Blood Conservation Operations in Pediatric Cardiac Patients: A Paradigm Shift of Blood Use

Transcription

1 Blood Conservation Operations in Pediatric Cardiac Patients: A Paradigm Shift of Blood Use Mohsen Karimi, MD, Ivan Florentino-Pineda, MD, Ted Weatherred, MD, Ahsan Qadeer, MBBS, Carol Ann Rosenberg, CCP, Andrea Hudacko, CCP, and Duchwan Ryu, PhD Department of Surgery, Division of Pediatric Cardiothoracic Surgery, Division of Pediatric Anesthesiology, Department of Pediatric Perfusion, and Department of Biostatistics, Georgia Health Science University, Children s Medical Center, Augusta, Georgia Background. Red blood cell transfusion is associated with high morbidity in pediatric patients undergoing cardiac operations. The aim of this study was to evaluate the clinical effects and outcomes of blood conservation for our pediatric patients undergoing cardiac operations. Methods. We retrospectively analyzed a collected database of 168 pediatric patients who underwent biventricular (BV) and univentricular (UV) cardiac operations from 2006 to Patients were grouped into no blood conservation (n 86 [BV 74, UV 12]) and blood conservation (n 82 [BV 68, UV 14]) cohorts. There were no statistical differences in age, sex, weight, and preoperative or postoperative hemoglobin levels in the BV groups. Results. Even though the blood conservation group had longer cardiopulmonary bypass (CPB) (p < ) and cross-clamp times (p < 0.002) with lower hemoglobin levels (p < ), there was a decreased need for intraoperative (p < ) and postoperative blood transfusions (p < 0.018), lower inotropic scores (p < ), a decrease in ventilator days (p < ), and a shorter length of hospital stay (p < ). In the UV blood conservation group, there were no statistical differences in age, sex, weight, CPB and cross-clamp times, preoperative and postoperative hemoglobin levels, and red blood cell transfusions despite lower intraoperative hemoglobin levels (p < ) and blood transfusion (p < 0.01) requirements. There were significantly lower inotropic scores (p < 0.001) and a trend toward a shorter duration of time on the ventilator (p < 0.07) in the blood conservation group. Logistic regression analysis demonstrated a significant correlation between intraoperative blood transfusion and increased inotropic score, longer duration on the ventilator, and increased length of hospitalization. Conclusions. Blood conservation in pediatric cardiac operations is associated with fewer ventilator days, lower inotropic scores, and shorter lengths of stay. These findings, in addition to attendant risks and side effects of blood transfusion and the rising cost of safer blood products, justify blood conservation in pediatric cardiac operations. (Ann Thorac Surg 2013;95:962 7) 2013 by The Society of Thoracic Surgeons Cardiopulmonary bypass (CPB) is associated with defective hemostasis, which results in bleeding and a requirement for allogeneic blood product transfusions in many pediatric patients undergoing cardiac operations. Conservation of blood has become a priority during surgical procedures because of the shortages of donor blood, the risks associated with the use of allogeneic blood products, and the costs of these products. Further, transfusions expose patients to a variety of potential cellular and humoral antigens, pose risks of disease transmission and immunomodulation, and may alone represent proinflammatory stimuli in the perioperative period. Multidisciplinary approaches are required when considering strategies to limit blood transfusions. Strategies to reduce bleeding and transfusion requirements during surgical procedures include recognizing risk factors, developing transfusion protocols, conserving red Accepted for publication Sept 7, Address correspondence to Dr Karimi, 310 Cedar St, Boardman Bdg#204, New Haven, CT 06510; mohsenkarimimd@hotmail.com. blood cells (RBCs), using new alternatives to RBCs, altering inflammatory responses, and improving anticoagulation reversal [1]. Children undergoing cardiac operations are frequently exposed to blood products. The effect of blood conservation in reducing postoperative morbidity and mortality has not been well elucidated in pediatric patients undergoing cardiac operations. We hypothesize that decreased exposure of children to RBCs during cardiac operations would potentially reduce associated complications and improve postoperative outcomes. Patients and Methods Study Design Retrospective analysis of 168 pediatric patients who underwent open cardiac operations from 2006 to 2010 was performed to compare the outcomes of blood conservation operations with traditional operations. The first group underwent surgical repair between 2006 and by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 Ann Thorac Surg KARIMI ET AL 2013;95:962 7 BLOOD CONSERVATION IN PEDIATRIC CARDIAC PATIENTS 963 (historical control) by a different surgeon