| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2003;126:1087-1094
© 2003 The American Association for Thoracic Surgery
Cardiopulmonary support and physiology |
a Oxford Heart Centre, John Radcliffe Hospital, Oxford, United Kingdom
Received for publication May 21, 2002; revisions received August 15, 2002; revisions received November 12, 2002; accepted for publication February 3, 2003.
* Address for reprints: David P. Taggart, Oxford Heart Centre, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom
david.taggart{at}orh.nhs.uk
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionConclusionReferences |
|---|
METHODS: A single-center, randomized, double blind, placebo-controlled trial of aprotinin administration in total arterial grafting was performed. The primary outcome variable was postoperative blood loss, and the secondary outcome variable was the number of units of donor blood or coagulant products transfused. The incidence of myocardial injury was determined from serial measurements of cardiac troponin T and creatine kinase-MB and renal injury from serum creatinine.
RESULTS: The placebo group (n = 34) and aprotinin group (n = 36) were similar with respect to all preoperative and intraoperative comparisons. One patient in each group underwent reexploration for bleeding. Open-label aprotinin was administered to 9 patients in the placebo group (26%) and to 2 patients in the aprotinin group (6%). There was a highly significant reduction in the median (interquartile range) blood loss in the aprotinin group compared with the placebo group (785 mL [590-1025 mL] vs 1525 mL [1175-1920 mL], respectively). Similarly, the aprotinin group demonstrated a marked reduction in the need for blood transfusion (77% vs 39%; P = .0001), the mean number of transfused blood units (2.6 vs 0.8, P < .001), and the number of patients requiring coagulant products (24% vs 3%; P < .001). There was no difference in myocardial injury in the 2 groups. Four patients in the aprotinin group had persistently elevated creatinine levels in the postoperative period (3 of whom had elevated preoperative creatinine levels and perioperative complications).
CONCLUSIONS: Aprotinin significantly reduces blood loss and the need for blood component transfusion in patients undergoing total arterial grafting without increasing the risk of myocardial injury. Aprotinin should be considered routinely in patients undergoing total arterial grafting but cautiously in patients with an elevated preoperative creatinine level.
It is the principal investigator's standard practice to perform CABG using BITA and radial artery grafts. However, in my experience (D.P.T.),4 the use of BITA grafts increases mean postoperative blood loss by approximately 400 mL in comparison with the use of a SITA graft (mean [SD] BITA vs SITA: 1324 mL [362 mL] vs 945 mL [330 mL]; P < .01). This is consistent with an increased surface area of chest wall dissection when harvesting BITA vessels.
Aprotinin is a naturally occurring serine protease inhibitor that has potent hemostatic properties when used in cardiac surgery. Recent advances in the understanding of the basic mechanism of aprotinin have shown that aprotinin may in fact also have antithrombotic properties by its selective blockade of protease-activated receptor-1 thrombin receptors.5
The major concern over the routine use of aprotinin is, however, an increased risk of graft occlusion and perioperative myocardial infarction (MI) through prothrombotic mechanisms. The International Multicenter Aprotinin Graft Patency Experience (IMAGE) trial was undertaken to resolve conflicting conclusions of studies that documented an increased risk of vein graft occlusion6-8 and those that did not.9 The IMAGE trial was conducted in 13 centers in the United States and Europe, randomizing 870 patients to aprotinin or placebo and assessing the graft patency using graft angiography.10 Unfortunately, whereas vein graft occlusion rates were similar in the 2 American groups (9%), they differed in the European groups (23% in the aprotinin group vs 12% in the placebo group). Even after adjustment for risk factors, the IMAGE trial suggested that aprotinin led to a slightly increased risk of venous graft occlusion when evaluated on per-patient or per-graft analysis.
We hypothesized that aprotinin would be efficacious in reducing blood loss in patients undergoing TAG without increasing the risk of myocardial injury, particularly because arterial grafts are inherently more resistant to thrombosis than vein grafts. Consequently, we performed a randomized trial of aprotinin in patients undergoing TAG to determine blood loss and need for blood products. The incidence of perioperative MI and myocardial injury was assessed by serial electrocardiograms and measurements of cardiac troponin T (cTnT) and creatine kinase (CK)-MB. Renal injury was assessed by serial measurement of serum creatinine.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
Patient information and consent
Consenting patients, fulfilling the inclusion and violating none of the exclusion criteria, provided written informed consent before participation.
