Age of transfused blood is not associated with increased postoperative adverse outcome after cardiac surgery

Transcription

1 British Journal of Anaesthesia 106 (5): (2011) Advance Access publication 16 March doi: /bja/aer029 CARDIOVASCULAR Age of transfused blood is not associated with increased postoperative adverse outcome after cardiac surgery M. McKenny 1 *,T.Ryan 1,H.Tate 4, B. Graham 2, V. K. Young 3 and N. Dowd 1 1 Department of Anaesthesia, 2 Blood Transfusion Department, and 3 Department of Cardiothoracic Surgery, St James s Hospital, James s Street, Dublin, Ireland 4 Helen Tate Statistics Ltd, Cambridge, UK * Corresponding author. mikemckenny@eircom.net Editor s key points Longer storage age of red blood cell transfusions may be associated with adverse outcome in cardiac surgery. An observational study of 1153 patients receiving 5962 units finds no association with storage age. Total number of units transfused had an association with outcome. There is a need for large prospective studies of this potential risk factor. Background. This study investigated the hypothesis that storage age of transfused red blood cells (RBCs) is associated with adverse outcome after cardiac surgery, and examined association between volume of RBC transfusions and outcome after cardiac surgery. Methods. Adult patients undergoing first time elective/urgent cardiac surgery who had received RBC transfusion perioperatively were included. Three prospective institutional databases were linked. Patients were grouped according to the oldest storage age of any RBCs transfused: those who received only RBCs stored for 14 days, only RBCs stored for.14 days, and a mixture of both ages of blood. The effect of RBC age on early mortality, postoperative ventilation 72 h, renal failure, pulmonary and infectious complications, length of intensive care stay, and postoperative ventilation time was examined using regression analyses with adjustment for confounding factors, including number of units transfused. Results. Data were analysed on 1153 patients who received a total of 5962 RBC units. There was no difference in adjusted odds of any outcome between the 14 days group and the group who received RBCs aged.14 days. Multivariate logistic regression analyses disclosed number of RBC units transfused as the most consistent factor associated with major postoperative complications, P, in all cases. A trend of increasing complication rate was observed with more units transfused. Conclusions. Storage age of RBC transfusion up to 35 days was not associated with increased postoperative adverse outcome after cardiac surgery. The number of RBC units transfused is consistently associated with adverse outcome. Keywords: blood transfusion; cardiac surgery; outcome; storage age; volume transfused Accepted for publication: 10 January 2011 The concept of universal benefit of red cell transfusion has been challenged in a number of clinical settings. 1 2 Blood transfusion has been associated with increased morbidity, particularly infection, and mortality in diverse patient groups. 3 6 In patients undergoing cardiac surgery, blood transfusion has been associated with an increased postoperative incidence of infection, prolonged ventilatory support, longer stay in the intensive care unit (ICU), new atrial fibrillation, new renal failure, cardiac complications, and neurological events Recent studies in trauma and cardiac surgery patients suggest that the association between blood transfusion and adverse outcomes relates to the storage age of transfused red blood cells (RBCs) Other studies have not found this correlation, so association between storage age of RBCs and outcome remains controversial Alternatively, it is suggested that the relationship reflects the volume of blood transfused rather than a specific age of blood effect, with a dose response relationship between the number of units transfused and major morbidity observed in a number of studies On the basis of the current data, the association between duration of storage of transfused RBCs and adverse outcome in adult patients undergoing cardiac surgery remains hypothetical. Limiting RBC transfusion to units stored for 14 or even 28 days would clearly impact greatly on blood transfusion practice and reduce availability of an already scarce resource. Cardiac surgery is an intrinsically good model for investigation of the relationship between transfusion and outcome. It is associated with significant transfusion requirements and well-established risk stratification models exist which facilitate adjustment between groups. Hard endpoints such as death and prolonged ventilation occur sufficiently often to facilitate statistical analysis. The aim of this study & The Author [2011]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please journals.permissions@oup.com

