Mechanical Thrombectomy for Aseptic, Atraumatic, Medically Refractory Cerebral Venous Sinus Thrombosis: a Systematic Review

Background Cerebral venous sinus thrombosis (CVST) follows a severe clinical course in 13.5% of cases. Practice guidelines recommend endovascular therapy (EVT); no randomised control trials (RCTs) exist comparing EVTs. Purpose To determine whether specific EVTs are superior to alternatives. Data Sources CENTRAL, Medline, Embase, five other databases and four clinical trials registers. Grey literature searches, reference checking, citation searching, and author contact. Study Selection All CVST cases treated with mechanical thrombectomy (MT) were includible. Paediatric, and trauma-related or infection-related thromboses were excluded. Data Analysis Standard Cochrane review procedures. Primary outcome measures; clinical efficacy (modified Rankin Score, mRS), technical efficacy (recanalisation), and clinical safety (procedure-related complications and death). Subgroup analyses were performed, comparing outcome measures between demographic groups, clinico-radiological severity, interventional strategies, and degrees of recanalisation. Data Synthesis In this study 124 papers were included (n = 486). All patients underwent MT, with 69.5% of patients receiving concomitant chemolysis. New/expanding intracerebral haemorrhage (ICH) occurred in 5.1%; non-haemorrhagic complications in 1.4%; 10.7% died. Predictors of poor efficacy included age >= 55 years, altered mental status (AMS), Glasgow Coma Scale (GCS) < 8. Predictive of poor safety outcomes included pre-existing ICH, deep system thrombosis, and AMS. Complete recanalisation was associated with improved clinical efficacy and safety outcomes. Limitations The review is based on case reports/series, increasing bias-risk. Myriad of potentially includible studies were necessarily excluded due to lack of requisite details. Conclusion Predictors of poor outcomes with medical therapy predict poor outcomes with MT; these measures should not dictate candidacy. Complete recanalisation predicts favorable clinical and safety outcomes. Local chemolysis is safe, improves recanalisation, and should be recommended, provided there is no contraindication. Clot maceration strategies and stent-retriever thrombectomy are associated with superior clinical efficacy and safety endpoints, as compared with balloon angioplasty and rheolysis.