Risk of cardiovascular diseases and gastrointestinal bleeding is comparable between celecoxib and non-selective non-steroidal anti-inflammatory drugs in patients with ankylosing spondylitis: a nationwide retrospective cohort study

OBJECTIVE: To compare the risk of cardiovascular disease (CVD) and gastrointestinal bleeding (GIB) between celecoxib and non-selective non-steroidal anti-inflammatory drugs (nsNSAIDs) in patients with ankylosing spondylitis (AS). METHOD: In this nationwide retrospective cohort study using the Korean Health Insurance Review and Assessment database, adult AS patients who received newly prescribed non-steroidal anti-inflammatory drugs (NSAIDs) continuously for ≥ 30 days (celecoxib or nsNSAIDs) between 2013 and 2017 were evaluated. The co-primary outcomes were the occurrence of composite CVD events, including hospitalization for myocardial infarction, ischaemic heart disease, stroke, transient ischaemic attack, heart failure, and coronary revascularization; and composite GIB, including hospitalization for upper and lower GIB. Propensity score (PS) matching was used to correct for baseline differences between the celecoxib- and nsNSAID-treated groups. RESULTS: We identified 3164 celecoxib-treated and 18924 nsNSAID-treated patients with AS. After 1:1 PS matching, 3047 patients with AS were assigned to each of the celecoxib- and nsNSAID-treated groups. The incidence of composite CVD and GIB was 18.2/1000 person-years and 6.5/1000 person-years in celecoxib-treated and 15.1/1000 person-years and 7.3/1000 person-years in nsNSAID-treated patients, respectively. Compared to the nsNSAID-treated group, the hazard ratios of composite CVD and GIB in the celecoxib-treated group were not significant, with values of 1.17 (p = 0.499) and 0.87 (p = 0.696), respectively. There were no significant differences in the risk of each component of the composite CVD and GIB between the two groups. CONCLUSION: We did not find significant differences in the risks of CVD and GIB between celecoxib and nsNSAIDs in AS patients.