Med Microbiol Immunol

DOI 10.1007/s00430-009-0127-4

ORIGINAL INVESTIGATION

Multicentre evaluation of the Elecsys® hepatitis B surface

antigen II assay for detection of HBsAg in comparison
with other commercially available assays
Ji-Dong Jia · Ma Hong · Lai Wei · Xin-Xin Zhang ·
Yuan-Li Mao · Lan-Lan Wang · Zhi-Liang Gao ·
Jin-Lin Hou · Jun Zhang

Received: 9 June 2009

© Springer-Verlag 2009

Abstract Screening for hepatitis B surface antigen
(HBsAg) demands highly sensitive and speciWc immunoas-
says with the ability to detect clinically relevant surface
gene mutants—the presence of which is an increasing con-
cern and can compromise test results. The objective of the
study was to compare the clinical and technical perfor-
mance of the fully automated Elecsys HBsAg II assay with
other widely used HBsAg immunoassays in China. This
was a multicentre study in which eight Chinese laboratories
compared the Elecsys HBsAg II assay with the Architect
HBsAg assay, AxSYM HBsAg assay, or generic microtitre
plate (MTP) enzyme-linked immunosorbent assays (manu-
factured in China) against preselected samples, including
recombinant surface gene mutants, and routine clinical
samples. Elecsys HBsAg II was the most sensitive assay for
detecting positive results in seroconversion samples: up to
14, 11, and 22 days earlier than the Architect, AxSYM, and
MTP HBsAg assays, respectively. It also detected all 211
preselected HBsAg-positive specimens and all 13 recombi-
nant HBsAg mutants; the AxSYM and MTP HBsAg assays
failed to detect 3 and 10 mutant samples, respectively. Sen-
sitivity was 100% for Elecsys HBsAg II with routine sam-
ples, compared with 99.1, 98.9, and 95.2% for the
Architect, AxSYM, and MTP HBsAg assays, respectively.
In conclusion, it was demonstrated that the Elecsys HBsAg
II assay is suitable for routine HBsAg screening in China.
Keywords Hepatitis B virus · Hepatitis B surface antigen
(HBsAg) · Immunoassay · HBV mutants
Introduction
Hepatitis B virus (HBV) surface antigen (HBsAg), a trans-
membrane glycoprotein, appears early in the natural history
of HBV. It is accepted as the Wrst immunological marker to

J.-D. Jia (&) · M. Hong L.-L. Wang

Beijing Friendship Hospital, Capital Medical University,
95 Yong-an Road, 100050 Xuanwu District,
Beijing, People’s Republic of China
e-mail: jia_jd@ccmu.edu.cn
L. Wei
Peking University People’s Hospital,
Peking University Hepatology Institute,
Beijing, People’s Republic of China
X.-X. Zhang
Ruijin Hospital, School of Medicine,
Shanghai Jiatong University, Shanghai,
People’s Republic of China
Y.-L. Mao
Beijing 302 Military Hospital of China,
Beijing, People’s Republic of China
West China Hospital Laboratory,
West China Medicine School, Sichuan University,
Chengdu, People’s Republic of China
Z.-L. Gao
The Third AYliated Hospital of Sun Yat-Sen University,
Guangzhou, People’s Republic of China
J.-L. Hou
Nanfang Hospital, Southern Medical University,
Guangzhou, People’s Republic of China
J. Zhang
Public Health Clinical Center, Fudan University,
Shanghai, People’s Republic of China

