Perioperative Renal Outcome in Cardiac Surgical Patients With Preoperative Renal Dysfunction: Aprotinin Versus Epsilon Aminocaproic Acid

ORIGINAL ARTICLES

Perioperative Renal Outcome in Cardiac Surgical Patients With Preoperative Renal Dysfunction: Aprotinin Versus Epsilon Aminocaproic Acid Andrew D. Maslow, MD,* Alyas Chaudrey, MD,† Arthur Bert, MD,* Carl Schwartz, MD,* and Arun Singh‡ Objective: The administration of aprotinin to patients with pre-existing renal dysfunction who are undergoing cardiac surgery is controversial. Therefore, the authors present their experience with the use of aprotinin for patients with preoperative renal dysfunction who underwent elective cardiac surgery requiring cardiopulmonary bypass (CPB). Design: Retrospective analysis. Setting: University hospital. Participants: Consecutive cardiac surgical patients with preoperative serum creatinine (SCr) >1.8 mg/dL undergoing nonemergent cardiac surgery requiring CPB. Interventions: None. Methods: One hundred twenty-three patients either received epsilon aminocaproic acid (EACA, n ⴝ 82) or aprotinin (n ⴝ 41) as decided by the attending anesthesiologist and surgeon. Data were collected from the Society of Thoracic Surgeons database and from automated intraoperative anesthesia records. Renal function was assessed from measured serum creatinine (SCr) and calculated creatinine clearances (CrCls). Acute perioperative renal dysfunction was defined as a worsening of perioperative renal function by >25% and/or the need for hemodialysis (HD). Analysis: Data were recorded as mean and standard deviation or percentage of population depending on whether the data were continuous or not. Data were compared by using an analysis of variance, chi-square analysis, Student paired and unpaired t tests, Fisher exact test, Wilcoxon rank sum test, and Mann-Whitney U test. A p value <0.05 was considered significant. Results: Overall, 32% and 41% of patients had acute perioperative renal dysfunction measured by CrCl and SCr, respectively. Seven patients required HD (5.7%). Six of these 7 had complicated postoperative courses. Of all the variables measured, only the duration of the aortic cross-

clamp (AoXCl) and CPB were significantly associated with acute perioperative renal dysfunction. Acute perioperative renal dysfunction was associated with increased intensive care unit and hospital stays, postoperative blood transfusion, dialysis, and major infection. Aprotinin patients were significantly older (75.2 v 70.2 years, p < 0.05), had lower left ventricular ejection fraction (44.4% v 49.2%, p < 0.05), a greater preoperative history of congestive heart failure (63 v 44%, p < 0.05), a greater renal risk score (5.8 v 4.9, p < 0.05), and underwent more nonisolated coronary artery bypass graft surgeries (77% v 29%, p < 0.0001). CPB time (126.0 v 96.5 minutes, p < 0.001) and AoXCl duration (100.9 v 78.0 minutes, p < 0.005) were longer in the aprotinin group. Diabetes (60.5% v 41.5%, p < 0.05) and hypertension (90.1% v 73.2%, p < 0.05) were more prevalent in the EACA group. Baseline renal function and renal outcomes were not significantly different between the aprotinin and EACA groups. Six of the 7 patients who required HD received EACA (p ⴝ 0.1). The earliest SCr recorded >3 months after surgery was significantly lower in the aprotinin group compared with the EACA group (1.8 v 2.2 mg/dL, p < 0.05). Conclusion: Acute perioperative renal dysfunction was associated with worse patient outcome and longer CPB and AoXCl times. Demographic and surgical variables indicated that the sicker patients undergoing more complex surgeries were more likely to be treated with aprotinin. Although aprotinin patients had a higher renal risk score, the administration of aprotinin did not negatively impact renal outcome. © 2008 Elsevier Inc. All rights reserved.

I

tality rate.8-11 In comparison, patients with preoperative renal dysfunction (serum creatinine [SCr] ⱖ1.5 mg/dL) are at greater risk for acute worsening of renal function, prolonged mechanical ventilation, increased intensive care unit (ICU) and hospital stays, and experience greater short- and long-term mortality.12-16 For nondialysis patients with SCr ⬎1.7 and ⱖ2.5 mg/dL, the perioperative mortality is incrementally increased and may be as high as 33%.15-18 Bleeding after cardiac surgery is also a major perioperative complication, and as many as 7% of patients in some practices require mediastinal re-exploration.2-4 As a result, the administration of an antifibrinolytic agent such as epsilon aminocaproic acid (EACA) or aprotinin is standard practice for surgical procedures requiring cardiopulmonary bypass (CPB) at some institutions. Agent selection depends on a number of demo-

N CARDIAC SURGICAL PATIENTS with normal preoperative renal function, acute perioperative renal dysfunction/ failure occurs in up to 17%, with less than 5% requiring renal replacement therapy (RRT; hemodialysis [HD] or hemofiltration),1-10 the latter being associated with a 40% to 60% mor-

From the Departments of *Anesthesiology, †General Surgery, and ‡Cardiac Surgery, Brown University Medical School, Rhode Island Hospital, Providence, RI. Address reprint requests to Andrew D. Maslow, MD, 63 Prince Street, Needham, MA 02492. E-mail: [email protected] © 2008 Elsevier Inc. All rights reserved. 1053-0770/08/2201-0004$34.00/0 doi:10.1053/j.jvca.2007.07.017 6

KEY WORDS: creatinine, creatinine clearance, cardiopulmonary bypass, hemodialysis, aprotinin, epsilon aminocaproic acid

Journal of Cardiothoracic and Vascular Anesthesia, Vol 22, No 1 (February), 2008: pp 6-15

RENAL OUTCOME: APROTININ VERSUS EACA

7

graphic and surgical variables and is determined on a case-bycase risk-benefit analysis. The decision to administer aprotinin to patients with preoperative renal dysfunction remains controversial.12,19,20 A retrospective analysis of the authors’ consecutive patients with preoperative renal dysfunction (SCr ⱖ1.8 mg/dL) scheduled for nonemergent cardiac surgical procedures requiring CPB is presented. It was hypothesized that (1) the administration of aprotinin would not be associated with worsening of renal function when compared with EACA, and (2) without randomization, clinicians would use aprotinin more commonly in complex cardiac surgeries with patients who had more preoperative comorbidities. METHODS After approval from the hospital internal research review board, a retrospective analysis was undertaken of 144 consecutive patients with preoperative renal dysfunction (immediate preoperative SCr ⱖ1.8 mg/ dL) scheduled for nonemergent cardiac surgery from a single cardiac center between 2000 and 2007. Exclusion criteria included emergency surgery (n ⫽ 5), active endocarditis (n ⫽ 2), preoperative cardiogenic shock within 1 week of surgery (n ⫽ 2), use of intra-aortic balloon counterpulsation (n ⫽ 2), history of renal transplantation (n ⫽ 2), preoperative hemodialysis (n ⫽ 4), and planned use of deep hypothermic circulatory arrest (n ⫽ 4). The resulting study group consisted of 123 patients. Patients either received EACA (n ⫽ 82) or aprotinin (n ⫽ 41) as agreed on by the attending anesthesiologist and surgeon guided by a subjective assessment of the patient’s risk for bleeding and the planned surgical procedure. The administration of these medications was according to standardized departmental protocol. Dosing of EACA included an initial 10-g loading dose followed by an infusion of 1 g/h with an additional 10-g dose added to the CPB prime. The dosing of aprotinin included an initial 2 million KIU loading dose, followed by an infusion of 500,000 KIU/h, and an additional 2 million KIU added to the CPB prime. Demographic and perioperative data were collected from the Society of Thoracic Surgeons database and from automated intraoperative

