| ||||||||||||||
|
|
|||||||||||||
Department of Anesthesiology, Rush Medical College, Rush-Presbyterian-St. Lukes Medical Center, Chicago, Illinois
Address correspondence and reprint requests to Christopher J. OConnor, MD, Department of Anesthesiology, Rush Medical College, Rush-Presbyterian-St. Lukes Medical Center, Chicago, Illinois 60612. Address e-mail to coconnor{at}rpslmc.edu
| Abstract |
|---|
|
|
|---|
Implications: Because aprotinin is metabolized and eliminated in the kidney, its clearance may be reduced in patients with renal insufficiency. Our data suggest that aprotinin clearance is reduced, and aprotinin half-lives are prolonged in patients with renal insufficiency undergoing CPB. Dosing modification may therefore be indicated when aprotinin is administered to these patients for cardiac surgery.
| Introduction |
|---|
|
|
|---|
Unfortunately, the pharmacokinetics of aprotinin in patients undergoing cardiopulmonary bypass (CPB) are poorly defined, and data from individuals with renal insufficiency are lacking. A previous report determined that plasma aprotinin levels were elevated for prolonged periods and aprotinin clearance reduced when aprotinin was administered in standard doses to a single patient with end-stage renal disease (ESRD) (3). This finding suggested that altered aprotinin pharmacokinetics might produce unpredictable and excessively high plasma levels in patients with renal dysfunction using standard dosing schedules.
To clarify the influence of renal dysfunction on aprotinin elimination, we studied the pharmacokinetics of aprotinin in cardiac surgical patients with both normal and abnormal renal function undergoing CPB.
| Methods |
|---|
|
|
|---|
|
|
|
|
-stat pH management were used for all patients. Ultrafiltration was not employed. Porcine heparin was used for anticoagulation, and protamine was used for heparin reversal at the completion of CPB.
Blood samples for aprotinin concentrations were obtained at the following intervals: 30 min after completion of the loading dose (LD + 30); 30 and 60 min after the initiation of CPB (CPB + 30 and CPB + 60, respectively); at the completion of CPB (end-CPB); and at 8 (LD + 8), 24 (LD + 24), and 32 h (LD + 32) after completion of the loading dose. Samples were placed in tubes containing sodium ethylenediamine tetraacetate, immediately placed on ice, centrifuged, and the resultant plasma was stored at -20°C until assayed. Plasma samples were analyzed using a commercial chromogenic assay (Unicorn Diagnostics Ltd., London, UK). Plasma samples were first treated with acetone to remove the effect of serine protease inhibitors and were then diluted with buffers containing
2-macroglobulin and factor XIIa inhibitors. Purified kallikrein was then added to complex with aprotinin. Residual kallikrein activity was measured photometrically by the liberation of p-nitroaniline, whose concentration is inversely proportional to the concentration of aprotinin (5). Blood samples for thromboelastography (TEG) were drawn after the induction of anesthesia and at 24 h after completion of the aprotinin loading dose. TEG results were assessed for the development of a hypercoagulable state, as measured by the maximal amplitude, the reaction time, and the clot lysis index at 30 min.
The following preoperative demographic and laboratory data were recorded: patient age, weight, gender, coexisting medical conditions, the need for dialysis, and baseline levels of hemoglobin, serum creatinine, blood urea nitrogen (BUN), albumin, serum glutamic oxaloacetic transaminase, glutamic pyruvic transaminase, bilirubin, and CrCl. Intraoperative variables that were recorded included the cardiac index at the end of CPB, urine volume, the duration of aortic cross-clamping and CPB, the lowest bladder temperature and the duration of the hypothermia, the lowest pH during CPB, total fluids and blood products administered, and the total dose of aprotinin used. Variables recorded during the postoperative period included the BUN and creatinine on the first three mornings after surgery, 24-h blood product requirements, and the cardiac index 24 h after the completion of the loading dose. Patients with dialysis-dependent renal failure were dialyzed at the discretion of the nephrology service, and the timing and type of the first dialysis (peritoneal or hemodialysis) were recorded.
Pharmacokinetic variables were calculated using noncompartmental methods. Assuming first-order elimination, areas under the aprotinin concentration-time curve (AUC) were determined using the linear trapezoidal rule with extrapolation to infinity (6). The terminal elimination rate constant,
, was calculated from the 8, 24, and 32 h samples, and the terminal elimination half-life (T1/2) was determined by dividing 0.693 by
. Aprotinin clearance (ApCl) for each patient was calculated by dividing the total IV aprotinin dose by the corresponding AUC.
