ournal of Automatic Chemistry, Vol. 15, No. 6 (November-December 1993), pp.

189-208

Multicentre evaluation of the Boehringer

Mannheim/Hitachi 911 Analysis System
Z. Zaman, N. Blanckaert, Ch. Cobbaert than 5%. Sixty-six of the 77 method comparisons performed met
Clinical Chemistry Laboratory, University Hospitals Leuven, Capucynenvoer 33-35, the acceptance criteria. The deviations of the remaining 11 results
B 3000 Leuven, Belgium could be explained by differences in either calibration, application
P. Gillery or by the use of different methods.
Laboratoire Central de Biochimie, Hdpital Robert Debrg CHR Reims, Rue Alexis Practicability was assessed using a questionnaire which covered
Carrel, F 51092 Reims Cedex, France all of the important aspects of an analysis system in the clinical
P. Hagemann laboratory. Twelve groups of attributes out of 14 were rater higher
Zentrallaboratorium, Kantonsspital Miinsterlingen, CH 8596 Miinsterlingen, for the BM/Hitachi 911 than for the present situation in the
Switzerland laboratories concerned. Especially high scores were given for the
H. Luthe versatility group.
Abt. Klinische Chemie, Zentrum Innere .Medizin, Georg-August-Universitiit The acceptance criteria for the analytical performance of the
G6ttingen, Robert-Koch-Str. 40, D 37075 Ggttingen, Germany BM/Hitachi 911 analysis system were fulfilled in all laboratory
R. Motta, D. Patrono, M.-A. Jionsu6 segments with few exceptions. The practicability exceeded the
Laboratorio Centralizzato, Ospedale S. Orsola, USL 28-B0 Nord, Via requirements in most of the attributes. The results of the study
Massarenti No. 9, 1 40138 Bologna, Italy confirmed the usefulness of the system as a consolidated workstation
A. Torralba, M. J. Castifieiras, X. Fuentes-Arderiu in small- to medium-sized clinical laboratories and in S TA T
Servei de Bioquimica Clinica, Hospital Princeps d’Espanya, L’Hospitalet de laboratories, or as an instrument for special analytes like proteins
Llobregat, E 08907 Barcelona, Spain and drugs, or for urinalysis in large laboratories.
W. Bablok, I. Domke and W. Stockmann
Boehringer Mannheim GmbH, Sandhofer Str. 116, D 68298 Mannheim 31,
Germany Introduction
The analytical performance and practicability of the Boehringer
Mannheim (BM)/mtaci 91 analysis system have been assessed The Boehringer Mannheim (BM)/Hitachi 911 analysis
in a multicentre evaluation, which involved six laboratories from system is the most recent medium-sized analysis system
to be introduced to the market by Boehringer Mannheim
European countries. Analytes commonly used in classical clinical
chemistry were tested in a core programme, which mainly followed GmbH. In addition to the well-accepted analytical
lhe ECCLS guidelines. In addition, a satellite programme covered performance and reliability of previous BM/Hitachi
other analytes, such as proteins, drugs and urine analytes. In total, analysis systems, the new instrument has features making
the study comprised more than 100 000 data items collected over a it attractive to different sections of clinical laboratories.
three-month period. The evaluation was supported with ’Computer These are in routine and emergency analysis (STAT),
Aided Evaluation’ (CAEv) and telecommunications. homogeneous immunoassays for the determination of
Acceptance criteria for the results were established at the proteins and drugs and urinalysis. Therefore the BM/Hitachi
beginning of the study. Nearly all of the analytes met the imprecision 911 has to be extremely flexible. This is achieved by
limits." within-run imprecision (as CVs) was 2l/ofor enzyme and incorporation of features such as automatic recognition
substrate assays, l% for ISE methods and 5l/o for immunoassays; of four different barcodes for sample identification, use of
between-day imprecision was 3l/ofor enzyme and substrate assays, up to four reagents per test, variable reaction time,
2o//o for ISE methods and 10% for immunoassays. random analysis of serum and urine specimens using the
No relevant drift effects (systematic deviation >_ 3O//o) were same calibration curve, fully automated predilution of
observed over eight hours. The methods were linear over a wide specimens and automatic calibration.
range. Sample-related and reagent-dependent carry-over can be The versatility of the new instrument required the design
reduced to a negligible amount by integration of a software- of a comprehensive evaluation. Six European laboratories
controlled wash-step. participated in the multicentre evaluation. Analytes of
Endogenous interferences were found for creatinine classical clinical chemistry were tested in a core pro-
method) and uric acid assays (caused by bilirubin), for creatine gramme, which mainly tbllowed the ECCLS guidelines
kinase, creatine kinase MB isoform and 7-glutamyltransferase [1]. In addition, the behaviour of the system in different
(caused by haemoglobin), and for immunoglobulin A (caused by laboratory sections was tested in a satellite programme.
lipaemia) The evaluators ran a much less extensive satellite
Accuracy was checked by an interlaboratory survey, recovery evaluation in order to maintain an acceptable cost/benefit
studies in control materials and method comparison studies. The ratio. In total, the study included more than 100 000 data.
survey showed that, with the exception of cholesterol and iron in Processing and analysis of the large data volumes was
two laboratories, the recovery of analytes did not deviate by more managed with a program system called ’Computer Aided
Evaluation’ (CAEv) and telecommunications.
Correspondence about this paper should be addressed to Dr Z. aman at the CAEv integrates the definition of study protocols,
University Hospitals, Leuven. performance of experiments in the laboratory, online/

0142-0453/93 $i0.00 @) 1993 Taylor & F is Ltd.

189
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

offiine data transmission and the immediate assessment reagent carry-over. The pipetting cycle for photometric
and evaluation of the results [2]. The program runs on tests, for the automatic predilution and for the wash steps,
a standard PC under MS-DOS. It had previously been is 10 and for the ISE methods 20 s.
successfully applied to the multicentre evaluation of the
BM/Hitachi 747 analysis system [3]. Various measurements and calibration procedures that
can be applied are:
Telecommunications were used for the first time in an
international multicentre evaluation. Installation of the (1) Endpoint measurements with/without sample blank
necessary facilities in the participating laboratories within 16 min.
allowed rapid and simple transfer of data to the centre
coordinating the study. (2) Kinetic determinations with sample or substrate start
(49 measuring points within 16 min).
The assessment of practicability was a further goal of the
evaluation of the BM/Hitachi 911. The evaluators (3) ’Fixed-time’ measurements (two-point kinetic).
answered 200 questions each about the system. All results (4) Combination of kinetic and endpoint measurements
of the multicentre evaluation are presented in this paper. with ’twin tests’.

(5) Potentiometric measurements with the ISE units

Description of the instrument within 20 at 37C.

The BM/Hitachi 911 is a medium-sized selective access (6) Serum indices for the characterization of haemolytic,
analyser with a capacity for 35 different tests, including icteric and lipaemic sera.
three ion selective electrode (ISE) methods for sodium, (7) Prozone check (detection of antigen excess).
potassium and chloride. The instrument specifications are
listed in table 1. The throughput is 360 photometric (8) Linear and four nonlinear modes of calibration, the
tests/h; this is reduced by automatic sample predilution frequency of which is controlled by the instrument
or additional wash steps which may be needed to eliminate (auto-calibration).

Table 1. BM/Hitachi 911 instrument specifications.

Type of instrument Discrete selective multianalyser.
2 Test channels 32, with ISE-module 35.
3 Test procedures Endpoint, endpoint with sample blank, kinetic with serum or substrate start, fixed-time kinetic,
combination of two endpoint tests, endpoint and kinetic tests, two kinetics tests performing
two tests in one cuvette, two prozone check procedures, measurement with ISE, linear
calibration and four nonlinear modes of calibration, autocalibration, two-point recalibration,
isoenzyme calibration, serum indices indicating haemolytic, icteric and lipaemic specimens.
4 Throughput Maximum 360 photometric tests/h, with ISE-module 720 tests/h.
5 Sampling system Turntable with 115 positions in total, arranged in three concentric rings:
--50 positions in the outer ring for routine samples
--40 positions in the middle ring
3 for wash solutions
--20 for STAT samples
--17 for standards/calibrators
--25 cooled positions in the inner ring for 17 standard and 8 control samples.
Primary tubes from 13 to 16 mm diameter and 75 to 100 mm length, secondary sample cups
with 2 ml and microcups with 0"5 ml maximum volume. Barcode identification of primary
tubes. Codes: code bar NW 7, code 32, 2 out of 5 interleaved, code 128. Different codes and
different tube sizes can be used within one sample disk.
6 Sample pipettor 3 to 50 gl (in steps of gl), imprecision < 1; for ISE same pipettor 15 gl for the three
determinations of Na, K, C1.
7 Reagent cooling Cold water circuit, refrigerator temperature 5 to 15C.
8 Reagent bottles 20, 50 and 100 ml sizes, 2 reagent disks with 33 bottle positions, one position reserved for
cuvettes rinsing solution. Disk for reagent and 4, disk 2 for reagent 2 and 3.
9 Reagent dispenser 2 reagent pipettors for dispensing reagents to 4, 50 to 350gl (in steps of lgl).
10 Mixing procedure 2 stirrers mix the reaction solution independently after addition of each reagent.
11 Reaction rotor Turntable with 120 cuvettes; half rotation (60 cuvettes + 1) in 10s one working cycle.
12 Reaction cuvettes Special plastic cuvettes, semidisposable.
Volume required: minimum 250 lal, maximum 500
Optical path length: 6 mm.
13 Reaction cycle 3, 4, 5, 10 and 16 minutes reaction time corresponding to 10, 13, 15, 31 and 49 measuring
points. Average time between two measuring points: 20 s.
14 Temperature control Water-bath, 37 0"IC.