Power calculations
The sample size calculation was based on the primary efficacy variable and postoperative blood loss, and aimed to detect a difference between the aprotinin and placebo groups. From previous studies, blood loss seemed to have a log-normal distribution, so calculations were based on the geometric mean and SD. A placebo geometric mean of 1223.6 mL and geometric SD of 1.5 mL was assumed. Converting this to loge scale, this became 7.110 mL and 0.432 mL, respectively. The power was set at 90%, and a 2-sided significance level was set at 5%, with a clinically relevant difference of 30% reduction in the geometric mean postoperative blood loss in the active group. Thirty-two evaluable patients were therefore required in each treatment group. Allowing for an invalidity rate of 15%, a total sample size of 75 patients was required.
Patients
Eligible patients were those undergoing first-time isolated CABG who required at least 2 bypass grafts in which it was planned that revascularization would be performed only using arterial conduits. Patients could be undergoing elective or urgent, but not emergency (ie, immediate), surgery. Exclusion criteria were significantly impaired left ventricular function (ejection fraction estimated <25%), acute MI within 6 weeks, known or suspected allergy to aprotinin, possible previous exposure to aprotinin, coagulopathy or history of any bleeding diathesis, and use of any investigational drug within the previous 30 days. Patients could be withdrawn at any time for medical reasons, for example, excessive postoperative bleeding necessitating resternotomy. For patients who prematurely discontinued participation in the trial, a final assessment was completed, wherever possible, according to the schedule of the 120-hour visit.
Overall study design and plan
The study was a single-center, double-blind, parallel group, placebo-controlled comparison of patients undergoing total arterial revascularization conducted by a single surgeon.
Randomization
A predetermined randomization scheme was generated by the Biometry Department of Bayer, using a fixed block size. Patients were risk stratified as elective or urgent. Sealed code break cards were available for clinical reasons.
Drug dose
Patients were randomized 1 day before surgery to receive placebo or full-dose ("Hammersmith") aprotinin infusion, commenced after induction of anesthesia. After induction of anesthesia, and before sternotomy, a test dose (5.0 mL of 1.4 mg/mL or 10,000 kallikrein inactivation unit/mL) of aprotinin or placebo solution was administered over several minutes through the central line. If there was no evidence of hypersensitivity after 10 minutes, the remaining 195 mL of the loading dose was administered over 20 to 30 minutes using an infusion pump. After completion of the loading dose, a maintenance infusion of 50 mL/h was continued for 4 hours. A further 200 mL was added to the pump prime of the bypass circuit.
Anesthetic protocol
Patients were premedicated with papaveretum and hyoscine. Anesthesia was induced using a combination of fentanyl and etomidate and maintained with nitrous oxide/air and halothane or isoflurane. Anesthesia was maintained on cardiopulmonary bypass (CPB) with a propofol infusion. Antibiotic prophylaxis was gentamicin and flucloxacillin or cefuroxime.
Cardiopulmonary bypass
The bypass circuit was primed with 1.5 L Hartmann's solution, and a membrane oxygenator was used. Flow rates were maintained to achieve 2.4 L/m2, and the patient's temperature allowed to decrease to 34°C. Continuous flow and alpha-stat control of acid base balance was used. Perfusion pressures were maintained between 40 and 90 mm Hg, with vasoconstrictor or vasodilator agents as necessary.
Anticoagulation on bypass
Prolongation of the activated clotting time (ACT) by aprotinin can lead to an overestimation of the degree of anticoagulation, leading to inadequate anticoagulation. Consequently, a kaolin-activated ACT greater than 600 seconds of heparin was obtained before institution of CPB with a loading dose of 300 U/kg of heparin. The ACT was rechecked periodically on CPB, and further heparin was administered if the ACT decreased less than 600 seconds. Heparin was reversed with 1 mg of protamine per 100 units of heparin, and additional protamine was administered depending on the measured ACT.
