MICROBIOLOGY

M. Brian Fennerty, MD

A Review of Tests for the

Diagnosis of Helicobacter
pylori Infection

Downloaded from https://academic.oup.com/labmed/article/29/9/561/2504085 by guest on 02 June 2024

ABSTRACT The rediscovery and description of
As recently as 20 years ago, Helicobacter pylori was Helicobacter pylori 15 years ago have resulted in a
misidentified as Pseudomonas spp or a species of paradigm shift in the understanding of the pathogenesis and
another genus. It was believed to be merely a con-
treatment of many gastrointestinal diseases. Successful cure
taminant in gastric mucosa samples until Barry
Marshall and Robin Warren "rediscovered" of the infection depends on an accurate diagnosis.
H pylori and described it in 1983. Their work rev- Numerous endoscopic and nonendoscopic H pylori
olutionized the treatment of many gastrointesti- diagnostic tests can detect H pylori infection, and testing
nal diseases and resulted in a paradigm shift in our has moved from the endoscopy unit to the primary care
understanding of the pathogenesis of these dis- physician's office. This article discusses the clinical aspects of
eases. Infection with H pylori has been proven to
these tests and the advantages, disadvantages, and
be an etiologic factor in peptic ulcer disease, his-
tologic chronic gastritis, and gastric cancer.1 This recommendations regarding appropriate use of these tests.
organism is also a possible factor in nonulcer dys-
pepsia, coronary artery disease, and even sudden Endoscopic Tests
infant death syndrome.2,3 During upper intestinal endoscopy, gastric juice From the
The treatment of H pylori infection has or biopsy specimens of gastric mucosa can be eas- Department of
evolved considerably since the initial therapeutic Medicine, Oregon
ily obtained for direct tests of H pylori infection, o
Health Sciences '£
trial reported in 1987. Currently, three or four such as histology, culture, and polymerase chain University, Portland, 3
drug regimens used for 7 to 14 days cure infection reaction (PCR), and for indirect tests, such as Ore. £
E
in 85% to 95% of patients.4 At the same time, urease testing. Each of these diagnostic methods Reprint requests to o
0
treatment is no longer exclusively administered has clinical utility but varies considerably in accu- Dr Fennerty, m/c PV- o
by gastroenterologists but rather is most com- racy, cost, and ease of use. 310, Oregon Health
monly performed by internists, family practition- Sciences University,
3181 SW Sam u
ers, and other primary care providers. Direct Endoscopic Tests Jackson Park Rd,
<fl
Concurrent with this evolution in the therapy for
H pylori infection is an evolution in the diagnos-
Histology
Histologic demonstration of H pylori remains the
Portland, OR 97201-
3098 or e-mail:
O c
[email protected] o
tic approach. Originally, histologic examination diagnostic gold standard to which other tests are
tS
of gastric biopsy specimens (often supplemented compared.5'6 However, the finding of superficial 0
with special stains) was required to demonstrate 10
chronic active gastritis is almost characteristic of
infection with H pylori. Rapid, simple, and reli- H pylori infection, regardless of whether the
able nonendoscopic tests now usually preclude organism is identified.7 In all patients with H
the need to perform endoscopy to document pylori infection, superficial chronic active gastritis
infection with this organism. This article dis- develops to some degree; thus, the absence of
cusses the advantages, disadvantages, and clinical chronic inflammation on mucosal biopsy effec-
use of various H pylori tests. tively excludes infection (sensitivity of histology,
95%). However, similar histologic changes are