The relationship between ApCl and CrCl, and between T1/2 and CrCl, were evaluated by linear regression analysis. In addition, CrCl was compared with ApCl and T1/2 according to subgroups of CrCl, as follows: Group 1, CrCl 0 mL/min (patients with ESRD requiring dialysis); Group 2, CrCl 125 mL/min; Group 3, CrCl 2650 mL/min; and Group 4, CrCl > 50 mL/min. Values for ApCl, T1/2, perioperative variables, and TEG measurements are reported as mean ± SD and are compared among and within groups using analysis of variance. Post hoc testing was performed using independent sample (among groups) and paired sample (within groups) Students t-tests corrected for multiplicity at
= 0.05 when analysis of variance identified differences with P < 0.05.
| Results |
|---|
|
|
|---|
There was a significant direct relationship between CrCl and ApCl (r = 0.75, P < 0.01) (Figure 1) and a significant, albeit weaker, inverse relationship between CrCl and T1/2 (r = 0.42, P < 0.05) (Figure 2). ApCl was significantly greater (P < 0.002) in patients with a CrCl > 50 mL/min compared with the remaining groups, whereas T1/2 was significantly prolonged in patients with ESRD compared with patients with a CrCl > 50 mL/min (19.9 vs 7.8 h, P < 0.05) (Figure 3). Plasma aprotinin concentrations for each sampling interval are shown for each CrCl group in Figure 4.
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Aprotinin is a small, basic polypeptide completely filtered in the glomerulus and metabolized by lysosomal enzymes in the proximal convoluted tubule (7,8). Filtered aprotinin is completely metabolized in the proximal tubule, because no biologically active metabolites are found in urine (2,9). Additionally, only 4% of radiolabeled aprotinin is found in the liver, which suggests that hepatic metabolism is probably insignificant. Elimination is therefore almost entirely renal, and the presence of renal dysfunction should predictably diminish ApCl and prolong elimination T1/2. Unfortunately, there is little information regarding aprotinin pharmacokinetics in cardiac surgical patients with normal renal function undergoing CPB and none for cardiac surgical patients with renal insufficiency or ESRD. Although the safe use of aprotinin has been reported in a small series of cardiac surgical patients with ESRD (10), neither plasma levels nor pharmacokinetic data were provided. We previously demonstrated reduced aprotinin clearance in a patient with ESRD undergoing CPB (3), a finding that prompted our current investigation of aprotinin elimination pharmacokinetics in patients with normal and abnormal renal function undergoing CPB.
Levy et al. (11) assessed pharmacokinetic variables after a single IV aprotinin dose of 1 or 0.5 million KIU administered to 28 patients with normal renal function 48 hours before cardiac surgery. They reported a mean elimination clearance of 35.5 mL/min (11). Schall et al. (12,13) reviewed the pharmacokinetics of aprotinin in 47 patients undergoing gynecologic surgery. After loading doses of 1 or 2 million KIU, they reported mean ApCl values between 3950 mL/min and elimination T1/2 between 5.3 and 8.3 hours. In another evaluation of 131I-labeled aprotinin administered to healthy volunteers, Kaller et al. (14) noted an elimination T1/2 of seven hours. These figures are consistent with data obtained in our study from patients with preserved renal function (CrCl > 50 mL/min), where mean ApCl and elimination T1/2 were 53 mL/min and 7.8 hours, respectively.
In contrast to healthy subjects, Bianchi et al. (9) employed very small doses of 99mTc-labeled aprotinin to image the kidney in 22 patients with varying degrees of renal insufficiency. Using a two-compartment model of aprotinin elimination, they reported ApCl values and a linear regression analysis of ApCl and glomerular filtration rate, that were nearly identical to results we obtained in this study. Rustom et al. (15) investigated the elimination of radiolabeled aprotinin in 9 patients with renal insufficiency (mean CrCl 40 mL/min) and found a mean ApCl of 38 mL/min, which is also similar to the value of 34 mL/min we observed in Group 3 patients (mean CrCl 44 mL/min). Finally, data from a brief report by Müller et al. (16), on the pharmacokinetics of aprotinin in two patients with renal insufficiency undergoing gynecologic surgery, revealed T1/2 of 13.3 and 14.9 hours, respectively. Thus, our findings are in agreement with elimination pharmacokinetic values obtained in noncardiac surgical patients with renal insufficiency receiving a single dose of aprotinin.
Our investigation reveals a linear relationship between CrCl and ApCl (Figure 1). However, in an effort to establish possible values for CrCl that would be clinically useful as a guide to possible dosing alterations, we further analyzed the data according to subgroups of CrCl. This subgroup analysis suggests that for CrCl > 50 mL/min, alterations in dosing are probably unnecessary. Although our study cannot define a lower threshold of CrCl for dose modification, a reduction in dosing appears to be appropriate for patients with severe renal impairment in whom clearance is reduced significantly and T1/2 is markedly prolonged.
