Ann Thorac Surg 2008;86:560-567. doi:10.1016/j.athoracsur.2008.04.048
© 2008 The Society of Thoracic Surgeons
Original Articles: Adult Cardiac
Impact of Aprotinin on Adverse Clinical Outcomes and Mortality up to 12 Years in a Registry of 3,337 Patients
Stephen A. Olenchock, Jr, DOa,*,
Peter H.U. Lee, MD, MPHa,
Tamar Yehoshua, CCPa,
Sabina A. Murphy, MPHb,
James Symes, MDc,
George Tolis, Jr, MDa
a Division of Cardiothoracic Surgery, Caritas St. Elizabeth's Medical Center and Tufts University School of Medicine, Boston, Massachusetts
b Cardiovascular Division, Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
c Division of Cardiothoracic Surgery, University of Miami School of Medicine, Miami, Florida
Accepted for publication April 14, 2008.
* Address correspondence to Dr Olenchock, Division of Cardiothoracic Surgery, Caritas St. Elizabeth's Medical Center, 11 Nevins St, Suite 306, Boston, MA 02135 (Email: olenchock{at}pol.net).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
 |
Abstract
|
|---|
Background: Recent studies have suggested increased renal complications and long-term mortality with aprotinin use in coronary artery bypass grafting (CABG) patients. However, these studies have been criticized for including multiple centers and different dosing strategies. We analyzed prospectively collected registry data from a single center hospital utilizing a full-dose aprotinin regimen to evaluate if aprotinin was associated with increased mortality and adverse outcomes compared with Amicar.
Methods: Data were prospectively collected from 1994 to 2006 at a teaching hospital. Long-term mortality was collected from a Social Security database. To account for differences between aprotinin and Amicar-treated patients, a propensity score was generated and propensity-stratified multivariate model for mortality were performed.
Results: Compared with Amicar-treated patients (n = 1,830), aprotinin-treated patients (n = 1,507) were older, more often female, had lower creatinine clearance, and more baseline risk factors. Blood loss was lower in aprotinin-treated patients (median 715 mL vs 918 mL, p < 0.001). Postoperative renal failure was significantly higher in aprotinin patients (6.2% vs 2.7%, p < 0.001). At median 5.4-year follow-up (up to 12.2 years), aprotinin-treated patients had higher mortality versus Amicar-treated patients (Kaplan-Meier failure rates 43.5% vs 23.7% at 8 years, p < 0.0001). In a propensity-stratified model with multivariate adjustment, aprotinin remained associated with increased mortality (hazard ratio 1.62, 95% CI 1.39 to 1.90, p < 0.001). There was a stepwise relationship between weight-based aprotinin dose and mortality (p-trend < 0.001).
Conclusions: Among patients undergoing CABG in this registry, aprotinin use was associated with increased renal failure and higher mortality through 12 years in a propensity-stratified analysis. The increased mortality may be related to higher concentrations of aprotinin received.
|
Introduction
|
|---|
Aprotinin, a serine protease inhibitor, is indicated for prophylactic use to reduce perioperative blood loss and the need for blood transfusion in patients undergoing cardiopulmonary bypass in the course of coronary artery bypass graft (CABG) surgery who are at an increased risk for blood loss and blood transfusion [1]. As stated in the aprotinin prescribing information, the benefit of aprotinin to patients undergoing primary CABG surgery should be weighed against the risk of anaphylaxis associated with any subsequent exposure to aprotinin [1]. High dose aprotinin has been shown in several studies to decrease postoperative bleeding and transfusion of blood products in patients undergoing CABG [2–4].
More recently, safety concerns regarding the possible side effects of aprotinin have been raised [5–9]. In recent reports, the risk of renal failure, stroke, and myocardial infarction (MI) was higher in patients who had received aprotinin [5–8]. Additionally, Mangano and colleagues [6] reported that aprotinin use was associated with an increased risk of long-term mortality after CABG surgery. These findings are in contrast to the randomized trials, meta-analysis, and Cochrane Collaboration summary, which showed no significant increase in mortality, MI, or renal failure [10–12]. Some critiques of these published manuscripts have been the inclusion of multiple centers in multiple countries with variable dosing strategies [13].
