Multiple Factors Contribute to Positive Results
for Hepatitis A Virus Immunoglobulin M Antibody
Adnan Alatoom, MD, PhD; M. Qasim Ansari, MD; Jennifer Cuthbert, MD
? Context.—In the United States, a successful vaccination
program for hepatitis A virus (HAV) infection has decreased
both its incidence and the true positive rate for diagnostic
Objective.—To survey positive results of HAV IgM tests
and determine the effect of changing ordering options.
Design.—We reviewed all positive results for IgM
antibody to HAV between January 2007 and December
2010. Patient demographics, clinical history, and laboratory
data were recorded and the encounter, order, and reason for
test reviewed. Each result was categorized as indicating
acute, recent, resolved, or indeterminate HAV infection.
Results.—A total of 10735 tests were performed; 35
patients had 49 positive results. Most positive test results
were associated with outpatient visits and were ordered in
the assessment of patients with liver disease, but not
clinical acute hepatitis. In the final analysis, 4 patients had
acute hepatitis A and 20 individual patients had recent
and/or resolved hepatitis. All but 1 of the remaining 11
patients had another established cause of liver disease with
a positive IgM HAV antibody test result; data to determine
causality were insufficient. The total number of tests
requested annually decreased more than 35% with the
introduction of computerized physician order entry.
Conclusions.—Current assays for IgM HAV antibodies
are overused in the absence of clinical acute hepatitis;
future clinical decision support may improve patterns of
order entry. Most patients have findings consistent with
HAV exposure but not acute hepatitis; dormant viral
infection may be a continuing source of antigen.
(Arch Pathol Lab Med. 2013;137:90–95; doi: 10.5858/
viral hepatitis in the United States.1Since the introduction
of a safe and effective vaccine and its widespread adoption,
the actual number of cases has decreased to approximately
10% of the peak level.2Of paramount importance, any bona
fide new diagnosis of acute hepatitis A has public health
implications. Cases are reported to the local health
departments to identify a possible common source of
infection and provide postexposure prophylaxis to contacts
of the index patient.
Testing for immunoglobulin (Ig) M antibodies against
HAV (IgM anti-HAV) is the mainstay of the serologic
diagnosis for acute hepatitis A infection and has been highly
sensitive and specific.3–5The presence of any detectable
antibody (total anti-HAV) plus the absence of high-titer
IgM-specific antibodies is used to differentiate between past
and current infection. In the past few years, isolated reports
have documented an increase in the number of positive
results for IgM anti-HAV in patients whose illnesses were
epatitis A virus (HAV) infection contributes approxi-
mately half of the clinically apparent cases of acute
not consistent with the case definition of hepatitis A in
conjunction with the decrease in true incidence.6–8
We conducted a retrospective study to determine how
commonly we may be encountering potential false-positive
results after noticing the occurrence of positive IgM anti-
HAV test results in the absence of clinical findings
suggesting acute viral hepatitis. We explored possible
underlying reasons and the effect of changing ordering
MATERIALS AND METHODS
An institutional review board approved a retrospective medical
record analysis for all patients with positive results for IgM anti-
HAV antibodies between January 2007 and December 2010.
Patients were identified by querying the laboratory information
system (Cerner Millennium, Cerner Corporation, Kansas City,
Missouri) to generate the list of samples and order type (individual
test or group) associated with positive results. Results from internal
and external quality control assessments were excluded.
Patient demographics, clinical history, and laboratory data were
obtained through review of electronic and paper medical records.
Patient age, sex, and race/ethnicity were recorded for each patient.
Laboratory data included test results for bilirubin, aspartate
aminotransferase (AST), alanine aminotransferase (ALT), all viral
hepatitis serologic assays, rheumatoid factor, and human immu-
nodeficiency virus antibody at the time of diagnosis.
Test order information was obtained from the electronic medical
record and from review of paper medical records. The ordering
Accepted for publication March 30, 2012.
From the Department of Pathology, University of Texas South-
western Medical Center, Dallas.
The authors have no relevant financial interest in the products or
companies described in this article.
