Perinatal outcomes of symptomatic chikungunya, dengue and Zika infection during pregnancy in Brazil: a registry-based cohort study

Abstract

The previous literature shows mixed conclusions regarding the risk of adverse perinatal outcomes in pregnant women with symptomatic chikungunya, dengue, and Zika. We investigated this topic using a linked population-based Brazilian cohort from 2015 to 2020. The study included 6,993,395 live births. Among these, 6066 ( < 0.1%) mothers were notified with chikungunya during pregnancy, 19,022 (0.3%) with dengue, and 8396 (0.1%) with Zika. Symptomatic maternal chikungunya was associated with an increased risk of preterm birth (Hazard ratio: 1.10, 95%CI 1.01-1.22), low Apgar score 5' (1.44, 1.14-1.82), and neonatal death (1.50, 1.15-1.96). Symptomatic maternal dengue was associated with preterm birth (1.07, 1.02-1.12), low birth weight (1.10, 1.04-1.15), congenital anomalies (1.19, 1.03-1.37), and low Apgar score 5' (1.26, 1.09-1.45). Symptomatic maternal Zika was associated with all adverse birth outcomes, particularly congenital anomalies, which were over twice the risk (2.36; 1.91-2.67) compared to the unexposed group. This study provides evidence of the adverse consequences of arbovirus infections during pregnancy, including critical time windows by trimester. Our findings emphasise the importance of implementing effective measures to prevent chikungunya, dengue, and Zika infections during pregnancy and the associated adverse birth and neonatal outcomes, which may have long-term health consequences for mothers and their children.

Fig. 1

Selection of study participants.

Fig. 2. Symptom onset by gestational week.

Distribution of pregnant women with arbovirus infection by gestational week, stratified by virus. Blue represents the number of pregnant women with symptomatic chikungunya infection; red represents the number of pregnant women with symptomatic dengue infection; green represents the number of pregnant women with symptomatic Zika infection.

Fig. 3. Overall effect estimates.

Estimated adjusted hazard ratios (birth outcomes) and adjusted risk ratios (neonatal death), comparing groups exposed and unexposed to arbovirus infection during pregnancy by outcome. Error bars represent the 95% confidence interval. NCE negative control exposure.

Fig. 4. Trimester effect estimates.

Estimated adjusted hazard ratios (birth outcomes) and adjusted risk ratios (neonatal death), comparing groups exposed and unexposed to arbovirus infection during pregnancy by trimester of disease and outcome. The number of pregnant women with symptomatic chikungunya, dengue and Zika infection by trimester is displayed in Supplementary Table 5. Error bars represent the 95% confidence interval.