Given the conflicting studies and recent uncertainty surrounding aprotinin use, we analyzed prospectively collected registry data from a single center hospital utilizing a full-dose aprotinin regimen to evaluate if aprotinin was associated with increased mortality and adverse outcomes when compared with Amicar.
|
Material and Methods
|
|---|
Consecutive patient data were collected from 1994 through 2006 for all patients undergoing CABG-only surgery at a single institution (Caritas St. Elizabeth's Medical Center, Boston, MA). The analysis included patients who underwent a single CABG-only surgery and received either aprotinin or Amicar during the surgery. Decision of which (if any) agent to use was at the discretion of the surgeon. Patients who received other antifibrinolytic agents or neither agent were not included in the analysis. Institutional Review Board approval was obtained for data collection and to perform the present study. Individual patient data were prospectively collected.
Definitions are consistent with the Society of Thoracic Surgeons Registry for Adult Cardiac Surgery. Preoperative cardiac status evaluated included MI prior to or during the current hospitalization, cardiogenic shock (clinical state of hypoperfusion at the time of procedure, defined as either systolic blood pressure less than 80 mm Hg; cardiac index less than 1.8 despite maximal treatment; or need for intravenous inotropes or intraaortic balloon pump to maintain systolic blood pressure greater than 80 mm Hg or cardiac index greater than 1.8), angina pectoris, and arrhythmia (defined as sustained ventricular tachycardia, ventricular fibrillation, atrial fibrillation, atrial flutter, third degree heart block). Preoperative renal failure was defined as a documented history of renal failure or a history of creatinine greater than 2.0 mg/dL. Postoperative complications evaluated included reoperation for bleeding, infection, neurologic (postoperative stroke greater than 72 hours, transient neurologic deficit, or continuous coma 
24 hours), pulmonary (prolonged ventilation, pulmonary embolism, or pneumonia), and renal failure (increase of serum creatinine to > 2.0 mg/dL and 2 times the most recent preoperative creatinine level or newly required dialysis). Additionally, patients were evaluated for readmission to the hospital within 30 days. Long-term mortality was collected using the Social Security Administration Death Master File (http://Ancestry.com) [14–16].
Amicar was administered as a 10 g bolus followed by 2 g/hour infusion during the case. Aprotinin was administered according to the standard "Hammersmith" full-dose protocol. In order to evaluate the effect of weight-based dosing on outcomes, we calculated the total dose of aprotinin given by determining the sum of the pump prime (280 mg), plus initial load of the medication (280 mg), plus the infusion during the total surgical time (70 mg/hour). This total calculated aprotinin total dose was then divided by the patient weight, giving a weight-based dose of aprotinin received. Patients were then divided into tertiles of weight-based dosing received (low-dose, intermediate dose, and high-dose).
Statistical Analysis
Baseline and procedural characteristics are presented as frequencies for categoric variables and medians and interquartile ranges for continuous variables. Comparisons between baseline and procedural characteristics for patients who received aprotinin versus Amicar were made using the 
2 test for categoric variables and Wilcoxon rank for continuous variables.