Reprints: M. Qasim Ansari, MD, Department of Pathology,
University of Texas Southwestern Medical School, 5323 Harry Hines
Blvd, Dallas, TX 75390-9073 (e-mail: qasim.ansari@utsouthwestern.
90 Arch Pathol Lab Med—Vol 137, January 2013IgM Anti-HAV Testing in the 21st Century––Alatoom et al
user, ordering provider, reason for test, and any associated code
from the International Classification of Diseases-Ninth Revision
were recorded for each positive test result. Until 2009, all orders
were hand-written by providers or conveyed as a telephone order
to authorized staff and then entered into the electronic medical
record by clerks (Figure). For outpatients, the serologic tests for
viral hepatitis were together in 1 section on a paper order form. In
2007, there were 2 groups of hepatitis tests, as well as individual
tests. The groups consisted of acute hepatitis serology tests (IgM
anti-HAV, hepatitis B surface antigen, total anti–hepatitis B core
with reflex IgM if positive, anti–hepatitis C virus) and chronic
hepatitis serology tests (hepatitis B surface antigen, total anti–
hepatitis B core, and anti–hepatitis C virus). Paper forms contained
an individual check box for IgM HAV testing but did not contain a
check box for total HAV antibody.
Starting in May 2008, all IgM anti-HAV tests were ordered
individually. There was only 1 group available: hepatitis panel
(hepatitis B surface antigen and anti–hepatitis C virus), which did
not include HAV testing. Providers also hand-wrote nonstandard
orders. These included ‘‘hepatitis labs’’ and ‘‘hepatitis serologies.’’
They were interpreted by order entry staff choosing from the
available tests for clerk order entry. Starting in November 2008 and
continuing throughout 2009 and 2010, computerized physician
order entry was introduced to the hospital system. Its implemen-
tation in the Emergency Services Department (November 2008)
and inpatient areas (late April 2009) was completed en bloc,
whereas in outpatient clinics, the process was slow. Some outreach
programs are still using paper forms.
IgM Anti-HAV Measurements
All assays were performed with the VITROS anti-HAV IgM assay
on the VITROS ECI System (Ortho-Clinical Diagnostics, Rochester,
New York) that uses the antibody class capture technique and
chemiluminescence. Results were calculated as a normalized signal
relative to the signal/cutoff (S/CO) value, with S/CO below 0.8
interpreted as negative; S/CO of at least 0.8 but below 1.2, as
borderline; and S/CO of 1.2 or above, as positive.
Data Analysis Definitions
After review of all available information, each positive test result
was associated with a specific diagnostic category. (1) Acute
hepatitis A: elevated aminotransferase levels with peak value above
500 U/L and negative serology test results for hepatitis B and
hepatitis C infections; (2) Recent acute hepatitis A: elevated
aminotransferase levels with peak value below 500 U/L and either
evolving S/CO (positive to borderline or negative) or resolving
aminotransferase values (normal AST level); (3) Resolved acute
hepatitis A: aminotransferase levels normal; (4) Indeterminate:
elevated aminotransferase levels below 500 U/L; neither evolving
S/CO nor resolution of abnormal aminotransferase values.
Between January 2007 and December 2010, a total of
10735 IgM anti-HAV tests were performed (Table 1). The
percentage of positive test results during this time period
was 0.5%; 35 individual patients tested positive for IgM anti-
HAV. In addition, 3 of the patients had a borderline test
result on blood taken 6 months previously (n ¼ 1) and 1 or
13 days later (n¼1 each). The mean age of the patients was
41 years (median, 43 years; range, 9–81 years); 20 of 35 were
male; 13 were black, 14 were Hispanic, 6 were non-Hispanic
white, and 1 was Asian.
A total of 49 separate tests had positive results. Eight
patients had 2 or more positive test results, of which 2 were
inadvertent duplicate orders during a single hospital
admission. One patient had 5 positive test results, 3 patients
had 3 positive test results, whereas the remainder had 2
positive results. Signal/cutoff results for the 2 duplicate
orders were essentially identical (S/CO of 2.70 and 7.02,
compared with 2.44 and 6.95, respectively). The demo-
graphic, laboratory, and assay data are shown in Table 2 for
each diagnostic category of HAV.
