Contribution of Serologic Assays in the Evaluation of Influenza Virus Infection Rates and Vaccine Efficacy in Pregnant Women: Report From Randomized Controlled Trials

Background. The utility of serologic testing to evaluate vaccine efficacy of seasonal inactivated influenza vaccine (IIV) is controversial. We aimed to evaluate the efficacy of IIV against serologically diagnosed influenza infection (SDI) and reverse-transcription polymerase chain reaction-confirmed influenza illness (PCR-CI) in women vaccinated during pregnancy. Methods. We undertook a post hoc analysis of 2 randomized clinical trials evaluating IIV efficacy among human immunodeficiency virus (HIV)-uninfected and HIV-infected pregnant women. SDI was defined as >= 4-fold increase in paired hemagglutinin antibody inhibition titers from 1 month postvaccination until end-of-study participation. PCR-CI was defined as molecular diagnostic evidence of influenza virus in pharyngeal specimens collected during clinical illness. Results. Among placebo recipients, the respective incidence of PCR-CI and SDI was 5.6% and 35.0% in HIV-uninfected women and 20.5% and 43.6% among HIV-infected women. Vaccine efficacy in HIV-uninfected women was similar for PCR-CI (66.9%; 95% confidence interval [CI], -20.1% to 90.9%) and SDI (59.2%; 95% CI, 37.0%-73.5%); however, fewer women required vaccination to prevent 1 episode of SDI (5; 95% CI, 3-9) than PCR-CI (27; 95% CI, 12-infinity). Also, vaccine efficacy was similar for PCR-CI (61.2%; 95% CI, 10.7%-83.2%) and SDI (60.9%; 95% CI, 33.9%-76.9%) in HIV-infected women, with 2-fold fewer women needing to be vaccinated to prevent SDI (4; 95% CI, 3-8) than PCR-CI (8; 95% CI, 4-52). Conclusions. Although vaccine efficacy was similar when measured for PCR-CI or SDI, IIV vaccination prevented a greater number of SDI than PCR-CI; the clinical relevance of the former warrants interrogation.