| ||||||||||||||
|
|
|||||||||||||
-Aminocaproic Acid with Postoperative Decrease in Creatinine Clearance in 1502 Coronary Bypass Patients

Departments of Anesthesiology
* and Medicine (Division of Nephrology)
, Duke University Medical Center, Durham, North Carolina
Address correspondence and reprint requests to Mark Stafford Smith, FRCPC, Department of Anesthesia, Box 3094, Duke University Medical Center, Durham, NC 27710. Address e-mail to staff002{at}mc.duke.edu
| Abstract |
|---|
|
|
|---|
-aminocaproic acid (EACA) have therefore raised concerns for renal safety. Since EACA renders these markers unreliable, we used perioperative change in creatinine clearance (DCrCl) to test the hypothesis that EACA is associated with greater reductions in creatinine clearance after heart surgery, particularly for patients with renal disease. We evaluated data from all elective primary coronary bypass patients during EACA introduction at our institution (July 1, 1991December 31, 1992; 10 g iv bolus pre-cardiopulmonary bypass, then 1 g/h for 5 h). DCrCl was calculated using preoperative (CrPre) and postoperative peak serum creatinine values, using the Cockroft-Gault equation. Patients with CrPre
133 µmol/L were also separately analyzed. Evaluated patients (n = 1502, ±EACA; 581/905, 16 exclusions) included 233 with CrPre
133 µmol/L (±EACA; 98/135). Multivariate analyses confirmed several known risk factors, but no association between DCrCl and EACA in all patients (P = 0.66), and the subgroup with CrPre
133 µmol/L (P = 0.42).
Implications: In a large population of primary Coronary Artery Bypass Graft including a subset with preoperative renal dysfunction, there were no postoperative reductions in creatinine clearance attributable to
-aminocaproic (EACA) administration. This retrospective study suggests that moderate
-aminocaproic acid dosing during cardiac surgery is safe for the kidney; however, this inference is based on a single marker of renal dysfunction and requires prospective confirmation using a variety of tests of renal function.
| Introduction |
|---|
|
|
|---|
-aminocaproic acid (EACA) therapy have lead to widespread use of this agent for cardiac surgery patients (19). Therefore, reports linking EACA with postoperative proteinuria and hyperkalemia in coronary artery bypass graft (CABG) surgery patients have raised serious concerns (1012). Renal dysfunction is an important complication of cardiac surgery, associated with major increases in postoperative morbidity and mortality (1315). Of the 800,000 patients who undergo CABG surgery worldwide each year (16), up to 8% will suffer a significant renal insult, and
1% will require dialysis (1315). Patients with preexisting renal disease are at a particularly increased risk for renal dysfunction after cardiac surgery (13,14,17). The nephrotoxic potential of EACA in cardiac surgery patients has caused concern among clinicians, pitting the benefits of improved hemostasis against the fear of increased renal insult; this dilemma is particularly evident in patients with preexisting renal disease. The clinical use of EACA (a lysine analog) relates to its inhibition of fibrinolysis; however, this agent also has several reversible effects that resemble those from an IV lysine infusion, including displacement of intracellular potassium and inhibition of renal tubular reabsorption of small filtered proteins (1820). These properties of EACA complicate its study during cardiac surgery by making hyperkalemia and tubular proteinuria unreliable markers of subtle renal injury (1821). Reports of ß2-microglobulinuria (10,11) and hyperkalemia (12) in cardiac surgery patients receiving EACA therapy may represent important renal injury; however, "lysine-like" effects of this agent may also explain these abnormalities. A more reliable measure of renal insult in this setting is estimated creatinine clearance (CrCl), a validated measure of renal function after cardiac surgery (22). Therefore, we used data from consecutive patients during the period when EACA therapy was introduced at our institution to test the hypotheses that EACA is associated with larger reductions in CrCl after CABG surgery: 1) in all patients, and 2) in the subset with preoperative renal disease.