190
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 1 (continued).
15 Photometer --Single beam photometer with 12 available wavelengths: 340, 376, 415, 450, 480, 505, 546,
570, 600, 660, 700 and 800 nm; mono- or bichromatic measurement with free selectable
wavelengths combination.
--Light source: halogen lamp. Detector- silicone photodiode.
--Wavelength adjustment: fixed, via a grating chromator, inaccuracy at 340 nm 2 nm and
at 405 to 800 nm + 5 nm.
--Half band width: 4 nm in the UV range
10 nm in the visible range
--.Photometric range of linearity: A -0- 2"5 at 340 nm
--Photometric resolution: A--0"0002
--.Photometric inaccuracy: max. 1 at 2"0 abs.
16 Ion-selective electrodes --Indirect potentiometry; flow-through electrode with liquid membrane.
--.Reference electrode" liquid membrane
---Dilution ratio" 1" 31
--Incubation temperature: 37 0"IC
--Calibration: 2-point with compensation
Measuring cycle: 20
--.Measuring range in serum in urine
Na / 80-180 10-250 mmol/1
K / 1"5-10 1-100 mmol/1
C1- 60-120 10-250 mmol/1
--Consumption of ISE solution per sample:
diluent" 450 Itl
internal standard: 1050 Itl
KC1 solution 130 Itl
17 Data processing --.Data input: via alpha numeric keyboard and item select keys.
---Data control: CRT 14 inches colour monitor.
-Data output: matrix printer, 80 characters per line, 220 characters per second.
-2 floppy disk drives, 3"5 inches, system disk and data disk for storage of 800 routine
samples and 200 STAT samples.
--Interface" RS 232 C. Bidirectional link with a host computer.
18 Water supply --Internal reservoir with an external supply.
Quality" _< ItS
Consumption during operation: 30 1/h
19 Ambient temperature 15 to 32C
20 Relative humidity 45 to 85
21 Physical dimensions Analyscr unit Operator unit
Width 1"03 m 0"48 m
Depth 0"76 m 0"65 m
Height 1-17 m 1"25 m
22 Weight Approx. 350 kg

The manufacturer provides an application disk to load cuvette. The average reagent consumption is 300 lal per
and store the application settings for the barcoded system determination.
reagents. These can be inserted in any free position of the
reagent disk. For the use of non-barcoded reagents, the Up to four reagents can be pipetted per test. For several
operator must define the appropriate application settings. STAT tests the second reagent is added 1"3 min after the
sample pipetting and for routine tests after 5 min. If
Specimens (3-50 lal volume per test) are processed either necessary, a fourth reagent can be pipetted after 10 min.
from primary tubes (5-10 ml), secondary cups (2 ml) or The results of eight STAT tests are available within 6 min.
microcups (500 lal). The primary tubes can be identified
by tbur different types of barcodes. The RS 232 interface
allows a bidirectional communication to a host computer.
One hundred and twenty plastic cuvettes (d 6 mm) are Materials and methods
arranged on a rotor which rotates in a water bath at 37C
(__0"IC). The cuvettes pass through the beam of the Inslrumenls and reagenls
photometer every 20 (12 fixed wavelengths between 340
and 800 nm, mono- or bichromatic). The photometric The methods and instruments used in the study are listed
range of linearity at 340 nm extends to an absorbance in table 2. The same lot of reagents was used in all
of 2.5. evaluation centres for each method. Reagents ti0r the
BM/Hitachi instrument were available in system packs
Two pipettors (50-350 gl) transtir the reagent into a containing barcoded bottles.
191
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 2. Analytes, methods and comparison instruments.

Abbrevi- Method on
Analyte ation BM/Hitachi 911 Comparison instruments Comparison methods
Pancreatic PAMYL Ethyliden-G7 PNP as substrate (EPS) BM/Hitachi 704, 747 Same as BM/Hitachi 911
a-amylase
Aspartate amino- ASAT According to IFCC and standard BM/Hitachi 704, 747 Same as BM/Hitachi 911
transferasc optimized
Creatine kinase CK Standard optimized BM/Hitachi 717, 737, 747 Same as BM Hitachi 911
Creatine kinase CK-MB Standard optimized, BM/Hitachi 717, 747 Same as BM Hitachi 911
MB isoform immunoinhibition
7-Glutamyl- GGT L-7-glutamyl-3-carboxy-4-nitroanilide BM/Hitachi 717, 737, 747 Same as BM Hitachi 911
transfcrase as substrate
fl-N-Acetylgluco- BNAG Chlorophenol red (CPR)-NAG as COBAS (R) Fara (Hoffmann- Same as BM/Hitachi 911
saminidase substrate LaRoche)
Calcium CA o-Cresolphthalein complexone BM/Hitachi 747, Same as BM/Hitachi 911
Atomic absorbance Atomic absorption
spectrometer spectroscopy
Perkin Elmer 1100 B
Cholesterol CHOL Cholesterol oxidase/ BM/Hitachi 704, 747 Same as BM/Hitachi 911
p-aminophenozone
Crcatininc CREA Jaff6 (kinetic) without deproteinization BM/Hitachi 704, 717,747, Same as BM/Hitachi 911
CX3 (Beckman)
Iron FE FerroZine without deproteinization BM/Hitachi 704, 747, Same as BM/Hitachi 911
COBAS (R) Fara (Hoffmann-
LaRoche)
F ruc tosamine FR U CrI" Reduction of nitroblue tetrazolium Isamat (R) Same as BM/Hitachi 911
chloride
Total protein TP Biuret BM/Hitachi 704, 717, 747 Same as BM/Hitachi 911
Uric acid UA Enzymatic PAP BM/Hitachi 704, 717, 747 Same as BM/Hitachi 911
C-Reactive CRP Turbidimetry Behring Nephelometer Fixed-time nephelometry
protein
Fcrritin FT Turbidimetry Stratus (R) II analyser Fluorescence immunoassay
Immuno- IGA Turbidimetry Behring Nephelometer Fixed-time nephelometry
globulin A
Transferrin TF Turbidimetry ARRAY (R) analyser Rate nephelometry
(Beckman)
Albumin in MAU Turbidimetry COBAS (R) Fara (Hoffmann- Turbidimedtry
urine LaRoche),
ARRAY (R) analyser Rate nephelometry
(Beckman)
Manual Radioimmunoassay
Phenobarbital PHEBA Homogeneous immunoassay, Abbott TDx Fluorescence polarization
CEDIA (R) Phenobarbital immunoassay
Phenotoin PHENY Homogeneous immunoassay, Abbott TDx Fluorescence polarization
CEDIA Phenytoin immunoassay
Thcophyllinc THEO Homogeneous immunoassay, Abbott TDx Fluorescence polarization
CEDIA ’- Theophylline immunoassay
Digoxin DIG Homogeneous immunoassay, Abbott TDx Fluorescence polarization
CEDIA (R) Digoxin immunoassay
Sodium NA ISE BM/Hitachi 747, CX3 ISE
(Beckman),
Flame spectrometer SCM Flame atomic emission
6341 (Eppendorf) spectrometry
Potassium K ISE BM/Hitachi 747, CX3 ISE
(Beckman),
Flame spectrometer SCM Flame atomic emission
6341 (Eppendorf) spectrometry
Chloride CL ISE BM/Hitachi 747, CX3 ISE
(Beckman),
Chloride analyser 925 Coulometry
(Corning)

192
Z. Zarnan el al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Control materials Evaluation protocol

Imprecision and quality control studies were performed The versatility of the BM/Hitachi 911 analysis system was
with lyophilized control sera (Boehringer Mannheim tested in a core programme and a satellite programme.
GmbH) and control urines (Boehringer Mannheim The core program had been used for evaluations of
GmbH, Bio-Rad). Detailed information is shown in the the BM/Hitachi 704, 717 and 747 analysis systems
appendix (table 10). [-5].
The 13 analytes tested in the core programme were
Calibration pancreatic 0c-amylase, aspartate aminotransferase, creatine
During the familiarization period, a fixed-factor was kinase, 7-glutamyltransferase, calcium, cholesterol, crea-
determined for the enzyme tests in three independent tinine, iron, total protein, uric acid, sodium, potassium,
calibration runs per day on three consecutive days. and chloride. This programme included a familiarization
The same lot of the calibrator for automated systems period, an initial trial and a main trial. The protocol of
(Boehringer Mannheim GmbH) was used for this purpose. the main trial is shown in table 3. The main trial was
The mean of the nine calibration runs was then taken as split between two groups, each consisting of three
the factor, provided that the range of all results did not laboratories, which tested the same set of analytes. For
exceed 3 of the mean. the studies of linearity, drift and sample-related carry-
over, the different methods were divided between the six
During experiments for the between-day imprecision and evaluators.
method comparison (21 working days) autocalibration of
the analytical system was tested. The autocalibration is The satellite programme covered other analytes such as
triggered by an analyte-dependent calibration interval. proteins (C-reactive protein, ferritin, immunoglobulin A,
In order to avoid any additional effects on the results, a transferrin), drugs (phenobarbital, phenytoin, theophyl-
’start-up’ calibration was activated before the remaining line, digoxin) and urinalysis (albumin, /-N-a.cetyl-
experiments. Detailed information about the calibrators glucosaminidase, creatinine, sodium, potassium, chloride).
employed is shown in the appendix (table 10). In addition, creatinine kinase MB isoform and fruc-

Table 3. Protocol of the multicentre evaluation (core programme, main trial)*.

