BACKGROUND: Despite the increased risk of hemorrhage and deteriorating neurological function of once-bled cerebral cavernous malformations (CM), the management of eloquently located CMs remains controversial. METHODS: All eloquently located CMs (n = 45) surgically treated between 03/2006 and 04/2011 in our department were consecutively evaluated. Eloquence was characterized according to Spetzler and Martin's definition. The following locations were approached: brainstem, n = 16; sensorimotor, n = 8; visual pathway, n = 7; cerebellum (deep nuclei and peduncles), n = 7; basal ganglia, n = 4, and language, n = 3. Follow-up data was available for 41 patients (91 %) with a median interval of 14 months. Outcomes were evaluated according to the Glasgow outcome and the modified Rankin scale. RESULTS: Immediately after surgery, 47 % (n = 21) had a new deficit. At follow-up, 80 % (n = 36) recovered to at least preoperative status or were better than before surgery, 9 % (n = 4) exhibited a slight, and 7 % (n = 3) had a moderate neurological impairment. Only two cases (4 %) with a new permanent severe deficit were observed, both related to dorsal brainstem surgery. The outcome after the surgery of otherwise located brainstem CMs was as beneficial as that for non-brainstem CMs. Patients with initially poor neurological performance fared worse than oligosymptomatic patients. CONCLUSIONS: Despite the high postoperative transient morbidity, the majority improved profoundly during follow-ups. Compared with natural history, surgical treatment should be considered for all eloquent symptomatic CMs. Dorsal brainstem location and poor preoperative neurological status are associated with an increased postoperative morbidity.
BACKGROUND: Despite the increased risk of hemorrhage and deteriorating neurological function of once-bled cerebral cavernous malformations (CM), the management of eloquently located CMs remains controversial. METHODS: All eloquently located CMs (n = 45) surgically treated between 03/2006 and 04/2011 in our department were consecutively evaluated. Eloquence was characterized according to Spetzler and Martin's definition. The following locations were approached: brainstem, n = 16; sensorimotor, n = 8; visual pathway, n = 7; cerebellum (deep nuclei and peduncles), n = 7; basal ganglia, n = 4, and language, n = 3. Follow-up data was available for 41 patients (91 %) with a median interval of 14 months. Outcomes were evaluated according to the Glasgow outcome and the modified Rankin scale. RESULTS: Immediately after surgery, 47 % (n = 21) had a new deficit. At follow-up, 80 % (n = 36) recovered to at least preoperative status or were better than before surgery, 9 % (n = 4) exhibited a slight, and 7 % (n = 3) had a moderate neurological impairment. Only two cases (4 %) with a new permanent severe deficit were observed, both related to dorsal brainstem surgery. The outcome after the surgery of otherwise located brainstem CMs was as beneficial as that for non-brainstem CMs. Patients with initially poor neurological performance fared worse than oligosymptomatic patients. CONCLUSIONS: Despite the high postoperative transient morbidity, the majority improved profoundly during follow-ups. Compared with natural history, surgical treatment should be considered for all eloquent symptomatic CMs. Dorsal brainstem location and poor preoperative neurological status are associated with an increased postoperative morbidity.