New Oral Anticoagulants And The Risk Of Vasculitis: A Nested Case-Control Study


      To estimate the risk of vasculitis associated with new oral anticoagulants (apixaban, dabigatran and rivaroxaban) as compared to warfarin.


      We conducted a nested case-control study using a 10% random sample of IMS LifeLink data (2011-2015). We constructed a nest of patients who had a primary or secondary ICD-9 diagnosis for nonvalvular Atrial Fibrillation in an inpatient or outpatient setting, followed by prescriptions for warfarin or any new oral anticoagulants (NOACs). We designated the date of first prescription as nest entry date. We required our nest of patients to be continuously enrolled for at least 6 months before the nest entry date to include only incident users. We identified cases using primary or secondary ICD-9 diagnosis for primary systemic vasculitis or leukocytoclastic vasculitis. Diagnosis date was designated as event date. We used incidence density sampling to select controls for each case, matched on age, gender and nest entry date. We used 1:n variable matching without replacement technique to select controls. We looked back 90 days from the event date to check for prescriptions for NOACs (exposed) or warfarin (unexposed). We did a sensitivity analysis by restricting the look-back period to 30 days. We calculated both unadjusted and adjusted odds ratio (OR) using conditional logistic regression. Adjustment was done for CHA 2DS2-VASc score.


      Our nest consisted of 13,295 patients. 80 cases were identified, which were matched to 7,605 controls, resulting in a total of 7,685 patients in our final sample. Unadjusted OR was calculated to be 0.91 (95% CI: 0.52-1.58). After adjustment for CHA 2DS2-VASc score, OR was found to be 0.90 (95% CI: 0.52- 1.57). Our estimates did not change in the sensitivity analysis: unadjusted OR (0.91; 95% CI: 0.52-1.58) and adjusted OR (0.90; 95% CI: 0.52-1.57).


      NOACs were not associated with an increased risk of vasculitis when compared to warfarin.