anesthesia records. Patients were managed and monitored per institutional routine including standard American Society of Anesthesiologists noninvasive monitors, intra-arterial catheters, and pulmonary artery catheters. Transesophageal echocardiography (TEE) was performed for all valve and aortic surgeries. Thereafter, the decision to perform TEE was determined on a case-by-case basis. CPB consisted of a closed-system membrane oxygenator with centrifugal pumps with normothermic perfusate at 37°C and cold cardioplegic arrest as previously described.21,22 Cardiopulmonary bypass was managed with a systemic flow rate of 2.5 L/min/m2. Hematocrits were maintained above 18% during CPB and then ⬎23% after separation from CPB. An intravenous infusion of phenylephrine or inhaled isoflurane was administered to maintain a mean arterial pressure between 55 and 80 mmHg. All patients received between 0.25% and 0.5% inspired isoflurane during CPB for amnesia. Separation from CPB was achieved with a mean arterial pressure ⬎55 mmHg and a cardiac index ⱖ2.5 L/min/m2. The choice of vasoactive medications was left to the discretion of the attending anesthesiologist. Demographic data included age, sex, body surface area, the presence of hypertension, diabetes, tobacco use, history of congestive heart failure, history of previous cardiac surgical procedures, recent cardiac catheterization (⬍48 hours), and recorded left ventricular ejection fraction (either from cardiac catheterization or from echocardiography). A scoring system to predict acute perioperative renal failure was used to help define risk of acute renal dysfunction between groups. The scoring system was based on more than 15,000 patients, with an acute perioperative renal dysfunction incidence of 1.7%.23 This was then validated with another 15,000 patients. Renal function was assessed from measured SCr and calculated serum creatinine clearances (CrCl) by the following equation developed by Cockroft and Gault24: CrCl ⫽ (140 ⫺ age) ⫻ Wt (kg)/(SCr ⫻ 72) (⫻0.85 for females). Renal dysfunction was defined as a worsening (pre- to postoperative) in SCr or CrCl by ⱖ25% or by the need for HD. Indications for HD included worsening metabolic acidosis (decreasing pH), hyperkalemia, and/or congestive heart failure in conjunction with an elevated or rising SCr and/or oliguria/anuria. The need for hemodialysis was the decision

Table 1. Surgical Procedures All Patients (N ⫽ 123) (%)

Surgical procedure Reoperation CABG alone Valve alone Valve/other CABG CABG/AVR CABG/MVR CABG/AVR/MVR AVR MVR AVR/MVR Aortic surgery Aortic surgery/AVR MVR/TVR PVR/TVR

EACA (n ⫽ 82) (%)

Aprotinin (n ⫽ 41)

p Value

⬍0.0001 12 (9.8) 74 (60.1) 20 (16.2) 29 (23.6) 75 (61.0) (1 reop) 17 (13.9) (1 reop) 4 (3.2) 1 (0.8) 12 (9.7) 7 (5.7) 1 (0.8) 2 (1.6) 2 (1.6) (1 reop) 1 (0.8) (reop) 1 (0.8)

2 (2.4) 63 (76.8) 8 (9.8) 11 (13.4) 63 (77) 7 (8.5) 2 (2.4) 5 (6.1) 2 (2.4) 1 (1.2) 1 (1.2) 1 (1.2) (reop)

10 (24.4) 11 (26.8) 12 (29.2) 18 (43.9) 12 (29) (1 reop) 10 (24.4) (1 reop) 2 (4.9) 1 (2.4) 7 (17.1) 5 (12.2) 2 (4.9) 1 (2.4) (reop) 1 (2.4)

NOTE. Data are listed as number (percentage). The difference in surgical procedures was significantly different (p ⬍ 0.0001) with respect to the performance of nonisolated CABG surgery, valve surgeries, and combination (valve ⫹) surgeries. Abbreviations: AVR, aortic valve replacement; MVR, mitral valve repair or replacement; TVR, tricuspid valve repair; VR, pulmonic valve replacement; reop, reoperation; EACA, epsilon aminocaproic acid.

8

MASLOW ET AL

of the same renal consult team. Renal tests were obtained daily after surgery. The earliest SCr ⱖ3 months after surgery was also recorded. Intraoperative data included surgical procedures, CPB time, aortic crossclamp (AoXCl) time, use of aprotinin or EACA, use of vasoactive

medications to maintain hemodynamics, use of diuretics, intraoperative urine output, and cardiovascular hemodynamics. Postoperative data included surgical re-exploration rate for bleeding, SCr (highest and final pre-discharge), need for RRT, duration of

Table 2. Demographic Variables

Sex (M/F) Age (y, SD) Weight (kg, SD) Height (cm, SD) BSA (m2, SD) CHF (n, %) HTN (n, %) LVEF (%, SD) PVD (n, %) DM (n, %) All IDDM Tobacco (n, %) Preop ACEi (n, %) NYHA (U, SD) Preoperative MI (n, %) LVH (n, %) Preop valve dysfxn AS, severe (n, %) AI (0-4, SD) MR (0-4, SD) Pre-CPB hemodynamics mBP (mmHg, SD) CVP (mmHg, SD) mPAP (mmHg, SD) CO (L/min, SD) Dopamine (ⱕ5 ␮g/kg/min) (n, %) CPB management (SD) Peak serum glucose (mg/dL) First post-CPB glucose (mg/dL) Insulin during CPB (U) Nadir hematocrit (%) Mean systemic BP (mmHg) Temperature CPB (°C) Post-CPB hemodynamics mBP (mmHg, SD) CVP (mmHg, SD) mPAP (mmHg, SD) CO (L/min, SD) Post-CPB vasoactive medications Epinephrine (n, %) PDEi (n, %) Dopamine (n, %) Norepinephrine (n, %) Nitroglycerin (n, %) CPB time (min, SD) AoXCl time (min, SD)

Total (N ⫽ 123)

EACA (n ⫽ 82)

Aprotinin (n ⫽ 41)

90/33 71.7 (11.6) 79.7 (17.4) 168.7 (9.1) 1.91 (0.24) 62 (50.4) 104 (85) 47.6 (13.2) 45 (36.6)

57/25 70.2 (10.6) 80.4 (17.9) 169.5 (9.5) 1.90 (0.24) 36 (44) 71 (90.1) 49.2 (12.5) 31 (40.7)

