Clinical Chemistry 52, No.
3, 2006 523
ization of LN status in CRC. An automated, multimarker
QRT-PCR assay for characterization of LNs from CRC
patients could be very useful for improved staging and
treatment decision-making in CRC.
This work was supported in part by Grant CA-01958
from the National Institutes of Health (to S.J.H.), and by
a cooperative research and development grant from
Cepheid (to T.E.G.).
References
1. Liefers GJ, Cleton-Jansen AM, van de Velde CJ, Hermans J, van Krieken JH,
Cornelisse CJ, et al. Micrometastases and survival in stage II colorectal
cancer. N Engl J Med 1998;339:223– 8.
2. Godfrey TE, Raja S, Finkelstein SD, Gooding WE, Kelly LA, Luketich JD.
Prognostic value of quantitative reverse transcription-polymerase chain
reaction in lymph node-negative esophageal cancer patients. Clin Cancer
Res 2001;7:4041– 8.
3. Sakorafas GH, Geraghty J, Pavlakis G. The clinical significance of axillary
lymph node micrometastases in breast cancer. Eur J Surg Oncol 2004;30:
807–16.
4. Doubrovsky A, de Wilt JH, Scolyer RA, McCarthy WH, Thompson JF. Sentinel
node biopsy provides more accurate staging than elective lymph node
dissection in patients with cutaneous melanoma. Ann Surg Oncol 2004;11:
829 –36.
5. Coello MC, Luketich JD, Litle VR, Godfrey TE. Prognostic significance of
micrometastasis in non-small-cell lung cancer. Clin Lung Cancer 2004;5:
214 –25.
6. Bilchik AJ, Saha S, Wiese D, Stonecypher JA, Wood TF, Sostrin S, et al.
Molecular staging of early colon cancer on the basis of sentinel node
analysis: a multicenter phase II trial. J Clin Oncol 2001;19:1128 –36.
7. Wiese DA, Saha S, Badin J, Ng PS, Gauthier J, Ahsan A, et al. Pathologic
evaluation of sentinel lymph nodes in colorectal carcinoma. Arch Pathol Lab
Med 2000;124:1759 – 63.
8. Saha S, Monson KM, Bilchik A, Beutler T, Dan AG, Schochet E, et al.
Comparative analysis of nodal upstaging between colon and rectal cancers
by sentinel lymph node mapping: a prospective trial. Dis Colon Rectum
2004;47:1767–72.
9. Noura S, Yamamoto H, Ohnishi T, Masuda N, Matsumoto T, Takayama O, et
al. Comparative detection of lymph node micrometastases of stage II
colorectal cancer by reverse transcriptase polymerase chain reaction and
immunohistochemistry. J Clin Oncol 2002;20:4232– 41.
10. Rosenberg R, Friederichs J, Gertler R, Hoos A, Mueller J, Nahrig J, et al.
Prognostic evaluation and review of immunohistochemically detected dis-
seminated tumor cells in peritumoral lymph nodes of patients with pN0
colorectal cancer. Int J Colorectal Dis 2004;19:430 –7.
11. Bustin SA, Dorudi S. Gene expression profiling for molecular staging and
prognosis prediction in colorectal cancer. Expert Rev Mol Diagn 2004;4:
599 – 607.
12. Raja S, Luketich JD, Kelly LA, Ruff DW, Godfrey TE. Increased sensitivity of
one-tube, quantitative RT-PCR. Biotechniques 2000;29:702, 704, 706.
13. Raja S, Luketich JD, Kelly LA, Gooding WE, Finkelstein SD, Godfrey TE.
Rapid, quantitative reverse transcriptase-polymerase chain reaction: appli-
cation to intraoperative molecular detection of occult metastases in esoph-
ageal cancer. J Thorac Cardiovasc Surg 2002;123:475– 83.