using conventional CPB without the availability and use of our current blood conservation equipment, techniques, and RBC transfusion protocol. The second group underwent surgical repair between 2008 and 2010 by the current surgeon, incorporating blood conservation equipment, techniques, and institutional policies and procedures for RBC transfusion triggers. The groups were further subcategorized into biventricular (BV) and univentricular (UV) physiologic types matched by the anatomic defect and surgical procedure. The type of procedure and the Risk Adjustment in Congenital Heart Surgery (RACHS) risk category [2, 3] are shown in Table 1. Patients were further subgrouped into no blood conservation (n 86 [BV 74, UV 12]) and blood conservation (n 82 [BV 68, UV 14]) cohorts. There were no changes in clinical personnel as far as anesthesia, perfusion, intensive care, or cardiology caregivers for both groups. A comprehensive database including demographic, intraoperative, and postoperative data was created for all patients in the cohort using electronic medical records. There were no patients who required extracorporeal membrane oxygenation before or after operation in the study groups. The institutional review board exempted patient consent and approved the study. Intraoperative variables included CPB and aortic crossclamp times (minutes). Preoperative hemoglobin (grams per deciliter) values were defined as the patient s hemoglobin measured by the blood gas analyzer before CPB. When the patient was placed on bypass, hemoglobin (grams per deciliter) was defined as the hemoglobin value immediately after initiation of CPB. The hemoglobin (grams per deciliter) value after bypass was defined as the hemoglobin value before the patient left the operating room suite. Intraoperative RBC transfusion (ml) was defined as the total amount of RBCs that the patient received from the time of arrival to the operating room until leaving the operating room, including the prime volume for the CPB circuit. Postoperative RBC transfusion was the total amount of the RBC transfusion given in the first 48 hours after the cardiac operation in the pediatric intensive care unit. All the patients received irradiated and leukocyte-depleted RBCs based on the institutional blood bank protocol. Inotropic scores, duration of mechanical ventilation, length of hospital stay, and postoperative complications were outcome variables evaluated in this study design. Peak inotropic scores were calculated for the first 48 hours based on the preestablished formula (dopamine [milrinone 10] [epinephrine 100]) measured in micrograms per kilogram per minute [4, 5]. Duration of mechanical ventilation was defined as number of days the patient was intubated in the pediatric intensive care unit until the day of successful extubation. The hospital length of stay was determined to be the total number of days that the patient was hospitalized from the day of operation to the time of discharge. The complications were based on the morbidities that were previously associated with RBC transfusion: infection, renal failure, end-organ injury, and so on. Table 1. Distribution of Biventricular and Univentricular Procedures and RACHS Risk Category RACHS Risk Category Cardiac Procedures (n 86) (n 82) Biventricular physiology Atrial septal defect Ventricular septal defect Atrioventricular canal Tetralogy of Fallot Double-outlet right 4 4 ventricle 2 Subaortic membrane Pulmonary valve 16 9 replacement/conduit Univentricular physiology 2 Bidirectional Glenn 6 8 procedure 3 Fontan operation 6 6 RACHS Risk Adjustment in Congenital Heart Surgery. A strategic protocol by the pediatric perfusion program at our institution was formulated and agreed upon by a multidisciplinary faculty within our institution to achieve a reduction in hemodilution and RBC transfusion. The formulation of the plan was divided into 2 areas equipment and technique. Equipment The Terumo Advanced Perfusion System 1 (Terumo Cardiovascular, Ann Arbor, MI) was modified and positioned close to the operating table to reduce tubing lengths and diameter. Five different arteriovenous loop circuits were customized specific to the weight of the patient. The Terumo FX05 (weight 15 kg) and FX15 (weight 16 kg) oxygenators (Terumo Cardiovascular) with integrated arterial filter were used for the CPB runs. The Terumo Capiox CP50 (Terumo Cardiovascular) was configured for the administration of cold cardioplegia and modified ultrafiltration (MUF). The Hemocor HPH 400TS (Minntech Corp, Minneapolis, MN) was used to remove excess fluid from the circuit. The Fresenius C.A.T.S Continuous AutoTransfusion System (Terumo Cardiovascular, Ann Arbor, MI) has allowed for successful return of shed blood during and after operation. Continuous arterial and venous blood gas monitoring (CDI 500; Terumo Cardiovascular) and cerebral and somatic saturation monitoring (Somanetics INVOS 5100 C system; Somanetics Corp, Troy, MI) provided additional hemodynamic information regarding adequacy of patient oxygenation and perfusion to tailor the need for blood transfusion. Use of point-of-care testing with i-stat (Abbott Point of Care, Inc, Princeton, NJ) and the Hemochron Signature Elite (ITC, Edison, NJ) allowed for microsampling of 0.5 ml of patient blood throughout operative and postoperative management. The differences in perfusion technique and equipment as well as

