Secondary Logo

Journal Logo

RAPID BLOOD TEST (BM® TEST) IS UNRELIABLE TO DETECT H. PYLORI INFECTION IN CHILDREN WITH ABDOMINAL PAIN

Kindermann, A.; Faus-Keler, T.ß; Ballauff, A.; Findeisen, A.; Laske, G.; Koletzko, S.

Journal of Pediatric Gastroenterology & Nutrition: May 1999 - Volume 28 - Issue 5 - p 560
E S P G H A N 32nd Annual Meeting; Warsaw, Poland, June 2-5, 1999
Free

Study center: Kinderklinik, Ludwig-Maximilians-University, Munich, Germany

    Abstract 65

    Several rapid tests for detection of antibodies against H. pylori (HP) in whole blood or serum are offered for clinical use. We evaluated the BM®-test (Boehringer Mannheim, Germany, identical to Helisal Rapid Blood test) in children presenting with abdominal pain. Methods: BM®-test and 13C-urea breath test (UBT) were performed on the same day in 195 children (4 - 18 y) from 4 centres. A detailed history was obtained on a structured questionnaire. Upper endoscopy was performed in all patients with a positive UBT, and in UBT negative children with suspected organic disease. A child was considered infected (HP+) when at least two applied methods (UBT, histology, urease test) or culture were positive, and considered non-infected (HP-) with concordant negative tests. Results: Eight children with previous therapy for HP-infection and one patient with undetermined HP-status were excluded from analysis. Of the remaining 186 children, 61 (33 %) were HP+ and 125 HP-. Serological testing in relation to HP-status revealed a sensitivity of 54 %, specificity of 90 %, positive predictive value of 72 %, and negative predictive value of 80 % (table). Within the group of HP+ children, the 33 patients with correct positive test results were not significantly different from the 28 children with false negative BM®-tests regarding age (median: 11.2 vs. 11.8 y) or δ-over-baseline values in the UBT (20.6 ‰ vs. 16.4 ‰). However, they had a longer history of abdominal pain (12 vs. 4.5 months in false negatives, p = 0.0112). The accuracy of the BM®-test was independent of ethnicity, sex, family history for ulcer disease or HP infection, and type of blood sampling (capillary vs. venous). In the 8 previously treated children (4/8 were still infected) the rapid blood test gave one false positive and 3 false negative results. Conclusions: Almost half of HP infected children and 10 % of non-infected children are misdiagnosed by the BM®-test. This poor performance makes this test unsuitable for paediatric use. False negative test results are not related to young age or a certain ethnic group.

    Table

    Table

    Section Description

    GASTROENTEROLOGY

    © 1999 Lippincott Williams & Wilkins, Inc.