San Bruno, CA - Concerns over the safety of aprotinin (Trasylol, Bayer) will be fueled by a new analysis published this week suggesting that the drug, used to stop bleeding in cardiac surgery, is associated with an increased risk of long-term mortality [1].

The analysis, published in the February 7, 2007 issue of the Journal of the American Medical Association, was conducted by a team led by Dr Dennis Mangano (Ischemia Research and Education Foundation, San Bruno, CA). Mangano's first study in this field, published in the New England Journal of Medicine (NEJM) last year, showed aprotinin to be associated with serious end-organ damage, particularly renal failure, in the perioperative period [2]. But an FDA advisory committee almost completely discounted that study because of queries over the methodology and claims that the FDA had not been able to independently validate the data. However, just days after the FDA advisory committee meeting, another study conducted by Bayer emerged that showed similar adverse events with aprotinin and seemed to validate Mangano's results.

Mangano's new analysis is based on the same set of patients as the first study, but whereas the first study addressed perioperative outcomes, this new analysis followed patients up to five years after surgery and found an increased risk of death in those given aprotinin.

Mangano commented to heartwire: "The findings in this second study raise more serious issues than in the first study—it shows a substantial increase in mortality with aprotinin, in the hospital period, in the first year, and out to five years. The other lysine analogs available that limit blood loss during surgery had similar long-term mortality to controls, so they appear to be safe. But aprotinin does not appear to be safe."

He says these new results should finally convince surgeons not to use aprotinin apart from in exceptional circumstances. "I would use one of the other two drugs if I were concerned about bleeding. If bleeding were still a problem after one of the other agents, then I would consider aprotinin. But this is just a very small group of patients—about 5%. For 95% of patients undergoing CABG, it is not appropriate to use aprotinin, in my view," Mangano told heartwire.

But still, not everyone is convinced that aprotinin causes harm. The author of an accompanying editorial [3], Dr Bruce Ferguson (Brody School of Medicine at East Carolina University, Greenville, NC), commented to heartwire: "While Mangano's studies do bring new and important information about aprotinin to the table, I don't think they demonstrate conclusively that aprotinin is harmful, as there are still questions about the methodology used, and until these are addressed it is impossible to say that no one should use aprotinin anymore."

In the paper, Mangano et al report that this observational study, conducted between 1996 and 2006, prospectively assessed survival at six weeks, six months, and annually for five years after CABG surgery among 3876 patients enrolled from 62 centers. The associations of survival with hemorrhage-sparing medications—aprotinin, aminocaproic acid, and tranexamic acid—or no antibleeding agent were compared with multivariable analyses, including propensity adjustments.

Results showed that during five years of follow-up, the death rate in the aprotinin-treated patients was nearly two thirds greater than in patients who received no antibleeding agent (controls), Mangano et al report. Covariate-adjusted survival analyses demonstrated significant association with death for aprotinin, but not for aminocaproic or tranexamic acid.

The authors add that in multivariable logistic regression, either with propensity adjustment or without, aprotinin was independently predictive of five-year mortality among patients with diverse risk profiles, as well as among those surviving their index hospitalization.

"These observations suggest that the deleterious perioperative safety findings previously reported for aprotinin are not self-limited, as has been suggested, but that their consequences likely continue over months to years following administration," they write. They conclude: "Use of aprotinin among patients undergoing CABG surgery does not appear prudent because safer and less-expensive alternatives (ie, aminocaproic acid and tranexamic acid) are available."

They estimate that over the past year, aprotinin was prescribed worldwide to at least 200 000 cardiac-surgery patients having a profile similar to patients in this study, and they add: "Thus, in 2006 alone, had aprotinin been replaced with either aminocaproic acid or tranexamic acid, we estimate that approximately 2000 deaths per year for the next five years (or 10 000 total deaths) might have been avoided."

