Evaluation of a locally produced rapid urease test for the diagnosis of Helicobacter pylori infection

Abstract

Background. The rapid urease test (RUT) is used at Groote Schuur Hospital for diagnosing Helicobacter pylori infection. This is an in-house method, which has not been validated.

Objective. To validate our practice of reading the RUT immediately after endoscopy (RUT0), by comparing this with a reading at 24 hours (RUT24) and with histological analysis.

Design. Ninety consecutive patients undergoing upper endoscopy over a 6-week period from October 2005 to November 2005, and in whom rapid urease testing was indicated, were included in the study. Patients with recent exposure (within 2 weeks of endoscopy) to proton pump inhibitors (PPIs), histamine receptor antagonists (H2RAs) and antibiotics (confounders) were noted and included in the cohort. Two antral and two body biopsies were taken for histological examination and a third antral biopsy was placed in the RUT bottle. Both haematoxylin and eosin and modified Giemsa staining methods were used to identify H. pylori. The RUT was read immediately (within 5 minutes of upper endoscopy) (RUT0), as per our current practice, and each specimen was re-read at 24 hours (RUT24). Sensitivity, specificity, positive and negative predictive values and the impact of confounders were calculated.

Results. Of the 90 patients undergoing rapid urease testing, 39% were male and 61% were female, with a mean age of 55 years (range 22 - 79 years). Histological examination revealed H. pylori in 67.8% (N=61) of the biopsy specimens. In the 65 patients without confounders, the sensitivity and specificity of the RUT0 were 65.9% and 100%, and 90.9% and 100% for RUT24. After including the 25 patients with confounders, the sensitivity and specificity were 68.8% and 100% for RUT0, and 90.1% and 100% for RUT24 respectively. Thirteen RUT0 specimens (30.9%) that were initially negative became positive at the RUT24 reading. There were 6 (9.8%) RUT0- and RUT24-negative but histology-positive specimens. Four of these 6 false-negative RUT24 results could be accounted for by a low H. pylori density on histological analysis (2 patients were taking PPIs). Confounders did not alter the sensitivity and specificity outcomes or impact on the number of false-negative RUTs.

Conclusions. Our locally prepared RUT is a specific test for the detection of H. pylori infection. The sensitivity is greatly enhanced by reading the test at 24 hours. The use of PPIs, H2RAs and antibiotics preceding endoscopy did not impact significantly on the results.