Sensitivity and Specificity of the Quidel Sofia Influenza A+B FIA Rapid Influenza Detection Test in Long-Term Care Facilities

BACKGROUND: Influenza is a significant pathogen for long-term care facility (LTCF) residents. As part of a randomized controlled trial to assess early detection of influenza in LTCFs, we deployed rapid influenza detection tests (RIDTs) at intervention LTCFs. Our primary objectives for this interim analysis were to evaluate the sensitivity and specificity of the Quidel Sofia® Influenza A+B Fluorescent Immunoassay RIDT in a high-risk, nontraditional population, and to describe the virology of acute respiratory infections (ARI) in LTCF residents. METHODS: Personnel at LCTFs identified cases of ARI, collected nasal specimens, and ran RIDTs from 10/21/2016 to 4/28/2017. The residual nasal swab and leftover lysis buffer were placed into a viral transport medium tube and sent to the Wisconsin State Laboratory of Hygiene for confirmatory influenza RT-PCR testing. In addition, all specimens were tested for other viruses using the Luminex NxTAG® Respiratory Pathogen Panel. Sensitivity and specificity of the Sofia RIDT were calculated using RT-PCR results as the reference standard. RESULTS: Specimens were collected from 228 residents (mean age = 71.3 ± 22.4 years). The mean time from symptom onset to specimen collection was 1.4 ± 1.6 days (range: 0-7 days). Respiratory viruses were identified in 134/228 cases (58.8%); influenza viruses (A: 7.5% and B: 14.5%) were the most commonly detected virus by PCR, followed by rhinovirus/enterovirus (13.2%), RSV (11.0%) and coronaviruses (10.1%). The sensitivities of Sofia RIDT for influenza A and influenza B were 77.8% (95% CI: 52.4–93.6%) and 80.0% (95% CI: 61.4–92.3%), respectively, with specificities of 98.4% (95.3–99.7%) and 97.1% (93.4–99.1%), respectively. Overall performance assessment for influenza A or B yielded a sensitivity of 79.2% (65.0–89.5%) and specificity of 96.1% (91.7–98.6%). The estimated likelihood of discovering one of the first two influenza cases at a LTCF using this RIDT is estimated to be ≥95.7%. CONCLUSION: Although a wide constellation of respiratory viruses cause ARIs within LTCF populations, influenza is very common. Early ARI recognition in residents, with testing shortly after symptom onset, likely contributed to high performance of the Sofia RIDT. Use of RIDTs allows early identification of influenza with high sensitivity and specificity in elderly LTCF residents. DISCLOSURES: J. Temte, Quidel: Investigator, Research support