Outcome of fetuses with congenital parvovirus B19 infection: systematic review and meta-analysis

Objective: To explore the outcome of fetuses affected by congenital parvovirus B19 (PB19) infection, with or without signs of hydrops on ultrasound. Methods: PubMed, EMBASE and CINAHL databases were searched for studies reporting on prenatal diagnosis and outcome of fetal PB19 infection. The outcomes explored were miscarriage, perinatal death (PND), intrauterine death, neonatal death, spontaneous resolution of hydrops or fetal anemia, need for intrauterine transfusion (IUT), resolution of hydrops or anemia after transfusion, fetal loss following transfusion, abnormal brain scan after birth and abnormal neurodevelopmental outcome. Outcomes were reported according to the presence or absence of signs of hydrops on ultrasound. A subgroup analysis was performed including hydropic and non-hydropic fetuses diagnosed at < 20 weeks and ≥ 20 weeks of gestation. Meta-analyses of proportions and meta-analyses using individual-data random-effects logistic regression were performed to analyze the data. Results: Thirty-five observational studies were included, involving 611 fetuses affected by PB19 infection. The risks of miscarriage (odds ratio (OR), 11.5; 95% CI, 2.7–49.7) and PND (OR, 4.2; 95% CI, 1.6–11.0) were higher in fetuses with PB19 infection presenting, compared with those not presenting, signs of hydrops on ultrasound. In fetuses affected by hydrops, spontaneous resolution of the infection, defined as disappearance of hydrops without need for IUT, occurred in 5.2% (95% CI, 2.5–8.8%) of cases whereas, in the group of fetuses not affected by hydrops, infection resolved in 49.6% (95% CI, 20.7–78.6%) of cases. IUT was performed in 78.7% (95% CI, 66.4–88.8%) of hydropic and in 29.6% (95% CI, 6.0–61.6%) of non-hydropic fetuses affected by congenital PB19 infection and resolution of the infection after IUT occurred in 55.1% (95% CI, 34.0–75.3%) and in 100% (95% CI, 57.3–100%) of cases, respectively. The risk of fetal loss after IUT was higher in fetuses affected compared with those not affected by hydrops (OR, 9.8; 95% CI, 2.8–34.6). The prevalence of abnormal brain imaging was 9.8% (95% CI, 2.5–21.0%) in fetuses affected and 0.0% (95% CI, 0.0–7.0%) in those not affected by hydrops, whilst the corresponding figures for abnormal neurodevelopmental outcome were 9.5% (95% CI, 2.6–20.2) and 0.0% (95% CI, 0.0–7.5), respectively; however, statistical power to assess these outcomes was inadequate due to the small number of included cases. Conclusions: Hydrops is the main determinant of mortality and adverse perinatal outcome in fetuses with PB19 infection. Perinatal outcome in non-hydropic fetuses is generally favorable.