This news release touts an analysis of Medicare data for 1,500 patients who suffered a major bleeding event while taking anticoagulants for atrial fibrillation. Researchers found that patients who resumed anticoagulants after a hemorrhage had better stroke and mortality rates than those who did not, and those who resumed treatment with a newer drug, dabigatran, had fewer recurrent hemorrhages than those who restarted treatment with the old standby warfarin. The news release lays out relative differences in outcomes but doesn’t include absolute risks and benefits that would convey the true significance of these findings. Nor does it mention the wide cost difference between the two drugs or the fact that prescribing patterns at the time the data were collected may have skewed the findings.
Bottom line: this is an observational study, not a controlled clinical trial. It is not appropriate to be making recommendations about one drug or another based on such a study.
This study sheds light on the dilemma of whether patients at risk for blood clots should resume taking anticoagulants after a major hemorrhage. Less than half of patients who survived a bleeding incident resumed taking anticoagulant drugs, apparently because their doctors perceived a high risk of recurrent bleeding. But the authors say their finding that the risk of stroke outweighs the risk of recurrent bleeding should encourage clinicians to resume anti-clotting drugs. And while the analysis shows better outcomes with a new drug compared with an old one, the authors acknowledge that finding needs to be validated in other patient groups and with more recent data. In an observational study such as this, one must be very cautious with the conclusions.
There’s no mention of the significant cost difference between the two anticoagulants. A one-month supply for the highest recommended dosage runs about $400 for dabigatran, marketed as Pradaxa, versus $11 for warfarin, which has additional monitoring costs.
The news release does not provide sufficient data to assess risks and benefits. It states: “The risk of dying from any cause or having a stroke was 23 to 34 percent higher in patients who discontinued anticoagulation therapy compared to those who resumed it. People who took dabigatran after their initial bleeding event had nearly half the risk of another major bleeding event within one year compared with those who took warfarin.” These numbers mean little without the actual percentages of patients who experienced adverse events, which would allow readers to assess the whether there are meaningful differences between taking an anticoagulant or not and between one drug versus the other.
The news release deals with bleeding, which is the most concerning adverse effect of blood thinners. But there are other harms including joint pain, bruising and dizziness. Warfarin requires dietary restrictions and regular monitoring with blood tests.
The news release does not mention several limitations described in the study, particularly that clinicians may have been reluctant to prescribe dabigatran to patients who were most at risk for recurrent bleeding during the period that the data were collected, from 2010 to 2012. The study says that’s because of safety warnings after the drug’s approval in 2010, clinicians’ perceptions that they have more control over the coagulation status of patients on warfarin, and the lack of an antidote to reverse dabigatran’s anticoagulant effects. An antidote for dabigatran, idarucizumab, was approved in October 2015. The study (but not the release) states: “Therefore, it will be important to repeat analyses similar to ours because newer Medicare Part D data that represents the period after the approval of idarucizumab become available.”
So, again, this was not a controlled trial and these patient groups were likely different in ways that could affect the findings. No conclusions should be drawn about comparative effectiveness or harms, but the release encourages such conclusions and doesn’t caution readers about key limitations.
The news release does not engage in disease mongering.
The news release states the study’s funding sources: the Commonwealth Foundation, the Agency for Healthcare Research and Quality, the National Institute of Mental Health, and La Caixa foundation. However, it does not state that one of the authors, Samir Saba, M.D., received research support from Boston Scientific, Medtronic, and St. Jude Medical, all of which make mechanical heart valves that require the use of anticoagulants to lower the post-operative risk of complications from blood clots. Since the potential conflict here is not direct, we will give the benefit of the doubt and award a Satisfactory rating.
The release focuses on the potential for superior outcomes with a new anticoagulant compared with warfarin, the traditional medication. However, there are other new anticoagulants on the market and they were not mentioned.
The news release states that warfarin has been used as a medication since 1954 and dabigatran was approved by the U.S. Food and Drug Administration in 2010. It could have added that both are usually covered in drug plans, although the co-pay for dabigatran may make it less feasible for some patients.
The news release overstates the novelty of this study by calling it “the first analysis of how to treat patients on anticoagulants who suffer a major bleeding event.” There have been other analyses of this question, for example:
The study itself describes the novelty of the research more precisely: “To the best of our knowledge our study is the first real-world analysis comparing clinical outcomes after a major hemorrhage among patients who reinitiated anticoagulation therapy with dabigatran or warfarin and those who never resumed anticoagulations.” They say the analysis “contributes significantly to the existing literature” by comparing patients who used different anticoagulant drugs after a hemorrhage.
Given that many people read only the headline, we think this one should have been far more nuanced. “Dabigatran superior to warfarin when anticoagulation is resumed after bleeding” should be qualified with the fact that this was the finding of one study, which needs to be repeated and confirmed with other data. The results should have been presented as very preliminary, but they were described in much more definitive terms.