RISKS OF SURGICAL-MANAGEMENT FOR CAVERNOUS MALFORMATIONS OF THE NERVOUS-SYSTEM

OBJECTIVE: As more information evolves regarding the natural history o f cavernous malformations (CMs), the risks of operative intervention m ust be balanced against nonoperative management. in an attempt to bett er delineate the surgical risks for operable CMs, we undertook a retro spective analysis of 94 patients with 97 CMs surgically excised at the Massachusetts General Hospital. METHODS: Data regarding surgical comp lications and outcome measures, including neurological status and seiz ure outcome, were analyzed. RESULTS: The incidence of transient neurol ogical morbidity was 20.6%, but only 4 of the 97 operations (4.1%) res ulted in persistent disabling neurological complications and 2 (2.1%) in nondisabling deficits. There was no operative mortality. Brain stem lesions (n = 14) were associated with the highest incidence of neurol ogical complications, both transient and persistent (odds ratio, 4.8; 95% confidence interval, 1.5-15.7). The overall neurological outcome w as excellent or good in 89.7% of all lesions: 96.8% of lobar CMs (n = 63), 64.2% of brain stem CMs (n = 14), 87.5% of cerebellar CMs (n = 8) , 100% of cranial nerve CMs (n = 4), and 75% of spinal cord CMs (n = 8 ). Patients with brain stem and spinal cord CMs were in poorer preoper ative neurological condition than were patients with CMs in other loca tions and therefore had a significantly reduced level of function afte r surgery (P < 0.01). There was improvement in 35.7% of the patients w ith brain stem lesions and 62.5% of the patients with spinal cord lesi ons after surgery. In the 38 patients presenting with seizures, 97% we re seizure-free after surgery. CONCLUSION: The risks of operative mana gement of CMs varies based on location. When evaluating patients with operable CMs for surgery, the incidence of complications as well as fi nal neurological outcome should be carefully weighed against the exist ing knowledge of the natural history of lesions managed expectantly.