14. Raja S, El Hefnawy T, Kelly LA, Chestney ML, Luketich JD, Godfrey TE.
Temperature-controlled primer limit for multiplexing of rapid, quantitative
reverse transcription-PCR assays: application to intraoperative cancer diag-
nostics. Clin Chem 2002;48:1329 –37.
15. Raja S, Ching J, Xi L, Hughes SJ, Chang R, Wong W, et al. Technology for
automated, rapid, and quantitative PCR or reverse transcription-PCR clinical
testing. Clin Chem 2005;51:882–90.
16. Tassone F, Hagerman RJ, Taylor AK, Gane LW, Godfrey TE, Hagerman PJ.
Elevated levels of FMR1 mRNA in carrier males: a new mechanism of
involvement in the fragile-X syndrome. Am J Hum Genet 2000;66:6 –15.
17. Xi L, Luketich JD, Raja S, Gooding WE, Litle VR, Coello MC, et al. Molecular
staging of lymph nodes from patients with esophageal adenocarcinoma. Clin
Cancer Res 2005;11:1099 –109.
18. Ferris RL, Xi L, Raja S, Hunt JL, Wang J, Gooding WE, et al. Molecular staging
of cervical lymph nodes in squamous cell carcinoma of the head and neck.
Cancer Res 2005;65:2147–56.
19. Schroder CP, Ruiters MH, de Jong S, Tiebosch AT, Wesseling J, Veenstra R,
et al. Detection of micrometastatic breast cancer by means of real time
quantitative RT-PCR and immunostaining in perioperative blood samples
and sentinel nodes. Int J Cancer 2003;106:611– 8.
20. Takes RP, Baatenburg de Jong RJ, Wijffels K, Schuuring E, Litvinov SV,
Hermans J, et al. Expression of genetic markers in lymph node metastases
compared with their primary tumours in head and neck cancer. J Pathol
2001;194:298 –302.
21. Mitas M, Cole DJ, Hoover L, Fraig MM, Mikhitarian K, Block MI, et al.
Real-time reverse transcription-PCR detects KS1/4 mRNA in mediastinal
lymph nodes from patients with non-small cell lung cancer. Clin Chem
2003;49:312–5.
22. Bonner CA, Loftus SK, Wasmuth JJ. Isolation, characterization, and precise
physical localization of human CDX1, a caudal-type homeobox gene. Genom-
ics 1995;28:206 –11.
DOI: 10.1373/clinchem.2005.062844
Cell-Free Plasma DNA: A Marker for Apoptosis
during Hemodialysis, Johanna Atamaniuk,1* Katharina
Ruzicka,1 Karl M. Stuhlmeier,2 Alireza Karimi,1 Manfred
Eigner,3 and Mathias M. Mueller1 (1 Institute of Laboratory
Diagnostics, and 3 First Medical Department, Dialysis
Unit, Kaiser Franz Josef Hospital, Vienna, Austria; 2 Lud-
wig Boltzmann Institute for Rheumatology, Vienna, Aus-
tria; * address correspondence to this author at: Institute
of Laboratory Diagnostics, Kaiser Franz Josef Hospital,
Kundratstrasse 3, A-1100, Vienna, Austria; fax 43-60191-
3309, e-mail johanna.atamaniuk@wienkav.at)
Background: We evaluated whether cell-free plasma
DNA might be an appropriate marker for cell damage
during hemodialysis (HD) and whether it correlated
with annexin V expression and 7-amino-actinomycin D
(7AAD) nuclear staining of blood leukocytes.
Methods: Circulating DNA, annexin V, and 7AAD were
measured in HD patients before HD, 20 min after start
of HD, and after HD had ended. Healthy volunteers
provided control measurements. Necrosis and apoptosis
were monitored by gel electrophoresis.
