Implementation of a Molecular Diagnostic Test for Pediatric Acute Gastroenteritis: The FilmArray GI Panel IMPACT Study

Background: New multiplex molecular tests can improve the detection of gastrointestinal pathogens, but the impact on clinical outcomes is unknown. We tested whether implementation of the FilmArray Gastrointestinal (GI) Panel (~1 hr turnaround time multiplexed PCR detecting 22 pathogens) improved outcomes for children presenting to the emergency department (ED) with acute gastroenteritis.

Methods: The GI IMPACT study is a multicenter stepped-wedge trial at 5 US children’s hospitals. Children <18 years presenting with acute gastroenteritis had clinical and epidemiologic data collected at baseline and day 7-10. During the pre-intervention period (PRE), standard of care (SOC) tests for GI pathogens were performed at the discretion of the provider; the FilmArray GI Panel was performed but the results were not reported. During the post-intervention period (POST), the FilmArray GI Panel was performed in real time and results were reported to the providers and families. The primary endpoint was the number of additional healthcare encounters within 10 days following the enrollment visit in the pre- and post-intervention periods; other outcomes were measured.

Results: 1043 children have completed the study to date; 571 children in the PRE period and 472 in the POST period. Enrollment is complete at 3 hospitals; enrollment in the POST period is ongoing at 2 hospitals (completion by July 2016). During the PRE period, 375 (66%) children submitted a stool sample; 23 (6%) were positive for ≥1 GI pathogens by SOC testing.  At least one pathogen was detected in 360/472 (77%) in the post-intervention period by FilmArray (P<0.001; Table). In the PRE period, a treatable pathogen was identified in 13/571 (2%) children by SOC methods (76 tested) and in 65/375 (17%) by FilmArray; during the POST period, a treatable pathogen was identified in 63/472 (13%) by FilmArray. In an unadjusted analysis, there was no difference in the average number of additional healthcare encounters between the periods (pre 0.39±0.67 vs. post 0.38±0.67; P=0.8).

Conclusion: Implementation of a rapid, multiplex diagnostic test for GI pathogens in the pediatric ED improved the identification of treatable pathogens but did not reduce subsequent healthcare encounters in this interim analysis.