Dose-dependency of diclofenac's cardiovascular risks: a series of nationwide emulated trials

Abstract Background The dose dependency of the adverse effects of diclofenac remains poorly understood. Purpose To examine the dose-related cardiovascular risks associated with diclofenac initiation Methods We used Danish nationwide health registries (1999–2018) to conduct a series of emulated trials (n=285). Eligible adults had no recent NSAID fillings, contraindications, or conditions with low adherence. Individuals eligible for inclusion were ≥18 years with (1) ≥90 days continuous prescription records prior to diclofenac initiation (baseline); (2) no NSAID prescriptions ≤90 days before enrollment, and (3) no exclusion criteria. Exclusion criteria reflected likelihood of low adherence to treatment (dementia, schizophrenia, or antipsychotic drug use) and labeled contraindications (ulcer disease/anti-ulcer drugs, gastrointestinal bleeding, inflammatory bowel disease, thrombocytopenia, or heart failure). Initiators of diclofenac were compared to healthcare-seeking non-initiators, but also head-to-head for initiators of high (≥75 mg pills as proxy for ≥150 mg/daily) vs. low dose (≤50 mg pills as proxy for <150 mg/daily). Cox proportional-hazards regression was used to compute the intention-to-treat hazard ratio, as measure of the incidence rate ratio (IRR), of major adverse cardiac and cerebrovascular events (MACCE) within 30 days from initiation. Results Among 3,177,484 diclofenac initiatiors, 31% used high and 69% used low dose. Compared with non-initiators, diclofenac initiatiors had a 70% increased rate for MACCE (IRR 1.70, 95% CI: 1.55–1.86), reflecting for the individual MACCE components an increased IRRs of 1.66 (95% CI: 1.54–1.79) for myocardial infaction, 1.32 (1.20–1.45) for ischemic stroke, and 1.69 (1.54–1.86) for cardiac death. The effect for MACCE did not differ between initiators of high (IRR 1.73, 95% CI: 1.51–1.97) and low dose (IRR 1.68, 95% CI: 1.52–1.86) (Figure 1). When comparing high and low dose diclofenac head-to-head, we found no meaningful difference in the IRR for MACCE (1.03, 95% CI: 0.89–1.19), MI (0.99, 0.87–1.11), ischemic stroke (0.95, 0.81–1.11) or cardiac death (1.04, 0.90–1.21) (Figure 2). Conclusion Initiation of low- and high-dose diclofenac was associated with a consistent and comparable increase in cardiovascular risk. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): Novo Nordisk Foundation