Results: Plasma DNA concentrations were not signifi-
cantly different between controls and patients before
HD. Circulating DNA increased significantly (P <0.05)
after 20 min of treatment with HD. Post-HD concentra-
tions of DNA were significantly higher compared with
pre-HD and controls (P <0.005). Agarose gel electro-
phoresis showed ladders typical of apoptosis in
post-HD samples. Two subpopulations of CD45ⴙ leu-
kocytes were defined by flow cytometry: annexin Vⴙ/
7AADⴙ population for apoptosis, and annexin Vⴙ/
7AADⴚ for early apoptosis. Compared with healthy
controls, mean fluorescence (MF) of 7AADⴙ apoptotic
cells in the annexin Vⴙ/7AADⴙ subpopulation before
HD was not significantly increased. HD increased MF of
7AADⴙ cells in the annexin Vⴙ/7AADⴙ subpopula-
tion. In this subpopulation, MF of annexin Vⴙ cells was
significantly higher (P <0.01). MF of annexin Vⴙ cells
in the annexin Vⴙ/7AADⴙ subpopulation increased
during HD.
Conclusions: During HD, cell-free plasma DNA con-
centrations, annexin V expression, and 7AAD uptake in
524 Technical Briefs
leukocytes increases. The increase in plasma DNA,
appearing as ladders typical of apoptosis, and the 7AAD
uptake in leukocytes demonstrate that the predominant
portion of circulating DNA in HD patients originates
from apoptotic leukocytes.
© 2006 American Association for Clinical Chemistry
Membranes used in hemodialysis (HD) are made of
cellulose (e.g., cuprophan), modified cellulose (e.g., cellu-
lose acetate), or synthetic polymers (e.g., polysulfone).
Studies based on a large number of patients have shown
that the mortality rate of individuals undergoing HD with
unsubstituted cellulose membranes was higher than with
synthetic or modified cellulose membranes (1 ). During
HD, blood-membrane interactions lead to activation of
circulating cells, plasma proteins, (2 ), and the comple-
ment system (3 ). Furthermore, contact of cells with less
biocompatible membranes in vitro can lead to apoptosis
(4 ). Exposure of normal neutrophils to uremic plasma
accelerates in vitro apoptosis compared with cells incu-
bated with normal plasma (2 ). The apoptosis-inducing
activity of uremic plasma is modulated by use of dialyzers
with different degrees of biocompatibility (2 ). Dialysis
membranes can promote neutrophil apoptosis directly as
well as through their interactions with monocytes (5 ).
Stimulation of mononuclear cells is likely caused by the
interaction of cell-surface proteins with the dialysis mem-
brane (6 ).
For detecting early and late apoptosis in leukocytes,
complex methods are required, such as flow cytometric
measurements of annexin V expression in combination
with 7-amino-actinomycin D (7AAD) nuclei staining (6 ).
In early apoptosis, phosphatidylserine (PS) is translocated
from the inner to the outer surface of the plasma mem-
brane. In the presence of Ca2⫹, annexin V has a high
affinity for PS and binds only to PS that has been exposed
(7 ). Cells in the late stages of apoptosis and dead cells
have lost plasma membrane integrity and are permeable
for 7AAD (8 ). Increased apoptosis in leukocytes after HD
was reported when they were cultured for 12 to 48 h
(2, 6, 9 ).
Cell-free DNA concentrations are sensitive indicators of
cellular damage originating from apoptosis or necrosis
(10 –12 ). The aim of this study, therefore, was to evaluate
whether circulating DNA may be an appropriate marker
to demonstrate apoptosis during HD and whether it
correlates with annexin V expression and 7AAD nuclear
staining of blood leukocytes.
Blood samples were drawn from a control group of 30
healthy donors (10 men and 20 women; age range, 25–75
years) and from 10 HD patients (5 men; age range, 61–74
years; 5 women; age range, 46 – 80 years) before HD, 20
min after the start of HD, and after the end of HD. All
patients had been treated in the dialysis program for more
than 2 weeks; the longest period of dialysis treatment was
42 months. In the HD group, 8 patients were treated 3
times and 2 patients 2 times per week. The length of time
per HD procedure was 3.5– 4.5 h. Informed consent was
obtained from all patients.