33/8 75.2 (10.4) 78.1 (16.5) 168.2 (8.6) 1.94 (0.23) 26 (63.4) 29 (73.2) 44.4 (14.1) 14 (34.1)

NS ⬍0.05 NS NS NS ⬍0.05 ⬍0.05 ⬍0.05 NS

66 (53.6) 28 (22.7) 77 (62.6) 40 (32.5) 3.66 (0.52) 74 (60.2) 28 (22.8)

48 (60.5) 22 (27.0) 52 (63.4) 27 (32.9) 3.68 (0.5) 51 (61.7) 17 (21.0)

16 (41.5) 6 (14.6) 25 (59.5) 12 (28.6) 3.63 (0.5) 23 (56.1) 10 (24.4)

⬍0.05 NS NS NS NS NS NS

30 (24.4) 0.5 (1.1) 0.8 (1.3)

13 (16) 0.3 (0.9) 0.5 (1.2)

17 (41.5) 0.9 (1.4) 1.2 (1.6)

0.002 0.004 0.01

74.7(9.6) 7.6 (3.0) 20.8 (6.0) 4.44 (1.2) 55 (45)

75.2 (9.7) 6.9 (2.3) 19.6 (5.5) 4.55 (1.2) 29 (35.8)

73.6 (9.5) 9.1 (3.6) 23.4 (6.1) 4.23 (1.2) 26 (63.4)

NS 0.0002 0.0008 NS 0.003

366.5 (70.5) 285.7 (68.4) 29.2 (17.2) 21.2 (4.1) 62.0 (7.2) 36.4 (0.2)

364.3 (79.8) 294.8 (69.4) 26.8 (15.0) 21.8 (4.4) 62.2 (7.7) 36.5 (0.3)

371.7 (45.4) 272.6 (65.8) 32.0 (21.1) 20.4 (3.0) 61.5 (6.3) 36.4 (0.2)

NS NS NS NS NS NS

67.3 (7.5) 7.6 (2.7) 19.5 (4.5) 5.87 (1.4)

67.8 (8.1) 7.2 (2.6) 18.6 (4.2) 6.06 (1.4)

66.1 (6.0) 8.2 (2.9) 21.1 (4.8) 5.55 (1.5)

NS 0.06 0.003 0.09

34 (28) 98 (80) 82 (67) 50 (41) 61 (50) 106.4 (46.4) 85.8 (38.8)

15 (18.5) 62 (75.3) 55 (66.7) 25 (30.9) 38 (46.9) 96.5 (40.9) 78.0 (35.7)

18 (43.9) 36 (87.8) 29 (70.7) 24 (58.5) 22 (53.7) 126.0 (50.8) 100.9 (40.5)

0.003 NS NS 0.003 NS 0.0008 0.002

p Value

NOTE. p Values represent comparisons between the EACA and aprotinin groups. The data are presented as continuous variables and standard deviation (SD) or as number and percentage. Abbreviations: BSA, body surface area; CHF, congestive heart failure; HTN, hypertension; LVEF, left ventricular ejection fraction; PVD, peripheral vascular disease; DM, diabetes mellitus; IDDM, insulin-dependent diabetes mellitus; ACEi, angiotensin-converting enzyme inhibitor; MI, myocardial infarction; NYHA, New York Heart Association Risk Score; LVH, left ventricular hypertrophy; AS, aortic valve stenosis; AI, aortic insufficiency; MR, mitral valve regurgitation; mBP, mean systemic blood pressure; CVP, central venous pressure; mPAP, mean pulmonary artery pressure; CO, cardiac output; PDEi, phosphodiesterase inhibitor; AoXCl, aortic crossclamp time; EACA, epsilon aminocaproic acid; NS, not significant.

RENAL OUTCOME: APROTININ VERSUS EACA

9

Table 3. Measurements of Renal Function and Renal Outcomes Total (N ⫽ 123)

Acute renal index (U, SD) (Thakar et al23) Creatinine (mg/dL, SD) Preoperative Highest postoperative Final postoperative ⱖ3 months after surgery Changes in creatinine (n, %) 25% increase pre-high 50% increase pre-high 25% increase pre-final 50% increase pre-final Creatinine clearance (CrCl) (mL/min, SD) Preoperative Lowest postoperative Final postoperative ⬎3 months after surgery Changes in CrCl (n, %) 25% decrease pre-low 50% decrease pre-low 25% decrease pre-final 50% decrease pre-final Hemodialysis (n, %) Infection (n, %) Leg infection Sternal superficial Sternal deep Septicemia Stroke (n, %) Intraoperative PRBCs (n, %) Postoperative PRBCs (n, %) Intraoperative platelets (n, %) Postoperative platelets (n, %) Intraoperative FFP/cryo (n, %) Postoperative FFP/cryo (n, %) Re-exploration bleeding (n, %) ICU (h, SD) Mech vent (h, SD) Mech vent (⬎1 day) (n, %) Hospital stay (days, SD) Death (n, %)

EACA (n ⫽ 82)

5.20 (2.05)

4.89 (1.93)

2.34 (1.0) 2.87 (1.4) 2.16 (1.1) 2.08 (0.9)

2.42 (1.2) 2.90 (1.5) 2.16 (1.2) 2.20 (1.0)

Aprotinin (n ⫽ 41)

5.8 (2.16) 2.20 (0.6) 2.82 (1.1) 2.16 (1.0) 1.81 (0.5)

p Value

⬍0.05 NS NS NS ⬍0.05

51 (41.5) 24 (19.5) 13 (10.6) 6 (4.9)

32 (39.0) 15 (18.3) 7 (8.5) 3 (3.6)

19 (46.3) 9 (22) 6 (14.6) 3 (7.3)

NS NS NS NS

34.9 (12.3) 30.5 (13.7) 41.4 (19.9) 42.0 (20.4)

35.5 (13.1) 31.4 (14.2) 42.5 (19.6) 41.5 (19.6)

33.9 (10.8) 28.7 (12.7) 39.1 (20.4) 43.1 (22.1)

NS NS NS NS

40 (32.5) 7 (5.7) 8 (6.5) 4 (3.2) 7 (5.7)

24 (29.3) 4 (4.9) 4 (4.9) 3 (3.7) 6 (7.3)

16 (39) 3 (7.3) 4 (9.8) 1 (2.4) 1 (2.4)

NS NS NS NS NS

0 1 (2.4) 1 (2.4) 3 (7.3) 2 (4.9) 37 (90.2) 31 (75.6) 22 (53.7) 7 (17.1) 13 (31.7) 16 (39.0) 1 (2.5) 184.6 (242.4) 25.2 (26.8) 12 (31.7) 15.6 (9.6) 4 (9.8)