123

appear following infection and persists throughout the
course of chronic HBV infection [1], although in clinical
practice it is not always measured at the earliest stages of
infection. Its measurement is used to aid clinical diagnosis
of HBV infection, to monitor the eYcacy of antiviral ther-
apy, to screen blood and organ donors for the presence of
HBV, as well as to permit surveillance of individuals at risk
of either acquiring or transmitting the disease [2–4].
A number of diVerent assays are used to screen for
HBsAg in serum samples, most employing polyclonal or
monoclonal HBs antibodies to capture antigen in patients’
samples. Despite HBsAg testing being identiWed as the
method of choice to prevent transfusion transmitted infec-
tions [5], many weak HBsAg tests are still used routinely
[6]. Recent large scale comparisons of assays routinely
used in Germany highlighted that HBsAg screening tests
had not improved signiWcantly over the past decade and
that a minimum standard was required. This has been
recognised in Europe where a recent Directive speciWes that
HBsAg assays used routinely in diagnostic settings should
meet the same levels of analytical sensitivity as those used
in blood screening devices [7].
In China, where the present study was conducted, the
most commonly used HBsAg screening tests are the Archi-
tect® and AxSYM® HBsAg assays (Abbott Laboratories,
Diagnostics Division, Abbott Park, IL 60064, USA), as
well as generic microtitre plate (MTP) enzyme-linked
immunosorbent assay (ELISA) manufactured in China.
Various factors can inXuence the sensitivity and speciWcity
of HBsAg immunoassays. These include not only assay
format and analytical detection limit, but also patients’ dis-
ease status (early acute, resolving, or occult infection) and
degree of hepatic dysfunction [8]. The presence and titre of
HBs antibodies as well as haemoglobin or bilirubin can also
inXuence the test result [8]. In addition, the presence of
HBsAg mutants undetectable by many assays is now
emerging as a major concern in routine HBsAg screening
[8–10]. The most relevant HBsAg mutations are amino acid
substitutions at positions 145, 141, 131 in the major ‘a’
determinant as well as insertions between amino acids 122
and 123 [8, 11, 12]. In addition, mutations near the ‘a’
determinant or in regulatory elements of the surface gene
may also inXuence expression and in some cases (when
L-protein is overexpressed) on secretion [8].
The Elecsys® HBsAg II assay (Roche, Penzberg,
Germany), the successor to the Elecsys HBsAg assay, is a
newly developed screening assay with an improved ability
to detect HBsAg mutants. It uses both monoclonal and
polyclonal antibodies for HBsAg detection, which may
confer advantages for mutant detection over assays that use
monoclonal capture and tracer antibodies alone [8].
In this multicentre study, we compared the clinical and
technical performance of the Elecsys HBsAg II assay as a
123
Med Microbiol Immunol
screening test for HBsAg with other tests widely used in
China focusing in particular on the detection of prevalent
HBsAg variants, early detection of newly acquired infec-
tion, and routine HBsAg screening.
Materials and methods
Study design
The study was carried out at eight major centres across
China that included both hospital and university laborato-
ries involved in HBV diagnosis and screening. Using pres-
elected samples, together with serum samples from routine
clinical practice, each centre compared the performance of
the new Elecsys HBsAg II assay with their routine method
for HBsAg testing (Table 1).
Immunoassays
Across the eight centres, the Elecsys HBsAg II assay was
compared with the Architect HBsAg assay, the AxSYM
HBsAg assay or the Chinese MTP ELISA (Table 1). For
the Elecsys HBsAg II assay, measurements were performed
on the Elecsys 2010 (rack and rotor system) or on a modu-
lar analytics E170 analyzer. The Architect HBsAg,
AxSYM, and Elecsys HBsAg II assays were carried out
according to standard procedures [13–15]. The Chinese
MTP ELISA detected HBsAg levels using mouse monoclo-
nal antibodies to HBsAg. Peroxidase-conjugated mouse
monoclonal HBsAg antibodies were added to test or control
samples that had been incubated with mouse monoclonal
antibodies in a microtitre plate. In order to quantify HBsAg
levels by spectrophotometry (wavelength 450 nm), samples
were incubated in working Chromagen solution before
being stopped by acidiWcation.
All discrepant samples were conWrmed by HBsAg con-
Wrmation assay based on the principle of speciWc antibody
neutralisation. BrieXy, samples were pre-treated with the
conWrmatory reagent and control reagent before running the
Elecsys HBsAg II assay. In order to conWrm a positive
result, the cut-oV index generated for the sample had to be
<50% of that with the control agent, which had to have a
cut-oV index of ¸0.9 with the Elecsys HBsAg II assay.
Serum samples
Samples included three preselected sets of sera (Table 1),
the Wrst of which was a panel of eight commercially avail-
able seroconversion sample series (ZeptoMetrix Corpora-
tion; SeraCare Life Sciences Inc)—designated BCP 6281,
BCP 6272, PHM 903, PHM 908, PHM 915, PHM 920,
PHM 929, and PHM 934. All were sequential follow-up
Med Microbiol Immunol

