INTRODUCTION: Intraventricular/germinal matrix hemorrhage affects 7-30% of premature neonates, 25-80% of whom (depending on the grade of the hemorrhage) will develop hydrocephalus requiring shunting. Predisposing factors are low birth weight and gestational age. MATERIAL: There is increasing evidence for the role of TGF-β1 in the pathogenesis of hydrocephalus, but attempts to develop treatment modalities to clear the cerebrospinal fluid (CSF) from blood degradation products have not succeeded so far. Ultrasound is a valuable screening tool for high-risk infants and magnetic resonance imaging is increasingly utilized to differentiate progressive hydrocephalus from ex vacuo ventriculomegaly, evaluate periventricular parenchymal damage, decide on the surgical treatment of hydrocephalus, and follow up these patients in the long term. Treatment of increasing ventriculomegaly and intracranial hypertension in the presence of hemorrhagic CSF can involve a variety of strategies, all with relative drawbacks, aiming to drain the CSF while gaining time for it to clear and the neonate to reach term and become a suitable candidate for shunting. Eventually, patients with progressive ventriculomegaly causing intracranial hypertension, who have reached term and their CSF has cleared from blood products, will need shunting. CONCLUSION: Cognitive long-term outcome is influenced more by the effect of the initial hemorrhage and other perinatal events and less by hydrocephalus, provided that this has been addressed timely in the early postnatal period. Shunting can have many long-term side effects due to mechanical complications and overdrainage. In particular, patients with posthemorrhagic hydrocephalus are more susceptible to multiloculated hydrocephalus and encysted fourth ventricle, both of which are challenging to treat.