Diagnostic Performance of Gastrointestinal Pathogen Panels in Patients With Suspected Infectious Gastroenteritis: A Systematic Review and Meta-Analysis

Abstract

Introduction: Recently, the Food and Drug Administration has approved several multiplex molecular diagnostic tests (gastrointestinal pathogen panels [GIP]) for evaluating patients with suspected infectious gastroenteritis, which can simultaneously and rapidly detect 4-22 potential bacterial, viral and/or parasitic pathogens. We reviewed the overall diagnostic performance and detection rate of GIPs, specifically it's performance against traditional stool cultures for common bacterial pathogens to identify rates of discordance. Methods: We performed a systematic literature review from January 2011 to April 2017, to identify comparative studies, in which patients with suspected infectious gastroenteritis underwent testing with both FDA-approved GIPs and stool cultures, to assess rates of positivity and discordance, using random-effects meta-analysis. Results: From 1599 studies, we identified 13 cross-sectional comparative studies (9070 patients); only 5 studies were reportedly performed in patients with acute diarrhea, whereas clinical status was not reported in others. Overall, 47.7% patients (95% CI, 34.3-61.5) tested positive for at least one possible pathogen using GIPs, including 7.2% patients (95% CI, 3.6-13.8) who tested positive for 2 or more co-infections. In 5 studies limited to patients with ‘acute diarrhea', 74.1% (95% CI, 55.3-86.8) were positive on GIPs, including 14.5% patients (95% CI, 3.9-41.6) with co-infections. To assess comparative diagnostic performance of GIPs against a gold standard of bacterial stool culture, we limited further analyses to only specific bacterial pathogens in GIPs, which are detectable through traditional stool cultures (Campylobacter, Shigella, Salmonella, E.coli, Yersinia, Vibrio). On this analysis, 19.4% patients (95% CI, 11.4-30.9) tested positive on GIPs, as compared to 13.5% patients (95% CI, 8.1-21.7) who tested positive on stool cultures. Discordance was observed in 4.5% cases - 3.8% patients (95% CI, 1.5-9.2) tested positive for candidate bacterial species on GIPs but not on stool cultures (possible false positives); 0.7% patients (95% CI, 0.3-1.7) tested positive on stool cultures but negative on GIPs. Conclusion: While GIPs are able to identify potential etiologies for diarrhea in more patients than traditional methods, there is potential for false positive results. Longitudinal studies on impact of GIPs on medical decision-making, patient outcomes and public health relevance of findings are warranted.