3 964 KARIMI ET AL Ann Thorac Surg BLOOD CONSERVATION IN PEDIATRIC CARDIAC PATIENTS 2013;95:962 7 differences in trigger points for blood transfusion for the 2 eras are depicted in Table 2. Perfusion Technique Our current practice allows the primary perfusionist to customize the patient circuitry with 5 available tubing packs and 2 oxygenators. These selections provided optimal circuit configuration based on patient size, cardiac anatomy, and corrective procedure and also decreased hemodilution while working safely within the US Food and Drug Administration s product specifications. The addition of a second pediatric perfusionist allowed the primary perfusionist to manage the physiologic changes of the patient on bypass while the second perfusionist facilitated the processing of shed blood to the autotransfuion equipment and blood gas management. In addition, incorporating vacuum-assisted drainage has made it possible to increase the height of the oxygenator to the level of the patient, decreasing the arterial and venous tubing length and significantly reducing hemodilution. Retrograde arterial priming has been instrumental in displacing the crystalloid priming volume of the circuit with the patient s own blood, reducing the hemodilutional effect at the initial stage of CPB. We also aggressively removed the added crystalloid volume to the circuit to maintain an even fluid balance throughout the operation. This was accomplished by hemoconcentrating and increasing the patient s urine output using mannitol. In addition, we performed venovenous or venoarterial MUF on patients weighing less than 20 kg at the conclusion of CPB. We routinely performed ultrafiltration on the remainder of the volume in the circuit and checked the hematocrit to ensure it was greater than or equal to the patient s most recent hematocrit value before reinfusion. This whole blood containing clotting factors and the cell-saving device processed blood were given to the patient before leaving the operating room. Statistical Analysis Standard descriptive statistics were used for patient demographic information. Values were calculated as mean standard deviation. Comparisons between the 2 groups were made using 2-sample t tests for continuous variables. The folded F test was used for equal variance, and if data did not support the different variances of 2 samples at less than the 5% significant level, pooled variance for the t test was used. We also assumed different variances for each sample and have used Satterthwaite s approximation to the degree of freedom of the t test if the same variance was suspected. For the counting data, the 2 test was used. Pearson and canonical correlation analyses were performed for the correlation analysis of the variables. Logistic regression analysis was used to determine the independent predictors influencing the outcome variables. Apvalue less than 0.05 was determined to be statistically significant. All analyses were performed by the Department of Biostatistics at our institution using SAS 9.3 software (SAS Institute, Cary, NC). Table 2. Perfusion Technique, Equipment, and Transfusion Comparison of the 2 Groups Perfusion Technique NC BC Circuit prime volume (ml) Integrated arterial filter with No Yes oxygenator Cerebral and somatic No Yes saturation Retrograde arterial priming No Yes Pediatric cell salvage No Yes Point of care blood No Yes microsampling MUF Yes Yes Serum lactate analysis No Yes Inline blood gas analyzer Yes Yes Mixed venous saturation Yes Yes Pole mounted vents No Yes Electronic charting No Yes Biventricular hematocrit 30% 21% during CPB Univentricular hematocrit 40% 25% during CPB Biventricular hematocrit 30% 21% after CPB Univentricular hematocrit after CPB 40% 35% BC blood conservation; CPB cardiopulmonary bypass; MUF modified ultrafiltration; NC no blood conservation. Results There were statistically no discernible differences in age, sex, weight, and preoperative and postoperative hemoglobin levels between the 2 groups for BV repair (Table 3). The blood conservation group had longer CPB (p ) and cross-clamp times (p 0.002), with lower hemoglobin levels (p ) during the surgical procedure. There was less need for intraoperative (p ) and postoperative RBC transfusions (p 0.018), and there were lower inotropic scores (p ), shorter ventilator days (p ), and shorter lengths of stay (p ) (Figs 1A, 2A). In the UV groups, there