Transfusion policy
Doctors and nurses, independent of the trial, determined the routine postoperative management of the trial patients. If any patient experienced excessive postoperative bleeding (blood loss of >200 mL in the first 2 consecutive hours in patients who were otherwise stable and in whom the bleeding was expected to stop without need for rethoracotomy), then, at the discretion of the investigator, double-blind medication was stopped, and an open-label aprotinin infusion started according to routine clinical practice.
In general terms, hemostatic factors were requested when bleeding was approximately 200 mL/h in otherwise stable patients. During the preoperative and postoperative periods, blood was transfused if the hemoglobin decreased to less than or was equal to 9 g/dL. Further units were administered to ensure that hemoglobin was maintained greater than 9 g/dL.
Analysis
The study was analyzed on an intention-to-treat (ITT) basis and included those patients who received open-label aprotinin. All patients receiving trial treatment were included in the safety analysis.
Trial objectives
The primary aim of the trial was to assess the efficacy of aprotinin in reducing postoperative blood loss after TAG and the need for transfusion of blood and blood products. In addition, the frequency of resternotomy, time to extubation, and total duration of hospital stay was recorded. Myocardial injury was assessed by serial electrocardiographic changes and serum CK-MB and cTnT (calculated as an area under the curve [AUC] from serial measurements) and renal function by measurement of serum creatinine.
Statistical analyses
Summary statistics for the demographic data include the arithmetic mean and SD (normally distributed data), the geometric mean and SD (log-normally distributed data), and the median and range (non-normal data) or frequency counts (categoric data, in which the number of categories were small).
Treatment groups were compared at baseline to check for balance in prognostic factors using the Student 2-sample t test (normally distributed variables), Wilcoxon 2-sample test (non-normally distributed variables), and Fisher's exact test (categoric variables).
The primary efficacy parameter, postoperative blood loss, was tested for normality using the Shapiro-Wilk test, and because it was approximately normally distributed after applying log transformation, the treatment groups were compared using the Student 2-sample t test.
Because the number of units of donor blood and other blood products and the frequency of resternotomy were categoric, non-normal data, the Wilcoxon 2-sample test was used to compare treatment groups, and the Cochrane-Mantel-Haenszel test was used when the number of categories was small. The number of patients receiving a blood transfusion at any time during the study was summarized using frequency counts and percentages, and treatment groups were compared using Fisher's exact test.
For cTnT and CK-MB, the AUC was computed for each patient. By providing an integrated value for the release of a chemical or metabolite over a given time frame, the AUC increases the chances of detecting minor differences between 2 groups while minimizing the chances of a false-positive finding by making multiple comparisons of individual time points. Furthermore, the AUC allows comparison of the data from all patients in a trial, including those with minor elevations in any chemical or metabolite, even when they remain within the normal range.
The AUC was calculated over 8 time points from the preoperation baseline (0 hours) to 123 hours after baseline by using the trapezoidal rule with interpolation for missing intermediate time points and then dividing by 123 to give units in nanograms per milliliter. Where a measurement at the final time point was missing, the AUC was calculated over the time points present and then divided by the appropriate time interval to give units in nanograms per milliliter. The AUC data for cTnT and CK-MB were compared between the 2 treatment groups using the Wilcoxon 2-sample test.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
|
Demographic features
The groups were similar with respect to age, gender, height, weight, and urgency of operation (Table 2).
In both groups, approximately one third of patients underwent urgent operation for unstable angina and 15% of all patients were treated for diabetes.
|
|
|
|
|
|
Renal injury
Serial changes in serum creatinine in both groups are presented in Figure 3.
Overall, there was no significant difference between the groups. There were, however, 4 patients in the aprotinin group and 2 patients in the control group whose postoperative creatinine exceeded 2 mg/L on day 5. Of particular interest, 3 of these 4 patients in the aprotinin group had elevated preoperative creatinine levels (1.7, 1.7, 1.5 mg/L, respectively), whereas both the control patients had normal preoperative creatinine values. In addition, the 4 patients in the aprotinin group had other complications that may have contributed to renal impairment, namely, longstanding type I diabetes mellitus, severe prolonged postoperative ileus, perioperative MI, and "prophylactic" intra-aortic balloon pump insertion for severely impaired left ventricle function. In the control group, 1 patient had no obvious explanation for a precipitous increase in postoperative creatinine, whereas the other patient had a hypovolemic cardiac arrest when a clip on the stump of the proximal right ITA eroded into the subclavian artery requiring emergency reexploration.