Imprecision
Within-run
On three different days, each day one run with 21 aliquots
--:Fhree control materials with different concentrations of the analyte.
--.One human serum pool at the decision level.
Between-day
Three control materials with different concent’rations of the analyte over 21 days
(evaluation and comparison instrument).
Drift

Two control sera and the calibrator were analysed every 30 min over 8 h to test eight methods (aspartate aminotransferase,
creatine kinase, 7-glutamyltransferase, calcium, iron, creatinine, total protein and uric acid).
-At zero hour, triplicate measurements were performed and the median was taken as the base value.
The percentage recovery of the base value was taken for assessing drift effects.

Analytical range limits [7]

--A high level sample was diluted with a low level sample to obtain a series of eleven concentrations with two being the baseline
samples and nine intermediate concentrations.
--Triplicate measurements on the 11 concentration levels were performed and the median for each level was calculated.
--The regression line (Passing/Bablok regression) was calculated using a range covering five concentration levels which was
assumed to be linear.
--The target values for all concentration levels were calculated from the regression lines.

Carry-over
Sample-related
Model of Broughton [ 11]:
---Measurements of five aliquots of a high-concentration sample (hi"" "hs) are followed by
--Measurements of five aliquots of a low-concentration sample (11" "15). The experiment is repeated 10 times.
If a carry-over effect exists, 11 is the most influenced, 15 the least influenced aliquot.
Reagent-dependent:
One control material was used.
Assay A influences assay B.
* For details, see reference [3]. (continued)
193
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 3 (continued).
--.Carry-over caused by the cuvettes
In a first step, reagents for assay A were requested 21 times. Just after the dispensing of the reagents for the 21st aliquot, the
analyser was stopped. In a second step, reagents for assay B were requested 42 times. The first 21 determinations were performed
in the cuvettes which previously had contained the reagents for assay A. These determinations would show carry-over effects
whereas the last 21 determinations would be uninfluenced. The difference of the medians of both series was the carry-over.
--.Carry-over caused by reagent probes and stirrers
Assay B was carried out 21 times. In the second step test A and B were alternately performed 21 times. The carry-over was the
difference between the medians of both series.

Interference

Protocol of Glick [8]

A specimen with concentrations at the decision level was spiked with the interfering substance and 10 serial dilutions were prepared
with the same baseline specimen. The different analytes were measured in triplicate. The percentage recovery of the baseline value
for each concentration level was calculated.

Accuracy
Calibration
The calibrators of BM/Hitachi 911 and of the comparison instrument were both run on each instrument.
Quality control in three control materials
--Median, calculated from the second of duplicate measurements over 21 days.
Interlaboratory survey
--One control material with concentrations not known to the evaluators.
--Median, calculated from the second of duplicate measurements over 10 days.
Method comparison in fresh human specimens
--10 to 15 specimens were analysed each day for 10 days on the BM/Hitachi 911 and on the comparison instruments. The total
number of specimens covered the entire analytical range.
--Comparison of the methods by calculation of the Passing/Bablok regression line [10-].
In the satellite programme within-run imprecision was determined using one or two control materials and one human pool. Between-day
imprecision was studied over 10 days with one or two control materials.
Drift, linearity, carry-over and interference studies were carried out only for selected analytes.

tosamine assays were run in this part of the evaluation to 6"7’meets the requirements’, and scores from 6"8 to l0
study. The results of these analytes are presented together ’exceeded the requirements’.
with those of the core programme.
The protocol included quality specifications which were
agreed at the evaorators’ first meeting: these are described Results
later in this paper.
Imprecision
Assessmenl of praclicabilily Acceptance criteria for imprecision were based on
Practicability was assessed with the aid of a recently statistical error propagation [3]. For within-run impre-
published questionnaire [6] comprising about 200 ques- cision in participating laboratories the median of the CVs
tions which covered all important aspects of an analysis should not exceed 2 for classical clinical chemistry
system in the clinical laboratory. The questions were analytes (enzymes and substrates) in serum (only in the
summarized into 14 groups, as shown in table 4. They tables is a differentiation made between serum and
were related to the installation of the analyser, organiza- plasma) and urine at all concentrations tested. The
tion of work, quality assurance and miscellaneous accepted CV was reduced to 1 for the determination of
characteristics. A first version of the questionnaire had electrolytes by ISE. Taking into consideration the
already been used for the assessment of practicability of problems associated with immunoassays, such as nonlinear
calibration or analytical sensitivity, it was agreed that
BM/Hitachi 747 [3].
CVs of less than 5 would be acceptable for these assays.
Grading was in comparison with the evaluators’ present
laboratory situation. The assessment was based on a scale Within-run imprecision data based on results from all the
from 0 to 10: a score of 0 meant unimportant, useless or control sera from all laboratories are shown in figures 1-3,
poor, and a score of 10 absolutely necessary or excellent. and data for individual control sera are given in the
A score of 5 could be interpreted as being acceptable or appendix (table 11). The medians of all analytes met the
comparable with the present laboratory situation. The acceptance criteria, except for phenytoin and digoxin in
grading was divided into three classes. Scores of up to 3"3 the low level control. In individual control sera, results
meant ’did not meet the requirements’, scores from 3"4 exceeding the acceptance limits were obtained for
194
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 4. Assessment of practicability--groups of attributes. range

PAMYL
median
Number Contents ASAT
of items CK
CK MB
Installation
GGT
(1) Environment 10 Installation conditions
cA
(space, power and
water supply, waste, CHOL
noise) CREA
(2) Spatial arrangements 14 Location of instrument FRUCT
components
E
(3) Training/operation 14 Operation (demand on
skills) T
Training course uA
Instruction manual

Organization of work Ck
(4) Start-up/shut-down 5 Duration
Availability cv [%1
(5) Sample processing 31 Tray
Loading Figure 1. Within-run imprecision for classical clinical chemistry
Vessels analytes and electrolytes, based on control material data from all
Predilution laboratories in the study.
Volumes
Volume detection
Identification
(6) Reagent handling 17 Preparation high
Loading
Identification
Stability
Exchange
Volume
(7) Work flow 23 Available tests
Test request
STAT procedure
Rerun
Result presentation
(8) Timing 7 Throughput
Startup time
STAT time [%1
Calibration
Operator time Figure 2. Within-run imprecision for homogeneous immunoassays,
based on control material data from all laboratories.
Quality assurance
9) Monitoring Sample processing
Reagent processing CREA
range
Measuring process median

Temperature control

-
MAU
Photometer control
10) Calibration 23 Materials NAG
Monitoring
Procedure NA
Frequency
11) Quality control 12 Number of controls K
Statistics
Real time/daily/ CL
cumulative
Alarms
Presentation cv [%1
Figure 3. Within-run imprecision for analytes in urine, based on
Miscellaneous control material data from all laboratories.
(12) Data processing 15 Hardware
environment Software
Communication cholesterol, iron, sodium, potassium, chloride and for
(13) Versatility Analysis modes albumin in urine.
Ease of applications
(14) Maintenance/ 15 Duration and frequency All analytes which had not met the quality specifications
troubleshooting Combinations of in control materials showed acceptable CVs in human
maintenance functions materials (table 5). Only the CV of creatine kinase MB
Volumes tbr priming isotbrm exceeded the acceptance criteria.
Trouble-shooting
Alarms Between-day imprecision was defined as acceptable if
Availability of technical median CVs for electrolytes were _< 2, for homogeneous
service
Power break-down immunoassays _< 10 and for the remaining analytes
investigated in serum and urine _< 3.
195
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 5. Within-run imprecision in a normal human serum, PAMYL range

plasma or urine pool (N 21). The results of the pools with ASAT
median

concenlrations close to the decision level are shown. CK

GGT
Analyte Unit Mean CV CA-
CHOL
Pancreatic a-amylase U/1 71 1.7 CREA
Aspartate aminotransferase U/1 30 1.3 FRUCT
Creatine kinase U/1 149 0.8 FE
Creatine kinase, MB isoform U/1 10 3.1 TP
7-Glutamyltransferase U/1 46 0.9
UA-
Calcium mmol/1 2"2 0.9
NA--
Cholesterol mmol/1 5.1 1.7
K
Creatinine gnol/1 99 1.6
Fructosamine gmol/1 250 1.4
Iron gmol/1 13 1.8 cv [/4
Total protein g/1 70 0.4
Uric acid gmol/1 350 1.0 Figure 4. Between-@ imprecision for classical clinical chemistry
Sodium mmol/1 141 0.5 analytes and electrolytes, based on control material data from all
Potassium mmol/1 4.0 0.6 laboratories.
Chloride mmol/1 101 0.6
C-Reactive protein mg/1 14 3"2
Immunoglobulin A g/1 1.8 2"5
Ferritin gg/1 239 2"2
Transfcrrin g/1 2"6 2"0
Phenobarbital gg/ml 16 2"8
Phenytoin gg/ml 8"6 2"9 IGA
high

Theophylline gg/ml 12 3.0 FT

Digoxin ng/ml 1.9 4.0
fi-N-Acctylglucosaminidase* U/1 10 0.6 T
Creatinine* mmol/1 1.0 1.0
Albumin* mg/1 33 1.4
Sodium* mmol/1 110 0.4
Potassium*
Chloride*

*Analytes in urine.
mmol/1
mmol/1
35
96
0.5
0.5
THE O
t
" cv [%]
10 11 12

Figure 5. Between-day imprecision for homogeneous immuno-

The medians and ranges of CVs in control materials assays, based on control material data ,from all laboratories.
analysed on BM/Hitachi 911 are shown in figures 4-6;
results obtained in individual control sera are presented
in the appendix (table 11). The medians met the defined
quality specifications in all cases. Individual CVs exceeding
range
the acceptance criteria were obtained for calcium, CREA
median
cholesterol, creatinine, sodium, chloride, phenytoin, MAU
digoxin in serum and tbr albumin in urine. The results
of the comparison instruments were of the same order as NAG

those of BM/Hitachi 91 (data not shown).