NS NS NS NS NS NS NS NS NS ⬍0.05 ⬍0.05 NS 0.004 NS ⬍0.05 NS NS

4 (3.2) 3 (2.4) 2 (1.6) 8 (6.4) 6 (4.8) 112 (91.1) 95 (77.2) 65 (52.8) 29 (23.6) 23 (18.7) 38 (30.9) 3 (2.4) 121.0 (170.3) 25.2 (37.2) 27 (21.9) 13.6 (12.2) 10 (8.1)

4 (4.9) 2 (2.5) 1 (1.2) 5 (6.2) 4 (4.9) 75 (91.4) 64 (78.0) 43 (52.4) 22 (26.8) 10 (12.2) 22 (26.8) 2 (2.5) 73.4 (45.4) 25.52 (43.7) 11 (16) 12.8 (13.3) 6 (7.3)

NOTE. p Values represent comparisons between EACA and aprotinin groups. The data are presented as continuous variables and standard deviation (SD) or as number (n) and percentage (%). Abbreviations: Mech vent, mechanical ventilation; PRBCs, packed red blood cells; FFP, fresh frozen plasma; EACA, epsilon aminocaproic acid.

mechanical ventilation, blood transfusion, occurrence of infection, ICU and hospital stays, and mortality. Data were recorded as mean and standard deviation or percentage of population depending on whether the data were continuous or not. Comparisons between different groups were compared by using an analysis of variance. Within-group and data correlations were assessed by using chi-square analysis, Student paired t tests, Fisher exact test, Wilcoxon rank sum test, and Mann-Whitney U tests. Multivariate analysis was performed. A p value ⬍0.05 was considered significant. RESULTS

One hundred twenty-three patients were included in the analysis (Tables 1 and 2). Overall, acute perioperative renal dysfunction (ⱖ25% decrement in renal function) or the need

for HD occurred in 41.5% (based on SCr) and 5.7%, respectively (Table 3). Thirty-two percent of patients had acute perioperative renal dysfunction as measured by CrCl. A preoperative SCr of ⱖ2.5 mg/dL had a trend toward acute perioperative renal dysfunction (p ⫽ 0.08) and the need for HD (13.6% v 4.0%, p ⫽ 0.08). The mortality for patients with a preoperative SCr ⱖ2.5 mg/dL was 13.6% compared with 6.0% for those with preoperative SCr between 1.8 and 2.5 mg/dL (p ⫽ not significant). A preoperative CrCl ⱕ30 mL/min was not predictive of acute perioperative renal outcome or of mortality. Additional analyses were unable to show a significant predictive value for CrCl of ⱕ25, 35, or 40 mL/min. The preoperative SCr and CrCl for patients with and without acute perioperative renal dysfunction were similar.

10

MASLOW ET AL

Table 4. Change in SCr From Preoperative Period to the Highest Postoperative SCr ⬍25% Change (n ⫽ 72)

Age (y) CHF (n, %) IDDM (n, %) DM (n, %) ARI (U, SD) HTN (n, %) PVD (n, %) LVH (n, %) Tobacco (n, %) EACA/aprotinin (n/n) NYHA (U, SD) Preoperative LVEF (%, SD) Nonisolated CABG (n, %) CPB time (min, SD) AoX time (min, SD) CPB management Peak serum glucose (mg/dL, SD) First post-CPB glucose (mg/dL, SD) Insulin during CPB (U, SD) Nadir hematocrit (%, SD) mBP (mmHg, SD) Temperature CPB (°C, SD) ICU stay (h, SD) Vent (h, SD) Vent ⬎1 day (no, %) Surgical stay (d, SD) Serum creatinine (mg/dL, SD) Preoperative Highest postoperative Final postoperative ⱖ3 months after surgery Changes in creatinine (n, %) 25% increase pre-high 50% increase pre-high 25% increase pre-final 50% increase pre-final Creatinine clearance (mL/min, SD) Preoperative Lowest postoperative Final postoperative ⱖ3 months after surgery Changes in CrCl (n, %) 25% decrease pre-low 50% decrease pre-low 25% decrease pre-final 50% decrease pre-final Septicemia (n, %) Dialysis (n, %) Death (n, %)

71.8 (11.4) 34 (47) 13 (18.3) 36 (48.6) 5.1 (2.0) 61 (84.7) 28 (38.9) 14 (18.1) 45 (63.4) 50/22 3.64 (0.54) 49.1 (13.1) 22 (30.1%) 98.6 (40.4) 78.1 (32.5) 351.3 (60.0) 280.7 (60.5) 28.3 (14.3) 21.9 (5.2) 63.0 (5.0) 36.5 (0.3) 75.6 (51.5) 24.7 (41.1) 11 (15.3%) 10.5 (5.5) 2.37 (1.2) 2.31 (1.0) 1.80 (0.7) 1.86 (0.8) – 0% 0% 0% 35.7 (13.1) 36.2 (13.7) 47.0 (20.9) 47.2 (22.1) 0 0 0 0 2 (2.8) 0 3 (4.2)

ⱖ25% Change (n ⫽ 51)

p Value

71.5 (12.1) 28 (56) 14 (29.2) 30 (62.0) 5.3 (2.1) 41 (84.0) 19 (38.0) 14 (28.0) 31 (64.6) 32/19 3.70 (0.50) 45.4 (13.3) 18 (34.7) 117.7 (52.3) 97.1 (44.5)

NS NS NS NS NS NS NS NS NS NS NS NS NS ⬍0.05 0.008

380.6 (75.6) 292.3 (70.1) 30.3 (20.7) 21.1 (2.5) 62.0 (6.0) 36.2 (0.2) 192.2 (251.5) 26.0 (30.7) 15 (29.4%) 18.3 (16.9)

NS NS NS NS NS NS 0.003 NS ⬍0.05 0.0005

2.31 (0.8) 3.68 (1.4) 2.67 (1.4) 2.43 (0.9)

NS ⬍0.0001 ⬍0.0001 0.001

– 22 (46.0) 11 (24.0) 5 (10)

⬍0.0001 ⬍0.0001 0.006

34.1 (11.5) 22.3 (8.8) 33.2 (15.0) 33.4 (13.5)

NS ⬍0.0001 0.0001 0.0005

40 (78.0) 6 (12.0) 7 (14.0) 3 (6.0) 6 (12.0) 7 (14.0) 7 (13.7)

⬍0.0001 0.002 0.0009 ⬍0.05 ⬍0.05 ⬍0.001 NS

NOTE. Preoperative medications, valve dysfunction, and surgical procedures were not significantly different. The data are presented as continuous variables and standard deviation (SD) or as number and percentage. Abbreviations: BSA, body surface area; CHF, congestive heart failure; HTN, hypertension; LVEF, left ventricular ejection fraction; PVD, peripheral vascular disease; DM, diabetes mellitus; IDDM, insulin-dependent diabetes mellitus; ACEi, angiotensin-converting enzyme inhibitor; MI, myocardial infarction; EACA, epsilon aminocaproic acid; NYHA, New York Heart Association risk score; LVH, left ventricular hypertrophy; AS, aortic valve stenosis; AI, aortic insufficiency; MR, mitral valve regurgitation; mBP, mean systemic blood pressure.

