Laboratory Performance in Neonatal Bilirubin Testing
Using Commutable Specimens
A Progress Report on a College of American Pathologists Study
Stanley F. Lo, PhD; Bernadine Jendrzejczak, BA; Basil T. Doumas, PhD
● Context.—In 2003 the Chemistry Resource Committee of
the College of American Pathologists introduced a com-
mutable specimen in the Neonatal Bilirubin Surveys. This
specimen was intended to help evaluate all bilirubin meth-
ods.
Objective.—To evaluate the effect of commutable spec-
imens on the performance of selected clinical analyzers in
measuring neonatal bilirubin from 2003 through 2006.
Design.—A human serum–based specimen enriched
with unconjugated bilirubin in human serum has been in-
cluded since 2003 in the Neonatal Bilirubin Surveys. The
bilirubin values of these specimens were determined by the
reference method and used to evaluate results reported by
various chemistry analyzers.
Results.—Coefficients of variation for College of Amer-
ican Pathologists All Data ranged from 4.9% to 6.2% for
the Neonatal Bilirubin Survey. However, coefficients of
I n 2003, the Chemistry Resource Committee of the Col-
lege of American Pathologists (CAP), acting on a report
regarding inaccuracies in the measurement of bilirubin in
clinical laboratories,1 introduced a new specimen in the
Neonatal Bilirubin Surveys (NB-Surveys). This specimen,
human serum enriched with unconjugated bilirubin, was
expected to be commutable (free of matrix effects) because
it closely resembles specimens drawn from healthy neo-
nates. Properly calibrated clinical analyzers were expected
to be accurate, providing total bilirubin (TBIL) values close
to those obtained by the reference method. The noncom-
mutable or conventional NB-Survey specimens consisted
of unconjugated bilirubin and ditaurobilirubin in bovine
serum.1
A guideline issued by the American Academy of Pedi-
atrics recommended a blood exchange transfusion for in-
fants older than 48 hours when bilirubin levels near 25
Accepted for publication April 2, 2008.
From the Reference Standards Laboratory, Department of Pathology
and Laboratory Medicine, Children’s Hospital of Wisconsin, Milwau-
kee (Drs Lo and Doumas and Ms Jendrzejczak); and the Department
of Pathology, Medical College of Wisconsin, Milwaukee (Drs Lo and
Doumas).
The authors have no relevant financial interest in the products or
companies described in this article.
Reprints: Stanley F. Lo, PhD, Reference Standards Laboratory, Chil-
dren’s Hospital of Wisconsin, 9000 W Wisconsin Ave, Wauwatosa, WI
53226 (e-mail: slo@mcw.edu).
Arch Pathol Lab Med—Vol 132, November 2008
variation for the 4 major instrument groups (Dimension,
Olympus, Synchron, and Vitros), which report 65% of all
results, varied from 2% to 3%. College of American Pa-
thologists All Data mean bilirubin values were within 0.46
mg/dL (7.8 ␮mol/L) of the reference method mean in
2003; in subsequent years these differences became larger,
peaking at 1.87 mg/dL (32 ␮mol/L) in 2005.
Conclusions.—The large systematic error of bilirubin
measurements is due primarily to failure of instrument
manufacturers to produce reliable bilirubin calibrators.
Primary calibrators should consist of human serum en-
riched with unconjugated bilirubin. Bilirubin values must
be assigned by the reference method, the performance and
robustness of which are reported in this article. Secondary
calibrators distributed to users must be traceable to pri-
mary calibrators.
(Arch Pathol Lab Med. 2008;132:1781–1785)
mg/dL (428 ␮mol/L) cannot be reduced by photothera-
py.2 Because this is considered a critical bilirubin value,
the CAP Chemistry Resource Committee decided to adjust
the bilirubin level in the human serum–based specimen
near 25 mg/dL (428 ␮mol/L).
Results from the 2003 human serum–based specimen
(NB-02) were very encouraging. The grand mean value for
CAP All Data was only 2.4% higher than the reference
method mean and the coefficient of variation (CV) % of
5.7.
This communication is a progress report on the perfor-
mance of laboratories participating in the NB-Surveys for
the years 2004 through 2006. The practice of including the
special, commutable NB-Survey specimen in the Chemis-
try Survey (C-Survey), as specimen C-96 or C-97, contin-
ued until the end of year 2006.
MATERIALS AND METHODS
The preparation and distribution to participating laboratories
of the human serum–based specimens has been described pre-
viously.3 All NB-Survey and C-Survey specimens (from 2003
through 2006) were analyzed for TBIL by the reference method4
at the Reference Standards Laboratory of the Children’s Hospital
of Wisconsin, Milwaukee. All other bilirubin results shown in this
article have been reported in the participant summary report.