Analytical range limits

It was desirable for the measuring range to cover the 12

greatest part of the physiological and pathophysiological cv [%1

range. Analytical range limits were established by Figure 6. Between-day imprecision for analytes in urine, based
determining the difference between the measured values on control material data from all laboratories.
and the target values from the dilution experiments as
described by Bablok [7]. In the upper range, a method
was defined to be linear if this difference was less than
5. The lower range was judged on the basis of such chemistry analytes in serum (table 6). The lower limit
pragmatic considerations as absolute differences between tested for creatinine and calcium was found to be
measured and target values, and the diagnostic relevance 30 gmol/1, and 0"3 mmol/1, respectively.
of these differences. In the case of multi-point calibration,
a method could be called linear if a change in the target
In urine, the required high linearity ranges were reached
for all analytes. Linearity in the low range was analysed
concentration led to a proportional change in the
for albumin and sodium only. The albumin method was
measured concentration.
found to be nonlinear below 25 mg/1 and the sodium
A wide linearity range was obtained for all clinical determination below 10 mmol/1.

196
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 6. Analytical ranges on the BM/Hitachi 911.

Range
Sample
Analyte material Unit Tested between Found between

Aspartate aminotransferase serum U/1 14 870 14 870

(IFCC method)
Creatine kinase serum u/ l0 -1520 l0 -1520
7-Glutamyltransferase serum u/1 l0 -1250 l0 -1250
Pancreatic a-amylase serum u/1 16 -2400 16 -2400
Calcium serum mmol/1 0 6 0.3- 6
Cholesterol plasma mmol/1 2"9- 21.5 2"9- 21.5
Creatinine serum gmol/1 0 -2000 30 -2000
Iron serum gmol/1 0.9- 20 0.9- 20
10 215 10 215
Total protein plasma g/1 8 175 8 175
Uric acid plasma gmol/1 0 -1200 0 -1200
C-Reactive protein* serum mg/1 0 25O 0 25O
Immunoglobulin A serum g/1 6.3 6.0
]3-N-Acetylglucosaminidase urine U/1 0 78 0 78
Creatinine urine mmol/1 1.5- 23 1-5- 23
Albumin urine mg/1 0 5O0 2O 5OO
Sodium urine mmol/1 0 2OO 10 2OO
Potassium urine mmol/1 0 115 0 115
Chloride urine mmol/1 0 215 30 215
*Multi-point calibration, see discussion in text.

resulted in an elevation of lipase activities by about

Drift effects were not accepted if a systematic deviation
400 U/1 and of lactate dehydrogenase activities by about
from the initial value exceeded 3. Over an eight-hour- 30 U/1. Activation of a software-controlled wash-step for
these combinations reduced the carry-over effects to a
period, no drift effects were observed in any of the 10
methods tested. negligible amount < 10 U/1).

Calibration stability Interferences

Calibration stability based on individual calibration According to Glick et al. [8], a method is resistant to
frequency claims was observed over a period of 21 working interferences if the deviation between the baseline value
days. No deviations in recovery exceeding 3 were found. and the measured value is less than 10. Only methods
not fulfilling this acceptance criterion are shown in the
interferograms ofbilirubinaemia, lipaemia and haemolysis
Carry-over (figures 7-9).
Carry-over effects were assessed on the basis of the Of the 13 methods tested only two showed interference
observed change in recovery of an analyte. Instead of by bilirubin. Decreased values were found for the
adapting an individual deviation for each analyte it was creatinine (Jaffa method) and for the uric acid assays at
decided to use the within-run imprecision system perform-
ance. This was defined as a change of less than twice the
standard deviation being acceptable. 200 200

180 180
Sample-related carry-over was only tested for analytes 160 160
with a large physiological range and between urine and 140 140
serum specimens. For the potassium assay, a slight 120 120
carry-over effect (0"17 mmol/1) was observed from urine 100
to serum with a concentration ratio of77:1 (table 7). This 80 80
was rated as being of no clinical relevance. 60 60
CREA
40
Previous experience with BM/Hitachi systems had revealed
20
reagent dependent carry-over caused by the cuvettes for
the combination triglycerides/lipase. This effect was 50 100 150 200 250 300 350 400

avoided by activating a special wash solution step within Bilirubin [[Jmol/I]

the normal cleaning procedure for the cuvettes on
Figure 7. Interference of bilirubin. The hatched zone represents
BM/Hitachi 91 1. 90 to 110% recovery of the baseline value. Analytes tested were:
Carry-over caused by the reagent probes and the stirrers creatine kinase, 7-glutamyltransferase, creatinine, total protein, uric
was tested for the combinations triglycerides/lipase and acid, sodium, potassium, chloride, C-reactive protein, immuno-
aspartate aminotransferase/lactate dehydrogenase. This globulin A, theophylline, phenobarbital, and phenytoin.
197
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 7. Sample-related carry-over.

An aly e Median M edian Standard
(sample of high of low Ratio Median of deviation
materials) Unit concentration concentration h" 11 -15 13, 1,, 15
Creatine kinase U/1 6 282 12.0 524 0.30 0.42
(serum/serum)
Ferritin gg/1 5 480 196 28 0.80 12.8
(serum/serum)
Creatinine gmol/1 35 100 58.1 604 0.40 1.25
(urine/plasma)
Albumin mg/1 54 300 16.5 3 290 0-60 0.28
(serum/urine)
Albumin mg/1 39 400 21.3 850 -0.05 0.21
(plasma/urine)
Potassium mmol/1 224 2.9 77 0.17 0.02
(urine/serum)
Potassium mmol/1 161 4.0 4O O.O4 O.O2
(urine/plasma)
h, specimen of high, low concentration.
11 specimen of low concentration, influenced by h.
13, 14, ls specimen of low concentration, uninfluenced.

200 200 bilirubin concentrations above 120 gmol/1 and 140 gmol/1,
180 180 respectively (figure 7). Similar effects have been observed
160 160 on other BM/Hitachi instruments [3].
140 140

120 120 Haemolysis caused interference with four of the methods

100 100 tested (figure 8). Activities ofcreatine kinase and creatine
80 kinase MB isoform were increased while that of 7-
GGT
60 60 glutamyltransferase decreased. The protein assay gave
40 40 positive bias with haemolysis at high haemoglobin
20 20 concentrations > 4 g/l).
0
Two of the 16 methods evaluated were susceptible to
Haemoglobin [g/I] interference by lipaemia (figure 9). While immuno-
Figure 8. Interference of haemolysis. The hatched zone represents globulin A concentration was increased, an opposite effect
90 to 110//o recovery of the baseline value. Analytes tested were: was observed on the activities of aspartate aminotrans-
creatine kinase, crealine kinase MB isoform, 7-glulamyltransferase, ferase. A slight increase was also observed in concen-
crealinine, total protein, uric acid, and fruclosamine. trations of C-reactive protein (+ 12) at high concen-
trations of triglycerides > 1500 mmol/1).

_
Accuracy
Interlaboratory survey and quality control
200 200
180
Five of the six laboratories participated in an inter-
180

160 160
laboratory survey in which the concentrations of 15
140 140
analytes were determined on BM/Hitachi 911 and on the
0i CRP 120
comparison instruments using one control serum (see table
100 100
10, appendix). Results were defined as being acceptable
80 80
if their deviations from the target values did not exceed
60 60
by more than 5 for classical analytes and __+ 10 for
40 40
the homogeneous immunoassays.
20 20
As shown in figure 10, all values measured on BM/Hitachi
2.3 4.6 6.8 9.1 11.4 13.7 16.0 18.3 911 fulfilled the acceptance criteria except cholesterol,
Triglycerides [mmol/I]
iron and uric acid. Decreased values were found for
cholesterol and iron in two laboratories on BM/Hitachi
Figure 9. Interference of lipaemia. The hatched zone represents 911, whereas acceptable results were obtained on the
90 to 110% recovery of the baseline value. Analytes tested." comparison instruments. Slightly increased uric acid
pancreatic amylase, aspartate aminotransferase, calcium, cholesterol, values + 6) were measured in one of three laboratories
iron, sodium, potassium, chloride, C-reactive protein, immuno- on BM/Hitachi 91 1.
globulin A, transferrin, ferritin, phenobarbital, phenytoin,
lheophylline, and digoxin. Similar results were obtained in the quality control study