Table 4 presents comparisons between patients with and without acute perioperative renal dysfunction based on measured SCr. Demographic variables, surgical procedures, and management during CPB were not significantly different

between those patients with or without acute perioperative renal dysfunction. Of the variables studied, only prolonged CPB and AoXCl times were associated with a ⱖ25% increase in SCr (CPB

RENAL OUTCOME: APROTININ VERSUS EACA

117.7 v 98.6 minutes, p ⬍ 0.05; AoXCl 97.1 v 78.1 minutes, p ⬍ 0.01) and a ⱖ25% decrease in CrCl (CPB 118.2 v 99.5 minutes, p ⬍ 0.05; AoXCl 97.6 v 80.4 minutes, p ⬍ 0.01). Increased CPB and AoXCl times were also associated with greater transfusion of packed red blood cells (p ⫽ 0.0004), fresh frozen plasma (p ⬍ 0.0001), cryoprecipitate (p ⫽ 0.001), and platelets (p ⫽ 0.0004); greater administration of post-CPB epinephrine (p ⬍ 0.05) and norepinephrine (p ⬍ 0.05); duration of mechanical ventilation (p ⬍ 0.05); and length of hospital stay (p ⬍ 0.05). Outcomes based on CrCl were similar to outcomes based on SCr. For both measures, a 25% decrement in renal function was associated with increased adverse outcome. Patients with acute perioperative renal dysfunction had longer ICU stays, hospital stays, greater postoperative blood transfusions, a higher incidence of septicemia, and a greater need for hemodialysis (Table 4). Acute perioperative renal dysfunction was associated with a 14% to 15% incidence of HD compared with 0% to 1.2% for patients without acute perioperative renal dysfunction (p ⱕ 0.001). Although mortality for acute perioperative renal dysfunction was 12.5% to 13.7% (based on SCr and CrCl, respectively), this was not statistically significantly different for patients without acute perioperative renal dysfunction (4.2%6.0%). Seven patients required hemodialysis. Compared with those patients who did not require HD, the preoperative SCr (⫹HD 2.7 ⫾ 1.3 v ⫺HD 2.3 ⫾ 1.1 mg/dL, p ⫽ 0.4) and CrCl (⫹HD 31.4 ⫾ 11.8 v ⫺HD 35.2 ⫾ 12.4 mL/min, p ⫽ 0.4) were not significantly different. Three of these 7 died (43%), 2 did not require prolonged dialysis, and 2 were discharged while still receiving dialysis. Six of these 7 had complicated postoperative periods including 4 patients with acute respiratory distress syndrome and sepsis syndrome. A fifth patient had prolonged vasoplegia (⬎48 hours), and a sixth patient required mediastinal re-exploration for hypotension and tamponade. A seventh patient had a preoperative SCr of 5.5 mg/dL. These 7 patients had 4-fold longer ICU stays (446 v 112 hours, p ⬍ 0.01), a 3-fold longer duration of mechanical ventilation (75 v 24 hours, p ⬍ 0.05), and 4-fold longer hospital stays (47 days v 12 days, p ⬍ 0.0001). Patients requiring HD had a 57% incidence of major infections compared with 3.5% for the remainder of the study patients (p ⬍ 0.0001). Mortality was 43% (3/7) for patients requiring HD compared with 4.3% for patients not requiring HD (p ⬍ 0.0001). Eighty-two patients received EACA, and 41 received aprotinin (Tables 2 and 3). Patients receiving aprotinin were significantly older (75.2 v 70.2, p ⬍ 0.05), had lower preoperative left ventricular ejection fraction (44.4 v 49.2, p ⬍ 0.05), a greater incidence of congestive heart failure (63.4 v 44.0%, p ⬍ 0.05), and more frequent coexistent valvular dysfunction (Tables 2 and 3). The incidences of diabetes mellitus and hypertension were higher in the EACA group. The acute renal index described by Thakar et al23 was significantly greater for the aprotinin patients (5.8 v 4.9, p ⫽ 0.02). There were no significant differences in other demographic variables between these 2 groups. Intraoperative data are presented in Table 2. CPB (126.0 v 96.5, p ⬍ 0.001) and AoXCl (100.9 v 78.0 minutes, p ⬍ 0.005) times were significantly longer in the aprotinin patients. The

11

aprotinin group more frequently underwent nonisolated coronary artery bypass graft (CABG) surgical procedures (77.0 v 29.0%, p ⬍ 0.0001) and combination (valve ⫹) procedures (43.9% v 13.4%, p ⬍ 0.0001). Before and after CPB, aprotinin patients tended to have higher central filling pressures (mean pulmonary artery pressure [mPAP]) and a trend toward lower post-CPB cardiac outputs despite a greater use of post-CPB inotropic support (Table 2). Aprotinin patients had greater durations of mechanical ventilation and ICU stays (Table 3), and received a greater amount of component blood therapy (fresh frozen plasma/cryoprecipitate) (Table 3). The transfusion of packed red blood cells, platelets, and the need for re-exploration were not different between the 2 groups. Hospital stay and mortality were not significantly different between the 2 groups. Baseline preoperative and postoperative SCr and CrCl were not significantly different between the EACA and aprotinin groups (Table 3). Postoperative worsening of renal function was not associated with the use of EACA or aprotinin. The need for HD was greater for the EACA group (7.3% v 2.4%) but did not reach statistical significance (p ⫽ 0.1). Serum creatinine values obtained 3 months or longer after surgery were significantly lower in the aprotinin patients (1.8 v 2.2 mg/dL, p ⬍ 0.05). Nonrenal outcomes are presented in Table 3. There were 6 perioperative strokes (4.8%), 3 of whom died. The surviving 3 had complete or near-complete resolution of their deficits. Four of these patients received EACA, and 2 received aprotinin. All but 1 involved valve surgery. There were 10 deaths (8.1%). Isolated CABG surgery was performed in 4 of these patients. Six patients who died received EACA (7.3%), whereas 4 received aprotinin (9.8%). Nonisolated CABG surgery required longer CPB (129.8 ⫾ 51 v 90.8 ⫾ 35.2 minutes, p ⬍ 0.0001) and AoXCl (103.3 ⫾ 43.1 v 73.8 ⫾ 30.9 minutes, p ⬍ 0.0001) times, had a higher incidence of stroke (10.2 v 1.3%, p ⫽ 0.005), and greater mortality (12.2% v 5.4%). These patients received more blood transfusions, had longer durations of mechanical ventilation

Table 5. Renal Data and Outcome for Patients Younger Than 75 Years Compared With Those Aged 75 and Older Age (y)

Weight (kg) Creatinine (mg/dL) Preoperative Highest postoperative Final postoperative ⱖ3 months after surgery Creatinine clearance (mL/ min) Preoperative Lowest postoperative Final postoperative ⬎3 months after surgery Dialysis Death

⬍75 (n ⫽ 57)

ⱖ75 (n ⫽ 62)

p Value

81.9 (19.4)

77.8 (15.5)

NS

2.54 (1.36) 2.97 (1.45) 2.20 (1.11) 2.17 (1.00)

2.18 (0.56) 2.72 (1.18) 2.04 (1.0) 1.99 (0.84)

0.06 NS NS NS

39.0 (14.2) 34.8 (16.3) 48.4 (3.2) 49.0 (25.5) 3 (5.3%) 5 (8.8%)

31.5 (9.2) 27.1 (9.7) 35.9 (11.4) 36.5 (12.1) 4 (6.5%) 5 (8.1%)

0.0008 0.002 0.0005 0.001 NS NS

NOTE. Data are displayed as either mean (SD) or number (percentage of group); p values compare with the 2 age groups.

