A Comparison of the Effect of Aprotinin and ε-Aminocaproic Acid on Renal Function in Children Undergoing Cardiac Surgery

A Comparison of the Effect of Aprotinin and ␧ -Aminocaproic Acid on Renal Function in Children Undergoing Cardiac Surgery Galina Leyvi, MD, Olivia Nelson, MD, Adam Yedlin, MD, Michelle Pasamba, Peter F. Belamarich, MD, Singh Nair, and Hillel W. Cohen, PhD, MPH Objective: To assess the incidence of renal injury among pediatric patients who received aprotinin while undergoing cardiac surgery compared with those who received ␧-aminocaproic acid (EACA). Design: A retrospective observational study. Setting: A single academic center. Participants: Pediatric cardiac patients who had cardiopulmonary bypass and received aprotinin or EACA. Intervention: Patients undergoing pediatric cardiac surgery received aprotinin from 2005 to 2007 and EACA from 2008 to 2009. Measurements and Main Results: The primary outcome was acute kidney injury (AKI) defined as serum Cr elevation at discharge more than 1.5 times the baseline value. Secondary outcomes included bleeding, blood transfusion, and the volume of chest tube drainage in the first 24 hours postoperatively. One hundred seventy-eight patients met inclusion criteria; 120 patients received aprotinin, and 58 patients received EACA. These 2 groups did not differ sig-

nificantly in age, weight, or duration of cardiac bypass. Logistic regression analysis, adjusted for confounding variables (ie, baseline Cr, sex, age, CPB time, inotropic support and vasopressors), showed a higher odds of suffering AKI at discharge with the usage of aprotinin (odds ratio ⴝ 4.7; 95% confidence interval, 1.1-19.5; p ⴝ 0.03). The volume of the first 24 hours of chest tube drainage was not significantly different between groups, as well as packed red blood cells and cryoprecipitate units. However, fresh frozen plasma and platelets showed statistically significant differences with more transfusion in the EACA group. Conclusion: In this retrospective study, the authors observed a higher odds of AKI for aprotinin usage compared with EACA, suggesting that the known concern for adults with adverse kidney effects with aprotinin is also appropriate for pediatric patients. © 2011 Elsevier Inc. All rights reserved.

A

of patients treated with aprotinin and 1.8% of patients who were treated with alternative antifibrinolytic agent.1 In pediatric patients, the use of aprotinin and other antifibrinolytics is associated more commonly with renal failure, with reported incidences as high as 9.6% and 4.1%, respectively.4 The incidence of renal dysfunction after pediatric cardiac surgery has been reported to vary from 16% to 36%, and some of this variation may depend on the definition of renal dysfunction that was used.4,7,8 In most of the previous studies on this subject, the providers administered aprotonin or other antifibriolytic agents according to their own clinical judgment, therefore lacking a standardized protocol for the administration of these agents. The present authors examined the association between aprotinin administration and acute kidney injury (AKI) in pediatric patients undergoing cardiac surgery at their institution during 2 distinct time periods in which the institutional protocol dictated the use of the antifibrinolytic agent. This gave the authors an opportunity to conduct a retrospective study with minimal selection bias.

PROTININ IS a serine protease inhibitor that was used widely in cardiac surgery as an antifibrinolytic to reduce postoperative bleeding. This agent is no longer in use or available because in adult populations it was found to be associated with renal dysfunction as well as increased risk of myocardial infarction and mortality.1,2 However, in pediatric cardiac surgery, the association between aprotinin use and renal dysfunction has not been shown.3-5 Aprotinin has a high affinity for the kidneys. It passes freely through the glomerulus, binds selectively to the brush border of the proximal tubule membrane, enters by pinocytosis, and accumulates within the cytoplasm. Aprotinin inhibits tubule protease secretion, prostaglandin and renin synthesis, prostasin secretion, and bradykinin release.6 These unfavorable tubular effects are complicated by aprotinin dose-dependent renal afferent vasoconstriction, thereby impairing deep cortical and medullary perfusion and leading to focal tubular necrosis.6 Establishing a clear causal relationship between aprotinin use and kidney injury is difficult because pediatric kidney injury during cardiac surgery could be multifactorial, potentially involving cardiopulmonary bypass (CPB), cardiogenic shock, vasopressor, and inotropic support as well as other agents given during the perioperative period. Severe kidney injury during cardiac surgery in adults requiring renal replacement therapy (such as dialysis) is rare. It was reported in 5.5%

