Background Pharmacological treatment of ostheoartritis (OA) usually include analgesics, non-steroidal anti-inflammatory drugs (NSAID) and symptomatic slow-acting drugs in OA (SYSADOA). The association between these groups of drugs and the risk of ischaemic stroke has not been properly addressed.

Objectives To analyse the risk of stroke in patients using analgesics, NSAID and SYSADOA drugs.

Methods We used a population-based patient hospital registry to identify all patients with a first-ever stroke discharge diagnosis between 2009–2015. Cases were matched to controls obtained from the Information System for Research in Primary Care (SIDIAP) database. Information on drug exposure was obtained from invoice data from pharmacies. Crude and adjusted odds ratios (OR, OR_adj) and their 95% confidence interval (95% CI) were calculated using multivariate models of conditional logistic regression for the next pharmacological groups and individual agents of each group: acetic acid derivatives, oxicams, propionic acid and derivatives, coxibs, SYSADOA and analgesics (opiods, metamizole and paracetamol). A cardiovascular risk score was calculated for each subject based on comorbidities.

Results 12 616 cases were matched to 1 25 264 controls by gender, age and geographic area. Among cases, 43% were women. The mean age was 72.6 (IQR 65–82) years and more cases were classified as high cardiovascular risk patients (n=2,511, 19.9%) than controls (n=12,467, 10.0%). Mortality in the following year after the index date was higher for cases (n=2,633, 20.9%) than for controls (n=8,168, 6.5%). A higher percentage of cases had a previous diagnostic of ischaemic heart disease, 13.8% (n=1,745) vs 7.7% (n=9,656) of the controls. OA diagnosis was present in 2823 (22.4%) cases and in 29 098 (23.2%) controls. Paracetamol was the most used drug (n=106,515, 77.3%) followed by ibuprofen (n=84,790, 61.5%).

Current users of acetic acid derivatives showed an increased risk of stroke [OR_adj, (95% CI) 1.10 (1.01–1.19)], so did diclofenac [OR_adj1.14 (1.04–1.25)]. Among the propionic acid derivatives, current users showed an increased risk [OR_adj1.24 (1.27–1.32) ], dexketoprofen showed the highest risk [OR_adj1.42 (1.25–1.60)] and naproxen the lowest [OR_adj1.18 (1.02–1.37)]. The SYSADOA group did not show any increased risk for any type of exposure, with a decreased risk of 17% [OR_adj0.83 (0.77–0.91)] for lifetime glucosamine exposure and a 22% [OR_adj0.84 (0.78–0.90)] for chondroitin sulfate. Analgesics were the most consumed drugs, and cases were more exposed to all subgroups of analgesics than controls. While the non-adjusted OR showed an increased risk of stroke for lifetime exposure for all agents in this group, this association was not observed with the adjusted OR_adj for fentanyl, neither buprenorphine, but it showed a risk for current users of metamizole [OR_adj of 1.67 (1.56–1.80)], tramadol [OR_adj1.15 (1.06–1.24)] and paracetamol [OR_adj1.43 (1.37–1.51)].

Conclusions Current exposure to NSAIDs, tramadol, metamizole and paracetamol is a risk factor for ischaemic stroke. Exposure to chondroitin sulphate and glucosamine are associated with a lower risk of ischaemic stroke.

Disclosure of Interest None declared