| Methods |
|---|
|
|
|---|
133 µmol/L). The subgroup with increased CrPre was selected for secondary analysis due to their increased risk for postoperative reduction in CrCl (13,15,17). No antifibrinolytic drugs other than EACA were used at the institution during the study period. Anesthesia was managed per the attending anesthesiologists preference; use of drugs with renal effects (e.g., IV dopamine, furosemide) was not regulated; however, 50 g of 20% mannitol solution (Abbott Laboratories, North Chicago, IL) was routinely added to the cardiopulmonary bypass (CPB) circuit priming solution. Hypothermic CPB was performed between 25°C and 32°C nasopharyngeal temperature using a Cobe CML membrane oxygenator (Cobe Laboratories, Lakewood, CO), Sarns 7000 MDX pump (Sarns Inc., Ann Arbor, MI), and Pall SP 3840 (Pall Biomedical Products Co., Glencove, NY) 40 µm arterial line filters.
|
-Aminocaproic Acid Data
Perioperative Renal Data
Blood samples were obtained preoperatively and daily postoperatively until hospital discharge per institutional routine to assess serum creatinine values. CrPre was obtained within 2 weeks of surgery, and defined as the value recorded closest to surgery, but not on the day of the procedure. Peak serum creatinine (CrPost) was the highest of the daily in-hospital postoperative values. The Cockroft-Gault equation (23) was selected as the most rigorously evaluated and consistently favored algorithm (2225) to calculate pre- and postoperative CrCl:
Men: CrCl = ((140 - age) x weight)/(serum Cr x 72)
Women: CrCl = [((140 - age) x weight)/(serum Cr x 72)] x 0.85
Units: weight - kg, age - yr, serum Cr - (mg/ dL) (88.4 µmol/L = 1 mg/dL)
Perioperative change in CrCl (DCrCl) was the difference between these values (DCrCl = CrPreCl - CrPostCl). Demographic variables gathered for each patient were chosen with reference to previously reported renal risk factors (14,15), including age, gender, weight, height, duration of CPB, history of Type I or Type II diabetes requiring pharmacotherapy, first intraoperative and post-CPB cardiac indexes, and use of an intraaortic balloon pump (IABP) after separation from CPB. Weight and height values used for analyses were those recorded on the anesthesia record.
Statistical Analysis
A similar statistical approach was taken in the analysis of both the whole patient group and the subgroup with preoperative renal insufficiency (i.e., CrPre
133 µmol/dL). Specifically, the association of demographic and perioperative variables with EACA use was initially compared using Students t-test for continuous variables and
2 test for categorical variables. The association of EACA use with DCrCl was then examined using a linear regression multivariate analysis, adjusting for CrPre and allowing for potential effects of recorded variables, and two-way interactions between known renal risk factors. Due to the potential for nonnormal distribution of CrCl values, and to assess generalizability and robustness of results, these analyses were then repeated on ranked data. Finally, to evaluate the possible occurrence of infrequent but important major renal injury attributable to EACA, multivariate logistic regression analysis of the entire patient group was performed to assess the association of EACA administration with two recently published definitions of moderate renal failure (13,14). Analyses were performed using SAS software version 6.12 (SAS Institute, Inc., Cary, NC); significance was judged at
= 0.05.
