Variability in the diagnostic performance of a bedside rapid diagnostic influenza test over four epidemic seasons in a pediatric emergency department

Abstract

We wanted to determine the diagnostic performance of a rapid influenza diagnostic test (RIDT) used bedside in a pediatric emergency department (PED). This was a prospective study over four consecutive winters (2009-2013), comparing the results of a RIDT (QuickVue®) with RT-PCR in children admitted to a PED. Among the 764 children included, we did not observe any significant differences in the diagnostic performance of RIDT except during the H1N1 pandemic. The overall sensitivity of the test was 0.82; the specificity 0.98; the positive and negative likelihood ratios 37.8 and 0.19. The positive and negative post-test probabilities of infection were 98% and 17%. The diagnostic performance was increased for influenza B cases (P = 0.03). RIDTs are suitable for use every winter with few differences in its diagnostic value, except during specific pandemic periods. This test could limit unnecessary complementary exams and guide the prescription of antivirals during influenza epidemic periods in PEDs.

Keywords: Diagnostic performance; Epidemic seasons; RIDT; Sensibility.