For HD, the same synthetic polymer membranes [Fre-
senius Polysulfone Capillary dialyzers (F6) low-flux] were
used, and all patients were dialyzed through graft arte-
riovenous fistulas. The dialysate fluids and the water
purity were routinely tested for bacteriologic contamina-
tions and heavy metals. We administered the following
medications to patients: parenteral iron [100 mg iron(III)-
saccharose]; oral vitamins B1, B2, B6, and C, folic acid,
biotin, and nicotinamide; and subcutaneous erythropoie-
tin (1000 –15000 IU). Exclusion criteria for our study were
status post-kidney transplantation, autoimmune disease,
malignancy, or acute infection.
We isolated DNA from 800 ␮L of plasma and measured
it as published previously (10 ). After isolation and mea-
surement of plasma DNA, agarose gel electrophoresis
was performed. Agarose, at a final concentration of 1.5%
dissolved in Tris-borate-EDTA buffer (pH 8.0), was used
for electrophoresis, and after electrophoresis, gels were
stained with Vistra Green (Amersham Pharmacia Bio-
tech). In cases where higher sensitivity was needed,
polyacrylamide gel electrophoresis was performed. Iso-
lated plasma DNA samples were collected from 20
healthy volunteers and used after concentration (Micro-
con YM-100). The samples were separated on a 6%
polyacrylamide gel [30% acrylamide/bis (Bio-Rad/Labo-
ratories); 5⫻ buffer (50 mmol/L Tris, 380 mmol/L glyc-
erin, 2 mmol/L EDTA); 100 ␮g/mL ammonium persul-
fate; 0.85 ␮L/mL N,N,N⬘,N⬘-tetramethylethylenediamine].
All chemicals were purchased from Sigma (Germany).
Both gels were scanned on a FluorImager 595 (Amersham
Biosciences).
To 0.5 mL of EDTA blood (4 °C), we added lysis buffer
(8.99 g of ammonium chloride, 1.00 g of potassium
hydrogen carbonate, 0.037 g of titriplex-III, and 100.0 mL
of doubly distilled water) diluted 1:10 with doubly dis-
tilled water. After incubation (5 min at room temperature
in the dark) and centrifugation (300g for 5 min at 4 °C) to
pellet the leukocytes, the cells were washed at 4 °C with
phosphate-buffered saline (Dulbecco’s W/O sodium bi-
carbonate buffer; Gibco) supplemented with 10 mL/L
fetal calf serum (PromoCell®), and then washed with
ice-cold calcium buffer (10 mmol/L HEPES/NaOH, pH
7.4; 140 mmol/L NaCl; 2.5 mmol/L CaCl2; IQ-Products)
and adjusted to 1.5 ⫻ 103 cells/L.
We incubated 100 ␮L of cell suspension for 20 min in
the dark at 4 °C with 10 ␮L of combined antibodies.
Antibodies conjugated commercially to fluorescein iso-
thiocyanate (FITC), phycoerythrin, peridinin-chlorophyll-
protein (PerCP), and allophycocyanin (APC) dyes were
used. The following tubes were provided: isotype control;
anti-annexin V/FITC (IQ Products), 7AAD/PerCP (BD
Pharmingen), and CD45/APC (Becton Dickinson); and
anti-annexin V/FITC, CD45/phycoerythrin, 7AAD/
PerCP, and CD14/APC. After incubation, the cells were
washed and resuspended in ice-cold calcium buffer.
Surface marker analysis was performed by fluores-
cence-activated cell sorting with an FACSCalibur (Becton
 Clinical Chemistry 52, No. 3, 2006 525

from samples after HD revealed the typical apoptotic

Fig. 1. Agarose gel separation of plasma DNA samples during HD (A),
and polyacrylamide gel separation of concentrated cell-free plasma
DNA samples (B).
