|
|
||||||||
Ann Thorac Surg 2006;81:1650-1657
© 2006 The Society of Thoracic Surgeons
a Department of Cardiothoracic Anesthesia, The Cleveland Clinic Foundation, Cleveland, Ohio
b Department of Quantitative Health Sciences, The Cleveland Clinic Foundation, Cleveland, Ohio
c Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio
Accepted for publication December 8, 2005.
* Address correspondence to Dr Koch, Department of Cardiothoracic Anesthesia (G-3), The Cleveland Clinic Foundation, Cleveland, OH 44195 (Email: kochc{at}ccf.org).
| Abstract |
|---|
|
|
|---|
METHODS: The US Social Security Death Index was used to ascertain survival status for 10,289 patients who underwent isolated coronary artery bypass grafting from January 1, 1995 through June 28, 2002. The outcome measure was all-cause mortality during the follow-up period. Unadjusted survival estimates were performed using the Kaplan-Meier techniques. Survival curves for transfusion status were compared with the log-rank test. The parametric decomposition model was used for risk-adjusted survival. A balancing score was calculated for each patient and forced into the final model.
RESULTS: Survival among transfused patients was significantly reduced as compared with nontransfused patients. The instantaneous risk of death displayed a biphasic pattern: a declining hazard phase from the time of the operation (early hazard) up until 6 months postoperatively and then a late hazard that continued out until about 10 years. Transfusion of red cells was associated with a risk-adjusted reduction in survival for both the early (0.34 ± 0.02, p < 0.0001) and late phases (0.074 ± 0.016, p < 0.0001).
CONCLUSIONS: Perioperative PRBC transfusion is associated with adverse long-term sequela in isolated CABG. Attention should be directed toward blood conservation methods and a more judicious use of PRBC.
| Introduction |
|---|
|
|
|---|
| Material and Methods |
|---|
|
|
|---|
Fifty-three patients had two surgeries during the follow-up period; hence, only the second surgery was kept in for the analysis. The remaining 11,910 patients were followed up with the United States Social Security Death Index to ascertain survival status during the follow-up interval. Registry records were linked to the death data based on key identifiers such as patient name, date of birth, social security number, gender, date last known alive, and state of last known residence. Because our institution is a large international referral center, 1,592 patients did not have social security numbers and their death information was considered missing. Another 29 patients did not have complete surgical information regarding operative dates for accurate vital status follow-up and were excluded from the analysis. The final dataset had 10,289 patients. The outcome measure for this study was all-cause mortality during the follow-up period. Baseline characteristics, clinical presentation, transfusion status, and operative variables are described for the patient population in Tables 1 and 2.
|
|
2 or Fisher's exact tests for categoric variables and the Wilcoxon rank sum test for continuous variables where appropriate.
Survival
Overall unadjusted survival estimates were performed using the Kaplan-Meier techniques. Survival curves for transfusion status were plotted and compared with the log-rank test. The plots displayed an early phase within about six months after surgery with a much higher risk for death than the late phase with lower risk after six months. To account for nonproportionality, we applied the decomposition method proposed by Blackstone and colleagues [13] to the whole dataset. Because the hazard function is time-varying as well as the influence of specific risk factors associated with the events, we used this model to examine survival after surgery. This method allowed for simultaneous modeling of the hazard functions and varying hazard ratios on both phases.
We considered preoperative and intraoperative variables as well as a number of transformations and interaction terms as potential prognostic factors for survival (Tables 1 and 2). Large scale variable selection was done by bootstrap aggregation (bagging) methods with 200 bootstrap samples out of the original dataset with replacement [14]. Stepwise model selection was applied to each sample with an entry p value of 0.10 and retention p value of 0.05. Variables selected in more than half of the samples were retained. In order for the linearity assumption of the regression model to hold, continuous and ordinal variables underwent various transformations and were entered into the model selection procedure. During the initial analysis of the data we found that body mass index appeared to have a bowl-shaped relationship with the long-term survival. This was first identified as a statistically significant quadratic term for body mass index in the model. The bottom of the bowl shape was found to be near 30. In order to reflect this relationship with some statistical parsimony, we adopted a regression spline type approach; for example, allow different slopes for body mass index less than 30 and greater than 30. In this way the model was simple, easy to interpret, and yet revealed the nonlinear relationship underneath. The list of variables was then submitted to the decomposition model for final analysis. Interaction terms between blood product variables were entered into the final model, but none were found to be significant at the 0.05 level.
