Bridging EIA and PCR testing results for the detection of rotavirus gastroenteritis in a US pediatric population, 2011-2016.

BACKGROUND: Rotavirus vaccine has greatly decreased rotavirus illness since being introduced in 2006 in the United States. Post vaccine introduction, enzyme immunoassay (EIA) rotavirus testing false positivity rates rose, and reverse transcription-polymerase chain reaction (RT-PCR) testing became prominent. RT-PCR testing is highly sensitive and can detect low quantities of viral shedding that could be clinically irrelevant. OBJECTIVES: We compared RT-PCR cycle threshold (Ct) values to EIA tests to bridge EIA and RT-PCR testing results. METHODS: Stool specimens from 2325 children with acute gastroenteritis (AGE) and 544 healthy controls (HCs) enrolled in the New Vaccine Surveillance Network during 2011-16 were tested for rotavirus via both EIA and semiquantitative RT-PCR (qRT-PCR) to compare rotavirus vaccine effectiveness (VE) and calculate Youden's index (sensitivity + specificity - 1) for various Ct cutoffs. RESULTS: The Youden's index indicated a Ct value < 29 was optimal for comparing EIA and qRT-PCR rotavirus testing; Ct value cutoffs ranging from < 27 to < 29 produced comparable sensitivity, specificity, and rotavirus VE estimates. Ct values were lower among AGE children than HCs. Rotavirus VE ranged from 85.5% among participants positive via both EIA and qRT-PCR (Ct < 40) to 59.4% among participants with a Ct value ≥ 29. CONCLUSIONS: A qRT-PCR Ct value cutoff of < 27 to < 29 was most comparable to EIA for rotavirus VE estimates. The Youden's index indicated a Ct value < 29 was the best single cutoff for rotavirus test positivity to provide an accurate estimation of clinically relevant rotavirus cases.