Proposed use of prophylactic decompressive craniectomy in poor-grade aneurysmal subarachnoid hemorrhage patients presenting with associated large sylvian hematomas.

Objective: As a group, patients who present in poor neurological grade after aneurysmal subarachnoid hemorrhage (SAH) often have poor outcomes. There may be subgroups of these patients, however, in which one pathological process predominates and for which the initiation of specific therapeutic interventions that target the predominant pathological process may result in improved outcome. We report the use of prophylactic decompressive craniectomy in patients presenting in poor neurological condition after SAH from middle cerebral artery aneurysms with associated large sylvian fissure hematomas. Craniectomy allowed significant parenchymal swelling in the posthemorrhagic period without increased intracranial pressure (ICP) or herniation syndrome.
Methods: Eight patients (mean age, 56.5 yr; age range, 42-66 yr) presented comatose with SAH (five Hunt and Hess Grade IV, three Hunt and Hess Grade V). Radiographic evaluations demonstrated middle cerebral artery aneurysm and associated large sylvian fissure hematoma (mean clot volume, 121 ml; range, 30-175 ml). Patients were brought emergently to the operating room and treated with a modification of the pterional craniotomy and aneurysm clipping that included a planned craniectomy and duraplasty. A large, reverse question mark scalp flap was created, followed by bone removal with the following margins: anterior, frontal to the midpupillary line; posterior at least 2 cm behind the external auditory meatus; superior up to 2 cm lateral to the superior sagittal sinus; and inferior to the floor of the middle cranial fossa. Generous duraplasty was performed using either pericranium or suitable, commercially available dural substitutes.
Results: All of the eight patients tolerated the craniectomy without operative complications. Postoperatively, all patients experienced immediate decreases in ICP to levels at or below 20 mm Hg (presentation mean ICP, 31.6 mm Hg; postoperative mean ICP, 13.1 mm Hg). ICP control was sustained in seven of eight patients, with the one exception being due to a massive hemispheric infarction secondary to refractory vasospasm. Follow-up (> or = 1 yr, except for one patient who died during the hospital stay) demonstrated that the craniectomy patients had a remarkably high number of good or excellent outcomes. The outcomes in the hemicraniectomy group were five good or excellent, one fair, and two poor or dead.
Conclusion: The data gathered in this study demonstrate that decompressive craniectomy can be performed safely as part of initial management for a subcategory of patients with SAH who present with large sylvian fissure hematomas. In addition, the performance of decompressive craniectomy in the patients described in this article seemed to be associated with rapid and sustained control of ICP. Although the number of patients in this study is small, the data lend support to the hypothesis that decompressive craniectomy may be associated with good or excellent outcome in a carefully selected subset of patients with SAH.