Unfortunately, neither the optimum dose of aprotinin nor the appropriate therapeutic concentrations required for effective hemostasis have been definitively established. Although theoretical calculations from experimental in vitro models suggest that levels of approximately 200 and 50 KIU/mL are necessary to inhibit plasma kallikrein and plasmin, respectively (17,18,2), the effective concentrations required in vivo may be substantially different depending on the complex interactions of the hemostatic system (17). In fact, significant blood-sparing effects and improvements in platelet function and indices of fibrinolysis have been demonstrated using small-dose aprotinin (11,1926). We chose the standard loading dose of 2 million KIU to rapidly achieve satisfactory plasma concentrations of aprotinin. The reduced infusion rate of 0.25 million KIU/h (the so-called "half-Hammersmith" infusion rate), along with exclusion of the pump prime dose, seemed appropriate dosing modifications for our study because of the potential for impaired elimination of aprotinin in patients with renal dysfunction. The levels we obtained (Figure 4) were consistent with values from patients with normal renal function reported by Von Oeveren et al. (27) and Dietrich et al. (28) using full- and half-Hammersmith dosing schedules, respectively, although the extreme variability of reported aprotinin plasma levels using different dosing regimens and assay techniques makes valid comparisons with other studies problematic (11,28,29,30,31).
Although the findings from previous investigations of healthy volunteers and patients with renal insufficiency agree with our own data, they were obtained after the administration of only a single dose of aprotinin to subjects undergoing noncardiac surgery. They are not strictly comparable to cardiac surgical patients because of the absence of possible confounding effects of pharmacokinetic alterations induced by CPB. However, the similarity of our elimination kinetics to those obtained in these studies suggests a negligible influence of CPB on aprotinin elimination, and a consistent reduction in clearance caused by changes in glomerular filtration rate.
The impact of CPB on drug elimination is complex, and the dynamic interplay of hemodilution, hypothermia, alterations in protein binding, drug sequestration by the lungs, changes in renal and hepatic blood flow, and altered volumes of distribution caused by hemorrhage and fluid administration may unpredictably alter drug elimination during cardiac surgery (32,33). To control the possible influence of these variables on ApCl, baseline measures of weight, the duration of CPB, the extent and duration of hypothermia, acid-base changes during CPB, perioperative blood and fluid administration, and perioperative cardiac index (as a surrogate for renal blood flow), were compared between groups to exclude factors other than CrCl that may have influenced aprotinin elimination (Table 1). This analysis revealed albumin concentrations that were predictably lower in patients with ESRD, although the impact of protein binding on plasma aprotinin levels should be minimal because little aprotinin is protein bound (7). As anticipated, perioperative BUN and creatinine values were significantly higher in patients with ESRD at each day but were unchanged within groups from preoperative values up to three days postoperatively. All other variables were similar between groups except for CPB duration, which was significantly longer in Group 2 versus Group 3 patients, and intraoperative packed red blood cells transfusion, which was greater in Group 2 versus Group 4 patients. While the latter difference might be expected to diminish aprotinin levels (via dilutional changes) and thus increase the calculated ApCl in Group 2, the opposite effect was observed. The impact of a longer duration of CPB in this same group is, however, unclear.
The pharmacokinetic analysis of drug elimination during and after CPB is restricted by the absence of true steady-state conditions, which are considered important for accurate measurements of drug elimination and the creation of pharmacokinetic models (33). However, we measured aprotinin concentrations and elimination parameters for 26 to 30 hours after discontinuation of the aprotinin infusion and completion of CPB, thereby encompassing a period of roughly three to five half-lives of aprotinin. Although this approach permits reliable measurement of terminal elimination T1/2, it precludes determination of the initial distribution T1/2, which would instead require frequent sampling of plasma concentrations immediately after administration of the loading dose (6). It also limits any conclusions regarding the appropriate application of a two- or three-compartment model to the pharmacokinetics of aprotinin. Clearance, however, can be accurately calculated without using compartmental models. Instead, noncompartmental models can be used to determine clearance by dividing the total drug dose by the AUC from zero to infinity (6).