Downloaded from https://academic.oup.com/labmed/article-abstract/29/9/561/2504085

by guest SEPTEMBER 1998 VOLUME 29, NUMBER 9 LABORATORY MEDICINE S81
on 28 May 2018
 sometimes noted in patients without infection. inhibitory concentration. Whether agreement
Thus, the sensitivity of the finding of chronic gas- between the E-test and more traditional methods
tritis is high but the specificity is lower (80%).7 of determining antibiotic sensitivity is good
Because of the suboptimal specificity and remains unclear. Furthermore, in vitro sensitivity
lower positive predictive value for the finding of testing does not always predict successful in vivo
chronic active gastritis, some investigators believe therapy with an agent. The Food and Drug
that identification of the organism on histologic Administration is generating guidelines to stan-
examination remains important.6 The sensitivity dardize antimicrobial sensitivity testing of H
of routine H&E stains for organism identification pylori cultures. If culture is not routinely done
is high in expert hands but may be substantially locally, the specimen should be placed in saline or
lower with pathologists who have less experience. Stuart's medium and packed in ice for trans-

Downloaded from https://academic.oup.com/labmed/article/29/9/561/2504085 by guest on 02 June 2024

Reported sensitivities of this finding have been as portation to the reference laboratory. This should
high as 90% to 95%, and specificities of 99% have allow the organism to remain viable for at least 24
been described.5 Other stains have been used to hours after the specimen is collected.
improve detection of H pylori. Warthin-Starry,
Giemsa, and Genta (H&E, Steiner silver, and Polymerase Chain Reaction
Alcian blue combined) stains all increase the sen- Analysis of gastric mucosal biopsy tissue or aspi-
sitivity of histology but add substantial cost and rates of gastric juice by PCR has been described
effort to this diagnostic test.5 Sensitivity of any of for detecting H pylori DNA.10 This is a very sen-
these techniques in detecting H pylori depends on sitive and specific diagnostic method, but it is
the density of the organism in the biopsy speci- expensive, is performed only in reference labora-
men (recent use of antibiotics or proton-pump tories, and is subject to error if not done with
inhibitors can decrease density), skill of the careful attention to technique. Currently, PCR
pathologist, and a technically sufficient stain. The testing for detection of H pylori remains a
yield is maximized by examination of biopsy research tool only.
specimens from both the antrum and fundus.
Indirect Endoscopic Tests: Urease Tests
Culture Urease testing is an indirect method for detecting
H pylori can be cultured from gastric mucosal H pylori in gastric mucosal biopsy specimens.
biopsy specimens using various media, such as H pylori possesses a potent urease; if urease is pre-
chocolate, brucella, and egg yolk-emulsion agars. sent, the organism converts urea to ammonia,
However, the technique is tedious (12 days for a increasing the pH of the medium. If a pH color
negative result, although growth occurring as indicator is added, a subsequent color change indi-
quickly as 3 days is possible) and is subject to rectly documents the presence of H pylori.
errors in storage and handling; as a result, the Three biopsy urease tests are available in the
sensitivity is decreased (60%-90%), while speci- United States. The CLOtest (Delta West, Perth,
ficity remains high (100%).5'8 Clinically, culture Australia) and Hpfast (GI Supply, Camp Hill, Pa)
of gastric mucosal biopsy specimens is useful for consist of urea and a color indicator in an agar
determining resistance to antimicrobial agents gel. PyloriTek (Serim Research, Elkhart, Ind) is a
(metronidazole, clarithromycin, and amoxacillin) urea-impregnated substrate pad; the pH indica-
to indicate appropriate treatment.9 Sensitivity tor is located in a reaction pad. When a biopsy
testing is usually done at reference laboratories, specimen containing H pylori is placed in the agar
and results vary according to the technique used. or on the pad, the generation of ammonia results
Discordant results in sensitivity testing may be in a distinctive red or blue-purple color change,
obtained depending on whether an agar, broth, or depending on the system used.
disk diffusion testing method is performed.5'8 The sensitivity and specificity of the urease
The epsilometer test (E-test [AB Biodisk, Solna, tests are similar (approximately 90% and 95%,
Sweden]) has recendy been introduced; this test is respectively).5 The pad test is quicker (maximum
a "strip" diffusion method that provides a direct sensitivity at 1 hour), whereas the gel test's sensi-
measure of an H pylori strain's1 minimal tivity at 1 hour is approximately 70%. By 4 hours,
however, the sensitivity of all three tests is the
same. Increasing the number of biopsy specimens
or warming the tests shortens the time to positiv-
ity only slightly.5,8,11

Downloaded from https://academic.oup.com/labmed/article-abstract/29/9/561/2504085

by guest5 6 2 LABORATORY MEDICINE VOLUME 29. N U M B E R S SEPTEM BER 1998
on 28 May 2018
 Fable 1. Which Test To Use?