Number of positive test results for hepatitis A virus (HAV) immunoglobulin M (IgM) for each year in the 2007–2010 period, and method of test
ordering. Abbreviations: CPOE, computerized physician order entry; ED, emergency department.
Table 1.Results of Hepatitis A Immunoglobulin M Tests Each Year in 2007–2010 Period
Year TotalPositive Low Levela
Arch Pathol Lab Med—Vol. 137, January 2013IgM Anti-HAV Testing in the 21st Century––Alatoom et al 91
In 2008, the highest number of positive tests (n¼21) was
obtained. This included 1 patient with classical acute
hepatitis, 1 patient with prior symptoms consistent with
acute hepatitis, 2 patients (3 tests) with potential false-
positive results (Table 3), and 10 low-level positive results.
In Dallas County, a few more cases of acute hepatitis A were
reported in 2008 (n¼46) than in 2007 (n¼42) or 2009 (n¼
40), potentially leading to a greater number of positive
The order-associated data for the 49 encounters with the
35 patients are depicted in Table 4. Between 2007 and 2008,
orders were initially written on paper by providers and
entered into the electronic medical record by clerical staff
(Figure). In 3 instances, no order was found in the medical
record in association with the encounter information. The
commonest encounter site was an outpatient clinic (33 of 49,
67%). More than half of the orders (26 of 49, 53%) were
either acute hepatitis serology tests or the individual test.
When the individual test was ordered, all providers also
ordered hepatitis B and hepatitis C tests (data not shown).
In most instances, the test was ordered for either laboratory
evidence of liver disease (elevated aminotransferase level) or
as a repeat after an earlier positive result; together, these
accounted for 39 of 49 orders (80%).
The results of liver tests that are frequently abnormal in
patients with acute viral hepatitis (aminotransferases and
bilirubin) are shown in Table 5 in association with the
reason for ordering the IgM anti-HAV test. The peak values
around the time of the initial test order are depicted. Four
patients had aminotransferase levels greater than 500 U/L;
they presented with jaundice (n¼3) or abdominal pain (n¼
1). Three of these patients were hospitalized. For 11
patients, all test results were normal. The patient who was
tested because of household exposure to an index case had
normal AST and bilirubin levels.
Other Serologic and Immunologic Markers
Other immunologic test results are shown in Table 3. All
patients were tested for hepatitis B surface antigen and
hepatitis C antibody. One patient was confirmed positive for
hepatitis B surface antigen, whereas 2 had false-positive
results. Three patients had a false-positive hepatitis C virus
antibody result by enzyme-linked immunosorbent assay.
There were 8 other patients with a positive antibody test
result for hepatitis C virus infection. Hepatitis C viremia was
confirmed in 5 patients, one of whom had a positive
rheumatoid factor. The other patients with chronic hepatitis
C were not tested for rheumatoid factor.
Patients were classified as having acute hepatitis A, recent
hepatitis A, or resolved hepatitis A by the combination of
peak aminotransferase levels and the change in amino-
transferase levels and S/CO over time. We also divided the
cohort into 2 groups, by the absolute S/CO value (Table 6).
The value of 1.90 was used to divide the group into higher
(n ¼ 26) and lower (n ¼ 31) S/CO because this value was
below that observed for all patients with symptoms, signs,
and laboratory test results consistent with acute hepatitis.
In the final analysis, acute hepatitis A was the diagnosis
for 4 patients with 5 positive test results. Another 8 patients
(12 results) had findings consistent with recent hepatitis A,
while 14 patients (17 results) were classified as having
resolved acute hepatitis A. Of the latter group, 2 were also in
the recent hepatitis A group at an earlier time point. The
final 11 patients (15 results) could not be classified. All had
abnormal aminotransferase levels and 10 of 11 had reasons
for hepatic enzyme elevation other than hepatitis A. The
remaining patient, a 38-year-old homeless man (S/CO,
1.36), had no follow-up data and no alternative diagnosis.
The findings in the 4 patients with acute hepatitis A were
the same as those reported previously from our institution.5
The only difference between the past period and the present
period was the incidence. Between 1997 and 1998, 13
patients were hospitalized with acute hepatitis A, as
compared with 3 patients during the past 4 years.