198
Z. Zaman el al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

115
Assessment of electrolyte concentrations in serum is
especially critical. Due to the narrow physiological range
110
of these analytes, a relatively large confidence interval
for the regression line was obtained. Therefore, method
105
comparisons were judged by the concentration range in
which the difference between the methods was less than
100 5%. The comparisons were accepted if less than 5%
deviations were obtained within the following concen-
tration ranges:
(1) Sodium 120-170 mmol/1
(2) Potassium 2- 10 mmol/1
(3) Chloride 80-130 mmol/1.
PAMYL
ASAT
CK
GGT
CA
CHOL
CREA
FE
TP
UA
NA ISE
K ISE
CI ISE
CRP
IGA As examples, six method comparisons comprising three
Laboratory 50 [] BM/Hitachi [] comp. instr.
constituents of serum (creatine kinase, iron, potassium),
one of urine (sodium) and two homogeneous immuno-
Figure 10. Inlerlaboralory survey using a control serum with assays (one turbidimetric protein assay and one CEDIA "
concentrations unknown to the evaluators. Each symbol represents assay) are shown in figure 11.
lhe median from measurements performed over 10 days in one
Method comparisons not fialfilling the acceptance criteria
laboratory. Results on BM/Hitachi 91"1 (dashed bars) and on the are listed in table 8. For classical clinical chemistry
comparison instrument (open bars) are shown. analytes in serum, seven comparisons out of 33 did not
meet the acceptance criteria. Two out of 12 method
with three control sera (see table 11, appendix). In comparisons in urine were unacceptable. Comparisons of
addition, the recovery of pancreatic amylase in one homogeneous immunoassays included seven protein
control (106"6%) exceeded the predefined limit. assays and 13 drug assays. All method comparisons
fulfilled the acceptance criteria except one tbr digoxin.
In the quality control study, recovery experiments with
the CEDIA > assays tbr therapeutic drug monitoring, and One of the 12 method comparisons for electrolytes in
with urine as sample material, were also included. The serum (sodium) did not meet the acceptance criteria. The
recovery of phenobarbital and phenytoin was within the range in which the results obtained on BM/Hitachi 911
acceptance criterion of 10 at the target value, whereas and on the comparison instrument deviated less than 5
that of theophylline was 88 in one control serum with was 127-194 mmol/1 instead of the required concentration
a concentration below the therapeutic range, in one of range of 120-170 retool/.
three laboratories on both BM/Hitachi 911 and on the
comparison instrument. This effect was not confirmed in
a satellite study in which a recovery of 98"4 was obtained Reliability
using the same control material. Reliability during the evaluation phase was rated with
the aid of a logbook in which any breakdown, deiict,
In control urines, all analytes met the acceptance criteria, malfunction or incident of the analysis system was
except creatinine, for which, in one control urine, a recorded. Some technical problems occurred during the
recovery of 106"9o was obtained on BM/Hitachi 911 and evaluation.
ot" 110"4 on the comparison instrument. In the other
two control urines, the recovery on BM/Hitachi 911 was The ISE unit at most evaluation sites had problems, ti0r
between 90 to 95. example noise alarm, air bubbles and crystallization of
potassium chloride at the sipper probe of the dilution
vessel. These resulted in an increased imprecision. A major
Method comparison modification of the ISE unit at the beginning of the main
In total, 74 method comparison studies were performed trial, which caused some delay in several laboratories,
using fresh human sera or urines; the resulting regression improved the performance.
equations are shown in the appendix (table 12). All
method comparisons were presented graphically and No other systematic malfunctions were observed. Single
assessed by the evaluators. malfunction episodes occurred with the sample probe and
with the stirrer resulting in both situations in a wrong
For the determination of enzymes and substrates in serum home position and a stoppage of the analyser. Software
and urine, agreement with the comparison method was malfunctions reported during the multicentre evaluation
rated as being sufficient if the slope of the regression were corrected in two new system software releases.
equation did not deviate more than 53/o from unity
and the intercept was less than 5 of the diagnostic-
ally important decision levels. The acceptance criteria
Assessment of practicability
were increased to -+-10 slope deviation tbr the homo- The practicability of BM/Hitachi 911 was judged in
geneous immunoassays. Depending on the analyte, comparison with the present situation in the individual
either the detection limit or a deviation of 10 from laboratories. The median of all laboratories was calculated
the relrence value were tolerated as intercept for these from the mean of all scores obtained for each group of
assays. attributes. These results are shown in figure 12.

199
 oo

200

oo

,300- 600 900 1200 1500

CK (BM/Hitachi 717) [U/l] FE (Centrifugal analyser) [pmol/I]

250

200

o o

o . .: o

5O

50 00 150 200 250

0
NA in urine (ISE; direct method) [retool/I]
K (FEP)[mmol/I]

600

500

400

300

200

1oo
- }’_’"
o

100 200 300 400 500 6OO

FT (FIA) [/Jg/I] DIG (FPIA) [ngiml]
Figure 11. Method comparison for some of the analytes in human serum/plasma/urine analysed on BM/Hitachi 91/ and comparison
instruments. Solid lines represent regression equations calculated according to Passing and Bablok [10] and dotted lines y x.
The following regression equations were obtained: CK (N 150), y l’OOx + 0"0; FE (N 150)y l’OOx O’l& K (N 150)
y l’OOx 0"07; NA in urine (N 100)y 0"98x 1"35; Fr (v oo)y 0"93x + 0"33; .DIG (N 100)y 1.03x 0.14.
2oo
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 8. Individual method comparison results exceeding the acceptance limits.

Regression anlysis
Comparison
Analyte U nit Slope In tercep method/instrument
7-Glutamyltransferase U/1 0.91 0"20 BM/Hitachi 737
Calcium mmol/1 1.06 0.11 Atomic absorption spectroscopy
Cholesterol mmol/1 1.08 0.33 BM/Hitachi 704
Creatinine gmol/1 1.11 -8.81 BM/Hitachi 717
Iron tmol/1 0"92 0.17 BM/Hitachi 704
Iron gmol/1 0.93 0.37 BM/Hitachi 747
Fructosamine gmol/1 0.98 17.35 Reduction of nitroblue
tetrazolium chloride; longer
incubation time
Albumin in urine mg/1 1.60 2.46 Rate nephelometry
Albumin in urine mg/1 0.73 1.50 Radioimmunoassay
Digoxin ng/ml 0.75 0.06 Fluorescence polarization
immunoassay

When the medians of the scores for BM/Hitachi 911 were

compared with those of the present laboratory situation, Grading
12 of 14 groups of attributes were rated higher for the
new analysis system. Especially high.scores were given for
Environment
spatial arrangement, sample processing, reagent handling,
workflow, calibration and versatility. The medians of the Spatial Arrangements
scores for training/operation and maintenance/trouble-
Training Operation-
shooting were identical for BM/Hitachi 911 and the
comparison instruments. None of the attributes was rated
lower on BM/Hitachi 911.
Sample Processing-

Reagent Handling--
More detailed information on the distribution of scores in
Workflow
relation to the main topics is shown in figure 13. Higher
scores (8 to 10) were given more frequently for BM/Hitachi Ti mi ng
911 than for the existing laboratory situation. For
Monitoring
BM/Hitachi 911, out of 194 attributes 12 were rated with
a score of 0 to (each by only one evaluation site) and Calibration
93 attributes with a score of 9 or 10 (each by one or
Quality Control--
several laboratories).
Data Processing--

Versatility

Maintenance Troublesh.-
Discussion

The versatility of the BM/Hitachi 911 analysis system was

tested in the multicentre study in different laboratory
sections--routine, STAT, homogeneous immunoassays Figure 12. Assessment of practicability of the BM/Hitachi 911
and urinalysis. The results were assessed on the basis of analysis system (solid line) compared to the laboratory situation
quality specifications tbr the various performance charac- (dashed line). The median of the scores is shown.
teristics previously agreed upon by the group of evaluators.
A good pertbrmance was found for most of the analytes
in all laboratory sections using different sample materials
of serum, plasma and urine. Although some analytes did
Within-run imprecision
not fulfill one or the other acceptance criterion, none
could be rated unacceptable. The results outside the The results for control and human materials (see table 5)
acceptance criteria are discussed below with the exception show. that overall the instrument had given a good
of the enzymatic methods for determination of sodium, performance. The individual CVs of a few analytes were
potassium and chloride. The performance of these slightly higher than the acceptance limits. These were
methods was unacceptable on the BM/Hitachi 911. New found mainly in one laboratory and in one control
standardization procedures for these tests are under material. Variable results were obtained for urinary
development. albumin in a control material with a concentration below

201
Z. Zaman el al. M-ulticentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Installation Organization of work

% Scores % Scores
35 35

15

0 3 4 5
Grading Grading

Quality Assurance Miscellaneous

% Scores % Scores
50 40-.

4.0

30

2o ,__., d4 >’//

10 i

10
0 3 5 7 8 9 10
Grading
Grading

Figure 13. Assessment of practicabili.. Distribution of scores for the main groups installation, organization of work, quality assurance
and miscellaneous. Dashed bars, represenl the grading for the BM/Hitachi 911 and open bars that of the present laboratory situation.

the decision level (CV vMues of 2"1I, 2"2: and 9"2% at quality specifications were not met with the calcium assay
15 mg/1). This variation in results could not be reproduced (interim specification 1"5, BM/Uitachi 911 1"6%),
in a human urine pool with’ an albumir concentration of creatinine assay (specification 2"2, BM/Hitachi 911
33 rag/1. With this material:,, a CV of 1"4 was obtained 2"3%), sodium assay (interim specification 0"7%, BM/
(see table 5). Hitachi 911 1"4%) and the chloride assay (interim
specification 1"0%, BM/Hitachi 911 1"7%). However,
since the imprecision of the calcium and the creatinine
assay were only 0"1 higher than the specifications, the
Belween-day imprecision results obtained can be judged acceptable. Owing to
Individual CV values of several analytes exceeded the the low biological variation of sodium and chloride, the
acceptance limits. For calcium, cholesterol and creatinine imprecision specifications were set very low: 0"3 for
the maximum CVs ot" 3"1 and 3"3 may still be seen as sodium and 0"7% for chloride. Since no available tech-
borderfline cases. The maximum CVs of 3"73/0 for sodium nology is capable of producing such a low imprecision at
and 2"3 for chloride reflected the condition of the ISE a reasonable cost, interim specifications with 0"7 and
unit which had to be modified during the evaluation. In 1"0 were proposed. Even these values are unlikely to be
spite of this modification, an improvement was not easily achieved with the existing technology.
observed in all laboratories. In one laboratory, the
CEDIA :"> Phenytoin and Digoxin assays, and the urinary
albumin assay in the low level control, also showed CV
values slightly higher than 10. Analylical range limits
The acceptance criteria tbr linearity of the measuring
range were fulfilled for all analytes, except for urinary
albumin at concentrations below 25 mg/1. This non-
Quality speccalions
linearity resulted in an underestimation of about 3 mg/1
Comparing the between-day imprecision results of the albumin at the decision level of 20 mg/1. This is within
BM/Hitachi 911 with the proposed quality specifications the range found during the multicentre evaluation tbr the
tbr the imprecision of analytical systems tbr clinical albumin test [14]. In view of the large biological variation
chemistry [12, 13], it was concluded that the BM/Hitachi of this analyte, this deviation was considered acceptable.
911 analysis system achieved these specifications for nearly During the evaluation, single-point calibration had been
all analytes evaluated in the study (table 9). The proposed used for the C-reactive protein assay. This mode of