2 BJA McKenny et al. was to investigate the association between the storage age of transfused RBCs and major morbidity and mortality after cardiac surgery, and in addition to examine whether increasing numbers of RBC transfusions is associated with increased risk for morbidity and mortality in this group of patients. Methods The population investigated in this single-centre study included patients aged 18 yr or older undergoing first time elective/urgent coronary artery bypass grafting (CABG), cardiac valve surgery, or a combination of CABG and valve surgery, between January 2002 and December 2007, who had received red blood cell transfusion during the intraand postoperative period. Patients undergoing emergency or salvage surgery were excluded. Three institutional databases were accessed and linked to retrieve relevant patient characteristic, surgical, and postoperative variables. Electronic recordings of intra- and postoperative intensive care variables stored on Carevue TM database system were retrieved and matched with hospital transfusion records and a surgery database. The hospital s blood bank database was accessed to determine the number and age of RBC units, and quantity of other blood products transfused to each patient. The Patient Analysis and Tracking System (PATS, Dendrite Clinical Systems, UK), an international standardized database for adult cardiac surgical patients, was accessed to determine patient and surgical characteristics. PATS is managed by experienced personnel trained in database management and contains prospectively collected patient data. The Institutional Ethics Committee approved the use of these databases and waived the need for individual patient consent. Standard practice in the unit includes the administration of tranexamic acid to reduce intraoperative bleeding, and return of cardiotomy and pump blood to the patient. No strict protocol exists for perioperative blood transfusion in the unit, but it is limited where possible. When blood is ordered for transfusion, blood bank standard practice is to issue the oldest available RBC units in the first instance. Leucodepletion of all transfused blood used at this institution commenced before Blood is stored only until it is 35 days old; it is then discarded. The patient study groups were defined according to the oldest storage age of any RBCs transfused to a patient. We allocated the patients into three groups: those who received exclusively RBCs stored for 14 days or less ( 14 day group), those who received exclusively RBCs stored for more than 14 days (.14 day group), and those who received a mixture of RBCs stored for 14 days or less and greater than 14 days (Mixed group). The cut-off of 14 days was chosen because the putative storage lesion in blood becomes apparent after 2 weeks. A number of endpoints were examined; these included the binary endpoints of postoperative mortality (30 day mortality or death in hospital), prolonged postoperative ventilation ( 72 h), new renal failure requiring dialysis, pulmonary complications, and infectious complications. Pulmonary and infectious complications were defined as described in the PATS database system. A composite of these five endpoints was also analysed. The continuous endpoints of length of stay in intensive care and the actual postoperative ventilation time were also examined. Statistical analysis Patient and baseline characteristics and details of transfused blood were described using summary statistics for each of the three groups. The effect of the RBC age on the binary outcomes of postoperative mortality, pulmonary complications, new renal failure, infectious complications and prolonged ventilation, and on the composite outcome of one or more of these serious postoperative complications was examined in general using logistic regression analysis. Because the three patient groups differed in terms of their characteristics and prognostic status, three logistic regression analyses were undertaken for each binary endpoint; an analysis using RBC age group only, an analysis with