12 RANUCCI ET AL Ann Thorac Surg HEMODILUTION DURING CPB AND MORBIDITY RISK 2010;89:11 8 Abbreviations and Acronyms CI confidence interval CPB cardiop

Transcription

1 ORIGINAL ARTICLES: CARDIOTHORACIC ANESTHESIOLOGY: The Annals of Thoracic Surgery CME Program is located online at To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal. Hematocrit on Cardiopulmonary Bypass and Outcome After Coronary Surgery in Nontransfused Patients Marco Ranucci, MD, Daniela Conti, MD, Serenella Castelvecchio, MD, Lorenzo Menicanti, MD, Alessandro Frigiola, MD, Andrea Ballotta, MD, and Gabriele Pelissero, MD Departments of Cardiothoracic and Vascular Anesthesia and Intensive Care Unit and Cardiac Surgery, and Scientific Directorate, IRCCS Policlinico S. Donato, Milan, Italy Background. Preoperative anemia and the lowest registered hematocrit value on cardiopulmonary bypass are recognized risk factors for morbidity and mortality after coronary operations. A low hematocrit often results in blood transfusions with all of the associated possible complications. The relative contribution of these three factors to long-term outcome is still not well established. This study aimed to identify the role of preoperative anemia and hemodilution during cardiopulmonary bypass as determinants of morbidity and mortality after coronary operations. Methods. A consecutive series of 3,003 patients was analyzed. They had all undergone isolated coronary operations without receiving blood transfusions during their hospital stay. The preoperative hematocrit and the lowest hematocrit on cardiopulmonary bypass were analyzed in a multivariable model as predictors of major morbidity and operative mortality. Results. After adjustment for the other explanatory variables, both the preoperative hematocrit and the lowest hematocrit on cardiopulmonary bypass were found to be independent risk factors for major morbidity, but not for operative mortality. However, low values of preoperative hematocrit were not associated with an increased morbidity, provided that the lowest hematocrit on cardiopulmonary bypass was maintained above 28%. Median values of the lowest hematocrit on cardiopulmonary bypass below 25% were associated with an increased major morbidity rate. Conclusions. Excessive hemodilution during cardiopulmonary bypass is a risk factor for major morbidity even in the absence of blood transfusions. Techniques that aim to reduce the fall in hematocrit during cardiopulmonary bypass, including blood cardioplegia, may be useful, especially in patients with a low preoperative hematocrit. (Ann Thorac Surg 2010;89:11 8) 2010 by The Society of Thoracic Surgeons In recent years, many articles have identified the value of the lowest hematocrit (HCT) on cardiopulmonary bypass (CPB) to be a determinant of morbidity and mortality after cardiac operations [1 9]. Among the various postoperative complications related to this, acute renal failure seems to be the most prevalent [3 6, 8], but a higher incidence of stroke [3, 9], low cardiac output [3, 7], and other organ dysfunction [1 3] were also found to be associated with the degree of hemodilution during CPB. However, the lowest HCT on CPB is closely dependent on the preoperative HCT and the degree of intraoperative hemodilution. Preoperative anemia is an Accepted for publication July 30, Address correspondence to Dr Ranucci, Department of Anesthesia and Intensive Care, IRCCS Policlinico S. Donato, Via Morandi 30, San Donato Milanese, Milan, Italy; cardioanestesia@virgilio.it. independent risk factor for morbidity and mortality after cardiac operations, and specifically for coronary operations [10 14]. The consequence of low values of HCT before or during the operation is, of course, a higher risk of receiving allogeneic blood products during or after the operation. In fact, the preoperative HCT is the major determinant of transfusions in cardiac surgery [15 18]. Allogeneic blood transfusions are associated with an increased morbidity and mortality after cardiac operations [19 22]. Because of the correlation