Assessment of BioFire Gastrointestinal Panel unconfirmed positive results for enteropathogenic bacteria and Norovirus and accuracy of a post-analytical algorithm to mitigate unconfirmed results

ABSTRACT This study investigated unconfirmed-positive rates with the BioFire FilmArray Gastrointestinal Panel (FAGP) for bacterial targets and Norovirus and determined if melt curve analysis can mitigate unconfirmed results. Positive FAGP results for bacteria were classified as culture-positive or culture-negative and as melt curve-specific or curve-nonspecific. Culture-negative samples were prospectively confirmed with the Xpert GI PCR Panel. Norovirus-positive results were classified as melt curve-specific or curve-nonspecific and confirmed with the Xpert Norovirus PCR. The primary aim was to measure the accuracy of FAGP with and without melt curve analysis. Culture-negative rate was 24.6% (100/407) for Campylobacter jejuni , 90.1% (100/111) for Campylobacter 2 targeting Campylobacter coli and Campylobacter helveticus/upsaliensis , 58.1% (100/172) for Shigella /enteroinvasive Escherichia coli (EIEC), and 38.5% (100/260) for Salmonella spp. Positive predictive value (PPV) of FAGP based on culture and Xpert GI PCR Panel was 100% (42/42) for C. jejuni , 16.7% (4/24) for Campylobacter 2, 94.1% (16/17) for Shigella /EIEC, and 73.2% (30/41) for Salmonella spp. Inclusion of melt curve analysis had no impact on the sensitivity but significantly improved the specificity for Campylobacter spp. from 75% (60/80) to 100% (80/80; P < 0.001), Shigella /EIEC from 98.0% (49/50) to 100% (50/50; P > 0.05), and for Salmonella spp. from 78.0% (39/50) to 96% (48/50; P < 0.01). PPV for Norovirus was 7.6% (7/92) for Norovirus 1 and 76.0% (73/96) for Norovirus 2. Melt curve analysis improved accuracy for Norovirus 2 (sensitivity 100% [73/73] to 98.6% [72/73; P > 0.05]; specificity 79.8% [91/114] to 98.2% [112/114; P < 0.001]) but not for Norovirus 1 (sensitivity 100% [7/7] to 28.6% [2/7; P < 0.01]; specificity 52.0% [92/177] to 100% [177/177; P < 0.001]). FAGP has a high unconfirmed-positive rate for Campylobacter spp., Shigella /EIEC, and Salmonella spp. and Norovirus, which can be partially mitigated with melt curve analysis. IMPORTANCE The BioFire FilmArray Gastrointestinal Panel is a widely used syndromic panel for rapid diagnosis of gastroenteritis, but recent studies have shown an alarming rate of false-positive results for Norovirus, and concerns have been raised for Campylobacter spp. and other bacterial targets. This study found substantial rates of unconfirmed positive results with the FilmArray Gastrointestinal Panel for Campylobacter spp., Shigella /enteroinvasive Escherichia coli (EIEC), and Salmonella spp. in addition to Norovirus. Through defining specific melt curve parameters, this study shows that it is possible to mitigate unconfirmed positives and assure the quality of results fully for Campylobacter spp., Shigella /EIEC, and Salmonella spp. and partially for Norovirus.