Diagnostic Accuracy of Novel and Traditional Rapid Tests for Influenza Infection Compared With Reverse Transcriptase Polymerase Chain Reaction: A Systematic Review and Meta-analysis.

BACKGROUND

Rapid and accurate influenza diagnostics can improve patient care.

PURPOSE

To summarize and compare accuracy of traditional rapid influenza diagnostic tests (RIDTs), digital immunoassays (DIAs), and rapid nucleic acid amplification tests (NAATs) in children and adults with suspected influenza.

DATA SOURCES

6 databases from their inception through May 2017.

STUDY SELECTION

Studies in English, French, or Spanish comparing commercialized rapid tests (that is, providing results in <30 minutes) with reverse transcriptase polymerase chain reaction reference standard for influenza diagnosis.

DATA EXTRACTION

Data were extracted using a standardized form; quality was assessed using QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) criteria.

DATA SYNTHESIS

162 studies were included (130 of RIDTs, 19 of DIAs, and 13 of NAATs). Pooled sensitivities for detecting influenza A from Bayesian bivariate random-effects models were 54.4% (95% credible interval [CrI], 48.9% to 59.8%) for RIDTs, 80.0% (CrI, 73.4% to 85.6%) for DIAs, and 91.6% (CrI, 84.9% to 95.9%) for NAATs. Those for detecting influenza B were 53.2% (CrI, 41.7% to 64.4%) for RIDTs, 76.8% (CrI, 65.4% to 85.4%) for DIAs, and 95.4% (CrI, 87.3% to 98.7%) for NAATs. Pooled specificities were uniformly high (>98%). Forty-six influenza A and 24 influenza B studies presented pediatric-specific data; 35 influenza A and 16 influenza B studies presented adult-specific data. Pooled sensitivities were higher in children by 12.1 to 31.8 percentage points, except for influenza A by rapid NAATs (2.7 percentage points). Pooled sensitivities favored industry-sponsored studies by 6.2 to 34.0 percentage points. Incomplete reporting frequently led to unclear risk of bias.

LIMITATIONS

Underreporting of clinical variables limited exploration of heterogeneity. Few NAAT studies reported adult-specific data, and none evaluated point-of-care testing. Many studies had unclear risk of bias.

CONCLUSION

Novel DIAs and rapid NAATs had markedly higher sensitivities for influenza A and B in both children and adults than did traditional RIDTs, with equally high specificities.

PRIMARY FUNDING SOURCE

Québec Health Research Fund and BD Diagnostic Systems.