Intraventricular haemorrhage (IVH) is associated with a poor outcome. Simple external ventricular drainage has not modified the high morbidity and mortality of these patients. Our objective was to review our experience using intraventricular urokinase (UK) in treating patients with moderate to severe IVH. Prospective analysis of medical records of 14 patients diagnosed with spontaneous IVH who received ventriculostomy and intraventricular infusion of UK from January 2002 to December 2005. Patients with the following characteristics were included: 18-70 years of age, GCS between 5 and 14, and moderate to severe IVH (Graeb > or = 6) without simultaneous intraparenchymal haematoma > 30 ml. The final results were compared to historic control group (14 patients) treated between January 1999 to December 2001 with ventriculostomy alone. All 28 patients accomplished the inclusion criteria. Patient age, initial GCS and Graeb classification of IVH were similar in the two groups of treatment. There was higher ventriculostomy obstruction rate in the non-UK group (33.3 vs. 0%; p > 0.05), a higher rate of intracranial hypertension in the non-UK group (66.6 vs. 16.6%; p = 0.036) and a lower mortality rate in the UK group (25 vs. 58.3%, p > 0.05). There was no rebleeding associated with UK treatment. Intraventricular UK appears to be a safe treatment. It is effective in the prevention of catheter blockage, speeding the clearance of IVH, and it is associated with lower rate of intracranial hypertension and death.