were no statistical differences in age, sex, weight, CPB and cross-clamp times, preoperative and postoperative hemoglobin levels, and RBC transfusions (Table 4), despite lower intraoperative hemoglobin levels (p ) and fewer blood transfusion (p 0.01) requirements in the blood conservation group (Fig 1B). However, there was a significantly lower inotropic score (p 0.001) and a trend toward a shorter duration of ventilation (p 0.07) in the blood conservation group (Fig 2B). In general, there was a higher number of complications associated with the no blood conservation group (n 26) than the blood conservation group (n 2) (Table 5). Pearson correlation analysis revealed a significant association between a higher incidence of intraoperative RBC transfusions and higher inotropic support (correlation coefficient 0.45; p ), longer ventilator days (correlation coefficient 0.33; p ), and longer lengths of stay (correlation coefficient 0.22; p 0.003). Canonical correlation analysis of the linear

4 Ann Thorac Surg KARIMI ET AL 2013;95:962 7 BLOOD CONSERVATION IN PEDIATRIC CARDIAC PATIENTS 965 Table 3. Perioperative Data for Patients With Biventricular Repair Perioperative Data (NC, n 74) (BC, n 68) p Value Age at operation (y) Weight at operation (kg) Male, n (%) Bypass time (min) Cross-clamp time (min) Preoperative Hgb On bypass Hgb Post-bypass Hgb Intraoperative RBC Postoperative RBC Inotropic score Ventilator duration (d) Hospital length of stay (d) BC blood conservation; Hgb hemoglobin; NC no blood conservation; RBC red blood cell. stay, and worse 0- to 6-year survival [9]. Reports of bloodless operations in children weighing less than 5 kg are rare and do not include trials or studies. Since a great number of pediatric cardiac operations are performed during the neonatal period, homologous blood components are essential and blood conservation is a main concern. Nevertheless, blood conservation in pediatric cardiac operations has to be considered within the context of a risk-benefit strategy and not as an ultimate endpoint. In this study, we investigated the effect of blood conservation strategy intraoperatively and postoperatively on surgical outcome as it relates to length of stay, ventilator days, inotropic scores, and complication rate. We have adapted novel techniques in CPB by miniaturizing the circuit and using factors such as mixed venous saturation, regional cerebral and somatic saturations, serum lactic acid levels, and inline continuous blood gas analyzer measurements to tailor our decision about RBC transfusion. The bypass circuit was primed with the patient s own blood by performing retrograde arterial priming once the aortic cannula was in place. We also performed aggressive hemofiltration during the bypass run and performed venovenous or venoarterial MUF after termination of CPB to remove excessive intravascular volume. Using MUF has been shown to remove vasoreactive substances and significantly decrease the need for RBC transfusion after CPB in pediatric cardiac combination of the outcome variables also demonstrated a significant correlation between more intraoperative blood transfusions and higher inotropic scores, longer duration of ventilation, and longer lengths of stay (correlation coefficient 0.47; p ). Multiple logistic regression analysis revealed no statistically significant differences in demographic variables (age, weight, sex, and RACHS categories) between the blood conservation and no blood conservation groups. Intraoperative RBC transfusion was a strong independent factor for longer ventilator days (p ), higher inotropic scores (p ), and longer lengths of stay (p 0.035). The surgeon, who performed and implemented blood conservation operations, was also a strong independent predictor in reduced ventilator days (p ), lower inotropic scores (p ), and shorter lengths of stay (p ). Comment Blood conservation has been an ongoing quest since pediatric cardiac surgery was in its very infancy. Recent decades have witnessed many improvements in all stages of cardiac operations, allowing for a move toward bloodless surgical procedures in adult patients [6] and even in pediatric patients in some specific fields [7]. Moreover, blood transfusion has been associated with an increased mortality rate [8], and it remains unclear if it improves survival. Low hemoglobin levels during CPB in adults have been shown to be an independent predictor of operative mortality, prolonged intensive care unit stay, postoperative hospital Fig 1. Intraoperative and postoperative red blood cell (RBC) transfusion for (A) biventricular and (B) univentricular cohorts in no blood conservation (NC) and blood conservation (BC) cohorts. There were significantly fewer intraoperative and postoperative RBC transfusions in BC group for biventricular repair and significantly fewer intraoperative RBC transfusions in BC group for univentricular palliation.