Using logistic regression modeling, a propensity score was developed for the use of aprotinin versus Amicar. The propensity score is the conditional probability of receiving a treatment (aprotinin vs Amicar) given the observed covariates included in the model. The model included baseline and clinical characteristics of age, gender, ejection fraction (log transformed), body mass index, diabetes, hypertension, smoking, history of renal failure, family history of coronary artery disease, hypercholesterolemia, peripheral vascular disease, cerebrovascular disease, previous cardiovascular intervention, MI, congestive heart failure, angina, cardiogenic shock, arrhythmia, left main disease, time period of surgery, and surgeon. The overall C-statistic for the logistic model was 0.83. We used an iterative process to define and verify the following: (1) the strata were balanced on the score (ie, there was no association between the predicted probability from the model and treatment with aprotinin versus Amicar within any stratum); and (2) the strata were balanced on covariates (ie, there was no association between any covariate included in the propensity score model and treatment with aprotinin versus Amicar within any stratum). Using these iterative methods of definition and verification, the predicted exposure levels were divided into 9 propensity strata, which achieved balance on all 21 covariates.
For the outcomes of postoperative renal failure and need for blood products, a Cochran-Mantel-Haenszel test stratifying by the propensity stratum was used to estimate the pooled odds ratio (OR) effect of aprotinin treatment across the stratum.
Given the long-term and varying lengths of follow-up, survival analysis methods were used to evaluate differences in mortality between aprotinin versus Amicar. Mortality event rates are presented as Kaplan-Meier (KM) failure rates at 5 years (for purposes of comparison to earlier studies [6]) and at 8 years and comparisons are made for the full length of follow-up using the log-rank test. For survival analysis, multivariate models are presented using Cox proportional hazard models with the treatment effect within each propensity score stratum measured and averaged across strata to estimate the effect of aprotinin treatment on mortality. These models also include risk adjustment for covariates related to mortality, including age, gender, ejection fraction, body mass index, diabetes, hypertension, smoking, history of renal failure, hypercholesterolemia, peripheral vascular disease, cerebrovascular disease, congestive heart failure, cardiogenic shock, arrhythmia, and left main disease. Sensitivity models were performed including creatinine clearance, which was available in 1,488 patients, and procedural characteristics of perfusion duration and cross-clamp duration.
For the weight-based dose analysis and long-term mortality, a test for trend of the survivor function across ordered groups was used to evaluate whether there was a stepwise increase in mortality with increasing weight-based doses of aprotinin. A Cox proportional hazard model was also performed, including risk adjustment for the covariates related to mortality listed in the previous paragraph, with the exception of body mass index because weight was used for the calculation of the aprotinin dose received. All analyses were conducted using Stata/SE 9.2 statistical software (StataCorp, College Station, TX).
|
Results
|
|---|
The median length of follow-up was 5.4 years, with the longest follow-up of 12.2 years; one-quarter of patients had follow-up beyond 8 years. Use of aprotinin increased over time, with aprotinin used in 28% of patients during the period from 1994 to 2000, 49% of patients from 2001 to 2003, and 93% of patients from 2004 to 2006.
Baseline Characteristics
Compared with Amicar-treated patients (n = 1,830), aprotinin-treated patients (n = 1,507) were older (median age 72 years vs 68 years), more often female (33.7% vs 26.3%), had lower creatinine clearance (median 66.6 mL/minute vs 78.1 mL/minute) and more baseline risk factors, including hypertension and renal failure (Table 1). Additionally, they were more frequently in New York Heart Association class IV heart failure (31.7% vs 21.6%) and more often presented with left main disease 50% or greater (35.3% vs 25.2%).
Procedural Characteristics and Postoperative Complications
Surgical duration was slightly longer in patients who received aprotinin, although perfusion duration was slightly shorter (Table 2). Blood loss was lower in the aprotinin-treated patients compared with the Amicar-treated patients, both for loss by 6 hours (median 230 mL vs 255 mL, p < 0.0001) and for total blood loss (median 715 mL vs 918 mL, p < 0.0001; p = 0.0003 in propensity-stratified analysis). On univariate analysis, use of postoperative blood products was higher in the aprotinin-treated patients (46.9% vs 40.0%, p = 0.0003); however, in the propensity-stratified analysis there was no difference in need for blood products (odds ratio [OR] 1.03, 95% confidence interval [CI] 0.86 to 1.22, p = 0.77). Median intensive care unit duration was slightly longer in the aprotinin group (27 hours vs 24 hours), as was total postoperative length of hospitalization (Table 2).
