Reliability of two commercial serological kits for serodiagnosing Helicobacter pylori infection.

ABSTRACT

Two commercial serological kits, Pylori-set (Orion Diagnostica, Finland) and Helico-G (Cambridge Biomedical Ltd, UK), and an in-house ELISA were evaluated with sera from 24 Helicobacter pylori-positive and 146 H. pylori-negative dyspeptic patients. Sensitivity, specificity, positive and negative predictive values of Pylori-set were lower than that of Helico-G and in-house ELISA. Helico-G was more sensitive (91.7%) than in-house ELISA (83.3%) and both had comparable negative predictive values of 98.3% and 97.3%, respectively. However, specificity (97.9%) and positive predictive value (86.9%) of an in-house ELISA were much higher than specificity (80.1%) and positive predictive value (43.1%) of Helico-G. Kappa index of agreement between the three serological tests (Pylori-set, Helico-G or in-house ELISA) and the presence of H. pylori in antral biopsies was very low (k = 0.13; z = 1.9; p > 0.05), moderate (k = 0.49; z = 7.1; p < 0.0001), or substantial (k = 0.82; z = 10.8; p < 0.0001), respectively. Overall, statistical evaluations demonstrated that both commercial kits were not as reliable as the in-house ELISA for serodiagnosing H. pylori infection.