PURPOSE: We sought to evaluate the long-term functional outcomes and identify the potential risk factors for rebleeding in patients with brain stem cavernous malformations (BCMs) who presented with hemorrhages and were surgically or conservatively treated and prospectively monitored. METHODS: From January 1990 to July 2015, we included patients with first hemorrhagic episodes secondary to single BCMs. Modified Rankin score (mRS) was used for neurological status assessment. Univariate and multivariate regression statistics were used to identify the risk factors for rebleeding. RESULTS: A total of 99 patients with BCMs hemorrhages were included (59 [59.6%] women, mean age 37± 13 years). As initial treatments, 37 patients (37.4%) underwent surgery and 62 (62.6%) received conservative treatment. The median follow-up was 3.33 years (interquartile range 1.16-7 years; 408.3 patient/years). The rebleeding rate by patient/year was 10% in conservatively treated patients. Deterioration was significantly more frequent in patients with rebleeding (p = 0.0001). At the end of the follow-up, the mRS were favorable in 49 patients (65.3%) without rebleeding, whereas only 8 (33.3%) with rebleeding evolved to favorable outcomes (p = 0.006). Lesion size >18 mm (hazards ratio, HR 3.34, 95% CI 1.54-7.26; p = 0.0001) and ventral location or crossing the brain stem's midpoint (HR 2.5, 95% CI 1.14-5.46; p = 0.022) were associated with a major risk of rebleeding in the univariate analysis, but only a lesion >18 mm remained statistically significant (HR 2.7, 95% CI 1.2-6.21; p = 0.016) in the multivariate analysis. CONCLUSION: A lesion size >18 mm was the principal factor associated with hemorrhage recurrence. The overall functional outcome was good. However, significant morbidity was attributable to rebleeding.
PURPOSE: We sought to evaluate the long-term functional outcomes and identify the potential risk factors for rebleeding in patients with brain stem cavernous malformations (BCMs) who presented with hemorrhages and were surgically or conservatively treated and prospectively monitored. METHODS: From January 1990 to July 2015, we included patients with first hemorrhagic episodes secondary to single BCMs. Modified Rankin score (mRS) was used for neurological status assessment. Univariate and multivariate regression statistics were used to identify the risk factors for rebleeding. RESULTS: A total of 99 patients with BCMshemorrhages were included (59 [59.6%] women, mean age 37± 13 years). As initial treatments, 37 patients (37.4%) underwent surgery and 62 (62.6%) received conservative treatment. The median follow-up was 3.33 years (interquartile range 1.16-7 years; 408.3 patient/years). The rebleeding rate by patient/year was 10% in conservatively treated patients. Deterioration was significantly more frequent in patients with rebleeding (p = 0.0001). At the end of the follow-up, the mRS were favorable in 49 patients (65.3%) without rebleeding, whereas only 8 (33.3%) with rebleeding evolved to favorable outcomes (p = 0.006). Lesion size >18 mm (hazards ratio, HR 3.34, 95% CI 1.54-7.26; p = 0.0001) and ventral location or crossing the brain stem's midpoint (HR 2.5, 95% CI 1.14-5.46; p = 0.022) were associated with a major risk of rebleeding in the univariate analysis, but only a lesion >18 mm remained statistically significant (HR 2.7, 95% CI 1.2-6.21; p = 0.016) in the multivariate analysis. CONCLUSION: A lesion size >18 mm was the principal factor associated with hemorrhage recurrence. The overall functional outcome was good. However, significant morbidity was attributable to rebleeding.