Instruments specifically considered in this report include the BR2
(Advanced Instruments Inc, Norwood, Mass), Advia (Siemens
Medical Solutions Diagnostics, Tarrytown, NY), Aeroset and Ar-
chitect (Abbott Diagnostics, Abbott Park, Ill), Cobas Integra,
Laboratory Performance in Neonatal Bilirubin Testing—Lo et al 1781
 Table 1. College of American Pathologists (CAP) All
Data and Reference Method Mean Bilirubin Values
and Coefficient of Variation (CV)% for the Neonatal
Bilirubin Survey and Chemistry Survey Specimens
CAP All Data Reference
Mean Bilirubin Method Value, No. of
Specimen Value, mg/dL CV% mg/dL Laboratories
2003
NB-02 19.90 5.7 19.44 1743
C-96 19.92 5.6 19.44 4982
2004
NB-06 23.69 5.2 22.06 1784
C-97 23.41 6.4 22.06 4757
2005
Figure 1. College of American Pathologists (CAP) Neonatal Bilirubin
NB-05 26.25 6.2 24.38 1848
and Chemistry Surveys 2003–2006. Differences between CAP All Data
C-96 26.35 7.0 24.38 4552
and reference method mean values in milligrams per deciliter (1 mg/
2006 dL bilirubin ⫽ 17.1 ␮mol/L bilirubin).
NB-01 22.96 5.2 21.19 1890
C-97 22.44 5.5 21.19 4775
NB-11 22.32 4.9 21.04 1718
tween the NB-01 and NB-11 CAP All Data bilirubin mean
values (Figure 2) is due to large shifts in the mean values
reported by the Cobas Integra, Dimension, Roche Modu-
Roche Modular/Hitachi (Roche Diagnostic Systems, Branchburg, lar/Hitachi, and Vitros, which constitute 60% of the par-
NJ), Dimension (Dade Behring, Newark, Del), Olympus (Olym-
pus America Inc, Melville, NY), Synchron (Beckman Coulter Inc,
ticipating laboratories. Discrepancies between NB-01 and
Fullerton, Calif), Unistat (Leica Microsystems Inc, Buffalo, NY), C-97 (Figure 2) are primarily due to the large percentage
and Vitros (Ortho-Clinical Diagnostics, Raritan, NJ). of Vitros participants in the C-Survey.
According to the participant summary report, methods based Figure 3 shows differences between bilirubin mean val-
on the coupling of bilirubin with a diazo compound are used in ues for selected field methods and the reference method.
the Advia, Aeroset, Architect, Cobas Integra, Dimension, Roche With 3 exceptions, Advanced BR2, Aeroset (2003–2004),
Modular/Hitachi, Olympus, Synchron, and Vitros (TBIL slide). and Architect (2006), mean bilirubin values for all other
Methods based on the oxidation of bilirubin are used in the Ad- instrument groups are higher than those of the reference
via (vanadate oxidation) and Synchron (bilirubin oxidase). Meth- method. Note that there were no data from the Abbott
ods based on direct spectrophotometry are used in the Aeroset,
instruments in 2005, and both Abbott instruments use the
Architect, Advanced BR2, Unistat, and Vitros neonatal bilirubin
(BuBc) slide, which measures the sum of unconjugated bilirubin same diazo and direct spectrophotometry methods. The
and conjugated bilirubins (bilirubin monoglucuronide and diglu- positive bias ranges from less than 1 mg/dL to 4 mg/dL
curonide). According to Abbott, both the Aeroset and Architect (17.1 to 68.4 ␮mol/L). The Synchron group consistently
methods use a diazo method (diazotized 2,4-dichloroaniline), but shows a small positive bias. The negative bias of the BR2
for the NB-Survey users prefer direct spectrophotometry. has decreased over time and is smallest in 2006.
Reference methods are considered the gold standard
RESULTS against which field methods are compared and evaluated.