There was no difference in the frequency of reoperation for bleeding, postoperative infection complications, or perioperative MI (Table 3). However, there were increases among aprotinin-treated patients in pulmonary complications (7.4% vs 4.8%, p = 0.002) and postoperative renal failure (6.2% vs 2.7%, p < 0.0001). The increase in postoperative renal failure remained associated with aprotinin use in the propensity-stratified analysis (OR 2.00, 95% CI 1.30 to 3.06, p = 0.0017) as well as when patients with preoperative renal failure were excluded from the analysis (5.2% vs 2.6%, OR 1.78, 95% CI 1.13 to 2.81, p = 0.014). Aprotinin-treated patients were also more frequently readmitted within 30 days of surgery as compared with Amicar-treated patients (4.4% vs 2.0%, p < 0.001).
Mortality
On univariate analysis, mortality was higher in the aprotinin-treated patients compared with the Amicar-treated patients both early (by 30 days) and late (Fig 1,
p < 0.0001; at 30 days, KM failure rates 4.1% vs 1.0%; at 5 years, KM failure rates 26.8% vs 12.8%; at 8 years, KM failure rates 43.5% vs 23.7%). In a propensity-stratified model with multivariate adjustment, aprotinin remained associated with increased mortality compared with Amicar (hazard ratio [HR] 1.62, 95% CI 1.39 to 1.90, p < 0.001). Results were similar in a sensitivity model that also included creatinine clearance (HR 1.69, 95% CI 1.23 to 2.31, p = 0.001). Likewise, results were similar when including primary surgeon as a covariate in the model (HR 1.61, 95% CI 1.38 to 1.89, p < 0.001) and when including the procedural characteristics of perfusion duration and cross-clamp duration as covariates in the model (HR 1.53, 95% CI 1.30 to 1.80, p < 0.001). Results were consistent among key subgroups, including age, gender, diabetes, presence of left main disease, ejection fraction less than 0.40 (Fig 2). However, a significant interaction (p interaction = 0.024) was observed with presence of preoperative renal failure; in this subgroup of patients, use of aprotinin was associated with a more than twofold excess in mortality (HR 2.57, 95% CI 1.66 to 3.96, p < 0.001; Fig 2).
View larger version (17K):
[in this window]
[in a new window]
|
Fig 1. Mortality by aprotinin and Amicar use. Kaplan-Meier failure curves of mortality for patients who received aprotinin compared with Amicar. (— = aprotinin; - - - = Amicar.)
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig 2. Mortality in key subgroups. Hazard ratio plot from propensity-stratified multivariate models for mortality comparing aprotinin with Amicar in key subgroups. (EF = ejection fraction.)
|
|
Mortality by Weight Dose Group of Aprotinin
When categorizing patients by dose of aprotinin received per unit of body weight, the tertiles ranged from 5.2 to 10 mg/kg in the lowest tertile, 10.1 to 12.3 mg/kg in the middle tertile, and greater than 12.3 to 35 mg/kg in the highest tertile. There was a stepwise relationship between weight-based aprotinin dose and mortality (Fig 3;
p trend < 0.001). In a multivariate model, the high-dose aprotinin group trended to have a higher mortality than the low-dose group (HR 1.26, 95% CI 0.97 to 1.64, p = 0.084), while there was no difference between the intermediate compared with the low dose groups (HR 1.05, 95% CI 0.80 to 1.38, p = 0.73).