Environment Primary Test

The sensitivity of the urease test depends on Pretreatment diagnosis
the density of H pylori? Therefore, biopsy speci- During endoscopy Rapid urease test Qualitative s e r u m
mens from a site with few or no organisms can or w h o l e b l o o d
result in a false-negative test result. This occurs in Office-based Qualitative s e r u m Urea breath test
the setting of widespread gastric atrophy or gas- or w h o l e b l o o d
tric intestinal metaplasia, in which the organisms
Posttreatment diagnosis
are not found (H pylori resides only in apposition
During endoscopy Rapid urease test Histology
to normal gastric epithelium), or in the setting of
recent use of antibiotics or proton-pump Office-based Urea breath test
inhibitors, both of which can inhibit the organ- (alternate: ELISA*)
ism and lead to a false-negative result. ELISA indicates enzyme-linked immunosorbent assay.

Downloaded from https://academic.oup.com/labmed/article/29/9/561/2504085 by guest on 02 June 2024

The advantages of the urease tests are their * If quantitative ELISA is used as the primary test and serum is stored
simplicity, Clinical Laboratory Improvement for paired analysis.
Amendments of 1988 (CLIA '88) waiver, cost
(approximately $10 per test), specificity, and are cross-reactive antibodies for similar bacteria
rapidity of results. The disadvantage is the subop- (other Helicobacter species and Campylobacter
timal sensitivity in certain clinical situations. jejuni). Recent improvement in the antigens used
(eg, purified proteins and use of multiple strains)
Endoscopic Tests Summary have minimized this variability, but sensitivity
The diagnostic test to be performed at endoscopy and specificity of the various antibody tests
depends on the information desired by the clini- remain problematic.
cian (Table 1). If the presence of H pylori is the Enzyme-linked immunosorbent assay
only data needed, then a biopsy urease test is the (ELISA) detects H pylori IgG antibodies in serum
optimal first choice. Antibody testing of serum or and usually provides a quantitative antibody titer.
blood would be a backup to a negative test result The sensitivity and specificity of ELISA vary
in order to maximize sensitivity (see below). If (approximately 85%-90%).7 These tests are usu-
histologic features are clinically important, then ally done by a reference laboratory, and the delay
biopsy specimens from the antrum and fundus in receiving the result (days to weeks) and the
«
should be obtained. In experienced hands, special expense ($50-$100) make their use difficult to c
o
stains appear to be unnecessary for most cases justify. Although titers usually decrease after cure 5
-
but may have value in selected circumstances. If of H pylori infection, most patients continue to u
'£
antibiotic sensitivity testing is required, culture of demonstrate antibody to H pylori.14'15 z
biopsy specimens is appropriate. E
Office-based antibody testing of serum or E
whole blood is now widely available.16 These tests o
U
Nonendoscopic Testing are usually latex agglutination or solid-phase u
•£
Testing for H pylori by nonendoscopic means ELISAs that produce a characteristic color change \
requires an indirect diagnostic test. Antibodies to when the antigen combines with the antibody. s
o
H pylori can be detected in blood, serum, and They do not provide a titer and thus are qualita- W
saliva, and urea breath and blood tests detect the
organism's urease activity.
tive tests. Qualitative antibody tests were origi-
nally thought to have less sensitivity and
O c
0
specificity than laboratory-based quantitative
Antibody Tests ELISAs, but recent data indicate that most of I
(fl
H pylori induces both a local and a systemic anti- these tests have sensitivity and specificity similar
body response, producing IgM, IgA, and IgG anti- to those of the laboratory-based tests.17'18 Thus,
bodies.12 Antibody testing for IgG appears to be their accuracy, immediacy of results (10 min-
the most sensitive and specific antibody method utes), and cost ($10—$15) make them ideal
for detecting H pylori; testing for IgA or IgM is screening tests for H pylori in a primary care set-
unnecessary.13 The individual magnitude of the ting. This office-based testing method has been
antibody response to infection varies according
to the antigen being tested for. In addition, there