Table 1 shows CAP All Data mean bilirubin values and It is important, therefore, to establish the long-term per-
CV% for the NB-Survey and C-Survey specimens consist- formance characteristics of the reference method for serum
ing of unconjugated bilirubin in pooled human sera sent TBIL. From September 2003 to April 2007, the Reference
to participating laboratories during 2003–2006; also shown Laboratory of the Children’s Hospital of Wisconsin pre-
are values obtained by the reference method. For each year pared 11 bilirubin standard solutions by adding National
from 2003 through 2006, one commutable sample was pre- Institute of Standards and Technology Bilirubin, Standard
pared and sent out as a specimen in the NB-Survey and Reference Material 916, to pooled human sera to a con-
C-Survey; thus, in 2003, NB-02 was identical to C-96; in centration of 20 mg/dL (342 ␮mol/L). Following prepa-
2004, NB-06 was identical to C-97; in 2005, NB-05 was ration, these solutions were analyzed by the reference
identical to C-96; and in 2006, NB-01, C-97, and NB-11 method and the molar absorptivity (␧) of the alkaline azo-
were identical. Shown in Figure 1 are differences between pigment was compared with the values established in 2
mean values of CAP All Data and those of the reference round robin studies with participation from national and
method for NB-Survey and C-Survey specimens 2003– international laboratories.5,6 Collaborating laboratories in
2006. The best overall laboratory performance was the 1988 round robin study were from the United States
achieved in 2003. The CAP All Data mean value for NB- (3) and The Netherlands (2); in the 1998 round robin
02 was identical to that for C-96 and the difference, 0.46 study, laboratories were from the United States (3), Ger-
mg/dL (7.9 ␮mol/L) or 2.4%, between the CAP All Data many (2), and The Netherlands (1). Data from the 2 round
mean and that of the reference method was the smallest robins and from the Children’s Hospital of Wisconsin Ref-
in 4 years.3 erence Laboratory, shown in Table 2, indicate that the es-
In 2006, the commutable specimen was mailed to par- tablished molar absorptivity of the alkaline azopigment,
ticipants on 3 occasions, in April and November with the because of its reproducibility and narrow range, can and
NB-Survey (specimens NB-01 and NB-11, respectively) should be used to assess the accuracy of bilirubin calibra-
and in June with the C-Survey (specimen C-97). The un- tors consisting of unconjugated bilirubin in human sera.
usually large difference, 0.64 mg/dL (10.9 ␮mol/L), be- The 20-mg/dL (342-␮mol/L) stock solutions are dis-
1782 Arch Pathol Lab Med—Vol 132, November 2008 Laboratory Performance in Neonatal Bilirubin Testing—Lo et al
Figure 2. College of American Pathologists Neonatal Bilirubin Surveys 2006. Shown are differences between total bilirubin mean values (milli-
grams per deciliter) for specimens NB-01 and NB-11 (yellow bars) and between specimens NB-01 and C-97 (red bars) (1 mg/dL bilirubin ⫽ 17.1
␮mol/L bilirubin).
Figure 3. College of American Pathologists (CAP) Neonatal Bilirubin Surveys 2003–2006. Shown are differences between total bilirubin mean
values (milligrams per deciliter) for selected field and reference methods. The numbers on the horizontal axis identify different instruments: 1
indicates CAP All Data; 2, Advanced BR2; 3, Advia; 4, Aeroset (2003 and 2004)/Architect (2006); 5, Cobas Integra; 6, Cobas Integra, DZ Salt; 7,
Dimension; 8, Hitachi, Roche Modular; 9, Olympus; 10, Synchron Diazo; 11, Synchron BO; 12, Unistat; 13, Vitros 250/350; 14, Vitros 950; and
15, Vitros 5,1 FS (1 mg/dL bilirubin ⫽ 17.1 ␮mol/L bilirubin).

pensed in airtight plastic tubes and kept at ⫺70⬚C until
use. Each time the reference method is used, the stock
solution is analyzed to verify that its ␧ value is within the
accepted limits (mean value ⫾ 3 SD). Next, the stock so-
lution and 3 dilutions (2.5 mg/dL [42.7 ␮mol/L], 5.0 mg/
Arch Pathol Lab Med—Vol 132, November 2008
dL [85.5 ␮mol/L], and 10 mg/dL [171 ␮mol/L]) are an-
alyzed to construct a calibration curve, the regression
equation of which is used to calculate the results of the
specimens analyzed. Shown in Table 3 are molar absorp-
tivities, slopes, and intercepts of calibration curves ob-
Laboratory Performance in Neonatal Bilirubin Testing—Lo et al 1783
 Table 2. Molar Absorptivity of Bilirubin Alkaline
Azopigment Results From 2 Round Robins and the
Reference Laboratory of Children’s Hospital of
Wisconsin (CHW)*
n ␧ (L· mol⫺1 · cm⫺1)†
Round Robin 1988 (5 labs) 10 76 490
Round Robin 1998 (6 labs) 12 76 640
Reference Lab CHW 11 76 869
* n indicates number of determinations; ␧, molar absorptivity at 598
nm.