(A), lane 1, molecular size ladder (100 –1000 bp); lane 2, apoptosis control;
lanes 3 and 4, plasma DNA samples immediately before dialysis and after 20
min of HD, respectively, showing only slightly increased ladders. Lane 5, for
apoptosis, typically ladders appear in the DNA samples at the end of HD. (B),
lane 1, molecular size ladder (100 –1000 bp); lane 2, plasma DNA samples from
20 healthy volunteers concentrated into 1 sample. The pattern in lane 2 is typical
of apoptotic ladders.
Dickinson). We acquired 10 000 events using CellQuest
software (Becton Dickinson). We classified leukocytes as
“normal” (annexin V⫺ and 7AAD⫺), “early apoptotic”
(annexin V⫹ and 7AAD⫺), and “apoptotic” (annexin V⫹
and 7AAD⫹).
Data were analyzed with STATISTICA for Windows,
Ver. 6.0. Descriptive data are reported as the mean,
median, and SD. Statistical significance was determined
by Wilcoxon matched-pairs test for nonparametric vari-
ables. Statistical significance was defined as P ⬍0.05.
Before HD, plasma DNA concentrations in patients
were slightly increased [mean (SD), 13.89 (5.22) pg/␮L]
compared with controls [12.53 (4.70) pg/␮L]. After 20 min
of HD, plasma DNA concentrations [20.45 (10.27) pg/␮L]
were significantly increased (P ⬍0.05) compared with
pre-HD samples. During HD, plasma DNA continued to
increase, with the highest concentrations at the end of HD
[129.44 (83.38) pg/␮L; P ⬍0.005].
Agarose gel electrophoresis of plasma DNA obtained
ladders. In pre-HD samples and in samples taken after 20
min of HD, we could detect only attenuated ladders (Fig.
1A).
To demonstrate whether the cell-free DNA of healthy
individuals originates from apoptosis or necrosis, we
performed polyacrylamide gel electrophoresis, using con-
centrated DNA extracts from 20 healthy individuals.
These experiments again demonstrated typical apoptosis
ladders (Fig. 1B). Thus, the predominant part of circulat-
ing DNA in healthy individuals originates from apopto-
sis.
To detect the mean fluorescence (MF) of annexin V and
7AAD, we monitored 2 distinct subpopulations of leuko-
cytes: the annexin V⫹/7AAD⫹ (apoptotic), and annexin
V⫹/7AAD⫺ (early apoptotic) populations (Table 1).
The MF of 7AAD in the apoptotic cell population
showed no significant differences between the controls
and HD patients before start of HD. During HD, the MF
of the 7AAD⫹ cells increased from 710.1 to 998.9 in
patients’ samples. In addition, the MF of annexin V⫹ cells
from patients before HD was significantly higher (P
⬍0.01) than in controls. During dialysis, we again de-
tected significant changes in the MF of annexin V⫹ cells
(P ⬍0.05) in patients. The early apoptotic cell subpopula-
tions in controls and dialysis patients showed annexin V⫹
MF of 578.2 (131.5) and 662.2 (73.0), respectively (P ⬎0.05).
Uremia is associated with a state of immune dysfunc-
tion and increasing infection. Possibly, apoptosis relates
to dysregulation of the immune system (13, 14 ). Cell-free
plasma DNA has been found in many cases in which
apoptosis or necrosis was involved, suggesting that such
events are the main source for its presence. Measurement
of circulating DNA has been used as a prognostic tool in
the posttreatment monitoring of transplant patients (12 ).
In addition, it has been shown that within 15 min to 3 h
after major bodily injury, circulating DNA concentrations
in the peripheral blood of trauma patients developing
posttraumatic organ failure are significantly increased
(15 ). In our previous study, we showed increased concen-
trations of plasma DNA immediately after exhaustive
exercise and its disappearance within 2 h after the partic-
ipants had stopped running (10 ). The rapid appearance

Table 1. Flow cytometry results for CD45ⴙ apoptotic leukocytes.