In his editorial, Ferguson points out that there are still some uncertainties about Mangano's data. These include the fact that the reasons that patients received any specific agent (or no agent) are not clear. He also points out that aprotinin is generally used in higher-risk patients and that in Mangano's study more patients who received aprotinin underwent complex surgery than in the other groups. "Importantly, these biases, at the level of the surgical team, were not captured in the extensive patient-level data collection process nor in the analysis," he writes.

To heartwire, Ferguson commented: "We don't know for sure that aprotinin caused the worse outcomes, as these were selected patients and there is no way to assess why they got the drug. Okay, statistical techniques have been used to try to account for confounding factors, but you can't account for confounders that you don't know about. We don't know why aprotinin was being used in these patients, we don't know whether it was administered correctly, if it was given for a labeled indication or not. Since we don't know the answers to all these questions, there could be many other mitigating factors that could have contributed to the increase in mortality."

But Mangano argues that because he and his colleagues collected so much data about the patients, they can actually control for almost all confounders. "We had 7500 fields of data per patient. We know more about these patients than has ever been the case in any other observational study. I don't understand how people can still question this," he told heartwire. And other independent observers have supported this view, as two separate independent editorials published alongside Mangano's first aprotinin study in the NEJM last year both praised his data, one referring to his study as "a model for the future" and the other saying that the large number of patients and information gathered about those patients gave the study credibility.

Ferguson also made the point that many cardiac surgeons will continue to use aprotinin simply because it is so effective. "There is no equivalent drug to aprotinin in terms of ability to control bleeding at the time of surgery. Aprotinin is hands-down more effective than anything else available. For this reason, many cardiac surgeons are still using aprotinin in high-risk patients, as their clinical experience does not bear out the findings of these registry studies."

But Mangano fires back on this point that it is precisely because it is so effective at stopping bleeding that concerns about adverse effects have arisen. "The obvious concern about any drug that stops bleeding so well is that it will be likely to increase the risk of thrombosis. This is a natural worry that is directly related to the fact that it is so effective." He added: "It's all very well to say that cardiac surgeons like this drug, but they can act only on their own surgical experience. Have they followed their patients out to five years to see whether they have died or not? I don't think so." He also noted that sales of aprotinin were down 37% last year after his first study was published, showing that a significant number of surgeons were taking notice of his data.

It's all very well to say that cardiac surgeons like this drug. . . . Have they followed their patients out to five years to see whether they have died or not? I don't think so.

But both Ferguson and Mangano agree on one point: this issue has highlighted the current inadequacy of the postmarketing process for evaluating the safety of new drugs and devices. Ferguson commented: "The same thing has happened with drug-eluting stents and with Vioxx, and in all these cases we still don't know for sure if these drugs/devices are harmful or not. This is particularly shocking for aprotinin, as it has been on the market for 13 years. If Mangano is right, although I personally do not believe that he is, there are 200 000 people getting a drug who shouldn't be."

Mangano expanded on this point to heartwire: "The main problem is that randomized trials conducted for approval are done in relatively young and healthy patients. But as a result, we don't really know how drugs work in older, sicker patients, and these are the people who need drug treatments the most. The safety of drugs should be assessed aggressively and early in the populations receiving them."

He argues that safety should be treated conservatively. "If there is a serious safety signal such as the ones we have found, everyone should pull back, and it should be up to the company to prove in a substantial study that the drug is safe. I believe aprotinin should now be suspended except for rescue use and that Bayer should have to do a 10 000-patient study to prove it is safe. But this is not happening. Our whole system is wrong.

"This problem could be addressed overnight, but unfortunately we have a $250-billion-a-year industry that contributes toward funding the FDA. This is wrong. The FDA should be there to protect the public. It is no good asking companies to do this. Safety does not fit into a business model. There has to be an independent body to do this."

Mangano rejects the claim that he refused to allow the FDA full access to his data and says he feels he is being persecuted for raising a safety issue. "I object to the way I was made to look as if I were holding up the review. I have nothing to hide. I'm not gaining anything financially from this. I am worried about aprotinin and just want to find out the truth. All I am asking is that we should be able to do this. Don't slaughter us just because we raise a question about a drug."