12

MASLOW ET AL

(37.5 ⫾ 52.2 v 17.4 ⫾ 20.1 hours, p ⬍ 0.05), and longer hospital stays (16.1 ⫾ 9.4 v 12.1 ⫾ 15.4 days, p ⫽ 0.08). Age (ⱖ75 or ⱖ80 years old) was not associated with renal outcomes or death. The older patients (ⱖ75 years) had lower SCr (p ⫽ not significant) at all measured points (preoperative, highest postoperative, final postoperative, and ⱖ3 months after surgery); the calculated CrCl was also significantly (p ⬍ 0.005) lower for these patients compared with younger patients (Table 5). DISCUSSION

As defined in this study, acute perioperative renal dysfunction occurred in up to 41% of patients and was associated with adverse outcome. Of the variables studied, only CPB and AoXCl times were significantly associated with acute perioperative renal dysfunction. In comparison to the EACA group, patients receiving aprotinin were older, had greater ventricular dysfunction, underwent more nonisolated CABG surgery, experienced longer CPB and AoXCl times, and started surgery with a higher renal risk score. Despite these adverse biases, acute renal outcome was not significantly different between the aprotinin and EACA groups. Although no significant differences in acute perioperative renal outcome were found between these 2 groups, the SCr measured ⱖ3 months after surgery was significantly lower in the aprotinin group. Regarding patient outcome, the measurement of renal function using CrCl did not contribute beyond that obtained from the perioperative SCr data. It did, however, show that the preoperative assessment of renal function based on SCr alone may not adequately describe the extent of renal dysfunction present, as displayed in Table 5. This has been previously reported and described as “occult renal dysfunction” and was more likely found in elderly patients with lower body mass.9 Patients considered as high risk (preoperative CrCl ⱕ60 mL/min) for acute renal dysfunction were sought for the present study.5,25-27 Because the surgical database included data on SCr and not CrCl, patients were enrolled based on the preoperative SCr. Previous literature describes an incremental surgical risk for nondialysis patients with preoperative SCr between 1.8 mg/dL (approximately 150 ␮mol/L) and 2.5 mg/dL and then an SCr ⬎2.5 mg/dL.15-18,28 Grayson et al28 found a significant demarcation in adverse renal and patient outcome between patients with preoperative SCr less than or greater than 150 ␮mol/L (approximately 1.8 mg/dL).28 Therefore, patients with preoperative SCr ⱖ1.8 mg/dL were included in this study. In this dataset, only 4 patients had a preoperative CrCl ⬎60 mL/min (62.5, 66.3, 70.6, and 74.0 mL/min). The latter 3 of these received EACA. For these 4 patients, the perioperative period was uneventful. Excluding these 4 patients would not change the study’s results. The renal outcome data presented here are similar to previous investigations in regard to the impact of preoperative and acute perioperative renal dysfunction on patient outcome.5,12,23,25,29 In the present study, the overall need for HD was 5.7%, and the group’s mortality was 8.1%. For patients with a preoperative SCr ⱖ2.5 mg/dL, the need for HD and mortality were both 13.6%, compared with 3.1% and 6.0%, respectively, for those with preoperative SCr between 1.8 and 2.5 mg/dL. Of patients with acute renal dysfunction, 14% required HD and 13.7% died, compared

with 0% and 4.2%, respectively, for patients without acute renal dysfunction. Three of 7 (43%) patients requiring HD died. Previous investigations have reported a 9% to 13% incidence of HD for patients with a preoperative SCr between 133 and 300 ␮mol/L (approximately SCr 1.4-2.1 mg/dL).9,18,28 Mortality for patients with preoperative renal dysfunction varies from as low as 4% to 11% for patients with preoperative SCr between 1.5 and 2.5 mg/dL to 17% to 33% for nondialysis patients with preoperative SCr ⱖ2.5 mg/dL.15-17 Depending on its definition, the mortality associated with acute perioperative renal dysfunction ranges from 20% to 41% compared with less 5% for those without acute renal injury.7,8 The mortality associated with acute renal dysfunction requiring HD may be as high as 60% to 70%.7,8,11,17,18 Acute renal dysfunction, in this (25% decrement) and other studies (20-100% decrement), was associated with a host of adverse outcomes including prolonged ICU and hospital stays, blood transfusions, infections, HD, and a trend toward increased mortality.7-10,15-18 The likelihood of developing acute perioperative renal dysfunction after cardiac surgery is increased with preoperative renal dysfunction, poor cardiac function during and/or after surgery, and/or a complicated perioperative course.5-11,15,16,23,26,28,30,31 These risk factors include but are not limited to the need for urgent/emergent surgery, perioperative hemodynamic dysfunction (reduced left ventricular ejection fraction, need of inotropes, intraaortic balloon pump, and low cardiac output), prolonged CPB (⬎2 hours) and AoXCl times, and nonisolated CABG surgery.8-11,15,16,25,28 In the present study, the durations of CPB and AoXCl were associated with significant worsening of renal function. Because patients with and without acute perioperative renal dysfunction did not differ regarding baseline renal function, demographics, or surgical procedures, the authors speculate that perioperative events, such as CPB and AoXCl times, and complicated postoperative courses are more significant contributors to renal outcome than the administration of antifibrinolytics. Six of the 7 patients requiring HD had a complicated postoperative course. This is similar to the data reported by Mora-Mangano et al,11 in which 80% of patients with acute severe postoperative renal dysfunction had experienced perioperative hemodynamic instability or hemorrhage.11 These data highlight the importance of perioperative events as they contribute to acute renal dysfunction.5,11,23 Aprotinin and its effects on renal function have been the subject of recent intense debate and controversy.12,20,32,33 Aprotinin has a high affinity for renal tissue and is rapidly eliminated from the circulation by glomerular filtration (t1/2 ⫽ 1-2 hours). After 4 hours, 80% to 90% of the administered dose is stored in the proximal convoluted tubular cells before excretion as active protein. Whether or not aprotinin has been associated with renal dysfunction may depend on the experimental design and/or definition of renal dysfunction.12,34-38 Animals receiving aprotinin have shown reductions in renal blood flow and glomerular filtration rate,34 whereas transient increases in SCr (⬎0.5 mg/dL) in patients receiving aprotinin have been reported in humans requiring CPB.35 Although these differences were statistically significant, it is unclear if these changes were clinically significant.19