From the Albert Einstein College of Medicine/Montefiore Medical Center, Bronx, NY. Address reprint requests to Galina Leyvi, MD, Department of Anesthesiology, Albert Einstein College of Medicine, Montefiore Medical Center, 111 East 210th Street, 4th Floor, Silver Zone, Bronx, NY 10467. E-mail: [email protected] © 2011 Elsevier Inc. All rights reserved. 1053-0770/2503-0003$36.00/0 doi:10.1053/j.jvca.2011.01.015 402

KEY WORDS: antifibrinolytic agents, pediatric cardiac surgery, creatinine, acute kidney injury

METHODS The participants in this study were patients who had undergone pediatric cardiac surgery using CPB between May 2005 and July 2009. This retrospective study was approved by the Institutional Review Board, and informed consent was waived. Data were collected from patients’ medical records and the hospital medical information system. Aprotinin (Trasylol; Bayer Healthcare Pharmaceuticals, Pittsburgh, PA) containing 10,000 KIU/mL was used according to the hospital protocol from May 2005 to December 2007 and administered in doses of 20,000 IU/kg followed by a continuous infusion of 20,000 IU/kg/h throughout the procedure. Four thousand international units per kilogram of aprotinin also were added in the pump prime. In November 2007, Bayer pharmaceuticals in conjunction with the Food and Drug Administration decided to stop worldwide marketing of aprotinin.9 Thereafter, the authors’ institution began replacing aprotinin with ␧-aminocaproic acid (EACA). EACA was used from January 2008 to July 2009 in doses of a 150 mg/kg bolus followed by a 15-mg/kg/h

Journal of Cardiothoracic and Vascular Anesthesia, Vol 25, No 3 (June), 2011: pp 402-406

EFFECT OF APROTININ AND EACA

403

Table 1. Pediatric Modified RIFLE (p-RIFLE) Criteria to Determine the Extent of AKI pRIFLE

Risk Injury Failure Loss End-stage

Change in GFR From Baseline

Cr Elevation From Baseline

Decrease ⬎25% More than 1.5 times Decrease ⬎50% More than 2 times Decrease ⬎75% More than 3 times Persistent failure for 4 weeks Persistent failure for ⬎3 months

Abbreviation: GFR, glomerular filtration rate. Data from Zappitelli et al,8 Ricci et al,11 and Venkataraman and Kellum.12

