity. We evaluated the ability of an ultrafast SPE/MS/MS system to simultaneously analyze tacrolimus, everolimus, sirolimus, and cyclosporin A in whole blood. MS methods for tacrolimus, everolimus, sirolimus, and cyclosporin A and their corresponding internal standards were optimized for analysis by QQQ MS. Calibration standards for each analyte were prepared in bovine whole blood. The whole blood samples were mixed with water and precipitated using a zinc sulfate and methanol solution containing the internal standards. Precipitated samples were gently mixed and then centrifuged. Following centrifugation, supernatants were transferred to a 96-well plate for analysis. Samples were analyzed using an Agilent RapidFire high-throughput mass spectrometry system coupled to a QQQ mass spectrometer. A C18 column was used for online SPE. Prepared calibration standards were run in triplicate over a series of days to establish both intra- and inter-day precision and accuracy. Cyclosporin A had both intra- and inter-day accuracies within 15% and CV values less than 6% for all concentrations within the linear range (7.8-1000 ng/mL). The method for all four analytes had excellent linearity within their respective measured ranges with an R2 value greater than 0.995. Signalto-noise ratios were calculated by looking at peak to peak height and found to be greater than 40:1 at the limit of quantitation for all four analytes. To further evaluate this method, identical human samples were analyzed by RapidFire and a traditional LC/MS/MS method. Excellent correlation was found for the two methods. Based on these results: tacrolimus, everolimus, sirolimus, and cyclosporin A can be accurately and precisely measured in whole blood. All four immunosuppressant drugs were simultaneously analyzed in a 12 MRM panel in less than 13 seconds per sample using ultrafast SPE/MS/MS. While the analytical results were comparable to LC/MS/MS, the analysis time was approximately 10 times faster. This methodology is capable of throughputs >270 samples per hour.
S440

T299 POLYADENYLATED SEQUENCING PRIMERS ENABLE COMPLETE READABILITY OF SHORT PCR AMPLICONS ANALYZED BY DIDEOXYNUCLEOTIDE SEQUENCING M. Beranek(1), M. Drastikova(1), J. Petera(2), F. Gabalec(3)
1
T300 CELLULARITY CONTROL AND MICROBIOLOGICAL SAMPLE SUITABILITY: KEY-ROLE OF REAL TIME-PCR IN SEXUALLY TRANSMITTED DISEASES REPORTING L. Bianchi, Z. Napoli, S. Donati, C. Sebastiani, S. Fabbri, M. Niccolai, R. Lari U.O. Analysis Laboratory ASL 3 Pistoia, Italy Background. The use of Real-Time PCR (RT-PCR) to diagnose sexually transmitted diseases (STD) such as C. trachomatis (CT), Human Papillomavirus (HPV), U. urealyticum (UU), U. parvum (UP), M. hominis (MH) and M. genitalium (MG) infections, needs an adequate standardization during pre/postanalytical phases. Aim of this study was to standardize: 1) pre-analytical phase, with a correct DNA sample extraction modality from biological matrix like urine (UR), seminal fluids (SF), urethral swabs (US) and cervical swabs (CS) using internal control (IC) and endogenous control (EC); 2) postanalytical phase, using cellularity cut-off and range cycle threshold (Ct) to define bacterial/viral load. Methods. DNA from 933 samples (135 UR, 110 SF, 135 US and 553 CS) was extracted with EZ1 DNA Tissue or Virus kit (Qiagen) and amplified by RT-PCR to detect CT (artus, Qiagen), MH, MG, UP and UU (Nuclear Laser); HPV detection and genotyping was performed by end-point PCR to amplify L1 gene (Innolipa HPV Genotyping Extra, Innogenetics) and by microarray technology targeting E1 gene (INFINITI HPV Genotyping, Autogenomics); sample cellularity was calculated by quantitative RT-PCR amplifying human HPRT1 gene (Cell Control, Argene). Results. CT, MH, MG, UU, UP and HPV positivity respectively was: 8.6%, 8.8%, 2.25%, 9.1%, 27.8% (8% with Ct>30) and 17.33%. Average cellularity/PCR in UR, US and CS was respectively: 1,661384, 5,6291,174 and 8,068990. Samples with less than 300 cells/PCR for UR and 750 cells/PCR for US and CS are classified unsuitable. L1 High Risk (HR)-HPV positivity was 12,1% vs 15,9% of E1 positivity. Conclusions. Reported data demonstrate: 1) using IC and EC enables pre-analytical optimization with an analytical quality increase; 2) cellularity/PCR allows to evaluate sample suitability while a range of Ct is useful to evaluate bacterial/viral load; 3) results interpretation is fundamental to diagnose HRHPV (with integrated DNA) and to give an appropriate clinical significance to low viral loads; 4) more data are required to standardize sample cellularity control.
Institute of Clinical Biochemistry and Diagnostics, Faculty of Medicine and University Hospital Hradec Kralove, Czech Republic 2 Department of Clinical Oncology, Faculty of Medicine and University Hospital Hradec Kralove, Czech Republic 3 4th Department of Internal Medicine, Faculty of Medicine and University Hospital Hradec Kralove, Czech Republic Background: Dideoxynucleotide DNA sequencing is one of the principal procedures in molecular biology. Loss of an initial part of nucleotides behind the 3 end of the sequencing primer limits the readability of sequenced amplicons. We present a method which extends the readability by using sequencing primers modified by polyadenylated tails attached to their 5 ends. Methods: In the study, seventeen samples were tested. Performing PCR, we amplified eight amplicons of six human genes (AMELX, APOE, HFE, MBL2, SERPINA1 and TGFB1) ranging from 106 bp to 680 bp. For sequencing we used BigDye Terminator v3.1 Cycle Sequencing Kit, Applied Biosystems. Standard or polyadenylated sequencing primers for each gene were used in parallel. The purified extension products (BigDye XTerminator Purification Kit, Applied Biosystems) were separated using an ABI 3130 Genetic Analyzer with POP7 polymer. Results: When standard primers were present in the sequencing mixture, the loss of nucleotides fluctuated between 20 and 53 according to the amplicon. Polyadenylation of the sequencing primers minimized the loss of bases in all amplicons. Complete sequences of shorter products (AMELX 106 bp, SERPINA1 121 bp, HFE 208 bp, APOE 244 bp, MBL2 317 bp) were obtained. Also in the case of TGFB1 products (366 bp, 432 bp, and 680 bp, respectively), the lengths of sequencing readings were significantly longer if adenylated primers were used. Conclusions: Single strand dideoxynucleotide sequencing with adenylated primers is a universal, easier, cheaper, and less time-consuming way to achieve complete or near complete readability of short PCR amplicons when compared to the standard double strand sequencing. The study was supported by the research projects NT113344/2010 and NT11344-4/2010 from the Ministry of Health, Czech Republic.
S441

T301 VIRTUAL CROSS-MATCH ON SINGLE ANTIGEN LUMINEX ASSAY IS PREDICTIVE OF COMPLEMENTDEPENDENT CYTOTOXICITY CROSSMATCH IN KIDNEY TRANSPLANTATION. M. Francone, G. Romeo, R.E. Saladino, S. Autolitano, F. Turiano, N. Imbesi, A. Caccamo Centro Regionale di Tipizzazione Tissutale - Azienda Ospedaliera Bianchi Melacrino Morelli, Reggio Calabria, Italy Background: Detection of antiHLA donor-specific antibodies (DSA) was performed using CDC (complement-dependent cytotoxicity) up to the introduction of solid phase based assays. ELISA and, more recently, Luminex technology based on polystyrene microspheres coated by HLA molecules, has highly increased DSA detection sensitivity. Virtual crossmatch (VXM), based on Single Antigen (SA) Luminex assay, is used to compare HLA donor specific antigens and recipient antiHLA antibodies specificities. Therefore it may significantly decrease time of pre-transplant immunogenetic tests. In addition the results are more reliable than CDCXM (complementdependent cytotoxicity crossmatch). Aim of this study is to verify whether VXM result is predictive of CDCXM outcome. Material and Methods: 403 CDCXM combinations performed between 2009 February and 2012 March (184 sera samples from patients in renal transplant waiting list tested against lymphocytes from 43 deceased donors) were analyzed to assess whether VXM is predictive of CDCXM outcome.VXM was carried out using antiHLA specificities assigned performing a Luminex SA assay. Results: 275 sera samples were negative at Luminex screening for research of HLA antibodies, 256 were CDCXM negative, 19 were positive; 15 of 19 positive sera had negative CDC reaction with Dithiothreitol addition, bearing out hypothesis that CDCXM positivity is due to antiHLA IgM or autoantibodies presence. Both CDCXM and VXM were performed on 128 positive sera samples, 81 with concordant XM results(63,3%); 47 positive-positive: in 35 (74%) SA assay showed at least one DSA with MFI (Mean Fluorescence Intensity) value 5000, supporting that the higher is MFI Ab value, the more predictive is VXM on CDCXM outcome. 47 samples (36,7%) XM contrasting: 33 CDCXM- /VXM+ is probably due to either non fixing complement anti HLA Ab of clinical relevance, or Luminex higher sensitivity than CDCXM; 14 CDCXM+/VXM-, is maybe due to non HLA Ab : further analysis should therefore be carried on to consider eligible donor an otherwise excluded one. Conclusions: VXM, especially with high DSA MFI, is predictive of CDCXM outcome. T302 VITAMIN D LC-MS/MS: OPTIMIZATION OF CHROMATOGRAPHIC PARAMETERS TO INCREASED 60% OF COMMERCIAL KIT THROUGHPUT M. Cant, F. Keller Ente Ospedaliero Cantonale, Ospedale San Giovanni, Dipartimento di Laboratorio Background: In the last 5 years many companies developed In Vitro Diagnostic (IVD) Kits for the quantitation of Vitamin D3 and D2 (Vit-D) for Liquid Chromatography coupled to Tandem Mass Spectrometry (LC-MS/MS) instrumentation. This analysis is particular difficult for the presence of several metabolic Isomers and contaminants. The IVD kit ensures to the clinical laboratory a correct quantification validating the peaks separation between the Vit-D and their contaminants. All Vit-D commercial kits need 5 minute analysis time with a throughput of 96 samples/8 hours (h). In the present work we evaluate an implementation of IVD commercial Kit to obtain an increased throughput of up to 160 samples per days. Methods: The LC-MS/MS are composed by 1290 chromatographic system and 6460 triple quadrupole (Agilent Technologies). All the analysis was performed with the Chromsystems MassChrom Vitamin D3/D2 kit. In the present work we: 1-increase the flow rate of trapping/Washing samples 2- increase the flow rate of washing analytical column; 3overlap the analysis.. To guarantee the chromatographic validation we dont change the elution and, the reduction in time of modified steps, are compensated by the increasing of flow. This compensation permit to maintain the same washing mobile phases volume without change in Vit-D peaks width, separation between isomers/contaminants, etc. Results: The evaluation of chromatographic parameters was performed on deuterated Vitamin D3 (internal Standard - ISTD). The base width of ISTD is 0.20 min and the separation between ISTD and its contaminant peak is 0.20 min in both methods. Our experience highlight a strong stability of the Chromsystems modifies method. The calibration are stable up to 9 months with a CV% 100%20% for both high and low level internal Quality Controls. Conclusions: The Chromsystems MassChrom Vitamin D3/D2 kit was optimized to obtain a 3 minutes method without any change in chromatographic peaks separation without loose the original kit validation. The LC-MS/MS modified Chromsystems assay is a very robust method with up to 9 months stable calibration. The increment of 60% in throughput of this new method allow a better use of LC-MS/MS instrumentation in clinical laboratory hospital
S442

T303 EMIT ASSAY INCREASE RISK OF REJECTION IN PATIENTS TREATED WITH MICOPHENOLIC ACID? M. Cant, R. Della Bruna, F. Keller Ente Ospedaliero Cantonale, Ospedale San Giovanni, Dipartimento di Laboratorio Background: Mycophenolic Acid (MPA) is the third most widely immunosuppressant in solid organ transplantation. Several studies highlight the necessity of an accurate and sensitive drug monitoring. The assay to monitor MPA are Tandem Mass spectrometry coupled with Liquid Chromatography (LCMS/MS), High Performance Liquid Chromatography (HPLC) and Immunoassay like Enzyme Multiple Immunoassay Technology (EMIT). Several studies show an EMIT overestimation of MPA between 14% to 35%. These value are calculated as media of the Altman-Bland test of all samples. In the present study we want to evaluate the EMIT overestimation at different MPA concentration ranges and highlight the bias between the therapeutic range (1-3.5 mg/L), the extratherapeutic range (>3.5 mg/L) and low level therapeutic range (1-2 mg/L). Methods In this study we evaluated 43 serum samples from patients who had different organ transplantation. The serum was analysed with MassTox Series A MPA (Chromsystems GmbH) for MPA. The same samples were analysed with Syva EMIT kit (Siemens Healthcare Diagnostic GmbH). All statistical analysis were performed with Analyse-it v2.20; the data comparison was calculated with Deaming Fit, the bias was calculated with Altman-Bland Test. Results LCMS/MS show an interassay CV=3.94% and an inaccuracy of 1.09% at 1.94 mg/L. The comparison with EMIT was LCMS/MS=0.85*EMIT+0.32 with a positive bias for EMIT of 33.5%. If we divide the samples by EMIT concentration range, the EMIT positive bias are: 1-2 mg/L: 46.4% SD: 11.7; 2-3 mg/L: 33.6% SD: 16.8; 3-4 mg/L: 20.2% SD: 15.4; >4 mg/L: 21.4% SD: 7.0. The EMIT bias in the therapeutic range (1-3.5 mg/L), is 35.2% (SD: 16.9). The EMIT bias for samples with higher concentration (>3.5 mg/L) is 19.8% (SD: 7.8). Conclusions The overall correlation between EMIT and LCMS/MS show an overall EMIT overestimation of 33.5% coherent with data published in literature; but, at low concentration range 1-2 mg/L show a bias of 46.4%. The adeguate MPA minimum level, with cyclosporin coadminstration, is fixed at 1.3 mg/L for Kidney and 1.4 mg/L for Hearth transplantation (HPLC assay). In both cases an EMIT concentration of 1.5 mg/L overestimate the real concentration of 1.0 mg/L (HPLC or LC-MS assay) with increased risk of rejection T304 INTER-BATCH VARIATION OF THE BINDING SITE FREELITE LIGHT CHAIN ASSAYS D.J. Matters, L.J. Woollatt, H.D. Carr-Smith The Binding Site Group Ltd Introduction: Analysis of serum free light chains utilising the Freelite assay is commonplace for identifying and monitoring patients with B cell disorders. This study compares inter-batch variation for 3 batches of the Freelite assay on 8 different platforms within a single laboratory (Beckman Coulter AU400, Siemens Advia 1650, Roche C6000, Roche Modular P, Beckman Coulter Immage, Roche Integra 400, The Binding Site SPAPLUS and the Siemens BNII). Methods: 3 batches of reagent for each platform were used to test two sets of samples; 1) >30 serum samples from individuals with both normal and elevated light chain levels; 2) >45 serum samples from healthy adult individuals. Sample set 1 was compared between batches, on the same platform using results from a single batch as the predicate values and correlating the remaining two batches results. Sample set 2 was compared to the manufacturers 95th percentile reference normal ranges of 3.3-19.4 mg/L for kappa light chains and 5.7126.3 mg/L for lambda light chains. Analysis was performed on SPSS and Analyse-it. Results: For sample set 1, Passing and Bablok regression slopes for kappa were 0.97x-0.14 / 0.97x-0.08 (AU400), 1.11x0.81 / 1.11x-0.04 (Advia), 1.01x-0.62/1.10x-1.85 (C6000), 1.11x+0.06 / 0.98x+0.84 (Modular P), 1.01x-0.03 / 0.94x+0.23 (Immage), 1.00x-0.38 / 1.00x-0.15 (Integra), 0.91x+0.3 / 0.89x+0.16 (SPAPLUS), 1.01x-0.49 and 0.96x-0.19 (BNII). Passing and Bablok slopes for lambda were 1.00x+0.16 / 1.14x+0 (AU400), 1.00x-0.16 / 0.94x-0.19 (Advia), 1.08x-0.53 / 1.14x-1.84 (C6000), 0.87x+0.26 / 0.94x-0.02 (Modular P), 1.15x-4.44 / 1.05x-2.27 (Immage), 0.93x+1.59 / 0.95x+0.93 (Integra), 0.99x-0.19 / 0.93x+0.03 (SPAPLUS), 0.96x-0.66 and 0.97x-1.26 (BNII). For sample set 2 the median values and distributions were not significantly different (kappa p=0.404 and p=0.461, lambda p=0.4 and 0.461 respectively) with an overall mean 95th percentile range of 4.40-15.33 mg/L for kappa and 7.05-20.02 mg/L for lambda. Conclusion: Interbatch agreement for the last 3 batches produced was within acceptable limits. Furthermore, reference range comparison showed that there was no difference between the results returned. These results support the role of Freelite measurement in prolonged patient monitoring.
S443

T305 ASSESSING ANALYTICAL QUALITY OF GLUCOSE MONITOR: A COMPARISON BETWEEN POCT SYSTEMS AND MULTIPARAMETER ANALYZER M. Casati(1), F. Lorenzini(1), A. Cappellani(1), S. Ippolito(1), S. Melzi(1), V. Perlangeli(1), S. Pittalis(2), L. Carati(1)
1 Laboratorio Analisi Chimico Cliniche, Azienda Ospedaliera San Gerardo, Monza, Italia 2 Laboratorio di Patologia Clinica, Azienda Ospedaliera della Provincia di Lodi, Lodi, Italia
T306 SWITCHING FROM IMMUNOASSAY TO LIQUID CHROMATOGRAPHY-TANDEM MASS SPECTROMETRY (LC-MS/MS) FOR IMMUNOSUPPRESSANT DRUGS DETERMINATIONS C.N. Castellana, D. Gallesi, A. Malagnino, S. Pisa, B. Poppi, A. Tomasi Toxicology and Clinical Pharmacology Laboratory - Azienda Ospedaliero-Universitaria di Modena Policlinico Background: The Toxicology and Clinical Pharmacology Laboratory performs about 14.000 immunosuppressant drugs determinations every year, about 50 test in a day, concluded by 3 pm of the day of sample receiving. Methods: Until 2011 the utilized methods were Immunoassay: CMIA (Chemiluminescent Microparticles Immuno Assay) on Architect (Abbott) for ciclosporin, tacrolimus and sirolimus; QMS (Quantitative Microsphere System) on CdX90 (Tema Ricerca) for everolimus; CEDIA (Chemiluminescent Enzyme Donor Immuno Assay) on Aries (IL-Instrumentation Laboratory) for Mycophenolic acid. From 2012 the utilized method is LCMS/MS (Liquid Cromatography/Mass/Mass) on UPLC-TQD (Ultra Performance Liquid Cromatography-Triple Quadrupole Detector) Waters with Chromsystems assay kit (column, deutered internal standard, calibrators, quality control and mobile phase) Immunoassay sensivity is 25 ng/mL for ciclosporin, 1.5 ng/mL for tacrolimus, 2 ng/mL for sirolimus, everolimus and 0.3 ug/mL for mycophenolic acid. LC-MS/MS sensivity is 5ng/mL for ciclosporin, 0.5 ng/mL for tacrolimus, sirolimus, everolimus and 0.1 ug/mL for mycophenolic acid. Even if LC-MS/MS is the reference method, comparison between LC-MS/MS and immunoassay patients data was necessary to support the Clinicians for patient therapeutic drug monitoring, during method switching. Results: The linear regression LC-MS/MS versus immunoassay and correlation coefficient are: Ciclosporin y=0.7791x+16.725 / r =0.990991, Tacrolimus y=0.821x+0.5924 / r=0.929904, Sirolimus y=0.7041x-0.1783 / r=0.916859, Everolimus y=1.0727x+1.0434 / r=0.774135, Mycophenolic acid y=0.8059x-0.2485 / r=0.948176. Conclusions: Our LC-MS/MS ciclosporin, tacrolimus, sirolimus and mycophenolic acid results are lower than the immunoassay results (ciclosporin -18%, tacrolimus -20%, sirolimus -25% and mycophenolic acid -25%. Only the LCMS/MS everolimus results are higher than the immunoassay results (+13%), because of a known undervalue of QMS immunoassay. After the results discussion with the Clinicians, we switched from immunoassay to LC-MS/MS for all immunosuppressant determinations, with the warranty to make available all results every day, by 3 pm of the day of sample receiving.
Background: The blood glucose test is one the most frequent in the hospital and it is a cornerstone in the diagnosis and monitoring of diabetes and management of hyperglycemia control in critical and non critical settings. Inside our hospital, on bedside glucose test takes account of about 50% of inpatient glucose measurements and different instruments for measuring glycaemia are used. The aim of the study was to demonstrate whether Roche Point of Care instruments for glucose measurements have the required accuracy compared with reference method (hexokinase) on multiparametric laboratory analyzer. Methods: Following the protocol of evaluation of Mahoney (2007) of split sample design methodology, glucose from arterial blood gas analysis (left over) was measured at the same time in double on the following systems produced by Roche Diagnostics (Mannheim, Germany): multiparameter analyzer Cobas C8000 (hexokinase, reference method), blood gas analyzer Cobas b123 (glucose oxidase), glucometer AccuChek Inform II Professional (glucose dehydrogenase modified). The results were evaluated according to the CLSI EP9-A Method Comparison. Results: We evaluated 90 samples with glucose values ranged between 1.06 and 30.61 mmol/L. (haematocrit range: 22.150.6%) The comparison between methods showed excellent correlations: in particular between Cobas b123 and Cobas C8000 (y =0.93+0.24, r =0.99, Standard Error Estimated = 0.715) and between Accu-Chek Inform II and Cobas C8000 (y = 0.98+0.28, r =0.998, Standard Error Estimated =0.281). The analysis with the Consensus Error Grid shows that 100% of the results fall within the acceptable range (Zone A. No effect on clinical action) in both of comparison. Conclusions: Our study shows an excellent correlation between the three Roche system methods for blood glucose measurement. For an analytical point of view, the Roche POC glucose testing systems are suitable for use in ICUs and other medical and surgical areas, according to the range of measurement considered, in interchangeable way.
S444

T307 AUTOMATED BLOOD CELL COUNTS: PRELIMINARY EVALUATION L. Cerutti(1), L. Ciardelli(1), L. Scudeller(2), E. Genini(1), G. Merlini(1) Clinical Chemistry Laboratory, IRCCS Fondazione Policlinico San Matteo, Pavia 2 Scientific Direction, IRCCS Fondazione Policlinico San Matteo, Pavia Background: The today automation facilities applied to haematological laboratory systems have changed the diagnostic scenario and help the role of Clinical Pathologist in improving test and supporting clinical decision. These results are achieved also applying the sophisticated analytic philosophies coming from the today advanced scientific research to the routine activity. The aim of our investigation is to evaluate and compare, on the same samples, the analytical performances of six automated blood cell counters. Methods: For this preliminary evaluation, a random 20 samples (out of 300 from those obtained in the routine) were selected. Each sample has been processed on the following hematologic instruments: 1) Abbott Cell-Dyn Sapphire, 2) Siemens Advia 2120, 3) ABX Pentra DX 120, 4) Beckman Coulter DXH 800, 5) Sismex Xe 5000, 6) Abbott Cell-Dyn Ruby, and the results obtained on leukocyte, lymphocyte, erythroblast and reticulocyte absolute count were used. The concordance Lins Coefficient for each pairwise comparison and the graphical Bland and Altman analysis were applied. Results: the range of measurements (from instrument 1) was 1.22-28.5 for leucocytes, 0.75-16.7 for lymphocyte, 0-56.4 for erythroblast, 0.002-0.27 for reticulocytes. The best/worse coefficient (for each comparison) was as follows: leucocyte 0.99 (5 vs 6)/0.21 (1 vs 6), lymphocyte 0.82 (4 vs 6)/0.11 (1 vs 4), erythroblasts 0.96 (2 vs 5)/0.01 (2 vs 3), reticulocytes 0.92 (2 vs 5)/0.59 (3 vs 4). Conclusions: For the evaluated parameters, the comparisons reveal a relatively low concordance between the hematology analyzers, probably due to the different technologies employed, particularly for leukocytes and lymphocytes. On the contrary we have observed a good concordance for erythroblast and reticulocyte absolute count. Our results, though drawn from a very small sample, suggest that several issues must still be resolved, for instance the analytic quality of the count of certain populations (namely: leukocytes) probably due to technological limits in searching for morphologic abnormalities of immature or atypical cells.
1
T308 ABILITIES OF THE FACTOR FOR CORRELATING DIMENSION JAFFE CREATININE METHOD TO THE IDMS REFERENCE MEASUREMENT PROCEDURE IN ASSESSMENT OF CHRONIC KIDNEY DISEASE A. Chittamma(1), S. Vanavanan(1), C. Kitiyakara(2), M. Rochanawuttanon(1)
1
Department of Pathology, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand 2 Department of Medicine, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand Background: To improve the performance of serum creatinine (sCr) measurement for the reliable of estimated glomerular filtration rate (eGFR), all major sCr assays have been standardized to the isotope dilution mass spectrometry (IDMS). However, this was not claimed by the Dimension Jaffe method but the manufacturer provided a correction factor to achieve IDMS traceable results. We compared sCr values derived from the manufacturers factor to IDMS traceable enzymatic method. Agreement of CKD classification based on the eGFR was also assessed. Methods: sCr was measured in 2,660 patients on the Dimension RxL MAX analyzer using the non-IDMS modified Jaffe method (non-IDMS sCr), then a correction factor (-0.168 mg/dL) was used to convert the results to IDMS creatinine values (corrected sCr). The samples were also performed by IDMS traceable enzymatic method (IDMS sCr) on the Vitros 350 analyzer. The CKD-EPI equation was used to calculated eGFR. Results: According to a factor derived from creatinine within the range of 0.30-2.5 mg/dL, 2,636 samples were used in data analysis. The median creatinine concentration with a range for corrected sCr (0.70 mg/dL, 0.27 to 2.31 mg/dL) was significantly lower than non-IDMS (0.87 mg/dL, 0.44 to 2.47 mg/dL) and IDMS sCr (0.78 mg/dL, 0.31 to 2.25 mg/dL), P <0.05. The bias was -0.08 mg/dL. The Passing-Bablok regression showed a slope of 1.0374 (95% confidence interval [CI], 1.021 to 1.053) and intercept of -0.124 mg/dL (95% CI, 0.136 to -0.111) with a high Spearmans rank correlation coefficient of 0.927 (95% CI, 0.922 to 0.932), P <0.0001. The limit of agreement was -0.23 to 0.06 mg/dL. Based on the eGFR corresponding to CKD stage 1 to 5 defined by Kidney Disease Outcome Quality Initiative (K/DOQI), the corrected sCr classified patients into CKD stages for 82.9%, 14.0%, 3.8%, 0.3%, 0% while the IDMS sCr were 73.6%, 21.3%, 4.7%, 0.4%, 0%, respectively. The overall consistency patients was 87.5% which demonstrated good agreement between both methods (kappa=0.657). Conclusions: The corrected sCr derived from factor for correlating Dimension Jaffe creatinine method to the IDMS method correlated and agreed well with the IDMS traceable enzymatic method. No patient was discordantly classified by more than one group of any CKD stages.
S445

T309 MEASUREMENT OF 1,25-DIHYDROXYVITAMIN D2 AND D3 IN BLOOD BY LC-MS/MS UTILIZING ION FUNNEL TECHNOLOGY P. Christensen, V. Starkie Agilent Technologies, Inc., Cheadle, UK Liquid chromatography triple quadrupole mass spectrometry (LC-MS/MS) has become an essential tool for small molecule quantitation due to its high sensitivity and specificity, excellent reproducibility and the ability to perform simultaneous analysis of multiple analytes. 1,25-dihydroxyvitamin D a metabolite of vitamin D has proven to be a challenging compound to analyze due to the low pg/ml range relevant to clinical research. Quantitative analysis was performed using multiple reaction monitoring (MRM) transition pairs for each analyte and internal standard. Final concentrations were calculated by comparing the response of the analyte to a known concentration of internal standard and plotting the result on a calibration curve developed using stripped human plasma spiked with standards. A UHPLC system configured for sample enrichment was also explored. This system consisted of two pumps, a 2 position/6 port valve, an enrichment column and a reversephase C18 analytical column. 1,25-dihydroxyvitamin D3 in extracted serum samples displayed a low pg/ml limit of quantitation (LOQ). The calibration curves shows excellent linearity and an R2 >0.998. The LOQ displayed good reproducibility and precision with a CV of <20%. A sensitive method has been developed for the quantitation of 1,25dihydroxyvitamin D2 and D3 in blood using ultra high performance liquid chromatography (UHPLC) and triple quadruple (QQQ) mass spectrometry enhanced with dual ion funnel technology. T310 SIMULTANEOUS MEASUREMENT OF THYROXINE (T4), TRIIODOTHYRONINE (T3) AND REVERSE TRIIODOTHYRONINE (RT3) BY TWO DIMENSIONAL LIQUID CHROMATOGRAPHY AND TANDEM MASS SPECTROMETRY IN NEGATIVE ION MODE D. Yang, A. Fandino, P. Christensen Agilent Technologies, Inc., Cheadle, UK Accurate and precise measurement of thyroid hormones in serum or plasma is important for clinical research. In this work, we present a reliable and sensitive method for the measurement of thyroid hormones. Liquid chromatography with online sample cleanup was employed to minimize manual sample preparation. This was coupled to a triple quadrupole mass spectrometer utilizing dual ion funnel technology to increase ion sampling and attain the highest sensitivity. The LC configuration comprising a loading and an analytical column allows cleaning hydrophilic unretained matrix components while trapping the target analytes at low organic mobile phase composition during the loading process. During the analysis process, the column valve is switched and the mobile phase strength is increased to transfer the target analytes onto a narrow bore sub-2 m analytical column. The LC configuration also enables injecting a large sample volume. Thyroxine (T4), triiodothyronine (T3) and reverse triiodothyronine (r-T3) were spiked in charcoal stripped serum after removing thyroid binding globulin and other large proteins. In negative ion mode, r-T3 can be quantified using its two unique multiple reaction monitoring (MRM) transitions at 649.8>604.6 and 649.8> 462.8. T4 can be quantified using the MRM transition at 775.8 >126.9 and T3 can be quantified using the MRM transition at 649.8>126.9. Critical aspects of LC/MS/MS method development affecting sensitivity are identified for quantifying thyroid hormones at the low pg/mL level. Under optimal LC and MS/MS conditions, LODs of 1 pg/mL for T3 and 2 pg/mL for T4 and rT3 as well as LLOQs of 2 pg/mL for T3 and <5 pg/mL for T4 and rT3 were obtained. The linear correlation coefficients in the range from 2 pg/mL to 100 pg/mL were >0.99 for the three analytes. Free thyroid hormone concentrations can be reliably quantified at the low pg/mL level by LC/MS/MS. The advantage of using mass spectrometry consists not only better specificity, but also capability of quantifying multiple analytes within one experiment.
S446

