Open Access Original Article

Cord Blood As Early Predictive Marker of Neonatal Sepsis Pak Armed Forces Med J 2020; 70 (3): 824-29

COMPARISON OF PROCALCITONIN AND HEMATOLOGICAL RATIOS IN CORD

BLOOD AS EARLY PREDICTIVE MARKER OF NEONATAL SEPSIS
Sadia Israr, Asma Hayat, Tariq Mahmood Ahmad, Numan Majeed, Sakeena Hussain Naqvi*, Saema Tehseen*
Army Medical College/National University of Medical Sciences (NUMS) Rawalpindi Pakistan, *Pakistan Aeronautical Complex,
Kamra Pakistan

ABSTRACT
Objective: To evaluate the predictive value of neutrophil to lymphocyte ratio (NLR) and platelet to lymphocyte
ratio (PLR) with procalcitonin in cord blood.
Study Design: Case control study.
Place and Duration of Study: Study was conducted in Army Medical College and Pak Emirates Military Hospital
Rawalpindi, Jul 2018 to Mar 2019.
Methodology: Those mothers, having deliveries with early rupture of membranes (EROM), premature rupture of
membranes (PROM), preterm, dai handled, meconium and failure of induction have been included. Total 60
neonates were included in this study. Nineteen neonates were taken as a continuous clinical unsteadiness with
a clearly documented suspicion of sepsis and two neonatologists decided sepsis within 1-3 days of life. While
remaining neonates were control who have no infection.
Results: Laboratory values show there was a gross difference in mean values of case and control for white blood
cell (WBC), platelets, neutrophil, platelet to lymphocyte ratio and procalcitonin with statistical significance except
lymphocyte count and neutrophil to lymphocyte ratio. In combined ROC curve. The cutoff of procalcitonin was
calculated to be 0.4ng/ml with an area under curve of 84.5%. Similarly, cutoff of neutrophil to lymphocyte ratio
was determined to be 1.39 with an area under curve of 65.1% at sensitivity of 63% and specificity of 58.5%. Cutoff
of platelet to lymphocyte ratio was determined to be 47.48 with an area under curve of 66.8% at sensitivity of
78.9% and specificity of 49%.
Conclusion: Combined hematological markers neutrophil to lymphocyte ratio and platelet to lymphocyte ratio
with procalcitonin in cord blood could be used as a simple, sensitive and predictive parameter for identifying
neonates susceptible to sepsis.
Keywords: Early rupture of membrane (EROM), Neutrophil to lymphocyte ratio (NLR), Platelet to lymphocyte
ratio (PLR), Prolonged rupture of membrane (PROM).

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

INTRODUCTION based on temperature, heart rate, respiratory rate,

Establishing a globally accepted case and white blood cell count, and hypotension. In
definition in the newborn is tricky due to lack of 2016, this definition was redefined and updated
consensus, lack of optimal diagnostic tests, varied criteria of sepsis was stated by European society
pathologies and nonspecific clinical signs. Inter- of intensive care medicine (ESCIM) and Society of
national consensus definition, new guidelines, critical care medicine (SCCM) consensus task
and thoughtful practical processes would ensure force that life threating organ dysregulation
meaningful reporting and improved patient caused by host response failure to infection2.
outcomes1. The clinical presentations are variable In early onset neonatal infections, 85% of early
and nonspecific. According to international sepsis onset neonatal sepsis occurs within 24 hours, 5%
definition conference 2014, sepsis is a clinical present between 24-48 hours, and a smaller
syndrome that shows the presence of both percentage present between 48-72 hours3. The
infection and systemic inflammatory response early-onset sepsis occurs within first three days of
life (American Academy of Pediatrics definition).
Correspondence: Dr Sadia Israr, House No. C-17, AMF Colony,
Kamra Pakistan (Email: [email protected]) World Health Organization (WHO) reports
Received: 30 Oct 2019; revised received: 10 Jan 2020; accepted: 15 Jan that approximately 5 million neonatal deaths
2020