Table 1 Study centres, assay procedures, and samples tested

Study centre Elecsys system Comparative routine assay Samples tested

Beijing Friendship Hospital, E 2010 12000 Architect 36 preselected HBsAg-positive sera

Capital Medical University, 317 routine samples
Beijing
Ruijin Hospital Laboratory, E 2010 12000 Architect 37 preselected HBsAg-positive sera
Shanghai Jiatong University, 3 native HBV mutants
Shanghai 318 routine samples
Peking University People’s Hospital, MODULAR Analytics E 170 12000 Architect 44 preselected HBsAg-positive sera
Beijing 326 routine samples
Beijing 302 Hospital, Beijing E 2010 Chinese MTP ELISA 20 preselected HBsAg-positive sera
528 routine samples
West China Hospital, E 2010 Chinese MTP ELISA 57 preselected HBsAg-positive sera
Sichuan University, 315 routine samples
Chengdu
Nanfang Hospital, E 2010 AxSYM 305 routine samples
Southern Medical University,
Guangzhou
Guangzhou Sun Yat-Sen Hospital, MODULAR Analytics E 170 AxSYM 293 routine samples
Sun Yat-Sen University,
Guangzhou
Shanghai Public Health Clinical Centre, MODULAR Analytics E 170 AxSYM 17 preselected HBsAg-positive sera
Fudan University, Shanghai 357 routine samples

samples from plasmapheresis donors who had donated
blood regularly and within short intervals over a period of
up to 115 days. The second set, provided by Roche Diag-
nostics, consisted of 13 recombinant mutant samples based
on naturally occurring HBV mutants. The mutations were
located in the highly variable ‘a’ region of the HBV surface
antigen and reXected clinically relevant native mutants
found in patients from all over the world as documented in
the literature. Recombinant mutants were generated by
transient expression in HepG2 cells. Mutants were secreted
into the cell culture supernatant. The subgenotypes of
recombinant samples R1, R5, R6, R7, R9, R10, R11, R12,
R13 were ayw3; R2 was adw; R3, R4 were ayw; R8 was
adw2.
The Wnal set, provided by study investigators, consisted
of diagnostically challenging samples, such as HBsAg-pos-
itive samples from particular patient cohorts or potentially
cross-reacting samples. In addition, each centre tested 200–
500 routine clinical specimens that formed part of the labo-
ratory’s normal sample throughput. All samples had tested
positive during routine assessments.
Results
Preselected samples
In the six study centres in which the seroconversion panels
were tested, the Elecsys HBsAg II assay was the most sensi-
tive assay for early detection of HBsAg—deWned as the
Wrst consistent positive test result from time of Wrst blood
sample (Table 2). In comparison with the Architect HBsAg
assay, the Elecsys HBsAg II assay detected HBsAg in 3 out
of 5 seroconversion samples tested (based on the presence
of HBV DNA) by as much as 14 days earlier. A similar
trend was seen in comparison with the AxSYM HBsAg
assay, although these data need to be interpreted with
caution as sample quality can aVect AxSYM assay results.
Because of the age and turbidity of the seroconversion
panel, results sometimes diVered between centres, and there
was also evidence of false-positive results. Compared with
the Chinese MTP ELISA, the Elecsys HBsAg II assay
detected more positive samples as well as much earlier. In
samples consistently positive with the Chinese MTP
ELISA, the Elecsys HBsAg II assay detected these as much
as 22 days earlier.
Results for the preselected study centre samples were
similar to those obtained with the seroconversion panel.
Again, the Elecsys HBsAg II assay demonstrated excellent
sensitivity. All 211 preselected HBsAg-positive specimens,
as deWned by the routine test procedure used by each centre,
were detected by the Elecsys HBsAg II assay.
All 13 recombinant mutants yielded positive test results
in the Elecsys HBsAg II assay (Table 3). Similar results
were obtained with the Architect HBsAg assay, although
with weakly positive results for recombinant mutants R1,
R2, and R11. There was no signiWcant diVerence between
samples. The AxSYM HBsAg assay performed less well,