Downloaded from https://academic.oup.com/labmed/article-abstract/29/9/561/2504085

by guest SEPTEMBER 1998 VOLUME 29, NUMBER 9 LABORATORY M E D L I N E
on 28 May 2018
 Fig 1. A, Gastric
brushing sample
showing slightly
cyanophilic curved
organisms of further facilitated by the recent CLIA waivers
Helicobacter pylori. granted for most of these tests.
Papanicolaou, high
Antibodies to H pylori can also be detected in
power. B, Gastric
biopsy sample from saliva. Initial studies indicate that the presence of
the same patient salivary IgG, but not IgA, has reasonable sensitiv-
with gastric foveolar ity and specificity.19'20 This test may have clinical
glands and application in certain situations (eg, pediatrics) if
numerous curved
spiral organisms of
it is as accurate as blood antibody tests. Currently,
H pylori in overlying however, it remains a research tool.
mucus. Giemsa, Antibodies to various other H pylori proteins
medium power. have also been investigated. H pylori strains with

Downloaded from https://academic.oup.com/labmed/article/29/9/561/2504085 by guest on 02 June 2024

Images from
the cagA gene have been implicated as being asso-
Edmonds PR.
Cytologic diagnosis ciated with a higher incidence of ulcer disease and
of Helicobacter gastric cancer. An IgG antibody to the product of
pylori and this gene (cagA protein) can be detected by sero-
associated gastric logic testing and is commercially available.21 The
diseases.
Cytopathology
clinical utility of this finding is questionable.
Check Sample, Cll
96-6. Chicago, III: Urea Breath Tests
ASCP; 1995:79-91. Although antibody testing can be considered a
"whole organism" test, it relies on a surrogate
marker of infection and thus suffers from prob-
Table 2. Tests for Diagnosis of Helicobacter Pylori lems with both sensitivity and specificity. Urea
Test Sensitivity/Specificity (%) Cost ($)T breath tests were derived as a real-time test to
Endoscopic tests document the presence of urease-producing bac-
Histology 95/80-99 50-100 teria in the stomach (ie, H pylori).9 Urea breath
tests are based on the principle that a urease-con-
Special stains 90/99 200*
taining organism releases labeled carbon from
Culture 60-90/100 150 urea and that this labeled carbon can then be
Urease tests 90/95 10
detected in C 0 2 collected from expired air. These
tests are designed to be performed in a physician's
Nonendoscopic tests office in less than 30 minutes.
Antibody tests
Two isotopes of carbon have been extensively
Quantitative ELISA 85-90/85-9C 50-100 studied in urea breath tests.12'22'23 C-13 is a non-
Qualitative serum 90/90 10-15 radioactive isotope detected by mass spectrogra-
ph^ whereas C-14 is radioactive and is detected
Qualitative whole blood 85-90/90 10-15
by a scintillation counter. Both tests have high
Urea breath tests sensitivity and specificity (95% for both). 22-25
C-13 95/95 280 Sensitivity and specificity data for the tests dis-
C-14 cussed in this review is compiled in Table 2. C-14
95/95 60
is inexpensive, and scintillation counters are
ELISA indicates enzyme-linked immunosorbent assay. widely available in most hospitals. In contrast, C-
* Sensitivity and specificity data from Cohen and Laine,5 Megroud, 6 13 is more expensive and requires mass spectrog-
Cutler, et a l 7 1 4 Fabre, et al, 1 0 Laine, et al, 11 Atherton, 12 Evans, et al, 13
Taka, et al, 1 6 Graham, et al, 17 Knigge, et al, 18 Logan, et al, 22 Slomianski, raphy; as a result, the sample usually must be sent
et al, 23 and Hahn M, Faigel DO, Magaret N, et al. Can noninvasive tests to a reference laboratory.
substitute for histology in the diagnosis of Helicobacter pylori (HP)
infection [abstract]. Gastrointest Endosc. 1998;47:A385.
t Costs for histology, special stains, and cultures are based on wholesale
charges to clinical laboratories in the US Northwest. For nonendoscopic
and urease tests, costs are based on charges from national reference
laboratories.
$ Includes cost of routine histology.

