Neurodevelopmental consequences of preterm isolated cerebellar hemorrhage: A systematic review

CONTEXT: The effect of neonatal cerebellar hemorrhage on neurodevelopmental outcome (NDO) in the absence of supratentorial injury is still largely unknown.

OBJECTIVE: To evaluate the influence of isolated neonatal cerebellar hemorrhage on cognitive, motor, language, and behavioral NDOs and assess the effect of location and size on outcome.

DATA SOURCES: Embase, Medline, and Scopus were searched from inception to September 30, 2017.

STUDY SELECTION: Studies in which a diagnosis of isolated cerebellar hemorrhage was reported in preterm infants (<32 weeks' gestation) with a standardized NDO at ≥12 months of age were included.

DATA EXTRACTION: Patient characteristics, location, and size of bleeding and NDO (defined as severe [yes or no] on the basis of given cutoff points) in 4 domains were extracted.

RESULTS: Of the 1519 studies identified, 8 were included in final analyses. Of infants with isolated cerebellar hemorrhage, 128 were described (cumulative incidence: 2.3%). The incidence of severe delay in cognition, motor, language, and behavioral development was 38%, 39%, 41%, and 38%, respectively. The overall incidence of severe neurodevelopmental delay in ≥1 domain ranged from 43% to 75% and was most seen in infants with vermis involvement (87%-93%) and with large bleeds (46%-82%).

LIMITATIONS: Different neurodevelopmental scales lead to data heterogeneity, and reporting of data on a group level limited possibilities for an outcome description on an individual level.

CONCLUSIONS: Of infants with isolated cerebellar hemorrhage, 43% to 75% were severely delayed in cognition, motor, language, and/or behavioral development, with the highest incidence with vermis involvement and with large bleeds.