RENAL OUTCOME: APROTININ VERSUS EACA

13

Other experimental data have shown that aprotinin protects renal tissue by decreasing tubular cellular apoptosis in a renal ischemia/reperfusion rat model.36,37 These conclusions would be in stark contrast with a recently published observational study that associated aprotinin use with a doubling in the risk of renal failure requiring dialysis.12 The displacement of intracellular potassium and inhibition of renal tubular reabsorption of small filtered proteins have been reported with EACA administration.39 Reports of acute renal failure40 and severe proteinuria41 associated with the use of EACA have also been published. Furthermore, EACA is associated with a significant increase in excretion of B2 microglobulin, a protein linked to specific tubular injury and damage.42 A small study comparing the effects of aprotinin and EACA on outcome after aortic surgery requiring deep hypothermic circulatory arrest found an increase in renal dysfunction for patients receiving EACA.43 Nevertheless, larger investigations did not show an association between the use of EACA and renal outcome.12,44-47 These varied findings could reflect differences in study methodologies as well as the heterogeneity of patients and surgical procedures.12,19,36-38,47,48 The present study enrolled patients with preoperative renal dysfunction (SCr ⬎1.8 mg/ dL) from a single site for which intraoperative and postoperative care is given by protocols. The previously published investigations by Mangano and colleagues12,48 enrolled adult cardiac surgical patients regardless of preoperative renal function from approximately 69 sites around the world, introducing a number of different variables regarding perioperative care. Although Mangano et al, in an attempt to compare their study groups, used various statistical methods to account for the very large number of demographic and perioperative variables, the present study made no such attempt. One goal of the present study was to assess for demographic and perioperative differences between patients who received EACA and those who received aprotinin. These findings reflect day-to-day decision-making by the clinicians at this institution. Patients receiving aprotinin were older, had greater cardiac dysfunction, and underwent a greater percentage of nonisolated CABG surgery, the latter of which was characterized by longer CPB and AoXCl times. The longer CPB and AoXCl times were further associated with greater transfusion and ICU stays. These data confirm that patients with more comorbidities and patients undergoing more complex surgeries received aprotinin preferentially by clinicians at this institution. Hence, these more complicated patients would be expected to have greater renal and other perioperative end-organ morbidity, regardless of which antifibrinolytic was selected. There was an insignificant 7% difference in acute renal injury seen between the EACA and aprotinin groups and a trend toward a greater need for HD in the EACA group. Midterm SCr (ⱖ3 months after surgery) was significantly lower for aprotinin patients. These findings occurred despite the

increased renal risk in aprotinin patients. Although these data may suggest a protective renal effect from aprotinin, alternative explanations may include differences in etiology of preoperative renal dysfunction and/or different effects on renal hemodynamics associated with the varied cardiac surgical procedures. For example, renal dysfunction caused by severe valvular dysfunction would be expected to improve after corrective cardiac surgery, whereas renal dysfunction caused by chronic hypertension and/or diabetes may not. Such conclusions, however, are only speculative and based on assumptions made from differences in demographics and surgical procedures. The number of patients studied is a significant limitation, especially when considering the retrospective and nonrandomized methodology of this study. This is further compounded by the demographic and surgical differences between the 2 groups. The small patient population may contribute to the lack of statistically significant differences in renal outcome (7% difference) and mortality (2.5% difference) between EACA and aprotinin groups. However, a similar statement could be made with regard to HD, which occurred 4.9% more in patients receiving EACA. Furthermore, the midterm SCr was lower in the aprotinin patients. When these data are considered, along with the greater risk for renal injury for the aprotinin patients, it might be speculated that a renal benefit may exist with the use of aprotinin. Nevertheless, the lack of power makes statements regarding the renal safety of aprotinin, or even EACA, premature. Instead, the authors suspect that renal outcome is more related to other perioperative events. There is little argument that a double-blinded, randomized controlled study could provide a direct comparison between aprotinin and EACA. For reasons described earlier, this was not a randomized prospective study nor were statistical analyses used to attempt to control the study populations; rather, it was an attempt to analyze the day-to-day practice at 1 institution. Therefore, it reports on a “real-world” analysis of 1 hospital’s experience in caring for patients with preoperative renal dysfunction undergoing cardiac surgery with CPB. In this setting, patients are at increased risk for renal injury based on demographic, surgical, and perioperative variables and not likely on the choice of antifibrinolytic. It is concluded that patients who received aprotinin were different from those who received EACA. On a day-to-day basis, clinicians administered aprotinin to patients who were at greater risk for renal injury based on demographic and procedural variables. Despite these differences in risk, acute renal outcome was not associated with the choice of antifibrinolytic agent but instead with duration of CPB and AoXCl and postoperative complications. The results of the present study show that direct comparisons of renal outcome between EACA and aprotinin are difficult to make based on nonrandomized and/or retrospective data.

REFERENCES 1. Loef BG, Epema AH, Smilde TD, et al: Immediate postoperative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival. J Am Soc Nephrol 16:195-200, 2005

2. Lytle BW, Loop FD, Cosgrove DM, et al: Fifteen hundred coronary reoperations: Results and determinants of early and late survival. J Thorac Cardiovasc Surg 93:847-859, 1987

