European drugs agency withdraws antiobesity drugBMJ 2008; 337 doi: https://doi.org/10.1136/bmj.a2301 (Published 29 October 2008) Cite this as: BMJ 2008;337:a2301
All rapid responses
Should Rimonabant stay in the market?
Note: Letter sent to Lancet at 03 December 2007, and not published
The meta-analysis by R. Christensen et. al., Lancet, Nov. 17 2007,
shows that Rimonabant caused significantly more adverse events than
placebo: number needed to harm (NNH) 25, and more serious adverse events:
NNH 59 with a relative risk increase of 40% or 1.4. Patients were also
more likely to discontinue the treatment because of depressive mood
disorders: NNH 49. The FDA in USA found that 26% of patients who took
Rimonabant had some kind of adverse psychiatric event versus 14% with
placebo (NNH 9).
At the other hand, in the RIO-Lipids Study the criteria for diagnosis
of the metabolic syndrome, a significant clinical endpoint, fell to 25.8%
with Rimonabant versus 41% with placebo, a number needed to treat (NNT) of
If the studies with the most significant clinical endpoints show a
relative risk reduction of 30% or more (not less than 20%), an absolute
risk reduction of 3% or more (not less than 2%), and a NNT for benefit of
30 or less (not over 50), what I call the "Rule 3-30 of Evidence Based
Medicine", and, at the other hand, if the relative risk increase of
adverse drug reactions if over 30% or 1.3, and the NNH for serious adverse
events is less than 300 for a drug with non vital indications as in the
case of Rimonabant, and more so if the NNH is so close to the NNT (note:
even for vital indications, the NNH for serious adverse events should be
not less than the NNT and rather at least 3 to 10 times over the NNT), and
since other much safer methods (diet, exercise, lifestyle), and at least
relatively safer and better known drugs for this indications exist, could
it be the time to reevaluate if Rimonabant stays in the market?.
Prof. Enrique Sánchez-Delgado, M.D.
Director of Medical Education
Hospital Metropolitano Vivian Pellas
Competing interests: No competing interests