between the preoperative HCT, the lowest HCT on CPB, and the higher rate of transfusions, the specific role of each of the three factors in independently determining a poor outcome after cardiac operations is still unclear. The majority of the relevant studies did not account for the preoperative HCT as a determinant of the lowest HCT on CPB by The Society of Thoracic Surgeons /10/$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 12 RANUCCI ET AL Ann Thorac Surg HEMODILUTION DURING CPB AND MORBIDITY RISK 2010;89:11 8 Abbreviations and Acronyms CI confidence interval CPB cardiopulmonary bypass EuroSCORE European System for Cardiac Operative Risk Evaluation HCT hematocrit IQR interquartile range OR odds ratio STS The Society of Thoracic Surgeons TRACK Transfusion Risk and Clinical Knowledge Moreover, separating the role of anemia from the impact of allogeneic blood products is a difficult exercise, and although some authors applied sophisticated statistical techniques to account for this interaction [3, 7, 8, 14], clear conclusions are, as yet, unavailable. The present study aims first, to verify if the preoperative HCT and the lowest HCT on CPB are independent risk factors for morbidity and mortality after isolated coronary operations, and second, to assess the relative contribution of the two factors in determining morbidity and mortality. Patients and Methods Patient Population From our institutional database (2000 to 2008), we selected the patients who underwent an isolated coronary operation, retrieving a cohort of 5,809 patients. Subsequently, we selected a cohort of patients who did not receive any allogeneic blood products during their hospital stay, obtaining a final cohort of 3,003 patients (51.7% of the coronary patients). The local Ethics Committee approved this study and use of our institutional database and waived the need for patient consent. Clinical Management All the patients were operated on under moderate hypothermia (34 C) and alpha-stat ph management. Roller or centrifugal pumps were used, with standard or biocompatible (heparin or phosphorylcholine treated) circuits and hollow-fiber oxygenators; the CPB circuit was primed with colloid solutions at variable volumes, ranging from 900 ml to 1,200 ml. No retroprime was used. Pump flow was set between 2.0 and 2.8 L min -1 m -2, according to the core temperature. Cardiac arrest was achieved with antegrade cold crystalloid cardioplegia or blood cardioplegia for patients with an ejection fraction less than 0.3. Hemofiltration during CPB was used to correct low values of HCT whenever feasible. During the operations in patients treated with conventional open CPB circuits, pericardial blood was collected into a reservoir and reinfused into the patient without being processed; conversely, this blood was not readmitted to the circulation in the patients treated with closed circuits, unless previously processed by a cell-saving device. After the operation and during the intensive care unit stay, mediastinal blood collected into a reservoir was not reinfused to the patients. Table 1. Demographics, Clinical Profile, and Operative Details of Patient Population (n 3,003) p Values for Association With Factor Number (%) or Mean (SD) Major Morbidity Mortality Age (years) 64.2 (8.9) Male 2,563 (85.3) Body surface area (m 2 ) 1.86 (0.17) Preoperative hematocrit (%) 40.4 (4.1) Left ventricular ejection fraction 52 (10) Recent ( 90 days) myocardial infarction 656 (21.8) Unstable angina 255 (8.5) Previous vascular surgery 128 (4.3) Previous cardiac surgery 42 (1.4) Chronic obstructive pulmonary disease 211 (7.0) Previous cerebrovascular accident 110 (3.7) Serum creatinine value (mg/dl) 1.16 (0.75) Diabetes mellitus, on medication 476 (15.9) Urgent procedure 97 (3.2) EuroSCORE (logistic) 3.4 (4.2) TRACK score 5.6 (4.7) Cardiopulmonary bypass time (min) 58 (22) Lowest hematocrit on pump (%) 27.8 (3.5) Lowest temperature on pump (%) 32 (1.4) EuroSCORE European System for Cardiac Operative Risk Evaluation; TRACK Transfusion Risk and Clinical Knowledge.