T311 DETERMINATION OF OPIATES IN DRIED BLOOD SPOTS USING NOVEL FLOW-THROUGH TECHNOLOGY COUPLED TO LC/MS/MS P. Christensen, K. McCann, K. Lewis, D. Nagtalon Agilent Technologies, Inc., Cheadle, UK The use of dried blood spot (DBS) technology in clinical research and pharmacokinetic studies has significant advantages to conventional plasma sampling because it allows sampling of small blood volumes, easy sample shipping and storage, and alleviates many concerns related to the handling of biohazardous materials. DBS coupled to highly sensitive LC/MS/MS systems promises significant advantages in bioanalytics both in the toxicology and pharmaceutical industry. A novel, fully automated online card extraction system coupled to a triple quadrupole mass spectrometer has been demonstrated for the analysis of opiates. This enables efficient, highly sensitive and specific, automated flow-through analysis of DBS cards by LC/MS/MS. This integrated analytical system greatly reduces analysis time and manual experimental errors. This work describes the online analysis of DBS cards for the determination of opiates. Excellent sensitivity down to 1ng/mL, linearity with an R2 >0.99, and precision with CV <15% were demonstrated for the online extraction method. The quantitative performance capabilities of the online extraction method was evaluated and compared to that of a conventional offline extraction hole punching method and consistent results were observed. The flow-through analysis of DBS cards by LC/MS greatly reduces analysis time and manual experimental errors. Excellent sensitivity, linearity, dynamic range, precision, accuracy, and reproducibility, and the quantitative performance capabilities of the online extraction method were demonstrated. This DBS analysis approach can be readily applied to clinical research, forensic toxicology and pharmaceutical, research studies. T312 EVALUATION OF IMMUNOGLOBULIN FREE LIGHT CHAIN (FREELITETM) ASSAYS ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER D.G. McEntee, D.J. Matters, M.D. Coley, S.J. Harding, P.J. Showell The Binding Site Group Ltd, Birmingham, UK Introduction: Here we describe the development of serum immunoglobulin free light chain (FreeliteTM) assays for The Binding Sites Next Generation Protein Analyser (NGPA). The NGPA analyser is a random-access bench top turbidimetric analyser capable of a wide range of on board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour, and multiple methods of antigen excess detection. Precision is promoted by single use cuvettes which are automatically loaded and disposed of. Method: NGPA kappa and lambda FLC assays were compared to existing Freelite (BNIITM) assays for 141 kappa (79 normal, 62 monoclonal, mean 535.20 mg/L, range 1.13-17,868.19 mg/L) and 129 lambda (79 normal, 50 monoclonal, 275.06 mg/L, range 0.52-3833.94 mg/L). Precision by repeat measurements was assessed at high (134.01 mg/L and 129.75 mg/L) and low levels (6.47 mg/L and 7.56 mg/L) (kappa and lambda respectively). Linearity was determined by serial dilutions from 30.99 mg/L to 3.10 mg/L (kappa) and 30.52 mg/L to 0.76 mg/L (lambda), measured in triplicate with results compared to expected values. 71 kappa and 79 lambda samples identified for the purpose of antigen excess testing were analysed. Results: 149 kappa samples mean 535.20 mg/L (range 1.1317,868.19 mg/L) compared to mean 540.77 mg/L (range 2.37-20,000.00 mg/L) on the predicate device (R2 =0.98). Similarly 129 lambda samples gave a mean value of 275.06 mg/L (range 0.52-3833.94 mg/L) compared to mean 288.83 mg/L (range 0.51-4480.00 mg/L) on the predicate device (R2=0.96). Kappa returned a precision CV at the high level of 2.2% (mean 134.01 mg/L, range 130.59-140.38 mg/L) and the low level of 4.9% (mean 6.47 mg/L, range 6.09-6.91 mg/L), whilst lambda returned 1.0% (mean 129.75 mg/L, range 126.61-131.32 mg/L) and 3.7% (mean 7.56 mg/L, range 7.147.96 mg/L). Both kappa and lambda gave acceptable linearity (R2=0.99 and 0.98 respectively) with a maximum deviation of 11.9% from the expected result. 9/9 kappa and 9/9 lambda samples identified for antigen excess were automatically rediluted on the NGPA. Conclusions: We conclude that these assays are rapid, accurate and precise and protected from false low results by the instruments automatic antigen excess check function.
S447

T313 EVALUATION OF A C4 ASSAY FOR USE ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER P.S. Patel, F. Murphy, M.D. Coley, P.J. Showell, S.J. Harding The Binding Site Group Ltd, Birmingham, UK Introduction Serum complement consists of around 30 proteins that have a fundamental role in immune system functionality. Inherited deficiencies in C4 are associated with an increased risk of developing systemic lupus erythematosus (SLE). Conversely, high levels of circulating immune complexes in SLE can reduce serum levels of complement components. C4 deficiency is also associated with glomerulonephritis and vasculitis. Here we describe the development of serum C4 assay for Binding Sites Next Generation Protein Analyser. The instrument is a random-access bench top turbidimetric analyser capable of a wide range of on-board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour. Precision is promoted by single-use cuvettes which are automatically loaded and disposed of, whilst the utility is enhanced through host interface capability, primary sample ID and bar coded reagent management systems.The assay has a measuring range of 0.064 0.9 g/L at the standard 1/10 sample dilution, with sensitivity of 0.0064 g/L. High samples are remeasured at a dilution of 1/20 with an upper measuring range of 0.128-1.8 g/L. Method: Correlation to the Binding Site C4 assay for the SPA PLUS was performed using 33 samples (mean 0.344; range 0.152 0.7160). Intra-run precision was assessed by measurement of twenty replicates of samples at 0.787 g/L and 0.105 g/L. Furthermore, precision was assessed at the clinical decision point of 0.154 g/L. Linearity was assessed by assaying a serially-diluted patient sample pool across the width of the measuring range and comparing expected versus observed results (0.001 0.882 g/L). Results: Correlation with the C4 SPA PLUS assay demonstrated good agreement when analysed by PassingBablok regression; y=1.08x - 0.03. The assay was shown to be linear over the range of 0.001 0.882 g/L; y=1.013x + 0.0177 mg/L (R = 0.998). Intra-run precision produced the following results: sample 1 (0.787 g/L) CV of 1.59%, sample 2 (0.154 g/L), CV of 2.17% and sample 3 (0.105 g/L), CV of 2.83% Conclusions: We conclude that the C4 assay for the Binding Site next generation protein analyser is reliable, accurate and precise and shows good agreement with existing assays. T314 DEVELOPMENT OF AUTOMATIC ANTIGEN EXCESS DETECTION PARAMETERS FOR IMMUNOGLOBULIN FREE LIGHT CHAIN (FREELITETM) ASSAYS ON THE ROCHE COBAS C501 M.D. Coley, D.J. Matters, P.J. Showell, S.J. Harding The Binding Site Group Ltd, Birmingham, UK Background: Studies have shown that immunoglobulin free light chain (FLC) measurements aid in the diagnosis and monitoring of patients with AL amyloidosis, non-secretory myeloma and light chain myeloma. The inherent nature of monoclonal FLC makes antigen excess a possibility even for polyclonal based assays. Here we describe the development of automatic antigen excess protection for use on existing FLC assays (FreeliteTM) on the Roche cobas c501. Method: Reaction kinetics of monoclonal patient sera prone to antigen excess (5 kappa, mean c501 result 5,345.12 mg/L, range 502.62-12,672.00 mg/L; and 4 lambda, mean c501 result 4,046.5 mg/L, range 1,590.00-6,213.00 mg/L) and 4 blood donor sera (mean kappa 10.51 mg/L, range 9.14-13.18 mg/L; mean lambda 11.41 mg/L, range 10.90-12.14; mean ratio 0.93, range 0.79-1.21) were analysed to set threshold limits for antigen excess protection. These thresholds were then validated using 67 blood donor serum, 68 kappa monoclonal and 32 lambda monoclonal patient sera. Results: Samples in antigen excess were typified by a high initial rate of reaction, which rapidly slows as the detecting antibody becomes saturated (early delta OD 125.5, late delta OD 14.0). This compares to non-antigen excess samples which show a slower, more sustained rate of reaction throughout the assay time (early delta OD 28.0, late delta OD 38.7). 22/40 kappa samples (mean 314.64 mg/L, range 2.64-4,919.00 mg/L) gave a false low result when assayed without antigen excess protection whilst for lambda this was 7/32 samples (mean 879.01 mg/L, range 7.68-4,738.00 mg/L). When assayed with antigen excess thresholds active 0/40 kappa samples gave a false low result (mean 3,103.82 mg/L, range 20.06-40,930.00 mg/L) whilst 0/32 lambda samples gave a false low result (mean 3,548.30 mg/L, range 7.68-56, 949.00 mg/L). Conclusions: All samples previously identified as being in antigen excess on the c501 analyser were automatically rediluted using these thresholds. Implementation of these parameters will improve throughput and prevent samples being mis-reported.
S448

T315 VERIFICATION OF ANALYTICAL PERFORMANCE AND COMPARABILITY OF BLOOD GAS ANALYZERS L. Contardi(1), C. Uhelly(2), D. Larosa(2), F. Parodi(1), S. Quiroga(1), V. Correa(1) CEMIC - Departamento de Anlisis Clnicos, Buenos Aires 2 ROCHE Diagnostics Argentina Background: Verification is confirmation, with objective evidence, that specific requirements have been fulfilled, difficult in blood gas analyzers due to low samples stability and aqueous control solutions. If two instruments are available, comparability must be ensured. Results difference from same patient may have clinical impact. So, we verified performance and assured comparability of two blood gas analyzers Cobas b 221, Roche. Methods: Repetibility (CLSI EP-15): 3 control levels tested automatically in an AUTO QC MODULE. Accuracy: linear regression between own results and EQAS Riqas consensus mean, bias was calculated at different concentrations. Total Error (TE) compared to Quality Requirement (TEa) RCPA: TE <TEa. Linearity (CLSI EP-6): 7 solutions (COOX/MSS) with known levels of O2, N2 and CO2, salts, buffers and metabolites. Uncertainty: applying Nordtest to Riqasresults. Comparability (EP-C54-A): with patients samples. Acceptability criterion: RCPA. Results: We verified precision manufacturers claims for: pH, pCO2, pO2, Hematocrit (Hto), Ionized calcium (Cai), Sodium, Potassium, Chloride, Bilirubin (Bb), Glu, Lac, Total Hemoglobin (Hbt), Oxygen Saturation, Oxyhemoglobin (O2Hb), Carboxyhemoglobin (COHb), Reduced Hemoglobin (HbH) and Methemoglobin (METHb). Linearity range: pH 6,87-7,69; pCO2: 13-130 mmHg; Hto: 22-75%; Cai: 0,43-2,52 mmol/L; Na: 90-173 mEq/L; K: 2,0-8, 8 mE/L; Cl: 68-132 mEq/L; Hbt: 6,2-23g/dL; O2Hb: 35-93,8%; COHb:2,8-28,2%; HbH:2,022,3%; METHb: 1,6-14,6%; Bb: 4,0-23,8 mg/dL; Glu: 29-460 mg/dL; Lac: 1,4-13,3 mmol/L.. We verified accuracy and acceptable performance regarding TEa for: pH, pCO2, Lac, Cai, Na. K, Cl and Glu in the three levels. We couldnt verify pO2 linearity and accuracy. Uncertainty %: pH=0,2; pCO2:6,21; pO2:13,05; Cai:2,51; Na:1,48; K:2,29; Cl:4,71; Glu:14,69; Lac:15,91. 3 level results comparability was obtained for pH, pCO2, pO2, Na, K, Cl y Cai. Conclusions: both instruments meet quality requirements and comparability for all parameters except pO2, due to possible contamination with ambient air.
1
T316 AQT90 FLEX - PRACTICABILITY AND IMPACT ON TNI TAT IN AN EMERGENCY LABORATORY B.C. Creanza, N. Dipace, M. Guida, G. Dirienzo ASL Bari, Struttura Complessa di Patologia Clinica, Ospedale Civile Umberto I, Altamura, Bari Background: The goals of this study were to evaluate the analytical performance of TnI (Troponin I) on whole blood AQT90FLEX analyzer (Radiometer ApS),random access POC (Point Of Care), and impact on TnI TAT (Turn Around Time) though its implementation in a emergency section of central laboratory. The practicability of AQT90FLEX has been analyzed with a Test Operators Evaluation. Methods: Correlation and concordance studies were performed collecting blood sample from ED (Emergency Department) and Cardiac Intensive Care Unit (CICU). Samples were drawn in duplicate: whole blood EDTA (K3EDTA) to be tested on AQT90FLEX and serum tube to be tested on ARCHITECT (Abbott-i2000SR). Total imprecision was determined using 3 levels of quality control materials and 2 plasma pools. The TnI TAT of the sample from ED analyzed with AQT90 FLEX was compared to the TnI TAT from ARCHITECT.TAT data were collected for several days in the slot time 8.30 AM02.30 PM. Practicability of AQT was evaluated through a questionnaire of 22 parameters, data were collected from different professional role. Results: Correlation of patient between the AQT and the ARCHITECT demonstrated the following regression: AQT cTnI = 0.1276 ARCHITECT+ 0.0073 r =0.989 (95% Confidence 0.9819 to 0.9938); concordance was 96%; day to day imprecision was CV% between 3.08 and 8.38 for concentration between 0.035 and 1.4 ng/mL. TAT analysis: AQT90 FLEX TnI <45=32 samples, between 45 and 60=15 , >60=11; Abbotti2000SR TnI <45=15 samples, between 45 and 60=14 , >60=28, average TAT on AQT90 44, Abbott- i2000SR 67. Test Operators Evaluation: Device design 4/5, Infection prevention 5/5, Practicability 5/5, Measurement 4/5, Overall 5/5. Conclusions: Troponin I is a very reliable marker that allows early diagnosis of ACS (Acute Coronary Syndrome), but requires short TAT.Both good practicability of the random access analyzer AQT90FLEX, with ability to work from capped primary whole blood sample tube, and the high degree of correlation with laboratory results allow to TnI TAT to be improved by an emergency section of a central laboratory. The safe and simple technology is suitable for measurements of cardiac troponin also by unskilled personnel in other settings such as emergency department and cardiac care unit.
S449

T317 METHOD COMPARISON OF FOUR AUTOMATED CA 15-3 IMMUNOASSAY SYSTEMS N. Cunha, S. Carreiro, M. Costa, J. Pimenta, M.A. Mendes Servio Patologia Clinica, IPO Coimbra FG, Portugal Background: Measurements of serum cancer antigen (CA) 153 are used to monitor tumor recurrence and treatment of patients with advanced breast cancer. Most systems use two monoclonal antibodies against (MAbs) 115D8 and DF3, to detect two epitopes in the molecular structure of the polymorphic epithelial mucin MUC1, commonly referred as CA 15-3. In the present study, we compared the methods of 4 automated CA 15-3 immunoassay systems. Methods: Results obtained in individual immunoassay systems, Centaur CA 15-3 (Siemens Healthcare Diagnostics), Cobas e411 CA15-3 (Roche Diagnostics) and Architect i1000SR CA15-3 (Abbott Laboratories), with the same set of 142 samples were compared with the Kryptor CA 15-3 (Thermo Scientific) system as our reference method using PassingBablok regression and differential diagrams according to Bland-Altman. Results: Poor agreement [y=1.09x+0.62 (r = 0.98; n = 142; 95% confidence intervals for slope and intercept, 1.01-1.14 and -1.47-3.51)] was found between Cobas e411 CA 15-3 (y) and Kryptor CA 15-3 (x). On the other hand, the comparison between Centaur CA 15-3 (y) and Kryptor CA 15-3 (x) revealed the highest degree of agreement with a slope of 1.03 (95% confidence interval, 0.95-1.12), and a negligible intercept (-0.49 U/mL; 95% confidence interval, -3.94 to 2.21) with r = 0.97. Acceptable agreement was observed [y = 1.05x-2.26; r = 0.96; n= 142; 95% confidence intervals for slope and intercept, 0.98 to1.13 and -4.68 to 0.61)] between Architect i1000SR CA15-3 (y) and Kryptor CA 15-3 (x). When method comparison studies were analysed using difference plots with Kryptor CA 15-3 in the abscissa, the mean differences were: 11.38 (Cobas e411 CA 15-3), 1.35 (Centaur CA 15-3) and 15.64 U/mL (Architect i1000SR CA15-3). Conclusions: In our study Cobas e411 CA 15-3 method had the highest slope with Passing-Bablok analysis. Data from Bland-Altman differential plots suggest significant individual differences among individual samples, mainly for high concentrations, which occur in patients with active breast cancer. Even for the method (Centaur) that showed the best agreement with Kryptor CA 15-3, substantial intermethod differences exists for some samples, indicating that redetermining the baseline is required when changing method. T318 EVALUATION AND COMPARISON BETWEEN METHODS FOR SERUM FREE LIGHT CHAIN IN PATIENTS WITH MONOCLONAL GAMMOPATHY M. Cuomo, V. Barchiesi, D. Cerasuolo, E. Cavalcanti Laboratory Medicine Unit, Istituto Nazionale Tumori IRCCS, Fondazione G. Pascale, Naples, Italy Background: The measurement of serum free light chains (FLC) is recommended for diagnosis, monitoring and prognosis of monoclonal gammopathies. In addition, specific assays on automated nephelometers and turbidimeters are available. Many factors need to be considered when deciding upon the most appropriate method for measuring FLC because the discovery of the monoclonal component can be unexpected and require further testing for confirmation. Methods: The aim of this study was the evaluation and comparison of two commercial kits: the new N Latex FLC (Siemens) and the FreeliteTM (Binding Site) assay, applied respectively on a BNP ProSpec, Siemens and Cobas C 6000, Roche, in accordance with manufacturers instructions. We included in this study 108 samples of patients with monoclonal gammopathies, admitted to the Istituto Nazionale Tumori, in order to compare methods and 60 serum samples from healthy donors to establish Reference interval with new N Latex FLC. Results: The T test shows significant statistical difference between methods for FLC (p=0.01) and / ratio (p=0.02), non significant for FLC (p=0.69). A moderate correlation was observed in the method comparison (Pearson coefficient) for FLC (r=0.89), for FLC (r=0.60) and / ratio (r=0.78). Cohen Method shows the concordance for FLC = 89%, for FLC = 65% and for / ratio=84%. A relevant difference in the results obtained from the two methods was observed in a patient with chain multiple myeloma. The FLC concentration was highly abnormal with the FreelightTM , whereas in the N Latex FLC assay the levels were within the reference ranges. This difference can depend to the specificities and affinities of the antibodies. The precision data, obtained according to CLSI, demonstrated the high precision for both assays with CV lower to 5% for FLC and in the within- day and lower 2% in the between- run. The Reference interval which we calculated for N Latex FLC are similar to those reported by manufacturers. Conclusions: Our study demonstrated that there is only a moderate concordance between the two FLC assay and further studies are needed in order to verify the performances of these methods in the diagnosis and monitoring of monoclonal gammopathies.
S450

T319 IMPLEMENTING EXACTIVE ULTRA HIGH RESOLUTION MASS SPECTROMETER AND EXACTFINDER DATA PROCESSING SOFTWARE IN TARGETED AND UNKNOWN SCREENING METHOD FOR URINE ANALYSIS M. Kozak(1), B. Duretz(1), M. Lee(2), C. Stefan(2), E. Wang(2), C. De Nardi(3)
1
T320 THERAPEUTIC DRUG MONITORING OF 8 NEW ANTICANCER AGENTS BY HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY-TANDEM MASS SPECTROMETRY I. Andriamanana(1), I. Gana(1), A. Hulin(1), B. Duretz(2), C. De Nardi(3) GH Henri Mondor, AP-HP, Universit Paris-Est Crteil, France 2 Thermo Fisher Scientific, Les Ulis, France 3 Thermo Fisher Scientific, Rodano, Italy Introduction: The treatment of some cancers has shifted from conventional chemotherapy drugs to chronic treatment with molecular targeted therapies. Targeted therapies include drugs such as Tyrosine kinase inhibitors (eg: Imatinib, Dasatinib, Nilotinib). A highly sensitive and selective method has been developed for the analysis of eight drugs using LC/MSMS. Methods: 50 L of plasma samples were extracted with methanol and the organic layer was diluted into the mobile phase. The injection volume was 15 L. Quantitation was performed using the imatinib-D8 as an internal standard. Chromatographic separation was achieved using a gradient on a C18 column. The mobile phases were the following: A was water containting 0.1% formic acid and 10 mM ammonium formate and B was acetonitrile containing 0.1% formic acid. Flow rate was 300 l/mn. Analytes were quantified using electrospray ionization in positive mode. Results: Calibration curves were established from 100 to 10000 ng/ml in human plasma, calculated and fitted by 1/x2 weighted linear regression. The square of the sample correlation coefficients, R2 were between 0.977 and 0.995 (n=10). Replicate analysis of quality control samples at the three concentrations (low, medium and high) was used for the intra and inter-assay precision and accuracy determination. Intraday and inter-day imprecision values were below 15%. The percentages of deviation between experimental and theoretical concentration were also below 15%. The specificity of the method was examined by analyzing different blank human plasma extracts with or without internal standard. No interferences were observed. The limits of quantitation were 100 ng/mL for all compounds except for dasatinib that was evaluated at 50 ng/mL. Conclusion: This method is the first broad range LC-MS/MS assay covering the major currently in-use TKIs. It has been validated according to international regulations and can be used to evaluate patient adherence to therapy and pharmacokinetic studies.
1
Thermo Fisher Scientific, San Jose, CA 2 Center of Addiction and Mental Health, Toronto, ON 3 Thermo Fisher Scientific, Rodano, Italy
Introduction: Fast methods allowing identification of unlimited number of compounds with capability of retrospective data analysis are required in toxicology labs for quick and confident analysis. Using ExactiveTM ultra-high resolution mass spectrometer with ExactFinder TM data processing software meets these expectations. Methods: Urine was spiked with internal standards, diluted 20 times with water and analyzed on PFP column with 15 min LC gradient. Mass spectrometer was equipped with ESI source. Full scan data in m/z range 100-1000 mu followed by Higher Energy Collisional Dissociation (HCD) was collected in positive and negative switching ionization modes. Resolution was 50K and 25K (FWHM) for full scans and fragmentation scans respectively. ExactFinder software identified target compounds base on exact mass, retention time, isotopic pattern, and presence of fragments. Unknown compounds were identified based on exact mass and isotopic pattern; proposed molecular formulas were searched against ChemSpider libraries. Results: Data collected with screening application on Exactive Orbitrap mass spectrometer and processed with ExactFinder software correlates very well with GC-MS and immunoassay data. All compounds identified with GC-MS and immunoassay were identified with our method. Additional compounds and more metabolites confirming presence of parent drug were identified. Collected data allowed for high confidence identification using m/z, retention time, isotopic pattern and all ions fragmentation spectra. We also found that information about retention time is not necessary required. Conclusion: Implementation of Exactive ultra high resolution mass spectrometer and ExactFinder software allowed us to develop efficient screening method which correlates well with traditional screening techniques, allows for unlimited number of target compounds, retrospective data analysis and unknown compounds identification.
S451

T321 COMPARISON OF A NEW HBA1C ANALYZER (HLC-723 GX) WITH TWO ESTABLISHED SYSTEMS B. Del Castillo-Figueruelo, I. Domnguez-Pascual, M. Conde-Snchez, J. Bobillo-Lobato, J.M. Guerrero, M.T. Herrera-Rey Department Of Clinical Biochemistry, Virgen Del Rocio University Hospital Background: Evaluation of glycated haemoglobin fraction (HbA1ac) levels is now widely accepted and performed in the monitoring of diabetic patients. High-performance liquid chromatography (HPLC) is considered the gold standard measurement procedure and this is the most frequently used technique in clinical practice. The objetive of this study was to compare three automated HbA1c tests using the same measurement principle, HPLC with cation-exchange columns. One of these is a new automated analyzer in its first evaluation study: HLC-723 GX Analyzer. Methods: One hundred and thirty six whole blood samples were collected by ramdon process for this study. The three automated tests were perfomed by Bio-Rad Variant II Turbo with 1.5 minutes HbA1c program, Bio-Rad Variant II with 3 minutes HbcA1c program and HLC-723 GX Tosoh Analyzer. 54 samples were analyzed by the three tests while 136 samples just by Variant II Turbo 1,5 min and Tosoh GX. A database of results was generated using Microsoft Excel and multiple regression analysis model was used for estimating the relationships among the results from the three tests. PassingBablok was used to obtain the correlation between Tosoh GX and Variant II Turbo 1,5 min; and Tosoh GX and Variant II 3 min. Results: Multiple regression analysis model generated a coefficient of determination (R2) of 99,65% giving a strong concordance between the results from the three tests. PassingBablok expressed a good correlation in the different test combinations: GX and Variant II Turbo 1,5 min (Intercept 0.10, 95% CI [-0,1561 to 0,1000]; slope 1.00, 95% CI [1,0000 to 1,0303]) ; GX and Varian II 3 min (Intercept 0.25, 95% CI [0,1929 to 0,2500]; slope 1.00, 95% CI [1,0000 to 1,0714]). Conclusions: We found excellent correlation between the three HPLC methods. This comparison study showed that a change test would not affect the monitoring of diabetes patient. The conclusions are particularly relevant to HLC-723 GX Analyzer as this is its first evaluation. T322 COMPARATIVE ANALYSIS OF HLA-B27 IMS-SANDWICH ELISA, FLOW CYTOMETRIC TEST AND PCR-SSP IN TYPING HLAB-27 P. Deo(3), J.L. Pathak(1), L. Hua(1), D. Dali(2), R.L. Shrestha(2)
1 Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China 2 Om Hospital and Research Centre, Department of Pathology, Kathmandu, Nepal 3 Department of Biochemistry and Molecular Biology,Monash University,Clayton Campus,Melbourne,Australia
Background: HLA-B27 antigen displays a degree of association with rheumatic diseases like ankylosing spondylitis(AS),Reiter syndrome and acute anterior uveitis. HLA-B27 typing is one of the useful tools for confirming association of AS with HLA-B27 phenotype in the clinical diagnosis. Most commonly used three types of test for HLA-B27 typing are 1. serological-flow cytometric (FC), 2. cytological- conventional microlymphocytotoxicity test (MLTC) and recently 3.Genotyping -polymerase chain reaction (PCR) with sequence specific primers (PCR-SSP) and PCR with sequence specific oligonucleotides (PCRSSO) . Aim of our study was to rule out the efficacy of IMS-sandwich ELISA for HLA-B27 typing in our setup. Methods: We compared results from three different methods Flow cytometric (FC), IMS-sandwich ELISA and PCR-SSP in 200 patients suspected of rheumatic diseases.Taiwan Advance Bio-Pharmaceutical HLA B-27 screening kit was used for IMSsandwich ELISA. Flow cytometric was performed by using HLA-B27 BD Biosciences screening kit and tools. DNA was extracted from the samples showing positive and ambiguous screening test results. DNA extraction was done by standard salting out procedure from whole blood. PCR-SSP was performed by using GmbH,Germany. Results: Out of 200 samples,IMS-sandwich ELISA showed 69 positive and 3 ambiguous test results while Flow Cytometric showed 70 positive and 2 ambiguous test results. Extracted DNA from these samples was further analyzed by PCRSSP.All72 samples(including positive and ambiguous test results from screening test) were HLA-B27 positive by PCRSSP by assigning cut off value CtB27 22.42 to discriminate HLA B-27 positive samples from negative. IMS-sandwich ELISA showed specificity of 100% and sensitivity of 95.8% while flow cytometric showed 100% specificity and 97.2% sensitivity with compared to PCR. Conclusion: PCR is most reliable among these three techniques employed ,but it is not practical in our set up due to lack of infrastructure,and limited number of sample load . PostPCR identification steps as gel electrophoresis, time -consuming procedure and prone to contamination are limitations of PCR, simple test procedure, cost effectiveness, 100% specificity and around 96% sensitivity makes IMSsandwich ELISA best option for our set up.
S452

