Efficacy and cost-effectiveness of the 13C-urea breath test as the primary diagnostic investigation for the detection of Helicobacter pylori infection compared to invasive and non-invasive diagnostic tests.

Background: Helicobacter pylori (H. pylori) is one of the most common bacterial infections in humans. There is a risk factor for gastric or duodenal ulcers, gastric cancer and MALT (Mucosa Associated Lymphoid Tissue)-Lymphomas. There are several invasive and non-invasive methods available for the diagnosis of H. pylori. The (13)C-urea breath test is a non-invasive method recommended for monitoring H. pylori eradication therapy. However, this test is not yet used for primary assessment of H. pylori in Germany.

Objectives: What are the clinical and health economic benefits of the (13)C-urea breath test in the primary assessment of H. pylori compared to other invasive and non-invasive methods?

Methods: A systematic literature search including a hand search was performed for studies investigating test criteria and cost-effectiveness of the (13)C-urea breath test in comparison to other methods used in the primary assessment of H. pylori. Only studies that directly compared the (13)C-urea breath test to other H. pylori-tests were included. For the medical part, biopsy-based tests were used as the gold standard.

Results: 30 medical studies are included. Compared to the immunoglobulin G (IgG) test, the sensitivity of the (13)C-urea breath test is higher in twelve studies, lower in six studies and one study reports no differences. The specificity is higher in 13 studies, lower in three studies and two studies report no differences. Compared to the stool antigen test, the sensitivity of the (13)C-urea breath test is higher in nine studies, lower in three studies and one study reports no difference. The specificity is higher in nine studies, lower in two studies and two studies report no differences. Compared to the urease test, the sensitivity of the (13)C-urea breath test is higher in four studies, lower in three studies and four studies report no differences. The specificity is higher in five studies, lower in five studies and one study reports no difference. Compared to histology, the sensitivity of the (13)C-urea breath test is higher in one study and lower in two studies. The specificity is higher in two studies and lower in one study. One study each compares the (13)C-urea breath test to the (14)C-urea breath test and the polymerase chain reaction (PCR) test, respectively, and reports no difference in sensitivity and specificity with the (14)C-urea breath test, and lower sensitivity and higher specificity compared to PCR. The statistical significance of these differences is described for six of the 30 studies. Nine health economic evaluations are included in the Health Technology Assessment (HTA) report. Among these studies, the test-and-treat strategy using the (13)C-urea breath test is compared to test-and-treat using serology in six analyses and to test and treat using the stool antigen test in three analyses. Thereby, test-and-treat using the breath test is shown to be cost-effective over the serology based strategy in three models and is dominated by a test-and-treat strategy using the stool antigen test in one model. A cost-effectiveness comparison between the urea breath test approach and the empirical antisecretory therapy is carried out in four studies. Of these, two studies report that the strategy using the urea breath test is cost-effective over the empirical antisecretory therapy. In two studies, test-and-treat using the (13)C-urea breath test is compared to the empirical eradication therapy and in five studies to endoscopy-based strategies. The breath test approach dominates endoscopy in two studies and is dominated by this strategy in one study.

Discussion: All included medical and economic studies are limited to a greater or lesser extent. Additionally, the results of the studies are heterogeneous regarding medical and economic outcomes respectively. Thus, the majority of the medical studies do not report the statistical significance of the differences in sensitivity and specificity. In direct comparisons the (13)C- urea breath test shows higher sensitivity and specificity than the IgG and stool antigen tests. In comparison to the urease test, results for sensitivity are inconsistent, and the specificity is slightly higher for the (13)C-urea breath test. There are not enough results for comparisons between the (13)C-urea breath test and the (14)C-urea breath test, histology and PCR to describe tendencies. The included economic studies suggest that the test-and-treat strategy using the (13)C-urea breath test is cost-effective compared to test-and-treat using serology as well as empirical antisecretory therapies. Due to a lack of valid studies, it is not possible to assess the breath test approach in comparison to test-and-treat using the stool antigen test and the empirical eradication therapy respectively, regarding the cost-effectiveness. The results of economic analyses comparing test-and-treat using the breath test to endoscopy strategies are too heterogeneous to draw any conclusions. Overall, none of the included economic models is able to completely capture the complexity of managing patients with dyspeptic complaints.

Conclusions/recommendations: Based on available medical and economic studies, there is no sufficient evidence to recommend test and-treat using (13)C-urea breath testing for the detection of H. pylori infection as the standard procedure for the management of uninvestigated dyspepsia in the German health care system. In addition, it must be considered that the DVGS guidelines of the Deutsche Gesellschaft für Verdauungs- und Stoffwechselkrankheiten (DVGS) recommend endoscopy based methods for the management of patients with dyspeptic complaints.