| Results |
|---|
|
|
|---|
|
|
|
88.4 µmol/L (
1.0 mg/dL) (13). Using this definition for renal insult, EACA administration was not independently associated with moderate acute renal failure (P = 0.47). A second analysis was performed using a definition of moderate acute renal failure from a publication by Mora-Mangano et al. (14). These authors defined moderate renal dysfunction in patients with CrPre <177 µmol/L (<2.0 mg/dL), as a CrPost increase of at least 62 µmol/L (0.7 mg/dL), with a peak serum creatinine exceeding 177 µmol/L. EACA administration was also not independently associated with this definition of moderate acute renal injury in the patients from our study with a CrPre <177 µmol/L (<2.0 mg/dL) (P = 0.34). Factors independently associated with the occurrence of moderate acute renal failure were similar to those identified in the linear multivariate analyses in the current study and in previously published analyses (13,14). | Discussion |
|---|
|
|
|---|
These data conflict with previous reports that have used markers of subtle renal insult to challenge the renal safety of EACA in the setting of cardiac surgery. Some factors strengthen the results of this study, compared with previous reports, including the use of a validated test to assess renal filtration, power to identify even small renal effects, analysis of both low- and high-risk patients, and confirmation of renal risk factors similar to those reported in other populations (14,15). The level below which perioperative renal insult becomes clinically insignificant has not been established. Although we believe the findings of this study offer some reassurance that moderate EACA administration during elective primary CABG surgery is not associated with major renal insult, this conclusion is based on retrospective evidence involving a single marker of renal dysfunction and requires prospective confirmation using a variety of tests of renal function.
A large body of nonsurgical experience has facilitated characterization of many effects and complications associated with EACA therapy. Large-dose EACA regimens for extended periods (i.e., 1830 g daily for 3040 days) (26,27) are used to treat refractory bleeding in the setting of abnormal hemostasis. In animal models of nephritis, EACA has shown protective effects; reducing albuminuria, attenuating renal histopathologic changes, and enhancing CrCl (28). However, several clinical reports have demonstrated an association between extended EACA therapy and renal dysfunction. Acute renal failure has resulted from EACA-related rhabdomyolysis (29), upper urinary tract obstruction due to clot (30), and intravascular thrombosis (31). EACA myopathy (a dose/duration-related phenomenon) (26,27) and clot-related obstructive uropathy are problems unlikely to complicate cardiac surgery. However, intravascular thrombosis is a potential complication of EACA therapy that raises serious concerns for cardiac surgeons and anesthesiologists despite the absence of definitive information (32).
The conclusions of our study differ from those of other reports regarding the renal risk of EACA for cardiac surgical patients. In a review, Royston (10) summarized data citing EACA-related ß2-microglobulinuria (i.e., tubular proteinuria) as evidence of nephrotoxic effects. Whereas in some settings tubular proteinuria may be the only evidence of subtle renal injury, it is an unreliable, marker with EACA use (21). Lysine and other lysine-analogs, including EACA, cause a reversible inhibition of small protein reuptake by the renal tubular brush border that is unrelated to renal injury (19,20). Garwood et al. (12) reported hyperkalemia after administration of EACA during cardiac surgery as additional evidence of EACA-related renal effects. Although this represents a larger dose than used in our study, Garwood clarifies that because EACA displaces intracellular potassium causing hyperkalemia (18), interpretation of these data as evidence of a nephrotoxic effect is problematic. In contrast, we use estimated CrCl to assess changes in renal filtration, a test that has been validated in postcardiac surgery patients (22). In addition, the large number of subjects available to test both our hypotheses gives some confidence that a major effect has not been missed. Interpretation of our study findings is not complicated by confounding effects of EACA on markers of renal injury; we present data derived using a valid measure of kidney function in a large population of postcardiac surgery patients that does not confirm previous reports of renal insult from EACA administration.