MF

7AADⴙ (annexin Vⴙ/7AADⴙ population) Annexin Vⴙ (annexin Vⴙ/7AADⴚ population)

Mean SD Median P Mean SD Median P

Controls (n ⫽ 20) 637.8 147.5 633.5 4031.4 627.8 4001.8
HD patients (n ⫽ 10)
Before HD 710.1 144.9 715.3 NSa,b 4914.9 822.3 4835.8 ⬍0.01b
20 min of HD 705.6 115.0 679.7 NSc 4638.2 746.9 4637.3 NSc
End of HD 998.9 428.9 832.0 0.022d 4814.1 898.1 4699.8 0.037d
a
NS, not significant.
b
Controls compared with before HD.
c
Before HD compared with 20 min of HD.
d
20 min of HD compared with end of HD.
526 Technical Briefs
and disappearance of circulating DNA seems to be a
characteristic phenomenon for the nature of this DNA.
Using DNA separated on a polyacrylamide gel, we
have demonstrated that circulating DNA in healthy vol-
unteers shows ladders typical for apoptosis. Circulating
DNA seems to be mainly a result of apoptosis. In support
of our hypothesis, Fournie at al. (16 ) concluded in a
previous study that degradation of leukocytes in the
artificial kidney was responsible for the increase in circu-
lation of extracellular DNA.
Our data for 7AAD and annexin V in CD45⫹ leuko-
cytes, measured immediately after blood draw, demon-
strate apoptotis of leukocytes in HD. Previous in vitro
studies (9 ) have demonstrated increased apoptosis of
leukocytes after cultivation mimicking uremic conditions.
In our study, increases in the MF of 7AAD in cell
populations and in circulating DNA during HD mainly
reflect apoptosis induced by contact with the dialysis
membrane.
The source of increased DNA concentrations seems to
be apoptotic leukocytes. Possibly, DNA fragments are
leaking from the nuclei of leukocytes. In our study, the
increase in uptake of 7AAD in leukocytes indicates in-
creased permeability. The rapid increase in circulating
DNA during HD and the increase in 7AAD in leukocytes
suggest that both are caused by apoptosis.
We thank Grandits Ernestine for excellent technical assis-
tance.
References
1. Hakim RM, Held PJ, Stannard DC, Wolfe RA, Port FK, Daugirdas JT, et al.
Effect of the dialysis membrane on mortality of chronic hemodialysis
patients. Kidney Int 1996;50:566 –70.
2. Jaber BL, Balakrishnan VS, Cendoroglo MN, Perianayagam MC, King AJ,
Pereira BJ. Modulation of neutrophil apoptosis by uremic plasma during
hemodialysis. Blood Purif 1998;16:325–35.
3. Germin Petrovic D. Comparison of biocompatibility of hemophane, cellulose
diacetate and acrilonitile membranes in hemodialysis. Acta Med Croatica
2004;58:31– 6.
4. Carracedo J, Ramirez R, Pintado O, Gomez-Villamandos JC, Martin-Malo A,
Rodriguez M, et al. Cell aggregation and apoptosis induced by hemodialysis
membranes. J Am Soc Nephrol 1995;6:1586 –91.
5. Nahar N, Shah H, Siu J, Colvin R, Bhaskaran M, Ranjan R, et al. Dialysis
membrane-induced neutrophil apoptosis is mediated through free radicals.
Clin Nephrol 2001;56:52–9.
6. Martin-Malo A, Carracedo J, Ramirez R, Rodriguez-Benot A, Soriano S, et al.
Effect of uremia and dialysis modality on mononuclear cell apoptosis. J Am
Soc Nephrol 2000;11:936 – 42.
7. Peng ZH, Xing TH, Qiu GQ, Tang HM. Relationship between fas/fasl
expression and apoptosis of colon adenocarcinoma cell lines. World J
Gastroenterol 2001;7:88 –92.
