. . . that the FDA approved this drug despite the lack of an antidote. So says Dr. Mark Mosley, an ED physician in Wichita, KS, in a New York Times article published yesterday (and riding high in their most-emailed list). The article’s thrust is that many people have suffered bleeding complications on the new anticoagulant and some ED docs and trauma surgeons feel that the safety profile in real life may not be as favorable as was seen in the pivotal RE-LY study. They interview the brother of a patient who died of hemorrhagic complications; he’s quoted as saying “My No.1 goal is to stop this insidious drug . . . ” Annals of Internal Medicine recently published an opinion piece that was less breathless but in the same vein.
As we discussed at grand rounds yesterday, our stroke group is not an early adopter of dabigatran. Just so, however, we shouldn’t too quickly pile on the criticism of the drug. Unlike the many “me-too” drugs that come on the market, this one’s appearance is a bona fide medical advance based on its superiority in preventing stroke and its easy dosing and lack of need for monitoring.
The safety concerns stem from a flurry of adverse event reports to the FDA. However, it’s important to consider that when an ED doctor or trauma surgeon encounters a dabigatran-associated hemorrhage, he/she may be more likely to report this potentially important new information to the FDA as compared to when he encounters a warfarin-associated hemorrhage, which is something everyone has seen many times before. In other words, passive surveillance for drug safety concerns may be biased against new drugs. To address this, the FDA has set up an active surveillance system called Mini-Sentinel. Using Mini-Sentinel’s health insurance and administrative claims data, the FDA found the following, which they published in a Drug Safety Communication yesterday:
For the populations in the Mini-Sentinel data assessment, the combined incidence rate (ICH and GIH events per 100,000 days at risk) was 1.8 to 2.6 times higher for new users of warfarin than for new users of Pradaxa. The incidence rate of GIH events only per 100,000 days at risk was 1.6 to 2.2 times higher for warfarin new users than for Pradaxa new users, and the incidence rate of ICH events only per 100,000 days at risk was 2.1 to 3.0 times higher with warfarin than with Pradaxa. The results indicate that the observed bleeding rates associated with new use of Pradaxa do not appear to be higher than the bleeding rates associated with new use of warfarin.
Emphasis added. There are caveats to this reassurance: We don’t know that the populations surveilled in Mini-Sentinel are comparable to the overall population of patients exposed to dabigatran. Also, as anyone who’s received one of those annoying coding queries is well-aware, there are problems with the accuracy of the medical coding that forms the basis for queries of databases such as Mini-Sentinel. Nevertheless, I value this information much higher than the sensationalist anecdotes of the New York Times.