continuous infusion in accordance with institutional practice protocol and literature recommendations.10 Both antifibrinolytic agents were started after anesthesia induction, but before surgery, and terminated at the conclusion of the surgery. Patients admitted on the day of surgery were premedicated as needed with oral midazolam, 0.5 to 1.0 mg/kg. After the placement of standard anesthesia monitors, anesthesia was induced by sevoflurane inhalation and intravenous agents. Endotracheal intubation followed, and invasive monitors were placed. General anesthesia was maintained by oxygen and inhalation anesthetics (sevoflurane or/and isoflurane), nondepolarizing muscle relaxants, and fentanyl titrated to hemodynamic response. The surgical technique included median sternotomy and aortic and venous cannulation (bicaval in most cases). Anticoagulation was achieved by the central administration of 300 U/kg of porcine heparin. Kaolin-activated coagulation time (ACT) values exceeding 480 seconds were confirmed before the initiation of CPB. Additional heparin was administered as necessary during CPB to maintain an ACT of 480 seconds. Nonpulsatile hypothermic CPB was performed using a non– heparin-coated circuit. The priming volume was chosen according to patients’ weight, with additional heparin in it. Packed red blood cells were added to the prime before the CPB or during the CPB as needed to achieve desirable hematocrit. Modified ultrafiltration was performed immediately after separation from CPB at the surgeon’s discretion. Protamine (3 mg/kg) was used to neutralize heparin upon completion of modified ultrafiltration. After confirmation of heparin neutralization by ACT, persistent bleeding was treated with transfusion of platelets followed by cryoprecipitate and fresh frozen plasma as deemed necessary by the attending anesthesiologist and surgeon. At the conclusion of surgery, patients were transported to the pediatric intensive care unit. Further management was at the discretion of the pediatric intensive care unit staff. Baseline creatinine (Cr) values were obtained either from the day before surgery or from a preoperative clinic visit. The discharge Cr value was defined as the last Cr obtained before discharge. To assess the incidence of AKI, the authors used RIFLE criteria. The adult RIFLE classification (Table 1), which was developed to assess the clinical significance of AKI based on serum Cr elevation, glomerulofiltration rate, and urine output,11,12 also was validated for pediatric patients.13 RIFLE defines 3 grades of increasing severity of AKI, risk, injury, and failure, and 2 outcome classes, loss and end-stage renal disease (Table 1). The authors used only Cr elevation criteria from RIFLE classification but not urine output or glomerulofiltration rate because they did not believe in the reliability of these parameters in a retrospective study. The discharge Cr elevation ratio was calculated as discharge Cr value divided by baseline Cr. Thus, the definition of the risk was a Cr elevation more then 1.5 times from baseline, injury was defined as Cr elevation more than 2 times the baseline, and failure was defined as Cr elevation more than 3 times the baseline.7,11-13 The grades of RIFLE classification (ie, risk, injury, and failure) were pooled into a

single outcome category defined as AKI. The 2 outcome classes (ie, loss and end-stage renal disease) were not considered because the authors are not planning to do a follow-up of the subjects. The effect of aprotinin on the incidence of AKI at the time of discharge was considered as the primary outcome. The chest tube drainage, incidence of bleeding, and blood transfusion were considered as secondary outcomes. Information about chest tube drainage was obtained from the nursing flow chart during the first 24-hour period and recorded as mL/kg/24 h. Postoperative bleeding was defined as a return to the operating room for the control of bleeding during the first 24 hours after surgery. Information about blood transfusion was taken from the hospital record as cumulative number of intraoperative and postoperative transfusions. The t test for normally distributed data or the Mann-Whitney U test for non-normally distributed data was used for the comparison of continuous variables between groups. The chi-square test was used to assess differences among the categoric variables. Statistical significance for all of these was defined as p ⬍ 0.05. Logistic regression models were constructed to assess the association of AKI with aprotinin use. In addition to aprotinin, variables included in the model as potential confounders of the association were baseline Cr, sex, age, duration of CPB, inotropic support (eg, epinephrine, milrinone, and dopamine), and vasopressors (eg, norepinephrine and vasopressin). The Hosmer-Lemeshow test was used to assess the goodness of fit with p ⬎ 0.05 as an indication the model was not inappropriate. Statistical analysis was performed by using SPSS 18 (SPSS Inc, Chicago, IL). RESULTS

Medical records of 193 cases were reviewed and met the criteria for entry into this study. Of these 193 cases, 11 received no antifibrinolytic agent and were excluded from further analysis. If patients had repeated surgery during the same hospitalization, the second procedure was excluded from analysis (4 cases) (Fig 1). In total, there were 120 aprotinin patients (67.6%) and 58 EACA patients (32.4%). Normal renal function

Fig 1.

A CONSORT diagram depicting the flow of study cases.