8. Schmid I, Krall WJ, Uittenbogaart CH, Braun J, Giorgi JV. Dead cell discrim-
ination with 7-amino-actinomycin d in combination with dual color immuno-
fluorescence in single laser flow cytometry. Cytometry 1992;13:204 – 8.
9. Majewska E, Baj Z, Sulowska Z, Rysz J, Luciak M. Effects of uraemia and
haemodialysis on neutrophil apoptosis and expression of apoptosis-related
proteins. Nephrol Dial Transplant 2003;18:2582– 8.
10. Atamaniuk J, Vidotto C, Tschan H, Bachl N, Stuhlmeier KM, Muller MM.
Increased concentrations of cell-free plasma DNA after exhaustive exercise.
Clin Chem 2004;50:1668 –70.
11. Stroun M, Lyautey J, Lederrey C, Olson-Sand A, Anker P. About the possible
origin and mechanism of circulating DNA apoptosis and active DNA release.
Clin Chim Acta 2001;313:139 – 42.
12. Lui YY, Dennis YM. Circulating DNA in plasma and serum: biology, preana-
lytical issues, and diagnostic applications. Clin Chem Lab Med 2002;40:
962– 8.
13. D’Intini V, Bordoni V, Fortunato A, Galloni E, Carta M, Galli F, et al.
Longitudinal study of apoptosis in chronic uremic patients. Semin Dial
2003;16:467–73.
14. Andrikos E, Buoncristiani E, D’Intini V, Bordoni V, Bonello M, Levin N, et al.
Effect of daily hemodialysis on monocytes apoptosis. Blood Purif 2005;23:
79 – 82.
15. Rainer TH, Chan LY, Lit LC, Cocks RA. Derivation of a prediction rule for
posttraumatic organ failure using plasma DNA and other variables. Ann N Y
Acad Sci 2001;945:211–20.
16. Fournie GJ, Lule J, Dueymes JM, Laval F, Delobbe I, Vernier I, et al. Plasma
DNA in patients undergoing hemodialysis or hemofiltration: cytolysis in
artificial kidney is responsible for the release of DNA in circulation. Am J
Nephrol 1989;9:384 –91.
DOI: 10.1373/clinchem.2005.058883
Calculating Uncertainty of Measurement for Serology
Assays by Use of Precision and Bias, Wayne Dimech,1*
Barbara Francis,1 Jennifer Kox,2 and Graham Roberts,2 for the
Serology Uncertainty of Measurement Working Party (1 Na-
tional Serology Reference Laboratory, Australia, Fitzroy,
Victoria, Australia; 2 National Association of Testing Au-
thorities, North Melbourne, Victoria, Australia; * address
correspondence to this author at: National Serology Ref-
erence Laboratory, Australia, 4th Floor, Healy Building,
41 Victoria Parade, Fitzroy, Victoria 3065, Australia; fax
61-3-9418-1155, e-mail wayne@nrl.gov.au)
Background: In many countries, regulatory authorities
that use International Organization for Standardization
Standards to assess laboratory competence require an
estimate of the uncertainty of measurement (MU) of
assay test results. This estimate can be determined by
identifying all sources of variation, calculating the ex-
tent of variation, and using established methods to
combine the uncertainty. Alternatively, laboratory staff
may use existing data generated from evaluations, pro-
ficiency testing, or external run controls to determine
MU.
Methods: A quality-control (QC) sample with low reac-
tivity was tested by laboratories participating in a na-
tional QC program. The results of testing the QC sample
were entered into a shared database by use of an
Internet-based program, EDCNet. Using a statistical
approach that accounts for imprecision and bias of test
results, we estimated the MU of the laboratories.
Results: A total of 2167 test results of a single QC
sample reported by 18 laboratories were analyzed, and
the MU of 1 laboratory was estimated by the statistical
model described.
Conclusion: Using peer-group run control data, MU of
serologic testing can be estimated by taking into account
both imprecision and bias.
© 2006 American Association for Clinical Chemistry
Most regulatory authorities that use International Orga-
nization for Standardization (ISO) Standards to assess