View larger version (17K):
[in this window]
[in a new window]
|
Fig 3. Mortality by weight-based dose of aprotinin. Kaplan-Meier failure curves of mortality for patients who received aprotinin by weight-based dose tertile. (— = high dose; - - - = mid dose; · · · = low dose.)
|
|
|
Comment
|
|---|
This observational study of more than 3,000 patients demonstrated an increased risk of renal failure for those who received aprotinin compared with Amicar. This finding is consistent with multiple studies, including a recent meta-analysis [7]. Additionally, the present study also showed an increased risk of death among patients treated with aprotinin through to 12 years, one of the longest follow-up periods reported to date that we could indentify in the published literature. Patients receiving aprotinin did have lower creatinine clearance and were of older age at baseline compared with the Amicar-treated patients. However, after multivariate adjustments and propensity score stratification, those that received aprotinin remained associated with increased renal failure and mortality. The higher mortality with aprotinin was particularly excessive in the subgroup of patients with renal failure prior to surgery.
Aprotinin during preclinical studies has been shown to produce deposits of proteins in the phagosomes of the epithelial cell of the proximal renal tubules, which are not excreted until the fifth to sixth day after intravenous administration [17]. High-dose aprotinin coupled with lower creatinine clearance and poorer preoperative renal function may account for some of the postoperative increase in renal failure seen in this study. Indeed, a significant interaction with respect to mortality was observed in the present study between aprotinin treatment and presence of renal failure prior to surgery. It has been demonstrated that the high-dose regimen produces more persistent changes in tubular function than low doses at 24 hours postoperatively. The mechanism may be explained by the high affinity for the renal tissue of the drug, which may result in saturation of the reabsorption mechanism at the proximal tubular level [17, 18]. While we did not record outcomes that account for the potential compounding effects of cardiopulmonary bypass, such as hypotension or use of vasopressors which may be additive in the effects of renal dysfunction, we did account for duration of time on cardiopulmonary bypass and cross-clamp time in the multivariate model.
The dosing strategy for aprotinin used in this evaluation was standard "Hammersmith" full-dose protocol. Plasma concentration of aprotinin is higher in those who receive full-dose as compared with half-dose strategies [19]. As early as 1994, only one year after Food and Drug Administration approval, alternate dose regimens for aprotinin were being evaluated, including those by Levy et al [20], who suggested a pharmacokinetics-based dosing scheme with aprotinin, which while likely more precise was not clinically feasible and was not adopted. Additionally, Beath and colleagues [19] demonstrated that there was a significant negative correlation between patient weight and plasma aprotinin concentration in patients receiving full-dose aprotinin (ie, the lower the body weight the higher the plasma aprotinin concentration). While a weight-based dosing regimen was suggested by the authors, full clinical evaluation of such a strategy was not undertaken; the only clinical outcome evaluated with such a strategy was thoracic drainage [21].
Using a weight-based model as a surrogate to plasma concentration, we hypothesized that higher doses of aprotinin per kilogram of body weight would be associated with increased mortality. Indeed, we observed a stepwise relationship between increase in the weight-based dose of aprotinin and mortality, most notably in the highest dose group, although the relationship was directionally but not statistically significant after multivariate adjustment. These findings suggest that some patients may be receiving excess plasma concentrations of aprotinin, which may have unknown deleterious effects, possibly including renal dysfunction, which subsequently contributes to the increased mortality. Further studies that attempt to elicit clinical outcomes attributed to aprotinin should take into account the plasma concentration of the medication in patients and consider dosing in a weight-based fashion.
