ROLE OF HELICOBACTER-PYLORI SEROLOGY IN SCREENING PRIOR TO ENDOSCOPY

Objective: To assess, prospectively, the value of a commercially available serological test for Helicobacter pylori in screening prior to endoscopy, and the effect of age and ethnic group on its performance. Design: A policy of screening by endoscopy was assessed prospectively in the direct access population. Subjects under the age of 45 years who were not taking non-steroidal anti-inflammatory drugs (NSAIDs) or who had negative serology were excluded. The most appropriate cut-off point was first determined based on the performance of the test in a clinic population, with validation against histology and biopsy urease (CLO) test. The serology used was an acid glycine extract of H. pylori. Setting: Norman Tanner Gastroenterology Unit, St George's Hospital, London, UK. Patients: Two hundred and ninety-five consecutive patients referred from clinic for endoscopy for dyspepsia and 367 consecutive patients directly referred by their general practitioner for dyspepsia. Main outcome measures: The number of patients requiring endoscopy was reduced and the percentage of pathology undetected was assessed, in the direct access population, using the cut-off point determined in the clinic population. Results: A cut-off point of 6.3 U/ml was selected as the most appropriate for screening. This resulted in a sensitivity and specificity of 98 and 75%, respectively, in the 295 clinic patients. The cut-off point at this level would have missed only one duodenal ulcer out of 22 peptic ulcers in the patients aged 45 years or under, and would have reduced the number of endoscopies in this age group by 42%. In the combined direct referral and clinic-referred populations, only two out of 38 peptic ulcers would have been missed and the number of endoscopies would then have been reduced by 41%. The performance of the test tended to improve with age in the total group of 662 patients (295 clinic and 367 direct access), In ethnic minorities, the test had a similar sensitivity but specificity was significantly reduced. Conclusion: H. pylori serology is valuable as a screening method prior to endoscopy. Age and ethnic background of patients needs to be taken into account when using serology.