Downloaded from https://academic.oup.com/labmed/article-abstract/29/9/561/2504085

by guest5 6 4 LABORATORY MEDICINE VOLUME 29. NUMBER 9 SEPTEMBER 1998
on 28 May 2018
 The C-14 test used in the United States be applicable in clinical situations where obtain-
(PYtest, Ballard Medical Products, Draper, Utah) ing serum would be easier than obtaining a
is a capsule-based test containing 1 u-Ci of C- breath sample (eg, with children, in clinics, and
14—labeled urea. The test originally had to be per- for screening).
formed at a site licensed by the Nuclear
Regulatory Commission (a nuclear medicine Clinical Use of t h e Laboratory Tests
department or laboratory), but the FDA In a patient with previous documentation of gas-
exempted this requirement in February 1998. The trointestinal disease associated with H pylori, a
test can now be performed in an office-based set- simple office-based screening test such as anti-
ting. Although C-14 has a radiologic half-life of body testing with a serum or whole-blood test
5,760 years, the radiation dose derived from this would be most appropriate. In a patient for

Downloaded from https://academic.oup.com/labmed/article/29/9/561/2504085 by guest on 02 June 2024

test is equivalent to the background radiation a whom test results are negative but the clinical sus-
person is exposed to during 24 hours of normal picion remains high, a urea breath test is an
daily activities. The test is performed by having appropriate backup test.
the patient swallow a capsule containing C-14 In patients undergoing endoscopy, urease test-
urea; a breath sample is collected in a balloon or ing of an endoscopically obtained gastric mucosal
vial 10 minutes later. The sample is then mixed biopsy specimen would be the most appropriate
with scintillation fluid and analyzed by a scintil- primary diagnostic strategy. In patients with sus-
lation counter. The test costs approximately $60. pected false-negative urease test results, testing of
The C-13 urea breath test (Meretek UBT, serum or whole blood for the presence of anti-
Nashville, Tenn) is a nonradioactive breath test. body is the most appropriate and cost-effective
This test uses C-13 urea contained in prinactin, a clinical backup strategy.27
clear, colorless solution. After fasting, the patient In patients who have been treated for H pylori
consumes a nutrient meal to slow gastric empty- infection in whom documentation of cure is
ing; a baseline breath sample is then collected. sought, urea breath testing done 4 or more weeks
The solution is then ingested and a 30-minute after conclusion of antimicrobial treatment is the
sample is obtained. These samples are transferred most appropriate strategy. Given the superior
to test tubes for shipping to the reference labora- sensitivity and specificity of this strategy, backup Fig 2. A patient uses
a urea breath test to
tory for mass spectrography. Although C-13 is testing should be unnecessary in patients whose detect Helicobacter c
0
normally present in breath, the slight increase in test results are negative. An alternative strategy in pylori infection. The a
the 30-minute C-13 concentration over the base- patients who have a previous quantitative ELISA test (Pytest, Ballard u
line concentration can be accurately detected by antibody titer is to wait 4 to 6 months and run a Medical Products, i
3
Draper, Utah) uses
mass spectrography; thus, both breath tests are paired serum sample (a pretreatment serum £
carbon-14 to E
similarly accurate. The C-13 test costs approxi- sample and a posttreatment sample obtained 4 to measure the o
mately $280. 6 months after therapy run on the same ELISA bicarbonate given 0
The accuracy of the urea breath tests is batch). A decrease in titer of 20% or more is off by the urea
affected by anything that affects the density of H approximately 90% sensitive for detecting cure produced by
H pylori.
pylori. Thus, recent antibiotic exposure or con-
current use of proton-pump inhibitors may result
in a false-negative test result.24'25 Therefore, it is
recommended that the test be delayed for at least
4 weeks after exposure to antibiotics and for 10
days after use of proton-pump inhibitors.22,25
i
False-negative test results in other situations are
infrequent. False-positive results are also rare and
occur secondary to urea hydrolysis by oral organ-
isms. A urea blood test has recently been
described.26 Similar to the urea breath test, the
blood test is used to detect labeled serum bicar-
bonate after ingestion of a labeled urea substrate.
This test is currently a research test only but may

