Rimonabant is a selective cannabinoid-1 receptor antagonist that has previously been shown to reduce waist circumference and body weight, in addition to improving metabolic factors such as highdensity lipoprotein-cholesterol, adiponectin, triglyceride, insulin and leptin levels. However, no previous trial has been adequately powered to assess whether this can lead to improved clinical outcomes.

The CRESCENDO (Comprehensive Rimonabant Evaluation Study of Cardiovascular Endpoints and Outcomes) trial was a double-blind, placebo-controlled trial that randomised 18 695 patients with previously manifest, or with an increased risk for, cardiovascular disease to either rimonabant 20 mg daily or matching placebo. The primary end point was the composite of cardiovascular death, myocardial infarction, or stroke.

At a mean follow-up of 13.8 months the trial was prematurely discontinued owing to concerns from three countries about suicide in patients taking the drug. At the close of the trial, no difference was seen in the composite primary end-point incidence (3.9% vs 4.0%; p¼0.68; see figure 1); however, gastrointestinal, neuropsychiatric and serious psychiatric side effects were all significantly increased with rimonabant.

Conclusion
This randomised controlled trial seeking to investigate the cardiovascular effects of rimonabant was halted early owing to an increased risk of serious psychiatric events figure 1.

• Topol EJ, Bousser M-G, Fox KAA, et al Rimonabant for prevention of cardiovascular events (CRESCENDO): a randomised, multicentre, placebo-controlled trial. Lancet 2010;376:517e23.