Chantix: Less Smoking, More Heart Attacks

Stop smoking? Absolutely. Easy as swallowing a pill? Think again.

A study just published in the Canadian Medical Association Journal by my colleague Dr. Sonal Singh indicates that Chantix can lead to "increased risk of a major harmful cardiovascular event" by 72% compared to placebo.

"People want to quit smoking to reduce the risk of cardiovascular disease but in this case they're taking a drug that increases the risk for the very problems they're trying to avoid," says Sonal Singh, M.D., M.P.H., an assistant professor of general internal medicine at the Johns Hopkins University School of Medicine and the lead author of the research.

How did they do the study?

Singh and his colleagues reviewed and analyzed 14 double-blind, randomized, controlled clinical trials involving more than 8,200 healthy people who received either varenicline (made by Pfizer and sold in the United States under the brand-name Chantix) or a placebo.

Were there warning signs about the heart risks? Yes:

Singh says questions about the drug's cardiovascular disease risks have been raised since varenicline went on the market in 2006, but no study has clarified the magnitude of these risks to the extent found in the new study. Singh says the FDA used a "fast-track" review process in allowing varenicline to be sold in the United States and would like regulators to take a new look.

Bottom line: Anyone wanting to quit smoking should approach treatments such as Chantix with considerable caution, discuss it with their physician, and consider safer alternatives.

Here is a brief video from coverage on CNN:

Despite claims by Pfizer, Dr. Singh and his colleagues did not "do the math wrong"!

Here's their data; any readers, especially statisticians, are welcome to decide for themselves.

Added July 7:

NOTE NBC Interview with Dr. Singh, July 6, 2011:


More like this

"healthy people"?

Isn't the real question whether any decrease in smoking brought about by the drug decreases cardiovascular *and other serious health risks* more than the drug increases them?

By DrugMonkey (not verified) on 06 Jul 2011 #permalink

Bottom line: In this meta study, patients taking Chantix for up to 52 weeks showed overall a 72% increase in cardiovascular events compared to placebo. Less smoking, yes, along with this risk.

Hmm. I had a look at the largest study in the table ((Tonstad, citation 26): ~600 patients in each wing, 4 CV events for Chantix, 0 CV events for placebo.)

Singh said they were only counting major cardiovascular events; ones which required hospitalization, such as heart attacks. Yet the results section in Tonstad states:
"Adverse events reported in the open-label period were mostly mild; no difference in adverse events between varenicline and placebo was observed during the double-blind period."

Either Tonstad or Singh is fibbing. That or there were so many mild adverse events that Tonstad didn't consider the major ones as worth mentioning. Unfortunately JAMA's link to the full version of the Tonstad paper keeps going to an error page, so I can't tell which.

Chantix is a very, VERY, dangerous drug. I took two scripts in '07 back/back, two hospital stays in '08. Please, believe me, this drug is for real. I, too, would be a doubter, had I experienced the side-effects first hand.

One hospital stay was for the heart attack symptoms.

One hospital stay was for serious suicide attempt.

It's TRUE, it DOES make smoking a chore and tasteless. And, that would be FINE...if it STOPPED there. But, it doesn't...the process of this drug snaking it's way thru your brain and attaching to receptors not even previously THOUGHT sneaky, slow, THOROUGH and all consuming!! TRULY, your thoughts, feelings, reactions, become distorted and weird. To have someone "watch you" and/or "watch yourself for any changes" is beyond laughable. Because, you don't even know yourSELF that something strange is happening.

Feared drugs as a teen...can describe first-hand at 55, in VIVID detail...hallucinations. Like everyone else who went to the trouble and expense to GET this prescription...I had HIGH hopes, delirously happy AND FINALLY put this stupidity to rest!

Almost put myself to rest instead.


"...had I NOT experienced the side-effects first hand."

On second thought (after posting #3:) Tonstad et al might have written "Adverse events reported in the open-label period were mostly mild; no difference in adverse events between varenicline and placebo was observed during the double-blind period." because the major adverse events didn't start happening until after the second 12 weeks (the double blind period) of varenicline dosing. I.e., during the follow up period. In which case they were just misdirecting/being skeevy, not actually fibbing.

Impugning the character of the authors of this study is not warranted or appropriate. After all, the idea of double-blind randomized controlled trials is to avoid any possibility of bias on the part of the researchers!

You did not find 4 events in the JAMA paper because these serious events were never reported in the JAMA paper, an not uncommon practice !

The study has an appendix that describes these 4 events and source of data.

Here is the link to the Pfizers study version of Tonstad with these 4 events during the double blind period of the trial- MI, MI, SVT, cerebral thrombosis in the Double Blind period of Page 9 occuring in Chantix arm. No CV events occur in placebo

In the overall study, The absolute difference is small 1.06 vs 0.82 or a difference of 0.24% but this due to Simpsons paradox. The relate risk of 72% is based on the relative risk in each study ( in some it was higher than 72% and some lower) and weighted for the study size ( larger size studies getting more wieght) in the meta-analysis.

One can argue about the relative and absolute effects but not about the fact that safety data should always be available to physicians and patients.

Thankfully the FDA has labelled Chantix with a warning for increased CV risk.

By Anonymous (not verified) on 06 Jul 2011 #permalink

In spite of the comments of the author of this post, it wasn't just Pfizer who criticized the analytical methods used by Singh et al. The comments on the article on the CMAJ website are from a diverse set of medical professionals, and are among the most blistering I have seen.

How does an article exposing excess cardiovascular risk in a widely used drug get published in CMAJ instead of a more prestigious journal? I would have to guess that the referees at NEJM and JAMA rejected the author's analysis.