There were several notable differences in higher baseline risk in the aprotinin group compared with the Amicar group, including older age, lower creatinine clearance, and more hypertension, heart failure, and left main disease. While propensity score stratification helps to balance the groups on the measured covariates, this study remains observational, and unmeasured confounders may impact the results. Although this analysis has its limitations and is not a randomized trial, it contributes additional novel information to the discussion around aprotinin use and dosing. As mentioned previously, the mortality data collected through up to 12 years is one of the longest reported in the literature at this time, with the recent manuscript by Mangano and colleagues [6] having follow-up of 5 years and data presented by Duke University [8] of up to 10 years, both of which showed higher mortality associated with aprotinin use. These data represent one of the largest single center evaluations of aprotinin use compared with a single antifibrinolytic, Amicar, currently reported. Individual surgeons were accounted for in a multivariate analysis for mortality and did not alter the overall results. Biases such as medication dosing and delivery have been reduced, as patients in the analysis received the "Hammersmith" or high-dose regimen, and patient care during cardiopulmonary bypass and during recovery in the postoperative period is standardized to one single institution. Additionally, the link between weight-based dosing and clinical outcome of increased mortality further support the notion that a "one-size-fits-all" dosing regimen may not be appropriate with this therapy. While aprotinin marketing is currently suspended pending final results of the "Blood conservation using Antifibrinolytics: a Randomized Trial in a cardiac surgery population" (BART) trial, additional consideration should be given to randomized trials of a weight-based dosing regimen of aprotinin, which may reduce the renal toxicity and overall mortality associated with the drug.
|
Discussion
|
|---|
DR ALESSANDRO PAROLARI (Milan, Italy): How do you explain the fact that up to 10 years the curves of late survival continue to diverge even if you correct by propensity score for all the other risk factors?
DR OLENCHOCK: That was one of the concerns we had as well. One of the things that we looked at was to try to determine if certain patients may be receiving excess amounts of this medication. The medication is currently served with a one-size-fits-all dosing. Patients in lower kilogram body weight or lower BMI may be receiving excessive amounts of concentration of this medication, and indeed we did find that when we separated them by tertiles these patients had a higher rate of renal dysfunction, which is probably what drives the excess long-term mortality in the higher dose groups. Even a small increase in renal dysfunction can likely have an impact on future survival for several years.
DR JOSEPH C. CLEVELAND, JR (Denver, CO): If I read your manuscript correctly, too, there was also an excess in pulmonary complications in the aprotinin group, and one of the questions, could you not even say that any aprotinin was associated with excess mortality? I like the idea of the dose basing, but if you look again, if you compare the Kaplan-Meier of the Amicar, even the low dose, is there a difference in those groups or not?
DR OLENCHOCK: That is true. Even if we consider a comparison of low dose aprotinin compared with Amicar, there was a difference in patients treated at all no matter which weight-based tertile they were in with aprotinin compared to Amicar, but the increase was less in the low dose tertile. The other thing is that these Kaplan-Meier curves do continue to diverge several years out, and our findings are consistent as far as long-term mortality. Our five-year mortality is 27%, which is very similar to other studies where Dr Mangano showed 21% and the group from Duke, who just presented at ASA, also showed that theirs was 36%. So it is true that no matter whether it was weight-based or not, there was increased risk, but it was less of an increase with the lower dose group.
DR JOHN W. HAMMON, JR (Winston-Salem, NC): Since you brought up Dr Mangano's name, I would like to respond to that. Propensity scoring does not correct for the wide disparity in risk that you had in your study. It could only be corrected for by a randomized prospective study. I think the point that you made about dosing is very valid. I hope there is somebody from Bayer in the audience that will listen to you.
DR OLENCHOCK: I agree. Thank you very much for your comment. Clearly a randomized prospective study is the gold standard for evaluating a therapy. That being said, observation and registry data can provide hypothesis generating information and hopefully this analysis does that.
|
References
|
|---|
- Trasylol prescribing safety informationhttp://www.fda.gov/Medwatch/safety/2006/safety06.htm#TrasylolAccessed January 2, 2008.
