High- vs. low-dose diclofenac and cardiovascular risks: a target trial emulation.

Abstract

To examine the dose dependency of diclofenac's cardiovascular risks. Using Danish health registries and the target trial emulation design, we conducted a series of 300 nationwide cohort studies during 1996-2020, each mimicking the strict design criteria of a clinical trial. Adults eligible for inclusion had no recent non-steroidal anti-inflammatory drug prescriptions, contraindications (gastrointestinal diseases, thrombocytopenia, or heart failure), or conditions with low adherence (dementia or psychiatric disease). Diclofenac initiators were compared to healthcare-seeking non-initiators and head-to-head using an approximated high dose of ≥150mg/day vs. low dose of <150mg/day. Cox regression was used to compute the incidence rate ratio (IRR) of major adverse cardiovascular events (MACE) within 30 days following initiation. We adjusted for age, sex, calendar period, comorbidity, comedication, and socioeconomic position. Compared with non-initiators (n=3789 617), diclofenac initiators (n=1 894 834) had an approximately 50% increased rate of MACE (IRR 1.53, 95% confidence interval [CI]: 1.43-1.63), reflecting IRRs of 1.54 (95% CI: 1.40-1.69) for myocardial infarction, 1.29 (1.14-1.45) for ischaemic stroke, and 1.92 (1.71-2.16) for cardiac death. The risk increase was observed for most conditions with chronic pain, in particular headache (IRR 5.10, 95% CI: 1.46-17.85). The risk increase was similar for initiators of high- (IRR 1.55, 95% CI: 1.40-1.71) and low-dose diclofenac (IRR 1.52, 1.41-1.63), which was confirmed in a head-to-head analysis (IRR 1.01, 95% CI: 0.90-1.12). Initiators of high- and low-dose diclofenac had comparably increased cardiovascular risks. This finding provides evidence against the assumption that low-dose diclofenac is risk-neutral.