3 Ann Thorac Surg RANUCCI ET AL 2010;89:11 8 HEMODILUTION DURING CPB AND MORBIDITY RISK Table 2. Outcome Data Outcome Variable Number and Incidence (%) Acute renal failure 19 (0.6) Stroke 11 (0.4) Mechanical ventilation (1.0) hours Surgical revision 13 (0.4) Mediastinitis 0 (0) Major morbidity 70 (2.3) Operative mortality 21 (0.7) Data Collection Two outcome variables were identified for the study: major morbidity, defined according to The Society of Thoracic Surgeons (STS) national database risk stratification subcommittee (one of the following: mechanical ventilation longer than 48 hours, surgical reoperation, mediastinitis, renal dysfunction or failure, stroke) and operative mortality (death in hospital or within 30 days of the operation after discharge). The renal morbidity endpoint was defined as acute postoperative renal insufficiency resulting in one or more of the following: an increase of serum creatinine to more than 2.0; 50% or greater increase in creatinine over baseline preoperative value; a new requirement for dialysis. Mortality after discharge was identified by a phone contact with the rehabilitation center accepting the patient, or with the patient or relatives when the patient was back home, after 30 days from the operation. The preoperative HCT (%) and the lowest HCT (%) on CPB were recorded and tested for any associations with the outcome variables. Other variables collected were: age (years), sex, weight, body surface area, left ventricular ejection fraction (%), recent (within 90 days before surgery) myocardial infarction, unstable angina, serum creatinine value (mg/dl), chronic obstructive pulmonary disease, diabetes mellitus (on medication), previous cerebrovascular accident, previous vascular surgery, previous cardiac surgery, urgent procedure, CPB duration (minutes), lowest temperature ( C) on pump, use of blood cardioplegia, and use of ultrafiltration during CPB. Patients were stratified for mortality risk according to the logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE) (23), and for transfusion risk according to the Transfusion Risk and Clinical Knowledge (TRACK) score (18). Statistics The univariate associations between the outcome variables and the preoperative and intraoperative risk factors were explored using a logistic regression analysis, producing an odds ratio (OR) and a 95% confidence interval (CI). Factors associated with the outcome variables (p 0.1) were entered into a multivariable logistic regression analysis (stepwise forward) to identify the independent risk factors for major morbidity and mortality. The Hosmer-Lemeshow goodness-of-fit test was used to assess the calibration properties of the models. Correlation between continuous variables were assessed using a linear regression analysis with r 2 correlation coefficient. For graphical purposes, the patient population was divided into deciles according to the values of preoperative HCT and lowest HCT on CPB; spline curves were used for the graphical representation of the relationship between the predictors and the outcome variables. To address the respective role of the preoperative HCT and the lowest HCT on CPB, a subgroup analysis was performed by dichotomizing the above variables according to the median value of the patient population. Data in tables and figures are presented as the number (%), mean (SD), or median (interquartile range [IQR] and absolute range). All tests were two-sided. A p value less than 0.05 was considered to be significant for all statistical tests. Statistical calculations were performed using a computerized statistical program (SPSS 11.0, SPSS, Chicago, IL). Results The demographics and preoperative risk profile of our patient population (n 3,003) are shown in Table 1. It was, broadly, a low-medium operative mortality risk population, with a mean EuroSCORE of 3.4 (95% CI: 3.2 to 3.5). The mean transfusion risk as assessed with the TRACK score was 35% (95% CI: 32% to 38%). The preoperative HCT had a mean value of 40.4%, with a median of 40% (IQR, 38% to 43%); the mean value of the 13 Table 3. Univariate Logistic Regression for Association of Outcome Data With Preoperative Hematocrit and Lowest Hematocrit on Cardiopulmonary Bypass Preoperative HCT (%) Lowest HCT on CPB (%) Outcome Variable OR (95% CI) p Value OR (95% CI) p Value Acute renal failure 0.92 ( ) ( ) Stroke 0.96 ( ) ( ) Mechanical ventilation 48 hours 0.92 ( ) ( ) Surgical revision 0.95 ( ) ( ) Mediastinitis Major morbidity 0.92 ( ) ( ) Operative mortality 0.92 ( ) ( ) CI confidence interval; CPB cardiopulmonary bypass; HCT hematocrit; OR odds ratio.