404

LEYVI ET AL

Table 2. Patients’ Age, Weight, Anesthesia Time, Surgical Time, Hospital Length of Stay, CPB Time, Aortic Cross-clamp Time, and Cr at Baseline

Aprotinin group (n ⫽ 120) EACA group (n ⫽ 58) p value

Age (y)

Weight (kg)

Hospital Stay (d)

Anesthesia Time (min)

Surgical Time (min)

CPB Time (min)

AXC Time (min)

Baseline Cr (mg/dL)

5.4 (0.9-11.1)

15 (8-34.6)

7 (4-13)

366 (311-453)

243 (183-315)

121 (86-174)

50 (32-130)

0.49 ⫾ 0.27

2.5 (0.8-10.1) 0.45

11.4 (7.6-27.2) 0.45

5 (4-8.5) 0.34

360 (318-462) 0.75

230 (186-310) 0.98

123 (92-175) 0.93

59 (38-139) 0.56

0.47 ⫾ 0.16 0.93

NOTE. Reported values correspond to median (25%-75%) for non-normally distributed data, or mean ⫾ standard deviation for normally distributed data; p values were obtained by the Student t test for normally distributed data and the Mann-Whitney U test for non-normally distributed data. Abbreviation: CPB, cardiopulmonary bypass; AXC, aortic cross-clamp time; EACA, ␧-aminocaproic acid.

was documented during the preoperative workup. There were 16 (13%) neonates in the aprotinin group and 5 (8.3%) in the EACA group. There was no significant difference in the proportion of patients who had prior corrective surgery (32.5% in the aprotinin group v 31.6% in the EACA group, p ⫽ 0.45) or received postoperative epinephrine, milrinone, dopamine, norepinephrine, or vasopressin (p ⬎ 0.54 for all of these). There was no significant difference between groups in the median values of the patients’ age, weight, anesthesia time, surgical time, length of hospital stay, CPB time, or aortic cross-clamp time as well as the average baseline Cr value (Table 2). There were 59 males (49%) and 61 females (51%) in the aprotinin group and 31 males (53%) and 27 females (47%) in the EACA group (p ⫽ 0.67). The surgical corrections for each group are presented in Table 3. According to the RIFLE criteria at discharge time, 20 patients (11.2%) had some degree of AKI: 9 patients (5%) at risk, 7 (3.9%) at injury, and 4 (2.2%) at failure. After adjusting for different confounding variables (ie, baseline Cr, sex, age, CPB time, inotropic support, and vasopressors), using a logistic regression model, the odds of AKI at discharge were 4.7 (95% confidence interval, 1.1-19.5; p ⫽ 0.03) in patients who received aprotinin compared with patients who received EACA. The incidences of bleeding (8.9% in the aprotinin group and 13.3% in the EACA group) and re-exploration (7.3% in the aprotinin groups and 5% in the EACA group) were not significantly different between groups (p ⬎ 0.36). The volume of the first 24 hours of chest tube drainage was not statistically significant between groups as well as packed red blood cells and cryoprecipitate units. However, fresh frozen plasma and platelets showed statistically significant differences with more transfusion in the EACA group. Detailed information regarding chest tube drainage, blood, and blood products transfusion is presented in Table 4. There were 4 (3.2%) deaths in the aprotinin group and 1 (1.7%) in the EACA group. Detailed information about mortality is presented in Table 5. There were 4 (3.2%) incidents of renal failure that required dialysis in the aprotinin group and no incidents of renal failure in the EACA group. Among these patients in the aprotinin group for whom dialysis was indicated, 1 patient died after the dialysis was initiated, and another patient died before dialysis was started.

DISCUSSION

This retrospective study assessed the association of AKI incidence among children receiving aprotinin or EACA while undergoing pediatric cardiac surgery. In contrast to previous pediatric studies, the authors found a statistically significant association of aprotinin usage and AKI as increased odds of AKI at discharge with aprotinin. In recent literature, only 1 prospective randomized study addressing safety and efficacy of aprotinin in a pediatric population was found.5 This study was

Table 3. Corrective Surgeries, Number of Patients in Each Group (% of Total) Fibrinolytic Category Dx/Procedure

TOF repair VSD/AVC repair ASD repair Valves repair BDG Fontan APVR repair DORV repair Aortic surgery RV-PA conduit Subaortic membrane ACA repair DCRV repair Central Shunt Rastelli procedure PA band TGA repair PA plasty Total

Aprotinin (%)

EACA (%)

Total (%)

