Correspondence
Red-Cell Storage and Complications of Cardiac Surgery
N Engl J Med 2008; 358:2840-2842June 26, 2008
- Article
To the Editor:
Koch et al. (March 20 issue)1 report increased risks of postoperative complications and death among patients undergoing cardiac surgery who were given blood that had been stored for more than 2 weeks, as compared with those given newer blood. These results are similar to those in a recently reported study; in that study, however, risk adjustment for confounding influences negated the storage effect.2 Koch et al. present predominantly non–risk-adjusted data, despite significant group differences. Risk-adjusted data are reported only for an inadequately described composite outcome, with greatly diminished significance (unadjusted P=0.001; adjusted odds ratio, 1.16 [95% confidence interval, 1.01 to 1.33]; P=0.03). Given the use of a composite end point and the opportunity for multiple comparisons, the clinical significance of this result is questionable, especially since many probable confounders are not considered, including emergency surgery,3 surgical blood loss and reexploration for bleeding,4 salvaged blood transfusion, aminocaproic acid and aprotinin use (withdrawn for adverse outcomes),5 and surgeon or year of surgery (to account for differences in the use of leukoreduced blood during the introduction of universal leukoreduction) and season of surgery (to account for variations in storage duration with seasonal blood shortages). It would be unfortunate if inappropriate lobbying by patients and their surgeons effected a major change in transfusion policy that was based on overinterpretation of unadjusted data.
5 ReferencesRichard J. Benjamin, M.D., Ph.D.
American Red Cross National Headquarters, Washington, DC 20006
benjaminr@usa.redcross. org Roger Y. Dodd, Ph.D.
American Red Cross Holland Laboratories, Rockville, MD 208531
Koch CG ,Li L ,Sessler DI , et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med 2008;358:1229-1239
Full Text | Web of Science | Medline2
van de Watering L ,Lorinser J ,Versteegh M ,Westendord R ,Brand A . Effects of storage time of red blood cell transfusions on the prognosis of coronary artery bypass graft patients. Transfusion 2006;46:1712-1718
CrossRef | Web of Science | Medline3
Nashef SA ,Roques F ,Michel P ,Gauducheau E ,Lemeshow S ,Salamon R . European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16:9-13
CrossRef | Web of Science | Medline4
Moulton MJ ,Creswell LL ,Mackey ME ,Cox JL ,Rosenbloom M . Reexploration for bleeding is a risk factor for adverse outcomes after cardiac operations. J Thorac Cardiovasc Surg 1996;111:1037-1046
CrossRef | Web of Science | Medline5
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
Full Text | Web of Science | Medline
To the Editor:
The study by Koch et al. suggests that storage of red cells for more than 14 days, as compared with a shorter storage period, is associated with poorer outcomes for patients undergoing cardiac surgery. Several questions regarding this retrospective study arise. How many patients in the database did not receive any blood products, and what was the outcome for these patients? The authors state that 2364 patients received a mixture of “newer and older” blood and were excluded. Why not report the outcomes for these patients as well (with the use of the same multivariate analysis) as a comparison? Does older blood negate the benefits of newer blood? Finally, although the two groups were said to have received “similar” numbers of units of blood per patient, the figures suggest otherwise: 8802 newer units in 2872 patients, or 3.06 units per patient, as compared with 10,782 older units in 3130 patients, or 3.44 units per patient, a difference of almost 13%. This difference is likely to be significant. The study reinforces the need for a prospective trial.
Scott W. Hall, M.D., Ph.D.
Blood Bank of Delmarva, Newark, DE 19713
shall@dclp.com To the Editor:
In the potentially practice-changing article by Koch and colleagues, we noted several issues that, if addressed, might mitigate or accentuate the reported adverse effects of transfusions of older blood. First, instead of causing complications, postoperative transfusions are frequently used after complications of cardiac surgery have occurred. Therefore, separate consideration of intraoperative and postoperative transfusion would be useful. Second, the authors note that rates of isolated coronary-artery bypass grafting (CABG) and isolated valve replacement were similar in the two groups. This does not ensure that complex and higher-risk procedures (e.g., CABG and valve replacement combined and double-valve replacement) were balanced or adjusted for in the analysis. Finally, restricting the comparison of patients to those who received only newer blood and those who received only older blood is fitting, but is it enough? With the use of the authors' approach, a hypothetical patient given 5 units of 15-day-old red cells and 1 unit of 31-day-old red cells will be classified as having received older blood than will another patient given 5 units of 30-day-old red cells. It would be reassuring if the results of the multivariate model were confirmed on the basis of a variable of “mean days of storage.”