To further control for confounding, a balancing score was calculated for each patient as the predicted value of red blood units in a linear model on all the preoperative and intraoperative variables. The balancing score was forced into the final models to further reduce bias. All analyses were performed using SAS 8.2 software (SAS Institute, Cary, NC).
| Results |
|---|
|
|
|---|
|
|
|
| Comment |
|---|
|
|
|---|
|
Additionally, variables associated with early hazard for death such as increasing age, peripheral vascular disease, left main trunk disease, reduced left ventricular function, increased preoperative creatinine and bilirubin, and reoperation have been described by others to be associated with an early hazard for death [16, 17].
Late Hazard
The association between PRBC transfusion and reduced long-term survival remained statistically significantly different from those who did not receive a PRBC transfusion throughout the follow-up period. The hazard for death associated with a 2 unit PRBC transfusion was similar in impact to the incremental risk of reoperation, reduced ejection fraction, or left main disease. The hazard for death associated with a 4 unit PRBC transfusion was similar in impact to the incremental risk of peripheral vascular disease, elevated preoperative creatinine, or history of heart failure.
Previous blood transfusion investigations have analyzed transfusion as a binary variable (yes/no), thereby limiting an ability to examine a dose-dependent relationship between each unit transfused and survival [12, 18]. Engoren and colleagues [12] reported a risk-adjusted reduction in late survival for CABG patients who were transfused PRBC. Their findings were similar to ours in that they reported a biphasic risk of death whereby early mortality was higher in the transfused group and decreased at one year, but still remained twice that of the in the nontransfused group at five years follow-up. Kuduvalli and colleagues [18] recently reported that perioperative PRBC transfusion was associated with a risk-adjusted increase in mortality during a one year follow-up period for patients undergoing isolated CABG.
The mechanism whereby PRBC transfusion influences adverse outcome remains speculative. The systemic inflammatory response may mediate the negative impact of transfusion on outcome [19, 20]. Patients undergoing cardiac surgery manifest a noninfectious inflammatory response reflected by increased concentrations of inflammatory mediators [21]. The degree of response exhibited by an individual patient may be related to intrinsic patient-related factors and degree of surgical trauma. Moreover, PRBC transfusion affects the release of inflammatory mediators [22]. Delivery of oxygen to tissues from stored blood may be limited due to severe depletion of 2,3 diphosphoglycerate [18]. Finally, there is evidence to support a role of immunomodulatory effects of blood transfusion in some of the observed adverse outcomes related to PRBC transfusion [23, 24]. A number of alterations in blood cells (such as decreases in the number of circulating lymphocytes, modification in the T-cell helper/suppressor ratio, and activation of immune cells) have been associated with PRBC transfusion. In addition, these changes may persist for months after a PRBC transfusion [23, 25].
Other Variables Related to Long-Term Survival
Additional variables related to reduced late survival in our study (see Table 3) have also been reported, in other investigations, to be associated with long-term survival [2630]. Weintraub and colleagues [30], in a 20-year follow-up study of CABG patients, reported reduced survival associated with age, female gender, hypertension, angina class, prior CABG, ejection fraction, number of vessels diseased, and increasing weight. Meyers and colleagues [17] reported late survival was related to heavier weight, prior myocardial infarction, diabetes, smoking, and use of vein grafts only in CABG patients from the Coronary Artery Surgery Study registry. Similarly, Thourani and colleagues [29] reported multivariate correlates of long-term mortality were diabetes, older age, reduced ejection fraction, hypertension, congestive heart failure, number of vessels diseased, and urgent or emergent operation in CABG patients.
Improved survival in our investigation was associated with a body mass index less than 30, increasing serum albumin, preoperative intraaortic balloon pump, hyperlipidemia, alcohol use, and internal thoracic artery grafting. The improved survival for patients with hyperlipidemia may be related to treatment with statin therapy. Recent investigations have demonstrated beneficial effects of statin therapy on reductions in adverse clinical events [31, 32]. The strong protective survival advantage of internal thoracic artery grafting has been well-established [33, 34].
Variation in Risk Factors With Follow-Up Time
We found a number of variables to exhibit different strengths at different follow-up times (see Table 3). Modulation of PRBC transfusion on early risk was far greater than late. Myers and colleagues [17] reported on the long-term survival after CABG from the Coronary Artery Surgery Study. They demonstrated time-varying effects of risk factors that were predictive of early, late, and constant phases of follow-up. Similarly, Gao and colleagues [35] recently reported on the time-varying effects in mortality risk factors for CABG patients. We reported that body mass index greater than 30, comorbid conditions, heart failure, and history of atrial fibrillation were variables related only to late risk; red cell transfusion, increasing age, creatinine, reoperation, reduced ejection fraction, peripheral vascular disease, and left main trunk disease conferred risk for both the early and late hazard for death.