We found that a positive result for IgM antibody to HAV is
now relatively rare (0.5% of all performed tests). Further-
more, only a minority of patients with a positive result have
a concomitant clinical diagnosis of acute hepatitis A (4 of 35,
11%). The number of assays decreased with the abandon-
ment of paper forms and introduction of physician order
entry. However, even in 2010, more than 2000 tests were
ordered. With operationalization of clinical decision support
systems, decreasing inappropriate test orders may be
possible. One intriguing unanswered question is whether
the persistence of IgM antibody correlates with persistence
of antigen in a clinically dormant state. While acute
symptomatic hepatitis A is now a much rarer disease than
in the past, it remains important to make this diagnosis,
which has significant public health implications.
Acute hepatitis A infection is diagnosed by the combina-
tion of an appropriate clinical presentation and laboratory
testing with IgM anti-HAV.1,4,5,8,9Possible explanations for
positive results without clinical acute hepatitis include
asymptomatic infection, previous infection with persistent
IgM antibodies, cross-reacting antibodies, or commercial
Table 2. Demographic, Laboratory, and Assay Data for 37 Positive Results
for Hepatitis A Virus (HAV) Immunoglobulin M
Acute HAV (n ¼ 4)
Recent HAV (n ¼ 8a)
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; S/CO, signal/cutoff value.
aTwo subjects appear twice, in both recent HAV and resolved HAV infection.
92 Arch Pathol Lab Med—Vol 137, January 2013IgM Anti-HAV Testing in the 21st Century––Alatoom et al
kits with a falsely low cutoff value. We found evidence
supporting these scenarios.
Clinical Acute Hepatitis
The classic IgM-specific assay acquires diagnostic sensi-
tivity for acute hepatitis A with substantial dilution of the
original serum or plasma sample.3Thus, the HAVAB-M
(Abbott Laboratories, Abbott Park, Illinois) radioimmuno-
assay introduced in 1981 diluted the sample 1:4000.3The
current assays use a similar IgM antibody capture system,
but detection of the antibody is made by enzyme-linked
chemiluminescence or another nonradioactive method.
Regardless, sample dilution is critical for diagnosis of
patients with the symptoms and signs of acute hepatitis.3
Occasionally, dilution may result in an initially negative test
result that becomes positive days or weeks later.10–12
The balance between sensitivity (correctly identifying all
patients with acute hepatitis A) and specificity (correctly
excluding all those without acute hepatitis A) is chosen to
allow optimum assay performance in a defined clinical
setting, acute hepatitis. Owing to the low levels of IgM
antibodies after vaccination, commercially available tests
rarely detect these antibodies1,9unless the sample dilution is
IgG antibodies to HAV appear soon after IgM, persist for
years after infection, confer lifelong immunity,1,5,9and are
not measured in the IgM-capture technique.3In the United
States, IgG-specific HAV antibody is not measured sepa-
rately; rather, all classes (IgM, IgG, and IgA) are measured
in the total HAV antibody assays. The sample is not
similarly diluted, thereby greatly increasing the sensitivity of
the assay. Over time, the IgM component decreases but
current clinical assays do not indicate the proportion of IgM
in the total. Research studies14,15have shown the potential
for differentiating prior infection by assessing IgG avidity,
but such assays are not commercially available.