123
 Med Microbiol Immunol

Table 2 Early detection of

Panel number Total days Mean time to consistent Wrst positive result (days)
HBsAg in preselected
(no. samples since Wrst
seroconversion panel series Elecsys HBsAg Elecsys HBsAg Elecsys HBsAg
per donation) blood sample
II vs. Architect II vs. AxSYM II vs. Chinese
HBsAg assay HBsAg assaya MTP ELISA

BCP 6281 (n = 12) 54 21 vs. 22 18 vs.16 19 vs. 47

BCP 6272 (n = 26) 115 94 vs. 102 94 vs. 98 94 vs. >115
PHM 908 (n = 8) 36 20 vs. 20 20 vs. 35
Where assays were compared at
PHM 915 (n = 13) 56 16 vs. 24 17 vs. >56
more than one centre the mean
across centres is provided PHM 920 (n = 6) 42 26 vs. 26 26 vs. 35
a PHM 903 (n = 4) 17 8 vs. 14
Excludes one apparent false-
positive test result with Elecsys PHM 929 (n = 9) 29 9 vs. 16
II and one with AxSYM HBsAg
PHM 934 (n = 6) 84 0 vs. 4
assay

detecting only 10 out of 13 recombinant mutants with any
consistency. All study centres using the AxSYM HBsAg
assay failed to detect recombinant mutants R9 and R11,
while two of the three centres also failed to detect recombi-
nant mutant R7. Of the four assays, the Chinese MTP
ELISA performed very poorly, detecting only 3 out of 13
recombinant mutants; moreover, of these R3 was only weakly
positive with absorbance readings just above the cut-oV value.
Preselected samples from Ruijin Hospital Laboratory,
Shanghai also included three rare native HBV mutants, two
of which tested negative with both the Elecsys HBsAg II
assay and the Architect HBsAg assay, while one was strongly
positive. The mutations occurred in the major antigenic part
of the ‘a’-loop between amino acids 126 and 134.
Routine samples
A total of 2,759 routine serum samples were screened for
HBsAg using the Elecsys HBsAg II assay and routine com-
parator assays. The Elecsys HBsAg II assay demonstrated
excellent sensitivity and speciWcity that were at least com-
parable to, if not better than, comparator assays (Table 4).
All 1,137 HBsAg-positive samples were identiWed by the
Elecsys HBsAg II assay on initial testing, giving an assay
sensitivity of 100%, compared with 99.1% with the Archi-
tect HBsAg assay (dropping to 93.2% at one study centre),
98.9% with the AxSYM HBsAg assay, and only 95.2%
with the Chinese MTP ELISA. With the exception of two
centres, where sample-to-sample carryover from positive
specimens may have contaminated some negative samples
leading to false-positive results, the Elecsys HBsAg II
assay proved highly speciWc for HBsAg detection.
Discussion
AVecting as many as two billion people worldwide, HBV is
a major public health problem [16]. This is especially true
in China, where the prevalence of HBsAg in the general
population is 7.18%, resulting in approximately 90 million
carriers [17]. Chronic HBV infection can lead to cirrhosis
and hepatocellular carcinoma that account for up to a 40%
lifetime risk of death [18].
Diagnosis and screening for HBsAg in China demands
tests with the highest sensitivity so that cases can be picked
up early. As our study has demonstrated, the Elecsys
HBsAg II assay was more sensitive for the detection of
early HBV infection in seroconversion samples than the
other assays tested. It detected positive samples approxi-
mately 2–14 days earlier than the Architect HBsAg assay
and by as much as 11 days earlier than the AxSYM HBsAg
assay. DiVerences with the Chinese MTP ELISA were even
more pronounced; the Elecsys HBsAg II assay detected
more positive samples among seroconversion samples and
by as much as 22 days earlier. The same was true for rou-
tine serum samples, where the Chinese MTP ELISA missed
approximately 5 in every 100 positive samples. Tested
under routine conditions, the Elecsys HBsAg II assay was
also more sensitive than the other assays, while providing
similar speciWcity. Failure to detect HBsAg-positive sam-
ples is critical in routine HBsAg screening because nega-
tive results are reported without conWrmation. Although
detection rates for both the Architect HBsAg assay and the
AxSYM HBsAg assay increased when repeat tests were
carried out, running duplicate tests on all routine specimens
as opposed to just initially reactive and borderline cases
adds considerably to costs and workload.
Today, the risk of missing HBsAg-positive samples is
also inXuenced by the emergence of HBV mutants [8–10].
Some native mutants lead to a loss of detection in all
HBsAg immunoassays, which may be due to a loss of anti-
genicity or to extremely low HBsAg concentrations—
below the detection level of the assay. HBsAg immuno-
assays vary in their overall sensitivity and in their ability to
detect certain HBsAg mutants [10]. It has recently been
reported that up to 20% of individuals with occult HBV do