202
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 9. Comparison of quality specifications for between-day imprecision and inaccuracy proposed by Fraser et al. [12 and 13] with
the results of the BM/Hitachi 911; in parentheses interim specifications. The data of BM/Hitachi 911 were obtained with Precinorm (R)
U, except for TF, IGA (Precinor,m (R) Protein) and drugs (Precinorm (R) TDM level 3).
Quality specifications BM/Hitachi 911 Quality specifications BM/Hitachi 911
between-day between-day inaccuracy mean recovery
Analyte CV () CV () ( deviation) (% deviation)
PAMYL 3.7 1.5 6.5 6.6
ASAT 7.2 1.9 6"2 3.4
CK 20.7 1.3 19.8 1.9
GGT 7.4 2"3 21.8 2.6
CA 0.9 (1.5) 1.6 0.7 (1.a) 1.8
CHOL 2.7 1.6 4.1 6.7
CREA 2.2 2.3 2.8 (4.4) 1.1
FE 15.9 2.1 8.9 4.5
TP 1.4 1.1 1.5 (2.8) 1.4
UA 4-2 1.3 4.0 (8.4) 5.9
NA 0.3 (0-7) 1.4 0.2 (0.6) 0.5
K 2.4 1.8 1.6 (4.8) 1.0
CL 0.7 (1.0) 1.7 0.5 (1.4) 0-6
TF 2.4 (4.0) 1.5 2.3 (4.8) 5.4
IGA 2.2 (3.8) 9.9 12.5 3.2
DIG 3.8 (4.7) 3.7 3.9 1.6
THEO 11.1 2.4 5.4 2.8
PHEBA 2.2 2.2 5.6 2.0
PHENY 3.6 3.3 4.1 4.3

calibration gave an upper limit of the analytical range of absorbance to exceed the photometric measurement range
80 mg/1 which is not sufficient for diagnostic purposes. of 3"3 absorbance units.
However, studies performed during the multicentre
evaluation of the C-reactive protein assay using multi-
point calibration showed a higher measuring range at Avcuracy
least up to 250 mg/1 [15-1, or even higher, depending on During the interlaboratory survey and the quality control
the standards used. Therefore, only multi-point calibration study a recovery ofless than 95 was found for cholesterol.
is recommended now by the manufacturer, and the results The assigned value of this analyte was confirmed by the
obtained during this evaluation are not reported. reference method of isotope dilution/mass spectrometry
[16]. Therefore, the effect could not be attributed to a
wrongly assigned value. In two of the three method
lnlerferences comparison studies a good agreement was found, whereas
Increased total protein values were measured at haemo- in the third increased cholesterol values were obtained on
globin concentrations above 4 g/1. This can be readily BM/Hitachi 911. These contradictory results could not
explained by the ttct that haemoglobin is a protein. The be explained. The effect was not observed in a satellite
activity of/-glutamyltransferase was decreased in haemo- study performed in one of the evaluation centres.
lytic samples. As a consequence, the 7-glutamyltransferase
test will be modified by the manufacturer in order to
Iron values below the acceptance criteria were found in
reduce this interference. Increased activities of creatine three of the four control sera used in the interlaboratory
kinase, and especially creatine kinase MB isoform, were survey and the quality control study. The effect was
found in the presence of haemoglobin. This is caused by confirmed in the method comparison study in two of the
the release of adenylate kinase from erythrocytes. Again, three laboratories (table 8). In the third laboratory, in
which the comparison instrument was a centrifugal
the manufacturer intends to modify the application for
the creatine kinase MB isoform assay, with the aim of analyser, a slope of 1-00 and an intercept of-0"15 gmol/1
iron were obtained in the regression analysis. The deviant
reducing the susceptibility of the method to interference results were probably caused by wrongly low values being
by haemoglobin.
assigned to the calibrators of the two comparison
Increased immunoglobulin A concentrations had been instruments.
measured in lipaemic samples. Theretbre, a modified If the results of the accuracy experiments are compared
immunoglobulin A assay is presently being developed to to the quality specifications proposed by Fraser el al. [12,
overcome this problem. The effect of turbidity on 13], all assays performed on BM/Hitachi 911 met the
aspartate aminotransferase activities has previously been requirements with the exception of cholesterol and
observed during other BM/Hitachi system evaluations transferrin.
I-3-1. It can be explained by the fact that the fresh reagents
tbr this assay have high initial absorbance and that the Eleven out of 74 method comparisons gave slope intervals
addition of a turbid specimen would cause the total and intercepts that were outside the acceptance limits

2O3
Z. Zaman el al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

(table 8). Differences in the 7-glutamyltransferase assay exceptions. Moreover, the practicability of this instrument
were caused by an incorrect temperature conversion factor exceeded the requirements for most of the attributes.
on the comparison instrument in one laboratory. The Owing to its versatility, the instrument is best placed as
method comparison tbr the calcium assay yielded a slope a consolidated workstation in the. small- to medium-sized
of 1"06. However, the differences between the values of laboratory, or as a back-up instrument for special
individual samples (from -0"13 to 0"18 mmol/1) were determinations like proteins, drugs or urinalysis in a
acceptable. In one of three laboratories a slope of 1"11 large laboratory.
was obtained in the method comparison for creatinine.
In spite of this, agreement was present at the decision
level. However, with this high slope, results of samples Acknowledgement
tiom dialysis patients are not correctly interpreted. In a
satellite study in one of the laboratories, the deviation was The authors wish to thank all their co-workers in the
confirmed. The manufacturer will change the application respective laboratories and departments participating in
in order to correct the discrepancy. The lack of agreement the study for their excellent support.
in the tiustosamine assay could be explained by the fact
that the comparison method was performed with a longer
incubation time. The evaluators accepted this deviation.
References
Albumin in urine showed discrepant results in two of the
three laboratories where the assay was compared with 1. ECCLS, Guidelines for the evaluation of analysers in clinical
nonturbidimetric methods. One out of the four digoxin chemistry. ECCLS Document, 3 (1986), 2
method comparisons was outside the accepted range. The 2. BABLOK, W., BAREMBRUCH, R., STOCKMAN, W., BRAVER, P., GRABER,
experiment had been pertbrmed in a laboratory where P., MICHEL, R and VONDERSCHMIDTa’, D., Journal of Automatic
Chemistry, 13 1991 ), 167.
many samples were derived from haemodialysis patients. 3. BONINI, P., CERIOTTI, f., KELLER, F., BRAVER, P., STOLZ, H.,
These samples might contain digoxin-like immunoreactive PASCUAL, C., GARCfA BEI:rRAN, L., VONDERSCHMIDTT, D. J., PEI,
thctors which are known to interfere digoxin assays [ 17]. P., BABLOK, W., DOMKE, I. and STOCKMANN, W., European Journal
The results ot’the C-reactive protein method comparisons of Clinical Chemistry and Clinical Biochemistry, 30 (1992), 881.
are not reported because single-point calibration used in
4. BAYER, P. M., KNEDEL, M., MONTALBETTI, N., BRENNA, S.,
PRENCIPE, L., VASSAULT, A., BAILEY, M., PHUNG, H. T., BABLOK,
the experiments had much narrower linear range than W., POPPE, W. and STOCKMANN, W., European Journal of Clinical
was originally thought. ThereIbre, the application was Chemistry and Clinical Biochemistry, 25 (1987), 919.
later changed to multi-point calibration. 5. BAADENHUIJSEN, H., BAYER, P. M., KELLER, H., KNEDEL, M.,
MONTALBETTI, N., BRENNA, S., PRENCIPE, L., VASSAULT, A., BAILEY,
M., PHUNG, H. T., BABLOK, W., POPPE, W. and STOCKMANN, W.,
Praclicabiliy European Journal of Clinical Chemistry and Clinical Biochemistry, 28
The outstanding tiature of the BM/Hitachi 911 analysis (1990), 261.
system is that it can be used as a consolidated workstation 6. STOCKMANN, W., BABLOK, W., POPPE, W., BAYER, P. M., KELLER,
F. and SCI-tWEmER, C. R., in Evaluation Methods in Laboratory Medicine,
tbr assays from diverse segments of a clinical laboratory
Ed. Haeckel, R. (VCH, Weinheim, 1993), 185.
(clinical chemistry, proteins, therapeutic drug monitoring). 7. BABLOK, W., in Evaluation Methods in Laboratory Medicine, Ed. Haeckel,
All tests necessary tbr a complete patient profile can be R. (VCH, Weinheim, 1993), 251.
pertbrmed using one primary tube without sample 8. GLICK, M. R., RYDER, K. W. and JACKSON, S. A., Clinical Chemistry,
splitting and additional manual workload. This simplifies 32 (1986), 470.
the distribution of samples and thus the organization of the 9. BABLOK, W., HAECKEL, R., MEYERS, W. and WosrqOK, W., in
laboratory. Hazards ot’sample mix-up as well as cost for Evaluation Methods in Laboratory Medicine, Ed. Haeckel, R. (VCH,
sample splitting are avoided. Moreover, analyses requiring Weinheim, 1993), 203.
a predilution step (tbr example, immunoglobulins) can 10. PASSiNg, H. and BABLOK, W., European Journal of Clinical Chemistry
also be handled easily without the need for high sample and Clinical Biochemistry, 21 (1983), 709.
volume. The analyser pertbrms the predilution itself. 11. BROV(ma’ON, P. M. G., GOWENIOCK, A. H., McCORMACK, J. J. and
NEILL, D. W., Annals of Clinical Biochemistry, 11 (1974), 207.
Urine samples can be run randomly in series with serum 12. FRASER, C. G., HYLTOFT PETERSON, P., RICOS, C. and HAECKEL,
or plasma specimens on BM/Hitachi 911. No relevant R., European Journal of Clinical Chemistry and Clinical Biochemistry, 30
carry-over effects between the different sample materials (1992), 311.
13. FRASER, C. G., HYLTOFT PETERSEN, P., RICOS, C. and HAECKEL,
were observed during the multicentre evaluation. An
R., in Evaluation Methods in Laboratory Medicine, Ed. Haeckel, R.
excellent analytical pertbrmance was obtained for urine (VCH, Weinheim, 1993), 87.
samples, showing that the instrument is well suited for 14. HuvcI-I, A., Wiener Klinische Wochenschrifi, 103, Suppl. 189 (1991),
urinalysis. 23.
15. BORQUE DE LARREA, L., Rus, A., CARLSTR6M, A., PIRA, U.,
Suitability of BM/Hitachi 911 for routine analyses DELOBBES, E., THIRY, P., FERRARI, L., BAROZZI, D., HAFNER, G.,
depends, of course, on the size of the laboratory. One of LOTZ, J., VAN OERS, e. J. M., LEERKES, g., SZYMANOWICZ, A.,
the evaluators considered the instruments well suited DUBOIS, H., HALLSTEIN, A. and DOKE, I., Klin. Lab., 39 (1993),
the routine requirements of his laboratory. The STAT 55.
facilities were rated positively in all evaluation centres. 16. COIEN, A., HERTZ, H.J., MANDEL, J., PANK, e. C., SCHAFFER, R.,
SNIEOSKI, L. T., SVN, T., WEICH, M.J. and WroTE, V. E., Clinical
The acceptance criteria tbr the analytical pertbrmance of Chemistry, 26 (1980), 854.
the Boehringer Mannheim/Hitachi 911 analysis system 17. ScI-IIEVSCn, H., JARAUSCH, J. and DOMKE, I., Wiener Klinische
were thlfilled tbr all laboratory segments with only a few Wochenschrift, 104, Suppl. 191 (1992), 59.