T323 ACCURACY OF THREE AUTOMATED 25HYDROXYVITAMIN D ASSAYS IN HAEMODIALYSIS PATIENTS B. Depreter(1), A. Heijboer(2), M. Langlois(1)
1
T324 DETERMINATION OF HYALURONIC ACID IN SERUM: ADAPTATION OF THE HA IMMUNOTURBIDIMETRIC KIT REAGENTS (CORGENIX, USA, DISTRIBUTED IN FRANCE BY ELITECH) ON UNICEL DXC 880I (BECKMAN COULTER) P. Desvignes(1), B. Vidal(1), A.M. Lorec-Penet(2), N. Luci(1), F. Hassanaly(1), H. Portugal(2) Laboratoire, Hpital de Martigues Laboratoire Central Sud, Assistance Publique-Hpitaux de Marseille, Portugal
2 1
Department of Laboratory Medicine, AZ Sint-Jan BrugesOstend AV, 8000 Bruges, Belgium 2 Endocrine laboratory, VU University Medical Center, 1081 HV Amsterdam, the Netherlands Introduction: Patients on dialysis therapy form an important subgroup for vitamin D measurements with high needs for accurate monitoring. Suboptimal 25 (OH)D concentrations increase the risk for hypocalcaemia, secondary hyperparathyroidism and mortality. Despite standardization efforts, unresolved discrepancies and unacceptable bias persist among immunoassays compared to reference methods, particularly in samples from diseased patients with atypical matrix composition such as haemodialysis patients. We evaluated the accuracy of 3 automated assays on Architect i2000sr (Abbott), ModulaE170 (Roche) and iSYS (IDS) analyzers for 25(OH)D measurement in comparison to a higher reference isotope dilution/online solid-phase extraction liquid chromatography tandem mass spectrometry (ID-XLC-MS/MS) method, in serum from haemodialysis patients. Material and methods: All three routine assays were heterogeneous, competitive immunoassays (Architect, iSYS) or vitamin D binding protein assay (ModularE170), measuring 25(OH)D2 and 25(OH)D3. We studied 99 haemodialysis patients (47 women, 52 men, age 24-94y) and a healthy control group of 50 blood donors (34 women, 16 men, age 20-65y). Measurements were double blind performed in three different laboratories with a different operator and aliquot for each method. Conclusion: Not all automated 25(OH)D assays equally accurate measure samples from haemodialysis patients compared to samples from healthy subjects. We suggest a possible role of matrix effects like elevated urea or other retained metabolites in haemodialysis sera, causing incomplete binding disruption between 25(OH)D and DBP. Architect shows the highest deviation from ID-XLC-MS/MS, almost consistently producing lower results, as well as for haemodialysis as healthy subjects. Architect falsely assigned 48.5% haemodialysis patients as having suboptimal (<30 ng/mL) and 12.2% patients as insufficient (<20 ng/mL) levels. Despite the lower degree of correlation with ID-XLC-MS/MS, iSYS showed a slightly better performance than ModularE170 in the clinically important concentration range 10-40 ng/mL. ModularE170 showed in overall the best fit crossed the line of identity and highest correlation coefficient, and is considered as the most reliable method to measure 25(OH)D in haemodialysis samples.
Background. Large unbranched glycosaminoglycan, the hyaluronic acid (HA) is widely distributed in the extracellular matrix and a small quantity is found in blood, essentially degraded in liver sinusods. Excellent direct biomarker of liver cirrhosis, it is used in combination with other blood markers in many algorithm-based scores, which provide an assessment of degree of fibrosis in chronic liver diseases (alternative to the invasive liver biopsy). Latex immunoturbidimetric assays to measure serum HA has been developed, adapted on automated analyzers, but no adaptation exist on Beckman Synchron Systems. Materials and methods. Then, we adapted the dosage of serum HA on a UniCel DxC 880i (method NIPIA at 940 nm) with the reagents of the HA Immunoturbidimetric Test Kit from Corgenix. Sera from hospitalised patients, harvested on dry tube with gel separator (Greiner Bio-One), were analyzed according to the protocol of validation of methods VALTEC (SFBC). Results. Good imprecision: intra-assay CVs were 4.2% at 70 g/L and 3% at 286 g/L, inter-assay CVs 5.2% at 50.7 g/L, 3.1% at 189 g/L and 2% at 484 g/L, with detection limit = 16.5 g/L and linearity >785 mg/L; no hook-effect until 50,000 g/L, no interference with bilirubin nor turbidity in the condition of protocol; the Bland Altman difference plot and the PassingBablock regression analysis (y = 0.97x + 5.44; r = 1.0; n = 42) show a good correlation with the ELISA method (HA test Kit from Corgenix). Stability of calibration was verified during 8 days with the reagents on-board and 30 days if stored stoppered at 2-8 C. Conclusion. The good performances of this method and its good correlation at every level with the ELISA method, already used in routine for the liver function tests, considered as reference method, thus allow us to use this turbidimetric adaptation on Synchron Systems, and to completing the range of parameters needed for the calculation of the fibrosis scores on these analyzers.
S453

T325 USE OF GEM PREMIER 3000 IN CARDIAC SURGERY UNDER EXTRACORPORAL CIRCULATION D. Dineva(1), I. Paskaleva(1), E. Doncheva(1), V. Kolarov(2), L. Bojadzhiev(2) National Cardiology Hospital - Laboratory Diagnostics Department 2 National Heart Hospital - Cardiac Surgery Clinic, Bulgaria Background: Use of extracorporal circulation (ECC) during cardiac surgery results in hemodynamic and metabolic alterations, disbalance of blood cell components and electrolytes. These dynamic changes require quick and precise measurements of blood gas parameters (, 2 and 2), electrolytes (Na+, K+, Cl), glucose, lactate and oximetry (hemoglobin and hematocrit). Aim of Study: Comparison of a cell analyzer GEM PREMIER 3000 (IL) with a conventional analyzer of blood-gas analysis and electrolytes OMNI C (Roche), as well as comparison of Hb and Hct measured with reference spectrophotometric and microfuge method, respectively. Methods: Precision of analyses was compared in parallel series on both analyzers (n=20). Regression analysis y-intercept, slope, bias, correlation coefficient, was used for statistics and assessment of difference. Blood arterial samples from patients cooled to 28 and 32oC, during aorta clamping, after rewarming to 36-37oC, and after ECC, were analyzed. Cardioplegic solution KIRKLIN (Laboratorium Dr. G. Bichsel AG) was used for cardiopulmonary bypass (CPB). Results: Mean values and bias did not show significant differences between pH, pCO2, pO2, Na+, K+, Cl- and glucose when measured with OMNI C and GEM. Lower values of hematocrit and hemoglobin were measured with GEM in comparison to OMNIC (Hct: R2 0.90; bias6.3; Hb: R2 0.89; bias29.3), with microfuge method for Hct (R2 0.94; bias 4.3) and spectrophotometric method for Hb, measured by Sysmex SF3000 (R2 0.73; bias40.2) in the arterial samples obtained during CPB. Conclusion: GEM measures the Hct by electric conductance technique, and the Hb is calculated from the measured Hct. Plasma impedance varies depending on alterations in ion strength, protein concentration, and colloids during the CPB, and the lower values of Hb and Hct might result in reluctant transfusions of erythrocyte concentrates.
1
T326 OPTIMIZATION OF AN INTACT PTH ASSAY FOR USE IN MONITORING THYROIDECTOMY PATIENTS. R. Djiana, D. Thibeault, D. Blank McGill University Health Centre - Royal Victoria Hospital Background: Determination of parathyroid hormone (PTH) is performed as an aid in diagnosing and monitoring bone disease. It is also of clinical relevance in assessing calcium disorders, and in managing patients undergoing thyroidectomy or parathyroidectomy. There are many assays for PTH; but because of a lack of standardization, low PTH levels frequently seen after surgery and the heterogeneity of circulating PTH entities, a careful method evaluation is warranted. The aim of this study was to evaluate the technical performance and the clinical validity of a routine intact PTH immunoassay. Methods: The evaluation was performed on two Beckman Coulter Access immunosystems, LXi 725 and DxI 800, (LXi/DXi). The imprecision study (within-run and between-day), was performed using quality control materials and patient plasma pools at two different concentrations. The correlation studies involved a series of samples selected to cover the analytical range, with an emphasis on samples at the lower end tested on the LXi/DXi and an Elecsys 2010 (Roche Diagnostics). Data was analyzed by Bland-Altman difference plot and Passing/Bablok regression analysis. Results: Imprecision studies yielded within-run and betweenday coefficients of variation of 3.2% -4.8% and 9.0% -16%, respectively at 1.9, 11 and 15 pmol/L. Using patient plasma pools, the total imprecision was respectively 15% and 10% for the around medical decision thresholds established at 1.8 and 3.0 pmol/L. LXi/DXi PTH results correlated variably with Elecsys PTH (r2 ~0.780) over the analytical range. The unweighted linear regression analysis yielded slopes of 0.793 1.026 and intercepts of -0.25 to 0.21 pmol/L. The mean difference for surgery patients was -0.25 to -0.18 pmol/L. Despite LXi/DXi and Elecsys Intact PTH assays are all one step sandwich chemiluminescence-immunoassays, correlation deviations were expected due to differences in immunoassay standardization. Conclusions: Our data suggest LXi/DXi intact PTH gives information regarding medical decision thresholds that are similar to that of the Elecsys 2010 assay, supporting the applicability of the LXi/DXi assay to monitor thyroidectomy and parathyroidectomy patients.
S454

T327 VITAMIN D3 A COMPLEX ANALYTE, SHIFTING FROM HPLC TO AN AUTOMATED METHOD B.P. Dos-Santos, R. Souto-Fernndez, F. FerndezRodrguez A Coruna University Hospital Complex Background: Vitamin D3 (VitD3) has become a household name in the clinical laboratory, but that doesnt mean the dust has setle on the debate as to which analite to use for patient management (D3, D2 or total) or which method to use to determine its value; analytical method choices ranges from the most complex in terms of time and lab equipment as RIA and HPLC, to the most straightforward molecular immunology methods where a broad range of automated solutions are comercialy available. As more and more determinations of VitD3 are requested from laboratories the need to shift from more time consuming methods like HPLC to automated ones with lateral processing capacities is difficult to evade. Methods: We conducted a method comparisong between Alligens Vitmaine D2/D3 HPLC vs. Siemens Centaur Vitamine D, Roches Elecsys Vitamine D and DiaSorins Liaison Vitamin D. Following the guidelines outline by the Clinical and Laboratory Standards Institute (CLSI) in 2002, that is we selected 40 patients serum samples with known prior values covering a broad range, but still within the values regularly observed in tested population, we then randomized the samples into four series, and processed them in two runs in parallel on every instrument. A conventional linear regresion plus the statisitcs recomended in the CLSI guideline was performed using MS Excel. Results: Linnear regression: Centaur y = 0.3823x + 5.6624 R = 0.4605 , Elecsys y =0.7595x - 4.9448 R =0.7569, y = y = 0.7295x - 5.222 R =0.7102, intra-series differences were assessed obtaining Centaur Tle 61,4, Elecsys Tle 50,06, and Liaison Tle 53,34, that is to say there were no signinificant intraseries differences, The acceptability limit for intermethod comparisons was for the Centaur DX 0,18, Elecsys DX 0,25 and for the Liaison 4,0, under those conditions no sample was rejected, and the linnear regresion data were accepted. Conclusion: The goodness of fit of the automated assays evaluated were not optimal but a trend is easly decirnible where the Elecsys is superior to the Liaison which in term is superior to the Centaur; only three methods were assessed, more are commercially available and laboratories should carefully consider which to choose to incorporate to their instruments and methods T328 DEVELOPMENT OF A RAPID AUTOMATED SPE PROCEDURE FOR THE MEASUREMENT OF PLASMA COTININE BY LC-MS/MS A. Dunlop, I. Clunie, D. Stephen, J. Allison Department of Clinical Biochemistry, NHS Grampian Background: Cotinine is the primary metabolite of nicotine and the preferred biomarker for assessing cigarette smoke exposure. Several liquid chromatography-tandem mass spectrometry (LC-MS/MS) methods have been described for measuring cotinine in biological fluids. Such analyses often require lengthy sample preparation protocols involving protein precipitation and solvent evaporation. We describe a novel LCMS/MS method for the measurement of cotinine using a simplified solid-phase extraction (SPE) procedure. Methods: Extraction of cotinine from plasma samples was carried out by an HTS-PAL autosampler robot, using disposable SPE cartridges. Cotinine was quantified by LCMS/MS with electrospray ionisation (ESI) using multiple reaction monitoring (MRM) and cotinine-d3 as internal standard. Results: The assay was linear over the analytical range 0.0001 mg/L 1 mg/L. Limits of detection and quantification were 0.14 ng/mL and 0.23 ng/mL, respectively. Intra- and inter-assay imprecision of cotinine in all samples was <5% relative standard deviation. The analytical recovery of cotinine spiked into plasma was >95%. Conclusion: We have developed and validated a rapid, sensitive and specific LC-MS/MS method for the determination of cotinine in plasma, using a straightforward automated SPE protocol. This method can be applied to routine monitoring of smoking cessation in both the clinical and forensic environment (e.g. psychiatric patients within the secure forensic unit).
S455

T329 STUDY COMPARING TWO CHEMISTRY ANALYZERS S. Esteve Poblador, C. Valldecabres rtiz, E. Prez Gmir, S. Grriz Pintado rea de Diagnstico Biolgico. Hospital Universitario La Ribera, Alzira, Espaa Background: When a measurement procedure replaces another, you have to prove it estimating the systematic error (SE). The aim of this study was to check the interchangeability of two procedures. Methods: The study was performed for 5 days, and 15 duplicate samples were processed. The reference procedure was a PPE Modular (Roche) and the evaluated analyzer was an Advia 1800 (Siemens). The serum parameters evaluated were: urea, creatinine, total bilirubin (TB), direct bilirubin (DB), total cholesterol (TC), triglycerides, HDL-cholesterol (HDL), uric acid (UA), phosphorus, calcium, magnesium, albumin, total protein (TP), sodium, potassium, chloride, aspartateaminotransferase (AST), alanine-aminotransferase (ALT), alkaline-phosphatase (ALP), gamma-glutamyltransferase (GGT), amylase, creatine-kinase (CK), lactate-dehydrogenase (LDH), cholinesterase, iron, glucose, transferrin, C-reactive protein (CRP), antistreptolysin O (ASO), rheumatoid factor (RF), LDL-cholesterol (LDL) and ferritin. The results were analyzed by: Analysis of differences. The differences (D) and relative percentage differences (RD) between procedures were calculated, describing themselves as the mean (Dm or RDm) and standard deviation (SD o SRD). The confidence interval (95% CI) of the mean were calculated. Linear regression. The results of both methods were represented, obtaining the slope (b), intercept (a) and 95% CI. Results: Analysis of differences. The systematic difference was constant for chloride (95% CI (Dm) did not include 0), and systematic difference was proportional for chloride and cholinesterase (95% CI (RDm) did not include 0). Linear Regression. The systematic difference was constant for DB, TC, HDL, phosphate, albumin, TP, chloride, AST, ALT, CK and CRP (a >0), and for creatinine, TB, UA, ALP, amylase, LDH, cholinesterase, iron, transferrin and ASO (a <0). Showed a systematic proportional difference, TB, triglycerides, UA, potassium, ALT, ALP, GGT, CK, LDH, cholinesterase, iron, ASO and glucose (b >1), and creatinine, TC, albumin, TP, AST, transferrin and RF ( b <1). Conclusions: In order to correct the proportional SE it is necessary to review the calibration process and also the reference values, especially for chloride and cholinesterase. T330 COMPARISON OF THE PT/INR ASSAY WITH USING TWO DIFFERENT REAGENT:QUIKCOAG PT AND TECHNOPLASTIN HIS ON CEVERON ALPHA COAGULATION ANALYZER A.A. Etem, S. Mizrak, G. Can, A. Salman, S. Unal Comparison methods, Behets disease, vitiligo disease Objective: Prothrombin time (PT) is a coagulation parameter that reflects the activity of the extrinsic coagulation pathway (factors VII, X, V and II, and fibrinogen), starting from tissue factor-induced activation of factor VII and ending in fibrin formation. Our purpose to compare PT/INR values with using BioMedica QuikCoag PT and Technoclone Technoplastin HIS (heparininsensitive) commercially available kits on Ceveron alpha coagulation analyzer (Austria). Materials and methods: The study was conducted in a large, multi-specialty general hospital (Usak State Hospital). We compared 78 patients plasma PT/INR values with using BioMedica QuikCoag PT Reagent (Canada) and Technoclone Technoplastin HIS PT Reagent (Austria) commercial kits on Ceveron alpha automated coagulation analyzer (Austria). 2,7mL venous blood samples were simultaneously collected from each subject with BD vacutainer tubes (0.109M sodium citrate) and centrifuged at 2,500xg for 15 min. The PT-INR was measured immediately in the laboratory with using both kits in consecutively on Ceveron alpha coagulation analyzer. Results from the BioMedica and Technoclone kits were compared using the two-sample t-test. p value <0.05 was considered to indicate statistical significance. Results: A total of 78 patients mean age was 5121 ( 32 male 46 female). The mean level of PT 14,775,80 with Technoplastin HIS and 20,6612,19 with QuikCoag PT. The mean level of INR 1,420,65 with Technoplastin HIS and 1,67 1,16 with QuikCoag PT. We obtained higher level of PT/INR test results with QuikCoag PT than Technoplastin HIS PT. We found statistically significant difference between PT and INR results with using two-sample t-test (respectively P=0,000 and P=0,000). Conclusion: The results obtained using BioMedica reagents are not equivalent to those obtained using Technoclone reagents.
S456

T331 EVALUATION OF PANACLEAR MMP-3 LATEX, A REAGENT FOR HITACHI LABOSPECT 008 E. Hamada, M. Maekawa Department of Laboratory Medicine, Hamamatsu University School of Medicine Background: MMP-3 (matrix metalloproteinase 3) is an enzyme produced in the synovial cells and chondrocytes of joints and is possibly most closely associated with collagen degradation and cartilage destruction. In recent years, reagents applicable to ELISA and automated analyzers have been developed and used frequently in the diagnosis of rheumatoid arthritis, prediction of responses of this disease for treatment and evaluation of the disease activity. An improved reagent with latex turbidimetric immunoassay method for LABOSPECT 008 has recently been launched and its basic performance was evaluated. Method: The reagent PANACLEAR MMP-3 Latex (Sekisui Medical Co., Ltd.) and the clinical chemistry analyzer LABOSPECT 008 (Hitachi) were used for the following evaluations: (1) reproducibility (2) dilution linearity (2 level of MMP-3 samples were diluted with physiological saline) and limit of detection (2.6SD method), (3) interference (evaluated with Interference Check (Sysmex) and ascorbic acid), (4) correlation (with an existing reagent on JCA-BM1650 (JEOL Ltd.) in combination), and (5) probe contamination test (evaluated on 42 parameters of the same pre-installed module). Results: (1) Within-run reproducibility of CV (n=20) :1.35% (Control L mean 110.9 ng/mL), 0.78% (Control H mean 435.4 ng/mL ), 1.62% (pooled serum mean 99.2 ng/mL) (2) Linearity was up to 1500 ng/mL and limit of detection was 9.85 ng/mL. (3) No influences were observed by bilirubin <200 mg/L, hemoglobin <5 g/L, RF < 550 U/mL or ascorbic acid <500 mg/L in sample. (4) Good correlation with the existing reagent was noted, with the regression formula being y = 0.991x13.9 and the correlation coefficient being 0.985 (N=92). (5) Probe contamination test: No influence from contamination was noted on any of the 42 parameters. Conclusion: Improved PANACLEAR MMP-3 Latex with LABOSPECT 008 was shown in the present study as having favorable performance. The improved reagent will be useful for measuring of MMP-3 in clinical laboratories. T332 EVALUATION OF A SINGLE LATEX-ENHANCED ASSAY FOR MEASURING COMBINED SERUM FREE LIGHT CHAINS IN CLINICALLY RELEVANT SAMPLES ON THE SPA PLUS TURBIDIMETRIC ANALYSER J. Faint, L. Assi, C. Buckley, A. Mushtaq, P. Showell, S. Harding The Binding Site Group Ltd., Birmingham Background: Recent studies indicate a relationship between diseases associated with B cell activation, renal dysfunction and inflammation and elevated levels of combined and polyclonal serum free light chains (cFLC). Furthermore, cFLC concentrations have been linked to adverse outcomes in both normal, chronic kidney disease and systemic lupus erythematosus (SLE) populations. Here we describe Combylite, a single, latex-enhanced, turbidimetric assay for the measurement of cFLC and its comparison to summation of results from individual and assays (Freelite) Methods: cFLC levels were measured in presentation and follow up sera from patients with a variety of disorders using Combylite on a SPA PLUS analyser (The Binding Site Group Ltd, TBS). The assay utilises polyclonal antibodies specific for and free light chains, and the results were compared with summated and . Monoclonal samples with abnormal / ratios (0.26-1.65) were excluded from the analysis. Results: The measuring range was 6.25-200 mg/L with a sensitivity of 0.625mg/L. Inter-assay CVs for samples with high (169.66 mg/L), intermediate (53.98 mg/L) and low (12.23 mg/L) cFLC concentrations were 5.5%, 5.5% and 14.4%, respectively. The assay was linear over a range of 6.05-223 mg/L with a regression of measured against expected values of y=1.03x-3.4, R2=0.99. Comparison was made with summated FLC values by measuring samples from controls (n=193, range 6.7-93.9 mg/L) and patients with SLE (n=372, 7.1-164 mg/L), rheumatoid arthritis (n=332, 8.5-108.1 mg/L), lymphoma (n=37, 6.2-252.6 mg/L), liver disease (n=361, 5.4337.4 mg/L), cardiovascular disease (n=406, 8.2-284 mg/L), and CKD (n=515, 20.8-259.1 mg/L). Passing-Bablok fit was y=1.02x-2.39 and the linear fit was y=0.98x-0.91 with an R2 of 0.94. Conclusions: Measuring cFLC levels may provide a hitherto unavailable insight into B cell activation. Combylite is a single, rapid and reproducible immunoassay and further work is required to establish its utility in a variety of disorders.
S457

T333 COMPARATIVE STUDIES OF THE ACCESS HCV AB PLUS ASSAY D. Nogues(1), A. David(3), R. Falcou Briatte(2) Bio-Rad, Steenvoorde, France 2 Bio-Rad, Marnes la coquette, France 3 Bio-Rad, Steenvoorde, France Background: The aim of this study was to assess the performance of the Access HCV Ab PLUS (Bio-Rad) on the Beckman Coulter Immunoassay UniCel DxI 800 system (Beckman Coulter Inc.) in terms of specificity and sensitivity using serum and plasma samples. All results were compared to Elecsys Anti-HCV II assay (Roche Diagnostics) and ADVIA Centaur HCV assay Siemens). Methods: Comparison studies were performed on UniCel DxI 800 system, MODULAR ANALYTICS E170 system (Roche Diagnostics) and ADVIA Centaur XP system (Siemens). 500 non-selected serum samples from a routine laboratory, 3 commercial HCV seroconversion panels and one anti-HCV low titer performance panel were used on each system for specificity and sensitivity testing. Results: 488 of the 500 non selected samples were true negative. The specificity was 99.59% (95% CI: 98.53-99.95%) for all assays. The 2 false positive samples were different on UniCel DxI 800, MODULAR E 170 and AVDIA Centaur XP. The concordance with MODULAR and ADVIA Centaur was 99.20%. The clinical sensitivity from 12 positive samples was 100% for all assays. Using 3 commercial seroconversion panels plus one anti-HCV low titer performance panel, Access HCV Ab PLUS and Elecsys Anti-HCV II assays showed good performance and were more sensitive than the ADVIA Centaur HCV assay. Conclusions: The performance of the Access HCV Ab PLUS assay on the UniCel DxI 800 immunoassay system was excellent in terms of specificity and sensitivity. The specificity and the clinical sensitivity were equivalent for the 3 assays. The seroconversion sensitivity was better on UniCel DxI and Modular than on ADVIA Centaur. Adapted for high throughput routine testing, the Access HCV Ab PLUS assay performed on UniCel DxI 800 immunoassay system is fully suited for the screening of HCV infection in diagnostic laboratories.
1
T334 LUMIPULSE G1200, AN AUTOMATED CLEIA PLATFORM SHOWING GOOD PERFORMANCE FOR DIFFERENT TUMOUR MARKERS R. Falzarano, V. Viggiani, S. Michienzi, S. Tudini, F. Longo, L. Frati, E. Anastasi Department Of Molecular Medicine, Sapienza University of Rome, Italy Background: Tumour markers are commonly used to detect a relapse of disease in oncologic patients during follow-up. It is important to evaluate new assay systems for a better and more precise assessment, as a standardized method is currently lacking. Aim of this study was to assess the concordance between an automated chemiluminescent enzyme immunoassay system (LUMIPULSE G1200) and our reference methods using 6 tumour markers. Materials: Serum samples from 821 oncologic patients, representing a variety of diagnoses, were analyzed using LUMIPULSE G1200 and our reference methods. Serum values were measured for the following analytes: PSA, AFP, CEA, CA125, CA15-3, and CYFRA. For the determination of CEA, AFP and PSA, an automatic analyzer based on chemiluminescence (Access2) was applied as reference method. To assess CYFRA, CA125, and CA15-3, an immunoradiometric manual system was used (CisBio). Results: The concordance between LUMIPULSE G1200 and both reference methods was as follows: PSA 97%; AFP 96%; CEA 95%; CA15-3 91%; CYFRA 96%. A lower concordance (82%) was found when CA125 was assessed. In this group, 9 samples showed high disagreement between LUMIPULSE G1200 (CLEIA) and the immunoradiometric manual system. After diluting the samples and retesting with the reference method, a higher concordance (98%) with undiluted LUMIPULSE G1200 values was obtained. These data demonstrate the presence of the hook effect. The precision of each assay was assessed by testing 6 serum samples. Each sample was analyzed for all tumour biomarkers in duplicate and in three different runs. The coefficients of variation (CVs) were less than 6.3% and 6.2% for within-run and between-run variation respectively. Conclusions: Our data suggest an overall good agreement between all methods. However, some artifacts were obtained with immunoradiometric system and suggest the presence of an artifact secondary to the hook effect. CLEIA automated assay showed a good reliability in all samples.
S458

T335 ISO LIKELIHOOD RATIO LINES: A TOOL FOR THE INTERPRETATION OF RECEIVER OPERATING CHARACTERISTIC CURVES J. Faro-Viana Servio de Patologia Clnica, Centro Hospitalar de Lisboa Ocidental EPE, Portugal Background: Receiver Operating Characteristic curves (ROC) are widely used as a measure of the accuracy of diagnostic tests. However, their interpretation in terms of post-test probability is not straightforward. To ease that interpretation we present a simple tool, based on Likelihood Ratios (LR). Introduction: ROCs are functions of the Sensitivity (Se) and Specificity (Sp), calculated for all the possible cut-off points. LRs are the ratios of the probabilities of a test result in a specified diseased population versus a non-diseased population. Each LR value is determined by Se and Sp pairs linearly related through a straight line that we propose to call Iso Likelihood Ratio line (isoLR). In dichotomous tests, chosen cut-off points define the positive and negative results, whose LRs are respectively called Positive Likelihood Ratio (LR+) and Negative Likelihood Ratio (LR-), both functions of the Se and Sp at the chosen cut-offs. It follows that it is possible to plot isoLRs in ROC graphs. Methods: Plot of isoLRs in ROC graphs. Results: The isoLRs for the LR+ and LR- are represented as straight lines passing through points 0,0 and 1,1 respectively, with slopes equal to their values. They define areas for LR+s higher or equal and LR- lower or equal then their respective values. The intersections of the ROC with the isoLR+ and isoLR- define cut-off points for the positive and negative tests respectively. The part of the ROC that will eventually fall between these points corresponds to a gray zone. For reference, we plot by default the IsoLR corresponding to LR values of 10, 4, 2 for LR+ and to their reciprocals (1/10, 1/4 and 1/2) for LR-. According to the specific clinical use of the tests, other LRs can be chosen and the corresponding isoLR are easily plotted. Conclusions: In ROC graphs, isoLRs lines are very easy to plot and allow the definition of cutoffs and areas with specified LR values. These areas can be directly interpreted in terms of posttest probability, using the respective LRs and pre-test probabilities. IsoLRs can also be useful in method comparison studies, if common LRs are established to standardize the interpretation. T336 RAPID DETERMINATION OF BILIRUBIN IN BIOLOGICAL SAMPLES BY COLORIMETRICSOLID-PHASE EXTRACTION-FIBER OPTIC REFLECTANCE SPECTROSCOPY H. Filik Istanbul University Background: Bilirubin is a product of red blood cell breakdown. Normally, bilirubin is carried in the blood and passes into your liver, where its removed and becomes part of bile. Bilirubin is not normally present in the urine. Only conjugated bilirubin is excreted into the urine and normally only trace amounts can be detected in urine. Normal bilirubin levels are less than 1.0 mg/dL. Bilirubin in your urine may indicate liver damage or disease. Analysis of urine is an important nursing procedure. It is relatively cheap and requires minimal amount of training. However, it gives vital objective information about the patients internal functioning. Method: A new and selective colorimetric solid-phase extraction (C-SPE) procedure for the determination of Bilirubin in urine samples was proposed. The solid-state sensor is based on the reaction of bilirubin or mesobilirubin in presence of an oxidizing agent to convert the blue color and the analytes in samples were extracted onto a solid sorbent matrix and then quantified directly on the adsorbent surface by using a miniature reflectance spectrometer. An oxidation method is generally considered to be more accurate than a diazo method for the determination of bilirubin in urine. The measurements were carried out at a wavelength of 642 nm since it yielded the largest divergence different in reflectance spectra before and after reaction with the bilirubin. Results: Under the optimum experimental conditions, a linear calibration curve for bilirubin was obtained over the 0.125.85 g L-1 concentration range studied and the limit of detection was 0.1 g mL-1. There were no significant differences between RSD (%) values for intra-day and inter-day precision, which indicates the method was reproducible. The proposed method was applied to the determination of BLB in urine samples with a recovery for the spiked samples in the range of 96-102%. Conclusions: The proposed sensor was applied for the detection of bilirubin in the urine sample and satisfactory results were obtained. The above observations showed that the developed sensor might be promising in the detection of bilirubin in clinical samples. The sensor is small and of lowcost. Furthermore, it can be made into a portable instrument for in situ measurements.
S459