123
 Med Microbiol Immunol

Table 3 Detection of recombinant HBV mutants

Recombinant mutant Subgenotype Elecsys HBsAg II Architect HBsAg AxSYM HBsAg Chinese MTP
assay; s/co (range) assay; IU/ml (range) assay; s/co (range) ELISA; s/co (range)

R1 (F8L/R24K/N40R/G43R/L94S/ ayw3 Positive (4.07–6.02) Positive (0.05–0.19) Positive (1.44–5.72) Negative (0.14–0.36)
M103/113A114/M133T/P142L/D144G)
R2 (I110L/S113T/T114S/T126I/N131T/ adw Positive (3.52–7.38) Positive (0.06–0.20) Positive (1.51–7.76) Negative (0.19–0.56)
F134Y/T143S/G145R)
R3 (S132Y/P142S/G145R) ayw Positive (10.07–13.59) Positive (0.86–1.08) Positive (6.94–12.96) Negative/Positive (0.15–1.01)
R4 (Q129P/F134R/P142L/ ayw Positive (13.93–19.68) Positive (0.45–0.59) Positive (2.78–7.09) Negative (0.03–0.08)
D144E/G145 K/S171F/L175S)
R5 (R122I) ayw3 Positive (2.79–3.32) Positive (0.27–0.34) Positive (1.64–3.60) Negative (0.01–0.78)
R6 (R122T) ayw3 Positive (9.8–12.02) Positive (1.48–1.53) Positive (5.36–13.69) Negative (0.89–0.23)
R7 (C124R) ayw3 Positive (3.16–4.59) Positive (15.95–20.47) Negative/Positive (0.98–2.36) Positive (16.41–22.13)
R8 (E122I) adw2 Positive (5.28–6.46) Positive (0.36–0.48) Positive (1.74–4.26) Negative (0.39–0.56)
R9 (T123N) ayw3 Positive (125.7–142.4) Positive (1.01–1.93) Negative (0.51–1.25) Positive (5.04–6.62)
R10 (G145K) ayw3 Positive (3.72–5.76) Positive (0.52–0.60) Positive (3.03–8.09) Negative (0.34–0.49)
R11 (122RA123) ayw3 Positive (3.35–4.48) Positive (0.05–0.10) Negative (0.51–1.17) Negative (0.59–0.60)
R12 (P142L/G145R) ayw3 Positive (1.70–2.40) Positive (0.33–0.42) Positive (2.64–5.97) Negative (0.16–0.56)
R13 (D144G) ayw3 Positive (3.95–4.02) Positive (0.12–0.13) Positive (1.19–3.12) Negative (0.47–0.49)
Elecsys HBsAg II s/co: <0.9 negative; 0.9–1.0 borderline; >1.0 positive
Architect HBsAg assay titre: <0.05 IU/mL negative; ¸0.05 IU/mL reactive
AxSYM HBsAg assay titre: <1.0 IU/mL negative; ¸1.0 IU/mL reactive or s/co: <2.0 negative; ¸2.0 positive
Chinese MTP ELISA s/co: <1.0 negative; ¸1.0 positive
Due to the small number of samples, statistical analysis was not possible

123
 Med Microbiol Immunol

not react to HBs antibodies and, therefore, go undetected

235/235 vs. 233/235 (100 vs. 99.15%)