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
Potential limitations of this trial
Although this was a trial of TAG, 4 patients in the placebo group (12%) and 2 patients in the aprotinin group (6%) received additional vein grafts. This reflected our "learning" phase with TAG and now occurs in less than 5% of our patients. In any event, it did not affect the results of the trial when analyzed on an ITT basis.
Blood loss, blood transfusion, and use of coagulation products
This trial demonstrates that aprotinin results in a dramatic reduction in blood loss and blood product requirements in patients undergoing TAG. Aprotinin almost halves blood loss, halves the number of patients requiring red blood cell transfusion, and significantly reduces the number of units transfused per patient and the need for coagulant products. Even so, our results may still underestimate the benefits of aprotinin; in the ITT analyses, 26% of patients in the placebo group received open-label aprotinin compared with 6% of patients in the aprotinin group.
Aprotinin and risk of myocardial injury
The overall AUC for both troponin and CK-MB release, reflecting a global index of myocardial injury, was lower in the aprotinin group, although this did not reach statistical significance.
There is conflicting evidence as to whether aprotinin increases the risk of perioperative myocardial injury in conventional CABG using an ITA and vein grafts. Cosgrove and colleagues6 reported an increased incidence of perioperative MI in patients undergoing redo CABG from 7% of patients (4/56) in the placebo group to 16% of patients (9/57) receiving high-dose aprotinin. Lemmer and coworkers7 also reported an increased risk of perioperative MI in patients receiving high-dose aprotinin (8.9% vs 5.6%), although this did not reach statistical significance. In contrast, the IMAGE study assessed 831 patients' electrocardiograms and cardiac enzymes and reported that aprotinin did not increase the risk of MI compared with the placebo group.10 Similarly, a multicenter study of redo CABG by Alderman and colleagues11 showed no difference in the incidence of MI in the placebo or aprotinin-treated groups (29%). These conflicting observations may be explained, in part, by the meta-analysis of 72 trials involving more than 8000 patients performed by Levi and colleagues.12 They reported that high-dose, but not low-dose, aprotinin increased the risk of perioperative MI.
It must be emphasized that these studies did not include patients undergoing TAG. Consequently, a secondary aim of our trial was to assess the effect of aprotinin on both MI and global injury. Perioperative MI (suggesting focal injury) was determined by a combination of serial electrocardiograms and cardiac-specific enzymes (cTnT, CK-MB), whereas global myocardial ischemia was determined from an integrated AUC for total enzyme release calculated from serial enzyme measurements at 8 time points over 5 days.
Our trial demonstrates that aprotinin does not increase the risk of perioperative myocardial injury in the setting of TAG. It is notable, however, that only 2 patients (3%) experienced a perioperative MI, according to electrocardiographic and biochemical criteria. Because most perioperative MIs are caused by acute graft failure,8 it is possible that this may represent a reduced risk of acute failure of arterial conduits in comparison with vein grafts. This is certainly consistent with angiographic data from several studies reporting significantly better patency of arterial conduits in comparison with vein grafts early after CABG,13 but in the absence of our own angiographic data this must remain speculative.
This is also consistent with various animal studies that have shown aprotinin to be cardioprotective particularly when used as a cardioplegia additive.