T337 EVALUATION OF THE HLC-723GX ANALYSER (TOSOH BIOSCIENCE) FOR THE DETERMINATION OF HBA1C M. Fonfrede(2), J. Couturier(2), J. Marlet(2) Biochemical laboratory, Pitie Salpetriere Hospital Background: HbA1c is the commonest marker for monitoring diabetes mellitus and treatment could be modified according to the results. Since a few years ADA and WHO proposed it as a test for the diagnosis of type 2 diabetes mellitus. For these 2 applications accuracy is necessary. When a new device is launched it is necessary to evaluate its features in order to be confident with the results.We evaluated the latest generation analyser of the Tosoh family: the HLC-723GX (GX). GX is a compact analyser which separates the haemoglobin fractions using the cations exchange HPLC method. Methods: Intra and inter assay precision was evaluated with control specimens. Accuracy was evaluated with specimen from the European Reference Laboratory (ERL). Correlation was performed with patient specimens after they were tested for HbA1c. Interference of glucose, cyanate, acetate normally eluting in the fraction called labile A1c was assessed by overloading and/or by removal.The possibility to correctly separate the most common variants was evaluated. Results: The intra assay and inter assay precision was in agreement with the specifications of the manufacturer: for example, coefficients of variation (CV) in the inter assay protocol, were respectively 0.89% and 0.28% for respective values 5.28% (34 mmol/mol) and 10.11% (87 mmol/mol). The reproducibility was also tested with patients samples with variant haemoglobins. The results were for HbS: CV= 0% for HbA1c at 5.6% (38 mmol/mol) and for HbC CV = 0% for an HbA1c at 4.8% (29 mmol/mol). The bias observed with specimens from ERL versus assigned values was 0.38%. We did not observe any interferences with stable HbA1c when labile A1c levels were below 5%, HbF below 45% or HbS below 35%. For the correlation to G8, 119 samples were tested. The regression line was GX = 0.99G8 +0.08. At least according to samples received every day, we observed that the most common haemoglobin variants were correctly separated. Conclusion: GX is a small analyser for the determination of HbA1c which presents all the features of a big one. This analyser allows laboratories with low test volumes to be confident in the results. We appreciate particularly the close to equal consumption of each buffer. T338 COMPARISON OF POLYCLONAL ANTIBODY-BASED ASSAYS FOR FREE LIGHT CHAIN KAPPA AND LAMBDA TO MONOCLONAL ANTIBODY-BASED ASSAYS F.J. Garca igo, S. Ocaa Lpez, M.L. Casas Losada, M. Morito Aguilar, S. Gutirrez Moreno, A. de Lzar de la Via, F. Cava Valenciano Hospital Universitario Fundacin Alcorcn Background: The measure of serum concentrations of free light chain (FLC) kappa and lambda is useful in diagnosis and follow up of plasma cell dyscrasia. Nowadays there are two different methods to determine them, both based on a nephelometric immunoassay, but they differ in the type of antibody used, one of them is a monoclonal antibody and the other is a polyclonal antibody. Methods: The study was performed with 49 serum samples from patients with monoclonal gammopathy. Freelite FLC Kit was compared against N Latex FLC, both using the nephelometric system Siemens BNII. Results: Correlations between both assays, using Spearmans rho, were 0.928, 0.901 and 0.923 for kappa, lambda and kappa/lambda ratio, respectively. In order to evaluate the concordance, we used Bland and Altmans graphic methods, this study showed a good concordance in low values (below 100), but this does not happen in the case of high values. Besides, we decide to categorize data in function of their reference intervals, so the estimated concordance indexes for kappa, lambda and kappa/lambda ratio were 0.829, 0.785 and 0.788 respectively. The distribution of conflicting cases for each parameter was symmetrical, using McNemar asymmetry test. Conclusions: The correlation between the two methods was good, and also the concordance with categorized data. For this reason both methods are reliable for the determination of free light chains in monoclonal gammopathies.
S460

T339 COMPARISON OF C REACTIVE PROTEIN ANALYSIS BY NEPHELOMETRY (BN II) AND TURBIDIMETRY (ADVIA 2400) V. Garcia Moreira, A. Fernandez Leivas, E. Fernandez Rodriguez Hospital De Cabuees Background: Ultrasensitive CRP kits that can quantify very low concentrations, are being increasingly used in the risk evaluation and prognostic value of heart diseases. Recently, a new CRP method, wide-range C-reactive protein (wrCRP), has been developed for ADVIA chemistry systems by Siemens to measure serum CRP levels in the low range of concentrations with good accuracy. The aim of this study is to compare the values of CRP concentration measured by two different methodology to assess the transferability of results: method 1BNII nephelometer hsCRP(Siemens)and method 2-wrCRP Advia2400 (Siemens). Material and methods: We analyzed 55 samples of serum by two methods: (y) hsCRP: immunonephelometry. With a detection limit of 2.97 mg/L. (x) wrCRP: enhanced immunoturbidimetric assay with particles of polyethylene glycol (PEG) coated with anti-CRP antibodies which agglutinate in the presence of CRP. Measure a wide concentration range, with a detection limit of 0.1 mg/L. Data were statistically analyzed using the MedCalc statistical package by Spearman correlation and Passing-Bablok regression. Significance was set at P <0.05. Results: There is a good correlation between both methods, with a coefficient of correlation r=0.981 (P <0.01). Median CRP concentrations for method 1(y) was 7.20 mg/L (range: 2.97285) and by method 2(x) was 9.21 mg/L (range:0.07-306). Passing-Bablok regression equation was y=0.721+2.06x (95%CIslope: 0.69-0.74; 95%CIintercept:2.03-2.45). There is constant and proportional error which generally yields higher concentrations with the wrCRP method and the results are not transferable. Is necessary to review the current reference values to match the new wrCRP method. Excluding values <3 mg/L (detection limit of hsCRP method), correlation improved significantly, finding a r=0.996 (P <0.0001). The new regression line was y=0.773x-0,148 (95%CIslope:0.753-0.839; 95%CIintercept:-1.01-0.245). The new relationship demonstrates that there is systematic proportional error but not constant. Conclusion: There is a good correlation between both methods. The turbidimetric wrCRP assay is a reasonable alternative to hsCRP nephelometric assay besides being cheaper and faster. T340 INFRARED ANALYSIS OF URINARY STONES: A VERY USEFUL TOOL FOR THE CLINICIANS. A. Primiano(1), S. Persichilli(1), A. Cocci(1), G. Gambaro(2), C. Zuppi(1), J. Gervasoni(1)
2 1
Laboratorio Analisi I Policlinico Gemelli Istituto di medicina interna e geriatria Policlinico Gemelli
Background: Kidney stone disease is a common illness with multifactorial etiopathogenesis. Methods for urinary stone analysis are classified in two main categories: semi-quantitative and quantitative. In our laboratory we use the semi-quantitative methods that are the most widespread into the laboratory for routine analysis. These last methods can only identify the presence of individual ions without differentiate mixtures and the results are operator dependent. The aim of this paper is to compare semi-quantitative method (DiaSys) with a quantitative method (FT-IR spectroscopy technique) for urinary stone analysis, in order to introduce in our laboratory a more reliable technique. Material and methods: We have analyzed 16 urinary stones arrived from Urology Department of our Hospital. The semiquantitative analyses were performed using Urinary Calculi Analysis kit (DiaSys) according to the manufacturers instructions. The same samples were analyzed by FT-IR using the Perkin Elmer Spectrum One FT-IR Spectrometer In this case each sample was prepared as follows: 3-5mg of the stone was powdered in a pestle, homogenized with 300 mg of potassium bromide (KBr) and converted into a pellet through a press. All FT-IR spectra of kidney stones were then computer matched against a library of spectra (NICODOM IR Library) so as to generate a precise report on the various stone components. Results and conclusions: A comparison of results obtained using the two methods from these 16 stones shows good overlapping of the results, in fact the main substances identified with the semi-quantitative method are also identified with the FT-IR technique. Instead, there are substantial differences on identification of substances present in trace amounts. in fact FT-IR technique shows a high sensitivity and permits to accurately identify stone composition. Identification of gallstones composition is essential for clinicians to find out the underlying cause of kidney stones and to decide whether to treat patients therapeutically or surgically. For these reasons, according to these preliminary tests, the introduction of the FTIR technique in clinical chemistry laboratory routine may be more responsive to clinicians expectations.
S461

T341 AUTOMATION AND CONSOLIDATION OF CLINICAL CHEMISTRY AND IMMUNOCHEMISTRY ASSAYS: THE EXPERIENCE OF AN ITALIAN HOSPITAL-BASED CLINICAL LABORATORY G. Gessoni, L. Penzo, S. Valverde Clinical Pathology Dept., Ospedale Madonna Della Navicella, Chioggia Background: In this study we evaluated change in efficiency and efficacy in our Laboratory after implementation of a total Laboratory automation for Clinical Chemistry and immunochemistry. Methods: We recently implemented a total laboratory automation for Clinical Chemistry and immunochemistry in our hospital-based clinical Laboratory: Thermo EnGen preanalytical automation with a track system connecting two Vitros Fusion 5.1 and two Tosoh AIA-2000 analyzers. Results: For Vitros Fusion 5.1 CV was <5% within-run and <6% inter-assay. For Tosoh AIA-2000 analyzers, within-run CVs were usually <7% and below <8% inter-assay. Functional sensitivity of ferritin was 5 mg/L with a CV of 18% and for glucose was 1.1 mmol/L with a CV of 12%. No drift effects were observed; neither there was any carry-over, due to samples or reagents. When evaluating the functionality of the whole automation system we observed a turn around time (TAT) 90% <10 min for tubes needing check-in and exit only. For tubes needing a check-in, centrifugation exit cycle, TAT 90% was <30 min. Tests for clinical chemistry and/or immunochemistry on-line analyzers showed a TAT 90% <105 minutes. Conclusions: We conclude that the adopted automation system effectively reduces the labor associated with specimen processing; decreases the number of laboratory errors that occur with specimen sorting, labeling, and aliquoting; and improves the integrity of specimen handling throughout the steps of specimen processing. T342 COMPARISON BETWEEN METHODS OF DIRECT AND CALCULATED LDL CHOLESTEROL S. Grriz Pintado, E. Prez Gmir, S. Esteve Poblador rea de Diagnstico Biolgico, Hospital Universitario La Ribera, Alzira, Spain Background: LDL cholesterol (LDL-C) is a known risk factor for cardiovascular disease. The reference method for measurement is ultracentrifugation. For the daily routine, an estimate is made using the Friedewald formula or direct measurement using automated methods. The aim of this study was to compare the direct measurement of LDL-C (dc-LDL) with the estimate LDL-C (cc-LDL) calculated. Methods: 1742 patients were studied. Measurements of total cholesterol (TC), HDL cholesterol (HDL-C), triglycerides (TG) and dc-LDL were performed in the autoanalyzer Advia 1800 (Siemens). The cc-LDL was calculated using the Friedewalds formula (=TC-HDL-(TG/5)). Patients were classified into five groups according to the TG values: 150, 151-200, 201-300, 301-400 and >400 mg/dL. The mean and standard deviation of biochemical parameters, and the percentage of patients with cardiovascular risk (LDL-C 130 mg/dL) were calculated. Students-t-test was achieved for statistical data analysis. Results: In relation to the concentration of TG we observed that dc-LDL and cc-LDL presented similar results. For TG 150 mg/dL was 11031 (dc-LDL) and 11433 (cc-LDL); for TG between 151-200 mg/dL was 11835 (dc-LDL) and 11837 (ccLDL); for TG between 201-300 mg/dL was 12234 (dc-LDL) and 11638 (cc-LDL); and for TG between 301-400 mg/dL was 11733 (dc-LDL) and 10435 (cc-LDL). The percentage of individuals with risk of cardiovascular disease in the general group was 30% (dc-LDL) and 31% (cc-LDL). If TG 150 was 27% (dc-LDL) and 29% (cc-LDL); TG: 151-200, 36% (dc-LDL) and 35% (cc-LDL); TG: 201-300, 35 % (dc-LDL and cc-LDL); TG 301-400, 38% (dc-LDL) and 15% (cc-LDL); and TG >400, 32% (dc-LDL) and 10% (cc -LDL). The correlation between dcLDL and cc-LDL in the general group was r2=0.844. For TG 150, 151-200, 201-300, 301-400 and >400, was 0.866; 0.848; 0.837; 0.803 and 0.696, respectively. Conclusions: For patients with TG <400 mg/dL the dc-LDL and cc-LDL methods presented similar results. In hypertriglyceric patients (TG >400 mg/dL) the correlation is weaker. The dcLDL method doesnt appear to offer any advantages over the cc-LDL, so from a practical and economic point of view, we consider best to use cc-LDL in patients with TG <400 mg/dL.
S462

T343 OPTIMIZED ROOM TEMPERATURE STABLE, READY-TO USE, ONE-STEP RT-PCR MIX A. Moiana, L.J. Turner, L. Ventura, M. Gramegna Sentinel CH. SpA, Milan, Italy Background: PCR Mixes, as well as enzymes and amplification mixes components, are usually stored at a -20 or +2/8C. As a consequence, the reproducibility of an assay prepared from freezed and thawed material is critical. The aim of the present work is the achievement of a freeze-dried mix to perform, in a single step, an amplification starting from a RNA template in a ready-to-use, pre-dispensed, flexible and room-temperature storable format. Methods: Each test tube of the one-step RT-PCR Mix contains: reaction buffer, dNTPs, MgCl2, Reverse Transcriptase, Hot Start DNA Polymerase, preservatives and stabilizers, in a freeze-dried form. Performances were evaluated on a 7500 Real-time PCR System (Applied Biosystems) by TaqMan Ribosomal RNA Control Reagents (Life Technologies). The thermocycling protocol was: 48 C 30, 1 cycle; 95 C 10, 1 cycle; 95 C 15, 60 C 1, 40 cycles. For the freeze-drying process an epsilon 2-12D freeze-dryer (Martin Christ) was used. Accelerated and real-time stability studied were conducted according to the internationally accepted guidelines. Results: An optimized one-step RT-PCR mix, evaluating different compositions was developed. The formulation showed good performances; a Ct of about 20 with 10 pg of Human Raji RNA was obtained. The performances before and after freezedrying were evaluated. The lyophilization procedure does not impact on mix functionality. Cts before and after freeze-drying were comparable, 19.87 and 19.88 respectively. Preliminary accelerated stability studies data allows to assign to this product at present a shelf-life of 10 months at room temperature. Real-time stability studies, are still ongoing. Conclusions: The one-step RT-PCR Mix is a new ready-to-use and room-temperature storable mix useful for efficient one-step RT-PCR. The format (one vial/one test) and the room temperature storage allow the reduction of the time needed for preparation of the amplification mix, and the reduction of risk of contamination. Moreover the dried format permits large flexibility in the volume of template added to the mix. This onestep RT-PCR Mix is universal, applicable to different fields, gene expression, virology, etc. This mix is a valuable tool for the development of molecular diagnostic tests. T344 INTERLABORATORY COMPARISON OF METHODS FOR 34 DIFFERENT ROUTINE BIOCHEMICAL PARAMETERS ON THE BECKMAN COULTER AU5800, ROCHE COBAS 8000, ROCHE COBAS 6000, ROCHE MODULAR PE, OR THE SIEMENS ADVIA 2400 ANALYTICAL SYSTEMS O. Gressner, U. Schallenberg, C. Knauff, F. Wisplinghoff laboratoriumsmedizin kln, Dres. med. Wisplinghoff and Collegues Background: A lack of reproducibility of results obtained in medical laboratories using different methods is still a major health-economic burden and not infrequently hinders the continuous and efficient treatment of the patient. We therefore performed an interlaboratory comparison of routine biochemical parameters between 4 different laboratories using different methods of analysis. Methods: 998 anonymized residues of serum samples submitted to our laboratory were analyzed for the concentrations/ enzyme activities of 34 routine biochemical parameters. Serum samples were selected to obtain groups with 6 different ranges of concentrations per parameter, each group containing a maximum of 20 samples. Samples were split, stored in aliquots at -20 C until measurement, and then analyzed in parallel in 5 different laboratories, using either the Beckman Coulter AU 5800, Roche Cobas 8000, Roche Cobas 6000, Roche Modular PE, or the Siemens ADVIA 2400 analytical systems. Statistical analysis (linear regression analysis) of the results was performed using SPSS Statistics. Results: 23193 individual analyses in 998 samples were performed on each analyzer, actual sample throughput varied significantly between analyzers. Overall, the interlaboratory comparison of methods revealed an acceptable reproducibility, however, in some cases, coefficients of determination R2 varied significantly depending on the compared analyzers (e.g. for Creatine Kinase-MB, lipase). The best coefficient of determination were calculated for gamma-glutamyl transpeptidase. Conclusions: Despite some outliers on the low side our data demonstrate an overall acceptable agreement in the measured values between the tested laboratories and analytical systems.
S463

T345 GREAT IMPACT OF THE BRAND-NEW SYSMEX XNSERIES AUTOMATED HEMATOLOGY ANALYZER ON WORKFLOW EFFICIENCY AND CLINICAL UTILITY E. Hamada, M. Maekawa Department of Laboratory Medicine, Hamamatsu University School of Medicine Introduction: Recently, higher-quality testing environment has been increasingly required. To counteract these demands, Sysmex has introduced the XN-3000 automated hematology analyzer. Here we report our evaluation of usefulness of the XN-3000 and the integrated hematology testing system. Method: (1) Basic performance: We evaluated precision, linearity, correlation with the manual method. (2) Usefulness of new function: Low WBC mode, PLT-F (fluorescence) channel, WNR channel (real time NRBC count), Body Fluid mode, Automatic re-testing capability (3) Usability: We evaluated the usability of newly introduced functions of CNA-NET, which include automated smear preparation, selection of specific sample and sample requiring manual count. Results: (1) Basic performance: Precision was 0.42-5.21 (CV%), and linearity was confirmed up to WBC 378109/L, RBC 7.811012/L, HGB 235 g/L, HCT 72.1%, and PLT 1,691109/L. The correlation coefficient of WBC classification ratio with the manual method was 0.612-0.987. (2) In the evaluation using samples with WBC 0.5109/L, WBC precision was 1.29-4.17(CV%) for Low WBC mode, and the correlation with the manual method was 0.999. In the evaluation using samples with PLT 50109/L, the correlation coefficient between PLT-F and the manual method was 0.980. Usefulness of PLT-F can provide data with high accuracy and precision for in the case of giant and small platelets, and interference particles like fragment red cell and small RBC. Furthermore, PLT-F methods provide immature platelet fraction. WNR channel and Automatic re-testing capability reduced manual retesting and examine few manual procedure. Body Fluid mode was excellent. (3) Usability: In term of time for reporting was 50 94% reduction. The number of analyzer operator for 500 samples/ day decreased from 1.5 to 0.8 person. Conclusion: Basic performance and new function was satisfactory. It is possible to return efficient report in a short time. Use of the automated re-testing function and the CNANET made reporting time and processing time reduced. The newly introduced integrated hematology testing system improved the quality of the clinical laboratory testing. T346 MULTIPLEX DETERMINATION OF NINE SECOND-LINE ANTI-TUBERCULOSIS DRUGS BY UPLC-TANDEM MASS SPECTROMETRY M. Han(1), S.H. Jun(1), S.H. Song(2), J.H. Lee(3), K.U. Park(1), J. Song(1)
1 Department of Laboratory Medicine, Seoul National University Bundang Hospital, Gyeonggi-do, Korea 2 Department of Laboratory Medicine, Seoul National University Hospital, Seoul, Korea 3 Department of Internal Medicine, Seoul National University Bundang Hospital, Gyeonggi-do, Korea
Background: Therapeutic Drug monitoring (TDM) antituberculosis (TB) drug concentrations is helpful for patients that show a slow response to treatment and especially for multidrug-resistant cases. We developed a rapid method that simultaneously measures the blood concentrations of nine major second-line anti-TB drugs, including streptomycin, kanamycin, clarithromycin, cycloserine, moxifloxacin, levofloxacin, para-aminosalicylic acid, prothionamide, and linezolid using ultra-performance liquid chromatographytandem mass spectrometry (UPLC-MS/MS). Methods: Serum samples were extracted with acidified methanol followed by neutralization with NaOH. A HSS T3 column and gradients of ammonium formate and acetonitrile in 0.1% formic acid were used for UPLC separation. Drug concentrations were determined by multiple reaction monitoring in positive ion mode, and assay performance was evaluated. We applied the devised method to TDM of each drug by analyzing random serum samples from patients treated with second-line drugs (n=62). Results: The preparation of samples including acidified methanol extraction followed by neutralization steps gave a relatively good recovery and ionization efficiency, and chromatographic separation was achieved within 3 min/sample. Within-run and between-run imprecisions were 2.811.1% and 3.711.6%, respectively at the concentrations of low and high levels for each nine drug. The lower limits of detection and quantitation were 0.050.5 g/mL and 0.255.0 g/mL, respectively. Linearity was acceptable at 5 level concentrations for each drug. Evaluation of ion suppression showed no effects, except streptomycin, kanamycin and cycloserine with close elution times to the void volume of the column. Pilot application of devised method on the limited number of samples from the patients showed dispersed distribution of serum concentration of each drug. Conclusions: The performance of our devised techniques for MS/MS detection was generally acceptable. The devised method allows for rapid, sensitive, and reproducible quantification of nine second-line anti-TBs drugs and should be helpful for drug monitoring in TB treatment.
S464

T347 QUATIFICATION OF HUMAN SERUM HEPCIDIN - 25: EVALUTION OF AN ENZYME-LINKED IMMUNOSORBENT ASSAY VS MASS SPECTROMETRY C. Hansson(1), G. Dahlfors(1), P. Barany(2), P. Stl(3), J. Marmur(3), S. Beshara(1) Department of Clinical Chemistry, Karolinska University Hospital, SE-171 76 Stockholm, Sweden 2 Department of Renal Medicine, Karolinska University Hospital, SE-171 76 Stockholm, Sweden 3 Department of Gastro Center, Karolinska University Hospital, SE-171 76 Stockholm, Sweden Background: Hepcidin-25 is a central regulator in the iron metabolism, holding potential in the diagnosis of iron-related disorders. Mass spectrometry methods for Hepcidin measurement are referred to as reference method; however, there is an increasing clinical need for specific immunoassays. Methods: Serum samples from 68 healthy subjects (HS), 65 patients with liver-disorders (LD) and 225 hemodialysis patients (HD) were stored at -70 C until analysis using a commercial enzyme-linked immunosorbent assay (ELISA), (Bachem, St.Helens, UK). The study aimed at evaluating ELISA-kit performance in terms of precision and accuracy. For precision, we applied constructed controls of low and normal levels while for accuracy, the results were compared to measurements using LC-MS/MS. Results: Using ELISA, the results in HS showed ranges of 968 g/L and 2-44 g/L in males and females respectively; which are in accordance with published data. Intra- and inter-assay results showed CV of 18% and 13% for low and normal controls and 18% and 19% respectively. The overall correlation between ELISA results to those obtained by LC-MS/MS was good (r =0, 9). Correlation between Hepcidin-25 and ferritin in HS was also satisfactory (r =0, 87). Hepcidin-25 levels showed lower values in HS (males median 18 g/L) than LD (males median 31 g/L). HD had significantly higher results (males median 150 g/L). They were also more prone to processing errors with LC-MS/MS resulting in 29 samples failing to get a result. In addition, there was a higher variation of LC-MS/MS results at low levels. Females showed lower median values than males in all subgroups except hemochromatosis patients (males median 26 g/L, females median 29 g/L). Conclusion: The overall correlation between both methods was good, showing that the studied ELISA-kit is a reliable method for quantification of Hepcidin-25. Close follow-up of precision should be considered. Results from ELISA had a tendency to show higher levels than results from LC-MS/MS, which became more prominent at high Hepcidin levels. This demonstrates the necessity of establishing local reference ranges. Considering ELISAs capacity to readily be set up in most laboratories it should become the routine method for quantification of Hepcidin.
1
T348 FAST AND EFFICIENT SAMPLE PREPARATION OF TOXIN-PRODUCING CYANOBACTERIA FOR qPCR ANALYSIS H. Hautala, M. Vehniinen Division of Biotechnology, Department of Biochemistry and Food Chemistry, University of Turku Background: Toxic cyanobacterial blooms are a serious risk to public health worldwide. Quantitative polymerase chain reaction (qPCR) is a practical tool for risk assessment and toxicological studies related to toxic cyanobacteria. Sample preparation is an important part of qPCR analytics to obtain reliable results template loss during this step has to be minimized. We have developed an efficient method based on cell lysis by heating, which has been evaluated with four different cyanobacterial genera, including microcystin, saxitoxin and cylindrospermopsin producers. Methods: Ten toxic and non-toxic strains of cyanobacteria belonging to the genera Anabaena, Cylindrospermopsis, Microcystis and Planktothrix were used in this study. Templates for qPCR from equal amounts of cells were produced using two sample preparation methods: 1) heat-treatment (80 C, 10 min) of cells collected on fiberglass filters and 2) the standard phenol-chloroform extraction of DNA. Three qPCR methods were used to determine template yields; toxin-specific (mcyB for microcystin-producing cyanobacteria) and genus-specific (phycocyanin or RNA polymerase I genes for other toxin producers as well as non-toxic strains). Results: In all the studied strains, the same amount of cells yielded either an equal (4/10) or higher (6/10) amount of template when using the simple heat-treatment method compared to the phenol-chloroform method. The biggest differences (up to 200-fold) were observed for Microcystis and Planktothrix. The average time to produce template from one sample was 25 minutes for heat-treatment and 10 hours for phenol-chloroform extraction. Conclusions: The heat-treatment method improves template yields and allows for considerably faster sample preparation than phenol-chloroform-based DNA extraction. The number of steps needed to carry out the heat-treatment is minimal, and is likely to contribute to the increased template yields. The method is easy to use also in field conditions, and is well suited to both small and large scale studies, reducing the time needed to prepare samples for analysis.
S465

T349 EVALUATION OF THE SERUM FREE LIGHT CHAIN (FLC) FREELITE ASSAY ON THE NEW SPAPLUS IMMUNOTURBIDIMETRY ANALYZER C. Henig, T. Adler, S. Ammouri, M. Barak Regional Laboratory, Haifa and Western Galilee, Clalit Health Services, Israel Background: Serum free light chain (FLC) assay has become a routine test available on a variety of chemistry analyzers. The International Myeloma Working Group (IMWG) has recommended to use this assay for monitoring oligosecretory myeloma and AL amyloidosis patients. Recent publications have shown several application limitations which may stem from the polyclonal nature of the assay antibodies. The assay lack of standardization may explain the non satisfactory harmonization between methods and should emphasis the importance of monitoring the patients by the same method. Methods: We evaluated the performance of the serum FLC assay on the SPAplus immunoturbidimetry analyzer and compared the results to a Beckman-Coulter AU analyzer used in our lab. Results: We analyzed 62 sera from out patient clinics and 3 external quality control materials. Regression analysis revealed good correlation between the two assays (kappa: y=0.91x-0.4, r=0.94, range: 3.6-112.9 mg/L, lambda: y=1.06x-1.5, r=0.98, range: 5.8-137.5 mg/L). Reproducibility varied from 1.6% to 4% and repeatability was less than 2.5% for both kappa and lambda fractions. Extremely high levels of free light chains could be correctly measured due to the SPAplus antigen excess detection ability. Serial dilutions showed mean recovery of 95% and 103% for the polyclonal and monoclonal kappa FLC. On the other hand, for the lambda FLC, increasing recovery was determined with increasing dilutions. Conclusion: Laboratory staff should be aware of the lambda FLC non linearity although its clinical relevance for patient monitoring is not clear. Nevertheless, the SPAplus immunoturbidimetry analyzer is as an acceptable platform for the serum free light chain (FLC) Freelite assays. T350 FIELD EVALUATION OF SPAPLUS SYSTEM FOR THE DETERMINATION OF FREE LIGHT CHAINS (FLC) IN SERUM I. Infusino, A. Dolci, M. Panteghini Clinical Biochemistry Laboratory, Luigi Sacco University Hospital, Milan, Italy Background: Measurements of serum immunoglobulin k and FLC and FLC ratio calculation are recommended for the evaluation of plasma cell disorders. Several practical issues for the analytical measurement of FLC have, however, been identified. Searching for a solution able to fulfil the performance goals for the effective use of FLC in clinical setting, we evaluated the suitability of SPAplus analyzer using Freelite reagents (both from The Binding Site) for FLC determination. Particularly, we compared the system performance with allowable goals for bias, imprecision (CV) and total error (TE) derived from biologic variation of FLC. Methods: We evaluated the performance of SPAplus FLC using data collected during a six-month time period of routine use, employing two different reagent lots. The two-level (N and H) liquid SPAplus control material was used for bias estimate by comparing the obtained long-term experimental means (n=34, both levels) with the corresponding assigned values. The protocol for CV evaluation employed the liquid-frozen Bio-Rad Liquichek Unassayed Chemistry Control, measured in each performed run for a total of 29 runs. Inaccuracy was checked by results from three UK NEQAS exercises (system-specific (SPAplus) consensus value as reference). Goals (desirable/minimum quality levels) for bias, CV and TE were 4.1%/6.1%, <4.0%/6.0% and 10.7%/16.1% for FLC and 7.1%/10.6%, <3.5%/5.3% and 12.9%/19.3% for FLC, respectively. In addition, CV and TE for FLC ratio should be <2.3%/3.4% and 7.7%/11.6%. Results: Average cumulative bias was -6.0% (control N) and 6.2% (control H) for FLC, and 4.3% (N) and 6.1% (H) for FLC, respectively. Overall CV resulted in 10.8% for FLC (mean 11.9 mg/L), 7.3% for FLC (mean 13.6mg/L) and 8.8% for FLC ratio (mean 0.9). On EQAS evaluation all FLC and two out of 3 results for FLC were within the minimum allowable TE, while the FLC ratio achieved the minimum goal only in one exercise. Conclusions: Considering our previous experience with other analytical systems, the SPAplus solution undoubtedly represents a significant step forward. A further improvement in measurement imprecision (priority) and method alignment is probably needed to fulfil the stringent analytical goals derived from biologic variation.
S466

