Spontaneous cerebellar hemmorhage: Treatment of 20 cases

Spontaneous cerebellar hemorrhage (SCH) accounts for 5-10% of intracerebral hemorrhages. Long standing arterial hypertension with degenerative changes in the vessel walls and subsequent rupture is believed to be the most common cause. Infrequently, tumors, coagulopathies, arteriovenous malformations and amyloid angiopathy can also provoke SCH. Even though it was first described more than a century ago, the criteria that determine conservative or surgical management, as well as the factors influencing the course of this entity are still to be established.

Usually SCH presents with headache, nausea and vomiting, nystagmus, dysarthria, typical cerebellar signs and symptoms (e.g. limb ataxia, inability to stand and walk), nuchal pain and rigidity, loss or alteration of consciousness even with sudden death.

Over a period of seven (7) years a series of twenty (20) consecutive patients with SCH were treated in the Department of Neurosurgery of Thriassio General Hospital, Elefsina, Greece. On admission, all patients underwent a standard neurological examination and routine laboratory tests. The diagnostic imaging protocol included primarily computed tomography (CT) scans of the brain. Eight (8) patients were comatose on admission. Seven (7) presented with clinical signs of intracranial hypertension, whereas in five (5) patients headache was the dominant complain. Nine (9) cases were treated surgically and the remaining eleven (11) conservatively. Surgical treatment was undertaken in patients with Glasgow Coma Scale score (GCS score) <13, in hemispheric cerebellar hematomas larger than 30*40 mm or vermian hematomas larger than 25*35 mm in diameter, and in hemorrhages that caused obstructive hydrocephalus. In three (3) patients surgical intervention took place three-six days postictus due to rapid deterioration in the neurological status. In all of the surgically treated patients ventricular drainage of the cerebrospinal fluid was placed. Outcome was assessed by applying Glasgow Outcome Scale score (GOS score).

The follow-up was twelve (12) months. Eight (8) patients died. Three (3) died due to different causes than the hematoma in the posterior cranial fossa itself (cardiac arrhythmia attributable to ischemic heart disease, cerebral infarct and pulmonary embolism). Lower respiratory tract infections occurred in seven (7) patients. Outcome was assessed one month and one year after the manifestation of the hemorrhage. It was considered as poor in patients with GOS score 1-3 and favorable in patients with GOS score 4-5. At one year postictus eleven (11) patients had poor outcome, whereas nine (9) had favorable outcome. It has to be emphasized that the four (4) patients who were operated later due to coagulation abnormalities all had good outcome.

With the availability of CT scan, patients with SCH with milder symptoms and smaller hematomas are increasingly detected. Patients with GCS score >13 should be initially treated conservatively even if the size of the hematoma is marginal for surgical intervention. Nevertheless, they must be carefully monitored in case anything changes in their clinical picture predominantly in the level of consciousness. Signs of brain stem compression or brain stem hemorrhage in an elderly suffering already from serious systemic diseases have a negative influence on the outcome. Hematomas arising from the midline (vermian) have higher morbidity and worse outcome than hemispheric hemorrhages.

To summarize, in SCH the choice of treatment remains controversial. However, the underlining cause of the hemorrhage, patients age, size and particularly location of the hematoma, level of consciousness, the presence of obstructive hydrocephalus and the severity of possible secondary phenomena are factors that play a significant role in the decision for conservative or surgical management, as well as in the outcome of the patients.

Key words: Spontaneous cerebellar hemorrhage, therapy, hydrocephalus, outcome