19 (10.7) 31 (17.4) 15 (8.4) 20 (11.2) 6 (3.4) 5 (2.8) 3 (1.7) 2 (1.1) 7 (3.9) 4 (2.2) 3 (1.7) 1 (0.56) 2 (1.1) 0 1 (0.56) 1 (0.56) 0 0 120 (67.4)

18 (10.1) 14 (7.9) 8 (4.5) 5 (2.8) 1 (0.6) 1 (0.6) 1 (0.6) 3 (1.7) 1 (0.6) 1 (0.6) 0 1 (0.56) 1 (0.56) 1 (0.56) 0 0 1 (0.6) 1 (0.6) 58 (32.6)

37 (20.8) 45 (25.3) 23 (13.0) 25 (14.0) 7 (3.9) 6 (3.4) 4 (2.2) 5 (2.8) 8 (4.4) 5 (2.8) 3 (1.7) 2 (1.1) 3 (1.7) 1 (0.56) 1 (0.56) 1 (0.56) 1 (0.56) 1 (0.56) 178 (100)

NOTE. Valves included aortic, mitral, tricuspid, and pulmonic repair/replacement; p ⫽ 0.25. Abbreviations: TOF, tetralogy of Fallot; VSD, ventricular septal defect repair; AVC, atrial-ventricular canal repair; ASD, atrial septal defect repair; valves included aortic, mitral, tricuspid and pulmonic repair/replacement; BDG, bidirectional Glenn; APVR, anomalous pulmonary venous return; DORV, double-outlet right ventricle; RV-PA, right ventricle pulmonary artery; ACA, anomalous coronary artery; DCRV, double-chamber right ventricle; PA, pulmonary artery; TGA, transposition of great arteries.

EFFECT OF APROTININ AND EACA

405

Table 4. Value for Packed Red Blood Cells, Fresh Frozen Plasma, Cryoprecipitate, and the First 24 Hours of Chest Tube Drainage

Aprotinin group EACA group p Value

PRBC

FFP

Platelets

Cryo

CT Drainage (mL/kg/24 h)

1 (0-3) 2 (1-3) 0.15

0 (0-1) 1 (0-1) 0.03

0 (0-1) 1 (0-1) 0.01

0 (0-0) 0 (0-1) 0.082

9.7 (6-16.2) 11.7 (6.4-23.3) 0.09

NOTE. Reported values correspond to the median (25%-75%) for non-normally distributed data; p values were obtained by the Mann-Whitney U test. Abbreviations: PRBCs, packed red blood cells; FFP, fresh frozen plasma; Cryo, cryoprecipitate; CT, chest tube.

prematurely interrupted by aprotinin suspension and showed only the effect on platelet volume transfused and no difference on Cr elevation; only 13 patients were enrolled in each group. One large prospective but not randomized study (the choice of aprotinin administration was at the discretion of the anesthesiologist) on pediatric cardiac patients showed increased incidences of dialysis and renal dysfunction in patients who received aprotinin, but propensity-adjusted risk ratios were not statistically significant and multiple logistic regression did not show a significant association between aprotinin and renal dysfunction or dialysis.4 The 3 other retrospective studies (the choice of aprotinin administration was at the discretion of the anesthesiologist)3,7,14 did not find an association of aprotinin use and kidney dysfunction by multivariate logistic regression analysis, but dysfunction was associated with the length of CPB, the use of deep hypothermic circulatory arrest, the presence of comorbidities, and younger age. Those studies in which the choice of fibrinolytic agent was made according to the practitioner’s decision emphasized differences between aprotinin and nonaprotinin groups in several aspects including a history of previous surgery, organ dysfunction, surgical complexity, age, CPB time, and so on.3,4,7,14 In comparison, the present study observed similarities of the groups for the measured characteristics. The RIFLE criteria were used to increase the clinical relevance of the present findings and to facilitate the comparison of the present results with other studies. The RIFLE classification was developed in 2004 to allow comparison across various studies examining renal function in adults, and then it was validated by Akcan-Arikan et al13 for use in pediatric populations. The classification is now the most widely used criterion