T351 INTEGRATION OF AN AUTOMATED SAMPLE PREPARATION WORKSTATION FOR THE ANALYSIS OF IMMUNOSUPPRESSANT DRUGS BY LC-MS/MS R. Zhang(1), M. Jarvis(2)
1 2
Beckman Coulter, Indianapolis, IN, USA AB SCIEX, Concord, ON, Canada
T352 FULLY VALIDATED METHOD FOR RAPID AND SIMULTANEOUS MEASUREMENT OF SIX ANTIEPILEPTIC DRUGS IN SERUM AND PLASMA USING ULTRA-PERFORMANCE LIQUID CHROMATOGRAPHY-ELECTROSPRAY IONIZATION TANDEM MASS SPECTROMETRY J. Kuhn, C. Prante, P. Kuzaj, D. Hendig, C. Knabbe Institut fr Laboratoriums- und Transfusionsmedizin, Herzund Diabeteszentrum Nordrhein-Westfalen, Universittsklinik der Ruhr-Universitt Bochum, Oeynhausen, Germany Background: Therapeutic drug monitoring (TDM) of antiepileptic drugs is very useful to establish optimal therapy regimes for individual patients and to study the variation in pharmacokinetics that occurs between individuals. Therefore, the development of a specific and sensitive method allowing the fast measurement of these drugs is of great interest in TDM. Here, we describe a simple rapid assay to measure the antiepileptic drugs lacosamide, lamotrigine, levetiracetam, primidone, topiramate, and zonisamide. Methods: After the addition of internal standards (ISs) and protein precipitation of patient serum or plasma, 1 l of supernatant sample was injected onto an Acquity UPLC system which was directly coupled to a Waters TQ tandem mass spectrometer. The chromatographic separation was performed on a 2.1 X 50 mm reverse phase column (Waters, Acquity UPLC BEH Phenyl, 1.7 m). Elution occurred using a linear gradient of methanol and water, each containing 0.1% formic acid and 2 mmol/L ammonium acetate. Analytes were then ionized and detected by electrospray ionization mass spectrometry with multiple reaction monitoring. Runtime was 2.5 minutes per injection. Ion suppression was characterized using post-column infusion. Results: The calibration curves of the 6 antiepileptic drugs were linear over the working range between 0.05 and 50 mg/L (r >0.99). The limit of detection (LOD), as well as the lower limit of quantification (LLOQ) of all drugs measured in the assay was <0.05 mg/L. The intraassay and interassay coefficients of variation for all compounds were <15% for very low concentration (0.1 mg/L) and <8% in the clinically relevant concentration range (>1.0 mg/L). Mean recoveries were between 85.0 and 110.7% for all drugs. There were no significant ion suppressions detected at the elution times of the analytes. The mean differences between serum and heparinized plasma values were less than 6% for the 6 antiepileptic drugs. All drugs were stable in serum at -20 C, 4 C, and even at RT for at least 2 months. Conclusions: In summary, a specific and sensitive stable isotope dilution UPLC-MS/MS method was developed and validated for routine clinical monitoring of lacosamide, lamotrigine, levetiracetam, primidone, topiramate, and zonisamide.
Background: For Research Use Only. Not For Use In Diagnostic Procedures. Liquid chromatography-tandem mass spectrometry technology provides laboratories with a powerful tool for robust, accurate, sensitive detection of a wide variety of analytes. Method automation reduces the possibility of human error at many different stages, including preparation of calibration standards, sample preparation, and data processing. The objective of this work was the automation of an LC-MS/MS method for the analysis of the immunosuppressant drugs Tacrolimus, Cyclosporine A, Sirolimus, and Everolimus, to eliminate human error, increase reproducibility, eliminate subjectivity during data processing, and save time. Methods: An LC-MS/MS method for the analysis of immunosuppressant drugs was developed, making use of commercially available whole blood calibrators and controls. In addition to manual preparation, all steps of sample processing could be automated using a BioMek NXP platform. The sample preparation consisted of a simple protein precipitation using ZnSO4 solution. After centrifugation, the clear supernatant was injected directly onto the LC-MS/MS system. Samples were loaded in test tube format and the final samples were prepared in a 96-well plate format. The LC-MS/MS data acquisition, processing, and reporting were performed using the Cliquid software. Results: The reproducibility of the automated protocol versus manual protocol was assessed by preparing and analyzing replicates of each calibration standard. The measured %CVs were at least equivalent between protocols over the entire concentration range covered by the assay. The method displayed good linearity for all four immunosuppressant drugs, with R >0.999. Conclusions: An automated sample preparation protocol has been developed for the analysis of four immunosuppressant drugs by LC-MS/MS, with performance better than or equal to the equivalent manual sample preparation.
S467

T353 PERFORMANCE EVALUATION OF ELECTROLYTE MEASUREMENTS ON THERMO SCIENTIFIC INDIKO AND INDIKO PLUS CLINICAL CHEMISTRY ANALYZERS K. Kurki, S. Riistama-Laari, H. Lampinen Thermo Fisher Scientific, Vantaa, Finland Background: Electrolytes help to maintain exchanges of osmotic pressure and have a central role in maintaining the normal distribution of water. Certain diseases can change the rate. Sodium, potassium and chloride are part of a screening malfunction of electrolytes, monitoring the ionic balance (acid or basic) states of hydration. Besides ISE measurements, the Indiko analyzers serve routine clinical chemistry tests and various dedicated specialty testing needs offering a complete system solution with the throughput of 200 photometric and 120 ISE tests/hour for Indiko and 350 photometric and 135 ISE tests/hour for Indiko Plus. Methods: Electrolyte measurements in Indiko and Indiko Plus analyzers are made with ion selective electrodes (ISE) directly without any dilution of the sample. The measurement cell consists of sodium, potassium, chloride and reference electrodes. The measured potential between each ISE and the reference electrode is related to the natural logarithm of the ionic activity according to the Nernst equation. The changes in potential are developed across the ISE membrane / sample interface. Serum and Li-heparin plasma can be used as sample types. Calibrators used in the fully automated measurement are NIST traceable ready for use liquids. Results: The measuring range is for sodium 100 - 200 mmol/L, for potassium 2.0 - 10.0 mmol/L and for chloride 60 - 150 mmol/L. The observed CV% for repeatability (within run precision) was 0.1% - 0.2% for Na (n = 84), 0.2% for K (n=84) and 0.2% for Cl (n = 84). The observed CV% for within device (total precision) was 0.8% for Na, 1.0% - 1.1% for K and 0.8 % - 1.1% for Cl. The method comparison studies were performed using Thermo Scientific Konelab PRIME 60i direct ISE methods as a reference. Conclusion: The results demonstrate that ISE measurements can be done precisely and easily using Thermo Scientific Indiko and Indiko Plus clinical chemistry analyzers. Ion selective electrodes are maintenance free and they are easy to install. T354 EVALUATION OF THE THERMO SCIENTIFIC INDIKO PLUS CLINICAL CHEMISTRY ANALYZER K. Kurki, S. Riistama-Laari, H. Lampinen Thermo Fisher Scientific, Vantaa, Finland Background: Thermo Scientific Indiko Plus is a new bench top clinical chemistry analyzer, especially suitable for small and medium laboratories or as a back-up analyzer for bigger ones. It is intended for colorimetric and turbidimetric assays as well for electrolytes employing ISE technology. Throughput of the analyzer is maximum 350 photometric tests/hour. The Indiko Plus analyzer is a complete system including the instrument, system reagents, calibrators and controls as well the CE marked applications. Methods: In this study we have evaluated the performance of one substrate assay Glucose (HK), one enzyme assay AST (IFCC) and one specific protein assay Immunoglobulin G (IgG). We have performed precision studies and method comparison studies for all of methods. The precision studies were made with three different Glucose and AST concentrations and with two different IgG concentrations. The method comparison studies were performed using Thermo Scientific Indiko clinical chemistry analyzer as a reference system. Results: The observed CV% for repeatability (within run precision) was 0.5% - 0.7% for Glucose (n = 80), 0.3% - 0.8 % for AST (n = 20) and 1.2% -2.2% for IgG (n = 20). The observed CV% for within device (total precision) was 1.2% - 1.3% for Glucose, 1.2% - 1.6% for AST and 3.7% - 4.8% for IgG. In the method comparison studies the observed results showed excellent correlation for all the tested analytes between the evaluated and the existing routine analyzer. Conclusion: The results in total demonstrate that the Indiko Plus is a precise, easy to use and reliable analyzer for routine biochemistry tests.
S468

T355 NEW CK(IFCC)AND CK-MB ASSAYS FOR THERMO SCIENTIFIC INDIKO AND KONELAB CLINICAL CHEMISTRY ANALYZERS M. Kurki, R. Raita, P. Pitknen, S. Riistama-Laari, S. Tikanoja, H. Lampinen Thermo Fisher Scientific, Vantaa, Finland Background: Creatine kinase (CK) is an enzyme which consists of isoenzymes mainly of the muscle (CK-M) and the brain (CKB). CK exists in the human body in dimeric forms as CK-MM, CK-MB and CK-BB. CK activity is elevated in many diseases, including those involving skeletal muscle, the heart, the central nervous system and the thyroid. CK-MB is present in low concentration in normal serum but it increases as a result of heart injury. Measurement of CK is used especially in conjunction with CK-MB for diagnosis and monitoring of myocardial infarction. Methods: CK and CK-MB are two part liquid ready to use assays using the Thermo Scientific Indiko and Konelab clinical chemistry analyzers. CK is determined using IFCC recommended method. CK-MB reagent utilizes an immunoinhibition method employing monoclonal antibodies to the CK-M monomer. The activity of the non-inhibited CK-B monomer is assayed. The measuring range for CK is 10 1000 U/l, up to 9000 U/l with automatic dilution, and for CK-MB 51000 U/l, up to 3000 U/l with automatic dilution. Results: The repeatability (within-run precision) for CK is 0.61.8 % (CV; n=84), and for CK-MB 0.73.0 % (CV; n=84). The within device (total) precison for CK is 1.23.6 % (CV; n=84), and for CK-MB 2.05.4 % (CV; n=84). The methods were compared with the previous generation CK and CK-MB Konelab system methods (dry powder). Linear regression for CK was y (Indiko)=1.10x-4.7 and r=0.999, and y (Konelab)=1.07x-6.3, and r=0.999, n=95. Linear regression for CK-MB was y (Indiko)=0.99x-8.4 and r=0.996, and y (Konelab)=1.00x-10 and r=0.996, n=84. CK-MB results were slightly lower than previously probably because of the better immunoinhibition ability of the new method. On board stability time for both reagents is 21 days. Conclusion: New liquid form CK(IFCC) and CK-MB reagents have a longer on-board stability than the dry powder format reagents. With these ready to use system reagents, CK and CK-MB analysis on Thermo Scientific Indiko and Konelab analyzers is quick and accurate. T356 A NEW CALCIUM TEST BASED ON THE CHROMOPHORE NM-BAPTA M. Mcke, H. Josel, H. Kytzia Roche Diagnostics GmbH, Penzberg, Germany Background: Calcium in serum must be measured with high accuracy and precision as its normal range is tightly controlled. The new test Calcium Gen.2 on Roche/Hitachi cobas c501 is presented. Compared to the preceding assay it shows improved stability, linearity and recovery of reference materials and reduced interference by magnesium and contrast media containing gadolinium. Methods: The test uses the novel indicator NM-BAPTA on the automatic analyzer cobas c501 (Roche Diagnostics, Mannheim, Germany). Reagent 1 consists of a CAPSO buffered solution (at pH 10) of NM-BAPTA and reagent 2 of an EDTA solution. Firstly sample and reagent 1 are mixed to form a colored complex. Secondly reagent 2 is added. EDTA binds all calcium and discolors the calcium-dye complex. The difference in absorbance is measured at 340 nm. The method is traceable to standard reference material SRM 956 level 2 and is suitable for serum, plasma and urine. Each reagent lot is calibrated only once. Results: The test is linear up to 5 mmol/L in serum and 7.5 mmol/L in urine. Typical within run and total precision are 1% CV in the normal range and 2 3% CV for very low samples (ca. 0.5 mmol/L). There is no significant magnesium interference up to 15mmol/L in serum and up to 60mmol/L in urine. Gadolinium containing contrast media do not interfere at therapeutic concentrations. Stability on board the instrument is 6 weeks without recalibration. The test shows no significant interference up to an I-index of 60 (60 mg/dL bilirubin), a Hindex of 1000 (1000 mg/dL hemoglobin) and a L-index of 1000 (the L-index corresponds to turbidity, not to triglyceride concentration). The Passing Bablok regression of the method comparison to the previous calcium test on the same system is y =0.920x + 0.136 mmol/L. The serum based reference material SRM 956c level 2 is recovered in a range of 100 +2%. Conclusion: Compared to the preceding assay Calcium Gen.2 shows improved linearity and stability on board the analyzer without recalibration resulting in a more stable daily mean recovery. Due to the reduced interference to magnesium and to gadolinium containing contrast media results especially in urine become more reliable.
S469

T357 HOMOGENEOUS SOLID PHASE HYBRIDIZATION ASSAY UTILIZING LANTHANIDE CHELATE COMPLEMENTATION: THE EFFECT OF LABEL MOIETY INTERCHANGE AND SPACER LENGTH S. Lahdenper, J. Manninen, L. Joki, U. Karhunen, T. Soukka Division of Biotechnology, University of Turku, Finland Background: Homogeneous solid-phase hybridization assay is a wash-free proximity-based assay utilizing lanthanide chelate complementation technology. It consist of two oligonucleotide probes each carrying a part of the label, either the lanthanide ion carrier chelate (europium probe) or the light absorbing antenna ligand (antenna probe). A luminescent complex is formed by self-assembly when these two subunits are brought into close proximity by adjacent hybridization on the target sequence. We have investigated the effect of label moiety interchange and spacer length between the immobilized probe and substrate on assay performance using Pseudomonas aeruginosa heat shock protein gene groESEL as a model assay. Methods: One of the complementary probes was immobilized via 3 end onto a bottom of a microtiter well through a biotinstreptavidin linkage while the other was free in solution. For investigating the effect of label moiety interchange in immobilized and free oligonucleotide probes the immobilized probe was labeled either with europium or antenna and the solution phase probe with antenna or europium, respectively. In order to investigate the effect of the length of spacers up to 67 thymidines was introduced at the 3 end of the immobilized antenna probe. The triethylene glycol (TEG) linker added additional 15 atoms into the spacer. After hybridization timeresolved fluorescence was measured by surface-scanning. Results: A 60% signal increase was obtained with a 5 thymidine spacer (T5) compared to the probe containing the TEG linker only (T0). Signal was doubled with a 13 thymidine spacer (T13) and tripled with 67 thymidines (T67). Signal development rates were constant for up to 30 min during which the signals increased 23-41%, the largest increase occuring with the shortest spacers. After 30 min the signal development rate decreased 30% being constant for up to 420 min. Label moiety interchange between the immobilized and free probes didnt affect the signals. Conclusions: A 13 thymidine spacer is sufficient for introducing enough spatial freedom for the probes in our assay. With longer spacers the signals can additionally be increased, but the increase is less drastic and the production costs of the long spacer probes increase remarkably. T358 THE EFFECT OF THE SIZE OF UPCONVERTING PHOSPHORS IN HOMOGENEOUS ASSAY FOR ESTRADIOL S. Lahtinen(1), L. Mattsson(1), R. Arppe(1), E. Harju(1), M. Tuomisto(2), M. Lastusaari(2), T. Soukka(1)
1
Division of Biotechnology, University of Turku, Finland Department of Chemistry, Laboratory of Materials Chemistry and Chemical Analysis, University of Turku, Finland
2
Background: Upconverting phosphor (UCP) particles have the unique ability to convert low-energy infrared light into emission at visible wavelengths. This property enables autofluorescence free measurements, which makes UCPs ideal donors for homogeneous resonance energy transfer (RET) based assays. However, the performance of UCPs as donors in the assay depends on the size of the particle, which is affected by UCP synthesis method and different parameters used in the synthesis. Methods: The effect of the size of UCP-particles was studied in upconversion resonance energy transfer (UC-RET) based assay for estradiol in which 25250 nm sized UCPs were used as donors and Alexa Fluor 546 as an acceptor. UCPs were synthesized by co-precipitation or in organic oils. A functionalized silica layer was polymerized on the surface of UCPs and used for antibody conjugation. Upconversion luminescence and UC-RET-signal were measured with a modified plate reader equipped with an infrared laser diode as the excitation source. Results: One of the requirements for RET is that the distance between the donor and the acceptor is below 10 nm. Whole volume of the particle produces upconversion luminescence, but only the parts of the particle that are in the required distance for RET can transfer energy to the acceptor. Thus, particles larger than 40 nm produced UC-RET signal-to-background ratios (S/B) less than 10, because of a high background signal originating from donor crosstalk and radiative energy transfer due to low surface-to-volume ratio. Particles smaller than 40 nm produced the most efficient energy transfer with S/Bs of 4070. Even so, the smallest particles in this study (25 nm) produced the best S/B but the UC-RET-signal was relatively weak due to low upconversion luminescence. Conclusions: The size of UCPs used as donors in a homogeneous assay for estradiol had a significant effect on assay performance. By reducing the size a bigger volume of the particle was in the required distance for RET and thus higher signal-to-background ratios were obtained.
S470

T359 EVALUATION OF BECKMAN COULTER DXH800 HAEMATOLOGY ANALYSER M.Y. Lee, C.W. Lam, S.H. Lee, S.Y. Tan, H.X. Ng Department of Laboratory Medicine, Alexandra Hospital, Singapore Background: Our objective was to verify the analytical performance of Unicel DxH800, the newest generation haematologic analyser from Beckman Coulter, for routine CBC, 5-part WBC differential and NRBC count. The analyser is fully automated and uses individual cell volume, high-frequency conductivity and laser-light scatter to measure WBC differential and NRBC. The CBC analysis is based on Coulter Principle. Methods: The analytical performance of the analyser was assessed for imprecision, linearity, carry-over and method comparison. Imprecision was studied by measuring three levels of QC material (Coulter 6C Cell Control) over five days in accordance to CLSI EP5-A2 guidelines. Linearity was determined using Coulter Linearity Control which consists of ten bottles of linearity material with different concentrations that span the assay. Carryover study was performed as per manufacturers recommendation using whole EDTA blood (H) and diluent (L) in the following sequence H1 H2 H3 L1 L2 L3. The accuracy of CBC, WBC differential count and NRBC were compared with our current system, Sysmex XE-2100 using 82 patient samples. Results: The total within-laboratory CV for all parameters ranging from 0.5% - 8.8% and were within manufacturers specifications. Linearity was verified with recoveries of 98102%. No significant carry-over was observed. The DxH800 provides accurate results for all parameters, which correlates well with Sysmex XE-2100. Conclusions: The new generation automated haematology analyser, DxH800, showed satisfactory performance in method evaluation verification study. T360 COMPARISON BETWEEN PERKIN-ELMER AND CHROMSYTEM VITAMIN D KIT ON TQ 5500 FROM AB SCIEX C. Le Goff, S. Peeters, Y. Crine, M. Netchacovitch, J. Kaux, E. Cavalier University Hospital of Lige Background: Twenty-five hydroxy-vitamin D (25(OH) D) determination is now routinely prescribed in the Laboratory. Recently, different new methods have been available for this determination. Among them, LCMS/MS methods have emerged in some laboratories. However these methods are generally home-brewed and an important variability between them can be seen on different external quality controls, mainly due to a lack of standardization. Recently, Perkin-Elmer (PE) (Turku, Finland) and Chromsystem (CS) (Grafelfing, Germany) launched a standardised method for 25(OH )D determination on LCMS/MS. The aim of our study was to compare these methods on the AB SCIEX TQ5500 (Framingham, Massachusetts, USA) LCMS/MS to measure 25(OH) D3. Methods: All the samples were treated according to our preanalyitical procedure: after sampling, they were spun at +4 C at 3500G, aliquoted and kept frozen at -20 C until determination. A method comparison was assessed with CS and PE for the measurement of the 25(OH)D3. We selected 110 remnant samples with 25(OH)D3 levels ranging from 1.6 to 136.7 ng/mL with the PE method to cover the range of usually values Slope and intercept were calculated using Passing and Bablock linear regression and we compared the methods with the Bland and Altman plots. Results: For CS, the method is linear up to 250 g/L, the LOQ is 3.6 g/L, the intra-assay CV is <5% and the inter-assay is < 7%. For PE, the method is linear up to 314 g/L, the LOQ is 3.4 g/L, the intra-assay CV is <7.8% and the inter-assay is < 8.5%. On the whole range of measure (n=110), the regression equation is PE = 0.8521+0.9226 (CS) (95%CI of the intercept: (-0.0048;1.37) and 95% CI of the slope (0.89;0.95). The Bland and Altman plot does not show any bias between the two methods (mean difference CS-PE= -2.5 ng/mL) and the standard deviation of the mean is 3,98 ng/ml. Conclusion: The performances of these methods are comparable on our new TQ 5500 from AB SCIEX. For now, there is no consensus on a reference method for vitamin D quantification. We notice only that the values obtained by CS are systematically a little bit lower than PEs values, especially for results below 20 ng/mL. However, we have no clear explanation for such behaviour.
S471

T361 HOMOGENOUS HLA-DQA1*05 PCR ASSAY BASED ON SWITCHABLE LANTHANIDE FLUORESCENCE PROBES A. Lehmusvuori(1), M. Kiviniemi(2), T. Soukka(1)
2 1
T362 PLASMA HOMOCYSTEINE MEASUREMENT BY KONELAB20XT : COMPARISON WITH IMMULITE2000 AND ION EXCHANGE CHROMATOGRAPHY (JEOLAMINOTAC) S.K. Lim, P. Pernet, M. Vaubourdolle Service de Biochimie A / Ple de Biologie Mdicale et Pathologie Hpitaux Universitaires Est Parisien (AP-HP) Hyperhomocysteinemia is regarded as an independent risk factor for cardiovascular diseases. A wide range of methods are now available for total plasma homocysteine measurement. Lately, colorimetric assays adaptable to routine laboratory analyzers have been developed. We evaluated an enzymatic Diasys Homocysteine FS assay implemented on a biochemical analyzer Konelab 20XT (ENZ-Hcy) and compared its performance with our reference method, the ion exchange chromatography (IEC-Hcy) and an immunoassay (LIA-Hcy). ENZ-Hcy is a tri-reagent assay in which free homocysteine produced is recycled to reinforce the signal. In the final step the decrease in NADH is measured by a colorimeter. In the IEC-Hcy method, homocysteine was measured by the amino acid analyzer Jeol-Aminotac after a two-step pretreatment: dithiothreitol reduction followed by sulfosalicylic acid deproteinisation. The LIA-Hcy method (Siemens-Immulite2000) is a competitive immunoassay using a built-in reduction step. 90 blood samples (lithium heparinate) collected from patients screened for hyperhomocysteinemia was used for comparaison studies. Within and between-run imprecisions were assessed not only in patient samples but also in commercial control samples. Test for linearity was performed on dilutions of 50 mol/l L-homocysteine solution and a high concentration plasma sample (homocysteine=45 mol/l). The within-run coefficient of variations (CVs) for medium and high levels were < 1.5% for ENZ-Hcy and <6% for LIA-Hcy. The between-run CVs were <6.5% for ENZ-Hcy and <9.5% for LIAHcy. ENZ-Hcy method is not linear in the concentrations around the upper reference limit of 15mol/l and the correlation with IEC-Hcy method is better using non-linear regression (second degree polynomial curve, r2>0.99) than linear regression (r2=0.9685). In contrast LIA-Hcy method is linear and well correlated using Demings linear regression (r2>0.99). Using a cut-off value of 15 mol/l to define hyperhomocysteinemia we report only one discrepancy for LIAHcy method and 9 for ENZ-Hcy method over 47 plasma samples. The Diasys Homocysteine FS assay adapted to Konelab 20XT is short of linearity and accuracy. Despite a low imprecision this method is not appropriate for screening mild hyperhomocysteinem
Division of Biotechnology, Immunogenetics laboratory, University of Turku, Finland
Background: Switchable lanthanide chelate complementation probe technology is a versatile tool for homogenous DNA detection assays due to low background fluorescence level and high specific signal generation. The complementation probes are able to detect low picomolar quantities of target DNA in a homogenous assay and enables remarkably high signal-tobackground (S/B) ratio (max 300) in real-time PCR assay. The technology is based on two non-fluorescent oligonucleotide probes, one carrying a lanthanide ion carrier chelate and the other, a light absorbing antenna ligand. When the probes are hybridized into adjacent positions to the target DNA, a highly fluorescent lanthanide chelate complex is formed. Suitability of the technology for genotyping was studied developing complementation probes based homogenous PCR assay for the detection of HLA-DQA1*05 alleles used in assessing the risk for type 1 diabetes and celiac disease. Methods: Probes were designed to detect DQA1*05 alleles according to two nucleotide adjacent polymorphism. One oligonucleotide probe was labeled with a nonfluorescent EuIII chelate and the other with an antenna ligand. The antenna probe contained the polymorphic site and was designed to hybridize only to the DQA1*05 alleles at the measurement temperature of 24 C. Previously developed PCR primers were used and EuIII time-resolved fluorescence was measured after 40 cycle amplification. Performance of the assay was compared to the heterogeneous DQA1*05 PCR assay based on same primers testing 149 blood samples. Results: The complementation probe based end-point DQA1*05 PCR assay correlated 100% with the reference assay. Samples with DQA1*05 allele (n=89) gave high signalto-background ratio in average of 60.7 whereas the samples containing other alleles (n=60) yielded S/B in average of 1.5. Conclusions: The developed DQA1*05 assay demonstrates the ability of the complementation probes to discriminate different alleles in closed-tube end-point PCR. Huge signal difference between DQA1*05 and non-target alleles led to definite results. Homogenous assay format is easy to use and eliminates the risk of PCR product contamination. Multiplexing should be possible using different lanthanide ions (TbIII, SmIII and DyIII) with suitable antenna ligands.
S472

T363 HOMOGENOUS QUANTITATIVE ANALYSIS OF HUMAN PARATHYROID HORMONE BY USE OF MAGNETIC MARKERS AND SQUID MAGNETOMETER C. Maeda(1), K. Hisamatsu(1), K. Mitsuse(1), S. Egashira(1), N. Hamasaki(1), K. Enpuku(2), H. Kuma(1)
1
T364 LANTHANIDE CHELATE COMPLEMENTATION FOR SENSITIVE HOMOGENEOUS PROTEIN DETECTION E. Malmi, H. Pkkil, T. Soukka Division of Biotechnology, University of Turku, Finland Background: Fluorescence-based homogeneous assays typically have limited sensitivity compared to heterogeneous assays because of interfering background signal. In a novel lanthanide chelate complementation technology the fluorescent lanthanide chelate is divided into two separate non-luminescent moieties: a lanthanide ion carrier chelate and a light harvesting antenna ligand. The fluorescent chelate complex is formed only when these two label moieties are brought together via two specific binding events. This results in a minimal background signal in the homogeneous assay enabling low limit of detection. In this study the feasibility of lanthanide chelate complementation for sensitive homogeneous protein detection was evaluated by using cardiac troponin I (cTnI) as a model analyte. Methods: Two cTnI specific antibody fragments (Fabs) were conjugated to separate oligonucleotides. In the assay these Fab-conjugated oligonucleotides were hybridized with complementary signal oligonucleotides labeled either with Eu3+-carrier chelate or antenna ligand, thus forming two oligonucleotide complexes with Fab in one end and the label moiety in the other. In addition the signal oligonucleotides contained a short complementary terminal signal sequence. In the presence of cTnI the simultaneous binding of the Fabs brings the oligonucleotide complexes in close proximity and enables annealing of the signal sequence, which further results in the formation of long-lifetime fluorescent chelate complex. The total assay time was 6 min and the fluorescence was measured in time-resolved mode. Results: The detection limit of the homogeneous cTnI-assay (0.37 ng/mL) was of the same order as in the heterogeneous reference assay based on the same Fabs (0.26 ng/mL). The linear range of the homogeneous assay was slightly over 2 orders of magnitude, which can potentially be further improved by shortening the signal sequence. The highest observed signal-to-background ratio was 50. Conclusions: Here we have demonstrated that lanthanide chelate complementation technology enables sensitive and rapid method for homogeneous protein detection as a result of minimized background signal.
Department of Clinical Chemistry and Laboratory Medicine, Faculty of Pharmaceutical Sciences, Nagasaki International University 2 Research Institute of Superconductor Science and Systems, Kyushu University Background: Immunoassays are one main detection system used in the field of clinical chemistry. Recent developments of a new detection method utilizing a magnetic marker and magnetic sensor have enabled rapid and sensitive immunoassay without the need for bound/free (BF) separation. Recently, we successfully performed a quantitative evaluation of some proteins (immunoglobulin E, interleukin 8) in phosphate buffer, human serum and human hemolysate, without BF separation. In this time, we report the quantitative analysis of human parathyroid hormone (pTH) using this magnetic system. Methods: Newly-synthesized conjugated streptavidin was used as the magnetic marker for quantitative analysis. Target antigens (human pTH) were catched to first anti-pTH antibodies immobilized to plate and were biotinylated by secondary antibodies. A superconducting quantum interference device (SQUID) sensor detected the magnetic fields from markers fixed to pTH by the sandwich method. Magnetic signals from unbound markers were nearly zero due to Brownian rotation. Results: Our magnetic immunoassay could detect 10 pg of hormone (pTH) in phosphate buffer within 60 min. This technique has considerable potential for use as a biological immunoassay. Also, this homogeneous immunoassay could quantify three hundred cells from the fungus Candida albicans in phosphate buffer. Conclusions: The present study demonstrates the ability of magnetic markers for measuring biological targets without BF separation. We newly indicated the useful for detection of hormones as well as proteins. This detection system has great potential for use as the next generations analytical system.
S473

