Medical and interventional outcome of dissection of the cervical arteries. Systematic review and meta-analysis.

INTRODUCTION:The management of cervical artery dissections (CAD) is poorly standardized given the scarce number of prospective studies comparing medical and interventional approach to CAD. The aim of the present study is to perform a systematic review and meta-analysis of studies on the treatments of CAD. METHODS:Systematic review and meta-analysis - pre-registered on PROSPERO (CRD42022297512) and performed according to the PRISMA guidelines searching in three different databases (PubMed, Embase and Cochrane Database) - of studies on medical or interventional approach to the CAD. Only prospective studies were selected in order to reduce the risk of bias for the primary meta-analysis. Secondarily retrospective studies were also included. The aim was to assess the rate of stroke and of stroke/death/bleeding (major or intracranial) by Der Simonian-Laird weights of random effects model. RESULTS:After screening 456 articles, 6 prospective and 22 retrospective studies were identified. Two randomized controlled trials and 5 retrospective studies comparing antiplatelet (APT) vs. oral anticoagulant therapy (OAC) for CAD were identified, as well as 4 prospective and 17 retrospective single-arm studies evaluating stenting for CAD. In the meta-analysis of RCTs comparing APT vs OAC, 444 patients were considered and a borderline significant association was identified in terms of stroke/death in APT vs OAC groups (OR 5.6; 95% CI: 0.94-33.38, P=.06, I2: 0%). No differences were found for the stroke/death/bleeding outcome OR: 1.25, 95% CI: 0.19-8.18, P=.81, I2: 0% between the two treatments. In the meta-analysis including also retrospective studies, the risk of bias was "serious" and 4104 patients were included with no differences in APT vs. OAC for stroke (OR: 1.06, 95% CI: 0.53-2.11, P=.29, I2: 18%); no other comparisons were possible. The pooled meta-analysis of prospective studies on stenting for CAD included 4 series, for a total of 68 patients, in whom stenting was adopted primarily after failed medical therapy or after traumatic dissection. The pooled rate of stroke/death was 7% (95% CI: 3%-17%, I2=0%). The analysis of moderators identified a significant inverse association between the percentage of traumatic dissection and a reduction in postoperative stroke, Y=-1.60-2.02X, P=.03. The pooled rate of the composite endpoint of stroke/death/intracranial/bleeding was 8% (95% CI: 3%-18%, I2=0%). Secondarily, the meta-analysis including also 17 retrospective studies with an overall of 457 patients showed a 2.1% pooled rate of stroke/death (95% CI: 1.0%-3.3%, I2=0%) and 3.2% stroke/death/intracranial/major bleeding (95% CI: 1.8%-4.7%, I2=0%) CONCLUSIONS: Few prospective studies on CAD treatment are present in the literature. APT and OAC seem to have similar efficacy in reducing the recurrence of stroke after CAD. No definitive conclusion can be drawn for stenting, due to the low number of studies available. More prospective studies are necessary to evaluate its potential additional value over medical therapy alone in the early phase after CAD.