Christopher P. Cannon, MD, Brigham and Women's Hospital, Boston presented the results of the cardiovascular outcomes Multinational Etoricoxib and Diclofenac Arthritis Long-term (MEDAL) program. The MEDAL program consisted of pooled data from three trials conducted at 1,380 sites in 46 countries: the EDGE trial (osteoarthritis, 7,111 patients), the EDGE II trial (rheumatoid arthritis, 4,086 patients), and the MEDAL trial (osteoarthritis and rheumatoid arthritis, 23,504 patients). The goal of this program was to determine whether cardiovascular event rates were similar in patients treated daily with a COX-2 inhibitor (etoricoxib) compared with those treated with a widely used traditional NSAID (diclofenac). Patients over 50 years of age with a diagnosis of either rheumatoid arthritis (RA) or osteoarthritis (OA) of the hand, hip, knee, or spine were eligible for participation. Patients meeting all eligibility criteria were randomized either to etoricoxib (60 or 90 mg/day for OA or 90 mg/day for RA) or diclofenac (150mg/day).

A total of 34,701 patients were enrolled in the program, 17,412 in the etoricoxib group and 17,289 in the diclofenac group. The mean (SD) duration of therapy was 18.2 months (11.7) for etoricoxib and 17.7 (11.9) months for diclofenac. The demographic characteristics were similar between the two treatment groups. “Over the three year period that the patients were followed, there was no difference over time in the risk of [cardiac] events with these two different agents” said Dr. Cannon. The primary outcome measure of thrombotic cardiovascular event rates had a hazard ratio (HR) of 0.95 [95% CI, 0.81–1.11].

Subgroup analyses were also performed and revealed no differences between treatments when factors such as age, gender, diabetes, established atherosclerotic cardiovascular disease (ASCVD), established ASCVD or ≥ 2 risk factors, low-dose aspirin usage, type of arthritis, or etoricoxib dose were examined. Rates of upper gastrointestinal tract events were significantly lower with etoricoxib (HR=0.69; [95% CI, 0.57–0.83]).

“Observational studies may lead us astray” in making therapeutic decisions, said Dr. Cannon, emphasizing the importance of controlled, randomized trials such as those in the MEDAL program. Other questions regarding cardiovascular events and the use of these medications remain unanswered, as this study had only one comparator and many others agents are routinely utilized. The choice of diclofenac as the comparator has also been criticized. “The investigators are justified in saying diclofenac is the most widely used traditional NSAID on the market worldwide and therefore it's a worthy competitor in a head-to-head comparison”, said Robert M. Califf MD of Duke University in his discussion of the trial data. He noted, however, that “this leaves open whether naproxen is actually cardio protective and would have come out significantly better than the COX-2 specific drug. It also leaves open the 10–20 other potential comparisons one can make.” For now, clinicians must continue to tailor arthritis pain pharmacotherapy after considering the needs, history, and risks of each individual patient. The results of this study were published online in The Lancet on November 13, 2006 (Cannon et al, www.thelancet.com; DOI: 10.1016/S0140-6736(06)69666–9).