Strengths and limitations
To the best of our knowledge, this study is the most extensive Brazilian cohort study of women with confirmed ZIKV infection during gestation. There are few large cohorts with robust estimates regarding ZIKV impact on pregnancy losses and fetal malformations. Two early studies of Brazilian cohorts reported very diverse results: -a high prevalence (42%) of abnormal clinical or imaging findings distributed across trimesters of maternal infection3; or -no microcephaly but minor ultrasonography findings aside from ophthalmologic and/or hearing loss in a third of infants8. It is possible that referral bias, the relatively small number of mother-infant pairs, and an extensive definition of adverse outcomes might have contributed to the overestimation of the reported effects. Our results were generated in a much more representative sample of the total population once it contained all pregnant women with confirmed symptomatic ZIKV infection living in a distinct geographic region. We used stricter definitions of abnormalities potentially related to ZIKV infection, systematically approached the gestational age of the maternal ZIKV infection as a continuous variable and explored additional etiologies for abnormal infants’ findings.
Our cohort set up which allowed obtaining birth data for all of the women with confirmed infection and evaluate most of their infants during the epidemics may be a model to investigate infection-related morbidity in other epidemics such as COVID-19. Further, it demonstrates the need to analyze the gestational age of infection to guide better clinical practice for ZIKV maternal infection and the importance of long-term follow-up for exposed infants, as known for other congenital infections.
Our study does have limitations. Only symptomatic women were selected, and not all infants were followed from birth. Nevertheless, its size is comparable to other country-specific studies, and it contained perinatal outcome data for all women with confirmed ZIKV infection. Infant findings potentially related to ZIKV infection were only considered after ruling out other conditions with a large subgroup of infants having complete clinical, neurological, and brain imaging data. Our results likely reflect the real effect of the maternal ZIKV-symptomatic infection in this population.