Clinical Implications
Well-entrenched beliefs and expert opinion guide current transfusion recommendations [37]. However, much of this is due to a lack of randomized controlled trials as well as a lack of measures to access a patient's ability to compensate for anemia [37]. A recent clinical study reported lowest hematocrit on cardiopulmonary bypass (less than 22%) to be associated with increased morbidity and predictive of worse 0 to 6 year survival [36]. Transfusion of PRBC is certainly beneficial for some patients; however, there is demonstrable evidence of its association with a number of adverse outcomes. Giving penicillin to all patients with a fever, "universal fever trigger," is perhaps the best demonstrable analogy to a transfusion trigger in response to a specific hemoglobin level; "...some patients are helped, for others it has no impact and for some it is harmful" [37].
Furthermore, PRBCs are a limited resource. Restrictive transfusion guideline strategies have been demonstrated to be at least as effective and possibly superior to a liberal transfusion strategy in critically ill patients as well as an effective tool in decreasing PRBC utilization [38]. Despite transfusion guidelines and perceptions that blood conservation methods are already in place, a large percentage of cardiac surgical patients continue to be transfused.
Limitations
Our investigation is faced with limitations specific to nonrandomized study designs. For example, whether transfusion was a marker for increased illness not captured in our database or is an independent predictor for reduced survival cannot be determined with certainty. However, risk factors associated with reduced survival in CABG have been extensively investigated in the cardiac surgical literature and included and controlled for in our modeling. In addition, detailed information regarding leukoreduction of PRBC was not available from our blood bank at the individual patient level. Universal leukocyte reduction was instituted in 2002. It has been reported that leukoreduction of PRBC may improve midterm survival after cardiac surgery [39]. Lacking from our investigation and others is an explanation of mechanisms associated with the adverse early and late outcomes observed with transfusion of PRBC. Further investigation is needed to explore the role of immunomodulation or inflammation in the observed adverse outcomes.
Conclusion
We report that PRBC transfusion is associated with a risk-adjusted reduction in survival for patients undergoing isolated CABG. Blood conservation methods should be implemented and enforced, and more restrictive transfusion guidelines based on randomized controlled trials must be put in place to assure a more judicious use of PRBC.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Engoren, R. H. Habib, J. Hadaway, A. Zacharias, T. A. Schwann, C. J. Riordan, S. J. Durham, and A. Shah The Effect on Long-Term Survival of Erythrocyte Transfusion Given for Cardiac Valve Operations Ann. Thorac. Surg., July 1, 2009; 88(1): 95 - 100. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. D. Surgenor, R. S. Kramer, E. M. Olmstead, C. S. Ross, F. W. Sellke, D. S. Likosky, C. A. S. Marrin, R. E. Helm Jr, B. J. Leavitt, J. R. Morton, et al. The Association of Perioperative Red Blood Cell Transfusions and Decreased Long-Term Survival After Cardiac Surgery Anesth. Analg., June 1, 2009; 108(6): 1741 - 1746. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. S. Murphy, E. A. Hessel II, and R. C. Groom Optimal Perfusion During Cardiopulmonary Bypass: An Evidence-Based Approach Anesth. Analg., May 1, 2009; 108(5): 1394 - 1417. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Engoren and C. Arslanian-Engoren Long-term Survival in the Intensive Care Unit After Erythrocyte Blood Transfusion Am. J. Crit. Care., March 1, 2009; 18(2): 124 - 131. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. Sakwa, R. W. Emery, F. L. Shannon, J. M. Altshuler, D. Mitchell, D. Zwada, and A. R. Holter Coronary artery bypass grafting with a minimized cardiopulmonary bypass circuit: a prospective, randomized trial. J. Thorac. Cardiovasc. Surg., February 1, 2009; 137(2): 481 - 485. [Abstract] [Full Text] [PDF] |
||||
![]() |
R Rimpilainen, F Biancari, J. Wistbacka, P Loponen, S. Koivisto, J Rimpilainen, K Teittinen, and J Nissinen Outcome after coronary artery bypass surgery with miniaturized versus conventional cardiopulmonary bypass Perfusion, November 1, 2008; 23(6): 361 - 367. [Abstract] [PDF] |