RBC age group and the number of units transfused and EuroScore, and a further analysis including RBC age group, the number of units transfused, and all the patient characteristic/baseline characteristics (Full model). The effect of the age of transfused RBC on the continuous variables of length of stay in ICU and the hours of ventilation was investigated using linear regression analysis. A similar approach to that taken for the binary endpoints was used for the effects of possible confounding variables. Pairwise comparisons were made between the groups primarily using 95% confidence intervals for the effects derived. In addition, all endpoints were examined using the calculated mean age of all the RBC units transfused to each patient, using the same modelling structure but replacing the RBC age group with the mean age of transfused RBC. The analysis was undertaken using SAS version 9.1. Results Data on 1153 patients who received a total of 5962 units of RBCs intraoperatively or after operation were analysed. Baseline and operative characteristics for the patients in each group are shown in Table 1. Data on the Mixed group are presented in the tables for transparency, however, further discussion will be limited to the 14 days and.14 days groups. The two groups are similar regarding age, gender, Euro- SCORE, and degree of dyspnoea. No patient in the 14 days group had poor left ventricular function compared with 69 (7.6%) in the.14 days group. More patients in the.14 days group (14.4%) underwent combined CABG and valve procedures compared with the 14 days group (8.8%) (P not significant). The cardiopulmonary bypass time was similar between the two groups (110.2 vs min in 14 days and.14 days groups, respectively). The maximum storage age of the transfused units was 35 days, and median age was 28 days (Fig. 1). More patients received older blood than younger blood. 644

3 Association of age of transfused blood with adverse outcome BJA Table 1 Baseline and operative characteristics Variable Group 14 days (n568) Group >14 days (n5904) Group mixed (n5181) Age (yr) [mean (range)] 66.8 (46 81) 67.4 (25 89) 68.6 (31 86) Female [n (%)] 26 (38.2) 343 (37.9) 71 (39.2) EuroScore [mean (SD)] 4.38 (2.46) 4.79 (2.58) 5.78 (2.93) Poor LVF (ejection fraction,30%) [n (%)] 0 (0) 69 (7.6) 24 (13.3) Dyspnoea [n (%)] None 7 (10.3) 71 (7.8) 7 (3.9) Slight 26 (38.2) 346 (38.3) 65 (35.9) Marked 28 (41.2) 387 (42.8) 78 (43.1) At rest 7 (10.3) 100 (11.1) 31 (17.1) Operation type CABG only [n (%)] 47 (69.1) 603 (66.7) 111 (61.3) Valve only [n (%)] 15 (22.1) 171 (18.9) 34 (18.8) Combined CABG and valve [n (%)] 6 (8.8) 130 (14.4) 36 (19.9) Operation priority urgent 49 (72.1) 604 (66.8) 130 (71.8) Aortic cross-clamp time (min) [mean (SD)] 67.2 (27.7) 64.9 (27.5) 71.6 (32.7) CPB time (min) [mean (SD)] (32.8) (52.2) (47.5) Numbers of transfused units [mean (SD)] 2.4 (1.5) 4.6 (3.9) 9.0 (10.3) Age of transfused RBC days [mean (SD)] 10.8 (2.6) 25.4 (4.8) 18.3 (4.0) Blood products use [n (%)] Octoplas 15 (22.4) 241 (26.6) 95 (51.9) Platelets 17 (25.3) 329 (36) 110 (60) Numbers of transfused units mean (SEM) Octoplas 4.7 (0.9) 4.5 (0.2) 6.3 (0.4) Platelets 1.7 (5) 1.5 (1) 2.9 (2) 500 Frequency of age of RBCs transfused to all patients Number of units transfused Age of blood in days Fig 1 Storage age of all transfused blood. 645

4 BJA McKenny et al. Table 2 Primary endpoints in relation to duration of blood storage Endpoint Group 14 days (n568) Group >14 days (n5904) Group mixed (n5181) New renal failure [n (%)] 1 (1.5) 47 (5.2) 28 (15.5) Infectious complications [n (%)] 6 (8.8) 103 (11.4) 46 (25.4) Pulmonary complications [n (%)] 18 (26.5) 324 (35.8) 89 (49.2) Mortality [n (%)] 0 (0) 27 (3.0) 23 (12.7) Prolonged ventilation (.72 h) [n (%)] 0 (0) 29 (3.2) 22 (12.2) Composite of serious postoperative complications [n (%)] 20 (29.4) 368 (40.7) 105 (58.0) ICU time days [mean (SD)] 1.6 (1.6) 2.9 (5.0) 7.6 (14.1) Ventilation time (h) [mean (SD)] 9.4 (4.5) 17.9 (38.5) 62.6 (162.0) Table 3 Comparison between groups for all endpoints using the Full model. N/A, not available absence of any outcomes in the 14 days or less group for mortality and ventilation score meant detailed logistic regression analyses of these outcomes was not possible; *, P-value significant for the effect of age group in the model Endpoint Pairwise comparison Odds ratio 95% confidence interval P-value for the effect of age group in model New renal failure 14 