for the assessment of AKI in pediatric patients including children undergoing cardiac surgery.3,4,7,8 There is a large range of normal values of Cr, and this leads to difficulty distinguishing when an increase in Cr represents an abnormal decrease in the glomerular filtration rate for a given patient. Indeed, the glomerular filtration rate must fall by more than half before the serum Cr increase rises above the upper limit of normal for a given age group (or the serum Cr increase may remain stable until more than 50% of kidney function has been lost). Therefore, the Cr increase of 1.5 times, which is fitting in the criteria of kidneys at risk, is clinically important. One of the strengths of this retrospective study is that all pediatric cardiac patients received aprotinin according to hospital protocol if they had surgery before 2008, independent of the complexity of the proposed procedure and expected coagulopathy. The same principle was applied to the patients who had surgery from the beginning of 2008 to July 2009; they all received EACA. Only 6% of patients fell off the protocol and did not receive any antifibrinolytic agent during the transitional period. The present authors believe that adherence to this protocol in their institution minimizes the potential for an indication bias toward aprotinin use for the higher-risk patients and makes the present study different from other studies investigating this area of interest.4,7,14 Additional strengths of this study include a lack of significant differences between the comparison groups in baseline characteristics such as height, weight, and baseline Cr. There were also no significant differences in the length of anesthesia time, CPB time, and aortic cross-clamp time. This supports the idea that the aprotinin and EACA groups were fairly similar at baseline and had similar operative courses in terms of objective

Table 5. Perioperative Mortality Details Discharge Cr–to–Baseline Ratio

Age

Procedure

Death on Postoperative Day

Aprotinin group 5 days 7 days

Arterial switch Interrupted aortic arch

0 13

NA 1.6

NA 3-d postop day

Dialysis

2 years

VSD repair

2

3.6

4 years

DORV

7

1.2

No t initiated, but was intended No

Central Shunt

4

4.3

No

EACA group 1 month

NOTE. The discharge Cr compared with baseline calculated as a discharge Cr–to– baseline Cr ratio.

Proposed Cause of Death

Inability to wean from CPB Complete heart block, circulatory arrest Kidney and liver failure, hyperkalemia Left ventricular failure, ECMO, Stroke Multiple-organ failure, SVC thrombosis

406

LEYVI ET AL

measures such as the length of procedure and the length of time of various critical parts of the surgery. The limitations of this study are its retrospective nature and relatively small number of patients because it was limited by the number of cases done during a certain period of time with certain antifibrinolytic agents. In a retrospective observational study, it is not known if there are unmeasured (latent) confounders. In addition, because the use of the antifibrinolytic agents changed at one point in time by hospital protocol, there may have been differences between the 2 groups because the surgeries took place during different time periods. Although the pediatric surgeon and the staff anesthesiologists remained the same, techniques and other standards of care naturally evolve over time, and this change may introduce unintentional time-based selection bias. Using RIFLE criteria, the authors found that overall AKI can be higher than 10%, which may have a significant impact on the future of the patients’ life and health care resources utilization. This study comes to a different conclusion compared with other works on this issue in pediatric patients. The studies that examined whether aprotinin usage is associated with renal dysfunction in pediatric cardiac surgeries were performed at institutions where aprotinin use was at the discretion of the anesthesiologist or other practitioner. This resulted in more aprotinin use during complex cases rather than a uniform protocol of use for all surgeries with CPB within a certain time period. Several studies with this design have found that the length of CPB accounted for any difference in renal dysfunction between groups.3,4,7,14 A study design in which the choice

