Surgical Craniotomy for Intracerebral Haemorrhage.

Craniotomy is probably indicated for patients with superficial spontaneous lobar supratentorial intracerebral haemorrhage (ICH) when the level of consciousness drops below 13 within the first 8 h of the onset of the haemorrhage. Once the level drops below 9, it is probably too late to consider craniotomy for these patients, so clinical vigilance is paramount. While this statement is only backed up by evidence that is moderately strong, meta-analysis of available data suggests that it is true in the rather limited number of patients with ICH. Meta-analyses like this can often predict the results of future prospective randomised controlled trials a decade or more before the trials are completed and published. Countless such examples exist in the literature, as is the case for thrombolysis in patients with myocardial infarction in the last millennium: meta-analysis determined the efficacy more than a decade BEFORE the last trial (ISIS-2) confirmed the benefit of thrombolysis for myocardial infarction. Careful examination of the meta-analysis' Forest plots in this chapter will demonstrate why this statement is made at the outset. Other meta-analyses of surgery for ICH have also indicated that minimal interventional techniques using topical thrombolysis or endoscopy via burrholes or even twist drill aspiration may be particularly successful for the treatment of supratentorial ICH, especially when the clot is deep seated. Ongoing clinical trials (CLEAR III and MISTIE III) should confirm this in the fullness of time. There are 2 exceptions to these generalisations. First, based on trial evidence, aneurysmal ICH is best treated with surgery. Second, cerebellar ICH represents a special case because of the development of hydrocephalus, which may require expeditious drainage as the intracranial pressure rises. The cerebellar clot will then require evacuation, usually via posterior fossa craniectomy, rather than craniotomy. Technical advances suggest that image-guided surgery may improve the completeness of surgical evacuation and outcomes, regardless of which surgical technique is employed.