Current issues of ACP Journal Club are published in Annals of Internal Medicine


Diagnosis

1 of 4 commercially available kits was more sensitive but less specific than biopsy for diagnosing Helicobacter pylori

ACP J Club. 1993 Sept-Oct;119:49. doi:10.7326/ACPJC-1993-119-2-049


Source Citation

Taha AS, Reid J, Boothmann P, et al. Serological diagnosis of Helicobacter pylori-evaluation of four tests in the presence or absence of non-steroidal anti-inflammatory drugs. Gut. 1993 Apr;34:461-5.


Abstract

Objective

To compare 4 commercially available test kits for the serologic diagnosis of Helicobacter pylori with or without nonsteroidal anti-inflammatory drug (NSAID) use.

Design

Blinded comparison of 4 commercial serologic test kits with standard biopsy-related tests.

Setting

Outpatient clinics of a hospital in Scotland.

Patients

64 patients (46 women) from a rheumatoid arthritis outpatient clinic who had been taking NSAIDs for ≥ 4 weeks (NSAID group) and 60 patients (41 women) from a gastroenterology outpatient clinic who had abdominal complaints and had not been taking steroids, cytotoxic or antiulcer drugs, antibiotics, or had not had previous gastroduodenal surgery (non-NSAID group).

Description of tests and diagnostic standard

The Pyloriset Latex test detects IgG antibodies by agglutination using latex particles coated with acid-extracted antigen of H. pylori. The Helico-G test uses ELISA to measure IgG antibodies directed against H. pylori cell-membrane-derived antigen. Biolab Malakit uses ELISA to detect IgG antibodies against H. pylori urease antigen. The Bio-Rad GAP Test IgG uses ELISA to detect IgG antibodies against H. pylori outer-membrane antigen. The gold standard was ≥ 2 positive results of the biopsy-related tests (histology, culture, and Camplybacter-like organism [CLO] test [placing the antral biopsy specimen into a gel pellet and reviewing the slide for red color at 3 and 24 hours]). {All tests were assessed blind to other results.}*

Main outcome measures

Sensitivity and specificity for all 4 commercial tests for both the NSAID and non-NSAID groups.

Main results

Prevalence of H. pylori was 53% in the NSAID group and 72% in the non-NSAID group (P < 0.05). The sensitivity for the tests with a NSAID (or without a NSAID) were Pyloriset Latex, 59% (60%); Helico-G, 79% (74%); Biolab Malakit, 85% (81%); and Bio-Rad GAP Test IgG, 100% (95%). All tests except the latter were less than the gold standard (P < 0.05). The corresponding specificities were 50% (71%), 47% (59%), 50% (65%), and 30% (29%), with all tests less than the gold standard (P < 0.05).

Conclusions

All commercial serologic tests had lower sensitivities (except the Bio-Rad GAP Test IgG) and specificities than did standard biopsy-related tests for diagnosing Helicobacter pylori. The Bio-Rad GAP Test IgG had the highest sensitivity and lowest specificity.

Sources of funding: In part, Orion Diagnostica; Porton Cambridge; Biolab; Bio-Rad (diagnostic kits).

For article reprint: Dr. A.S. Taha, Crosshouse Hospital, Kilmarnock, Ayreshire KA2 OBE, Scotland, UK. FAX 44-1563-577972.

*Information supplied by author.


Commentary

This study by Taha and colleagues compared 4 commercially available serologic tests with antral biopsy for culture and CLO test results and histologic findings for the detection of H. pylori.

Noninvasive detection methods to avoid endoscopy have been proposed to screen persons for H. pylori who are at risk for peptic ulceration.

The gold standard for the diagnosis of H. pylori required positive results from 2 biopsy-related tests. Biopsies of the gastric body were not obtained nor were other noninvasive tests, such as urea breath testing, done. These limitations may explain the observed low specificities of serologic methods compared with those found in other studies (1, 2). Serologic methods were also insensitive compared with biopsy-based techniques.

Caution is advised in applying these results to clinical practice, because this study reports decreased reliability of serologic techniques compared with previous studies (1, 2). Further studies of the sensitivity and specificity of commercial serologic tests compared with other invasive and noninvasive detection methods are required. Serologic findings only indicate present or past infection with the organism, and, given the large number of persons who are infected but have no apparent disease, serologic methods alone are inadequate to diagnose peptic ulcer disease.

James Schieman, MD
University of MichiganAnn Arbor, Michigan, USA


References

1. Evans DJ Jr, Evans DG, Graham DY, Klein PD. A sensitive and specific serologic test for detection of Campylobacter pylori infection. Gastroenterology. 1992;96:1004-8.

2. Perez-Perez GI, Dworkin BM, Chodos JE, Blaser MJ.Campylobacter pylori antibodies in humans. Ann Intern Med. 1988;109:11-7.