A comparison of three fingerstick, whole blood antibody tests for Helicobacter pylori infection: A United States, multicenter trial

OBJECTIVE: We compared three whole blood antibody tests for Helicobacter py lori (H. pylori) in a United States, multicenter trial. METHODS: Patients referred for EGD at three medical centers were recruited. During EGD, biopsies were taken for histology and rapid urease testing (RU T). Immediately after endoscopy, patients underwent the antibody tests (Fle xPack HP, Abbott Diagnostics; QuikVue, Quidel Corporation; AccuMeter, ChemT rak) using whole blood obtained by two to three fingersticks. Performance c haracteristics were calculated for each antibody test using the biopsy-base d methods as a gold standard. RESULTS: A total of 131 patients participated; 50 (38%) patients had histol ogical evidence of H. pylori infection. Using histology as a gold standard, the sensitivities of FlexPack HP, QuikVue, and Accumeter were 76%, 78%, an d 84%, respectively. Specificity was 79% with FlexPack HP and 90% with Quik Vue and Accumeter. There were no significant differences in the performance of the three antibody tests though there was a trend toward superior perfo rmance for AccuMeter compared to FlexPack HP (p = 0.019). However, RUT prov ed superior to FlexPack HP using histology as a gold standard (p = 0.008). Using either concordant histology and RUT results or a positive histology o r RUT to define active H. pylori infection, there was no statistically sign ificant difference between the antibody tests. CONCLUSIONS: There were no statistically significant differences in the per formance of the three antibody tests. These tests proved only marginally se nsitive in detecting patients infected with H. pylori. Clinicians should be aware of the limitations of these tests, particularly when using them as a sole means of testing for H. pylori. (Am J Gastroenterol 1999;94:1512-1516 . (C) 1999 by Am. Coll. of Gastroenterology).