To determine the association of antenatal magnesium sulfate with cerebellar hemorrhage in a prospective cohort of premature newborns evaluated by magnetic resonance imaging (MRI).Cross-sectional analysis of baseline characteristics from a prospective cohort of preterm newborns (<33 weeks gestation) evaluated with 3T-MRI shortly after birth. Exclusion criteria were clinical evidence of a congenital syndrome, congenital infection, or clinical status too unstable for transport to MRI. Antenatal magnesium sulfate exposure was abstracted from the medical records and the indication was classified as obstetric or neuroprotection. Two pediatric neuroradiologists, blinded to the clinical history, scored axial T2-weighted and iron susceptibility MRI sequences for cerebellar hemorrhage. The association of antenatal magnesium sulfate with cerebellar hemorrhage was evaluated using multivariable logistic regression, adjusting for postmenstrual age at MRI and known predictors of cerebellar hemorrhage.Cerebellar hemorrhage was present in 27 of 73 newborns (37%) imaged at a mean ± SD postmenstrual age of 32.4 ± 2 weeks. Antenatal magnesium sulfate exposure was associated with a significantly reduced risk of cerebellar hemorrhage. Adjusting for postmenstrual age at MRI, and predictors of cerebellar hemorrhage, antenatal magnesium sulfate was independently associated in our cohort with decreased cerebellar hemorrhage (OR, 0.18; 95% CI, 0.049-0.65; P = .009).Antenatal magnesium sulfate exposure is independently associated with a decreased risk of MRI-detected cerebellar hemorrhage in premature newborns, which could explain some of the reported neuroprotective effects of magnesium sulfate.