T365 MEASUREMENT OF IMMUNOSUPPRESSANTS BY LCMS/MS: WORKFLOW OPTIMIZATION TROUGH AUTOMATED PROCESSING OF WHOLE BLOOD SAMPLES M. Marinova(1), C. Artusi(1), L. Brugnolo(1), G. Antonelli(2), M. Zaninotto(1), M. Plebani(1) Department of Laboratory Medicine, University-Hospital, Padova, Italy 2 Department of Medicine, University of Padova, Italy Background: Immunosuppressant TDM is a crucial requirement in providing optimal patient care following organ transplantation. LC-MS/MS is an efficient technology for routine determination of immunosuppressant in whole blood, due to its high specificity and sensibility. However, time-consuming manual sample preparation remains a significant limitation of this technique. Aim of the present study was to develop an automated sample-preparation protocol for quantification of sirolimus (SIR), everolimus (EVE) and tacrolimus (FK-506) by LC-MS/MS using a liquid handling platform Tecan Freedom EVO 100. Methods: Six-level commercially available blood calibrators for SIR (0.229.0 g/L), EVE (0.223.2 g/L) and FK-506 (0.223.1 g/L) were used for assay development, while four QC materials with different concentrations and three blood samples from patients under immunosuppressants treatment were employed for imprecision evaluation. Barcode reading, sample resuspension, transfer of whole blood samples into 96well plates, addition of internal standard solution, mixing, and protein precipitation were performed by liquid handling platform. After plate filtration using vacuum block Te-VacS, the deproteinized supernatants were submitted to SPE on-line, using column switching prior to analysis. The only manual step within the entire process was the transfer of the well plate to the LC autosampler. Results: Calibration curves were linear throughout the selected ranges. The intra- and inter-assay CVs (<16%), the LLOQ (0.2 mol/L) and accuracy (bias%<10) for all analytes were highly satisfactory. Very good agreement between the results obtained after manual and automated sample preparation was observed (n=390, r2=0.92, P <0.001). In daily routine (about 100 patient samples) a typical comprehensive total turnaround time is less than 6 hours (40 min for sample preparation using the automated protocol and 5 h for chromatographic analysis). Conclusions: Our findings indicate that the proposed analytical system is suitable for routine analysis, since it is straightforward and precise. Furthermore, it minimizes substantial manual workload and the risk of mistakes in the quantification of whole blood immunosuppressant concentrations compared to conventional methods.
1
T366 ANALYTICAL VALIDATION AND DEVELOPMENT OF THE 25 OH VITAMIN D ASSAY FOR THE NEW UHPLC-MS/MS METHOD P. Meemaew, S. Vanavanan, A. Chittamma, M. Rochanawutanon Department of Pathology, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand. Background: 25-Hydroxy vitamin D (25-OH D) is an important marker for several diseases. Automated immunoassays from various manufactures have been available but they were unsatisfied. We developed and validated the new ultra-highperformance liquid chromatography mass spectrometry (UHPLC-MS/MS) method for the serum quantitative determination of 25-OH D2 and D3 . Methods: Spiked serum bovine albumin samples were used in both method development and validation. Serum (150 L) adding with 20 uL of internal standard (IS) were treated with zinc sulfate and acetonitrile for removing protein. Both 25-OH D2 and D3 were analyzed by UHPLC-MS/MS technique (Agilent Technology, Wilmington,DE,USA) which the chromatographic system consisted of an Agilent 1290 LC system with a 3.0x50 mm, 2.7 m Poroshell 120 EC-C18 column and a 4.6x12.5 mm, 5 m Zorbax Eclipse Plus-C18 guard column. A gradient of mobile phase A (water/ammonium formate (5M)/formic acid, 99.8/0.1/0.1,v/v/v) and mobile B methanol/ammonium formate (5M) / formic acid (99.8/0.1/0.1, v/v/v) was used at flow rate of 0.4 mL/min for pump A and 0.5 mL/min for pump B. The 3 L injection volume was set. The Agilent 6460 Triple Quadrupole MS coupled to electrospray ionization (ESI) source was used. We validated the method in terms of linearity of range, limit of detection (LOD), limit of quantitation (LOQ), %recovery and precision. Results: The mass (m)/charge (z) transitions of 413.3>355.3 (25-OH D2), 401.3>365.2 (25-OH D3), 416.3>358.3 (IS of 25OH D2) and 404.3>368.2 (IS of 25-OH D3) were detected in the MRM and can be achieved at the time between 3.2 to 3.8 minutes. Total run time was 4.5 minutes per sample. Chromatogram of all analyzes showed symmetry peaks. The linearity over the range of 5-100 ng/mL with r2 = 0.999 for both. The LODs and LOQs were 1.61,5.37 and 2.01,6.69 ng/mL for 25-OH D2 and D3, respectively. Recovery of 25-OH D2 and D3 from serum samples spiked with 10,30 and 80 ng/mL ranged between 94.83% to 100.76%. Inter and intra assay percentage relative standard deviation (%RSD) were less than 10%. Conclusions: This new UHPLC-MS/MS method with a simple of sample preparation provides a rapid, accurate, precise and suitable assay in routine use as a useful tool for evaluation of individual vitamin D status.
S474

T367 AUTOMATIC EVALUATION OF DNA DOUBLE STRAND BREAKS: WHEN RESEARCH MEETS DIAGNOSTIC LABORATORY A. Melegari(1), C. Bonaguri(2), D. Campioli(3), A. Russo(2), S. Canovi(1), B. Venturelli(1), R. Perini(2), G. Lippi(2), T. Trenti(1)
1 Diagnostic Laboratory Department NOCSAE Hospital, Modena , Italy 2 Diagnostic Laboratory Department Parma Hospital ,Italy 3 Diagnostic Laboratory Department Policlinico Hospital, Modena, Italy
T368 DETERMINATION OF INFLIXIMAB TROUGH LEVELS (IFX-TL) AND ANTIBODIES TO INFLIXIMAB (ATI) IN INFLAMMATORY BOWEL DISEASE C. Guiotto(1), L. Germano(1), M. Vizzini(1), R. Cerruti(1), F. Frigerio(2), M. Daperno(2), R. Rocca(2), M. Migliardi(1)
1 S.C. Laboratorio Analisi, A.O. Ordine Mauriziano di Torino, Turin, Italy 2 S.C. Gastroenterologia A.O. Ordine Mauriziano di Torino, Turin, Italy
Background: The number of DNA double-strand breaks (DSB) increases if a cell is exposed to ionizing radiation, chemotherapeutic substances or free radicals. H2AX is a histon protein, around which DNA wraps itself in the nucleus.In case of DSB,the histone gets phosphorylated (-H2AX) on a serin residue.-H2AX serves as highly sensitive marker for double-strand breaks and can be made visible by a specific antibody that is linked to fluorescent dye. Aim: indirect immunofluorescent evaluation of -H2AX foci in cell nuclei is available as a complete manual procedure, but it is cost and time consuming and poorly reproducible. The method requires isolation and counting lymphocytes from whole blood, followed by immunostaining with primary and secondary antibodies prior to final microscopic evaluation of nuclear foci. In order to improve standardization of the method , taking advantage of the technology available in our diagnostic laboratories, we replaced the two manual steps requiring subjective microscopic evaluation (cell counting and analysis of nuclear foci)with automatic procedures. Methods: the evaluation was conducted using peripheral lymphocytes of 5 healthy subjects .Isolated lymphocytes were counted manually and also with two automated differential counters: Beckman Coulter LH 780 and Sysmex XE-2100 .The measurement of -H2AX foci was performed on the AKLIDES platform and the associated software, developed for the automated analysis of cell-based immunofluorescence assays. Results: the calculation of the lymphocytes with automated counters was more accurate and faster than those manual. The analysis on AKLIDES system with sensitive y-H2AX signals yelded several quantitative parameters that characterized the degree of cell damage:number of cells (from 101 to 115), average number of foci per cell (from 0,000 to 0,049), percentage of cells with foci total damage (from 0.000 to 3,883),standardizing analysis of y-H2Ax foci in lymphocytes. Conclusions: Our preliminary evaluation on a few healthy subjects where cell damage is light, leads to support the usefulness of automation . This improves standardization and enriches both quantity and quality of the obtained data and makes possible transferring it in clinical application areas.
Background: Anti-TNF blocker (Infliximab,IFX) is approved and used in the treatment of Crohns disease (CD) and ulcerative colitis (UC). Therapeutic algorithms based on drug monitoring combined with anti-drug antibodies detection were proposed. Aim of this study was to compare IFX trough level (IFX-TL) and antibodies to IFX (ATI) with two different commercial kits. Furthermore, we evaluate association between IFX-TL and better disease outcomes and between ATI and adverse drug reactions. Methods: 46 inflammatory bowel disease outpatients (27 CD and 19 UC;15 with active disease and 31 with quiescent disease), undergoing IFX i.v. dosing, was prospectively enrolled. Serum samples were taken before planned IFX infusion and detailed clinical history was collected. Two different ELISA-sandwich tests were used in order to determine IFX-TL and ATI concentrations: Promonitor IFX Determination of Drug and Anti-Drug Antibodies Concentration (Menarini) and TNF Blocker Monitoring/Antibodies against TNF Blocker (ImmunDiagnostik). Results: 36 of 46 patients were tested, until now, with both methods. Good correlation (Spearmans rho 0.935, p <0.0001) and discrete agreement (y=1.68x+0.619) between IFX-TL methods was found. As regards ATI detection (3/36 cases, 8%), both assays showed similar results, correlated with loss of clinical response to treatment. ATI levels determined with both Promonitor and ImmunDiagnostik methods resulted significantly associated with the degree of disease (OR 18.053, IC95% 0.85-383.93, P <0.05 and OR 5.273, IC95% 0.84-33.01, P <0.05, respectively). Regarding the results obtained with Promonitor, median IFX-TL in CD and UC patients were 1,5 and 4,8 g/mL, respectively, while considering the results obtained with ImmunDiagnostik, median IFX-TL in CD and UC patients were 3.51 and 5.6 g/mL respectively. Conclusions: The results show similar performance for the two different assay methods. Measurement of IFX and ATI proves to be a useful tool to monitor IFX therapy, though clinical implications are still matter of debate before these tests can be proposed for routine clinical practice. Consolidation of results on a larger cohort of patients is warranted to confirm and extend the obtained results.
S475

T369 A COMPARISON OF TWO DIFFERENT LABORATORY METHODS FOR DETECTION OF IGF-1 M. Monari, M. Pedretti, I. Soliman, R. Assandri, A. Montanelli Humanitas Clinical and Research Center, Rozzano (Mi), Italy Background: Insulin like growth factor I (IGF1) is also called Somatomedin C. Analysis of IGF1 in serum has become an integral component in the diagnosis of growth hormone related disorders and it is suggested as the most reliable component for monitoring the success of therapeutic interventions in acromegaly. There are no international accepted guidelines for the related ages ranges of results. The aim of this study is to compare two different laboratory methods for the evaluation of IGF-I serum levels. Methods: Serum IGF1 has been measured in 44 samples of people with age between 15 and 80 years. All samples were analyzed by two different assays IGF1 Diasorin Liason and IGF1 IMMULITE 2000. The first is an automated 2-site sandwich immunoassay with a chemiluminescent detection ( calibrators 1st WHO international Standard IGF1 NISBC code: 02/254; linearity between 3 and 1500 ng/mL, CV intra-assay < 4.4%, CV inter-assay <8.5%, hook effect >11.000 ng/mL) It is unknown any interferences. IGF1Immulite 2000 (Immulite 2000, Diagnostic Products Corp, Los Angeles, CA) is an automated 2-site sandwich immunocemiluminescent assay (calibrators WHO international Standard 87/518, sensitivity 25 g/L, intra and inter-assay CV is <8 ). It is unknown any interferences. Results: All 44 samples were measured by both two methods. We have registered a totally concordance by two system and the procedure was conducted in double-blind way. Conclusions: No differences in IGF-1 levels was observed and there is an excellent correlation in order to give an equal diagnostic information to patients. So, to reduce the condition linked to an inappropriate detection of values of IGF-1 such as: an unknown subclinical thyroid dysfunction, use of medications like contraceptive drugs, and finally the body mass index of subjects, it is important to adopt the same calibrator in different system and it is desirable to create reference values of different commercial assays and for different ethnic groups. In our laboratory the attenction is focused, after these preliminary data, to collect all results for a possible match with the manifacturers guidelines to observe if, in our population, there are differences. T370 PERFORMANCE EVALUATION OF A NOVEL FULLY AUTOMATED HAEMATOLOGY ANALYZER MINDRAY BC6800 (MEDICAL SYSTEMS S.P.A GENOVA-ITALY) R. Ottaviano, D. Garelli, S. Pastori, G. Giuliani UOC di Medicina di Laboratorio Azienda Ospedaliera G.Salvini Garbagnate Milanese, Milan, Italy Background: An accurate evaluation of analytical performance of a new hemocytometer is necessary before it can be introduced for routine testing. The aim of this study was a preliminary evaluation of complete blood cell (CBC) count on Mindray BC-6800, now available on the Italian market. Methods: The imprecision of the BC-6800 was assessed according to the Clinical and Laboratory Standards Institute (CLSI) document EP5-A2 and haematology analyzers evaluation guidelines. Imprecision within-run was evaluated including 10 inpatient random samples from several wards of Rho Hospital (Milan), collected with various values of parameters and assayed for 20 times. For imprecision between-run three control materials were assayed in duplicate for 20 consecutive working days. The results were expressed as coefficient of variation, CV. All parameters of the Complete Blood Cell Counts (White blood cells,Red blood cells,Platelets, Mean Corpuscular Volume, Mean Corpuscular Hemoglobin, Mean Corpuscolar Haemoglobin Concentration and Red Distribution Width (WBC, RBC, PLT, Hb, Ht, MCV, MCH, MCHC, RDW) were investigated. The background check and carry over (in triplicate) in autoloading mode and open vial mode were also investigated, running a high concentration sample (high control) and then a low concentration sample (diluent). Throughput was evaluated by Chronometric measurement of the time passed from starting the instrument with the first sample to the last results produced on the 100th sample, performing all parameters. Results: Optimal imprecision both in within-run and betweenrun was observed for all the parameters tested, with coefficients of variations (CVs) always below 3%, except for platelet count within run(4%). The blank count and the carry over demonstrated a performance (0.02% and 0.08% respectively) always under requirements. The throughput was approximately 120 tests/h processing 8 parameters, 93 tests/h for all parameters including reticulocytes and erythroblasts. Conclusions: The preliminary evaluation of CBC on Mindray BC-6800 suggests that this novel hemocytometer may be suitable for routine hematological analysis of most blood specimens, including those of patients with haematological disorders
S476

T371 CORRELATION BETWEEN A NEW FULLY AUTOMATED HEMATOLOGY ANALYZER MINDRAY BC-6800 (MEDICAL SYSTEMS S.P.A.) AND SYSMEX XE2100 ANALYZER ( DASIT S.P.A.) R. Ottaviano, D. Garelli, S. Pastori, G. Giuliani UOC di Medicina di Laboratorio Azienda Ospedaliera G. Salvini Garbagnate Milanese, Milan, Italy Background: The aim of this study was a correlation between a complete blood cell (CBC) count on Mindray BC-6800 (Medical Systems S.p.A.) a novel hematological analyzer and our analyzer in use (Sysmex XE2100 ), looking for a new hematology analyzer that would allow quality results, sample stability performance, improvement of slides review rate, and efficiency Methods: A total of 100 blood specimens were selected from inpatients of Garbagnate Milanese Hospital (Milan). Fresh samples of peripheral blood (<6 hours from collection), collected in Vacutainer tubes (Becton Dickinson) containing tripotassium EDTAanticoagulant and stored at room temperature, were analysed both in XE2100 analyzer and BC 6800, for all parameters. Blood smears were evaluated and compared also with results of two instruments. For each smear, 2 independent technologists performed a 200-cell WBC differential according to the H-20 NCCLS guidelines. The comparison of results between BC 6800 and XE2100 for CBC count or manual microscopy for WBC differential, was assessed by Pearsons correlation. The bias [95% confidence interval (CI)] was calculated by means of the Bland-Altman analysis. Results: The comparison of WBC, RBC, PLT counts and Haemoglobin between BC-6800 and XE2100 produced limited biases, i.e., 0.21x109/L (95% CI 0.10 to 0.52x109/L) for WBC, 0.15x1012/L (95% C 25 to 0.04x1012/L) for RBCs, 12 (95% CI 5 to 19x109/L) for PLT, and 2.1g/L (95% CI 2.5 to 1.7 g/L) for Hb. We have found excellent agreement and limited bias with the current gold standard for WBC differential (i.e., manual microscopy performed according to the CLSI guidelines) after analysis of 100 specimens with mean biases of 0.21x109/L (95% CI 0.05 to 0.45x109/L) for neutrophils, 0.14x109/L (95% CI 0.28 to 0.00x109/L) for lymphocytes, 0.25x109/L (95% CI 0.07 to 0.47x109/L) for monocytes, 0.03x10/L (95% CI 0.01 to 0.06 x 109/L) for eosinophils and 0.02 (95% CI 0.04 to 0.00 x 109/L) for basophils. Conclusions:The preliminary correlation between Mindray BC6800 and XE2100 analyzer suggests that this novel hemocytometer, for quality results, sample stability performance, slide review rates and efficiency may be suitable for routine hematological analysis. T372 COMPARISON OF TWO EXTRACTION DEVICES FOR DETERMINATION OF FAECAL CALPROTECTIN ON IMMUNOCAP 250 M. Oyaert(1), C. Trouv(2), M. Langlois(2), H. Vanpoucke(1)
1 Department of laboratory Medicine, Clinical Chemistry, H. Hart Ziekenhuis Roeselare Menen vzw 2 Department of laboratory medicine, Clinical Chemistry, AZ. St. Jan Brugge
Background: Calprotectin is an acute-phase protein used in the assessment of inflammatory bowel diseases (IBD), such as Crohns disease and Ulcerative Colitis. The aim of our study was to evaluate the performance of two different extraction devices for detection of faecal calprotectin (Roche vs. Thermofisher) o n the Phadia Immunocap 250. Methods: We analysed 105 clinical samples, with a great diversity of stool form and covering a broad range of calprotectin levels (0-3112 g/g). Within- and between-run precision of ten different extractions of the same sample were evaluated for both devices. According to the weighed stool captured with each device, the theoretical and expected (weight-corrected) concentration of calprotectin was calculated and compared with the measured value on Immunocap 250. Correlations of both devices over the whole measuring range (0-3112 g/g) as well as in the low range (0-600 g/g) were performed. Results: Within-run coefficients of variation for the Roche (2,69%) and Thermofisher (3,01%) device were excellent. Acceptable between-run precisions were obtained; 14.25% (Roche) and 10.77% (Thermofisher). Pearson correlation coefficient of calprotectin levels obtained with both devices was 0,92 (intercept 0,27; slope 0,91). Comparison of measured and calculated value for each device showed a better correlation for the Roche than for the Thermofisher device; 0,99 (intercept - 6,75; slope 1,07) and 0,80 (Intercept -3,82; slope 1,98) respectively. Correlations between calculated and measured calprotectin levels in the low measuring range (<600 g/g) were 0,99 (intercept -2,07; slope 1,09) with Roche and 0,45 (intercept 11,94; slope 1,92) with Thermofisher device. Conclusion: We found discrepant results for the Thermofisher extraction device between measured and expected (weightadjusted) values on Immunocap 250. Nevertheless, correlations for the Roche device were excellent. Enormous variations in weight with the Thermofisher device resulted in poor correlations between measured and calculated calprotectin levels, more specifically in the low range, and attributable to the different stool forms which were extracted. We therefore do not recommend the Thermofisher device to extract calprotectin for measurement on Immunocap 250.
S477

T373 PROSTATE CANCER URINE PROTEIN PROFILING BY MALDI-TOF/MS: NORMALIZATION, REPRODUCIBILITY AND PRELIMINARY RESULTS A. Padoan(1), M. La Malfa(1), D. Basso(1), C.F. Zambon(1), S. Moz(1), D. Bozzato(1), P. Fogar(1), E. Greco(1), F. Navaglia(1), G. Pavanello(2), M. Plebani(1)
2 1
T374 STUDY OF THE INTERCHANGEABILITY OF NT-PROBNP RESULTS IN TWO ANALYZERS E. Prez Gmir, S. Esteve Poblador, S. Grriz Pintado rea de Diagnstico Biolgico. Hospital Universitario La Ribera, Alzira, Espaa Background: BNP is a natriuretic peptide. In patients with left ventricle dysfunction its levels increase like those of the biologically inactive prohormone, in this process the proBNP splits into the biologically active BNP and the N-terminal fragment (NT-proBNP). The NT-proBNP is used in the differential diagnosis of dyspnea (cardiac or respiratory origin), helping in the diagnosis and prognosis of cardiac insufficiency. Our aim was to verify the interchangeability of the results obtained by means of two analyzers. Methods: We determined the values of NT-proBNP in an analyzer Cobas e-411 (method in use) (immunoelectro chemiluminiscent assay) and in an Immulite 2000 (method in study) (immunochemiluminiscent assay). 102 patients samples were processed in both analyzers. The origin of the patients was: cardiology (82%), internal medicine (13%), pneumology (3%), urology (1%) and emergencies (1%). The results obtained were compared using the Spearmans correlation coefficient and the Passing-Bablock concordance test. Besides, the patients were classified according to their diagnosis of cardiac insufficiency, and ROC curves were performed for each analyzer. The statistical analysis was performed with the statistical program MedCal. Results: We obtained a correlation coefficient of 0.972 (CI 95 %: 0.959-0.981). The straight line of regression was obtained with arranged in the intercept: -37.7904 (CI 95%: -52.5302 to 23.6813) and slope: 0.9469 (CI 95%: 0.8966 to 0.9994). By means of ROC curves we obtained for diagnosis of cardiac insufficiency an area under the curve (AUC) for the method in use of 0.742 (CI 95%: 0.640-0.827). With a sensitivity of 82% and a specificity of 56% for a cutoff of 683 pg/mL. The AUC for the method in study was 0.727 (CI 95%. 0.624-0.815), with a sensitivity of 91% and a specificity of 47% for a cutoff of 361 pg/mL. Conclusions: The comparison between the equipments provided a good correlation, but the values obtained in both analyzers are not transferable. In spite of this, the two have a good diagnostic capacity for cardiac insufficiency, though it is necessary to consider different cut-offs.
Department of Medicine, University of Padova, Italy SIPRES, Gruppo Pavanello, Padova
Background: Only few studies have addressed MALDITOF/MS analytical variability for the urine matrix. We evaluated intra-(CVw) and inter-assay (CVb) variability of MALDITOF/MS urinary profiling using internal standard (IS), relative intensity (RI) and total ion current (TIC) normalizations. Results were applied on prostate cancer (PCa) protein profiling. Methods: Fresh urine from 10 healthy subjects were pooled, aliquoted and frozen at -80 C. To estimate CVb, every day and for 14 days one aliquot was thawed, dialyzed at 4 C (1kDa) before MALDI-TOF/MS analysis (Bruker Daltonics). For CVw one aliquot was thawed, dialyzed and then spitted in 20 new aliquots, which were independently analysed. Before dialysis, each urine aliquot was added with a 1589.9 m/z peptide at 12 pmol/L (IS). Flex Analysis was used for baseline subtraction and peaks detection. Peak intensities were divided by the intensity of the IS, the intensity of the most abundant peak (RI) or by TIC, obtained by averaging all spectras TIC. 178 urine samples (106 Reference subjects, 72 PCa patients) were dialyzed, analysed by MALDI-TOF/MS and normalized. Results: In a range of 1050 to 4000 m/z, we identified from the analysis of all 14 spectra for the CVb a total of 134 peaks, and a total of 81 peaks from the 20 spectra for CVw. With the IS, RI, TIC normalization, the mean CVb% [95%CI] were: 228.9 [214.1-243.6], 158.3 [137.0-179.5], 156.6 [135.3-177.8], while CVw% [95% CI] were: 199.9 [171.9-227.8], 151.0 [116.7185.2], 149.4 [115.1-183.7], respectively. We evaluated, for each normalization method, whether a detection cut-off limit might improve CVs by checking a 0.2% to 5% range values. The best combination between CVs and the number of detectable features were 4%, 2.5% and 0.5 % for the IS, RI and TIC respectively.TIC performed better, being the mean CVb% 39.0 [33.6-45.2] and the mean CVw% 36.9 [31.1-42.8], respectively. On applying TIC normalization at the 178 collected spectra from the protein profiling study, 2 features (m/z 1161 and 2016) were statistically different between References and PCa patients. Conclusions: For the MALDI-TOF/MS analysis of urine, TIC normalization perform better. Preliminary results from protein profiling identified 2 possible interesting features for PCa diagnosis
S478

T375 EVALUATION OF A COMMERCIAL KIT FOR STEROIDS DETERMINATION BY LC-MS/MS J. Gervasoni(1), C. Zuppi(1), A. Cocci(1), M. Teti(1), B. Zappacosta(2), S. Persichilli(1)
1
T376 ION CHROMATOGRAPHIC-AMPEROMETRIC DETERMINATION OF IODIDE AND TOTAL IODINE IN URINE M. Petrarulo, D. Marranca, R. Cerruti, E. Miotto, M.F. Quarone, V. Nannavecchia, M. Migliardi S.C. Laboratorio Analisi, A.O. Ordine Mauriziano di Torino, Turin, Italy Background: The determination of total urinary iodine is a valuable tool to diagnosing iodine status before radioablative treatment of patients with differentiated thyroid carcinoma. The classic colorimetric method for detecting iodide is known as the Sandell-Kolthoff. In this reaction Ce(IV) is reduced by As(III), and iodide is a specific catalyst. Total iodine can be measured in urine in the same way by oxidative digestion and reduction procedure, before applying the classical colorimetric method. We evaluated the ion chromatographic (IC) determination of total iodine in urine with electrochemical (EC) detection at the silver electrode after oxidative digestion and subsequent reduction. In order to obviate to the use of As(III), suspected as being unsafe, we used thiosulfate as the reducing agent. Iodide assay was performed on simply diluted urine. Methods: Isocratic ion chromatography using a AS4A (Dionex) anion separator and a neutral phosphate buffer as the eluent. EC potential at -0.07 V vs SHE (Decade II, VT03 Ag flow cell, Antec, USA). Total Iodine: Urine is digested at 100C with ClO3/ClO4 and then reduced with thiosulphate. 2 L of mixture are injected. Iodide: 2 L of tenfold water-diluted urine are injected. Run time: 8 min. Results: Stability of Ag Electrode: electrode response was stable after hundreds urine assays (losses of some 5% could be observed after 400 runs). Sensitivity: LOD: 5 g/L, LOQ: 10 g/L. Detection Selectivity: the symmetric peak obtained when urine samples were analyzed proved high specificity. Common reductants (cysteine, Fe (II), SCN- ) do not interfere. Diagnostic use: Urine concentrations exceeding 50 g/L correlate well with those obtained by using the colorimetric method. Imprecision (<4%) and accuracy (recovery 100%) made the procedure better than the colorimetric one. Cost: This HPLC-EC procedure uses stable solid-phase electrode. Very few sample amounts are injected (0.2 L of urine) i.e, very long column life. No further consumables are required. In summary it seems to be very cost-effective. Conclusions: This analytical procedure is a simple, rapid, costeffective tool for the assessment of iodide and total iodine in urine. The electrode response is stable, no interferences have been observed, no unsafe substances are required.
Laboratorio Analisi I, Policlinico Gemelli, Rome Laboratori e Servizi, Centro di Ricerca e Cura Giovanni Paolo II, Campobasso
2
Background: Steroids and their precursors have traditionally measured by immunoassay. These methods ensure adequate sensitivity but are strongly affected by interfering compounds. Analysis of steroids by LC-MS/MS offers much improved specificity and therefore sensitivity. The analysis of more steroids simultaneously, possible using mass spectrometry, can be useful to the clinician. In this work we evaluated a commercial kit by Perkinelmer for the simultaneous analysis of ten steroids (Aldosterone, Cortisol, Dehydroepiandrosterone sulfate, Corticosterone, 11-Deoxycortisol, 4-Androstene -3,17dione Testosterone, 17 - a - Hydroxyprogesterone, Dehydroepiandrosterone, Progesterone) in serum, by liquid chromatography tandem mass spectrometry (LC-MS/MS). Methods: The kit use a combined solvent extraction and protein precipitation method. Analyses were carried out using a ThermoFisher Scientific TSQ Quantum Access triple quadrupole mass spectrometer operating with atmospheric pressure chemical ionization in the positive mode for nine analytes and in negative mode for aldosterone. Separation was achieved, within 15 min, with a gradient elution with a Phenomenex Luna C8 (100x2.1 mm, 3m) column maintained at 40 C. MS conditions were optimized using the tuning solution, according to kit instructions. Two transitions were monitored for each analyte/internal standard. Results: The method, used with our entry-level instrument, shows an adequate sensibility for all the steroids except for aldosterone. Calibration curves were linear (for all analytes except for aldosterone) in the calibration ranges. Inter and intraassay imprecision, obtained by measuring in replicate (n=6) the QC solutions on the same day and in duplicate on seven different days, were lower than 10%. Conclusions: The kit is suitable for clinical evaluation of all steroids. Only for aldosterone, sensitivity is not adequate because of the instrument characteristics. Although MS methods are time-consuming, and require skilful personnel, they have the specificity, imprecision, and sensitivity necessary for the reliable measurement of steroids, giving to the clinician a powerful diagnostic tool especially when a complete profile was provided.
S479

T377 BILE ACIDS INTERFERENCE IN PSEUDOCHOLINESTERASE DETERMINATION ON ROCHE COBAS 8000 SYSTEM K. Pocino, P.D. Daloiso, P. Corsale, C. Zuppi, M. Antenucci General Hospital "A.Gemelli", Roma, Italy Background: Bile Acids and Pseudocholinesterase (CHE) measurements in serum turn out to be overall useful as tests of liver function. Methods:The assays of these molecules, in our laboratory, are carried out on Cobas 8000 system (Roche Diagnostics), composed by two lines. CHE is tested both on Cobas C502 of CC line, and on the second C502 line, named CCE, where are also determined bile acids, cardiac and sepsis markers. It is noteworthy that reagents used for Bile Acids determination are provided by Sentinel Diagnostics, while reagents used for CHE detection are supplied by Roche Diagnostics. Both analytes are subjected to internal quality control (IQC) previously carried out with controls provided by Roche. Actually, according to current guide lines IQC is performed using multiparameter third party control sera , at two concentration level (ChemGol1 and ChemGol2, BIO-DEV, Milan). Results:From the observation of Levey-Jennings charts we noticed an increase of violation of Westgard rules only in the controls referred to CCE line. For this reason we proceeded to a general recognition of methods performed. It is shown, CHE determination immediately follows that of Bile Acids. This led to the hypothesis of analytical interference supported by chemical composition of reagents used for Bile Acids determination . In fact the reagent 1A contains Na3PO4 that added to pyrophosphate buffer, present in the reagent 1 used for CHE detection, inhibits the enzyme activity, because it increases pH solution. Therefore we decided to predispose an extra washing of the reagent needles with an acid solution (Cell Wash Solution). This device showed a significant improvement of quality control results as well as that of patient sera (n=40; CHE mean value before=6360UI/L; CHE mean value after extra washing =6872UI/L) Conclusion: Multiparameter third party controls permitted us to individuate this new interference in a closed system, that generally limits the implementation of methods not provided by Roche. T378 AGREEMENT BETWEEN SFRI4000 AND PROLYTE ISE ANALYZERS IN THE DETERMINATION OF PLASMA SODIUM AND POTASSIUM O. Popoola, O. Oyedele, M. Okunola Chemical Pathology Department, University College Hospital,Ibadan,Nigeria Background: Clinical decision making is confirmed by laboratory test results. When dealing with sick patients, the speed and accuracy of tests to detect metabolic derangements is very important. It is, therefore, required that laboratory results be timely, accurate, reliable and fit for the purpose. We evaluated agreement between two ion selective electrode analyzers (SFRI 4000 and PROLYTE) used for the quantitation of plasma sodium and potassium in our laboratory. Materials and methods: 42 samples originally meant for electrolytes assay in our laboratory were employed for the study. After blood receipt the samples were spun at 4,000r.p.m immediately. Plasma sodium and potassium were analyzed simultaneously with Prolyte and SFRI ISE analyzers. The agreement between the two analyzers was assessed using Bland-Altman Method. Results: Na+ Coefficient of variation: PROLYTE 0.9%, SFRI, 1.1%, Regression Analysis: Correlation Coefficient r2=0.93, slope=1.08, intercept,-13.24, maximum difference, 4mmol/L, minimum difference,-4mmol/L, Bland-Altman analysis mean difference between PROLYTE and SFRI ion selective method is 1.43.42 with 95% confidence interval of -2.02 to 4.82; 2. K+ Coeffient of variation: PROLYTE, 3.0% SFRI, 2.3%, Regression Analysis; Correlation Coefficient r2=0.98, slope=1.1, intercept ,-0.2, maximum difference = 0.2, minimum difference,-0.5, Bland-Altman analysis mean difference between PROLYTE and SFRI ion selective method is 0.210.26 with 95% confidence interval of -0.050.47. Conclusion: Good degree of agreement was observed on comparing the two ISE analyzers for the measurement of sodium and potassium. The two analyzers can be used without fear of inaccuracy and imprecision.
S480