4 14 RANUCCI ET AL Ann Thorac Surg HEMODILUTION DURING CPB AND MORBIDITY RISK 2010;89:11 8 Fig 1. Major morbidity rate in the patient population according to the deciles of preoperative hematocrit (HCT) distribution, with cubic spline curves. lowest HCT on CPB was 27.8%, with a median of 28% (IQR, 25% to 30%). The postoperative outcome is reported in Table 2. There was a major morbidity rate of 2.3% and an operative mortality rate of 0.7%. Patients with major morbidity (70 cases) had a preoperative HCT of and a lowest HCT on CPB of ; patients with operative mortality (21 cases) had a preoperative HCT of and a lowest HCT on CPB of ; the other patients (2,933 cases) had a preoperative HCT of and a lowest HCT on CPB of At the univariate analysis (Table 3), the preoperative HCT was significantly associated with prolonged mechanical ventilation, major morbidity, and operative mortality; the lowest HCT on CPB was significantly associated with prolonged mechanical ventilation and major morbidity. The other perioperative risk factors were explored for univariate association with major morbidity and operative mortality (Table 1). Factors with a p value for association less than 0.1 were used for the subsequent multivariable analysis. The lowest HCT on CPB was Fig 2. Major morbidity rate in the patient population according to the deciles of lowest hematocrit (HCT) during cardiopulmonary bypass (CPB) distribution, with cubic spline curves. Table 4. Multivariable Logistic Regression Analysis for Major Morbidity and Operative Mortality Major Morbidity Model 1. Independent variable: preoperative HCT; dependent variable: major morbidity. Regression coefficient: 0.063; constant: 3.4; OR (95% CI): 0.94 (0.89 to 0.99); p Hosmer-Lemeshow goodness-of-fit test: ; degrees of freedom: 8; p Model 2. Independent variable: lowest HCT on CPB; dependent variable: major morbidity. Regression coefficient: 0.087; constant: 3.3; OR (95% CI): 0.92 (0.85 to 0.99); p Hosmer-Lemeshow goodness-of-fit test: ; degrees of freedom: 8; p Model 3. Independent variable: preoperative HCT; dependent variable: operative mortality. Regression coefficient: 0.036; constant: 4.4; OR (95% CI): 0.96 (0.87 to 1.07); p Model 4. Independent variable: lowest HCT on CPB; dependent variable: operative mortality. Regression coefficient: 0.048; constant: 4.4; OR (95% CI): 0.95 (0.84 to 1.08); p Models are adjusted for confounding factors listed in the text. CI confidence interval; CPB cardiopulmonary bypass; HCT hematocrit; OR odds ratio. significantly (r , p 0.001) correlated with the preoperative HCT. Being highly collinear, these two factors were not included in the same model. The graphical relationship between the two predictors (preoperative HCT and lowest HCT on CPB) and the major morbidity rate is reported in Figures 1 and 2, for each decile of distribution of the independent variables, with spline curves. Four separate multivariable risk models (stepwise forward logistic regression) were created for major morbidity and operative mortality including either the lowest HCT on CPB or the preoperative HCT as independent variables (Table 4). The two risk models for major morbidity (70 events) were adjusted for age, ejection fraction, serum creatinine value, and previous cerebrovascular accident because the number of events allowed up to seven independent variables to be included in the model without the risk of overfitting. Conversely, the two operative mortality risk models (21 events) allowed only two independent variables to be included. The model was therefore adjusted for the EuroSCORE, which includes four of the five factors identified in the univariate analysis. After adjustment for the other confounding factors, both the preoperative HCT and the lowest HCT on CPB remained as independent risk factors for major morbidity. In both models, the Hosmer-Lemeshow goodnessof-fit test did not detect significant differences between the expected and observed major morbidity rates. The multivariable analysis for operative mortality excluded the preoperative HCT and the lowest HCT on CPB as risk factors for operative mortality.

5 Ann Thorac Surg RANUCCI ET AL 2010;89:11 8 HEMODILUTION DURING CPB AND MORBIDITY RISK Fig 3. Major morbidity (gray bars) and operative mortality (black bars) rates in the four groups. (LHCT lowest hematocrit on cardiopulmonary bypass; PHCT preoperative hematocrit.) To address the respective effects of the preoperative HCT and the lowest HCT on CPB, four groups were created by dividing the patient population according to the median value. The resulting four groups were group I (preoperative HCT 40% and lowest HCT 28%), group II (preoperative HCT 40% and lowest HCT 28%), group III (preoperative HCT 40% and lowest HCT 28%), and group IV (preoperative HCT 40% and lowest HCT 28%). Because 112 patients were lacking the lowest HCT on CPB value, the total number of patients for this analysis is 2,891, not 3,003. Major morbidity and operative mortality rates are shown in Figure 3. There was no between-groups difference for