Cavernous malformations of the brainstem: experience with 100 patients

Object. In this study the authors review surgical experience with cavernous malformations of the brainstem (CMBs) in an attempt to define more clearly the natural history, indications, and risks of surgical management of thes e lesions. Methods. The authors retrospectively reviewed the cases of 100 patients (38 males and 62 females; mean age 37 years) harboring 103 lesions at treated a single institution between 1984 and 1997. Clinical histories, radiographs , pathology records, and operative reports were evaluated. The brainstem le sions were distributed as follows. pens in 39 patients, medulla in 16, midb rain in 16, pontomesencephalic junction in 15, pontomedullary junction in 1 0, midbrain-hypothalamus/thalamus region in two patients, and more than two brainstem levels in five. The retrospective annual hemorrhage rate was mos t conservatively estimated at 5% per lesion per year. Standard skull base a pproaches were used to resect lesions in 86 of the 100 patients. Intraopera tively, all 86 patients were found to have a venous anomaly in association with the CMB. Follow up was available in 98% (84 of 86) of the surgical pat ients. Of these, 73 (87%) were the same or better after surgical interventi on, eight (10%) were worse, and three (4%) died. Two surgical patients were lost to follow-up review. Incidences of permanent or severe morbidity occu rred in 10 (12%) of the surgically treated patients. The average postoperat ive Glasgow Outcome Stale score for surgically treated patients was 4.5, wi th a mean follow-up period of 35 months. Conclusions. The natural history of CMBs is worse than that of cavernous ma lformations in other locations. These CMBs can be resected using skull base approaches, which should be considered in patients with symptomatic hemorr hage who harbor lesions that approach the pial surface. Venous anomalies ar e always associated with CMBs and must be preserved.