174/177 vs. 177/177 (98.31 vs. 100%)

285/285 vs. 284/285 (100 vs. 99.6%)

127/129 vs. 129/129 (98.4 vs. 100%)

267/267 vs. 267/267 (100 vs. 100%)
265/265 vs. 265/265 (100 vs. 100%)

123/123 vs. 123/123 (100 vs. 100%)

140/140 vs. 140/140 (100 vs. 100%)
[19, 20]. In addition, the HBV genome, especially the enve-
lope gene, is mutated with unusually high frequency [21].
The S-gene contains an exposed major hydrophilic region
SpeciWcity Elecsys HBsAg (residues 111–155), which encompasses the ‘a’ determinant
II vs. comparator that is important for inducing immunity. Nucleotide substi-
tutions in this region are common and result in reduced
binding or failure to detect HBsAg in diagnostic assays. As
the present study has shown, only the Elecsys HBsAg II
Table 4 Sensitivity and speciWcity of the Elecsys HBsAg II, Architect, AxSYM, and Chinese MTP HBsAg assays in routine clinical samples (initial test results)

assay and Architect HBsAg assay detected all preselected

recombinant mutants. However, for several mutants the
Elecsys HBsAg II assay had values well above the cut-oV
182/182 vs. 180/182 (100 vs. 98.9%)

228/228 vs. 224/228 (100 vs. 98.3%)

263/263 vs. 257/263 (100 vs. 97.7%)
137/137 vs. 124/137 (100 vs. 90.5%)

153/153 vs. 153/153 (100 vs. 100%)

score, whereas the Architect HBsAg assay was only mar-
ginally positive. Mutations and insertions at positions 122
83/83 vs. 82/83 (100 vs. 98.8%)
59/59 vs. 55/59 (100 vs. 93.2%)
32/32 vs. 32/32 (100 vs. 100%)
Sensitivity Elecsys HBsAg II

and 124 appeared to lead to a loss of detection in compara-

tor assays, while only weakly positive results were reported
for R1 and R2 containing multiple mutations between
amino acids 8–145, and R11, an arginine-alanine insertion
vs. comparator

between amino acids 122 and 123.

Overall, our results indicate that the Elecsys HBsAg II
assay is a highly sensitive and speciWc immunoassay for
HBsAg detection in the Chinese population. Compared
with the other assays routinely used for HBV diagnosis and
screening, all of which gave false-negative results, it has
MODULAR Analytics E 170 and Architect

MODULAR Analytics E170 and AxSYM

MODULAR Analytics E170 and AxSYM

the potential to oVer improved sensitivity, especially for

detection of mutant strains and earlier detection of HBV
E2010 and Chinese MTP ELISA
E2010 and Chinese MTP ELISA

infection in seroconversion samples. Increased seroconver-

sion sensitivity is important because it reduces the diagnos-
tic window during which cases of HBV infection in
donated blood samples may be missed [9]. Our Wndings are
E 2010 and Architect
E 2010 and Architect

E2010 and AxSYM

consistent with results from a recent study in which the

Elecsys HBsAg II assay was also compared with the Archi-
tect HBsAg and AxSYM HBsAg immunoassays against a
Assays

similar series of test specimens [22]. Here, the Elecsys

HBsAg II assay showed a statistically signiWcantly greater
sensitivity with the seroconversion panel to the comparator
tests, as well as excellent sensitivity for prevalent native
Shanghai Public Health Clinical Centre, Shanghai (n = 357)

and recombinant HBsAg mutants [22].

Guangzhou Sun Yat-Sen Hospital, Guangzhou (n = 293)
Peking University People’s Hospital, Beijing (n = 326)

In conclusion, this study has demonstrated the high sen-

West China Hospital laboratory, Chengdu (n = 315)

sitivity and speciWcity as well as the reliable detection of

Ruijin Hospital Laboratory, Shanghai (n = 318)
Beijing Friendship Hospital, Beijing (n = 317)

HBV mutants of the Elecsys HBsAg II assay for screening

of HBsAg in routine clinical practice in China.
Nanfang Hospital, Guangzhou (n = 305)
Beijing 302 Hospital, Beijing (n = 528)

Acknowledgments The authors thank Elements Communications

for providing medical writing assistance supported by Roche Diagnos-
tics GmbH, Germany.