824
Cord Blood As Early Predictive Marker of Neonatal Sepsis Pak Armed Forces Med J 2020; 70 (3): 824-29

occur per year. The incidence of EONS is Procalcitonin (PCT) is protein in nature and
estimated as 1-2 cases per 1,000 live births in is an acute phase reactant. Procalcitonin is prime
United States of America while in developing biomarker which starts to rise in early sepsis.
countries it ranges from 2.2 to 9.8 per 1000 live There is rapid rise in the level of PCT within 2–
births4. A study in United Kingdom has shown 4h, reached its peak level at 6-8h after bacterial
that 50% of neonatal sepsis mortalities occur in endotoxin exposure8 and return to normal after
first 24 hours and half of these neonates die next 24h. Early rise PCT levels in neonatal sepsis
before transfer to NICU. In developing countries, makes it a good marker for early diagnosis of
neonatal sepsis is reported to be the leading cause sepsis in neonates as compared to CRP9. Procalci-
of neonatal mortality with a burden of death tonin has shown highest diagnostic value as
tolling almost 27% of the total neonatal deaths5. compared to CRP in sepsis10. Procalcitonin testing
In Pakistan, in past years most of the deliveries can be used to help clinicians diagnose bacterial
were conducted at home but after 1990 this trend infection (that can cause sepsis) and make
changed and approximately 2.5 folds increase in decisions on starting antibiotic treatment.
deliveries at healthcare facilities6. Cord blood is Numerous sepsis biomarkers have been used
the first available sample which help the clinician in routine for early detection of neonatal EONS,
to start therapeutic strategy as soon as possible. but to date, there is no single or even a group of
To evaluate for neonatal sepsis, complete biomarkers that fulfills all essential criteria. The
blood count (CBC), platelet count, and cerebros- neutrophil to lymphocyte ratio (NLR), platelet to
pinal fluid examination are practiced in some lymphocyte ratio (PLR)combines neutrophils
clinical settings. This hematologic scoring system, and platelets with lymphocytes in the calculation,
although a quick aid to screen for infection in is considered comparatively more stable than
neonates, has poor positive predictive value their values alone. These ratios are inflammatory
unless the score of these parameters are too high. markers that can be easily, rapidly, and
These hematological parameters with routinely inexpensively measured with routinely used
used biomarkers i.e. lactate, procalcitonin and C- instruments10.
reactive protein might be of value in susceptible The objective of this study was to evaluate
neonates in deciding need of antimicrobial agents the predictive value of the neutrophil to
or whether therapy can be safely discontinued. In lymphocyte ratio and platelet to lymphocyte ratio
CBC, most reliable marker is neutropenia for neo- in cord blood and comparison of these ratios with
natal sepsis, but different factors affect neutrophil procalcitonin levels in cord blood which is the
count such as gestational age, planned cesarean biomarker traditionally used for sepsis.
delivery has lower counts than vaginal delivery
METHODOLOGY
and at high altitude total neutrophil counts
are higher. Platelet count are also unreliable in It was a case control study, conducted in
diagnosis of neonatal sepsis7. A new emerging Department of Gynecology and obstetrics, Pak
ratio neutrophil to lymphocytes (NLR) and Emirates Military Hospital and Department of
platelet to lymphocyte (PLR) is better and easy Pathology Army Medical College, Rawalpindi.
approach for early prognosis and diagnosis of This study was started after approval of ethical
sepsis. Advantage of these ratios is that it can be review committee of the institute. Study popula-
measured by using conventional methods with- tion was neonates delivered at Department of
out any new sophisticated instrument or tools. Gynecology and obstetrics and neonates being
The simple and less invasive marker of inflamma- admitted in NICU Department of Pediatrics,
tion whenthere is an increase in neutrophil coun- Pak Emirates Military Hospital, Rawalpindi for
tand decrease in lymphocytes in the circulation, treatment of sepsis from July 2018 to March 2019.
resulting in high ratio, is a significant marker. Sample size according to WHO calculator was

