4.1 Breath tests

Since their introduction, both ¹³C and ¹⁴C urea breath tests (CUBTs) have been used widely in patients both before and after H. pylori eradication therapy.⁵⁵ An RCT has shown that, compared to an accepted gold standard, accuracy was 94.8% before antimicrobial therapy and 95.4% afterwards.⁵⁶ False negative results may occur and CUBT should not be performed within two weeks of proton pump inhibitor (PPI) or within four weeks of antibiotic therapy.⁵⁷ Evidence level 1++

4.2 Serological tests

4.2.1 HOSPITAL BASED TESTS

Enzyme-linked immunosorbent assays (ELISAs) are the most commonly used serological method for the detection of H. pylori infection, and a large number of commercial ELISA kits are now available. Evaluation of test kits is usually made by using an appropriate “gold standard” consisting of a combination of culture, histology, CLO-test and breath testing as no technique is 100% sensitive or specific for detecting H. pylori infection. Studies of the sensitivity and specificity of these tests have produced inconsistent results. One meta-analysis of 21 studies showed sensitivity and specificity of 85% and 79% respectively and concluded that there is little evidence to suggest that any of the common commercial serological kits is more accurate than another and that the overall accuracy of these kits may not be adequate for clinical decision-making in all patient groups. ⁵⁸ Evidence level 1+

The Medical Devices Agency also evaluated a large number of ELISA tests.⁵⁹ 588 samples of sera were evaluated with 16 different tests. The overall accuracy of the assays averaged 78% (range 68-82%) for all sera. A further meta-analysis gave broadly similar results but reached no conclusions as to whether the use of kits was appropriate.⁶⁰ Evidence level 1+

A large review, using different methodology, contradicted these results. In a comparison of 36 different commercially available H. pylori serology kits on 26,812 patients the median sensitivity and specificity were 92% and 83%.⁶¹ Kits measuring IgA, IgG, and IgM simultaneously or IgA alone did not perform as well as those measuring only IgG antibodies. This review concluded that the overall performance of commercially available serology kits is sufficiently accurate as a means of diagnosing H. pylori infection. Evidence level 2+

4.2.2 NEAR PATIENT TESTS

Near patient tests are whole blood tests developed to provide a rapid diagnosis of H. pylori infection in the clinic. They are technically simple to perform, most of those currently used are one-step tests that require a drop of whole blood, but others require separation of serum. Recent evidence accumulated from 3,805 patients in eight studies suggests that these tests have considerably lower sensitivity and specificity than originally assumed.⁶² The mean sensitivity (weighted for number of patients studied) was 71.1% and the specificity was 87.6%. Studies of the accuracy of several new rapid whole blood test kits have shown a sensitivity and specificity of 82-95% and 83-94%, respectively, with positive and negative predictive values of 89-91% and 93-97%.^{63, 64} Evidence level 2+

4.3 Faecal antigen tests

Enzymatic immunoassays that detect the H. pylori antigen in stool specimens are available. A large number of patients have been evaluated using the stool antigen test for the initial diagnosis of H. pylori infection and the mean sensitivity and specificity has been calculated at 93.1% and 92.8% respectively.⁶² Caution is needed following eradication therapy as omeprazole significantly reduces faecal antigen values, resulting in a decreased accuracy.^{65, 66} Evidence level 2+

4.4 Choosing a test

4.4.1 PRE-TREATMENT DIAGNOSIS

In the primary care setting, the ELISA (hospital based serology) and CUBT have been shown to be equally effective in excluding the diagnosis of infection with H. pylori in patients who have had dyspeptic symptoms for at least two weeks.⁶⁷ Whole blood tests produced inferior results.⁶⁷ A multicentre study suggests that the faecal antigen test is comparable with the CUBT in the initial detection of H. pylori infection.⁶⁸ Evidence level 2++

In trials that compare the use of endoscopy with H. pylori test and treat, some have used CUBT and others have used serology testing.^{43, 45} Outcomes from both types of test have been similar. In clinical practice therefore, the difference in accuracy between the two tests may not be significant.^{43, 45} Evidence level 2++

4.4.2 POST-ERADICATION TESTING

CUBT has been shown to be effective in detecting eradication of H. pylori infection. Seven studies using CUBT alone as a comparison found the sensitivity and specificity of both faecal antigen testing and CUBT was approximately 90%. There has however been some variability in the results of the faecal antigen test in the post-therapy setting so at present faecal antigen testing is not comparable to CUBT in the post-therapy setting.⁶⁸ Evidence level 2++

Serology may be used for monitoring the efficacy of H. pylori eradication therapy but both pre- and post-treatment samples must be analysed simultaneously and it is necessary to wait a minimum of three months following eradication therapy before repeating the test.⁶⁹ Evidence level 2++

The CUBT or faecal antigen tests are recommended for the pre-treatment diagnosis of H. pylori infection in the community. Less accurate, hospital-based serology tests have a place within the non-invasive test and treat strategy.

CUBT is the recommended test to determine whether H. pylori has been successfully eradicated.

The CUBT should not be performed within two weeks of proton pump inhibitor therapy or within four weeks of antibiotic therapy as false negative results may occur.