FDA: Chantix not associated with psychiatric hospitalizations

The smoking cessation drug varenicline (Chantix) got a black-box warning in 2009 after the FDA received >500 reports of suicidal thoughts and behaviors, and 32 completed suicides in the U.S. (This was out of several million prescriptions filled, though.)

After that postmarketing surprise, the FDA sponsored 2 observational studies in Chantix users: one at the VA (n=28,000+) and the other through the Department of Defense's database systems (n=35,000+). They were both retrospective cohort studies that used propensity-matching to compare Chantix users to similar users of nicotine replacement therapy. The outcome was psychiatric hospitalization within 30 days of starting varenicline (most adverse events have been reported soon after Chantix initiation) or nicotine replacement.

Out of more than 60,000 patients observed, there were only a total of 71 psychiatric hospitalizations that occurred in the 30-day window after therapy initiation, over more than one year of observation. Psychiatric hospitalization was not more common in the varenicline users than in the nicotine replacement users.

One limitation of the data was not including PTSD hospitalizations as psychiatric hospitalizations in the VA study -- it wasn't reported how many of these there were, or whether that would have changed the conclusions.

The FDA's conclusion and recommendation was to continue to follow the existing black box warning, unchanged:

People who are taking Chantix or Zyban and experience any serious and unusual changes in mood or behavior or who feel like hurting themselves or someone else should stop taking the medicine and call their healthcare professional right away.

Friends or family members who notice these changes in behavior in someone who is taking Chantix or Zyban for smoking cessation should tell the person their concerns and recommend that he or she stop taking the drug and call a healthcare professional right away.

Chantix had more bad press after this study suggesting an increased risk of cardiovascular events -- although some criticized the methodology and others argue the benefits of reduced cardiovascular risk from quitting smoking outweigh the drug's small (if any) heart risks.

Varenicline sure does work, with sustained quit rates ~20-30+% described in this NEJM review, and other articles PulmCCM.org recently reviewed (here and here).

FDA Drug Safety Communication: Safety review update of Chantix (varenicline) and risk of neuropsychiatric adverse events. (FDA.gov)