The use of retrospective data for our study presents some analytical limitations. Since the study was designed to include the brief transition period when administration of EACA changed from being uncommon to routine for CABG surgery patients at our institution, we anticipated that demographic variables would be similar between groups; this is generally confirmed by our analysis (see Tables 1 and 2). In particular, the highest risk patients with preoperative renal impairment are balanced between the groups, implying that this factor was not influencing the decision to use EACA. Additional demographic variables were similar between EACA and non-EACA groups, with the exception of duration of CPB, IABP use, and weight (Tables 1 and 2). Multivariate analysis accounts for differences between groups of these variables. Body weight is slightly greater in the overall EACA group, although this effect is not present in the subgroup with preoperative renal dysfunction. CPB duration and IABP use are greater for patients receiving EACA in both analyses. A 3-mo comparison of CPB duration between EACA and non-EACA patients revealed that this effect was marked at the beginning (157 vs. 112 min, P = 0.0006) and absent by the end of the study period (121 vs. 108 min, P = 0.16), suggesting that "sicker" patients may have been the first to receive EACA. Presumably the greater need for postbypass IABP in EACA patients also reflects this effect. Since "sicker" patients generally have higher renal risk, any unmeasured bias in our study group would be expected to increase the likelihood of postoperative reductions in CrCl associated with EACA. Despite the potential for increased renal dysfunction with patients receiving EACA, our multivariate analysis identified no difference in postoperative reductions in CrCl in the group of patients receiving EACA. Our data, however, confirm independent associations of several known renal risk factors (1315) with postoperative reductions in CrCl. Another limitation of this study is the limited information regarding the overall integrity of the kidney offered by a single study such as CrCl. CrCl is an adequate test of renal filtration function, but does not address the numerous other homeostatic roles of the kidney, including osmolality, electrolyte and acid/base regulation, excretion of metabolic endproducts and toxins, and the production and release of several enzymes and hormones. Although our study suggests that moderate EACA usage is safe for the kidney in the setting of primary CABG surgery, a prospective study using a variety of tests of renal function to address this issue and evaluating the use of EACA for higher risk cardiac surgeries (e.g., repeat sternotomy, open chamber procedures) is needed.
In summary, we did not demonstrate postoperative reductions in CrCl after elective primary CABG surgery attributable to intraoperative administration of moderate EACA doses; this finding was consistent in analysis of all patients, and the high-risk subset with preoperative renal dysfunction. The level below which changes in CrCl become clinically insignificant has not been established, but the large number of subjects available to test both our hypotheses provides some confidence that a major effect has not been missed. Although this retrospective study offers some reassurance that moderate EACA dosing is safe for the kidney in the setting of primary CABG surgery, this inference is based on a single marker of renal dysfunction and requires prospective confirmation using a variety of tests of renal function.
| Acknowledgments |
|---|
The authors thank L. H. Muhlbaier, PhD, William D. White, MPH, and Ms. Latanya Rhames.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
1- and ß2-microglobulinuria poor markers of post cardiac surgery renal dysfunction. Anesthesiology 1999; 90: 9289.[ISI][Medline]
This article has been cited by other articles:
![]() |
Y. Ozier and S. Schlumberger Pharmacological approaches to reducing blood loss and transfusions in the surgical patient: [Approches pharmacologiques de reduction des pertes sanguines et des transfusions en chirurgie] Can J Anesth, June 1, 2006; 53(6_suppl): S21 - S29. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Stafford-Smith Evidence-Based Renal Protection in Cardiac Surgery Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2005; 9(1): 65 - 76. [Abstract] [PDF] |
||||
![]() |
B. J. McCreath, M. Swaminathan, J. V. Booth, B. Phillips-Bute, S. T.H. Chew, D. D. Glower, and M. Stafford-Smith Mitral valve surgery and acute renal injury: port access versus median sternotomy Ann. Thorac. Surg., March 1, 2003; 75(3): 812 - 819. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Swaminathan, B. J. McCreath, B. G. Phillips-Bute, M. F. Newman, J. P. Mathew, P. K. Smith, J. A. Blumenthal, and M. Stafford-Smith Serum Creatinine Patterns in Coronary Bypass Surgery Patients With and Without Postoperative Cognitive Dysfunction Anesth. Analg., July 1, 2002; 95(1): 1 - 8. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Butterworth, R. L. James, Y. A. Lin, J. Bennett, and R. C. Prielipp Gender Does Not Influence Epsilon-Aminocaproic Acid Concentrations in Adults Undergoing Cardiopulmonary Bypass Anesth. Analg., June 1, 2001; 92(6): 1384 - 1390. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|