Accuracy of Rapid Point-of-Care Diagnostic Tests for Hepatitis B Surface Antigen-A Systematic Review and Meta-analysis.

BACKGROUND

Rapid point-of-care tests provide plausible diagnostic strategy for hepatitis B surface antigen (HBsAg) in low resource areas. However, their utility depends upon their overall performance. Our objective was to meta-analyze the diagnostic accuracy of rapid point-of-care tests for HBsAg.

METHODS

Literature search was done with the help of a metasearch engine Mettā, a query interface for retrieving articles from five leading medical databases. Studies that employed rapid point-of-care tests for detection of HBsAg and compared the results with reference test were included. Two reviewers performed quality assessment of the studies and extracted data for estimating test accuracy. Twenty-seven studies were meta-analyzed and stratified by multiple parameters.

RESULTS

Twenty-seven studies had evaluated 49 test brands and generated 76 data points. Sensitivity of individual tests varied widely and were heterogeneous (range 43.5%-99.8%); while specificity estimates were more robust and close to 100% (range 90%-100%). Overall pooled sensitivity, specificity, positive likelihood ratio (LR), negative LR and diagnostic odds ratio for all tests were 97.1% (95% CI, 96.1%-97.9%), 99.9% (CI, 99.8%-100%), 118.4 (CI, 84.7-165.5), 0.032 (CI, 0.023-0.045) and 4094.7 (CI, 2504.1-6600.8) respectively. This suggested high pooled accuracy for all studies. We found substantial heterogeneity between studies. Three factors (study location, reference standard and study score) appeared most strongly associated with test estimates and observed heterogeneity. The Determine test showed consistency in performance in studies done across developed and developing countries and the Determine and the BinaxNOW tests had significantly higher estimates than pooled estimates of remaining tests. Tests revealed analytical sensitivity of 4 IU/ml against manufacturer's claim of 0.5 IU/ml; reduced sensitivity with HBsAg mutants and poor performance in seroconversion panels.

CONCLUSIONS

Tests with better analytical sensitivity need to be developed and their feasibility and outcomes in various clinical settings need to be addressed.