operative mortality; major morbidity significantly (p 0.021) differed between groups, with group IV having a significantly higher major morbidity rate than group I (p 0.018). There was a significant difference (p 0.001) between groups for the preoperative HCT and the lowest HCT on CPB; the difference between the two (delta HCT) was significantly different for group II versus the others and group III versus the others (see Table 5). No differences between any of the groups were detected for priming volume, CPB duration, and use of ultrafiltration during CPB. Patients in groups I and II received blood cardioplegia at a significantly higher rate than patients in groups III and IV (group I, 92 patients, 10%; group II, 40 patients, 11%; group III, 26 patients, 5%; group IV, 67 patients, 7%; p between groups I II and III IV). Other preoperative and operative factors were investigated for association with the degree of intraoperative hemodilution (delta HCT). The delta HCT was negatively associated with the body surface area, male sex, and left ventricular ejection fraction. These factors were entered into a multivariable model (logistic regression) to identify whether the group classification based on preoperative HCT and lowest HCT on CPB was independent predictor of major morbidity. After adjustment for these confounders, patients belonging to group IV still had a significant higher rate of major morbidity than patients in group I (p 0.008) and group II (p 0.045). Comment The main findings of our study on isolated coronary operations in nontransfused patients are (1) both the preoperative HCT and the lowest HCT on CPB are 15 Table 5. Values of Preoperative Hematocrit and Lowest Hematocrit on Cardiopulmonary Bypass in the Four Groups Group Variable Mean (SD) Median (IQR) Range I(n 932) Preoperative HCT (%) 43.8 (2.3) 43 (3) Lowest HCT on CPB (%) 31.2 (1.8) 31 (2) Delta HCT (%) 12.6 (2.8) 12 (3) 4 25 II (n 385) Preoperative HCT 38.3 (2.6) 39 (2) Lowest HCT on CPB 30.3 (1.6) 30 (2) Delta HCT (%) 8.2 (2.3) 9 (3) 4 11 III (n 541) Preoperative HCT 42.3 (2.0) 43 (3) Lowest HCT on CPB 26.0 (1.9) 27 (3) Delta HCT (%) 16.9 (2.6) 17 (3) IV (n 1,033) Preoperative HCT 36.8 (2.9) 38 (4) Lowest HCT on CPB 24.9 (2.4) 25 (4) Delta HCT (%) 12.1 (3.1) 12 (4) Difference between groups Preoperative HCT p Lowest HCT on CPB p Delta HCT (%) p group II versus all the others p group III versus all the others CPB cardiopulmonary bypass; HCT hematocrit; IQR interquartile range.

6 16 RANUCCI ET AL Ann Thorac Surg HEMODILUTION DURING CPB AND MORBIDITY RISK 2010;89:11 8 independent risk factors for major morbidity; (2) neither the preoperative HCT nor the lowest HCT on CPB are independent risk factors for operative mortality, and (3) a preoperative HCT of 40% or less is not a risk factor for major morbidity if the lowest HCT on CPB is maintained at higher level than 28%. The combined action of a preoperative HCT of 40% or less and a lowest HCT of 28% or less produces a significant increase of major morbidity. Our results demonstrate that the deleterious effects of preoperative anemia and hemodilution during CPB are independent of the possible interaction with allogeneic blood products transfusions. This information is relevant, as no clear conclusion can be drawn to date from the existing literature. Some of the first authors [1, 2] who highlighted the association between the lowest HCT on CPB and morbidity and mortality failed to consider transfusions within their models. Others [4] included the number of packed red cells transfused within their logistic regression model, and concluded that, despite the fact that transfusions and lowest HCT on CPB were intercorrelated, the lowest HCT on CPB remained an independent risk factor for an increase in peak postoperative creatinine value. This is a sound statistical approach, but the nature of the outcome variable (peak postoperative creatinine value) does not allow to clearly identify whether a serious postoperative renal dysfunction was present. Karkouti and coworkers [6] analyzed the more relevant renal outcome data represented by acute renal failure (necessitating renal replacement therapy). They demonstrated an increased risk of acute renal failure when values of the lowest HCT on CPB were either lower than 21% or higher than approximately 25%. In their analysis, a lowest HCT on CPB below 21% (severe hemodilution) was an independent risk factor for acute renal failure, in a model adjusted for excessive perioperative blood loss (five or more units of blood transfused on the day of the operation). Unfortunately, they limited their analysis to the transfusions on the day of the operation (which was less than 10% of their patient population), whereas the acute renal failure could have occurred at any postoperative point in time. Therefore, transfusions after the first day may have had a confounding effect. A similar potential bias can be found in the study of Habib and associates [3], who only considered intraoperative transfusions in their risk model. So, as the