204
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Appendix
Table 10. Control and calibration materials.
Calibrator/control material Lot No.
Calibrator for automated systems, 172219
C.f.a.s. (lyophilized)
Calibrator for automated systems, 176 286
C.f.a.s (liquid)*
ISE Compensator 636238
Precimat (R) Fructosamine 172O78
Preciset (R) MAU 176872
Precimat (R) BNAG 628 755
’
Precimat CRP
Precimat (R) IGA
17463802
17420001
Precimat Transferrin 172644701
Preciset (R) Ferritin 173223
Calibrator low/high THEO A 7601 C/A 7602 C
Calibrator low/high PHEBA A 7735 B/A 7736 B
Calibrator low/high PHENY A 7502 B/A 7511 B
Calibrator low/high DIG A 7667 B/A 7668 B
Precinorm (R) U 169 151
Precipath e U 173579
Precipath (R) Evaluation* 174412
Precinorm Fructosamine 175 900
Precipath Fructosamine 171 630
Precinorm MAU
Precipath MAU
’ 175221
175 367
Precinorm :- NAG 628 757
Precinorm -’ Protein 177031
Precinorm TDM 17505090
Precinorm : TDM 3 17505290
(Boehringer Mannheim GmbH,
Mannheim, Germany)
Interlaboratory survey
Precinorm.-,’ U 168 588
C.f.a.s. Proteins 178 586
Lyphochek ,!’
Quantitative urine control normal 52201
level (Bio-Rad Laboratories,
M/inchen, Germany)
* Only for evaluation purposes.

Table 11. Imprecision and recovery of all laboratories in individual control materials on the BM/Hilachi 911.
Between-day Within-run

Mean recovery CV CV
Analytc Assigned
[unit] Control material value Median Range Median Range Median Range
PAMYL Precinorm" U 385 106.6 104.5-107-0 1.5 1.5-2.9 0.6 0.6-1.2
[U/1] Precipath" U 518 102.4 99.2-103.1 1.1 1.0-2-9 0.8 0.7-0.8
Precipath" Evaluation 504 100.4 99.3-102.0 1.7 1.3-2.4 1.0 0.8-1.1
ASNI" Precinorm’" U 53 103.4 97.3-104.6 1.9 1.3-2.7 1.6 1.2-2.2
[U/1] Precipath" U 125 100.5 94.8-100.7 1.4 1-0-1.5 0.8 0.8-0.9
Precipath" Evaluation 90 100.9 97.8-103.0 1.8 1.5-2.1 1.3 1.2-1.4
CK Precinorm" U 258 98.1 97.4-103-5 1.3 1.0-1.3 0.9 0.8-1.2
[U/1] Precipath" U 471 98.7 98.1-103.8 1.2 0.7-1.2 0.7 0.5-1.6
Precipath" Evaluation 322 102.2 100.6-106.8 1-2 0.8-1.5 1.1 0.7--1.6
GC;I" Precinorm" U 62 97.4 95.5- 9.8.2 2"3 1.6-2.4 1.1 1.0-1.4
[U/1] Precipath" U 189 99.0 97.4-101.0 1.7 1.4-2.0 O.5 O.5-0.8
Precipath" Evaluation 102 100.7 98.2-102.5 1.7 1.5-1.7 0.8 0.7-1.0
(continued)
2O5
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 11 (continued).
Between-day Within-run

Mean recovery CV CV
Analytc Assigned
[unit] Control material value Median Range Median Range Median Range
CA Precinorm U 2.3 101.8 100.4-103.5 1-6 1-4- 2.7 1.4 0.7-1.5
[mmol/1] ’
Precipath U
Prccipath (R) Evaluation
3.5
2.4
100.0
101.3
99.7-102.1
101.3-101.3
2-0
1.7
1.4- 2.5
1.4- 3.1
0.8
0-8
0.7-0-8
0.8-1.5
CHOL Precinorm R U 2.9 93.3 87.7- 95.5 1.6 1.2- 2.6 1.6 1.2-1.9
[mmol/1] Precipath U 3.2 89.3 80.3- 91.8 2-6 2.3- 3.1 1.6 1.1-1-8
Precipath Evaluation 4.0 96-1 92.9- 98.5 2.1 1.5- 2.4 2-0 1-5-2.1
CREA Precinorm (R) U 190 101.1 100.1-101.4 2.3 1.5- 2.5 1.6 0.9-1.6
[gmol/1] Precipath (R) U 344 98.3 97.1-100.2 2-7 1-7- 3.1 1.0 0.8-1.5
Precipath (R) Evaluation 163 94.8 94.4- 96.0 2.9 1.7- 3.3 1.4 1.4-1.5
FRUCT
[lamol/1]
Precinorm ’ Fructosamine
Precipath (R) Fructosamine
287
552
96-8
101.1
3.1
2.O
1.0
0.9
0.7-1.1
0.8-1-0
FE Prccinorm (R) U 19.5 85.5 84.7- 86.2 2.1 1-4- 2.2 1.8 1.3-3.1
[gmol/1] ’
Precipath U
Precipath (R) Evaluation
24.6
32.7
94.6
97.1
92.8- 95.9
96.5- 98.2
2-2
1.9
2.0- 2.3
1.3- 2.5
1.3
0.9
0.9-1-5
0.9-1.5
TP
[g/1]