Other Hepatitis A
Our finding of positive results for IgM-specific HAV
antibody in asymptomatic infection and recent infection is
expected. IgM antibodies will be produced regardless of the
disease severity. The apparent persistence of IgM-specific
HAV antibody in a small number of patients after the
resolution of all clinical evidence of hepatitis is well
documented.16,17The antibody levels in such patients are
generally low, similar to some of the findings reported
In 2005, the Centers for Disease Control and Prevention
(CDC) reported an increase in the number of persons ‘‘with
positive serologic tests for acute hepatitis A virus infection
whose illness was not consistent with the clinical criteria of
the hepatitis A case definition.’’8Data from state public
health officials in Connecticut (19 not consistent of 127 ¼
15%) and Alaska (10 not consistent of 37¼27%), as well as
Resolved HAV (n ¼ 14a)
Indeterminate (n ¼ 11)
MedianMean RangeMean Range
for Patients With 49 Positive Results
for Hepatitis A Virus (HAV) Immunoglobulin M (IgM)
Encounter and Order Information
Acute hepatitis serology
tests (¼ group)
HAV IgM and total
Other standard tests
Symptoms and signsb
Exposure to HAV
(n ¼ 49)
Reason for test
aOther standard tests: chronic hepatitis serology tests in 4 instances
and hepatitis B core IgM test in 1 instance. Nonstandard tests cannot
be selected and were interpreted by the entering user from the hand-
written order (‘‘Hepatitis A core IgM,’’ ‘‘Hepatitis labs’’). For 1 patient,
there was no provider order in the record; the clerk ordered HAV IgM
bJaundice (n ¼ 4), abdominal pain (n ¼ 2), other gastrointestinal
symptoms (n¼1). All 7 patients had a final clinical diagnosis of either
acute hepatitis (aspartate aminotransferase [AST] .500 U/L) or recent
hepatitis A (AST ,500 U/L).
cCirrhosis (n ¼ 2), hepatitis C (n ¼ 2), hepatitis B (n ¼ 1), elevated
hepatic enzymes (n ¼ 12).
dRepeat after previous positive test (n ¼ 13), duplicate order (n ¼ 1).
eProvider or clerical error (n ¼ 5), no note retrieved (n ¼ 2), homeless
outreach (n ¼ 3).
IgM Anti-HAV Testing in the 21st Century––Alatoom et al
Results for Patients With Positive Hepatitis A Virus
Other Serologic and Immunologic Test
Hepatitis A total antibody
Hepatitis B surface antigen
Hepatitis C virus antibody
aOne patient tested negatively for total hepatitis A virus (HAV)
antibody 6 months before the positive immunoglobulin M (IgM)
result; 1 patient tested negatively for total HAV antibody simulta-
neously with a positive IgM result.
bNot confirmed on neutralization.
cNegative recombinant immunoblot assay result; patient also had a
false-positive result for hepatitis B surface antigen.
Arch Pathol Lab Med—Vol. 137, January 201393
the Sentinel Counties Study (87 not consistent of 140 ¼
62%), were quoted. The positive results came from different
licensed IgM anti-HAV tests (Connecticut, n ¼ 3 tests;
Alaska, n ¼ 3 tests from 2 manufacturers). An editorial8
concluded that IgM anti-HAV testing should be restricted to
‘‘persons with clinical hepatitis,’’ not those with abnormal
liver enzyme levels. The Kentucky Department for Public
Health also had similar findings during a 3-year period.7
There were 156 diagnostic false-positive results reported
(156 of 269 ¼ 58%) in the absence of acute hepatitis,
compared with 113 confirmed cases.
Our data provide not only a numerator for IgM anti-HAV
positive results but also a unique denominator, all ordered
tests. In addition, the reason for ordering the test was
retrievable from the electronic health record, data not
previously obtainable in the public health studies. Only
when IgM anti-HAV tests were ordered for patients with
symptoms or signs suggesting acute hepatitis was the final
analysis acute hepatitis A. The percentage of patients with
acute disease (15%) is lower than that in the published
literature. The likely explanation is that the previous data
were pulled from public health notifications, not only from
Whether the persistence of IgM-specific antibody is
associated with persistent antigen stimulation is unclear.
In the report from the CDC,8the testing of IgM HAV
antibody–positive specimens for HAV RNA yielded only 1
positive result from 25 persons with no symptoms of acute
hepatitis, compared with 34 positive results for HAV RNA in
51 symptomatic cases.8The single positive result was
associated with a follow-up IgM anti-HAV result that was
negative, implying recent asymptomatic HAV. In hepatitis
B, IgM antibody and hepatitis B virus DNA ‘‘reappear’’ with
activation of dormant virus associated with immune
suppression. As with HAV, the diagnostic accuracy of IgM
testing for acute hepatitis B infection is dependent on
sample dilution. IgM-class antibodies continue to be
produced in chronic hepatitis B infection albeit usually at
lower levels undetectable in the diluted samples.18
False-Positive and Low-Positive Results
Rheumatoid factor can interfere with IgM anti-HAV
testing.19One patient with chronic hepatitis C had a high
titer for rheumatoid factor, which possibly interfered with
the IgM anti-HAV testing by bridging the capture antibody
and the signal antibody, leading to false-positive results.