14

3. Dacey LJ, Munoz JJ, Baribeau YR, et al: Northern New England Cardiovascular Disease Study Group: Reexploration for hemorrhage following coronary artery bypass grafting. Arch Surg 133:442-447, 1998 4. Zanardo G, Michielon P, Paccagnella A, et al: Acute renal failure in the patient undergoing cardiac operation. Prevalence, mortality rate, and main risk factors. J Thorac Cardiovasc Surg 107:1489-1495, 1994 5. Chertow GM, Lazarus JM, Christiansen CL, et al: Preoperative renal risk stratification. Circulation 95:878-884, 1997 6. Bucerius J, Gummert JF, Borger MA, et al: Stroke after cardiac surgery: A risk factor analysis of 16,184 consecutive adult patients. Ann Thorac Surg 75:472-478, 2003 7. Stallwood MI, Grayson AD, Mills K, et al: Acute renal failure in coronary artery bypass surgery: Independent effect of cardiopulmonary bypass. Ann Thorac Surg 77:968-972, 2004 8. Bovre T, Calabro MG, Landoni G, et al: The incidence and risk of acute renal failure after cardiac surgery. J Cardiothorac Vasc Anesth 18:442-445, 2004 9. Wijeysundera DN, Karkouti K, Beattie WS, et al: Improving the identification of patients at risk of postoperative renal failure after cardiac surgery. Anesthesiology 104:65-72, 2006 10. Provenchere S, Plantefeve G, Hufnagel G, et al: Renal dysfunction after cardiac surgery with normothermic cardiopulmonary bypass: Incidence, risk factors, and effect on clinical outcome. Anesth Analg 96:1258-1264, 2003 11. Mora-Mangano C, Diamondstone LS, Ramsay JG, et al: Renal dysfunction after myocardial revascularization: Risk factors, adverse outcome, and hospital resource utilization. Ann Intern Med 128:194203, 1998 12. Mangano DT, Tudor IC, Dietzel C, for the Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation: The risk associated with aprotinin in cardiac surgery. N Engl J Med 354:353-365, 2006 13. Anderson RJ, O’Brien M, Mawhinney S, et al: Renal failure predisposes patients to adverse outcome after coronary artery bypass surgery. Kidney Int 55:1057-1062, 1999 14. Nakayama Y, Sakata R, Ura M, et al: Long-term results of coronary artery bypass grafting in patients with renal insufficiency. AnnThorac Surg 75:496-500, 2003 15. Weerasinghe A, Hornick P, Smith P, et al: Coronary artery bypass grafting in nondialysis-dependent mild-to-moderate renal dysfunction. J Thorac Cardiovasc Surg 121:1083-1089, 2001 16. Penta de Peppo A, Nardi P, De Paulis R, et al: Cardiac surgery in moderate-to-end-stage renal failure: Analysis of risk factors. Ann Thorac Surg 74:378-383, 2002 17. Durmaz L, Buket S, Atay Y, et al: Cardiac surgery with cardiopulmonary bypass in patients with chronic renal failure. J Thorac Cardiovasc Surg 118:306-315, 1999 18. Durmaz I, Yagdi T, Calkavur T, et al: Prophylactic dialysis in patients with renal dysfunction undergoing on-pump coronary artery bypass surgery. Ann Thorac Surg 75:859-864, 2003 19. Cosgrove DM, Heric B, Lytle BW, et al: Aprotinin therapy for reoperative myocardial revascularization: A placebo-controlled study. Ann Thorac Surg 54:1031-1036, 1992 20. Karkouti K, Beattie WS, Dattilo KM, et al: A propensity score case-control comparison of aprotinin and tranexamic acid in hightransfusion-risk cardiac surgery. Transfusion 46:327-338, 2006 21. Singh AK, Feng WC, Bert AA, et al: Warm body, cold heart surgery. Clinical experience in 2817 patients. Eur J Cardiothorac Surg 7:225-230, 1993 22. Bert AA, Stearns GT, Feng W, et al: Normothermic cardiopulmonary bypass. J Cardiothorac Vasc Anesth 11:91-99, 1997 23. Thakar CV, Arrigain S, Worley S, et al: A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol 16:162-168, 2005

MASLOW ET AL

24. Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine. Nephron 16:31-44, 1976 25. Chertow GM, Levy EM, Hammermeister KE, et al: Independent association between acute renal failure and mortality following cardiac surgery. Am J Med 104:343-348, 1998 26. Holzmann MJ, Ahnve S, Hammar N, et al: Creatinine clearance and risk of early mortality in patients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 130:746-752, 2005 27. Asimakopoulos G, Karagounis P, Valencia O, et al: Renal function after cardiac surgery off- versus on-pump coronary artery bypass: Analysis using the Cockroft-Gault formula for estimating creatinine clearance. Ann Thorac Surg 79:2024-2031, 2005 28. Grayson AD, Khater M, Jackson M, et al: Valvular heart operation is an independent risk factor for acute renal failure. Ann Thorac Surg 75:1829-1835, 2003 29. Kuitunen A, Vento A, Suojaranta-Ylinen R, et al: Acute renal failure after cardiac surgery: Evaluation of the RIFLE classification. Ann Thoracic Surg 81:542-546, 2006 30. Corwin HL, Sprague SM, Delaria GA, et al: Acute renal failure associated with cardiac operations. J Thorac Cardiovasc Surg 98:11011112, 1989 31. Chukwuemeka A, Weisel A, Maganti M, et al: Renal dysfunction in high-risk patients after on-pump and off-pump coronary artery bypass surgery: A propensity score analysis. Ann Thorac Surg 80: 2148-2154, 2005 32. Body SC, Mazer CD: Pro: Aprotinin has a good efficacy and safety profile relative to other alternatives for prevention of bleeding in cardiac surgery. Anesth Analg 103:1354-1359, 2006 33. Beattie WS, Karkouti K: Con: Aprotinin has a good efficacy and safety profile relative to other alternatives for prevention of bleeding in cardiac surgery. Anesth Analg 103:1360-1364, 2006 34. Maier M, Starlinger M, Zhegu Z, et al: Effects of the protease inhibitor aprotinin on renal hemodynamics in the pig. Hypertension 7:32-38, 1985 35. D’Ambra MN, Akins CW, Blackstone EH, et al: Aprotinin in primary valve replacement and reconstruction: A multicenter doubleblind, placebo-controlled trial. J Thorac Cardiovasc Surg 112:10811089, 1996 36. Kher A, Meldrum KK, Hile KL, et al: Aprotinin improves kidney function and decreases tubular cell apoptosis and proapoptotic signaling after renal ischemia-perfusion. J Thorac Cardiovasc Surg 130:662-669, 2005 37. Schweizer A, Hohn L, Morel DR, et al: Aprotinin does not impair renal haemodynamics and function after cardiac surgery. Br J Anaesth 84:16-22, 2000 38. 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 128:442-448, 2004 39. Perazella MA, Biswas P: Acute hyperkalemia associated with intravenous epsilon-aminocaproic acid therapy. Am J Kidney Dis 33: 782-785, 1999 40. Biswas CK, Milligan DA, Agte SD, et al: Acute renal failure and myopathy after treatment with aminocaproic acid. Br Med J 281:115116, 1980 41. Cooksey MW, Knapp MS: Aminocaproic acid and proteinuria. Br Med J 1:769, 1968 42. Royston D: High-dose aprotinin therapy: A review of the first five years’ experience. J Cardiothorac Vasc Anesth 6:76-100, 1992 43. Eaton MP, Deeb G: Aprotinin versus epsilon-aminocaproic acid for aortic surgery using deep hypothermic circulatory arrest. J Cardthorac Vasc Anesth 12:548-552, 1998

RENAL OUTCOME: APROTININ VERSUS EACA

44. Gaudino M, Luciani N, Giungi S, et al: Different profiles of patients who require dialysis after cardiac surgery. Ann Thorac Surg 79:825-830, 2005 45. Stafford-Smith, Phillips-Bute B, Reddan DN, et al: The association of E-aminocaproic acid with postoperative decrease in creatinine clearance in 1502 coronary bypass patients. Anesth Analg 91:10851090, 2000 46. Lemmer JH, Stanford W, Bonney SL, et al: Aprotinin for

15

coronary artery bypass grafting: Effect on postoperative renal function. Ann Thorac Surg 59:132-136, 1995 47. Bidstrup BP, Harrison J, Royston D, et al: Aprotinin therapy in cardiac operations: A report on use in 41 cardiac centers in the United Kingdom. Ann Thorac Surg 55:971-976, 1993 48. 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