Conference Report
Aprotinin in Cardiac Surgery: A Summary of the FDA Panel Hearing to Assess Safety
September 12, 2007, Rockville, Maryland

Mary Thompson

Medscape Cardiology.  2007; ?2007 Medscape
Posted 09/27/2007

Summary

On September 12, 2007, members of the US Food and Drug Administration's (FDA) Cardiovascular and Renal Drugs Advisory Committee met in joint session with the agency's Drug Safety and Risk Management Advisory Committee to consider the fate of the protease inhibitor aprotinin (Trasylol; Bayer Pharmaceuticals, West Haven, Connecticut), a drug commonly used during on-pump coronary artery bypass graft (CABG) surgery to reduce the risk of bleeding and the need for transfusion. At issue were a number of observational studies suggesting the drug is associated with a higher risk of renal dysfunction and other adverse events, including myocardial infarction (MI), stroke, and death. In particular, the FDA was interested in an observational safety study conducted by Bayer that was not disclosed when the agency last considered the issue of aprotinin safety a year ago. The latest panel meeting was convened to discuss these data and provide more detailed analyses of previous studies in an effort to determine if the drug is safe enough for continued use, if additional labeling changes are warranted, and/or if randomized clinical studies are needed to further refine these risks.

During the panel meeting, numerous speakers raised serious questions about the validity of the observational study data, based on various confounders and suspect statistical methodologies, and the joint panel was in general agreement that the data appeared to be inconclusive for all but renal dysfunction. However, the panel members also agreed that the observational studies raise enough questions to warrant further clinical research investigating the risks associated with the use of aprotinin.

After reviewing all of the data and analyses, the panel voted 16 to 1 (with 1 abstention) to keep the drug on the market, although several panel members said they would like to see modifications or additional restrictions on the label limiting use to cardiac surgery patients who meet a specific high-risk profile or who are undergoing complex procedures. Expressing frustration at the inadequacy of the current safety data on aprotinin, the panel also voted unanimously in support of additional postmarket clinical studies, including randomized clinical trials, to further assess the risks and benefits of the drug; although the exact nature of these studies was the subject of much debate. The panel members were a bit more divided on the question of whether to include results from observational studies in the drug's label, voting 11 to 6 (with 1 abstention) against this proposal. However, the majority expressed a desire to see some type of general statement in the label about the safety concerns raised by these data.

This review provides an overview of the presentations and lively discussions from this recent FDA meeting.

Background

Excessive bleeding is a significant cause of morbidity and mortality in major surgical procedures, and numerous risks are associated with repeat surgery due to bleeding complications. Blood loss is also of concern, particularly in patients undergoing cardiac surgery, in which the need for transfusions in this patient population is associated with prolonged need for mechanical ventilation, impaired wound healing, multiple organ system failure, prolonged length of hospital stay, and increased postoperative mortality.

A number of pharmacologic strategies to reduce perioperative bleeding and the need for transfusions in patients undergoing cardiac surgery have been evaluated, including the use of aprotinin, a broad spectrum proteinase inhibitor derived from bovine lung that acts on multiple mediators associated with inflammatory response, fibrinolysis, and thrombin generation. When used in cardiac surgery patients on cardiopulmonary bypass, it reduces bleeding and decreases the need for blood transfusion.[1]

Although the use of transfused allogeneic blood products during on-pump CABG surgery is fairly common, it is associated with numerous risks, including transfusion-related acute lung injury, multi-organ failure, and death ( Table 1 ).[2]

Aprotinin was approved by the FDA in December 1993 to reduce perioperative blood loss and the need for blood transfusion in high-risk patients undergoing cardiopulmonary bypass (CABG) surgery. Data from randomized, placebo controlled clinical trials conducted during the premarket period, found no increased risk of in-hospital mortality, stroke, MI, or renal failure associated with the drug's use compared with placebo.[1] Furthermore, 2 of the pivotal studies that led to FDA approval found that significantly fewer patients treated with aprotinin needed at least 1 unit of blood compared with patients who received placebo.[1,3-5] In a press release announcing initial market approval, the FDA determined that the drug should be reserved for high-risk patients because severe allergic reactions were reported among patients who received a second dose of the drug.[3] However, in 1998, the FDA expanded the drug's indication to all CABG patients, eliminating this high-risk stipulation.