Chronic hepatitis C in 7 other patients, at least 4 of whom
were viremic, may also be associated with rheumatoid
factor; they were not tested. The presence of false-positive
results from cross-reactivity with other antigenic epitopes in
the assay is another potential explanation, particularly since
false-positive tests for hepatitis B surface antigen and
hepatitis C were encountered for 2 and 3 patients,
respectively. Cross-reactions among viral hepatitis antibod-
ies have been reported previously.20Finally, 1 patient had a
result that was either a false positive or the consequence of a
Immunoglobulin M Divided by Signal/Cutoff (S/CO) Value and Liver Enzymes
Final Analysis for 49 Positive Results With Hepatitis A Virus (HAV)
AST and/or ALT, U/L
. 500 Elevated (, 500)Normal
Higher S/CO (?1.90)a
Acute hepatitis A
Lower S/CO (,1.90)a
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase.
aThe lowest S/CO value for a patient with aminotransferase levels .500 U/L was 1.95. The value 1.90 was chosen to separate the positive results
into 2 nonoverlapping groups.
bOther potential causes of elevated aminotransferase levels: hepatitis C (n¼3), cirrhosis (n¼1). One patient had a total Hepatitis C Virus antibody
test performed simultaneously; the result was negative, indicating a probable false-positive immunoglobulin M (IgM) result.
cRheumatoid factor (n¼ 1 patient, 2 Hepatitis C Virus IgM tests), false-positive Hepatitis B surface antigen test result (n ¼3 patients, 4 Hepatitis C
Virus IgM tests), false-positive Hepatitis C Virus antibody test result (n ¼ 2 patients, 6 Hepatitis C Virus IgM tests).
dOther potential causes of elevated aminotransferase levels: hepatitis C (n¼1), cirrhosis (n¼5), other chronic liver disease (n¼4), insufficient data
(n ¼ 1).
94 Arch Pathol Lab Med—Vol 137, January 2013
Table 5.Levels of Liver Enzymes and Bilirubin and the Reason for Ordering Hepatitis A Virus Immunoglobulin M Test
Bilirubin, mg/dL AST, U/L
Mean – SD
5.4 – 4.5
1.0 – 0.6
4.1 – 6.4
0.4 – 0.1
MedianRangeMean – SD
1527 – 1541
122 – 119
84 – 63
29 – 22
Symptoms and signs (n ¼ 6)
Abnormal enzymes (n ¼ 8)
Chronic liver disease (n ¼ 7)
Unknown (n ¼ 10)
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase.
aFemales: 10–35 U/L and males: 10–50 U/L for both AST and ALT since mid 2009.
IgM Anti-HAV Testing in the 21st Century––Alatoom et al
technical error; this conclusion is based on the simultaneous
negative total HAV antibody and positive IgM antibody
results in a more dilute sample. Simultaneous measurement
of total antibody and IgM can be performed and may prove
to be beneficial in helping physicians correctly interpret
In conclusion, positive results for IgM anti-HAV testing
have multiple etiologies. Previous reports have recommend-
ed that clinicians limit laboratory testing for acute hepatitis
A infection to persons with clinical findings typical of
hepatitis A or to persons who have been exposed to settings
where HAV transmission is suspected.6,8However, an
approach targeting individual providers may not result in
a decrease in the number of positive results for IgM anti-
HAV in the absence of acute hepatitis. In contrast, system
approaches, whereby clinical decision support guides
appropriate testing, may engineer change. Restricting orders
to a subset with acute hepatitis (liver aminotransferase levels
.500 U/L) can also decrease the number of positive results
for HAV IgM. New tests that allow simultaneous measure-
ment of total antibody and the proportion of IgM may be
the most useful addition to the laboratory testing menu.
