Abstract 853: Short-Term Elevations in Blood Pressure Levels in Arthritis Patients Receiving Diclofenac or Etoricoxib in the MEDAL Program was not Associated with Subsequent Increased Risk of Thrombotic Cardiovascular Events
It has been postulated that the increased thrombotic cardiovascular (CV) risk observed in trials of COX-2 inhibitors vs. placebo, and the apparent similar risk with traditional NSAIDs, may be related to their potential to elevate BP. We evaluated the effect of baseline and change in BP on CV events in patients receiving an NSAID or a COX-2 in the MEDAL Program. Patients (mean age 63 years) randomized to etoricoxib 60 or 90 mg (n=17,412) or diclofenac 150 mg (n=17,289) daily for a mean duration of 18 months. The relationship of baseline BP and change in BP from baseline to Month 4 to the study primary endpoint (confirmed thrombotic CV events ([CVEs]) was examined using Cox proportional hazards model. To test consistency, CV death/MI/stroke, all-cause and CV/CHF mortality, and CHF subsequent to month 4 were also evaluated. The relationship of CVEs subsequent to Months 4, 8 and 12 to BP change ≥ vs < 10 mm Hg at those timepoints was explored using Poisson regression model. Baseline SBP was associated with higher risk of all events (Table⇓). Baseline DBP was inversely and associated with risk of all events except CV/CHF mortality. Risk ratios (based on event rates per patient year of follow-up) for CVEs at Month 4, 8 and 12 for change in SBP ≥ vs < 10 mm Hg were 1.07, 1.07 and 0.88 respectively. There was no meaningful interaction between change in SBP with treatment with etoricoxib or diclofenac in relation to CVEs. CHF risk, but not thrombotic endpoints, was associated with change in BP from months 0 to 4. Baseline SBP, but not short-term change in SBP, was significantly associated with risk of thrombotic CV events. Change in SBP on treatment was associated only with CHF. Results from this program, with mean and maximum treatment duration of 18 and 42 months, respectively, suggest that the increase vs. placebo in major CV events observed with COX-2s in prospective RCTs may be due primarily to causes other than the just the BP-elevating effects of these agents.