In-practice evaluation of whole-blood Helicobacter pylori test: its usefulness in detecting peptic ulcer disease.

Background: Approximately 10% of patients presenting with dyspepsia to the general practitioner have peptic ulcers; the large majority of which are related to infection with Helicobactor pylori. Office-based tests for H. pylori detection are generally validated and evaluated in selected patient groups.

Aim: To evaluate the clinical effectiveness of a whole-blood serology test for infection with Helicobacter pylori in detecting peptic ulcer disease (PUD) in daily general practice.

Method: A descriptive study of 171 primary care dyspepsia patients selected for open-access endoscopy in primary care and aged between 18 and 75 years, in 92 general practices in central, southern, and eastern parts of the Netherlands. H. pylori status was assessed using the BM-test Helicobacter pylori, which is identical to the Helisal test. Dyspepsia severity score was measured using a validated symptom score. Symptom characteristics and probability of relevant disease were assessed by the general practitioner. Endoscopy was carried out in local hospitals. Diagnostic outcome of both endoscopy and H. pylori reference test was supplied by local specialists. The BM-test was evaluated against endoscopic results.

Results: A high number (61.8%) of false-negative BM-tests resulted in a low sensitivity (95% confidence interval [CI] = 48-75%) for detection of H. pylori infection. Only 12 out of 32 patients with PUD had a positive BM-test, resulting in a positive likelihood ratio (LR) for PUD of 1.41 and a negative LR of 0.85.

Conclusions: This study confirms the relatively poor performance of the BM-test in daily general practice, and shows the limited diagnostic value of H. pylori office-tests for detecting PUD in primary care. The discriminative value of the test result is too small to support either a 'test-and-endoscope' of a 'test-and-treat' strategy in general practice.