825
Cord Blood As Early Predictive Marker of Neonatal Sepsis Pak Armed Forces Med J 2020; 70 (3): 824-29

taken as 60. Those mothers who had given scence technique. Lowest limit of detection for
voluntarily consent were included in the study. PCT concentrations was 0.02 ng/mL.
Consecutive sampling was done. Those having Statistical analysis was done by using SPSS
deliveries with early rupture of membrane, software version 20. Mean and Standard Devia-
preterm, diabetics and having other septic risk tion (SD) was calculated to assess the level of
factors were included an all healthy mothers, hematological indices, lactate, CRP and PCT. The
having deliveries at 39 weeks of gestational age difference in the level of all biochemical markers
were excluded from this study. among the case groups and control groups were
Immediately after birth while the placenta assessed using independent t-test. The difference
was still in situ, four Howard Kelly forceps were between the two groups was considered
placed on the cord to isolate a 20cm segment statistically significant, when the p-value was
in the middle. The isolated segment was cut ≤0.05 in independent t-test. Receiver Operating
between the two sets of clamps, baby and the Characteristics (ROC) curve was drawn to deter-
placenta still having a clamp in place. 7cc blood mine the sensitivity, specificity for the parameters
(from the placental end of the cord) was collected and Area Under the Curve (AUC) was calculated
in a syringe, from which 5ml cord blood was
taken in plain tube and 2ml cord blood
transferred in EDTA tube.
All the samples were centrifuged after
clotting, sample for CBC was processed, and
separated sera were stored at -20°C in Eppendorf
tubes prior to analysis. Follow up was done
within 72 hours. Neonates having a continuous
clinical unsteadiness, with a clearly documented
suspicion of sepsis confirmed by the neonato-
logist at 1-3 day of life were taken as cases. Only
those neonates which were labelled as having
Figure-1: Mode of delivery.
sepsis by two neonatologists independently were
taken in case group. Clinical instability was
defined as having any of the following clinical
signs in neonate: delayed cry, irritability, reluc-
tant to feed, fever or hypothermia, tachypnea or
hypotonia. Neonates who did not develop sepsis
within 72 hours of birth were considered as
controls.
All samples were analyzed in the Depart-
ment of Chemical Pathology, Army Medical
College, Rawalpindi. Complete blood count was
performed on 3-part differential automated cell Figure-2: ROC curve for procalcitonin, neutrophil to
counter Sysmex KX21N (Sysmex, Kobe, Japan). lymphocyte ratio and platelet to lymphocyte ratio.
PL= Platelet to Lymphocyte Ratio, Ratio=Neutrophil to Lymphocyte
For quality control, two levels of internal quality Ratio
control materials were used with each batch of and compared and the most accurate cut-off
analysis. Serum procalcitonin was measured by values were taken to predict sepsis.
Roche cobas c411 based on electrochemilumine

826
Cord Blood As Early Predictive Marker of Neonatal Sepsis Pak Armed Forces Med J 2020; 70 (3): 824-29

RESULTS p=.03. Procalcitonin values were also as

Study participants was included 60 neonates significant as WBC, platelets and neutrophil
delivered with high risk mothers, divided in two counts. Procalcitonin has mean value of 0.968 ±
groups. Nineteen neonates were taken as cases 0.13 ng/mL for cases and 0.0288 ± .0176 ng/mL
on neonatologist decision with positive clinical for controls, p 0.002. Cord blood Procalcitonin
presentation and lab results and being admitted levels were raised almost in all cases, so out of 19
in neonatal intensive care unit. Other group babies (thirteen males and six females) showed
included 41 neonates having no infection were presence of sepsis, 18 have raised procalcitonin
taken as controls. Some controls although have levels. Lab Values shows there was a gross
shown mild clinical unsteadiness so kept under difference in values of case and control for WBC,
observation butafter management discharged by platelets, neutrophil, lymphocyte, NLR, PLR and
neonatologist. Demographic data showed that procalcitonin with statistically significance except
male neonates delivered with emergency lymphocyte and NLR which did not show any
caesarian section in presumed mothers have significance between case and control groups.
higher risk of neonatal sepsis. The combined ROC curve for NLR, PLR and
Neonates included in study showed 45% PCT is shown in figure. The cutoff of PCT was
were born with normal vaginal delivery and calculated to be 0.04 with a CI of 95% (71.9%-

Table-I: Values of different parameters in cord blood.