Downloaded from https://academic.oup.com/labmed/article-abstract/29/9/561/2504085

by guest SEPTEMBER 1998 VOLUME 29. NUMBER 9 LABORATORY MEDICINE 565
on 28 May 2018
 of infection. The expense of this test, the long 11. Laine L, Chun D, Stein C, et al. The influence of size or
number of biopsies on rapid urease test results: a prospective
wait necessary to perform this test, and require- evaluation. Gastrointest Endosc. 1996;43:49-53.
ment of a paired, stored sample make this a cum- 12. Atherton JC. Non-endoscopic tests in the diagnosis of
bersome and little-used method of determining Helicobacter pylori infection. Aliment Pharmcol Ther.
1997;11:11-20.
H pylori cure. The availability of urea breath tests
13. Evans DJ Jr, Evans DG, Graham DY, et al. A sensitive and
at similar cost (C-14) should make this approach specific serologic test for detection of Campylobacter pylori
unnecessary. infection. Gastroenterology. 1989;96:1004-1008.
14. Cutler A, Schubert A, Schubert T. Role of Helicobacter
pylori serology in evaluating treatment success. Dig Dis Sci.
Conclusion 1993;38:2262-2266.
The many diagnostic tests available for H pylori 15. Cutler AF, Prasad VM. Long-term follow-up of Heli-
have different clinical indications and value. For cobacter pylori serology after successful eradication. Am ] Gas-

Downloaded from https://academic.oup.com/labmed/article/29/9/561/2504085 by guest on 02 June 2024

