PURPOSE The aim of this study is to assess the treatment results in patients with cerebellar infarctions and attempt to determine the features influencing the results. MATERIAL AND METHODS We have analysed 12 patients with cerebellar infarctions treated from 1987 to 1996. The patients were admitted to hospital in the period between 24 hours and 37 days after onset. The level of consciousness on admission was assessed according to the Glasgow Coma Scale and results of treatment were assessed according to the Glasgow Outcome Scale at the time of discharge. The CT, MR, angio-MR, cerebral arteriography, USG-Doppler and ECHO were the main diagnostic methods. All patients were treated medically; temporary external ventricular drainage or ventriculoperitoneal shunt were used when hydrocephalus had been found. The influence of the level of consciousness, stroke localisation and presence of hydrocephalus on the results of treatment were analysed. RESULTS 2 patients were in coma on admission, 3 patients scored 13-14 points in GCS and 7 patients did not demonstrate any consciousness disturbances. On neurological examination patients showed cerebellar signs in 9 cases, hemiparesis in 4 cases, lower cranial nerve dysfunction in 3 cases, eyeball movement disorders, compulsory gaze deviation, hemihypoaesthesia, trigeminal nerve dysfunction, peripheral facial palsy in 2 cases each and deafness in 1 case. Infarctions were localised in the cerebellar hemisphere only in 7 cases, in the cerebellar hemisphere with vermian extension in 2 cases and in 3 cases in cerebellum with brain stem extension. Based on additional investigations atherosclerotic occlusion in 3 cases and embolism mechanism in 4 cases were found as causes of the stroke; in 5 cases the cause of stroke remained unknown. The hydrocephalus resulting from the obstruction of cerebrospinal flow pathways was found in 5 patients. 1 patient died, permanent neurological deficits (GOS: SD) were found in 2 patients, the treatment results were successful (GOS: MD + GR) in 9 patients. The infarctions with brain stem extension had the greatest influence on treatment results among all analysed features. The level of consciousness on admission and presence of hydrocephalus had less influence on treatment results. CONCLUSIONS In all patients with consciousness disturbances the hydrocephalus was found. The stroke localisation had a great influence on treatment results: worst results were obtained in patients with infarcts extending onto the brain stem. The consciousness disturbances on admission and the presence of hydrocephalus had only a slight influence on the results of treatment.