investigators admitted, the role of transfusions was not fully elucidated in their study, and the use of a more liberal transfusion policy during CPB is cited as a possible measure to limit the deleterious effects of hemodilution. However, in another study from the same group [8], the investigators could demonstrate that transfusions are not the answer to hemodilution, and may even worsen the problem. Recently, we performed a study addressing the roles of the lowest HCT on CPB and the lowest oxygen delivery on CPB on the development of acute renal failure (5). In this study, we identified the lowest HCT on CPB value of 26% as an independent risk factor for acute renal failure. However, after adjustment for transfusions (at any point in time), the lowest HCT on CPB lost significance in a multivariable model. We believe that, despite statistical attempts to separate the consequences of a low HCT from those of transfusions, these two factors remain tightly correlated; moreover, it is difficult to analyze the time-related association between transfusions and the onset of the outcome event. Consequently, we decided to eliminate this complex interaction by only including nontransfused patients in our analysis. This approach allowed us to demonstrate that a low HCT before and during CPB is a risk factor per se for major morbidity. There are many hypotheses that could interpret this result. Fundamentally, they are related to the link between poor oxygen delivery and organ dysfunction. A low oxygen delivery rate may be dangerous for many organs, such as the brain, heart, liver, kidneys, and gut. The kidney seems to be particularly prone to ischemia: severe hemodilution may induce renal injury by augmenting the local renal inflammatory response subsequent to an ischemia-reperfusion phenomenon, particularly in the hypoxic environment of the renal medulla [4]. In-vitro studies [24, 25] have demonstrated that a slight fall in renal oxygen delivery may cause ischemic injury in susceptible areas of the kidney. This could be offset by an increased renal blood flow in the presence of severe hemodilution, leading to increased work and oxygen needs, and a decrease in plasma oncotic pressure, leading to fluid accumulation in the interstitial space, capillary closure and reduced oxygen delivery [6]. Similar mechanisms can be hypothesized for other complications, including ischemic cerebral events. Indeed, damage to the brain and kidney has been repeatedly associated with severe hemodilution during CPB, and our major morbidity index includes stroke and acute renal failure. The analysis of the respective roles of the preoperative HCT and the lowest HCT on CPB in determining a major morbidity offers further insights and clinical considerations. The only group that demonstrateds a significantly higher rate of major morbidity was group IV. The median value of the lowest HCT in this group was 25%. This value was previously identified as the cut-off for acute renal dysfunction [5, 6]. Patients in group III did not experience a significantly higher rate of major morbidity: in this group, the median value of lowest HCT on CPB was higher (27%) than in group IV. This finding leads to the conclusion that preoperative anemia may not cause major morbidity, provided that the lowest HCT on CPB was maintained at an acceptable value. Patients with a preoperative HCT of 40% or less had rates of major morbidity similar to those of patients with a preoperative HCT greater than 40% when the lowest HCT on CPB was maintained above 28%. This information highlights the concept of hemodilution during CPB: patients in group II could avoid severe hemodilution during CPB because the delta HCT was significantly lower than in the other groups (8.2% versus 12% to 16%). Therefore, preoperative anemia could not be a risk factor provided that adequate measures were taken to avoid

7 Ann Thorac Surg RANUCCI ET AL 2010;89:11 8 HEMODILUTION DURING CPB AND MORBIDITY RISK excessive hemodilution on CPB. These measures may include minimization of the priming volume of the circuit and oxygenator, a restrictive policy of fluid administration before and during CPB, the use of retroprime, the use of ultrafiltration during CPB, the use of blood cardioplegia, and the use of miniaturized, closed circuits. There are some limitations to the present study. The first is strictly related to the study design: including only nontransfused patients means selecting a population at low-medium mortality risk. Consequently, the morbidity and mortality rates are low. It is possible that both the preoperative HCT and the lowest HCT on CPB had no impact on mortality because of the limited number of events observed. A second limitation is that all our patients received a low priming volume (a policy that we have adhered to over the last 15 years). Therefore, extreme levels of hemodilution during CPB were rare events. It is possible that the inclusion of a higher number of these cases may result in different levels of significance, and may have a clear impact on mortality. There