Between May 1976 and October 1989, 246 patients with primary cerebellar hemorrhage were admitted to our university hospital and affiliated hospitals.
The patients were classified into four categories; benign type, moderate type, severe type and fulminant type, according to the grading of hypertensive cerebellar hemorrhage proposed by Matsumoto in 1982. We then decided the most appropriate therapy according to this grading, i.e., conservative treatment for benign type, either conservative treatment or ventricular drainage, or hematoma evacuation for moderate type, hematoma evacuation by suboccipital craniectomy or stereotactic aspiration surgery for severe type, and urgent surgical treatment for fulminant type if possible.
There were 76 patients (30.9%) with benign type, 51 (20.7%) with moderate type, 65 (26.4%) with severe type, and 54 (22.0%) with fulminant type.
One hundred and twelve patients (45.5%) underwent conservative treatment, 37 (15.1%) were given ventricular drainage, 59 (24.0%) underwent suboccipital craniectomy and 38 (15.4%) underwent stereotactic aspiration surgery.
The age of the patients at onset showed a peak in the seventh decade, the mean age being 67.2 years. The mean age of patients with fulminant type was significantly lower than that of patients with other types.
Patients with past history of hypertension numbered 132 (53.7%), those with cerebral infarction 28 (11.4%), and those with cerebral hemorrhage 11 (4.5%).
Nausea and vomiting were the most common initial symptoms of benign type hemorrhage followed by vertigo and dizziness. However, in patients with the fulminant type, disturbance of consciousness was most common.
Hematoma in the benign type was mostly located in the cerebellar hemisphere, but in most cases of the severe type and fulminant type it had invaded the vermis or was located within it.
Among patients given conservative therapy, the proportion with benign type showing a good outcome (ADL1 or 2) was 84, 7%, the corresponding figure for moderate type was 56.3%, but all patients with severe and fulminant type died except one.
Among patients given ventricular drainage, the proportion showing a good outcome for moderate type was better than that for those given conservative therapy. However proportion of cases of severe and fulminant type with poor outcome were 75.0% and 100% in each.
Among patients given surgical treatment, the proportion of patients with moderate, severe and fulminant types showing a good outcome were 70.8%, 39.2%, and 20.0%, respectively. In other words, the outcome for these types with surgical treatment was better than those with conservative therapy. However, the outcome in fulminant type cases was poor with any treatment.
Thus it appears that patients with benign type hemorrhage can be controlled with conservative therapy only. Patients with moderate type who do not have hydrocephalus and have hematoma less than 3 cm in diameter can be treated by ventricular drainage only, and those who have hematoma more than 3 cm in diameter must be undergone surgical removal of the hematoma. Patients with severe type should also be undergone surgical removal of the hematoma. If patients with the fulminant type hemorrhage are not in a deep coma, stereotactic aspiration may be indicative.
View full abstract