Because of the strong desire to reduce the bleeding risks associated with on-pump CABG surgery, aprotinin experienced rapid uptake among US cardiothoracic surgeons between 1998 and 2005. Two other antifibrinolytic drugs also can be used to reduce bleeding in CABG patients -- namely, aminocaproic acid (Amicar; Xanodyne Pharmaceuticals, New Port, Kentucky), which has a broad hemostasis indication, and tranexamic acid (Cyklokapron; Pharmacia and Upjohn, Division of Pfizer, Piscataway, New Jersey), which is indicated for hemophiliacs undergoing tooth extraction and is used less commonly in cardiac surgery on an off-label basis. However, neither has specific FDA approval for use in CABG surgery, nor are these alternatives as widely studied or accepted for this indication as aprotinin. The Society of Thoracic Surgeons' 2007 guidelines on blood transfusion and conservation in cardiac surgery endorse aprotinin as a Class I treatment for the reduction of blood loss, transfusion, and re-exploration for bleeding in cardiac surgery (both aminocaproic acid and tranexamic acid are also endorsed as Class I treatments to reduce blood transfusion and total blood loss, although the guidelines state that these drugs are slightly less potent than aprotinin and their safety profile is less well studied).[6]

Safety in Question

Although aprotinin is widely used in CABG surgery, safety has been an issue surrounding its use for some time. Due to several reports of life-threatening anaphylactic allergic reactions in patients receiving the drug, particularly those who received a second dose, a black-box warning was added to the drug's label in the 1990s. This was strengthened in December 2006, when additional text was added warning of the risk of anaphylaxis in patients receiving a second dose of the drug within 12 months of the initial exposure. In the 2006 labeling, known or suspected aprotinin exposure during the previous 12 months was listed as a contraindication for use.

The effect of aprotinin on renal function also has been of concern and was brought to the forefront recently with the publication of several observational studies linking aprotinin with renal dysfunction and renal failure.[7-10] The risk of renal dysfunction in these studies is presented in Table 2 .

In early 2006, the FDA initiated a year-long safety review to consider recent observational data, including 2 published studies by Mangano and colleagues[8] and Karkouti and colleagues,[9] respectively, linking aprotinin with increased risk for a number of adverse events, including kidney failure, MI, and stroke. The FDA's Cardiovascular and Renal Drugs Advisory Committee met to consider the accumulated safety data on September 21, 2006, and the FDA issued additional label changes in December of 2006, warning that aprotinin increases the risk of renal dysfunction and may increase the need for dialysis in the perioperative period. Additionally, the indications for aprotinin use were modified once again to restrict use to CABG patients who are at increased risk for blood loss and blood transfusion.

A few days following the September 2006 panel meeting, Bayer Pharmaceuticals disclosed that an additional observational safety study, known as the i3 study,[11] had been conducted by the company and preliminary results showed that aprotinin may increase the risk of death, serious kidney damage, congestive heart failure, and stroke. According to company officials, 2 Bayer employees had received information about this study before the September 2006 FDA panel meeting, but mistakenly chose not to share that information within the company or with the FDA due to potential problems with the study's design and methodology that raised questions about the validity of the findings. This disclosure prompted the FDA to issue a product warning and schedule another panel meeting, held on September 12, 2007, to delve more deeply into this issue.

Clinical Evidence

At the September 12th meeting, panel members heard from Bayer representatives, as well as Drs. Mangano and Karkouti, along with statisticians who re-analyzed and reviewed the observational data for the FDA. After hearing all of the evidence, panelists concluded that the product meets safety criteria to remain on the market, and that current safety warnings should not be changed, though they also instructed Bayer to conduct further randomized controlled trials to help identify why the various observational studies have reached different conclusions.

In several consecutive presentations, Bayer representatives pointed to the shortcomings of the i3 study,[11] which relied on a large, prospective, in-hospital database encompassing 78,199 CABG patients treated between April 2003 and March 2006. The study's primary findings demonstrated a 50% increase in the risk of dialysis and a 54% increase in the risk of mortality among patients treated with aprotinin compared with those given aminocaproic acid.

However, Bayer scientists questioned these findings, saying the database, which tracked discharge diagnoses, failed to measure important covariates, including preexisting renal failure, insulin-dependent diabetes, urgent vs elective surgery, and New York Heart Association functional class. Moreover, the data did not include laboratory results, medical history, or timing of diagnoses that emerged during hospitalization. As a result, Bayer believes the database is unsuitable to assess aprotinin safety. In addition, the statistical analysis methods used to analyze the data failed to adjust for confounding variables, Bayer said, calling the conclusions into question. A re-analysis performed by the company revealed sufficient statistical evidence to support higher rates of renal dysfunction with aprotinin, but insufficient evidence to link aprotinin to renal failure, stroke, MI, or perioperative mortality.

Bayer representatives also criticized the Mangano[8] and Karkouti[9] observational studies for failing to use adequate statistical methods to account for confounding variables, and FDA re-analyses of these studies appear to be largely in agreement. FDA statisticians performed a propensity score adjustment to account for imbalances in baseline risk factors and geographical regions in the Mangano study, which included prospective data on 4374 on-pump CABG patients treated at 69 centers worldwide who were treated with aprotinin (n = 1295), aminocaproic acid (n = 833), tranexamic acid (n = 822), or no agent (n = 1374).

Mangano and colleagues initially reported that there was a significantly increased risk of death, cardiovascular events, cerebrovascular events, and renal events among patients receiving aprotinin, but not among those who were treated with either of the other 2 drugs. However, the FDA re-analysis calls these conclusions into question. FDA statisticians found no statistically significant differences between patients receiving aprotinin and those who got no antifibrinolytic agent in terms of cardiovascular, cerebrovascular, and in-hospital death outcomes, although there was a significantly greater risk for renal failure among patients who received aprotinin. When the FDA statisticians compared aprotinin vs aminocaproic acid in a subgroup of patients treated at North American centers, the FDA researchers once again uncovered a significantly higher rate of adverse renal outcomes, including renal dysfunction but particularly renal failure, among patients treated with aprotinin. Moreover, the subgroup analysis also revealed a troubling trend toward greater long-term mortality in the aprotinin-treated patients.

In January 2007, Mangano and colleagues published a long-term mortality re-analysis of the data and reported significantly increased mortality (20.8% 5-year mortality) with aprotinin treatment compared with control (no drug), an effect not observed with either aminocaproic acid or tranexamic acid.[12] However, this analysis also has been widely criticized in the medical community for failing to account for numerous potential covariates that could impact mortality, including intraoperative hemoglobin levels, the extent of underlying disease, and the various surgical approaches used in the study. The fact that the FDA re-analysis revealed a potential mortality trend lends more credence to this possibility.

Similarly, in the Karkouti single-center study,[9] the FDA used a subgroup analysis with overlap in propensity scores for treatment comparison. This analysis revealed a statistically significant effect of aprotinin on renal dysfunction, some evidence for a renal failure effect, but no statistical evidence that aprotinin increases risk of MI, stroke, or in-hospital death.

Ongoing studies, including the randomized, controlled Blood Conservation using Anti-Fibrinolytics: A Randomized Trial in a Cardiac Surgery Population (BART) study, comparing aprotinin, aminocaproic acid, and tranexamic acid in 3000 high-risk cardiac surgery patients, hopefully will shed more light on safety. This study, the largest randomized trial of antifibrinolytic drugs in high-risk surgery ever conducted, has enrolled about 2100 patients to date. Interim results of the BART study were reported in  form last year.[13] The authors noted in their  that the rates of primary and secondary events, including massive postoperative bleeding (11.2%), respiratory failure (13.3%), renal failure (7.7%), and prolonged low output (15.0%) were higher than expected ( Table 3 ). These results, they suggest, "may be explained by the inclusion of a higher-risk population than initially anticipated or the lack of recent, good quality data from large patient cohorts in which to base assumptions." Comparison results for the drugs are not yet available.

Another interesting observation was brought to light at the panel meeting by Anthony P. Furnary, MD (Providence Health System, Portland, Oregon), who discussed results of a recent retrospective study that analyzed data from an international multicenter cardiac surgery database. The study, published in the September 11, 2007 issue of Circulation,[14] found that the increase in renal failure seen in patients given aprotinin was directly related to a greater number of transfusions in this high-risk group and was unrelated to aprotinin use.

Panel Deliberations

The FDA panel brought up a number of interesting points in their discussion. Although most agreed that further aprotinin safety studies are necessary, some expressed doubt that they would be conducted in a timely manner, given the fact that the FDA cannot mandate postmarket studies if the drug remains on the market. One panelist expressed frustration about whether "we will ever make sense of all this data."

Several panel members questioned why the reduction in the rate of transfusions associated with aprotinin did not translate into a favorable impact on mortality in any studies to date, especially since transfusions are known to put patients at higher mortality risk. Other members pointed out that mortality following CABG surgery has improved considerably over the years anyway, due to improvements in surgical technique and post-surgical care, so there are many variables that could affect this outcome.

Panel members also emphasized the need for a risk/benefit approach to this issue, but questioned whether this could be firmly established from observational studies. According to one panelist, "any drug that increases the risk of acute kidney injury has to be balanced to a rather significant benefit."

In the end, the panelists voted 16 to 1 (with 1 abstention) to allow aprotinin to remain on the market; however, when asked to describe any label modifications or restrictions they would like to see, several suggested a stronger emphasis on restricting use to high-risk cardiac surgery patients, and one emphasized a better, more specific definition of high risk in the labeling.

After a discussion of the value of observational data, the majority of panelists (11 vs 6 with 1 abstention) voted against including specific findings of observational study data in the product labeling, most stating that the data was not reliable or complete enough to include. However, the majority also agreed that they would like to see a general statement in the labeling informing physicians that questions have been raised from observational data about possible increased incidence of MI, stroke, and death with the use of aprotinin.

Finally, the panel voted 18 to 0 in favor of additional clinical studies, including randomized clinical trials, to further assess the risks and benefits of aprotinin. Although some expressed doubts that a long-term mortality study would be practical, given the number of confounding variables and the potential number of patients that would have to be enrolled, others strongly supported some type of mortality trial. According to one panelist, CABG surgery has changed significantly in the 15 years since this drug was approved, and there is a need for randomized data in a contemporary setting.

For their part, Bayer representatives at the meeting pledged to work closely with the FDA on product labeling changes and on trial design if a randomized trial is deemed appropriate, although they acknowledged many challenges to the latter undertaking. Bayer's Head of Global Development, Kemal Malik, MD, summed up the company's position on aprotinin safety earlier in the day, telling panelists "Bayer remains convinced that the totality of the data will continue to demonstrate a favorable risk/benefit profile for Trasylol when used according to the label."

Supported by an independent educational grant from Bayer Pharmaceuticals


Table 1. Mortality Due to Transfusion[2]


Cause of Death Deaths per Million Units Transfused
Red Blood Cells Platelets
Transfusion-related acute lung injury 10-20 10-20
Bacterial contamination 0.1 15-75 (uncultured)
4-15 (cultured)
Lipid enveloped viruses < 0.1 --
Transfusion errors ~1-2 --
Allergic reactions 5 5
 
Total mortality per million components 16-27 19-100

 

Table 2. Effects of Aprotinin on Renal Dysfunction: Outcomes From Recent Studies


Study No. of Patients Relative Risk or Odds Ratio
(95% CI)
Bayer RCT full-dose aprotinin: 2047
placebo: 1957
*1.41 (1.12, 1.79) OR
Brown 2007[7] high-dose aprotinin: 1778
low-dose aprotinin: 1041
*1.47 (1.12-1.94) high dose RR
1.01 (0.69-1.49) low dose RR
Mangano 2006[8] Primary CABG:
Control (no drug): 1022
Aprotinin: 796

Complex surgery:Control (no drug): 352 Aprotinin: 499
*composite renal outcome (renal failure or renal dysfunction):
OR 2.34 (1.27-4.31) for primary CABG and 2.59 (1.36-4.95) for complex surgery

1.26 (0.76-2.11) all patients RR (per FDA analysis)
Karkouti 2006[9] Aprotinin: 449
Tranexamic Acid: 449
*1.53 (1.11-2.12) RR
(per FDA analysis)
Coleman 2007[10] Control (no drug): 2986
Aprotinin: 362
*2.03 (1.37-3.01) OR

*statistically significant; CABG = coronary artery bypass graft; CI = confidence interval; FDA = US Food and Drug Administration; OR = odds ratio; RR = relative risk

 

Table 3. Major Outcomes Following High Risk Surgery: Interim Results of the BART Trial


Outcomes Patients
(N = 1210)
Primary Outcome
   Any massive bleeding events (%) 11.2
Secondary Outcomes
   Myocardial infarction (%) 4.4
   Cerebrovascular accidents (%) 3.3
   All cause death (%) 4.6
   Any event (%) 10.5
Other Serious Events
   Respiratory failure (%) 13.3
   Renal failure (%) 7.7
   Prolonged low output state (%) 15.0
   Any events (%) 22.4

 



References

  1. Levy JH. Overview of clinical efficacy and safety of pharmacologic strategies for blood conservation. Am J Health Syst Pharm. 2005;62:S15-S19.
  2. Goodnough LT. Risks of blood transfusion. Anesthesiol Clin North America. 2005;23:241-252.
  3. US Food and Drug Administration. Approval of aprotinin, press release 93-48, December 30, 1993. Available at: http://www.fda.gov/bbs/topics/NEWS/NEW00453.html Accessed September 25, 2007.
  4. Lemmer JH, Jr., Stanford W, Bonney SL, et al. Aprotinin for coronary bypass operations: efficacy, safety, and influence on early saphenous vein graft patency. A multicenter, randomized, double-blind, placebo-controlled study. J Thorac Cardiovasc Surg. 1994;107:543-551.
  5. Levy JH, Pifarre R, Schaff HV, et al. A multicenter, double-blind, placebo controlled trial of aprotinin for reducing blood loss and the requirement for donor blood transfusion in patients undergoing repeat coronary artery bypass grafting. Circulation. 1995;92:2236-2244.
  6. Ferraris VA, Ferraris SP, Saha SP, et al. for The Society of Thoracic Surgeons Blood Conservation Guideline Task Force. Perioperative blood transfusion and blood conservation in cardiac surgery: The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists Clinical Practice Guideline. Ann Thorac Surg. 2007;83:S27-S86.
  7. Brown JR, Birkmeyer NJ, O'Connor GT. Meta-analysis comparing the effectiveness and adverse outcomes of antifibrinolytic agents in cardiac surgery. Circulation. 2007;115:2801-2813.
  8. Mangano DT, Tudor IC, Dietzel C, for the Multicenter Study of Perioperative Ischemia (McSPI) Research Group and the Ischemia Research and Education Foundation. The risk associated with aprotinin in cardiac surgery. N Engl J Med. 2006;354:353-365.
  9. Karkouti K, Beattie WS, Dattilo KM, et al. A propensity score case-control comparison of aprotinin and tranexamic acid in high-transfusion-risk cardiac surgery. Transfusion. 2006;46:327-338.
  10. Coleman CI, Rigali VT, Hammond J, Kluger J, Jeleniowski KW, White CM. Evaluating the safety implications of aprotinin use: the Retrospective Evaluation of Aprotinin in Cardio Thoracic Surgery (REACTS). J Thorac Cardiovasc Surg. 2007;133:1547-1552.
  11. Schneeweiss S, Seeger J, Walker AM. Final report. Risks of renal failure and death following use of aprotinin or aminocaproic acid during CABG surgery. Part A: Report on computerized inpatient data from the Premier Perspective Comparative Database. August 7, 2007. Available at: http://www.trasylol.com/Trasylol_09_12_07_AddendumA-2.pdf Accessed September 25, 2007.
  12. Mangano DT, Miao Y, Vuylsteke A, et al; for the Investigators of The Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation. Mortality associated with aprotinin during 5 years following coronary artery bypass graft surgery. JAMA. 2007;297:471-479.
  13. Pretorius R, BART Investigators. Major outcomes following high-risk cardiac surgery. Can J Anesth. 2006;53:26231.
  14. Furnary AP, Wu Y, Hiratzka LF, Grunkemeier GL, Page US 3rd. Aprotinin does not increase the risk of renal failure in cardiac surgery patients. Circulation. 2007;116:I127-I133.

Mary Thompson, staff writer, Medscape Cardiology

Disclosure: Mary Thompson has disclosed no relevant financial relationships.