Methods: Search of MEDLINE, EMBASE, CINAHL, and CENTRAL to identify comparative observational studies (cohort, case-control, cross-sectional and ecological studies) reporting on outcomes of interest in pregnant women with evidence of influenza virus infection compared to non-pregnant patients. Outcomes included pneumonia, hospital admission, admission to intensive care units (ICU), receipt of mechanical ventilatory support, death, and a composite of ICU admission and/or death. A random effects model was used to obtain risk estimates.
Results: A total of 148 non-ecological and 10 ecological studies of pregnant women with evidence of influenza virus infection were included. There was a higher risk for influenza-associated hospital admission in pregnant versus non-pregnant patients in meta-analysis (odds ratio (OR) 2.94, 95% confidence interval 1.58-5.47). Meta-analysis found no effect of pregnancy on other influenza-associated outcomes: mortality (OR 1.03, 0.81-1.3), pneumonia (OR 1.80, 0.72-4.49), receipt of mechanical ventilatory support (OR 1.21, 0.70-2.08), or the composite of ICU admission and/or death (OR 0.95, 0.59-1.67). Influenza-associated ICU admission was less common in pregnant women when compared to non-pregnant women of reproductive age, only (OR 0.51, 0.42-0.62). In ecological studies, 4 of 8 studies reported higher mortality rates, 4 of 4 studies found higher hospital admission rates, and 2 of 2 studies found a higher admission rate to ICUs among pregnant women.
Conclusion: Meta-analysis of non-ecological studies demonstrated a higher risk of influenza-associated hospital admission for pregnant women, but no increase in other severe influenza events. In contrast, some ecological studies suggested a higher risk of death and ICU admission. Given the limitations of ecological study designs, influenza virus infection in pregnancy may not be associated with severe outcomes other than a higher likelihood of hospital admission.
J. Winkup, None
J. Ortiz, None
M. Loeb, None