† Corrected for purity (98.3%) of bilirubin.
tained during a 3.5-year period, and precision data on
controls prepared in our laboratory. The data in Tables 2
and 3 demonstrate the impressive stability and robustness
of the reference method in US and European laboratories.
In 2003, the CAP All Data mean values for NB-02 and
C-96 were identical and within 2.4% of the reference value
(Figure 1). In our opinion this was due to the use of cal-
ibrators with accurately assigned bilirubin values. In sub-
sequent years accuracy deteriorated. Results for 2006 are
strange; there is good agreement for CAP All Data means
between specimens C-97 and NB-11, but the mean value
for NB-01 is greater than NB-11 by 0.64 mg/dL (10.9
␮mol/L). The lower mean bilirubin value for NB-11 was
due to the downward shift of 4 instrument group mean
values (Cobas Integra, Dimension, Hitachi/Roche Modu-
lar, and Vitros, which represent 59% of the reported re-
sults) (Figure 2). The close agreement between the mean
values for C-97 and NB-11, 22.44 mg/dL (383.7 ␮mol/L)
and 22.32 mg/dL (381.7 ␮mol/L), indicates a small, if any,
loss of bilirubin, in the 6-month period (June–November).
The most logical explanation for the difference between
NB-01 and NB-11 would be a change in calibrators or re-
assignment of bilirubin values in the calibrators used by
these instruments. Consistent with this explanation, Dade
Behring and Roche Diagnostics were among the diagnos-
tic companies that lowered the assigned values to their
bilirubin calibrators (Diane Stille, Dade Behring Inc, New-
ark, Del, written communication, May 7, 2006, and Mela-
nie Swartzentuber, Roche Diagnostics Corporation, Indi-
anapolis, Ind, written communication, August 18, 2006).
In addition, the Roche line of instruments revealed dra-
matic improvement from 2005 to 2006 (Figure 3) and is
consistent with an earlier change to lower assigned bili-
rubin calibrator values (Melanie Swartzentuber, Roche Di-
agnostics Corporation, written communication, October 5,
2005). The large difference between C-97 and NB-01 mean
bilirubin values for the Vitros indicates a lapse in the cal-
ibration of the TBIL and neonatal bilirubin (Bu⫹Bc) meth-
ods; because NB-01 contained only unconjugated biliru-
bin, one would expect the TBIL and neonatal bilirubin val-
ues to be very close.
Precision
Coefficients of variation for CAP All Data increased in
2004 and 2005, but in 2006 returned to the level of 2003
(Table 1). Lack of improvement is attributed to the large
differences in the mean bilirubin values reported by the
various instrument groups (Figure 3). It should be noted
that the CVs for the major instrument groups (Dimension,
Olympus, Synchron, and Vitros), which report about 65%
of all results, vary from 2% to 3%.
1784 Arch Pathol Lab Med—Vol 132, November 2008
Table 3. Long-term Calibration Stability and
Precision of the Reference Method for Total Bilirubin
Calibration Curve
(September 15, 2003–April 23, 2007)
SD
␧ (L· mol⫺1 · cm⫺1)* Slope Intercept
610
Mean 76 554 0.0770 ⫺0.0003
601
SD† 377 0.0004 0.0018
589
N 45 46 46
Min 75 579 0.0759 ⫺0.0038
Max 77 257 0.0782 0.0043
* ␧ indicates molar absorptivity.
† One standard deviation for a single measurement.
Accuracy
The goal of limiting the total error to ⫾10% of the ref-
erence method value3 can be realized only by improving
accuracy. Since 2003 the gap between the reference values
and the CAP All Data has progressively increased (Table
1); this applies to most instrument groups with notable
exceptions of Synchron, and lately Advanced BR2 and
Cobas Integra. To state simply: Calibrators used by most
manufacturers are inaccurate, that is, the assigned values
are much higher than the actual bilirubin concentrations.
Because for most major instrument groups precision is
quite good (CVs, 2%–3%), further improvement in accu-
racy would depend on the use of accurate calibrators.