1. Koff RS. Hepatitis A. Lancet. 1998;351(9116):1643–1649.
2. Centers for Disease Control and Prevention (CDC), Division of Viral
Hepatitis. Hepatitis A FAQs for health professionals. http://www.cdc.gov/hepatitis/
HAV/HAVfaq.htm. Accessed September 1, 2011.
3. Decker RH, Kosakowski SM, Vanderbilt AS, Ling CM, Chairez R, Overby
LR. Diagnosis of acute hepatitis A by HAVAB-M, a direct radioimmunoassay for
IgM anti-HAV. Am J Clin Pathol. 1981;76(2):140–147.
4. Lemon SM. Type A viral hepatitis: epidemiology, diagnosis, and
prevention. Clin Chem. 1997;43(8, pt 2):1494–1499.
5. Cuthbert JA. Hepatitis A: old and new. Clin Microbiol Rev. 2001;14(1):38–
6. Castrodale L, Fiore A, Schmidt T. Detection of immunoglobulin M antibody
to hepatitis A virus in Alaska residents without other evidence of hepatitis. Clin
Infect Dis. 2005;41(9):e86–e88.
7. Tarkhashvili N, Thoroughman D, Humbaugh K. Positive hepatitis A IgM test
results among persons without symptoms or risk-factor history—Kentucky, 2002–
2004. Infect Dis Clin Pract. 2007;15(4):245–249.
8. Dembek Z, Hadler J, Castrodale L, et al. Positive test results for acute
hepatitis A virus infection among persons with no recent history of acute
hepatitis—United States, 2002–2004. MMWR Morb Mortal Wkly Rep. 2005;
9. Nainan OV, Xia G, Vaughan G, Margolis HS. Diagnosis of hepatitis A virus
infection: a molecular approach. Clin Microbiol Rev. 2006;19(1):63–79.
10. Shin HP, Lee JI, Jung SW, Cha JM, Joo KR, Kang SY. Factors for predicting
positive results for anti-HAV IgM retesting among initially seronegative patients.
Dig Dis Sci. 2010;55(12):3537–3540.
11. Petrov P, Nikolova A, Komitova R, et al. Acute hepatitis A infection with
negative anti-HAV IgM. In: Global Hepatitis A Meeting Program. 2007:137.
12. Chakvetadze C, Mallet V, Gaussec L, Hannoun L, Pol S. Acute hepatitis A
virus infection without IgM antibodies to hepatitis Avirus. Ann Intern Med. 2011;
13. Miller WJ, Clark W, Hurni W, Kuter B, Schofield T, Nalin D. Sensitive
assays for hepatitis A antibodies. J Med Virol. 1993;41(3):201–204.
14. Roque-Afonso AM, Grangeot-Keros L, Roquebert B, et al. Diagnostic
relevance of immunoglobulin G avidity for hepatitis A virus. J Clin Microbiol.
15. Syhavong B, Rasachack B, Smythe L, et al. The infective causes of hepatitis
and jaundice amongst hospitalised patients in Vientiane, Laos. Trans R Soc Trop
Med Hyg. 2010;104(7):475–483.
16. Liaw YF, Yang CY, Chu CM, Huang MJ. Appearance and persistence of
hepatitis A IgM antibody in acute clinical hepatitis A observed in an outbreak.
17. Kao HW, Ashcavai M, Redeker AG. The persistence of hepatitis A IgM
antibody after acute clinical hepatitis A. Hepatology. 1984;4(5):933–936.
18. Hoofnagle JH. Reactivation of hepatitis B. Hepatology. 2009;49(suppl 5):
19. Sacher RA, Peters SM, Bryan JA. Testing for viral hepatitis: a practice
parameter. Am J Clin Pathol. 2000;113(1):12–17.
20. Herremans M, Bakker J, Duizer E, et al. Use of serological assays for
diagnosis of hepatitis E virus genotype 1 and 3 infections in a setting of low
endemicity. Clin Vaccine Immunol. 2007;14(5):562–568.
Mean – SD
2216 – 2142
119 – 75
72 – 48
36 – 52
Arch Pathol Lab Med—Vol. 137, January 2013IgM Anti-HAV Testing in the 21st Century––Alatoom et al95
Copyright of Archives of Pathology & Laboratory Medicine is the property of College of American
Pathologists and its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email articles for