days/.14 days , days/mixed , days/mixed , 1.29 Infectious complications 14 days/.14 days , days/mixed , days/mixed , 1.17 Pulmonary complications 14 days/.14 days , days/mixed , days/mixed , 1.38 Mortality 14 days/.14 daysn/a 14 days/mixedn/a days/mixed , 0.83* Prolonged ventilation 14 days/.14 daysn/a 14 days/mixedn/a days/mixed , 1.38 Composite 14 days/.14 days , days/mixed , days/mixed , 1.21 Difference in means 95% confidence interval P-value for the effect of age group in model ICU time days 14 days/.14 days , days/mixed , days/mixed , 0.12 Ventilation time (h) 14 days/.14 days , days/mixed , days/mixed , 26.48* Table 4 Number or units of transfused RBCs and outcomes Endpoint 1 or 2 units 3 or 4 units 5 or 6 units >7 units 432 patients 265 patients 207 patients 249 patients New renal failure [n (%)] 6 (1.4) 8 (3.0) 15 (7.3) 47 (18.9) Infectious complications [n (%)] 34 (7.9) 19 (7.2) 24 (11.6) 78 (31.3) Pulmonary complications [n (%)] 115 (26.6) 84 (31.7) 90 (43.5) 142 (57.0) Mortality [n (%)] 7 (1.6) 6 (2.27) 11 (5.29) 28 (11.2) Prolonged ventilation [n (%)] 0 (0) 5 (1.9) 7 (3.4) 39 (15.7) Composite [n (%)] 134 (31.0) 94 (35.5) 100 (48.3) 165 (66.3) 646

5 Association of age of transfused blood with adverse outcome BJA A total of 166 RBC units stored for 14 days were given to 68 patients, and 904 patients received a total of 4166 RBC units stored for.14 days. The mean (SD) age of units transfused in the 14 days group was 10.8 (2.6) days, and for the.14 days group was 25.4 (4.8) days. Patients in the.14 days group received more blood than those in the 14 days group. The mean (SD) number of units transfused to patients in the.14 days group was 4.61 (3.90), and for the 14 days group was 2.44 (1.46) (P¼0.0013). Blood product (Octoplas and platelets) use was similar between the groups, as was the mean number of product units transfused per group. Having chosen a cut-off of 14 days, the data set clearly precluded having equally sized groups. After examining the data available post hoc, a cut-off of 21 days produced reasonably sized groups for analyses and model fitting. All analyses were repeated using this 21 day cut-off point to examine the hypothesis that older RBCs were associated with higher complication rates; the findings were unchanged. In total, there were 52 postoperative deaths giving an overall mortality rate of 4.5%. Fifty-one patients (4.4%) required ventilation for.72 h. New postoperative renal failure requiring dialysis occurred in 76 patients (6.6%), infectious complications occurred in 155 patients (13%), and pulmonary complications in 431 patients (37%) (Table 2). The.14 day group had higher rates of mortality, prolonged postoperative ventilation, new renal failure requiring dialysis, and pulmonary and infectious complications compared with the,14 day group. The composite outcome of multiple postoperative complications was also higher in the.14 days group. The mean (SD) length of stay (in days) in ICU was 1.63 (1.63) for the 14 days group and 2.85 (5.0) for the.14 days group. The mean ventilation time for the,14 days group was 9.4 (4.5) h and for the.14 days group was 17.9 (38.4) h. Although all of the observed differences indicated that those who received blood stored for.14 days fared worse, examining the odds ratios for the unadjusted pairwise comparisons between the groups, there was no statistical difference in the odds of developing any outcome between the 14 day group and the.14 day group. Multiple logistic regression analyses were undertaken to examine the effect of RBC age group, taking into account the potential confounding factors. Of particular concern was the effect of the number of units transfused and preoperative risk factors including EuroSCORE, poor left ventricular function, and dyspnoea grade. The absence of any outcomes in the 14 days group for mortality and prolonged ventilation meant detailed logistic regression analyses of these outcomes was not possible. Where possible we endeavoured to fit multiple logistic regression models to further examine potential predictive factors; this showed that there was no difference in any outcome between the two groups. Including all covariates in the model (the Full model), the adjusted odds for the individual complications or for the composite outcome demonstrated no difference between the groups with all the baseline covariates in the model (Table 3). Linear regression analyses with all covariates included