Radiosurgery versus observation for brainstem cavernous malformations: a 5-year multicentre cohort study

Abstract The role of radiosurgery in preventing haemorrhage in brainstem cavernous malformations remains a subject of debate. This study aimed to evaluate whether radiosurgery provides a protective benefit against haemorrhage in these patients. This multicentre, prospective observational study was conducted in 17 centres and enrolled eligible patients with brainstem cavernous malformations consecutively. Data collected included clinical baseline information, radiosurgery planning details, periodic follow-up evaluations and any adverse radiation effects. The primary outcome of the study was the incidence of first prospective haemorrhage, and the secondary outcome was the development of new or worsening neurological dysfunctions. The impact of radiosurgery was assessed using multivariate Cox regression analysis. From March 2016 to August 2018, the study enrolled 377 patients: 280 in the observation group receiving standard care alone and 97 in the radiosurgery group receiving both radiosurgery and standard care. The overall cohort consisted of 173 females (45.9%) with a mean age of 40.5 years (range, 18–68 years), and there were no significant differences in baseline characteristics between the two groups. After a median follow-up period of 70 months, haemorrhage occurred in 25.0% (n = 70) of patients in the observation group and 10.3% (n = 10) of patients in the radiosurgery group. Multivariate Cox regression analysis identified radiosurgery as an independent protective factor against haemorrhage (hazard ratio 0.379, 95% confidence interval 0.195–0.738, P = 0.004). Following 1:2 propensity score matching, the incidence of prospective haemorrhage was 24.9% (45/181) in the observation group compared with 10.3% (10/97) in the radiosurgery group (hazard ratio 0.379, 95% confidence interval 0.190–0.755, P = 0.006). Adverse radiation effects were observed in 12 patients (12.4%), with none being permanent. Additionally, new or worsening neurological dysfunctions were significantly more common in the observation group (28.9%) compared with the radiosurgery group (16.5%) (P = 0.016). These results suggest that radiosurgery is associated with a low rate of haemorrhage in patients with brainstem cavernous malformations and could provide a benefit in selected patients. However, further research is required to confirm these findings.