Robert H. Habib, Ph.D.
University of Toledo College of Medicine, Toledo, OH 43614
robert_habib@mhsnr.org Anoar Zacharias, M.D.
Saint Vincent Mercy Medical Center, Toledo, OH 43608To the Editor:
In their retrospective study, Koch et al. describe the effects of the duration of blood storage on morbidity and mortality among patients undergoing cardiac surgery. However, the interpretation of the data may be limited by the following facts. First, the distribution of individual blood groups differed between the study groups (e.g., for blood group O, 31.1% of patients received newer blood vs. 53.1% of patients receiving older blood). Second, transfused red cells were less often matched by blood group in patients receiving older blood. Third, other, non-ABO blood-group systems, as well as the number of (potentially) ABO-mismatched transfusions per group, were not specified. Fourth, the effects of anticoagulants, heparin antagonists (e.g., protamine), or blood-saving strategies (e.g., cell-saver technique) and intraoperative blood loss or preoperative autologous blood donation were not taken into consideration. Well-designed prospective, randomized trials are needed to explicitly clarify the effect of the duration of red-cell storage on morbidity and mortality among critically ill patients.
Tim Frenzel, M.D.
Walter Sibrowski, M.D., Ph.D.
Martin Westphal, M.D., Ph.D.
University Hospital of Muenster, 48149 Muenster, Germany
martin.westphal@gmx. net Author/Editor Response
Regarding the concerns of Benjamin and Dodd, our composite outcome consisted of complications defined by the Society of Thoracic Surgeons; since they are available on the society's Web site, we did not include detailed definitions of the complications. We illustrated findings from our analysis with non–risk-adjusted depictions, but our study inferences were based entirely on the risk-adjusted results. Of the five “confounders” that Benjamin and Dodd mention, two we did include: emergency surgery and leukocyte-reduction status. Aminocaproic acid was uniformly administered, and operative blood loss was never estimated. The true confounder, reexploration for bleeding, is quantitatively captured (we believe) by considering units transfused. We also addressed temporal trends in the modeling, but the date of operation was not a significant variable. It is unclear to us how “season of surgery” can confound the analysis, since storage duration is storage duration, regardless of the season when the blood was procured.
With regard to Hall's questions, we have previously examined the effect of red-cell transfusion by comparing patients who received transfusions with patients who did not.1-4 The 13% difference in units transfused, which he calculates, is based on Table 1 of our article. However, the numbers of units (8802 units of newer blood and 10,782 units of older blood) excluded units with missing blood types. The total numbers of units transfused were 9210 (average, 3.21) in the newer-blood group and 10,832 (average 3.46) in the older-blood group (Figure 1A of our article). The admixture group generally received more units of blood and had more complications. We do not know whether an admixture of newer with older blood negates the effects of older blood.
Regarding Habib and Zacharias's comments, the exact timing of complications in relation to transfusion is problematic, since for each, the date of occurrence is recorded rather than the specific hour. Of note, transfusion for our patients generally occurs early in the perioperative course; infection and prolonged ventilatory support are events that occur later. If a patient dies, this event certainly follows any transfusion. The expression of red-cell age was dichotomized at 14 days, but in the Supplementary Appendix of our article, red-cell age is expressed as the maximum age of any unit, and the nomogram shows that the older the red cells, the worse the survival rate.
Frenzel and colleagues inquire about our ABO blood-group data. These distributions were reflective of the diverse geographic distribution of our patients. In general, patients received type-specific blood. Preoperative autologous donation was rarely used; of note, autologous blood undergoes similar storage-related changes. Finally, we have no information suggesting that ABO blood type, anticoagulants, and cell-saver use are confounders.