UA
[-tmol/1-]
Precinorm

-
Precipath U

Precinorm (R) U
Precipath (R) U
U

Prccipath (R) Evaluauon

Precipath Evaluation
51
42
52
289
589
247
101.4
105-0
104-1
105.9
102.5
103.8
100.6-102.4
103.8-106-0
102.9-104.5
105.6-108.3
102.4-104.6
103.2-106-9
1.1
1.4
1.4
1-3
1-3
1.5
1.0- 1.1
1.1- 1.5
1.2- 1.5
1.2- 2.6
1.0- 1.7
1.3- 2.9
0.5
0.8
0.7
0.6
0-5
1-0
0.4-1.5
0.4-1.1
0.5-1.4
0.4-0.7
0.5-0.8
0.9-1.0
NA Precinorm U 129 99.5 98.2-101.3 1-4 1.4- 1.4 0-7 0.4-1.0
[mmol/1] Precipath U 135 100-1 100.1-101.7 1.4 1.3- 1.9 0.4 0.3-1.0
Precipath t Evaluation 84 101.7 96-4-101.7 2.6 2.3- 3.7 1-0 0-5-1-3
K Precinorm U 4.9 99.0 98.5-101.0 1.8 1-0- 2.0 0.8 0.3-1.1
[mmol/1] Precipath U 6.6 100.8 100.1-102.4 1.7 1.5- 1.9 0.6 0.4-1.4
Precipath(R) Evaluation 4.6 101.1 100.5-101.5 1.6 1.3- 2.0 0.7 0.5-1.3
CL Precinorm c-’ U 85 100.6 98.0-101.2 1.7 1-6- 2.3 1.1 0.5-1.2
[mmol/1] Precipath U 105 103.5 100.5-104.6 1.7 1.6- 2.3 0.5 0.4-0.7
Precipath Evaluauon 101 96.4 95.0- 97.3 1-8 1.6- 2.3 0.7 0.6-0.8
CRP Prccinorm Protein 49 102.7 100.7-104.6 2.6 0.8- 4.7 1.3 0.8-1.8
[mg/1]
IGA Precinorm (R) U 1.8 96.9- 97.8 1.3- 2.3 1-9-2.0
[g/1] Precinorm ’-" Protein 3.6 96.8- 96.8 0.7- 1.5 0-9-1.4
F’I’ Precinorm (R) Protein 95 102.0 2.8 1.6 1.3-3.5
[gg/1]
TF Precinorm (R) U 2.3 98.0 6.9 3.0 2.0-3.0
[-g/l] Precinorm (R) Protein 4.5 94.6 2.9 1.5 1.4-1-9
PHEBA Prccinorm TDM
(R)
6.1 103.3 96.7-103.3 6.0 3.1- 6.4 3.9 2.9-4.2
[gg/ml] Precinorm (R) TDM 3 49.2 97.8 95-6-102.0 2.2 1. l- 2-9 1.7 1-7-1.8
PHENY Precinorm TDM 4.4 106.8 95.5-106.8 6.6 4.6-11.7 5.9 5.6-8.4
[gg/ml-] Precinorm (R) TDM 3 25.4 95.7 94-9- 96-1 3.3 2-3- 4.0 2.6 2.1-4.5
THEO Prccinorm ’-e TDM 5.0 94.0 88-0- 96.0 4.2 3.6- 5.7 4.8 4.3-4.8
[gg/ml] Precinorm a’ TDM 3 25.3 97.2 93-7- 98.4 2.4 2.4- 2.5 2.1 2.0-2.3
DIG Precinorm : TDM 0.95 101.1 91.6-110.5 9.4 6.9-10.2 7.6 6.5-8.8
[ng/ml] Precinorm (R) TDM 3 3.2 98.4 95.0-104.5 3.7 1.6- 4.3 3.5 2.4-4.0
Prccinorm ’:e BNAG 1.1 0.9 0.8-1.1
[u/q
CREA Lyphochck* 7.7 101.4-106.9 1.6- 3,7 1.3-1.7
(.Urine) Precinorm ’-" MAU 16.1 90.7- 92.8 0.9- 3.3 1.1-1.2
[mmol/1-] Precipath MAU 4.3 93.5- 94.9 2-7- 4.0 1.3-1.4

206
Z. Zaman et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 11 (continued).
Between-day Within-run

Mean recovery CV CV
Analyte Assigned
[unit] Control material value Median Range Median Range Median Range
MAU Precinorm (R) MAU 15.5 90.9 89.7- 94.2 3.3 2.7-11.3 2"2 2.1-9.2
(Urine) Precipath (R) MAU 94.8 96.7 93.3- 97.8 1.6 0.8- 3.5 1.1 0.8-2.0
[mg/1]
NA Lyphochek (R) 60 99.8 1.6 0.4 0.4-0.5
(Urine)
[mmol/1]
K Lyphochek (R) 23 103.3 1.7 0.2 0.2-0.3
(Urine)
[mmol/1]
CL Lyphochek (R) 53 94.8 "2 0.4 0.4-0.5
(Urine)
[mmol/1]

Table 12. Method comparison for human serum/plasma/urine analytes assayed on BM/Hitachi 911 ) ,and,the comparison instrument
(x). Each line represents the result from one laboratory.
Regression analysis 95,?/o
y=bx+a Median
Sample Range distance
Analyte (Unit) material N (from x) b [9]
PAMYL (U/l) Serum 150 8 to 727 1.02 -1.27 6.3
Plasma 105 5 to 684 1.01 -0.79 17.3
Serum 225 28 to 2010 1-01 -2.59 32.9
ASAT (U/l) Serum 150 9 to 642 1-03 0.33 2"2
Plasma 105 10 to 549 0.99 -1.16 3.4
Serum 211 10 to 620 1.01 0.24 9.3
CK (U/l) Serum 146 3 to 763 1.01 1.45 5.6
Serum 150 6 to 1270 1.00 0.06 6.1
Plasma 150 5 to 1060 1.03 -0.39 8.7
CK-MB (U/l) Serum 150 3 to 116 0.95 1.02 2’8
Plasma 100 5 to 443 0.99 -0.10 10.7
GGT (U/l) Serum 147 3 to 1170 0.91 O.2O 3.4
Serum 149 0 to 593 1.00 0-00 0.1
Plasma 150 6 to 1225 1.04 0.43 15"5
CA (mmol/1) Serum 150 1.5 to 2.9 1.06 -0.11 0.09
Plasma 105 2"0 to 3.0 0"96 0.06 0.05
Serum 212 1.4 to 3.5 1.00 -,O-O2 0.11
CHOL (mmol/1) Serum 150 0.9 to 10.0 1.08 -0.33 0.15
Plasma 105 1.0 to 8.2 1.01 -0.01 O-25
Serum 150 1.4 to 13.9 1.01 0.00 0.24
CREA (lamol/1) Serum 149 23.9 to 1775 1.04 0.4 18.0
Serum 150 34.0 to 1169 1.11 -8.81 11.5
Plasma 150 47,0 to 648 i.02 -6.81 13.1
FE (gmol/1) Serum 150 1.0 to 45 0.92 0.17 0.5
Plasma 90 0.4 to 48 0.93 0.37 1.1
Serum 150 1.9 to 76 1,00 -0.15 1.1
FRUCT (gmol/1) Plasma 100 143 to 626 0.98 -17.35 19.4
TP (g/l) Serum 147 38 to 81 1-04 -0.30 1.5
Serum 150 42 to 117 1.04 0-02 2"2
Plasma 150 34 to 91 1.05 1.48 2.7
UA (gmol/1) Serum 149 59 to 987 1.01 12.03 2O’2
Serum 150 68 to 1034 1.05 -16.33 46.6
Plasma 150 76 to 834 1.00 -3.83 30’2
(continued)
207
Z. Zarnan et al. Multicentre evaluation of the Boehringer Mannheim/Hitachi 911 Analysis System

Table 12 (continued).
Regression analysis 95?/0
y=bx+a Median
Sample Range distance
Analyte (Unit) material 3/ (from x) b a fg]
NA (mmol/1) Plasma 90 130 to 149 1.24 -36"79 2.68
Plasma 90 129 to 149 1.18 -28.03 2"O9
Serum 75 119 to 152 1.07 -8"75 2.17
Serum 75 131 to 151 1.17 -26.35 1.67
K (mmol/1) Plasma 90 3.20 to 5"20 1.05 -0.36. 0.17
Plasma 150 2.91 to 5-22 1.00 -0-07 0-13
Serum 75 1.90 to 7.99 1.00 0.04 0.11
Serum 75 3.4 to 6.51 1.04 -0"2 0.39
CL (mmol/1) Plasma 90 93 to 114 1.03 -2"63 2"2O
Plasma 90 93 to 112 1.12 10.10 2.44
Serum 75 80 to 119 1.01 -0.63 1.88
Serum 74 84 to 112 1.13 -10.71 6.17
IGA (g/l) Serum 100 1.00 to 2.46 "00 0.03 0.03
Serum 99 0.42 to 64.32 "00 0.00 0.55
FT (gg/1) Plasma 100 6 to 518 0.93 0.33 25"9
TF (g/l) Plasma 90 1.00 to 6.00 0.95 0.14 O’28
PHEBA (gg/ml) Plasma 70 1"2 to 59.6 1.00 -0.60 1.63
Serum 98 1.0 to 43.3 0.95 -0.96 1.92
Serum 80 12.4 to 45.2 1.08 2.44 2"42
PHENY (gg/ml) Plasma 70 1.4 to 28.9 1.01 -0.59 1.46
Serum 99 0.3 to 60.2 1.00 0" 71 1’72
Serum 79 3.5 to 24.4 1.03 1.49 1.55
THEO (gg/ml) Plasma 70 2"0 to 62"0 1.06 -0.18 0"98
Serum 100 0.3 to 26" 7" 1.08 -0.21 0.77
Serum 50 2"1 to 23.4 1.04 -0"52 1.59
DIG (ng/ml) Serum 98 0"0 to 4.4 0.75 -0.06 0"76
Plasma 100 0.3 to 3.5 1.03 -0.14 0.18
Serum 100 0.1 to 4.3 O.96 O’O3 0.35
Serum 80 0.4 to 4.4 1.03 -0.09 0.40
BNAG (U/l) Urine 100 0.5 to 7.5 1.01 0.07 0.10
CREA (mmol/1) Urine 100 2"3 to 29"2 1.00 -0.15 O’22
Urine 100 1.5 to 18.8 1.00 -0"20 O.50
MAU (mg/1) Urine 100 1.5 to 1365 1.01 0.26 1.05
Urine 90 0.0 to 67 0"73 1.49 3.09
Urine 94 2"0 to 58 1.60 -2.46 2"65
NA (mmol/1) Urine 100 16 to 314 1.00 0.39 2"5
Urine 100 9 to 230 0.98 1.35 4.1
K (mmol/1) Urine 100 8 to 94 0.99 0.05 1.3
Urine 100 8 to 92 O’99 O.24 2"2
CL (mmol/1) Urine 100 12 to 286 1.01 -2"03 2"6
Urine 100 5 to 246 "00 5"90 11.5

208