Case Control
p-value
Mean ± SD Mean ± SD
WBC (x103 cells/µL) 14.79 ± 2.75 13.15 ± 2.95 0.04
Platelet (x103 cells/µL) 301.52 ± 42.12 207.46 ± 72.7 0.00
Neutrophil (x103 cells/µL) 8.35 ± 2.04 6.02 ± 2.24 0.00
Lymphocyte (x103 cells/µL) 4.96 ± 1.59 4.60 ± 1.82 0.47
Neutrophil to lymphocyte ratio 1.91 ± 0.93 1.54 ± 0.99 0.17
Platelet to lymphocyte ratio 69.90 ± 33.97 51.91 ± 26.60 0.03
Procalcitonin (ng/mL) 0.0968 ± 0.13 0.0288 ± 0.02 0.002
Table-II: Area under the curve for different parameters.
Asymptotic 95% Confidence Interval
Test Result Variable(s) Area
Lower Bound Upper Bound
Neutrophil to lymphocyte ratio 0.651 0.505 0.797
Platelet to lymphocyte ratio 0.668 0.530 0.805
Procalcitonin 0.845 0.719 0.971
55% via caesarean section. Neonates delivered 97.1%), with an AUC of 84.5%. Similarly, cutoff
through cesarean section had higher risk of of NLR was determined to be 1.39 with a CI of
developing neonatal sepsis as compared to 95% (50.5%-79.7%) with an AUC of 65.1% at
neonates delivered by vaginal deliveries. sensitivity of 63% and specificity of 58.5%. Cutoff
For cases, WBC count mean 14.79 ± 2.75 x 103 of PLR was determined to be 47.48 with a CI of
cells/µL p=0.04, platelets 301.5 ± 42.12 x 103 95% (53.0%-80.5%) with an AUC of 66.8% at
cells/µL, p 0.00, neutrophil count varied between sensitivity of 78.9% and specificity of 49%.
cases 8.35 ± 2.4 to 6.02 ± 2.22 x103 cells/µ for DISCUSSION
controls, p 0.00l ymphocyte count were 4.9 ± 1.59 In developing countries, it is affirmed that
and 4.60 ± 1.82 x 103 cells/µL for cases and the burden of neonatal sepsis is still high and the
controls, p=0.47, NLR 1.91 ± 0.93 for cases and clinical presentations; symptoms and signs are
1.54 ± 0.99 for controls, p=0.17, PLR for cases variable and non-specific, leading to delay in
and controls were 69.90 ± 33.9 and 51.91 ± 26.60,

827
Cord Blood As Early Predictive Marker of Neonatal Sepsis Pak Armed Forces Med J 2020; 70 (3): 824-29

diagnosis11-15. Thus, there is a dire need for these ratios in disease prediction is still
early diagnosis and timely administration of unknown18. A previous study showed PCT and
antibiotics. Blood culture-based diagnosis is no NLR have similar predictive values for sepsis21
doubt a first-class way to find the causative but our study didn’t show any significance of
organisms, but well-equipped laboratory and NLR in sepsis but showed strong significant of
trained personnel are required for this gold PLR with sepsis. These finding are consistent
standard that are lacking in most of the hospitals with studies which shows PLR is more valuable
in our population. Neonatal sepsis has rapid than either platelet or lymphocyte count alone in
course and multiple system involvement. To the prediction of various septic conditions22,23.
address the problems of delay and false negative Mean value for NLR, reported higher in
results associated with blood culture-based Non-Hispanic Blacks was 2.24, in Whites It was
diagnosis, an algorithm of simple, rapid and eco- 1.76 less than blacks in a study conducted in
nomical markers is required. Procalcitonin is now United States24. Another study conducted in
widely used in clinical laboratories in Europe Chinese population, mean baseline value for NLR
and USA with FDA recommendation for the was 1.5 ± 0.05 in healthy adults. Different studies
evaluation of risk factors as well as for diagnosis used different reference value but the use of
and monitoring of sepsis13. PCT cut-off limit >0.5 these arbitrary cut off points are characteristically
ng/mL indicates a two-fold probability of sepsis. misleading for risk stratification because NLR
In 2010, in a meta-analysis results, PCT have and PLR values are affected by age, environment,
shown a higher pooled sensitivity (72%) p=<0.0514 cardiovascular diseases etc25. So, there is a need
and PCT is a sensitive marker in detection of for establishment of reference value and cut off
early onset neonatal sepsis15,16. These finding are points according to local factors. In our study,
same as our results which showed the signifi- cutoff of NLR was determined to be 1.39 with a
cance of PCT AUC 84.5%, p<0.05. CI of 95% (50.5%-79.7%) with an AUC of 65.1% at
In addition, by using the multiple markers sensitivity of 63% and specificity of 58.5%. Cutoff
combining early sensitive marker with hemato- of PLR was determined to be 47.48 with a CI of
logical indices will further enhance the predictive 95% (53.0%-80.5%) with an AUC of 66.8% at
accuracy of these mediators for susceptible cases. sensitivity of 78.9% and specificity of 49%.
These tests are simple and cost effective so help These stable values, neutrophil to lympho-
clinicians to avoid unnecessary antibiotic thera- cyte ratio (NLR), platelet to lymphocyte ratio
pies. In these aspects, hematological ratio such as (PLR) combines neutrophils and platelets with
NLR and PLR are emerging in early detection of lymphocytes in the calculation, are comparable
neonatal sepsis because no extra blood sample or with procalcitonin. So, these ratios can be used
equipment is required for these ratios17. To our conveniently in combination with procalcitonin
knowledge there have been no such study in cord without requirement of additional sample or
blood which show the prediction value of NLR equipment.
and PLR in EONS. Platelet-lymphocyte ratio
CONCLUSION
appears to be more closely related to sepsis as
indicated in previous studies that platelets Combination of hematological markers, NLR
are tightly associated with severe inflammatory and PLR, with procalcitonin in cord blood could
diseases18-19. Previous studies have shown these be used as a simple, sensitive and predictive
ratios as predictor of birth outcome if neonate is parameter for identifying neonates susceptible to
presumed to be affected because neonatal NLR sepsis.
reflect the maternal chorioamniotic inflamma- CONFLICT OF INTEREST
tion11,20. NLR and PLR could be a predictor of This study has no conflict of interest to be
poor birth outcome. The prognostic impact of
declared by any author.

828
Cord Blood As Early Predictive Marker of Neonatal Sepsis Pak Armed Forces Med J 2020; 70 (3): 824-29

REFERENCES we stand? A review. Pediatr Neonatol 2016; 57(4): 265-73.