Instruments from the same manufacturer using meth-
ods based on different principles exhibit the widest vari-
ation in results. Examples are as follows: (1) By direct
spectrophotometry (neonatal bilirubin method), differenc-
es between the reference method results and those re-
ported by the Vitros are 2.74 mg/dL (49.9 ␮mol/L) (NB-
01) and 2.06 mg/dL (35.2 ␮mol/L) (NB-11), respectively,
but only 0.36 mg/dL (6.1 ␮mol/L) by the TBIL diazo
method; the reference method and the mean TBIL Vitros
values for C-97 were 21.04 mg/dL (360 ␮mol/L) and 21.40
mg/dL (366 ␮mol/L), respectively. (2) Synchron values
obtained by the diazo method differ by less than 1.0 mg/
dL from those of the reference method while those of the
bilirubin oxidase method differ (except for 2003) from 1.97
mg/dL (33.7 ␮mol/L) to 2.96 mg/dL (50.6 ␮mol/L) (Fig-
ure 3). (3) According to Abbott package inserts, 2 methods
are available for TBIL for both the Aeroset and Architect
analyzers. One is a diazo method (diazotized 2,4-dichlo-
roaniline) and the other a direct spectrophotometry. How-
ever, because these are ‘‘open platform’’ instruments, the
users may have chosen a variety of other methods (listed
in the participant summary report as ‘‘JG w/blank,’’
‘‘Spectrophot w/o blank’’ or ‘‘w/blank,’’ ‘‘DZ Salt,’’ ‘‘ox-
idation,’’ ‘‘vanadate oxidation’’), which yield very variable
results. It is important to emphasize that calibrators con-
taining ditaurobilirubin or made in a protein base other
than human serum may not be suitable for calibrating all
bilirubin methods,1 that is, assigned values to calibrators
may have to be different for methods based on different
principles. A calibrator suitable for all methods is one
made with human serum enriched with unconjugated bil-
irubin. This is because it has the same composition as
specimens obtained from healthy neonates.
COMMENT
Because of the importance of accuracy in the measure-
ment of bilirubin in neonates, we proposed setting a goal
Laboratory Performance in Neonatal Bilirubin Testing—Lo et al
for limiting the total error (inaccuracy plus imprecision)
to ⫾10% of the reference method value at bilirubin con-
centrations greater than 10 mg/dL (171 ␮mol/L).3
In 2003, the first year of the introduction of the com-
mutable survey specimen consisting of human serum en-
riched with unconjugated bilirubin, that goal appeared to
be within reach for at least 80% of the reported results.
That was presumably due to better bilirubin calibrators
used in that year as judged by the small difference be-
tween the CAP All Data and reference method mean val-
ue. In 2006, for specimen NB-01, only the Advanced BR2
and the Synchron CX4/5CE met the upper limit of the
⫾10% goal; all instruments met the lower limit except the
Architect. The ⫾20% error allowed by the Clinical Labo-
ratory Improvement Amendments of 1988 implies that ac-
ceptable values for a specimen having bilirubin concentra-
tion of 25 mg/dL (428 ␮mol/L) would be from 20 mg/
dL (342 ␮mol/L) to 30 mg/dL (512 ␮mol/L); this very
large window should be unacceptable to pediatricians and
neonatologists. To improve bilirubin assays, manufactur-
ers of clinical analyzers and reagent kits will need to im-
Arch Pathol Lab Med—Vol 132, November 2008
prove the quality of bilirubin calibrators. A bilirubin ref-
erence method and reference material, as well as an es-
tablished molar absorptivity for the alkaline azopigment,
can be used to assess the accuracy of calibrators made
with unconjugated bilirubin in human serum. These cal-
ibrators may then be used to assign values to secondary
calibrators for field methods.
References
1. Lo SF, Doumas BT, Ashwood ER. Performance of bilirubin determinations
in US laboratories—revisited. Clin Chem. 2004;50:190–194.
2. American Academy of Pediatrics, Provisional Committee for Quality Im-
provement and Subcommittee on Hyperbilirubinemia. Practice parameter: man-
agement of hyperbilirubinemia in the healthy term newborn. Pediatrics. 1994;94:
558–565.
3. Lo SF, Doumas BT, Ashwood ER. Bilirubin proficiency testing using speci-
mens containing unconjugated bilirubin and human serum: results of a College
of American Pathologists study. Arch Pathol Lab Med. 2004;128:1219–1223.
4. Doumas BT, Poon PKC, Perry BW, et al. Candidate reference method for
determination of bilirubin in serum: development and validation. Clin Chem.
1985;31:1779–1789.
5. Doumas BT, Perry BW, McComb RB, et al. Molar absorptivities of bilirubin
(NIST SRM 916a) and its neutral and alkaline azopigments. Clin Chem. 1990;36:
1698–1701.
6. National Institute of Standards and Technology. Certificate of Analysis. Stan-
dard Reference Material 916a, Bilirubin. Gaithersburg, Md: National Institute of
Standards and Technology; 2001.
Laboratory Performance in Neonatal Bilirubin Testing—Lo et al 1785