Because of findings obtained from a limited number of clinical investigations, controversy has developed regarding the potential nephrotoxic effects of aprotinin (34,35). However, data from several large clinical studies failed to demonstrate an adverse effect of aprotinin on renal function, even in patients with mild elevations in baseline serum creatinine (23,3638). In addition, the reported changes in serum creatinine from the former studies were conservatively estimated as >0.5 mg/dL increases in postoperative creatinine levels, an incremental difference unlikely to be of clinical significance. Based on prior clinical evidence that failed to demonstrate nephrotoxic effects of aprotinin, and based on the lower total aprotinin dose we administered, we considered our study protocol safe for patients with renal insufficiency. Our study contains too few patients to satisfactorily assess the impact of aprotinin therapy on postoperative renal function in patients with either normal or abnormal renal function. Moreover, this was a pharmacokinetic investigation that was not designed to assess the potential nephrotoxicity of aprotinin.
An additional concern in cardiac surgical patients with reduced aprotinin elimination was the potential development of a prothrombotic state caused by persistent biologically effective concentrations of aprotinin at 24 hours. Nonetheless, despite a prolonged T1/2 in Groups 1 and 2, TEG analysis revealed maximal amplitude, reaction time, and the clot lysis index at 30 minute values at 24 hours that were unchanged compared with baseline values. Although TEG may be an unproven measure of clinical hypercoagulability, it was reassuring to demonstrate an unchanged in vivo whole blood coagulation profile in patients with reduced ApCl. It is possible, however, that full-dose aprotinin therapy may produce abnormal TEG results if higher concentrations of aprotinin are achieved in patients with renal dysfunction. In addition, although it is unknown whether elevated plasma aprotinin concentrations associated with unmodified full-dose therapy in patients with significant renal dysfunction are associated with adverse sequela, less drug (and thus cost) is probably required to achieve plasma levels comparable to those obtained with standard dosing in patients with normal renal function.
The fundamental conclusions of our investigation are based on the significant relationship between CrCl and ApCl demonstrated by regression analysis (Figure 1), and are not dependent on specific subgroupings of CrCl. However, the small sample size precludes specific conclusions regarding dosing modification based on CrCl. Further study of a larger series of patients with renal dysfunction using predetermined target plasma aprotinin levels will be necessary to establish definitive dosing guidelines based on renal function.
In summary, our data demonstrate that ApCl is reduced and aprotinin elimination T1/2 is prolonged in cardiac surgical patients with impaired renal function undergoing CPB. Based on our findings, modification of standard cardiac surgical dosing may be appropriate in patients with significant renal dysfunction. A reduction in the infusion rate to 0.25 million KIU/h and elimination of the pump prime dose, as used in this study, appear to be acceptable modifications of standard dosing regimens.
| Acknowledgments |
|---|
The authors gratefully acknowledge the assistance of Robert McCarthy, PharmD, with the analysis of pharmacokinetic parameters.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. C. Body and C. D. Mazer Pro: Aprotinin Has a Good Efficacy and Safety Profile Relative to Other Alternatives for Prevention of Bleeding in Cardiac Surgery Anesth. Analg., December 1, 2006; 103(6): 1354 - 1359. [Full Text] [PDF] |
||||
![]() |
E. H. Kincaid, D. A. Ashburn, J. R. Hoyle, M. G. Reichert, J. W. Hammon, and N. D. Kon Does the Combination of Aprotinin and Angiotensin-Converting Enzyme Inhibitor Cause Renal Failure After Cardiac Surgery? Ann. Thorac. Surg., October 1, 2005; 80(4): 1388 - 1393. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. A. Nuttall, D. N. Fass, L. J. Oyen, W. C. Oliver Jr., and M. H. Ereth A Study of a Weight-Adjusted Aprotinin Dosing Schedule During Cardiac Surgery Anesth. Analg., February 1, 2002; 94(2): 283 - 289. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Royston, R. Cardigan, C. Gippner-Steppert, and M. Jochum Is Perioperative Plasma Aprotinin Concentration More Predictable and Constant After a Weight-Related Dose Regimen? Anesth. Analg., April 1, 2001; 92(4): 830 - 836. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Ekback, K. Axelsson, L. Ryttberg, B. Edlund, J. Kjellberg, J. Weckstrom, O. Carlsson, and U. Schott Tranexamic Acid Reduces Blood Loss in Total Hip Replacement Surgery Anesth. Analg., October 1, 2000; 91(5): 1124 - 1130. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Beath, G. A. Nuttall, D. N. Fass, W. C. Oliver Jr., M. H. Ereth, and L. J. Oyen Plasma Aprotinin Concentrations During Cardiac Surgery: Full- Versus Half-Dose Regimens Anesth. Analg., August 1, 2000; 91(2): 257 - 264. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. O'Connor, D. Roozeboom, R. Brown, and K. J. Tuman Pulmonary Thromboembolism During Liver Transplantation: Possible Association with Antifibrinolytic Drugs and Novel Treatment Options Anesth. Analg., August 1, 2000; 91(2): 296 - 299. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|