T379 SINGLE-WALLED CARBON NANOTUBES AS POTENTIAL AGENTS FOR THE CIRCULATORY FUNCTION CORRECTION UNDER PATHOLOGICAL STATES OF CENTRAL GENESIS Y. Prylutskyy(1), L. Shapoval(2), V. Sagach(2), O. Dmitrenko(2), L. Stepanenko(2), L. Pobigailo(2), U. Ritter(3), P. Scharff(3) Taras Shevchenko National University of Kyiv, Kyiv, Ukraine Bogomolets Institute of Physiology, National Academy of Sciences of Ukraine, Kyiv, Ukraine 3 Technical University of Ilmenau, Institute of Chemistry and Biotechnology, Ilmenau, Germany
2 1
T380 COMPARISON OF LIQUID BASED CYTOLOGY WITH CONVENTIONAL CYTOLOGY IN THE EVALUATION OF ABDOMINAL MASSES A. Rautela, M. Choudhury Lady Hardinge Medical College, University of Delhi, New Delhi Background: Liquid based preparations are used for gynaecologic and non-gynaecologic cytology including body fluids and FNA. These preparations have several preparatory, screening and diagnostic advantages making them an appropriate alternative to conventional cytology including uniform collection procedures, avoiding hazards of needle handling required and easy transportation to the laboratory. Immediate liquid fixation preserves morphology and residual material can be used to either process multiple slides & cell block for ancillary tests.Aims of this study were to study the role of Liquid Based Cytology in the diagnosis and characterization of abdominal masses,to compare the Liquid Based Cytology with Conventional smear Cytology and to correlate the results with either cell block preparations or histopathological sections and immunohistochemistry. Methods: In the study total of 30 cases with abdominal masses were included. Ultrasound guided fine needle aspirate and cell block was collected. The aspirate was spread and fixed for conventional smears. A second pass was made and rinsed into liquid based collection vial and processed in Thin Prep 2000( used for preparation of liquid based cytology smears). Slides from both the methods were compared for cellularity, adequacy, background (blood and necrotic cell debris), cellular architecture, maintenance of nuclear detail, cytoplasmic integrity and diagnostic accuracy.Cell block/biopsy correlation whichever available was done. Observation and Results: Cellularity of Conventional smears was superior to Thin Prep (P=0.025). Difference not statistically significant (P=0.112) regarding adequacy of smears. Maintenance of architectural pattern in Conventional smears was superior to Thin Prep (P <0.001). Reduction in background material of Thin Prep was superior to conventional smears (P <0.001). Cytoplasmic integrity in Conventional smears was superior to Thin Prep (P <0.001). Regarding nuclear details difference is not statistically significant (P=0.091) among the two preparations. The difference in the diagnostic accuracy between the two methods is not statistically significant (P value=0.226) Conclusion:The liquid based cytology technique is expensive for routine use if used alone than in conjunction with conventional cytology.
Background: The purpose of the work is to examine the hemodynamic effect of single-walled carbon nanotubes (SWCNTs) depending on the dose and technology of their administration in rats with genetically determined hypertension. Methods: Spontaneously hypertensive rats represent an animal laboratory model of hypertension in humans that is widely used to compare response patterns in rats with persistently increased and normal arterial pressure. In urethane anesthetized rats, SWCNTs were administrated in the medullary nuclei (nucleus of solitary tract, NTS; paramedian nucleus, PMn; lateral reticular nucleus, LRN; nucleus ambiguus, AMB) that are directly involved in the nervous control of the vascular tone and cardiac activity. The effectiveness of applied antihypertensive technology was assessed by analyzing the changes in the systemic arterial pressure (SAP) and the heart rate. Also we have studied in detail the toxic effect by analyzing the stability of the erythrocytes to acid hemolysis depending on the dose and the way of SWCNT administration. Results: SWCNT injections in the AMB resulted in the SAP drop by 14.4% (P <0.05), in the LRN - by 22.8% and in the NTS - by 21.6%. Hypotensive responses were characterized by rapid development with maximum in 10-20 s and they lasted 3 min. SWCNT-evoked responses in the PMn were mostly hypertensive. In spontaneously hypertensive rats, SWCNT injections resulted in the SAP lowering in the studied nuclei by 22.8%, 21.0% and 13.0% in the AMB, LRN and PMn, responsively, that is, in those animals responses were usually more significant compared to those in the control rats. Hypotensive responses in spontaneously hypertensive rats developed more slowly compared to those in the control group. Changes in the heart rate were insignificant at left side SWCNT injections, while there right side administration resulted in heart rate reduction. Biochemical analysis of the blood samples to evaluate the erythrocyte stability to acid hemolysis after SWCNT either intravenous or intramedullary administration showed that SWCNTs did not have toxic effect on the cardiovascular system. Conclusions: The data obtained suggest that SWCNTs are the promising class of pharmaceutical compounds to treat hypertension. Considering that the cardiovascular diseases, including hypertension, are often associated with excessive activation of the sympathetic nervous system, hypotensive effect of SWCNTs on the cardiovascular system can be realized via reducing the activity of the medullary sympathetic neurons through activation of nitric oxide synthesis. In addition, SWCNTs may be involved in inactivation of excessive free radical production observed in hypertension.
S481

T381 EVALUATION OF THE PERFORMANCE OF THE VIDAS ATG ASSAY IN SUBJECTS WITH CONFIRMED OR SUSPECTED THYROID AUTOIMMUNE DISEASE E.J. Rivers(1), F.S. Apple(2), T. Davis(3), A.W. Butch(4), D. Fuller(3), R. Buckner(3), J. Tolaini(1), B. Rice(1) bioMerieux, Inc., Durham, NC 27712 Hennepin County Medical Center, Minneapolis, MN 3 Wishard Health Services, Indianapolis, IN 4 UCLA Medical Center, Los Angeles, CA
2 1
T382 THERAPEUTIC DRUG MONITORING OF MYCOPHENOLIC ACID USING PHARMACOKINETIC MODELLING AND BAYESIAN ESTIMATION WITH THE MPAT DIMENSION ASSAY D. Richard(1), F. Libert(1), L. Roche(1), l. Samson(2), J. Debord(3), F. Saint-Marcoux(3)
1
Background: Anti-thyroglobulin (Anti-Tg) antibodies are often found in conjunction with anti-thyroid peroxidase antibodies in Hashimotos thyroiditis and Graves disease. The VIDAS ATG (bioMrieux, France) assay is an automated quantitative test for the detection of IgG anti-Tg autoantibodies in serum or plasma for use as an aid in the diagnosis of autoimmune thyroid disease. The study objective was to establish the performance of the VIDAS ATG assay compared to the Architect Anti-Tg assay (Abbott, USA) using samples from patients with Hashimotos thyroiditis or Graves disease, samples submitted for routine thyroid disease testing, and samples from subjects with non-thyroid autoimmune disease. Methods: The cut-off used for each assay was determined by bioMrieux according to National Academy of Clinical Biochemistry criteria. Single replicates of 317 samples were tested using the VIDAS ATG and Architect Anti-Tg assays. Sensitivity, positive percent agreement (PPA), negative percent agreement (NPA), and overall percent agreement (OPA) were determined. Results: The VIDAS ATG and Architect Anti-Tg demonstrated PPA of 90%, NPA of 96%, and OPA of 94%. In samples from subjects with Hashimotos thyroiditis, the VIDAS ATG and Architect Anti-Tg demonstrated 98% PPA, 93% NPA and 97% OPA. In samples from subjects with Graves disease, the VIDAS ATG and Architect Anti-Tg assays demonstrated PPA of 97% and NPA of 93% and OPA of 95%. The sensitivity of both assays in Hashimotos thyroiditis subjects was 77%. The sensitivities of the VIDAS ATG and Architect Anti-Tg assay in Graves disease subjects were 47% and 44%, respectively. Conclusions: The VIDAS ATG assay demonstrated equivalent sensitivity to the predicate device and high PPA, NPA, and OPA in all of the populations examined including subjects with Hashimotos thyroiditis, Graves disease, non-thyroid autoimmune disease and routine thyroid testing samples. Thus, the VIDAS ATG assay provides a rapid (25 minute) test for the presence of Anti-TG in support of a diagnosis of autoimmune thyroid disease. This product has not been evaluated by the FDA, and is not intended for sale in the United States
Department of Pharmacology Toxicology Clermont-Ferrand University Hospital, France 2 SIEMENS Healthcare Diagnostics, Saint-Denis (France) 3 Department of pharmacology Toxicology, Limoges University Hospital, France Mycophenolic acid (MPA) is the antimetabolite of choice in immunosuppressive protocols. MPA exhibits large interindividual pharmacokinetic variability due to numerous factors, such as liver and renal functions, serum albumin levels or associated drugs; (ii) MPA AUC0-12 h is better correlated with patient outcomes than any single concentration timepoint. Different consensus conferences have advised: (i) to perform drug dosing based on MPA inter-dose AUC when obtained using limited sampling strategies (LSS); (ii) to target an MPA AUC0-12h between 30 and 60 mg.h/L. For MPA, multiple Maximum a Posteriori Bayesian estimators (MAP-BEs) were previously developed and routinely used for the dose adjustment of MPA in transplant patients based on a limited sampling strategy (LSS) using 3 concentrations. The aims of the present study were to further evaluate the new Dimension MPAT assay compared with a reference HPLC-DAD technique for MPA PK studies in three different populations of heart, liver and kidney allograft, and use BE models for optimizing TDM with these two analytical methods. Plasma samples were obtained from 103 transplant patients. Patients plasma samples were thawed in batches and analyzed in parallel. A hybrid MAP-BEs specific to the MPAT assay was developed and validated. Precisely: (i)the error patterns of the MPAT assay were determined for each transplant type according to the recommendations, (ii)the equations derived from the regression analysis between the MPAT kit and HPLC were determined for each group of patients, (iii)100 MPA PK profiles were simulated for each group, (iv) each MAP-BE was evaluated by comparison of observed (ie trapezoidal rule) and estimated AUC using the predefined LSS and determination of the bias and Root Mean Squared Error (RMSE). 127 full kinetic profiles have been analysed. Correlation between different data were r = 0.9662, r = 0.9442 and r = 8766 for heart graft, liver graft and renal graft respectively. For each MAP-BE the mean bias between the estimated AUC and AUCtrap did not exceed 6%, with RMSE values less than 15%. The number of patients with an imprecision greater than 20% was less than 15%. We concluded that HPLC-DAD and the Dimension MPAT assay could be used for routine dose adjustments of MPA with MPABEs.
S482

T383 PERFORMANCE EVALUATION OF A NEW FERRITIN ASSAY FOR THERMO SCIENTIFIC KONELAB CLINICAL CHEMISTRY ANALYZER S. Riistama-Laari, M. Karppelin, S. Tikanoja, H. Lampinen Thermo Fisher Scientific, Vantaa, Finland Background: The determination of ferritin is important when diagnosing iron metabolism disorders, monitoring iron therapy, ascertaining the iron reserves in groups at risk and in differential diagnosis of anemias. While very low serum ferritin values are always indicative of iron deficiency, very high serum ferritin values have many implications. Increased serum ferritin can be suggestive of iron overload but is also seen in conjunction with liver parenchymal damage, infections, inflammation and malignant diseases without any quantitative relationship to the iron reserve. Thermo Scientific Konelab clinical chemistry analyzers (20, 20XT, PRIME 30, PRIME 60) are random access, fully automated, clinical chemistry systems. Colorimetric, turbidimetric and ISE methods are applicable and the analyzers are capable of handling routine and stat requests. Methods: Konelab (20XT, PRIME 30, PRIME 60) Ferritin method with system reagents, calibrators and controls is a particle enhanced immunoturbidimetric assay using latex particles coated with rabbit antibodies against human ferritin. Serum and Li-heparin can be used as sample types. The increase in absorbance caused by formation of immunocomplexes is recorded at 700 nm. Results: The assay measuring range is 7 350 g/L extended with automatic dilution up to 3500 g/L. The determination limit of the assay is 7 g/L. The repeatability (within-run precision) is from 2.5 to 4.4% (CV) for samples with ferritin concentrations from 28 to 270 g/L (N=80). The within device (total) precision is from 5.7 to 6.6% (CV) for samples with ferritin concentrations from 28 to 270 g/L (N=80). A method comparison study was performed using a commercially available particle enhanced immunoturbidimetric method as the reference. The Konelab method correlated well with the reference method. Linear regression was y = 1.02 x + 0.32 and r = 0.985 (N=98). Conclusion: The results demonstrate that ferritin can be analyzed accurately and easily using Thermo Scientific Konelab clinical chemistry analyzer. T384 PERFORMANCE EVALUATION OF A NEW D-DIMER ASSAY FOR THERMO SCIENTIFIC KONELAB CLINICAL CHEMISTRY ANALYZER S. Riistama-Laari, M. Karppelin, S. Tikanoja, H. Lampinen Thermo Fisher Scientific, Vantaa, Finland Background: D-Dimer is a small protein fragment, present in the blood after a blood clot is degraded in fibrinolysis by plasmin. Although many clinical conditions are associated with increased blood concentrations of D-Dimer, its testing has become a useful laboratory tool for the diagnosis of venous thromboembolism (VTE) because it has high negative predictive value when used in combination with pretest clinical probability. Other clinical conditions related to increased concentrations of D-Dimer are arterial thrombosis (including myocardial infarction and stroke), disseminated intravascular coagulation (DIC), recurrent thrombotic risk following anticoagulation, post operative state, significant liver disease, malignancy and normal pregnancy. Thermo Scientific Konelab clinical chemistry analyzers (20, 20XT, PRIME 30, PRIME 60) are random access, fully automated, clinical chemistry systems. Colorimetric, turbidimetric and ISE methods are applicable and the analyzers are capable of handling routine and stat requests. Methods: Konelab (20XT, PRIME 30, PRIME 60) D-Dimer method with system reagents, calibrators and controls is a particle enhanced immunoturbidimetric assay using latex particles coated with mouse anti-human D-Dimer monoclonal antibodies. Sample type is citrate plasma and the increase in absorbance caused by formation of immunocomplexes is recorded at 600nm. Results: The assay measuring range is 0.2 5.0 mg FEU/l, extended with automatic dilution up to 20 mg FEU/l. The determination limit of the assay is 0.2 mg FEU/l. Expected values <0.5 mg FEU/l. The repeatability (within-run precision) is from 0.9 to 2.9% (CV) for samples with D-Dimer concentrations from 0.57 to 3.85 mg FEU/l (N=80). The within device (total) precision is from 3.9 to 6.5% (CV) for samples with D-Dimer concentrations from 0.57 to 3.85 mg FEU/l (N=80). A method comparison study was performed using a commercially available particle enhanced immunoturbidimetric method as the reference. The Konelab method correlated well with the reference method. Conclusion: The results demonstrate that D-Dimer can be analyzed accurately and easily using Thermo Scientific Konelab clinical chemistry analyzer.
S483

T385 EVALUATION OF THE PERFORMANCE OF THE VIDAS ATPO ASSAY IN SUBJECTS WITH CONFIRMED OR SUSPECTED THYROID AUTOIMMUNE DISEASE E.J. Rivers(1), F.S. Apple(2), T. Davis(3), A.W. Butch(4), D. Fuller(3), R. Buckner(3), J. Tolaini(1), B. Rice(1) bioMerieux, Inc., Durham, NC 2 Hennepin County Medical Center, Minneapolis, MN 3 Wishard Health Services, Indianapolis, IN 4 UCLA Medical Center, Los Angeles, CA Background: Anti-thyroid peroxidase (Anti-TPO) antibodies contribute to thyroid autoimmune disease, a major factor underlying hypothyroidism and hyperthyroidism. The VIDAS ATPO (bioMrieux, France) assay is an automated quantitative test for the detection of IgG anti-TPO autoantibodies in serum or plasma for use as an aid in the diagnosis of autoimmune thyroid disease. The study objective was to establish the performance of the VIDAS ATPO assay compared to the Architect Anti-TPO assay (Abbott, USA) using samples from patients with Hashimotos thyroiditis or Graves disease, samples submitted for routine thyroid disease testing, and samples from subjects with non-thyroid autoimmune disease. Methods: Clinical cut-off for the VIDAS ATPO assay was determined according to National Academy of Clinical Biochemistry criteria (package insert value was used for Architect Anti-TPO assay). Single replicates of 317 samples were tested using the VIDAS ATPO and Architect Anti-TPO assays. Sensitivity, positive percent agreement (PPA), negative percent agreement (NPA), and overall percent agreement (OPA) were determined. Results: The VIDAS ATPO and Architect Anti-TPO demonstrated PPA of 74%, NPA of 96%, and OPA of 95%. In samples from subjects with Hashimotos thyroiditis, the VIDAS ATPO and the Architect Anti-TPO demonstrated 98% PPA, 86% NPA, and 97% OPA. In samples from subjects with Graves disease, the VIDAS ATPO and Architect Anti-TPO assays demonstrated PPA of 94%, NPA of 96%, and OPA of 95%. The sensitivity of both assays in Hashimotos thyroiditis subjects was 88%. The sensitivities of the VIDAS ATPO and Architect Anti-TPO assay in Graves disease subjects were 62% and 65%, respectively. Conclusions: The VIDAS ATPO assay demonstrated sensitivity comparable to that of the predicate device and high level agreement, in all of the populations examined including subjects with confirmed Hashimotos thyroiditis, Graves disease, or non-thyroid autoimmune disease and samples submitted for routine thyroid testing. Thus, the VIDAS ATPO assay provides a rapid (25 minute) test for the presence of AntiTPO in support of a diagnosis of autoimmune thyroid disease. This product has not been evaluated by the FDA, and is not intended for sale in the United States.
1
T386 HIGH GRADIENT MAGNETIC SEPARATION: A NOVEL METHOD IN THE SEPARATION OF UPCONVERTING LANTHANIDE-DOPED NANOPARTICLES O. Salovaara, R. Arppe, L. Mattsson, S. Lahtinen, T. Soukka Division of Biotechnology, University of Turku, Finland Background: High gradient magnetic separation (HGMS) is a widely used method from mining industry to clinical research. In biotechnology it has been used to collect and concentrate biomolecules and cells by binding them onto magnetic nanoparticles. Upconverting nanoparticles (UCNPs) are unique reporters utilized in assay and imaging technologies. They have a luminescent property of converting low-energy infrared light into visible emission. However, as the particle size of the UCNPs approaches the size of biomolecules, the handling of reporters becomes cumbersome with traditional purification methods. Methods: While the lanthanide-doped crystal structure of UCNPs is responsible for the upconversion luminescence, the lanthanide dopants also bring paramagnetic properties to the UCNPs enabling the use of HGMS as a method to capture them and their conjugates. The separation system consists of a pair of permanent neodymium super magnets and a column structure filled with ferromagnetic matrix (steel wool) placed in between them. The matrix is saturated with the magnetic field produced by the magnets and forms strong magnetic gradients on its surface. In these gradients the field is strong enough to capture even weakly paramagnetic UCNPs. Results: The working principle of HGMS with UCNPs was demonstrated by injecting the UCNPs to the system and using different magnetic field strengths to capture the particles from the sample. The system was able to capture more UCNPs with a stronger magnetic field. The magnetic selectivity of HGMS was demonstrated by introducing a mixture of non-magnetic blue latex particles and UCNPs to the system, which resulted in latex particles flowing through in the presence of the magnetic field while the UCNPs were eluted after the magnets were removed. Conclusions: By utilizing methods from industry applications we have developed a working solution for separation of UCNPs from liquids relying solely on the intrinsic paramagnetic properties of the lanthanide dopants. In this method there is no need to embed separate, optically inactive magnetic materials inside the UCNPs as the particle structure is paramagnetic by itself. The results indicate that HGMS could also be used with antibody-conjugated UCNPs used in bio-assays.
S484

T387 VALIDATION OF TWO NEW PNEUMATIC TRANSPORTATION SYSTEMS FOR BLOOD SAMPLES, COVERING HIGHEST(13 METER) AND LONGEST(500 METER) DISTANCE TO DATE. THE SYSTEM IS EASY TO HANDLE, AND THE SAMPLES ARE IN FEW SECONDS DELIVERED, READY TO PLACE AT THE PREANALYTICAL SYSTEM, CONNECTED TO THE CHEMICAL ANALYZERS A.D. Schroeder Aarhus University Hospital, Department of Clinical Biochemistry NBG, Denmark Presenting Author Anne Dilani Schroeder Background: The challenge is to quarantee fast analysis time for automated chemical analyzers. To optimize the proces rapid sample delivery via pneumatic tube transportation is attractive for reducing transport time to the laboratory. Methods: After drawing the blood sample, the tubes are placed in the Tempus600 rack and loaded in the Tempus600 launch unit, without packaging compared to conventional systems. The launch unit is placed in the emergency department and the point of delivery is in the laboratory, Department of Clinical Biochemistry: The Tempus600 system was examined with a view to assess to what extend the blood samples were affected by the transport. 50 patients participated in the study, chosen randomly among patients in the departments phlebotomy. Two blood samples were drawn from each patient. One sample tube was sent by routine courier transport to the laboratory (reference) and the second one was sent with Tempus600. The blood samples were tested for: Biochemical tests (potassium, lactatedehydrogenase, alkaline phoshatase, haemolytic index), Coagulation tests (international normalized ratio, activated partial tromboplastin time) and Haematology test (leucocytes, lymphocytes and trombocytes), using Roche Cobas6000, Stago STA-R and Sysmex XE2100 analyzers. Results: No significant differences were found between routine transport and Tempus600 for the tests (P-value >0,05), except the haemolytic index. When transporting the blood sample tubes using Tempus600 the hemolytic-index went up approx. 2,3 mg/dL compared to routine transport (min. -7 to max. 10 mg/dL). This increase did not affect the results, and the haemolytic-index were accepted for the blood sample testing. Results from the 500 meter pneumatic system are still preliminary and will be presented at the congress. Conclusions: The validation of Tempus600 has shown very satisfying results. Based on the findings it is recommended using Tempus600 for transporting blood sample tubes. T388 NEW LIQUID STABLE NA+-, K+- AND CL- ASSAYS FOR THE FAST AND EASY ASSESSMENT OF ELECTROLYTES ON CLINICAL CHEMISTRY ANALYZERS P. Schu(1), S. Hubbard(1), C. Fache(1), D. Thoenges(1), P. Mathe(2), D. Kaufman(2), E. Metzmann(1), M. Grimmler(1), T. Hektor(1)
2 1
DiaSys Diagnostic Systems GmbH, Holzheim, Germany Specialty Assays, Inc., Hillsborough, NJ
In all higher life forms a precise maintenance of an electrolyte balance is crucial to regulate pH and hydration of the body and is critical for nerve and muscle function. In patients electrolytes such as Na+, K+ and Cl- are frequently measured as part of the clinical routine to evaluate acute or chronic diseases and to monitor treatment of certain problems. This includes high blood pressure, heart failure, liver and kidney diseases or diabetesrelated complications. Here we present a new ready-to-use electrolyte reagent panel to measure Na+, K+ and Cl- on clinical chemistry analyzers (CCA). Established methods for the determination of electrolytes are flame emission spectroscopy (FES) and potentiometry with ion selective electrodes (ISE). These systems are cumbersome to integrate into routine testing or require a lot of regular maintenance to ensure a reliable performance. The liquid-stable DiaSys electrolyte tests are optimized for laboratories with small or midsized clinical analyzers without an ISE. The assessment of values presented here, are based on ion-dependent enzymes for Na+ and K+ and on a new colorimetric method for the detection of Cl- respectively. In particular sodium levels are detected using a Na+-dependent -galactosidase. This reaction is assessed by enzymatic release of o-nitrophenol from its substrate. Potassium is detected by a K+-dependent pyruvate kinase connected to a lactate dehydrogenase/NADH system. In contrast chloride values are assessed by a new colorimetric method using a specific Cldependent iron(III) chloride-complex. All three tests show a wide linear range and allow the robust determination of electrolyte values in serum or plasma samples on routine CCA without prior dilution. Using a DiaSys respons920 analyzer the tests demonstrated very strong correlation to the reference methods and an extraordinary precision of <2% within run and <3% between day. No significant interferences within 3% (Na+) and 4.5% limits (K+, Cl-) are given for all test assays. Our results demonstrate, that the DiaSys Na+, K+ and Cl- reagent panel offers a great opportunity for small and mid-sized labs to automate routine electrolyte diagnosis. The reagents can be used manually as well as on CCAs with a comparable performance to ISE or FES.
S485

T389 IMMUNOSUPPRESSANT THERAPEUTIC DRUG MONITORING WITH HPLC-MS/MS ARE ALL METHODS EQUAL? C. Seger, J. Taibon, A. Griesmacher Institute of Medical and Chemical Laboratory Diagnostics (ZIMCL), University Hospital Innsbruck Background: HPLC-MS/MS is often considered as golden standard method in clinical biochemistry. Its selectivity is praised, its accuracy, precision and sensitivity is considered to be equal or better than clinical chemistry routine methodologies in endocrinology or therapeutic drug monitoring (TDM); especially compared to immunoassays. On the other side however, it has been shown in immunosuppressant drug TDM proficiency testing (PT) schemes, that most often locally designed LC-MS/MS platforms show a higher degree of inter laboratory variability than automated commercial immunoassays. To allow a deeper insight into the factors contributing to the observed imprecision of LC-MS/MS assays, we combined published and widely used sample preparation protocols with the online-SPE-LC-MS/MS immunosuppressant drug TDM assay designed at our institute (ZIMCL method). Methods: Both qualitative and quantitative experiments were performed. Anonymized leftover routine whole blood samples were used. Extraction yields were evaluated by peak area comparisons for cyclosporine A, tacrolimus, everolimus and sirolimus. The corrective action of the routinely used internal standards (IS) in a quantitative assay setup was investigated for cyclosporine A (IS cyclosporine D) and tacrolimus (IS ascomycin) using identical calibrator materials. Results: The peak area comparison experiment did clearly show that several published sample preparation protocols had significantly lower analyte yields compared to the ZIMCL method. Comparison of extraction yields from calibrator and patient sample materials did unveil, that only two of eight protocols investigated did not discriminate these materials. Quantitative results comparison to the ZIMCL method did show significant deviations (10% to 30%) for some but not all of the sample preparation protocols. Conclusion: The increased inter-laboratory variability observed in PT schemes stems most likely from different extraction yields in the whole blood sample preparation and not from an interlaboratory calibration bias. Erythrocyte lysis prior to protein precipitation seems to be mandatory to achieve an identical analyte yield from patient and calibrator samples. T390 SERUM FREE LIGHT CHAIN ASSESSMENTS: COMPARISON OF PRECISION AND LINEARITY H. Sharrod-Cole, D. Matters, P. Showell, S. Harding The Binding Site Ltd, Birmingham, UK Background: Serum free light chain measurements utilising the polyclonal antisera based immunoassay (FreeliteTM, Binding Site, UK) have changed the diagnostic and monitoring paradigm for patients with B cell disorders; culminating in their inclusion in international guidelines. Recently, a monoclonal antibody immunoassay has become available (N latex FLC, Siemens, Germany). Here we compare the within batch imprecision and linearity. Methods: Comparison of the Freelite and N latex FLC assays was performed on a BNII nephelometer (Siemens, Germany). Precision was measured at 140-200% of the lower calibration range and 80-95% of the upper calibration points at standard dilutions. Due to different dynamic measuring ranges the concentrations of the analytes measured were: for FreeliteTM low 8.75-12.5 mg/L and 7.1-10.1mg/L, high 160-190 mg/L and 130.24-154.7 mg/L; N latex low 5.367.66 mg/L and 2.62-3.75 mg/L, high 48-57 mg/L and 98-116.4 mg/L. To assess linearity, monoclonal samples were analysed in their undiluted states in triplicate, and allowed to repeat as required (start dilutions for Freelite and were 1/100 and N latex and were 1/100 and 1/20 respectively). Dilutions were performed with Siemens N Diluent to achieve a dilution series of 0-100%. Results: Within batch CVs for lower end and upper end precision respectively were: for Freelite 5.1% and 4.1%, for 6.3% and 3.5%; for N latex FLC 1.3% and 3.3% and for 3.8% and 1.9%. Following linearity analysis the R2 values for Freelite and were 0.9881 and 0.9929 whereas for N latex R2 values were 0.9429 and 0.94. The maximum difference between the observed and expected values was -38.3% for N latex but only -18.3% for Freelite. The change between the observed and expected values shifted from -38.3% to 25.1% as measured concentrations exceeded the standard dilution on the N latex assay. Conclusion: Both the FreeliteTM and N latex FLC and assays demonstrate acceptable precision at the lower end of their respective measuring ranges at the standard dilution. Freelite showed a greater degree of linearity than the N latex assay, possibly reflecting matrix interference.
S486