||||
![]() |
W. Dietrich, M. Spannagl, J. Boehm, K. Hauner, S. Braun, T. Schuster, and R. Busley Tranexamic Acid and Aprotinin in Primary Cardiac Operations: An Analysis of 220 Cardiac Surgical Patients Treated with Tranexamic Acid or Aprotinin Anesth. Analg., November 1, 2008; 107(5): 1469 - 1478. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. McGrath, C. G. Koch, M. Xu, L. Li, T. Mihaljevic, P. Figueroa, and E. H. Blackstone Platelet Transfusion in Cardiac Surgery Does Not Confer Increased Risk for Adverse Morbid Outcomes Ann. Thorac. Surg., August 1, 2008; 86(2): 543 - 553. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-H. Yap, L. Lau, M. Krishnaswamy, M. Gaskell, and M. Yii Age of Transfused Red Cells and Early Outcomes After Cardiac Surgery Ann. Thorac. Surg., August 1, 2008; 86(2): 554 - 559. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. T. Newsome, R. S. Weller, J. C. Gerancher, M. A. Kutcher, and R. L. Royster Coronary Artery Stents: II. Perioperative Considerations and Management Anesth. Analg., August 1, 2008; 107(2): 570 - 590. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Lindvall, U. Sartipy, T. Ivert, and J. van der Linden Aprotinin is Not Associated With Postoperative Renal Impairment After Primary Coronary Surgery Ann. Thorac. Surg., July 1, 2008; 86(1): 13 - 19. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Benjamin, R. Y. Dodd, S. W. Hall, R. H. Habib, A. Zacharias, T. Frenzel, W. Sibrowski, M. Westphal, C. G. Koch, L. Li, et al. Red-Cell Storage and Complications of Cardiac Surgery N. Engl. J. Med., June 26, 2008; 358(26): 2840 - 2842. [Full Text] [PDF] |
||||
![]() |
M. S. Aapro and H. Link September 2007 Update on EORTC Guidelines and Anemia Management with Erythropoiesis-Stimulating Agents Oncologist, May 1, 2008; 13(suppl_3): 33 - 36. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. G. Koch, L. Li, D. I. Sessler, P. Figueroa, G. A. Hoeltge, T. Mihaljevic, and E. H. Blackstone Duration of Red-Cell Storage and Complications after Cardiac Surgery N. Engl. J. Med., March 20, 2008; 358(12): 1229 - 1239. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Westaby Aprotinin: Twenty-five years of claim and counterclaim J. Thorac. Cardiovasc. Surg., March 1, 2008; 135(3): 487 - 491. [Full Text] [PDF] |
||||
![]() |
A. DeAnda Jr. Aprotinin and Cardiac Surgery J. Thorac. Cardiovasc. Surg., March 1, 2008; 135(3): 492 - 494. [Full Text] [PDF] |
||||
![]() |
D. Pagano, N. J. Howell, N. Freemantle, D. Cunningham, R. S. Bonser, T. R. Graham, J. Mascaro, S. J. Rooney, I. C. Wilson, R. Cramb, et al. Bleeding in cardiac surgery: The use of aprotinin does not affect survival J. Thorac. Cardiovasc. Surg., March 1, 2008; 135(3): 495 - 502. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Rawn Blood Transfusion in Cardiac Surgery: A Silent Epidemic Revisited Circulation, November 27, 2007; 116(22): 2523 - 2524. [Full Text] [PDF] |
||||
![]() |
F. F. Immer, A. Ackermann, E. Gygax, M. Stalder, L. Englberger, F. S. Eckstein, H. T. Tevaearai, J. Schmidli, and T. P. Carrel Minimal Extracorporeal Circulation is a Promising Technique for Coronary Artery Bypass Grafting Ann. Thorac. Surg., November 1, 2007; 84(5): 1515 - 1521. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Bennett-Guerrero, T. H. Veldman, A. Doctor, M. J. Telen, T. L. Ortel, T. S. Reid, M. A. Mulherin, H. Zhu, R. D. Buck, R. M. Califf, et al. Evolution of adverse changes in stored RBCs PNAS, October 23, 2007; 104(43): 17063 - 17068. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. J. Gardner To Transfuse or Not to Transfuse Circulation, July 31, 2007; 116(5): 458 - 460. [Full Text] [PDF] |
||||
![]() |
A. Kulier, J. Levin, R. Moser, G. Rumpold-Seitlinger, I. C. Tudor, S. A. Snyder-Ramos, P. Moehnle, D. T. Mangano, and for the Investigators of the Multicenter Study of Impact of Preoperative Anemia on Outcome in Patients Undergoing Coronary Artery Bypass Graft Surgery Circulation, July 31, 2007; 116(5): 471 - 479. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. H. Tyson III, E. Rodriguez, O. C. Elci, T. C. Koutlas, W. R. Chitwood Jr, T. B. Ferguson, and A. P. Kypson Cardiac Procedures in Patients With a Body Mass Index Exceeding 45: Outcomes and Long-Term Results Ann. Thorac. Surg., July 1, 2007; 84(1): 3 - 9. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. G. Koch and E. H. Blackstone Reply Ann. Thorac. Surg., June 1, 2007; 83(6): 2260 - 2261. [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 |