were very few women in the sample, so these results may not apply to female patients. Finally, owing to the retrospective nature of our data, it is impossible to identify all the factors that led to different values of delta HCT. Because the CPB technique was the same in all patients in terms of priming volume, it is likely that other factors may have played a role, namely, the fluid management before and during CPB, possibly due to the different working practices and expertise of the anesthesiologists and perfusionists. Moreover, and again owing to the retrospective nature of the study, we could find associations, but we cannot claim causations. In conclusion, the preoperative HCT is a morbidity risk factor in coronary operations only if it triggers a low value of HCT on CPB. The adoption of adequate measures to contain the intraoperative decrease of the HCT seems advisable, and it is important to comment that the use of blood cardioplegia produced a protective effect in our patient population. References 1. Fang WC, Helm RE, Krieger KH, et al. Impact of minimum hematocrit during cardiopulmonary bypass on mortality in patients undergoing coronary artery surgery. Circulation 1997;96(Suppl 2): DeFoe GR, Ross CS, Olmstead EM, et al. Lowest hematocrit on bypass and adverse outcomes associated with coronary artery bypass grafting. Ann Thorac Surg 2001;71: Habib RH, Zacharias A, Schwann TA, Riordan CJ, Durham SJ, Shah A. Adverse effects of low hematocrit during cardiopulmonary bypass in the adult: should current practice be changed? J Thorac Cardiovasc Surg 2003;125: Swaminathan M, Phillips-Bute BG, Conlon PJ, Smith PK, Newman MF, Stafford-Smith M. The association of lowest hematocrit during cardiopulmonary bypass with acute renal injury after coronary artery bypass surgery. Ann Thorac Surg 2003;76: Ranucci M, Romitti F, Isgrò G, et al. Oxygen delivery during cardiopulmonary bypass and acute renal failure following coronary operations. Ann Thorac Surg 2005;80: Karkouti K, Beattie WS, Wijeysundera DN, et al. Hemodilution during cardiopulmonary bypass is an independent risk factor for acute renal failure in adult cardiac surgery. J Thorac Cardiovasc Surg 2005;129: Ranucci M, Biagioli B, Scolletta S, et al. Lowest hematocrit on cardiopulmonary bypass impairs the outcome in coronary surgery: an Italian multicenter study from the National Cardioanesthesia Database. Tex Heart Inst J 2006;33: Habib RH, Zacharias A, Schwann TA, et al. Role of hemodilutional anemia and transfusion during cardiopulmonary bypass in renal injury after coronary revascularization: implications on operative outcome. Crit Care Med 2005;33: Karkouti K, Djaiani G, Borger MA, et al. Low hematocrit during cardiopulmonary bypass is associated with increased risk of perioperative stroke in cardiac surgery. Ann Thorac Surg 2005;80: Higgins TL, Estafanous FG, Loop FD, Beck GJ, Blum JM, Paranandi L. Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients. A clinical severity score. JAMA 1992;267: Bell ML, Grunwald GK, Baltz JH, et al. Does preoperative hemoglobin independently predict short-term outcomes after coronary artery bypass graft surgery? Ann Thorac Surg 2008;86: Kulier A, Levin J, Moser R, et al. Impact of preoperative anemia on outcome in patients undergoing coronary artery bypass graft surgery. Circulation 2007;116: Zindrou D, Taylor KM, Bagger JP. Preoperative haemoglobin concentration and mortality rate after coronary artery bypass surgery. Lancet 2002;359: Karkouti K, Wijeysundera DN, Beattie WS. Risk associated with preoperative anemia in cardiac surgery: a multicenter cohort study. Circulation 2008;117: Litmathe J, Boeken U, Feindt P, Garms E. Predictors of homologous blood transfusion for patients undergoing open heart surgery. Thorac Cardiovasc Surg 2003;51: Magovern JA, Sakert T, Benckart DH, et al. A model for predicting transfusion after coronary artery bypass grafting. Ann Thorac Surg 1996;61: Alghamdi AA, Davis A, Brister S, Corey P, Logan A. Development and validation of Transfusion Risk Understanding Scoring Tool (TRUST) to stratify cardiac surgery patients according to their blood transfusion needs. Transfusion 2006;46: Ranucci M, Castelvecchio S, Frigiola A, Scolletta S, Giomarelli P, Biagioli B. Predicting transfusions in cardiac surgery: the easier, the better: the Transfusion Risk and Clinical Knowledge score. Vox Sang 2009;96: Leal-Noval SR, Rincón-Ferrari MD, García-Curiel A, et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest 2001;119: 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: 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: Engoren MC, Habib RH, Zacharias A, Schwann TA, Riordan CJ, Durham SJ. Effect of blood transfusion on long-term survival after cardiac operation. Ann Thorac Surg 2002;74: Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J 2003;24: Ratcliffe PJ, Endre ZH, Tange JD, Ledingham JG. Ischaemic acute renal failure: why does it occur? Nephron 1989;52: Brezis M, Rosen S, Silva P, Epstein FH. Renal ischemia: a new perspective. Kidney Int 1984;26: