Dronedarone and Clinical Guidelines: A Dark Chapter That Must Be Reread

John Mandrola


February 18, 2014

It's tough to write about dronedarone (Multaq, Sanofi). It's really not sportsmanlike to pile on. We all make mistakes; we all fall down. This science and medicine thing is a humbling exercise indeed.

What follows, therefore, is less a story about a flawed drug for atrial fibrillation than a tale of good intentions gone wrong. To be clear from the outset, dronedarone, if graded on the curve of existing AF drugs, looks only slightly awful.

The point of revisiting the sordid saga of how dronedarone rose to prominence in clinical guidelines tells a much larger story about an emerging patient safety issue—namely, the metastasis of the clinical guideline.

The irony, of course, is that expert guidelines were set out to help busy practitioners provide the most evidence-based care for their patients. Guidelines were supposed to offer guidance, but now they have morphed into de facto rules upon which quality officers determine good doctoring and malpractice juries determine right from wrong. As if the practice of medicine could ever be that binary. Hubris is a bite that stings—especially when it comes to medical care.

So I ask: What if said "rules" are flawed? What if these supposedly unbiased expert assessments of the evidence were not so unbiased and not so expert?

Could there be a more fundamental problem?

This week, in a special communication to JAMA Internal Medicine, a group of Italian investigators performed a systematic review of the evidence base for dronedarone and came to the conclusion that if a drug as dubious as dronedarone could find its way into clinical guidelines there must be a major problem with the entire process--a reliability problem[1].

The specifics of the Italian investigation were simple: first, they applied basic rules for interpreting clinical science to the evidence base for dronedarone, and second, they used criteria set out by the Institute of Medicine (IoM) to assess the quality of the AF treatment guidelines pertaining to the use of dronedarone[2]. In a sense, the researchers graded the graders.

What they found was striking. First . . . the assessment of dronedarone:

  • In five trials that measured mortality, there were 13 more deaths per 1000 patients in the dronedarone-treated group compared with placebo.

  • In three trials that measured adverse effects, there were 46 per 1000 more events in the dronedarone group compared with placebo.

  • In the one major trial that compared dronedarone's ability to suppress AF against amiodarone, there were 214 more AF recurrences per 1000 patients in the dronedarone group.

  • In the ATHENA trial[3], the only "positive" dronedarone trial, the benefit seen in the composite primary end point was driven by the soft surrogate, lower hospital admissions.

Second . . . grading the quality of the AF treatment guidelines (from table 2 of the JAMA-IM paper):

The IoM sets out eight quality domains for clinical guidelines. The researchers found that the US AF treatment guidelines[4] were lacking in five of the eight, including faulty management of conflict of interest, inadequate composition of the writing group, lack of systematic reviews, insufficient grading of the strength of recommendations, and only partial use of external reviews.

This led the authors to conclude that current recommendations in favor of dronedarone were not justified. More telling was this:

"O  ur study raises many concerns about these guidelines, including the rigor with which the available clinical evidence was reviewed and evaluated and the extensive financial associations between many members of the guidelines panels and the manufacturer of dronedarone."


We should congratulate this brave group of researchers. Then we should ask why this special communication wasn't published in a cardiology journal.

The story of how a marginally effective and perhaps even hazardous drug made it to first-line status of major guideline documents is a hard image to revisit. Also difficult to look back at were the lavish steakhouse dinners, the webinars, the "sponsored" programs, and then the result: legions of general cardiologists and primary-care doctors prescribing dronedarone en masse. With dronedarone, for a time at least, hype trounced science.

The one great thing about mistakes is how much they inform. History is always that way.

Most important here is not whether dronedarone should be removed from the market. In reality, it's just another cardiac ion-channel blocker that, in very small numbers of cases, may transiently suppress paroxysmal AF, and, like other antiarrhythmic drugs, if used in the wrong patient, increase the risk of death.

The really important message here concerns the dangers of letting flawed guidelines metastasize. The findings of this report should provide a jarring wake-up call to all those who magically believe that human-derived guidelines in their current form could ever equate to all that should be.

In 2009, DrAllen Sniderman (McGill University, Montreal, Canada) and Dr Curt Furberg (Wake Forest University, Winston-Salem, North Carolina) wrote in the Journal of the American Medical Association that guideline-making needed reform[5].They offered seven commonsense suggestions to improve the current morass. These included simple but obvious things such as bringing epidemiologists and statisticians into the fold, not publishing recommendations as unanimous unless they were unanimous, posting a preliminary version of the statement online for comments, providing independent scientific review, providing full and complete financial disclosures, and insisting on stricter rules governing conflict of interest for both individuals and medical societies.

It seems no one listened. It's time now. Better late than never.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.