Surgery for cerebral cavernous malformations: a systematic review and meta-analysis.

Harris, Lauren; Poorthuis, Michiel H F; Grover, Patrick; Kitchen, Neil; Al-Shahi Salman, Rustam

BACKGROUND We sought to quantify the risks of neurosurgical excision of cerebral cavernous malformations (CCMs) in a systematic review of cohort studies. METHODS We updated our previous systematic review by searching OVID Medline, OVID EMBASE, and the Cochrane Library from 1 January 2013 to 30 April 2019. The primary outcome was a composite of death attributed to CCM or surgery, non-fatal symptomatic intracerebral haemorrhage (ICH), or new or worsened persistent non-haemorrhagic focal neurological deficit (FND). RESULTS We included 70 cohorts, 67 reporting surgery alone, and three compared surgery to conservative management. A total of 5,089 patients (median age 36 years, 52% female) underwent surgery (total follow-up 19,404 patient-years). The annual rate of the composite outcome was 4.2% (95% CI 2.9 to 5.7; 46 cohorts; I2 = 93%), which was higher in cohorts reporting exclusively brainstem CCM (6.0%, 95% CI 4.1-8.3; 25 cohorts, I2 = 92%) versus predominantly supratentorial CCM (2.4%, 95% CI 1.3-3.8, 21 cohorts, I2 = 86%, phet = 0.001). The annual rate of the composite outcome was higher in cohorts with > 95% presenting with ICH (6.1%, 95% CI 4.2-8.4; 23 cohorts, I2 = 93%) versus others (2.3%, 95% CI 1.2-3.7; 23 cohorts, I2 = 83%, phet = 0.001). The incidence of the composite outcome did not change over time in cohorts of exclusively brainstem CCM (p = 0.7) or predominantly supratentorial CCM (p = 0.5). CONCLUSIONS The risk of death, ICH, or FND after CCM excision is ~ 4%. This risk is higher for brainstem CCM and CCM that have caused ICH but has not changed over time. TRIAL REGISTRATION This systematic review was registered (PROSPERO CRD42019131246).