Aprotinin and risk of renal injury
Although several studies have reported that aprotinin does not increase the risk of renal failure after cardiac surgery,14 others have documented an increase in postoperative serum creatinine levels. In a multicenter, randomized study of 212 patients undergoing valve replacement, postoperative increases in creatinine exceeding 0.5 mg/dL occurred in 30%, 14%, and 8% of the high-dose aprotinin, low-dose aprotinin, and placebo groups, respectively, and were statistically significantly different.15 In a randomized study by Cosgrove and associates,6 postoperative increases in serum creatinine of greater than 0.5 mg/dL occurred in 25%, 20%, and 18% in the high-dose aprotinin, low-dose aprotinin, and placebo groups, respectively, although these differences were not statistically significant. Similarly, a multicenter, randomized, controlled trial showed a modest postoperative increase in serum creatinine levels in 19% of the patients (20/108) in the aprotinin group compared with 12% of the patients (13/108) in the placebo group, although this was not statistically significant.14
Overall, as judged by median and interquartile ranges, our trial did not show any adverse effect of aprotinin on renal function as judged by serial serum creatinine levels. This is consistent with the recent trial by Schweizer and colleagues,16 who randomized 57 patients undergoing cardiac surgery with CPB to high-dose aprotinin and placebo and detected no difference in several sophisticated parameters of renal function between the groups. It should be noted, however, that 4 patients in the aprotinin group and 2 patients in the control group had postoperative creatinine levels exceeding 2 mg/L on day 5. In the control group, 1 patient had a hypovolemic cardiac arrest requiring emergency thoracotomy to control bleeding, and the other patient had no obvious explanation for the precipitous increase in creatinine. Of the 4 patients in the aprotinin group, 3 had elevated preoperative creatinine levels, and 4 had other complications that may have contributed to postoperative renal impairment. It is therefore uncertain whether it was the elevated preoperative creatinine level or a perioperative complication that contributed most to the postoperative elevation, but it would seem sensible to use aprotinin cautiously in those with preoperative renal impairment.
|
Conclusion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. L. Ngaage, A. R. Cale, M. E. Cowen, S. Griffin, and L. Guvendik Aprotinin in Primary Cardiac Surgery: Operative Outcome of Propensity Score-Matched Study Ann. Thorac. Surg., October 1, 2008; 86(4): 1195 - 1202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Sundt The demise of aprotinin: Our share of the blame. J. Thorac. Cardiovasc. Surg., April 1, 2008; 135(4): 729 - 731. [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Brown, N. J.O. Birkmeyer, and G. T. O'Connor Meta-Analysis Comparing the Effectiveness and Adverse Outcomes of Antifibrinolytic Agents in Cardiac Surgery Circulation, June 5, 2007; 115(22): 2801 - 2813. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. Edmunds Jr and R. W. Colman Thrombin During Cardiopulmonary Bypass Ann. Thorac. Surg., December 1, 2006; 82(6): 2315 - 2322. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
C. Berroeta, A. Benbara, S. Provenchere, N. Ajzenberg, J. Benessiano, J.-P. Depoix, J.-M. Desmonts, B. Iung, and I. Philip A Comparison of Bilateral with Single Internal Mammary Artery Grafts on Postoperative Mediastinal Drainage and Transfusion Requirement Anesth. Analg., December 1, 2006; 103(6): 1380 - 1385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Poston and O. Toshinaga Reply Ann. Thorac. Surg., November 1, 2006; 82(5): 1951 - 1952. [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Abu-Omar and C. Ratnatunga Cardiopulmonary Bypass and Renal Injury Perfusion, July 1, 2006; 21(4): 209 - 213. [Abstract] [PDF]
|
|
|
|
 |
|
 J. H. Fischer and M. Steinhoff Effects of aprotinin on endothelium-dependent relaxation of large coronary arteries Eur J Cardiothorac Surg, December 1, 2005; 28(6): 801 - 804. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Lindvall, U. Sartipy, and J. van der Linden Aprotinin Reduces Bleeding and Blood Product Use in Patients Treated With Clopidogrel Before Coronary Artery Bypass Grafting Ann. Thorac. Surg., September 1, 2005; 80(3): 922 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. van der Linden, G. Lindvall, and U. Sartipy Aprotinin Decreases Postoperative Bleeding and Number of Transfusions in Patients on Clopidogrel Undergoing Coronary Artery Bypass Graft Surgery: A Double-Blind, Placebo-Controlled, Randomized Clinical Trial Circulation, August 30, 2005; 112(9_suppl): I-276 - I-280. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