in the model showed no significant difference in ICU stay or time ventilated between the groups. The number of RBC units transfused and ejection fraction grade were predictors of prolonged ventilation time and prolonged ICU stay. Analysis of the complication rates for the binary outcomes, according to the number of units transfused, grouped into four categories showed that for all outcomes there is a trend of increasing complication rate with a larger number of units transfused (Table 4). Discussion This study found that the storage age of RBC transfusion up to 35 days was not associated with an increased risk of major postoperative complications or mortality after cardiac surgery. There was no difference in early postoperative mortality, prolonged postoperative ventilation, new onset renal failure, infectious or pulmonary complications, or a composite of these serious complications in patients who were transfused exclusively with RBCs stored for 14 days or less and patients who were transfused exclusively with RBCs stored.14 days, when potential confounding factors were taken into account. In addition, there was no difference in the duration of ventilation or length of ICU stay between these two groups. Increased storage age of transfused blood has been associated with increased morbidity and mortality in patients with sepsis and with trauma In cardiac surgery patients, transfusion of red cells stored for more than 2 weeks was associated with a significantly increased risk of postoperative complications and reduced survival compared with transfusion of red cells stored for 2 weeks or less. 4 However, like previous studies that used a range of definitions for old blood including mean unit age, age of the oldest unit transfused, or any transfusion over 30 days old, or blood stored for more than 18 days, we found no association between storage age and adverse postoperative outcome. Possible explanations of the putative deleterious effects of older blood include the storage lesion and alterations in red cell structure and function that occur during storage. The red cells are thought to change from biconcave discs to speculated, echinocytic red cells that are less pliable and have a tendency to aggregate and compromise microcirculatory perfusion. However, these alterations have been documented in whole blood rather than red cell concentrate, and recent data using laser technology have questioned storage-induced red cell deformation and the associated hypothesis of impaired oxygen delivery. 32 In addition, using gastric ph as a surrogate measure of tissue oxygenation, a deleterious effect of older blood (stored for.15 days) has been reported in one study 33 but was not reproduced in another. 34 The age-related 2,3-DPG reduction in transfused blood has also been implicated in its detrimental effect. While levels of 2,3-DPG are reduced, the red cell can 647

6 BJA McKenny et al. recover its normal 2,3-DPG level within 72 h of infusion and its subsequent function in vivo seems unaltered. 35 This may explain similarities in indices of global tissue oxygenation comparing patients who receive fresh blood with those who receive stored blood. Leucocytes contained in RBC infusions are implicated in the negative effects of older blood. 37 Leucodepletion is now performed on a universal basis in this country and this may further explain lack of association between age of transfused RBCs and postoperative outcome in our study. The evaluation of the independent role of the storage age of blood on postoperative outcome is not straightforward as many of the studies are small and have methodological limitations, or do not adjust for confounding factors including differences in baseline characteristics (ejection fraction or age). The study in cardiac surgery patients 4 did not appear to adjust for total transfusion volume. Although the distribution of the transfusion volume, as described by the mean, was similar between the study groups, it is possible that the total transfusion volume remained an important residual confounder. The number of RBC transfusions was the most consistent factor associated with major postoperative complications and mortality in this study. For all outcomes, there was a trend of increasing rate of complication with increasing number of units transfused, particularly 5 RBC units. This supports the idea that a dose response relationship exists between the number of units transfused and