- Bidstrup BP, Royston D, Sapsford RN, Taylor KM. Reduction in blood loss and blood use after cardiopulmonary bypass with high dose aprotinin (Trasylol) J Thorac Cardiovasc Surg 1989;97:364-372.[Abstract]
- Cosgrove 3rd DM, Heric B, Lytle BW, et al. Aprotinin therapy for reoperative myocardial revascularization: a placebo-controlled study Ann Thorac Surg 1992;54:1031-1036.[Abstract]
- Lemmer Jr JH, Stanford W, Bonney SL, et al. Aprotinin for coronary bypass operations: efficacy, safety, and influence on early saphenous vein graft patency. A multicenter, randomized, double-blind, placebo-controlled study. J Thorac Cardiovasc Surg 1994;107:543-551.[Abstract/Free Full Text]
- Mangano DT, Tudor IC, Dietzel C, Multicenter Study of Perioperative Ischemia Research Group; Ischemia Research and Education Foundation The risk associated with aprotinin in cardiac surgery N Engl J Med 2006;354:353-365.[Abstract/Free Full Text]
- Mangano DT, Miao Y, Vuylsteke A, et al. Mortality associated with aprotinin during 5 years following coronary artery bypass graft surgery JAMA 2007;297:471-479.[Abstract/Free Full Text]
- Brown JR, Birkmeyer NJ, O'Connor GT. Meta-analysis comparing the effectiveness and adverse outcomes of antifibrinolytic agents in cardiac surgery Circulation 2007;115:2801-2813.[Abstract/Free Full Text]
- Shaw AD, Stafford-Smith M, White WD, et al. The effect of aprotinin on outcome after coronary-artery bypass grafting N Engl J Med 2008;358:784-793.[Abstract/Free Full Text]
- Schneeweiss S, Seeger JD, Landon J, Walker AM. Aprotinin during coronary-artery bypass grafting and risk of death N Engl J Med 2008;358:771-783.[Abstract/Free Full Text]
- 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 2004;128:442-448.[Abstract/Free Full Text]
- Henry DA, Moxey AJ, Carless PA, et al. Anti-fibrinolytic use for minimizing perioperative allogenic blood transfusion Cochrane Database Syst Rev 2001;1CD001886.
- Furnary AP, Wu YX, Hiratzka LF, Grunkemeier GL, Page US. Aprotinin does not increase the risk of renal failure in cardiac surgery patients Circulation 2007;116(suppl I):I127-I133.[Medline]
- Ferraris VA, Bridges CR, Anderson RP. Aprotinin in cardiac surgery. Letter to the Editor. N Engl J Med 2006;354:1953-1954.[Free Full Text]
- Social Security Death Indexhttp://www.Ancestry.com 2006.
- Schisterman EF, Whitcomb BW. Use of the social security administration death master file for ascertainment of mortality status Population Health Metrics 2004;2:2.[Medline]
- Cowper DC, Kubal JD, Maynard C, Hynes DM. A primer and comparative review of major U.S. mortality databases Ann Epidemiol 2002;12:462-468.[Medline]
- Rustom R, Grime S, Maltby P, Stockdale HR, Critchley M, Bone JM. A new method to measure renal tubular degradation of small filtered proteins in man using radiolabelled aprotinin (Trasylol) Clin Sci 1992;82:289-294.
- Faulí A, Gomar C, Campistol JM, Alvarez L, Manig AM, Matute P. Kidney-specific proteins in patients receiving aprotinin at high- and low-dose regimens during coronary artery bypass graft with cardiopulmonary bypass Eur J Anasthesiol 2005;22:666-671.
- Beath SM, Nuttall GA, Fass DN, Oliver Jr WC, Ereth MH, Oyen LJ. Plasma aprotinin concentrations during cardiac surgery: Full- versus half-dose regimens Anesth Analg 2000;91:257-264.[Abstract/Free Full Text]
- Levy JH, Bailey JM, Salmenpera M. Pharmacokinetics of aprotinin in preoperative cardiac surgical patients Anesthesiology 1994;80:1013-1018.[Medline]
- Royston D, Cardigan R, Gippner-Steppert C, Jochum M. Is perioperative plasma aprotinin concentration more predictable and constant after a weight-related dose regimen? Anesth Analg 2001;92:830-836.[Abstract/Free Full Text]
Top
Abstract
Introduction