5 966 KARIMI ET AL Ann Thorac Surg BLOOD CONSERVATION IN PEDIATRIC CARDIAC PATIENTS 2013;95:962 7 Table 4. Perioperative Data for Patients With Univentricular Palliation Perioperative Data (NC, n 12) (BC, n 14) p Value Age at operation (y) Weight at operation (kg) Male, n (%) 10 (83.3) 7 (50) Bypass time (min) Cross-clamp time (min) Preoperative Hgb On bypass Hgb Post-bypass Hgb Intraoperative RBC Postoperative RBC Inotropic score Ventilator duration (d) Hospital length of stay (d) BC blood conservation; Hgb hemoglobin; NC no blood conservation; RBC red blood cell. operations [10 12]. We routinely autotransfused the patient throughout the operation as well as at the termination of the bypass procedure, and the salvaged blood was transfused to the patient before leaving the operating room suite. The benefit and safety of cell salvage has been well elucidated in the pediatric cardiac surgical population by reducing exposure to allogeneic blood transfusion [13]. In addition, the postoperative RBC transfusions were tailored based on institutional protocol and established trigger points for blood transfusion for a given repair and physiology. In general, we tolerated a hematocrit value as low as 21% for BV repair and 25% for UV palliation as long as the patient s hemodynamic measurements were within acceptable range. There were no patients in the blood conservation group who had adverse neurologic events by the change in RBC transfusion trigger point, as long as we maintained mixed venous saturation at greater than 60% and regional cerebral and somatic saturation at greater than the baseline level with no evidence of increasing lactic acidosis. There are mixed reports regarding the effect of a low hematocrit on adverse neurologic outcomes and psychomotor development, with other studies refuting the effect of low hematocrit values on neurodevelopmental or cognitive outcome in pediatric patients with cardiac conditions [14,15]. We continuously monitored regional cerebral saturation using near-infrared spectroscopy and maintained baseline saturation by increasing flow and cerebral vasodilation using ph-stat during cooling before giving the patient a blood transfusion. Although, there were no immediate neurologic insults with our bypass technique, the long-term neurologic sequelae need further investigation on longitudinal follow-up. There are 3 significant findings that can be extrapolated from this study. First, this study demonstrated that by using sophisticated equipment and miniaturizing the cardiopulmonary circuit, we can safely perform surgical repair and conserve RBC transfusion with no short-term harm or adverse events in our pediatric patients with cardiac conditions. This allowed us to decrease our RBC transfusion requirements during CPB and still regain our RBC count at the end of the operation close to the baseline starting Table 5. Number of Complications for the 2 Groups Complications (NC, n 86) (BC, n 82) Fig 2. (A) Significantly lower inotropic score, fewer ventilator days, and shorter length of stay (LOS) for blood conservation (BC) group in contrast to no blood conservation (NC) group for biventricular repair. (B) Significantly lower inotropic support for univentricular palliation in BC group and a trend toward fewer ventilator days and LOS in BC versus NC group. Pneumonia 4 0 Surgical wound infection 4 1 Stroke 4 0 Chylothorax 9 1 Deep venous thrombosis 1 0 Atrial flutter 1 0 Heart block 1 0 Death 2 0 Total 26 2 NC no blood conservation; BC blood conservation.

6 Ann Thorac Surg KARIMI ET AL 2013;95:962 7 BLOOD CONSERVATION IN PEDIATRIC CARDIAC PATIENTS 967 hemoglobin level. Second, we demonstrated that the need for inotropic support, ventilator days, and length of stay was significantly decreased in the blood conservation group despite lower intraoperative hemoglobin levels and longer bypass and cross-clamp times. These findings pertained more to the BV cohorts than to the UV groups, most probably because of the lower number of patients in the UV groups, which prevented reaching any statistical significance. This finding was somewhat surprising given the fact that longer bypass and cross-clamp times are in most part associated with an increased blood transfusion rate and related morbidity and mortality postoperatively. This may have an implication that the greatest benefit of blood conservation operations and less exposure to blood transfusions may have been potentially more significant if the blood conservation group had shorter pump runs and ischemic times. Finally, the complication rate in the blood conservation group was drastically less, although some of those adverse events potentially may not have been directly related to the amount of blood transfusion. Nonetheless, the strong correlation between RBC transfusion and associated risks has been well documented as well in other studies regarding the pediatric cardiac surgical population [16,17]. This study carries some