4 ReferencesColleen Gorman Koch, M.D., M.S.
Liang Li, Ph.D.
Eugene H. Blackstone, M.D.
Cleveland Clinic, Cleveland, OH 44195
kochc@ccf.org 1
Koch CG ,Khandwala F ,Li L ,Estafanous FG ,Loop FD ,Blackstone EH . Persistent effect of red cell transfusion on health-related quality of life after cardiac surgery. Ann Thorac Surg 2006;82:13-20
CrossRef | Web of Science | Medline2
Koch CG ,Li L ,Duncan AI , et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med 2006;34:1608-1616
CrossRef | Web of Science | Medline3
Koch CG ,Li L ,Van Wagoner DR ,Duncan AI ,Gillinov AM ,Blackstone EH . Red cell transfusion is associated with an increased risk for postoperative atrial fibrillation. Ann Thorac Surg 2006;82:1747-1756
CrossRef | Web of Science | Medline4
Koch CG ,Li L ,Duncan AI , et al. Transfusion in coronary artery bypass grafting is associated with reduced long-term survival. Ann Thorac Surg 2006;81:1650-1657
CrossRef | Web of Science | Medline
- Citing Articles (11)
Citing Articles
1
Nancy M. Heddle, Richard J. Cook, Donald M. Arnold, Mark A. Crowther, Theodore E. Warkentin, Kathryn E. Webert, Jack Hirsh, Rebecca L. Barty, Yang Liu, Connie Lester, John W. Eikelboom. (2012) The effect of blood storage duration on in-hospital mortality: a randomized controlled pilot feasibility trial. Transfusionno-no
CrossRef2
Julie Sanders, Sanjiv Patel, Jackie Cooper, Jennifer Berryman, Daniel Farrar, Michael Mythen, Hugh E. Montgomery. (2011) Red blood cell storage is associated with length of stay and renal complications after cardiac surgery. Transfusion 51:11, 2286-2294
CrossRef3
L. van de Watering. (2011) Red cell storage and prognosis. Vox Sanguinis 100:1, 36-45
CrossRef4
R. Henschler, M. M. Müller, H. U. Pfeiffer, E. Seifried, W. Sireis. (2010) Production of standard blood components. ISBT Science Series 5:n1, 190-195
CrossRef5
Gustaf Edgren, Mads Kamper-Jørgensen, Sandra Eloranta, Klaus Rostgaard, Brian Custer, Henrik Ullum, Edward L. Murphy, Michael P. Busch, Marie Reilly, Mads Melbye, Henrik Hjalgrim, And Olof Nyrén. (2010) Duration of red blood cell storage and survival of transfused patients (CME). Transfusion 50:6, 1185-1195
CrossRef6
John W. Eikelboom, Richard J. Cook, Yang Liu, Nancy M. Heddle. (2010) Duration of red cell storage before transfusion and in-hospital mortality. American Heart Journal 159:5, 737-743.e1
CrossRef7
Eleftherios C. Vamvakas. (2010) BLOOD COMPONENTS: Meta-analysis of clinical studies of the purported deleterious effects of “old” (versus “fresh”) red blood cells: are we at equipoise?. Transfusion 50:3, 600-610
CrossRef8
N. Ahrens. (2009) Transfusion-related immune reactions: pathogenesis and prevention. ISBT Science Series 4:n2, 230-235
CrossRef9
R. J. Benjamin. (2009) Evidence-based transfusion: young versus old blood as a case study. ISBT Science Series 4:n2, 323-328
CrossRef10
John G.T. Augoustides. (2009) The Year in Cardiothoracic and Vascular Anesthesia: Selected Highlights From 2008. Journal of Cardiothoracic and Vascular Anesthesia 23:1, 1-7
CrossRef11
Y. M. Bilgin, A. Brand. (2008) Transfusion-related immunomodulation: a second hit in an inflammatory cascade?. Vox Sanguinis 95:4, 261-271
CrossRef
