Antithrombotic Trends Before and After Publication of Randomized Clinical Trials in Cervical Artery Dissection: A Secondary Analysis of the STOP-CAD Study.

Background Two randomized clinical trials have failed to demonstrate the superiority of anticoagulation over antiplatelet therapy in acute cervical artery dissection, with many patients still receiving anticoagulation despite bleeding risks. In this analysis, we assessed temporal changes in antithrombotic strategies of cervical artery dissection following publication of these 2 trials.Methods This is a secondary analysis of a retrospective multinational observational cohort study evaluating outcomes related to antithrombotic treatment for acute cervical artery dissection (January 2015-December 2022). The odds of oral anticoagulant use (over single or combination antiplatelet) therapy across each year were estimated using multivariable logistic regression and joinpoint regression. Least absolute shrinkage and selection operator-adjusted regression was used to assess the odds of anticoagulation each year following 2016.Results Among 3345 included patients, 862 (25.8%) were treated with anticoagulation (n = 436 treated with anticoagulation and antiplatelet therapy). Nearly half (45.1%) were female with a median age of 46 years (interquartile range, 37-56). Compared with patients treated in 2015, those treated in subsequent years showed a stepwise decrease in treatment with any oral anticoagulation (odds ratio [OR] 0.91, 95% CI: 0.87-0.94). Independent predictors of increased anticoagulant use included female sex (adjusted OR [aOR] 1.22, 95% CI: 1.03-1.46), presence of infarct on imaging (aOR 1.45, 95% CI: 1.19-1.77), and partially occlusive thrombus (aOR 3.09, 95% CI: 2.38-4.01). Meanwhile, independent predictors of decreased anticoagulant use included lower baseline National Institutes of Health Stroke Scale score (aOR 0.95, 95% CI 0.93-0.97) and single vertebral artery involvement (aOR 0.63, 95% CI: 0.52-0.76). Meanwhile, independent predictors of decreased anticoagulant use included lower baseline National Institutes of Health Stroke Scale score (aOR 0.95, 95% CI: 0.93-0.97) and single vertebral artery involvement (aOR 0.63, 95% CI: 0.52-0.76). Each year following 2016 was associated with a lower rate of anticoagulation (average annualized percent change -13.91%, 95% CI: -18.73% to -9.10%). In 2021, there was a significant difference in the monthly rate of decline of anticoagulation use when compared with preceding years (P<0.001).Conclusion These real-world clinical practice data indicate a decline in the use and duration of oral anticoagulation for acute cervical artery dissection.