References
No. samples

1. Ganem D, Prince AM (2004) Hepatitis B virus infection—natural

history and clinical consequences. N Engl J Med 350:1118–1129
2. Hoofnagel JH, DiBisciglie AM (1991) Serologic diagnosis of
acute and chronic viral hepatitis. Semin Liver Dis 11:73–83

123
Med Microbiol Immunol
3. Frösner G, Schomerus H, Wiedmann KH, Zachoval R, Bayer B,
Bäcker U et al (1982) Diagnostic signiWcance of quantitative
HBsAg determination in acute and chronic hepatitis B infection.
Eur J Clin Microbiol 1:52–58
4. Centers for Disease Control, Prevention (1998) Recommendations of
the immunisation practices advisory committee: prevention of peri-
natal transmission of hepatitis B virus: prenatal screening of all preg-
nant women for hepatitis-B surface antigen. MMWR 37:341–355
5. Nick S, Scheiblauer H (2007) Sensitivities of CE-marked HIV,
HCV and HBsAg assays. J Medical Virol 79:S59–S64
6. Katsoulidou A, Moschidis Z, Sypsa V, Chini M, Papatheodoridis
GV, Tassopoulos NC et al (2007) Analytical and clinical sensitiv-
ity of the Procelix Ultrio HIV-1/HCVHBV assay in samples with
a low viral load. Vox Sanguinis 92:8–14
7. European Community (1998) EC directive 98/79/EC. OYcial
Journal of the European Communities. L331/1
8. Weber B (2005) Recent developments in the diagnosis and moni-
toring of HBV infection and role of the genetic variability of the S
gene. Expert Rev Mol Diagn 5:75–91
9. Dufour DR (2006) Hepatitis B surface antigen (HBsAg) assay—are
they good enough for their current uses? Clin Chem 52:1457–1459
10. Gerlich W (2004) Diagnostic problems caused by HBsAg
mutants—a consensus report of an expert meeting. Intervirology
47:310–313
11. Seddigh-Tonekaboni S, Waters JA, JeVers S et al (2000) EVect of
variation in the common ‘a’ determinant on the antigenicity of
hepatitis B surface antigen. J Med Virol 60:113–121
12. Weber B (2005) Genetic variability of the S gene of hepatitis
B virus: clinical and diagnostic impact. J Clin Virol 32:102–112
13. Roche Diagnostics Elecsys HBsAg II Package Insert (2007)
14. Abbott Diagnostics Architect Package Insert (2006)
15. Abbott Diagnostics AxSYM Package Insert (2008)
16. Kane M (1995) Global programme for control of hepatitis B infec-
tion. Vaccine 13(Suppl 1):S47–S49
17. Lok ASF, McMahon BJ (2007) Chronic hepatitis. Hepatology
45:507–539
18. Pan C, Jia J, Chu C-J, Hou J, Hu K-Q (2008) Epidemiology, pre-
sentation, and treatment of chronic HBV infection in mainland
China, in Taiwan, and in Chinese Americans in the United States.
Curr Hepat B Rep 2:159–165
19. Torbenson M, Thomas DL (2002) Occult hepatitis B. Lancet
Infect Dis 2:479–486
20. Fang Y, Teng X, Xu WZ, Li D, Zhao HW, Fu LJ et al (2009)
Molecular characterization and functional analysis of occult hepa-
titis B virus infection in Chinese patients infected with genotype C.
J Med Virol 81:826–835
21. Hollinger FB (2007) Hepatitis B virus genetic diversity and its im-
pact on diagnostic assays. J Viral Hepat 14(Suppl 1):11–15
22. Mühlbacher A, Weber B, Bürgisser P, Eiras A, Cabrera J,
Louisirirotchanakul S et al (2008) Multicenter study of a new
fully automated HBsAg screening assay with enhanced sensitivity
for the detection of HBV mutants. Med Microbiol Immunol
197:55–64
123