of the known limitations of a retrospective study design. It precludes accurate assessment of practice pattern and trigger points for RBC transfusion in the no blood conservation group. There were also differences in surgeons as well as perfusion techniques and equipment that collectively could confound the data analysis. Also, because of the lack of electronic charting and the absence of specific intraoperative measurements (ie, cerebral and somatic saturation, serum lactic acid level, and so on.), we could not perform any statistical comparison for some variables between these 2 groups. Despite these limitations, the results are consistent with other published data in children describing the association of RBC transfusion with poor outcomes. The information presented here may serve as important preliminary data in planning future studies to evaluate the influence of further optimizing blood conservation operations on outcomes in pediatric cardiac surgical patients. This study has shown that blood conservation in pediatric cardiac operations was associated with a decrease in postoperative inotropic needs, days on ventilation, and length of stay. These findings were more pronounced in 2-ventricle repair than in single-ventricle palliative surgery. Miniaturization of the CPB circuit, contemporary techniques and equipment, and institutional commitments and protocols were paramount in establishing a successful pediatric cardiac surgical blood conservation program. Future improvements in perfusion technology and blood conservation protocols in association with additional prospective randomized trials will further capitalize our understanding of the benefit of blood conservation operations in pediatric patients with cardiac conditions. References 1. Ferraris VA, Ferraris SP, Saha SP, et al. The Society of Thoracic Surgeons Blood Conservation Guideline Task Force and Spiess BD, Shore-Lesserson L, Stafford-Smith, et al; The Society of Cardiovascular Anesthesiologists Special Task Force on Blood Transfusion: Perioperative Blood Transfusion and Blood Conservation in Cardiac Surgery: The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists Clinical Practice Guideline. Ann Thorac Surg 2007;83: Jenkins KJ, Gauvreau K, Newburger JW, Spray TL, Moller JH, Iezzoni LI. Consensus-based method for risk adjustment for surgery for congenital heart disease. J Thorac Cardiovasc Surg 2002;123: Jenkins KJ, Gauvreau K. Center-specific differences in mortality: preliminary analyses using the Risk Adjustment in Congenital Heart Surgery (RACHS-1) method. J Thorac Cardiovasc Surg 2002;124: Bradley SM, Simsic JM, McQuinn TC, Habib DM, Shirali GS, Atz AM. Hemodynamic status after the Norwood procedure: a comparison of right ventricle-to-pulmonary artery connection versus modified Blalock-Taussig shunt. Ann Thorac Surg 2004;78: Wernovsky G, Wypij D, Jonas RA, et al. Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants. A comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation 1995; 92: Ovrum E, Holen EA, Lindstein Ringdal MA. Elective coronary artery bypass surgery without homologous blood transfusion. Early results with an inexpensive blood conservation program. Scand J Thorac Cardiovasc Surg 1991;25: Gombotz H, Rigler B, Matzer C, Metzler H, Winkler G, Tscheliessnigg KH. 10 years experience with heart surgery in Jehovah s witnesses [article in German]. Anaesthesist 1989;38: Rao SV, Jollis JG, Harrington RA, et al. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004;292: 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: Journois D, Pouard P, Greeley WJ, Mauriat P, Vouhé P, Safran D. Hemofiltration during cardiopulmonary bypass in pediatric cardiac surgery. Anesthesiology 1994;81: Wang MJ, Chiu IS, Hsu CM, et al. Efficacy of ultrafiltration in removing inflammatory mediators during pediatric cardiac operations. Ann Thorac Surg 1996;61: Draaisma AM, Hazekamp MG, Frank M, Anes N, Schoof PA, Huysman PA. Modified ultrafilteration after cardiopulmonary bypass in pediatric cardiac surgery. Ann Thorc Surg 1997;64: Golab HD, Scohy TV, de Jong PL, Takkenberg JJM, Bogers JJC. Intraoperative cell salvage in infants undergoing elective cardiac surgery: a prospective trial. Eur J Cardiothorac Surg 2008;34: Jonas RA, Wypij D, Roth SJ, Bellinger DC, Visconti KJ, du Plessis AJ. The influence of hemodilution on outcome after hypothermic cardiopulmonary bypass: results of a randomized trial in infants. J Thorac Cardiovasc Surg 2003;126: Ando M, Takahashi Y, Suzuki N. Open heart surgery for small children without homologous blood transfusion by using remote pump head system. Ann Thorac Surg 2004;78: Salvin JW, Scheurer MA, Laussen PC, et al. Blood transfusion after pediatric cardiac surgery is associated with prolonged hospital stay. Ann Thorac Surg 2011;91: Székely A, Cserép Z, Sápi E, et al. Risks and predictors of blood transfusion in pediatric patients undergoing open heart operations. Ann Thorac Surg 2009;87: