Aprotinin is Safe in Pediatric Patients Undergoing Cardiac Surgery

Backer CL,Kelle AM, Stewart RD, Suresh SC,Ali FN, Cohn RA, Seshadri R, Mavroudis C.

J Thorac Cardiovasc Surg 2007;134:1421-28

Reviewers: Denise Joffe, MD, Mike Eisses, MD
Seattle Children’s Hospital

Background:
In November 2007, Bayer suspended worldwide marketing of aprotinin due to concerns related to its safety in adult patients having cardiac surgery. This decision by Bayer (www.trasylol.com) was based upon FDA review of 2 studies conducted by Mangano et al, an internal review of their own database of aprotinin safety, and an early review of an ongoing randomized controlled trial (BART) in Canada comparing aprotinin with the lysine analog medications aminocaproic acid and tranexamic acid. All three of these reviews suggest that the use of aprotinin is associated with an increased risk of renal failure, cerebrovascular accidents and overall mortality. These concerns regarding safety are absent in the pediatric literature where the data on aprotinin predominantly centers on its efficacy in reducing blood loss. This is the first study to examine the effects of aprotinin on renal failure, neurological outcome and mortality in pediatric patients having surgery for congenital heart disease.

Design:
The study was a retrospective observational study that compared patients who had cardiac surgery without aprotinin at Children’s Memorial Hospital in Chicago during a six year period from 1994-1999, with a cohort of patients who received aprotinin and had surgery from 2000-2006. The study design was facilitated by the fact that after the year 2000, all patients at Memorial routinely received aprotinin.
The outcomes measured include operative and late mortality, biochemical acute renal failure (an increase in serum creatinine to twice or more of preoperative levels within 72 hours) and the need for temporary dialysis. Neurologic outcomes were also compared. Other predictors of outcome, apart from the use of aprotinin, included patient characteristics such as age, sex, body surface area, emergency status, preoperative ventilatory requirements, Aristotle score, preoperative creatinine levels and previous cardiac surgery. Intraoperative factors included the use of deep hypothermic circulatory arrest (DHCA), and bypass and aortic cross clamp time. As in Mangano’s study, complex statistical analysis was used to try and determine the effect of aprotinin alone on the outcomes measured. Aprotinin dosing was based on body surface area and was consistent with a high dose regimen.

Results:
A total of 2090 patients were included. 1083 patients received aprotinin and 1007 did not. Patients in the aprotinin group were considered more complex (Aristotle score 7.81+/- 2.31 versus 7.23 +/- 2.57), younger (3.49 +/- 1.84 versus 3.64 +/- 4.75 years) and had more emergent procedures (87 versus 49).

There was no statistically significant differences in mortality (aprotinin 3.8% versus 4.5%, p=.37). There was no difference in the incidence of biochemical acute renal failure (6% of the no-aprotinin patients versus 5.7% of the aprotinin patients), although data regarding serum creatinine was incomplete in 138 patients (97 in the no aprotinin group and 41 in the aprotinin group). Temporary dialysis was required in 0.49% in the no-aprotinin group and 0.96% in the aprotinin group (p=.166). Mortality in those patients requiring temporary dialysis was 61% and was not related to the use of aprotinin. Interestingly, no patient in this group required permanent dialysis. A more detailed subanalysis performed according to risk factor as measured by the Aristotle score did not identify the use of aprotinin as a predictor of renal failure.

There was no difference in the incidence of postoperative new onset neurologic deficits persisting at discharge between patients who received aprotinin (0.24%) and those who did not (0.24%). In addition, there was no difference in the incidence of new onset seizures between groups (no aprotinin 0.89%, versus aprotinin 1.1%).

Other interesting data presented include the incidence of adverse reactions to aprotinin in this cohort. No patient had a reaction to the test dose of aprotinin. Two patients had events, one cardiac and one pulmonary, which were attributed to aprotinin administration. They required treatment with pressors and discontinuation of aprotinin. Ninety four patients were re-exposed to aprotinin within one year of administration, without any complication.

Comment:
Although the current safety concerns regarding aprotinin can be considered legitimate for the adult populations given Bayer’s response to suspend marketing, how these concerns translate to pediatric populations is unclear. This study provides interesting data regarding safety in pediatric patients. However, despite the favorable picture painted for aprotinin in this study, there are important limitations pointed out by the authors as well as those participating in the discussion, most notably Dr. JamesTweddell, Chief of Congenital Heart Surgery at Children’s Hospital, Wisconsin.

The most significant limitations of this study are the difference in the time frame in which the cohorts are compared and the inevitable biases of performing an observational, nonrandomized study. Extensive statistical analysis, including propensity scoring can help reduce bias, but not eliminate it.
The no aprotinin cohort was operated on beginning in 1994. There are clearly significant differences in all aspects of patient management during this 12 year interval which have the effect of confounding the outcome results and cannot be controlled for. It is possible that those patients given aprotinin would have done even better had they not received it.

For example, the use of DHCA at some centers has been reduced over time with more utilization of intermittent or low flow cerebral perfusion. In this article, they point out that the use of DHCA was a predictor for both operative mortality and renal dysfunction, two of their primary outcome measures. Further, DHCA was utilized more in the no-aprotinin group, although it didn’t reach statistical difference (p=0.09). Would this have reached significance with more patients? In addition, the duration of DHCA may have had an effect on the outcome measures. Perhaps including an average time of DHCA would have been helpful. A subanalysis of data relating outcome measures to duration of DHCA, or alternatively a subanalysis for each group where DHCA wasn’t used may have helped tease out the effects of DHCA. This is only one example of a potentially confounding factor when examining such complex outcomes.

Dr. Tweddell pointed out the limitations of the definition of biochemical acute renal failure which was chosen. The study definition was a doubling of baseline creatinine. The authors struggled with their definition of biochemical acute renal failure and relied on expert opinion from their nephrologists as well as the definition used by the cardiac surgical database. However, using that definition resulted in renal failure occurring paradoxically more commonly in patients with lower baseline creatinine values. Given that renal outcome was a major focus of this paper and that the study was performed in a single institution, Dr. Tweddell suggested that perhaps the definition should have been the upper limit of normal creatinine values at Memorial hospital. This may have been a more sensitive indicator of renal injury.

The effect of aprotinin on neurologic outcome is difficult to measure since the incidence of neurologic complications (post operative seizures, and neurologic abnormalities at discharge) is so low. In addition, the study did not examine more subtle, but important neurological abnormalities. This would be an important area to explore given the results of the Mangano study. In contrast, to finding a negative effect on neurological outcome, there are others who suggest aprotinin may confer some neuroprotection based on its anti-inflammatory properties.

In Mangano’s first study, he describes the fact that it seems counterintuitive to use an antifibrinolytic at a time when the patient is having a thrombotic complication (myocardial infarction), even if it is being used during cardiac surgery to help reduce bleeding. He expected that all antifibrinolytics might have a negative impact on thrombotic complications, however, negative outcome results were only found in the aprotinin group and not in the aminocaproic acid and tranexamic acid groups. Although this pediatric study did not specifically look at thrombotic complications, thrombosis in these patients may effect outcome, especially in those with single ventricle physiology who have abnormal venous flow patterns. Without quoting any figures, the authors state that thrombosis is rarely a complication in their patients. However, a more detailed comparison of this complication might be important.

Conclusion:
Despite the limitations, the authors should be applauded for this study since there is no definitive data on the safety of aprotinin in the pediatric population. Extrapolating data from the adult literature is potentially problematic and unfair for dictating guidelines in pediatric patients. Given the current safety concerns of aprotinin and the suspension of marketing, it is likely impossible to perform a randomized controlled study at this time in pediatric patients to answer these questions. We know that bleeding in pediatric patients after bypass can be a significant complication and therefore the need for strategies to reduce bleeding is paramount. This study provides evidence that aprotinin may act differently in pediatric patients, and despite the limitations cited above, this study helped to characterize the risk benefit ratio. Unfortunately, for the pediatric patient, the decision that Bayer and the FDA make regarding its safety in adult patients will likely dictate whether it will be available to consider in our patients.

References:
1. Mangano DT, Tudor IC, Dietzel C; Multicenter Study of Perioperative Ischemia Research Group; Ischemia Research and Education Foundation. The risk associated with aprotinin in cardiac surgery. Engl J Med. 2006 Jan 26;354(4):353-65.

2. Mangano DT, Miao Y, Vuylsteke A, Tudor IC, Juneja R, Filipescu D, Hoeft A, Fontes ML, Hillel Z, Ott E, Titov T, Dietzel C, Levin J; Investigators of The Multicenter Study of Perioperative Ischemia Research Group; Ischemia Research and Education Foundation Mortality associated with aprotinin during 5 years following coronary artery bypass graft surgery. JAMA 2007;297(5):471-9

3. Sedrakyan A, Treasure T, Elefteriades JA. Effect of aprotinin on clinical outcomes in coronary artery bypass graft surgery: a systematic review and meta-analysis of randomized clinical trials. J Thorac Cardiovasc Surg 2004;128:442-8

Aprotinin is Safe in Pediatric Patients Undergoing Cardiac Surgery

© 2007 CONGENITAL CARDIAC ANESTHESIA SOCIETY