14. Diaconu C, Bălăceanu A, Moroşan E. Sepsis biomarkers: past,
1. Wynn JL. Defining neonatal sepsis. Current opinion Pediatrics present and future. Farmacia 2015; 63(6): 811-15.
2016; 28(2): 135-40. 15. Leli C, Ferranti M, Moretti A, Dhahab A, Salim Z, Cenci E, et al.
2. Plevin R, Callcut R. Update in sepsis guidelines: what is really Procalcitonin levels in gram-positive, gram-negative, and fungal
new? Trauma Surg Acute Care Open 2017; 2(1): e000088. bloodstream infections. Disease Markers 2015; 2015.
3. El-Amin Abdalla EO, Salih FA, Salih HF, Elamin OE, 16. Mithal LB, Palac HL, Yogev R, Ernst LM, Mestan KK. Cord
Gamaleldin MA, Mustafa BM. Procalcitonin in the diagnosis of blood acute phase reactants predict early onset neonatal sepsis
early-onset neonatal infection in resource-limited settings. in preterm infants. PloS One 2017; 12(1): e0168677.
Cogent Med 2017; 4(1): 1283085. 17. Akkar OB, Sancakdar E, Karakus S, Yildiz C, Akkar I, Arslan M,
4. Rivera TSL, Dubón GEF, García RAF, Mazariegos A. Early onset et al. Evaluation of maternal serum 25-hydroxyvitamin d,
neonatal sepsis; diagnostic value of some laboratory tests. Intl J paraoxonase 1 levels, and neutrophil-to-lymphocyte ratio in
Med Surg Sci 2017; 4(1): 1109-14. spontaneous preterm birth. Medical Science Monitor: Intl Med J
5. Mustufa MA, Sheikh MA, Raza SJ, Arshad MS, Akhter T, Exper Clin Res 2016; 22(1): 1238-44.
Arain GM, et al. Trajectory of cause of death among brought 18. Okoye HC, Madu AJ, Korubo K, Efobi C, Eze OE, Obodo O,
dead neonates in tertiary care public facilities of Pakistan: A et al. Correlates of neutrophil / lymphocyte, platelet /
multicenter study. World J Pediatr 2017; 13(1): 57-62. lymphocyte, and platelet / neutrophil ratios of neonates of
6. Pasha O, Saleem S, Ali S, Goudar SS, Garces A, Esamai F, et al. women with hypertensive disease of pregnancy with neonatal
Maternal and newborn outcomes in Pakistan compared to other birth outcomes. Hyperten Pregnancy 2019: 1(1): 1-6.
low and middle income countries in the Global Network’s 19. Wang X, Su S, Guo Y. The clinical use of the platelet to
Maternal Newborn Health Registry: an active, community- lymphocyte ratio and lymphocyte to monocyte ratio as
based, pregnancy surveillance mechanism. Reproductive Health prognostic factors in renal cell carcinoma: a systematic review
2015; 12(2): S15-20. and meta-analysis. Oncotarget 2017; 8(48): 84506.
7. Arunachalam AR, Pammi M. Biomarkers in earlyonset neonatal 20. Toptas M, Asik H, Kalyoncuoglu M, Can E, Can MM. Are
sepsis: An update. Ann Clin Med Microbiol 2015; 1(2): 1007-15. neutrophil/lymphocyte ratio and platelet/lymphocyte ratio
8. Pontrelli G, De Crescenzo F, Buzzetti R, Jenkner A, Balduzzi S, predictors for severity of preeclampsia? J Clin Gyne Obs 2016;
Carducci FC, et al. Accuracy of serum procalcitonin for the 5(1): 27-31.
diagnosis of sepsis in neonates and children with systemic 21. Zhang HB, Chen J, Lan QF, Ma XJ, Zhang SY. Diagnostic values
inflammatory syndrome: a meta-analysis. Bio Med Center of red cell distribution width, platelet distribution width and
Infectious Dis 2017; 17(1): 302-10. neutrophil lymphocyte count ratio for sepsis. Exper Therap Med
9. Van Herk W, Stocker M, van Rossum AM. Recognising early 2016; 12(4): 2215-19.
onset neonatal sepsis: an essential step in appropriate 22. Akboga MK, Canpolat U, Yuksel M, Yayla C, Yilmaz S, Turak O,
antimicrobial use. J Infec 2016; 72(1): S77-S82. et al. Platelet to lymphocyte ratio as a novel indicator of
10. Enguix-Armada A, Escobar-Conesa R, García-De La Torre A, De inflammation is correlated with the severity of metabolic
La Torre-Prados MV. Usefulness of several biomarkers in the syndrome: a single center large-scale study. Platelets 2016; 27(2):
management of septic patients: C-reactive protein, procalcitonin, 178-83.
presepsin and mid-regional pro-adrenomedullin. Clini Chem 23. Zencir C, Akpek M, Senol S, Selvi M, Onay S, Cetin M, et al.
Laboratory Med (CCLM) 2016; 54(1): 163-68. Association between hematologic parameters and in-hospital
11. Can E, Hamilcikan Ş, Can C. The value of neutrophil to mortality in patients with infective endocarditis. Kaohsiung J
lymphocyte ratio and platelet to lymphocyte ratio for detecting Med Sci 2015; 31(12): 632-38.
early-onset neonatal sepsis. J Pediatr Hematol/Oncol 2018; 40(4): 24. Azab B, Camacho-Rivera M, Taioli E. Average values and racial
e229-e32. differences of neutrophil lymphocyte ratio among a nationally
12. Sargın MA, Yassa M, Taymur BD, Celik A, Ergun E, Tug N. representative sample of United States subjects. PloS One 2014;
Neutrophil-to-lymphocyte and platelet-to-lymphocyte ratios: are 9(11): e112361.
they useful for predicting gestational diabetes mellitus during 25. Alexander N. Reference values of neutrophil-lymphocyte
pregnancy? Therap Clin Risk Manag 2016; 12(1): 657-3. ratio, platelet-lymphocyte ratio and mean platelet volume in
13. Cortese F, Scicchitano P, Gesualdo M, Filaninno A, De Giorgi E, healthy adults in north central Nigeria. J Blood Lymph 2016;
Schettini F, et al. Early and late infections in newborns: where do 6(1): 143-48.

829