Incidence and risk factors of cerebral sinovenous thrombosis in infants.

Aim: To describe the incidence of term and preterm neonatal cerebral sinovenous thrombosis (CSVT) and identify perinatal risk factors.

Method: This was a national capture-recapture calculation-corrected surveillance and nested case-control study. Infants born preterm and at term with magnetic resonance imaging-confirmed neonatal CSVT were identified by surveillance in all paediatric hospitals in Germany (2015-2017). Incidence was corrected for underreporting using a capture-recapture method in one federal state and then extrapolated nationwide. We reviewed PubMed for comparisons with previously reported incidence estimators. We used a population-based perinatal database for quality assurance to select four controls per case and applied univariate and multivariable regression for risk factor analysis.

Results: Fifty-one newborn infants (34 males, 17 females; 14 born preterm) with neonatal CSVT were reported in the 3-year period. The incidence of term and preterm neonatal CSVT was 6.6 (95% confidence interval [CI] 4.4-8.7) per 100 000 live births. Median age at time of confirmation of the diagnosis was 9.95 days (range 0-39d). In the univariate analysis, male sex, preterm birth, hypoxia and related indicators (umbilical artery pH <7.1; 5-minute Apgar score <7; intubation/mask ventilation; perinatal asphyxia), operative vaginal delivery, emergency Caesarean section, and pathological fetal Doppler sonography were associated (p<0.05) with neonatal CSVT. Multivariable regression yielded hypoxia (odds ratio=20.3; 95% CI 8.1-50.8) as the independent risk factor.

Interpretation: Incidence of neonatal CSVT was within the range of other population-based studies. The results suggest that hypoxia is an important perinatal risk factor for the aetiology of neonatal CSVT.