Surgical management of pediatric brainstem cavernous malformations

Abstract

The goal of this study was to evaluate surgical outcomes of pediatric brainstem cavernous malformations (CMs) and identify the risk factors associated with postoperative full recovery and rebleeding. The clinical charts and radiographs from a series of 52 pediatric patients (37 male and 15 female; mean age 12.2 years; range 1-17 years) who underwent surgery for brainstem CMs between 1996 and 2011 were reviewed. Follow-up evaluation measures were obtained retrospectively. Neurological function was evaluated using the modified Rankin Scale (mRS) score. The lesion locations among the 52 patients included the midbrain (n = 7, 13.5%), pons (n = 38, 73.1%), and medulla (n = 7, 13.5%). The mean duration of symptoms was 18.5 months, and the preoperative annual hemorrhage and rebleeding rates were 12.3% and 32.5% per patient-year, respectively. The mean lesion size was 2.1 cm. Gross-total resection without surgery-related death was achieved in 49 patients (94.2%). Immediate postoperative reduced neurological function was observed in 17 patients (32.7%). Surgical morbidities developed in 25 patients (48.1%) and remained in 11 patients (21.2%) after 7.9 years of follow-up. The mean mRS scores at admission, discharge after surgery, 3 and 6 months postsurgery, and recent evaluation were 2.0, 2.3, 2.0, 1.5, and 1.0, respectively. The postoperative mRS scores at 6 months (p < 0.001) and on recent evaluation (p < 0.001) were significantly lower than those at admission. Postoperative rebleeding occurred in 2 patients, and the postoperative annual rebleeding rate was 0.5% per patient-year. By the most recent evaluation, 10 patients (19.2%) had achieved full recovery and all patients were either improved (n = 32, 61.5%) or unchanged (n = 20, 38.5%). The adverse predictors for full recovery included age ≥ 12 years (HR 0.230, p = 0.021), ≥ 2 preoperative hemorrhages (HR 0.124, p = 0.048), and poor preoperative status (HR 0.197, p = 0.040). An HR < 1 predicted poor complete recoveries. The single risk factor predicting postoperative rebleeding was incomplete resection (χ2 = 4.340, p = 0.037). Fair outcomes for pediatric brainstem CMs could be obtained through surgery, but only a few patients achieved full recovery. Thus, to minimize surgical morbidity, surgical planning must be tailored to individual patients in all cases in which an operation is warranted. Complete resection must be attempted to reduce the risk of postoperative rebleeding. The predictors associated with complete postoperative recovery were referential for determining treatment.