troenterol. 1996;91:85-88.
the most part, pretreatment testing for H pylori is 16. Taha AS, Reid J, Boothmann P, et al. Serologic diagnosis
based on the endoscopic rapid urease test or on of Helicobacter py/on-evaluation of four tests in the presence
office-based serology or whole-blood antibody or absence of non-steroidal anti-inflammatory drugs. Gut.
1993;34:461-465.
testing. Posttreatment testing is primarily done 17. Graham DY, Evans DJ Jr, Peacock J, et al. Comparison of
by a urea breath test. However, extensive clinical rapid serological tests (FlexSure HP and QuickVue) with con-
investigation in the field of testing for H pylori ventional ELISA for detection of Helicobacter pylori infection.
Am J Gastroenterol. 1996;91:942-945.
diagnosis continues, and tests such as salivary 18. Knigge KL, Fennerty MB, Faigel DO, et al. An office-
antibody testing, urea blood tests, and variations based whole-blood test is preferable to a laboratory quantita-
on currently available antibody and urea breath tive serology test in the determination of Helicobacter pylori
infection documented by endoscopic biopsy [abstract]. Gas-
tests will probably become available in the near troenterology. 1998,114:A180
future. As was seen with H pylori therapy, a rapid 19. Patel P, Mendall PP, Khulusi S, et al. Salivary antibodies
evolution in the development of H pylori diag- to Helicobacter pylori: screening dyspeptic patients before
endoscopy. Lancet. 1994;344:511-512.
nostic tests is likely.© 20. Christie JML, McNulty CAM, Shepherd NA, et al. Is
saliva serology useful for the diagnosis of Helicobacter pylori'.
Gut. 1996;39:27-30.
References
21. Graham DY, Genta RM, Graham DP, et al. Serum CagA
1. Fennerty MB. Helicobacter pylori. Arch Intern Med.
antibodies in asymptomatic subjects and patients with peptic
1994;154:721-727.
ulcer: lack of correlation of IgG antibody in patients with pep-
2. Patel P, Mendall MA, Carrington D, et al. Association of
tic ulcer or asymptomatic Helicobacter pylori gastritis. / Clin
Helicobacter pylori and Chlamydia pneumoniae infections with
Pathol. 1996;49:829-832.
coronary heart disease and cardiovascular risk factors. BMJ.
1995;311:985. 22. Logan RPH, Poison RJ, Misiewicz JJ, et al. Simplified sin-
gle sample l3carbon urea breath test for Helicobacter pylori:
3. Pattison CP, Marshall BJ, Young TW, et al. Is Helicobacter
comparison with histology, culture, and ELISA serology. Gut.
pylori the missing link for SIDS? Gastroenterology.
1991;32:1461-1464.
1997;112:A254.
23. Slomianski A, Schubert T, Cutler AF. 13C urea breath test
4. Fennerty MB, DiTomasso J, Morales TG, et al. Helicobac-
to confirm eradication of Helicobacter pylori. Am J Gastroen-
ter pylori: A primer for internal medicine physicians. IM Inter-
terol. 1995;90:224-228.
nal Medicine. 1995; 16:32-50.
24. Chey WD, Spybrook M, Carpenter S, et al. Prolonged
5. Cohen H, Laine L. Endoscopic methods for the diagnosis
effect of omeprazole on the ,4C-urea breath test. Am J Gas-
of Helicobacter pylori. Aliment Pharmacol Ther. 1997;11:3-9.
troenterol. 1996;91:89-91.
6. Megraud F. How should Helicobacter pylori infection be
25. Laine L, Estrada R, Trujillo M, et al. Prospective evaluation
diagnosed? Gastroenterology. 1997;113:S93-S98.
of the effect of proton pump inhibitor therapy on Helicobacter
7. Cutler AF, Havstad S, Ma CK, et al. Accuracy of invasive
pylori in diagnostic testing [abstract]. Gastroenterology. In press.
and noninvasive tests to diagnose Helicobacter pylori infec-
26. Chey WD, Murthy H, Toskes P, et al. Validation of the Bc-
tion. Gastroenterology. 1995;109:136-141.
Urea blood test for H pylori infection: a US, multicenter trial
8. Megraud F. Resistance of Helicobacter pylori to antibiotics.
[abstract]. Gastroenterology. 1998;114:A91.
Aliment Pharmacol Ther. 1997; 11:43-53.
27. Hahn M, Faigel DO, Magaret N, et al. Can noninvasive
9. Fennerty MB. Should we abandon metronidazole-
tests substitute for histology in the diagnosis of Helicobacter
containing Helicobacter pylori treatment regimens? The clini-
pylori (HP) infection [abstract]. Gastrointest Endosc.
cal relevance of metronidazole resistance. Am / Gastroenterol.
1998;47:A385.
1998;93:1-3.
10. Fabre R, Sobhani I, Laurent-Puig P, et al. Polymerase
chain reaction assay for the detection of Helicobacter pylori in
gastric biopsy specimen: comparison with culture, rapid ure-
ase test and histopathological tests. Gut. 1994;35:905-908.

Downloaded from https://academic.oup.com/labmed/article-abstract/29/9/561/2504085

by guest5 6 6 LABORATORY MEDICINE VOLUME 2 9 , NUMBER 9 SEPTEMBER 1998
on 28 May 2018