T391 EVALUATION OF A C1 INACTIVATOR ASSAY FOR USE ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER S.K.I. Silvera, S.W. Chan, A. Kaur, H. Sharrod-Cole, S.J. Harding, P.J. Showell The Binding Site Ltd, Birmingham, UK Background: C1 Inactivator is an important regulator of the classical complement pathway. It acts by inhibiting the activity of C1s and C1r serine proteases, therefore limiting any potentially harmful effects of over-activation. Measurement of C1 Inactivator aids in the diagnosis of hereditary and acquired angioedema which are the most common complement deficiencies. Here we describe the performance characteristics of a C1 Inactivator assay for use on the Binding Sites next generation protein analyser. The instrument is a randomaccess bench top turbidimetric analyser capable of a wide range of on-board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour. Precision is promoted by single-use cuvettes which are automatically loaded and disposed of, whilst the utility is enhanced through host interface capability, primary sample ID and bar coded reagent management systems. The assay has a measuring range of 0.06 0.4 g/L at the standard 1/10 sample dilution, with sensitivity of 0.06 g/L. High samples are remeasured at a dilution of 1/20 with an upper measuring range of 0.12 0.8 g/L. Methods: Correlation to the Binding Site C1 Inactivator assay for the SPA PLUS was performed using 28 samples (range 0.081 0.406 g/L). Intra-run precision was assessed by measurement of ten replicates of samples at 0.108 g/L and 0.343 g/L. Furthermore, precision was assessed at the clinically relevant decision point of 0.169 g/L. Linearity was assessed by assaying a serially-diluted patient sample pool across the width of the measuring range and comparing expected versus observed results (0.068 0.338 g/L). Results: Correlation with the C1 Inactivator SPA PLUS assay demonstrated good agreement when analysed by PassingBablok regression; y=0.94x + 0.01. The assay was shown to be linear over the range of 0.068 0.338 g/L; y = 0.9526x + 0.01795 (R2 = 0.9961), maximum deviation from expected result was 15.8%. Intra-run precision produced the following results: sample 1 (0.108 g/L) CV of 0.85%, sample 2 (0.169 g/L), CV of 0.98% and sample 3 (0.343 g/L), CV of 0.59%. Conclusions: We conclude that the C1 Inactivator assay for the Binding Site next generation protein analyser is reliable, accurate and precise and shows good agreement with existing assays. T392 EVALUATION OF A CYSTATIN C ASSAY ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER S.L. Stone, A.J. Alvi, H. Sharrod-Cole, S.J. Harding, P.J. Showell The Binding Site Group Ltd, UK Background. Cystatin C is a 13kDa, non-glycosylated endogenous cysteine protease inhibitor. Here we describe the performance characteristics of a Cystatin C assay for use on the Binding Sites next generation protein analyser. The instrument is a random-access bench top turbidimetric analyser capable of a wide range of on-board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour. Precision is promoted by single-use cuvettes which are automatically loaded and disposed of, whilst the utility is enhanced through host interface capability, primary sample ID and bar coded reagent management systems. The assay has a measuring range of 0.402 8.04 mg/L at the standard 1/10 sample dilution, with a sensitivity of 0.04 mg/L. High samples are remeasured at a dilution of 1/20 with an upper measuring range of 0.804 - 16.08 mg/L. Methods. Correlation to the Binding Site Cystatin C assay for the SPA PLUS was performed using 35 samples (range 0.619 7.910). Intra-run precision was assessed by measurement of twenty replicates of samples at 6.2, 3.4, 1.0, 0.75 and 0.64 mg/L. Linearity was assessed by assaying a serially-diluted patient sample pool across the width of the measuring range and comparing expected versus observed results (0.402 8.04 mg/L). Antigen excess capacity was determined by extending the width of the curve through increasing calibrator concentration and looking for the hook effect. Results. Correlation with the Cystatin C SPA PLUS assay demonstrated good agreement when analysed by linear regression; y=0.94x + 0.004, R2 =0.996. The assay was shown to be linear over the range of 0.342 8.451 mg/L; y=0.9929x 0.1262, R2 =0.996, maximum deviation from expected result was 10.0%. Intra-run precision produced the following results: sample 1 (6.275 mg/L) CV of 1.81%, sample 2 (3.420 mg/L), CV of 1.50%, sample 3 (1.040 mg/L), CV of 1.99%, sample 4 (0.748 mg/L), CV of 2.26% and sample 5 (0.636), CV of 4.04%. The assay was proven to be antigen-excess proof to 46.53 mg/L (6 times higher than seen in chronic kidney disease stage 5). Conclusions. We conclude that the Cystatin C assay for the Binding Site next generation protein analyser is reliable, accurate and precise and shows good agreement with existing assays.
S487

T393 EVALUATION OF AN IGG ASSAY FOR USE ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER S. Kausar, A.J. Alvi, P.J. Showell, S.J. Harding The Binding Site Group Ltd, Uk Background: In normal adults, IgG constitutes approximately 75% of total serum immunoglobulin. IgG levels are decreased in primary immunodeficiency conditions, whilst levels are elevated in conditions such as multiple myeloma. Here we describe the performance characteristics of an IgG assay for use on the Binding Sites next generation protein analyser. The instrument is a random-access bench top turbidimetric analyser capable of a wide range of on-board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour. Precision is promoted by single-use cuvettes which are automatically loaded and disposed of, whilst the utility is enhanced through host interface capability, primary sample ID and bar coded reagent management systems. Methods: The assay has a measuring range of 1.65-36.40 g/L at the standard 1/10 sample dilution, with a sensitivity of 0.16 g/L. High samples are remeasured at a dilution of 1/40 with an upper measuring range of 6.60-145.60 g/L. Correlation to the Binding Site IgG assay for the SPA PLUS was performed using 31 samples (range 1.04-31.8 g/L). Intra-run precision was assessed by measurement of twenty replicates of samples at 33.1, 15.5, 6.3 and 3.1 g/L. Linearity was assessed by assaying a serially-diluted patient sample pool across the width of the measuring range and comparing expected versus observed results. Antigen excess capacity was determined by extending the width of the curve through increasing the calibrator concentration and looking for the hook effect. Results: Correlation with the IgG SPA PLUS assay demonstrated good agreement when analysed by PassingBablok regression: y=0.95x + 0.07. The assay was shown to be linear over the range of 1.9-39.3 g/L; y=1.02x + 0.03 (R2=0.9989), maximum deviation from expected result was 9.4%. Intra-run precision produced the following results: sample 1 (33.1 g/L) CV of 1.16%, sample 2 (15.5 g/L), CV of 1.98%, sample 3 (3.1 g/L), CV of 2.15% and sample 4 (6.3 g/L), CV of 1.55%. The assay was antigen-excess proof to 195.63 g/L. Conclusions: We conclude that the IgG assay for the Binding Site Next Gen is reliable, accurate and precise and shows good agreement with existing assays. T394 EVALUATION OF A C3C ASSAY FOR USE ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER L.W. Aston-Abbot, F. Murphy, P.J. Showell, S.J. Harding The Binding Site Group Ltd, UK Introduction: Serum complement consists of around 30 proteins that have a fundamental role in immune system functionality. Inherited deficiencies in C3c are associated with an increased risk of developing systemic lupus erythematosus (SLE). C3c deficiency may present with recurrent infections such as pneumonia, septicaemia and meningitis. Here we describe the development of serum C3c assay for Binding Sites Next Generation Protein Analyser. The instrument is a randomaccess bench top turbidimetric analyser capable of a wide range of on-board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour. Precision is promoted by single-use cuvettes which are automatically loaded and disposed of, whilst the utility is enhanced through host interface capability, primary sample ID and bar coded reagent management systems.The assay has a measuring range of 0.253.0 g/L at the standard 1/10 sample dilution, with sensitivity of 0.025 g/L. High samples are remeasured at a dilution of 1/20 with an upper measuring range of 0.5-6 g/L Method: Correlation to the Binding Site C3c assay for the SPA PLUS was performed using 49 samples (mean 1.214; range 0.492.15 g/L). Precision studies (CLSI EP5-A2) were performed at three levels in duplicate over 21 working days. Antigen levels of 2.418, 0.778 and 0.38 g/L were assessed for total, within-run, between-run and between-day precision, using one lot of reagent on three analysers. Linearity was assessed by assaying a serially-diluted patient sample pool across the width of the measuring range (0.3033.036 g/L)and comparing expected versus observed results. Results: Correlation with the C3c SPA PLUS assay demonstrated good agreement when analysed by PassingBablok regression; 1.02x + 0.05 g/L. The assay was shown to be linear over the range of 0.3033.036g/L; y=1.03x+0.017 g/L (R = 0.998). Precision produced the following results for total, within-run, between-run and between-day precision respectively: sample 1 (2.418 g/) 3.1%, 1.0%, 1.3% and 2.6%, sample 2 (0.778 g/L) 0.4%, 2.0%, 2.4% and 1.4%, sample 3 (0.38 g/L) 3.3%, 1.1% 1.3% and 2.8%. Conclusions: We conclude that the C3c assay for the Binding Site next generation protein analyser is reliable, accurate and precise and shows good agreement with existing assays.
S488

T395 EVALUATION OF AN IGG4 ASSAY FOR USE ON THE BINDING SITE NEXT GENERATION PROTEIN ANALYSER L.D. Southan, R.E. Grieveson, A. Kaur, S.J. Harding, P.J. Showell The Binding Site Group Ltd, UK Background: Recent research has shown that high polyclonal IgG4 levels are associated with autoimmune pancreatitis (AIP). Here we describe the performance characteristics of an IgG4 assay for use on the Binding Sites next generation protein analyser. The instrument is a random-access bench top turbidimetric analyser capable of a wide range of on-board sample dilutions (up to 1/10,000) and throughput of up to 160 tests per hour. Precision is promoted by single-use cuvettes which are automatically loaded and disposed of, whilst the utility is enhanced through host interface capability, primary sample ID and bar coded reagent management systems. Methods: The assay measuring range was 0.033 g/L at the standard 1/25 sample dilution, with sensitivity of 0.003 g/L. High samples were remeasured at a dilution of 1/500 with a measuring range of 1.560 g/L. Correlation to the Binding Site IgG4 assay for the SPA PLUS was performed using 16 samples (range 0.042.39 g/L). Antigen excess protection was assessed by measuring 18 analyte concentrations ranging 0.0382.37 g/L at the minimum sample dilution (equivalent to 0.9659.2 g/L at the standard 1/25 sample dilution). Intra-run precision was assessed by measurement of twenty replicates of samples; low (0.04 g/L), medium (1.23 g/L) and high (2.39 g/L). Linearity was assessed by assaying a serially-diluted patient sample pool and comparing expected versus observed results (range 0.032.39 g/L). Results: There was good agreement between assays (mean 0.56 g/L; range 0.042.37 g/L v mean 0.55 g/L; range 0.022.39 g/L), Passing-Bablok regression; y=0.98x + 12.44. The assay was linear over the range of 0.032.39 g/L; y=1.07x + 1.01 (R2=0.99), maximum deviation from expected was 15.9%. Intra-run precision produced the following results: low (0.04 g/L) CV of 6.82%, medium (1.23 g/L) CV of 2.66% and high (2.39 g/L) CV of 2.61%. All samples up to 2.37 g/L (equivalent to 59.2 g/L at the standard sample dilution) were correctly flagged by the antigen excess protection function and did not report results in antigen excess. Conclusions: We conclude that the IgG4 assay for the Binding Site next generation protein analyser is reliable, accurate and precise and is protected from antigen excess when challenged with the high IgG4 levels seen in AIP. T396 EVALUATION OF CUTOFF VALUE FOR HIGH-SENSITIVE TROPONIN T ASSAY IN RELATION TO PREVIOUSLY USED TROPONIN I ASSAY S. Hrabric Vlah, V. Supak Smolcic Clinical Department of Laboratory Diagnostics, Clinical Hospital Center Rijeka, Rijeka, Croatia Background: Cardiac troponin is considered a leading laboratory diagnostic tool for detection of myocardial damage. Until recently main diagnostic assay was measuring cardiac troponin I (cTnI) levels. However, development of highsensitive troponin T (hsTnT) assay has made its use available. The aim of this study was to determine cutoff value for hsTnT assay, which was introduced in our laboratory, in comparison to previously used cTnI assay. Methods: Total of 401 patient samples from routine laboratory workload was simultaneously analyzed for hsTnT (Roche, Basel, Switzerland) and cTnI (Siemens, Newark, USA) levels. Blood samples were collected in BD Vacutainer tube with gel separator and centrifugated for 10 minutes at 3500 rpm at least 30 minutes after collection and before analysis. Cutoff value of 0.2 g/L for cTnI was determined upon introducing method into routine laboratory work according to manufacturer recommendation and our internal validation study. Cutoff value for hsTnT was evaluated using receiver operating characteristic (ROC) curve with MedCalc statistical software (Mariakerke, Belgium) using cTnI cutoff value as differential criteria. Results: Pearsons correlation coefficient for hsTnT and cTnI measurements was r=0.939 (P <0.001; 95% CI 0.927-0.949) which indicates excellent correlation between tests. The area under the ROC curve (AUC) was 0.982 (P <0.001; 95% CI 0.964-0.993) for hsTnT assay. The ROC curve gained cutoff value for hsTnT of 0.049 g/L, with sensitivity of 96.2% (95% CI 89.2-99.2) and specificity of 91% (95% CI 87.4-93.9). Conclusion: Results demonstrated that the hsTnT concentration of 0.049 g/L could be used as cutoff value for detection of myocardial damage in emergency department. We find this cutoff in accordance with manufacturer recommendation (0.053 g/L). Therefore, introduction of new hsTnT method using proposed cutoff will not influence consistency of medical decisions.
S489

T397 VALIDATION OF A METHOD TO RULE OUT URINE ALBUMIN DETERMINATIONS S. Soto Fernndez(1), S.A. Lojo Rocamonde(2) M.D., staff technical specialist. Urinalysis Section, Central Laboratory. Clinical Universitary Hospital, Health Government Services. Santiago de Compostela, Spain 2 M.S.Chem., Ph.D., EurClinChem., chief. Urynalisis Department, Central Laboratory. Clinical Universitary Hospital, Health Government Services. Santiago de Compostela, Spain Aim: New generation dipsticks offer the albumin/protein-tocreatine ratio. We need to know their analytical behavior and cost-effectiveness. Question: Is it possible to avoid the exact (reference) determination of ACR/PrCR if dipstick value (field) is the semiquantitative maximum result? Methods (Siemens Healthcare Diagnostics): 1. Strip (proteinto-creatinine ratio, sqPrCR): Multistix PRO-12 in a Clinitek-ATLAS reader. Maximum category: 500 mg protein/g creatinine. 2. Creatinine: photometry in an ADVIA-2400. Jaff kinetic method with ID-MS calibration. 3. Albumin (albumin-tocreatinine ratio, ACR): nephelometry in a BN-II. Mouse monoclonal antibody. 4. Protein (protein-to-creatinine ratio, PrCR): nephelometry in a BN-II. TCA/20% aqueous solution. SAMPLES (from 01.01.09 until 31.10.12). Number of urine with sqPrCR strip results: 474932. Number or urine with PrCR/ACR results (reference): 8475. Female: 3597. Male: 4878. Number of urine with sqPrCR 500 mg/g: 1285. Female: 431. Male: 854. Results: a. JNC7 criteria number of urine samples with maximum sqPrCR by dipstick n=1285 PrCR ACR Normal <31 mg/g 0 0% 71 5.5% High 31-299 187 14.6% 14.6% 263 20.5% 26.0%. Very high >299 1098 85.4% 100% 951 74.0% 100%. b. ESH/ESC criteria i. female number of urine samples with maximum sqPrCR by dipstick n=431 PrCR ACR Normal <31mg/g 0 0% 45 10.4% High 31-299 79 18.3% 18.3% 103 23.9% 34.3% Very high >299 352 81.7% 100% 286 65.7% 100%. ii. male number of urine samples with maximum sqPrCR by dipstick n=854 PrCR ACR Normal <22 mg/g 0 0% 35 4.1% High 22-299 108 12.6% 163 19.1% 23.2% Very high >299 746 87.4% 100% 665 76.8% 100% Conclusions: 1. The goodness-of-strip is shown with 0% of normal PrCR if spPrCR500mg/g. 2. Nevertheless, this approach would not detect 26.0%, 34.3% or 23.2%, according to criteria, of urine samples with normal or high ACR. This is clearly unacceptable regardless of the criteria used. 3. In short, the answer is that we advocate to do not.
1
T398 DEVELOPMENT OF A PHOTON UPCONVERSION READER WITH NEAR-INFRARED LASER EXCITATION J. Terrijrvi(1), J. Pohjola(2), V. Haaslahti(2), T. Soukka(1)
1 2
Division of Biotechnology, University of Turku, Finland Hidex, Turku, Finland
Upconverting nanoparticles (UCNPs) have attracted an immense attention for their possible applications in biotechnology. These trivalent lanthanide ion doped materials are capable of absorbing near-infrared light and converting it into luminescence at visible wavelengths. The advantages of upconversion luminescence are the avoidance of autofluorescence at measurement wavelengths, reduced sensitivity to scattering and absorption as well as narrow emission peaks of lanthanide ions. Hindrance with the upconversion technology thus far has been the lack of commercial upconverting nano particles as well as a wellestablished detection platform with infrared laser irradiation source for the excitation of the UCNPs. Recently commercial UCNPs have been introduced so the only problem remains with the instrumentation. To solve this problem a new detection platform was developed in association with Hidex, Turku, Finland. This was based on Plate Chameleon V reader (Hidex) from which the light source was replaced with 976 nm laser system (4 W). To enhance the performance of the device the laser itself was equipped with a peltier element for temperature control. In this epifluorometer the laser light was collimated to 0.8 mm diameter beam, which was directed to a well with an aluminum mirror. Emitted light from the well was gathered with a lens, filtered to block the excitation light and to select a correct emission band, e.g. 550 nm and then detected with a photomultiplier tube (PMT). UCNP detection limit, with streptavidin coated 25 nm UCNPs, with this new device was measured to be approximately 500 000 particles/ml, which allows the performance of high sensitivity immunoassays. Also the dynamic range was 5 orders of magnitude. With the development of this detection platform, the first steps to standardization of UCNP assays have been taken and this is also one of the key issues in acquainting the technology to common knowledge.
S490

T399 ASSESSMENT OF EFFICACY OF DISINFECTION AND STERILIZATION PROCEDURES J. Tuteska(1), V. Stojkovski(2), M. Arapceska(1), E. Mitevska(1) Medical Nursing College, St. Kliment Ohridski University, Bitola, R. Macedonia 2 Faculty of Veterinary Medicine, Ss. Cyril and Methodius University, Skopje, R. Macedonia 3 Faculty of Biotechnical Sciences, St. Kliment Ohridski University, Bitola, R. Macedonia Background: Surgical procedures and larger number of invasive medical procedures involve contact by a medical device or surgical instrument with a patients sterile tissue or mucous membranes. A major risk of all such procedures is the introduction of pathogenic microbes, which can lead to infection. Aim of this study was assessment of the efficacy of disinfection and sterilization procedures in University Clinical Center in Skopje, Macedonia. Methods: For the period of one year in 18 departments of Clinical Center efficacy of disinfection and sterilization procedures was assessed with application of physical, chemical and biological indicators. Influencing factors that affect efficacy of disinfection/sterilization were cleaning of the object, presence of organic and inorganic load, type and level of microbial contamination, concentration and exposure time to disinfectant/sterilzant, nature of the object, temperature and relative humidity. Steam, heat, ethylene oxide and hydrogen peroxide were commonly used as sterilizing agents. Results: According obtained results 98.81% of sterilization procedures were performed successfully. From 7242 sterilization procedures in total, only 86 procedures were unsuccessfully (56 because of electric current failure, and 30 because of technical failure). Biological indicators were recognized as reliable sterility indicators. As indicator were used spores inoculated on carrier. Growth of microorganisms was determined through the color change in a pH indicator due to a pH change resulting from byproducts of microbial cell growth in the medium. Conclusions: Achieving disinfection and sterilization through the use of disinfectants and sterilization practices is essential for ensuring that medical and surgical instruments do not transmit infectious pathogens to patients. Biological indicators provide confidence that a sterilization process has been done successfully. Inactivation of biological indicator strongly implies that other potential pathogens during sterilization have been killed.
1
T400 MEASUREMENTS OF MINERAL ELEMENTS AND HEAVY METALS FROM HAIR CELLS BY IPC-MASS SPECTROMETRY A. Udristioiu, M. Cojocaru, F. Mitu Aurelian Udristioiu, Emergency County Hospital Targu Jiu Romania, Clinical Laboratory Medical Analyses, City Targu Jiu, Romania 2 Manole Cojocaru, Titu Maiorescu University, Medicine Faculty, Physiology, Bucharest 3 Florin Mitu, Mecro System SRL, Bucharest, Background: Aim of this work was to identify intracellular levels of principal mineral nutritive elements (Ca, Mg., Cu, Zn) and trace elements (Al, Pb, Hg), measured by inductively coupled plasma mass spectrometry (ICP-MS) instrument, to apparent healthy persons, with a good nutritional status and gold standard of lifestyle. Material and method: At a number of 75 persons, all adult females, in range age 30-55 years, from different regions of the country was taken the samples of 100mg hair, obtained by cutting the first 3 cm closest to the scalp. Protocol of work Was delivered 100milligrams hair, cutting from back scalp of subjects, in the special cuvettes IPCMS and, results were measured and interpreted after graphics IPC MS. Results: A number of 12 patients (16%) displayed high levels of intracellular Mg, (mean value = 1.2 mmol /L, high values of Ca, (mean value =0.72 mmol/L), but low values of report Ca/Mg, (mean value = 0.58) and six patients (8%) were exhibited low levels Mg (mean value = 0.004 mmol/L low values of Ca (0.04 mmoli/ L) but high report Ca/Mg (10). Any persons not presented signs of acute or chronic intoxication with havey metals. Conclusions: Environmental of life to habitants, from different regions of country, is reflected in hair cells, as so is in all cells of body, in concentrations from last months.
1
S491

T401 ANALYTICAL VALIDATION AND DEVELOPMENT OF A 25 OH VITAMIN D ASSAY BY UHPLC-MS/MS S. Vanavanan, P. Meemaew, A. Chittamma, N. Teerakarnjana, S. Lapanan, M. Rochanawutanon Department of Pathology, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand. Background: 25-Hydroxy vitamin D (25-OH D) is an important marker for several diseases. Although various automated immunoassays are available they are not satisfactory in certain aspects. We developed and validated a new ultra-highperformance liquid chromatography mass spectrometry (UHPLC-MS/MS) method for determination of 25-OH D2 and D3. Methods: Aliquots of 150 uL spiked bovine serum albumin were prepared by adding 20 uL of internal standard (IS) followed by treatment with zinc sulfate and then by acetonitrile to remove protein. Samples were then analyzed for 25-OH D2 and D3 using UHPLC-MS/MS. The chromatographic system consisted of an Agilent (Santa Clara, USA) 1290 LC system with a Poroshell 120 EC-C18 column (3.0x50 mm, 2.7 m) and a Zorbax Eclipse Plus-C18 guard column (4.6x12.5 mm, 5 m). A gradient of mobile phase A (water/5M ammonium formate/formic acid, 99.8/0.1/0.1, v/v/v) and mobile phase B (methanol/5M ammonium formate/formic acid, 99.8/0.1/0.1, v/v/v) was used at flow rate of 0.4 mL/min for pump A and 0.5 mL/min for pump B. The injection volume was 3uL. An Agilent 6460 triple quadrupole MS, with an electrospray ionization (ESI) source, was used in positive mode. The method was validated in terms of linearity of range, limit of detection (LOD), limit of quantitation (LOQ), recovery, and precision. Results: The mass /charge transitions of 413.3>355.3 (25-OH D2), 401.3>365.2 (25-OH D3), 416.3>358.3 (IS of 25-OH D2) and 404.3>368.2 (IS of 25-OH D3) were monitored by multiplereaction monitoring (MRM) between the time from 3.2 to 3.8 min (total run time, 4.5 min). The linearity was 5-100 ng/mL with r2 = 0.999 for both analytes. The LODs were 1.61 and 2.01 ng/mL while the LOQs were 5.37 and 6.69 ng/mL for 25-OH D2 and D3, respectively. Recovery of both 25-OH D forms ranged from 94.8% to 100.7% at concentrations of 10, 30 and 80 ng/mL. Inter- and intra- assay precision value were less than 10%. Conclusions: This new UHPLC-MS/MS method, with a simple sample preparation, provides a rapid, accurate, and precise assay suitable for routine testing as a tool for evaluation of individual vitamin D status. T402 COMPARISON OF TWO METHODS FOR MEASUREMENT OF 25-HYDROXYVITAMIN D . Vogrinc, M. Lovri University hospital centre Zagreb, Department of laboratory diagnostics, Zagreb, Croatia Background: Measurement of circulating 25-hydroxyvitamin D (25(OH)D) in blood is the most reliable indicator of vitamin D status in organism. There are currently two main types of measurement used routinely for measuring the main circulating metabolites of vitamin D, 25 (OH)D2 and 25 (OH)D3. These are competitive immunoassay and methods based on chromatographic separation followed by non-immunological direct detection. With increasing clinical demand for 25(OH)D assays, fast automated platform with immunoassays are attractive. We evaluated an automated immunoassay for the quantitative determination of total serum 25(OH)D and compared it with a selective and sensitive HPLC method with UV detection. Methods: 25 (OH)D from human sera (routine serum samples, N=48) was measured using two methods: fully automated competitive electrochemiluminescence immunoassay (Roche Elecsys Vitamin D total assay) and Chromsystems HPLC method for 25 (OH) D3/D2. The evaluation protocol for immunoassay consisted of within-run imprecision (10 sequential runs, two samples) and between-run imprecision (10 consecutive working days, 2 sequential runs) with commercial controls PreciControl Bone 1 and PreciControl Bone 2, inaccuracy (N=20), and method comparison. Methods were compared by Passing and Bablok regression and BlandAltman analyses. Results: Within-run imprecision for Roche Elecsys Vitamin D total assay was 2.56%, and between-run imprecision was 2.99% and 3.79%. Quality requirement for inaccuracy was fulfilled. The comparison with HPLC method demonstrated strong correlation (r=0.9581; y=0.918x-4.5082) and good agreement (bias = 1.96 SD). Conclusion: Roche Elecsys Vitamin D total assay showed good correlation and agreement with HPLC-UV method and represents an accurate and precise automated tool for serum total 25 (OH)D determinations. Measurement of 25 (OH)D by immunoassay is also the method of choice for reasons of convenience, speed, turnaround, and cost.
S492

T403 GENETICALLY ENCODED PROTEASE SUBSTRATE BASED ON LANTHANIDE-BINDING PEPTIDE FOR TIMEGATED FLUORESCENCE DETECTION J. Vuojola, M. Syrjnp, U. Lamminmki, T. Soukka Division of Biotechnology, University of Turku, Finland Background: The study of biomolecular interactions is at the heart of biomedical research. Fluorescence and Frster resonance energy-transfer (FRET) are potent and versatile tools in studying these interactions. Fluorescent proteins are genetically encodable, which facilitates their use in recombinant protein and in vivo applications. To eliminate the autofluorescence background encountered in applications based on fluorescent proteins, lanthanide labels can be used as donor fluorophores. Their long emission lifetime enables the use of time-gating that significantly improves assay sensitivity. The commonly used lanthanide chelate and cryptate labels, however, are not genetically encodable. This problem can be circumvented by using a lanthanide-binding peptide, which is a small genetically encoded peptide sequence that selectively binds lanthanide ions with high affinity. Methods: In this work we combined the favorable characteristics of an intrinsically fluorescent terbium-ion containing lanthanide-binding peptide (Tb3+-LBP) and green fluorescent protein (GFP) in a FRET-based homogeneous caspase-3 activity assay. Caspase-3 is a key member mediating the proteolytic cleavage cascade that is activated during apoptosis. The genetically engineered construct used in the assay had LBP and GFP sequences at adjacent ends of a linker that comprised a recognition sequence for caspase-3. Results: In the assay both the time-gated terbium donor signal at 545nm and the time-gated sensitized GFP acceptor emission at 520nm were monitored. The inhibitor doseresponse curves gave IC50 values of approximately 60 nM. Signal-to-background ratios were maximally 33. The coefficients of variation between replicate reactions were small, maximally 4.2%. Conclusions: We were able to demonstrate, for the first time, the applicability of a Tb3+-LBPGFP energy-transfer pair in a protease activity assay. The intrinsically fluorescent and genetically encodable components enable easy expression of the construct without the need of cumbersome chemical labeling. By varying the fluorescent protein and the protease specificity of the internal linker sequence the method can be applied for the detection of a wide variety of proteases. T404 EVALUATION OF AN AUTOMATED ASSAY FOR GALECTIN-3 M. Zaninotto(1), M.M. Mion(1), G. Bragato(1), A. Clerico(2), C. Prontera(2), M. Plebani(1)
1 Department of Laboratory Medicine, University-Hospital, Padova, Italy 2 Scuola Superiore SantAnna, Department of Laboratory Medicine, Fondazione G. Monasterio, Pisa, Italy
Background: Replacement of functional myocytes with crosslinked collagen is a final common pathway that is central to the progression of heart failure (HF), irrespective of etiology. In response to different mechanical and neurohormonal stimuli, macrophages secrete Galectin-3, that stimulates cell proliferation and secretion of procollagen I. This protein is then irreversibly crosslinked to form collagen and result in cardiac fibrosis. We report an initial assessment of a new assay for Galectin-3. Methods: The analytical characteristics of a fully automated immunoassay for Galectin-3 on the Abbott ARCHITECT platform have been evaluated and the assay has been compared with a conventional enzyme immunoassay (EIA, BG Medicine) on a reference population of healthy subjects and on patients with heart failure. Results: The limit of blank (LOB), limit detection (LOD) and limit of quantitation (LOQ) of the ARCHITECT assay were 0.58 ng/mL, 1.32 ng/mL and 4.33 ng/mL (5.57% CV), respectively. The total imprecison on three levels of assay controls ranged from 1.60% to 3.75% and the serum/ plasma equivalency on 44 samples from patients with myocardial infarction was satisfactory, with an average difference of 13.76%. On 136 healthy subjects the 97.5th percentile was 19.56 ng/mL (21.03 in males, 19.00 in females). On a 274 healthy and diseased subjects the correlation between ARCHITECT and EIA was very high (r2=0.94), with a positive proportional bias of the automated assay. Galectin-3 levels correlated with the severity of HF in 177 patients, with median values of 17.05ng/mL in NYHA-I, 28.10ng/mL in NYHA-II, 27.40ng/mL in NYHA III and 42.95 in NYHA IV patients. By the Mann-Whitney test the differences between NYHA I and II/III/IV and NYHA III and IV were statistically significant. By ROC curve analysis, the best discriminating value between healthy and diseased subjects corresponds to 16.40 ng/mL (sensitivity: 90.4%; specificity: 73.4%). Conclusions: The automated assay for Galectin-3 has shown good analytical performances and a good correlation with another method. Though Galectin-3 values appear to correlate with the severity of HF, additional clinical data are needed in order to introduce this new assay in clinical practice.
S493

EUROMEDLAB 2013 Posters

Uploaded by

Copyright:

Available Formats

EUROMEDLAB 2013 Posters

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EUROMEDLAB 2013 Posters

Uploaded by

Copyright:

Available Formats

EuroMEdLab 2013 - postErs

biochimica clinica, 2013, vol. 37, SS