of drug is at the discretion of the practitioners inherently contains a form of selection bias in which it is difficult to differentiate renal dysfunction caused by more complex procedures that necessitate longer CPB time from kidney damage caused by aprotinin. The inclusion of the Aristotle complexity score3,7 and the Risk Adjustment for Congenital Heart Surgery score14 in a multiple regression model may not offset influence of this differences. Although the present study design is not a randomized controlled trial, it reduced this form of selection bias to a large extent. In general, children who have had AKI are at risk of developing chronic kidney disease years afterwards.15 In particular, neonates are at a higher risk of developing chronic kidney disease, hypertension, and metabolic syndrome.15,16 Although follow-up studies on children who suffered acute renal failure are not abundant, they report 80% survival for 3 to 5 years.17 The number of patients showing some signs of kidney injury during follow-up varies from 8.8%18 to 59%,17 with a higher incidence of injury in patients who had primary renal conditions. This study aimed to determine if children who were exposed to aprotinin may have been at a higher risk of sustaining renal damage during their surgeries. The finding that children given aprotinin were at a higher risk of AKI is consistent with findings for adults. Although the effect of mild perioperative kidney injury on the development of renal dysfunction in the future is not examined in this study, it may have potential for further investigation.

REFERENCES 1. Mangano D, Tudor IC, Deitzel C: The risk associated with aprotinin in cardiac surgery. N Engl J Med 354:353-365, 2006 2. Mangano DT, Miao Y, Vuylsteke A, et al: Mortality associated with aprotinin during 5 years following coronary artery bypass graft surgery. JAMA 297:471-479, 2007 3. Backer CL, Kelle AM, Stewart RD, et al: Aprotinin is safe in pediatric patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 134:1421-1428, 2007 4. Szekely A, Sapi E, Breuer T, et al: Aprotinin and renal dysfunction after pediatric cardiac surgery. Pediatr Anesth 18:151-159, 2008 5. Williams GD, Ramamoorthy C, Pentcheva K, et al: A randomized, controlled trial of aprotinin in neonates undergoing open-heart surgery. Pediatr Anesth 18:812-819, 2008 6. Kramer HJ, Dusing R, Glanzer K, et al: Effects of aprotinin on renal function. Contrib Nephrol 42:233-241, 1984 7. Manrique A, Jooste EH, Kuch BA, et al: The association of renal dysfunction and the use of aprotinin in patients undergoing congenital cardiac surgery requiring cardiopulmonary bypass. Anesth Analg 109: 45-52, 2009 8. Zappitelli M, Bernier P-L, Saczkowski RS, et al: A small postoperative rise in serum creatinine predicts acute kidney injury in children undergoing cardiac surgery. Kidney Int 76:885-892, 2009 9. Bayer Healthcare Pharmaceuticals communication (Leverkussen GaWH, CT). Available at: http://www.trasylol.com. Accessed July 2009

10. Williams GD, Bratton SL Riley EC, et al: Efficacy of ␧-aminocaproic acid in children undergoing cardiac surgery. J Cardiothorac Vasc Anesth 13:304-308, 1999 11. Ricci Z, Cruz D, Ronco C: The RIFLE criteria and mortality in acute kidney injury: A systematic review. Kidney Int 73:538-546, 2008 12. Venkataraman R, Kellum JA: Defining acute renal failure: The RIFLE criteria. J Intensive Care Med 22:187-193, 2007 13. Akcan-Arikan A, Zappitelli M, Loftis LL, et al: Modified RIFLE criteria in critically ill child with acute kidney injury. Kidney Int 71:1028-1035, 2007 14. Guzzetta NA, Evans FM, Rosenberg ES, et al: The impact of aprotinin on postoperative renal dysfunction in neonates undergoing cardiopulmonary bypass: A retrospective analysis. Anesth Analg 108: 448-455, 2009 15. Mak RH: Acute kidney injury in children: The dawn of a new era. Pediatr Nephrol 23:2147-2149, 2008 16. Askenazi DJ, Bunchman TE: Pediatric acute kidney injury: The use of the RIFLE criteria. Kidney Int 71:963-964, 2007 17. Askenazi DJ, Feig DI, Graham NM, et al: 3-5 year longitudinal follow-up of pediatric patients after acute renal failure. Kidney Int 69:184-189, 2006 18. Shaw NJ, Brocklebank JT, Dickinson DF, et al: Long-term outcome for children with acute renal failure following cardiac surgery. Int J Cardiol 31:161-165, 1991