adverse postoperative outcomes The reason for this relationship is not evident from this study. It is possible that increased number of units transfused may identify more complex, sicker patients, or variations in patient response to the trauma of cardiac surgery. A larger study may help to answer this question. Our study defined the patient groups according to the oldest unit of blood that they received. This is logically consistent with the hypothesis that storage lesions, which develop in blood over time, are thought to cause adverse patient outcomes. Defining storage age of transfused blood by use of mean or median storage time makes the assumption that younger blood may somehow overcome any potential deleterious effect of older blood transfused to the same patient, but there is no known biological rationale for this assumption. The putative storage lesion in blood becomes apparent after 2 weeks and this was the rationale for using a cut-off of 14 days to discriminate the study groups. This time frame also facilitates comparison with other studies. However, this cut-off produced groups of unequal size for comparison, and the small number of patients in the 14 days group meant that comparisons between the study groups lacked power. The absence of any outcomes in the 14 days group for mortality and prolonged ventilation meant detailed logistic regression analyses of these outcomes was not possible. However, where possible, we endeavoured to fit multiple logistic regression models. In addition, a post hoc analysis determined that a cut-off of 21 days allowed reasonably sized groups for comparison and model fitting. When these groups were compared, we found no difference in any outcome between patients who received only blood stored for,21 days, and patients who received only blood stored for.21 days. The limitations of this study include its observational nature which did not allow equal sizes for the study groups and/or randomization of, and causality cannot be established. The regression analyses may not completely adjust for known confounding factors, and the single-centre design of this study may affect external validity. In conclusion, we found no correlation between duration of storage of RBCs and adverse outcome after cardiac surgery. The total number of RBC transfusions was the factor associated with risk of adverse events. For all outcomes, there is a trend of increasing complication rate with larger number of units transfused. Thus, there is no justification at present to change current allocation practices in cardiac surgery, but there is a definite need for prospective randomized trials on perioperative blood transfusion in cardiac surgical patients. Acknowledgements We thank Ms Mary O Connell and Ms Mairead Houlihan for their help coordinating databases. Conflict of interest None declared. References 1 Tinmouth A, Fergusson D, Chin Yee I, Hebert P. Clinical consequences of red cell storage in the critically ill. Transfusion 2006; 46: Marik P, Corwin H. Efficacy of red blood cell transfusion in the critically ill: a systematic review of the literature. Crit Care Med 2008; 36: Hebert P, Wells G, Blajchman M, et al. A multicentre, randomized, controlled clinical trial of transfusion requirements in critical care: Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999; 340: Koch CG, Li L, Sessler D, et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med 2008; 358: Vincent J, Baron J-F, Reinhart K, et al. Anaemia and blood transfusion in critically ill patients. J Am Med Assoc 2002; 288: Taylor R, Manganaro L, O Brien J, Trottier S, Parkar N, Vermakis C. Impact of allogenic packed red cell transfusion on nosocomial infection rates in the critically ill patient. Crit Care Med 2002; 30: Zacharias A, Habib R. Factors predisposing to median sternotomy complications: deep vs superficial infection. Chest 1996; 110: Leal-Noval S, Rincon-Ferrari M, Garcia-Curiel A, et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest 2001; 119: Banbury M, Brizzio M, Rajeswaran J, Lytle B, Blackstone E. Transfusion increases the risk of infection after cardiovascular surgery. J Am Coll Surg 2006; 202:

7 Association of age of transfused blood with adverse outcome BJA 10 Koch CG, Li L, Duncan A, 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: Koch CG, Li L, Duncan A, et al. Transfusion in coronary artery bypass grafting is associated with reduced long-term survival. Ann Thorac Surg 2006; 81: Kuduvalli M, Oo A, Newall N, et al. Effect of peri-operative red blood cell transfusion on 30-day and 1-year mortality following coronary artery bypass surgery. Eur J Cardiothorac Surg 2005; 27: Engoren M, Habib R, Zacharias A, Schwann T, Riordan C, Durham S. Effect of blood transfusion on long-term survival after cardiac surgery. Ann Thorac Surg 2002; 74: Surgenor S, Kramer R, Olmstead E, et al. The association of red blood cell transfusion and decreased long-term survival. Anesth Analg 2009; 108: Koch CG, Li L, Van Wagoner D, Duncan A, Gillinov M, Blackstone E. Red cell transfusion is associated with an increased risk for postoperative atrial fibrillation. Ann Thorac Surg 2006; 82: Zallen G, Offner P, Moore E, et al. Age of transfused blood is an independent risk factor for postinjury multiple organ failure. Am J Surg 1999; 178: Offner P, Moore E, Biffl W, Johnson J, Silliman C. Increased rate of infection associated with transfusion of old blood after severe injury. Arch Surg 2002; 137: Leal-Noval S, Jara-Lopez I, Garcia-Garmendia J, et al. Influence of erythrocyte concentrate storage time on postsurgical morbidity in cardiac surgery patients. Anesthesiology 2003; 98: Vincent JL, Sakr Y, Sprung C, Harboe S, Damas P. Are blood transfusions associated with greater mortality rates? Anesthesiology 2008; 108: Weightman W, Gibbs N, Sheminant M, Newman M, Grey D. Moderate exposure to allogenic blood products is not associated with reduced long-term survival after surgery for coronary artery disease. Anesthesiology 2009; 111: Vamvakas E, Carven J. Length of storage of transfused red cells and postoperative morbidity in patients undergoing coronary artery bypass graft surgery. Transfusion 2002; 40: Van der Watering L, Lorsinser 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: Claridge J, Sawyer R, Schulman A, McLemore E, Young J. Blood transfusions correlate with infections in trauma patients in a dose-dependent manner. Am J Surg 2002; 68: Sreeman G, Welsby I, Sharma A, Phillips-Bute B, Smith P, Slaughter T. Infectious complications after cardiac surgery: lack of association with fresh frozen plasma or platelet transfusions. J Cardiothorac Vasc Anesth 2005; 19: Spinella P, Perkins J, Grathwohl K, et al. The risks associated with fresh whole blood and RBC transfusions in a combat-support hospital. Crit Care Med 2007; 35: Purdy R, Tweeddale M, Merrick P. Association of mortality with age of blood transfused in septic ICU patients. Can J Anaesth 1997; 44: Weinberg J, McGwin G Jr, Marques M, et al. Transfusions in the less severely injured: does age of transfused blood affect outcomes? J Trauma 2008; 65: Keller M, Jean R, La Morte W, Millham F, Hirsch E. Effects of age of transfused blood on length of stay in trauma patients: a preliminary report. J Trauma 2002; 53: Vandromme M, McGwin G Jr, Weinberg J. Blood transfusion in the critically ill: does storage age matter? Scand J Trauma Resusc Emerg Med 2009; 17: Yap C-H, Lau L, Krishnaswamy M, Gaskell M, Yii M. Age of transfused red cells and early outcomes after cardiac surgery. Ann Thorac Surg 2008; 86: Raat NJ, Verhoeven AJ, Mik EG, et al. The effect of storage time of human red cells on intestinal microcirculatory oxygenation in a rat isovolemic exchange model. Crit Care Med 2005; 33: Solheim BG, Flesland O, Seghatchian J, Brosstad F. Clinical implications of red blood cell and platelet storage lesions: an overview. Transfus Apher Sci 2004; 31: Marik P, Sibbald W. Effect of stored-transfused blood transfusion on oxygen delivery in patients with sepsis. J Am Med Assoc 1993; 269: Walsh T, McArdle F, McLellan S, et al. Does the storage time of transfused red blood cells influence regional or global indexes of tissue oxygenation in anaemic critically ill patients? Crit Care Med 2004; 32: Weiskopf RB, Feiner J, Toy P. Anaemia-induced neurocognitive dysfunction: is oxygen the only player? Anesthesiology 2006; 106: Nielson HJ, Reimert CM, Pedersen AN, et al. Time-dependent, spontaneous release of white cell- and platelet-derived bioactive substances from stored human blood. Transfusion 1996; 36: Chelemer S. Association of bacterial infection and red blood cell transfusion after coronary artery bypass surgery. Ann Thorac Surg 2002; 73: Murphy G, Reeves B, Rogers C, Rizvi S, Culliford L, Angelini G. Increased mortality, postoperative morbidity and cost after red blood cell transfusion in patients having cardiac surgery. Circulation 2007; 116: