Autotransfusion After Coronary Artery Bypass Grafting Halves the Number of Patients Needing Blood Transfusion

Transcription

1 Autotransfusion After Coronary Artery Bypass Grafting Halves the Number of Patients Needing Blood Transfusion Henrik Schmidt, MD, Poul Erik Mortensen, MD, Soren Lars Folsgaard, MD, and Esther A. Jensen, MD Departments of Anaesthesiology, Cardio-Thoracic Surgery, and Clinical Chemistry, Gentofte Hospital, University of Copenhagen, Hellerup, Denmark Background. Several randomized studies about autotransfusion of shed mediastinal blood in patients undergoing coronary artery bypass grafting have resulted in divergent findings concerning reduction of the need for homologous blood transfusions. Most of these studies used less strict criteria for homologous blood transfusion than applied in daily clinical practice. Methods. A prospective, randomized, controlled study involving 120 patients having elective, uncomplicated coronary artery bypass grafting was performed. The autotransfusion group received transfusion of shed mediastinal blood for 18 hours. Criteria for homologous blood transfusion were hemoglobin concentration less than 5.0 mmol/l in the intensive care unit and less than 5.5 mmol/l during the rest of the hospital stay. Results. Twenty-eight percent of patients in the autotransfusion group received homologous blood transfusion versus 55% in the control group (p = 0.007). Ninetyfive percent of the shed mediastinal blood was transfused. In the autotransfusion group, a total of 26 units of homologous blood was used versus 78 units in the control group (p < 0.001). Conclusions. Autotransfusion of shed mediastinal blood in patients undergoing elective, uncomplicated coronary artery bypass grafting halves the number of patients needing homologous blood and reduces the amount of homologous blood given. (Ann Thorac Surg 1996;61: ) he need for blood transfusion in patients undergoing T coronary artery bypass grafting has decreased during the past 20 years [1, 2]. Several ways of reducing the requirement for homologous blood in cardiac surgery have been actively pursued because of the concern about infections and immunologic reactions from homologous blood transfusions [3-7]. These methods include predonation of blood prior to operation [8], intraoperative withdrawal of blood before cardiopulmonary bypass (CPB) [9], and intraoperative plasma separation with a cell-saving device [10]. The first report on autotransfusion was by Blundell [11] in The first autotransfusion in a human was reported by Duncan [12] in 1886, and reinfusion of shed mediastinal blood was first described by Schaff and associates [13] in Randomized, prospective investigations of the benefit of autotransfusion of shed mediastinal blood have yielded conflicting results. Three studies [14-16] showed a decreased requirement of banked blood transfusion and advocated that the method be used routinely in cardiac operations. Only one [16] of these randomized studies reported a reduction in the Accepted for publication Dec 8, Presented at the Forty-third Annual Meeting of the Scandinavian Asso~ ciation for Thoracic Surgery, Aug 24-26, Address reprint requests to Dr Schmidt, Department of Anaesthesiology, Gentofte Hospital, Niels Andersensvej 65, DK-2900 Hellerup, Denmark. number of patients receiving blood transfusions, ie, from 92% in the control group to 75% in the autotransfusion group. In that study, there were no fixed criteria for blood transfusion, and the average blood requirement in the autotransfusion group was 2.7 units. In the two other randomized studies [14, 15], autotransfusion did not reduce the number of patients needing homologous blood transfusion. On the other hand, five different randomized studies [17-21] concluded that autotransfusion did not decrease the need for homologous blood transfusion. The aim of our study was to investigate whether autotransfusion of shed mediastinal blood could reduce the number of patients needing homologous blood transfusion and reduce the amount of transfused homologous blood if fixed transfusion criteria were used. The present study is part of a larger prospective, randomized, and controlled study concerning autotransfusion of shed mediastinal blood in patients undergoing coronary artery bypass grafting. At the same time, we studied the coagulation system, renal function, oxygen capacity of the blood, and coronary enzymes. These data will be published separately. Material and Methods Between November 1992 and October 1993, 120 adult patients undergoing primary elective coronary artery 1996 by The Society of Thoracic Surgeons /96/$15.00 Published by Elsevier Science Inc PII S (96)

2 1178 SCHMIDT ET AL Ann Thorac Surg AUTOTRANSFUSION IN CABG PATIENTS 1996;61: bypass grafting entered the study. Accepting a risk of 5% of overlooking a true difference (type! error) and a risk of 10% of accepting a false difference (type II error), this number of patients would allow detection of a difference between the autotransfusion group and the control group of at least 15% in the number of patients receiving homologous blood. The setup was a prospective, randomized, controlled cohort investigation. The study was approved by the regional ethics committee, and all patients gave informed written consent. Not included in the study were patients with a bleeding time of greater than 10 minutes by the method of Ivy, anticoagulation treatment, preoperative left ventricular ejection fraction of less than 0.40, diabetes, and pulmonary or kidney disease. Study Protocol Anesthesia was induced and maintained with fentanyl, midazolam hydrochloride, pancuronium bromide, and enflurane in oxygen. Hemodynamic variables were monitored with radial artery, central venous, and pulmonary artery catheters. In all patients, operation was performed through a median sternotomy using standard techniques for CPB with a crystalloid prime. Cardiopulmonary bypass was conducted using an arterial roller pump with a Bard cardiotomy/autotransfusion reservoir and a Bard membrane oxygenator (William Harvey HF-570). St. Thomas' cardioplegic solution (4 C) was used for myocardial protection, and CPB was performed at moderate hypothermia (32 to 36 C). Anticoagulation was established with heparin sodium, 300 IU/kg, and was monitored throughout the operation. After CPB, heparin was neutralized with protamine sulfate. The activated prothrombin time was lower than 40 seconds before the patient entered the study. At the conclusion of the procedure, blood remaining in the bypass circuit was collected in bags for later transfusion. By the end of the operation, the mediastinal and pleural tubes were attached to the inlet port of the Bard cardiotorny/autotransfusion reservoir. Prior to transfusion, the shed mediastinal blood was filtered through the 40-~m filter in the cardiotomy reservoir. After arrival in the intensive care unit, the patients were randomly allocated (120 closed envelopes) to autotransfusion of shed mediastinal blood (autotransfusion group) or to the control group. Excluded from the study after randomization were patients who underwent a revisional procedure for bleeding within the first 18 postoperative hours and patients needing inotropic drugs or an intraaortic balloon pump to maintain a cardiac index higher than 2 L. rain -1. m 2. All the residual blood from the CPB circuit was transfused. In the autotransfusion group, shed mediastinal blood from the cardiotomy reservoir was transfused every hour for the first 18 postoperative hours if more than 20 ml of blood had accumulated. In the control group, the cardiotomy reservoir was used for mediastinal drainage only. The criteria for postoperative volume replacement were as follows: 1. Cardiac index less than 2.2 L rain 1. m- 2 and pulmonary capillary wedge pressure less than 18 mm Hg or 2. Mean arterial pressure less than 60 mm Hg or 3. Urine output less than 1 ml kg 1. h-l Postoperative volume therapy was performed with crystalloid solutions. Homologous blood transfusion was given in accordance with the following criteria: 1. Hemoglobin concentration less than 5.0 mmol/l during stay in the intensive care unit 2. Hemoglobin concentration less than 5.5 mmol]l during stay in the surgical ward. The difference in hemoglobin concentration reflects the clinical practice in our hospital and the state of hemodilution of patients the first day after CPB. The blood used for transfusion was packed red blood cells stored in saline solution, adenine, glucose, and mannitol, 350 ml/u. Blood samples were drawn preoperatively and I hour, 6 hours, 18 hours, 2 days, 6 days, 1 month, and 3 months after operation and analyzed for hemoglobin concentration, hematocrit, platelet count, reticulocyte count, and levels of aspartate aminotransferase, lactate dehydrogenase, and creatine kinase MB activity. The physicians in the intensive care unit prescribed blood products in accordance with these criteria but were not involved in the study. The surgeons prescribing blood products during the stay in the surgical ward were unaware of whether the patients had received autotransfusion or not during the stay in the intensive care unit. Postoperative bleeding was recorded every hour for 18 hours in all patients. The number of transfusions of packed red blood cells and other blood products was recorded for 3 months after the operation. Infections, number of perioperative infarctions, and mortality within 3 months after the operation were noted. Statistical Methods The amount of bleeding and the number of blood transfusions are presented as the median value and the range. All other data are presented as the mean ± the standard deviation. Statistical analysis was performed using the Mann-Whitney test, Kruskal-Wallis test, and Fisher's exact test. The difference between the two groups in the percentage of patients exposed to homologous blood products was calculated and called the therapeutic gain; the 95% confidence interval was calculated. A p value (type I error) of less than 0.05 was considered significant. Results Of the 120 patients enrolled in the study, 109 completed it according to protocol. Eleven patients were excluded during the study; 7 in the autotransfusion group and 4 in the control group. There was no difference between the groups in the number of patients excluded. Five patients required reoperation for bleeding within the first 18

3 Ann Thorac Surg SCHM1DT ET AL ;61: AUTOTRANSFUSION IN CABG PATIENTS Table 1. Demographic Characteristics of the Two Patient s '~,b Autotransfusion Control Variable (n = 53) (n = 56) Age (y) Sex (male/female) 46/7 51/5 LVEF 0.62 ± Height (cm) 174 _ _+ 6 Weight (kg) 82.5 _ _ Where applicable, data are shown as the mean -+ the standard deviation. t, There were no significant differences between the groups. LVEF = left ventricular ejection fraction. Number of patients O unit "T unit 2 unit 3 unit 4 unil 5 unit hours, 3 patients had low cardiac output syndrome, and 3 patients in the autotransfusion group were excluded for technical reasons (air leak in the cardiotomy reservoir). No patient required the use of an intraaortic balloon pump. There were no differences between the two groups regarding patient age, weight, height, preoperative left ventricular ejection fraction, and sex (Table 1). Duration of CPB, aortic cross-clamp time, number of grafts per patient, operative blood loss, and perioperative weight gain were similar in the two groups (Table 2). None of the patients received homologous blood transfusion during the operation. Before the start of the study period, ie, at the time of randomization, there was no difference in hemoglobin concentration between the two groups. We found no difference in postoperative bleeding between the autotransfusion group (median value, 760 ml; range, 295 to 2,065 ml) and the control group (median value, 820 ml; range, 300 to 2,670 ml). Ninetyfive percent of the shed mediastinal blood (median value, 720 ml; range, 255 to 2,015 ml) was transfused during the 18 hours of autotransfusion. There was no difference in Table 2. Operative Characteristics of the Two Patient s ",t' Autotransfusion Control Variable (n - 53) (n 56) Duration of CPB 108 +_ (rain) Aortic cross-clamp ± 36 time (rain) No. of grafts 4.2 ± ± 1.5 Mean operative blood loss (ml)* ( ) ( ) Weight gain during ± 1.1 operation (kg) APTT before _+ 6.1 entering study ' Where applicable, data are shown as the mean -+ the standard deviation. b There were no significant differences between the groups c Range is given in parentheses. APTT = activated prothrombin time; CPB - cardiopulmonary bypass. I EAutotransfusion Control Fig 1. Number of homologous blood units used postoperatively. hemoglobin concentration at the end of autotransfusion (18 hours postoperatively) between the groups. Fifteen patients (28%) in the autotransfusion group received homologous blood transfusion compared with 31 patients (55%) in the control group (p = 0.007) (Fig 1). The therapeutic gain was 27% (95% confidence interval, 10% to 46%). The total amount of homologous blood used in the control group was about triple that in the autotransfusion group (78 versus 26 units). There was no significant difference between the groups in the number of patients receiving homologous blood during the stay in the surgical ward. One patient in both groups received fresh frozen plasma and platelet transfusion. There were no significant differences between the two groups in hemoglobin concentrations and hematocrit values during the 3-month follow-up, and all the hemoglobin concentrations had normalized I month postoperatively (Fig 2). Three patients died within the 3-month observation period. One died 3 days after operation of cerebral edema caused by an unrecognized cerebellar tumor. Another patient died 6 weeks postoperatively of an undetected adenocarcinoma, and 1 patient died of cardiac arrest 2 months after operation. There were no differences between groups in the number of acute myocardial infarctions (autotransfusion group, four; control group, five) or sternal infections (autotransfusion group, one; control group, three). Sepsis did not develop in any patient. Comment This is among the first prospective, randomized studies using clearly defined criteria for transfusion and volume replacement that demonstrate that postoperative autotransfusion of shed mediastinal blood after elective, uncomplicated primary coronary artery bypass grafting reduces the number of patients needing homologous blood. We reinfused 95% of the shed mediastinal blood,

4 1180 SCHMIDT ET AL Ann Thorac Surg AUTOTRANSFUSION IN CABG PATIENTS 1996;61: Hemoglobin mmol/l 12 lo! ~ Hematocrit fraction pre Oh lh 6h 18h 2 day 7 day 1 month 3 month I [~Autotransfusion: Control: HEMOGLOBIN ] Autotransfusion: ~ Control: HEMATOCRIT Fig 2. Hematologic data during 3-month follow-up. There were no significant differences between the groups. (Pre = preoperative; Oh before start of autotransfusion; lh = 1 hour postoperative before start of autotransfltsion; 6h, 18h - hours after start qf autotransfusion; 2 day, 7 day, 1 month, 3 month = postoperative time.) which resulted in a 50% reduction in the risk of exposure to homologous blood (from 55% of patients in the control group to 28% in the autotransfusion group). In two other studies [14, 15] showing that autotransfusion reduced homologous blood requirements, this did not lead to a reduction in the number of patients receiving blood transfusion. The risk of receiving homologous blood was 70% to 85% in both the autotransfusion and control groups [14, 15]. Schaff and associates [14] were able to transfuse only 50% of the shed mediastinal blood, and their criterion for transfusion was a hematocrit of less than 35% (hemoglobin concentration less than 7.3 mmol/ L). Eng and co-workers [15] transfused only 40% of the shed mediastinal blood and used the same transfusion criteria we did in this study. However, their use of homologous blood was much higher than ours, which may be explained by the lower amount of autotransfusion. Only one randomized study [16], involving 130 patients, has previously shown a reduction in the number of patients receiving homologous blood transfusion (from 95% in the control group to 75% in the autotransfusion group). No fixed criteria for transfusion and no information about hemoglobin values (or hematocrit values) before, during, and after autotransfusion were reported. The period of autotransfusion was 6 hours, and the amount of autotransfusion was 336 ml per patient (65% of the postoperative bleeding). The average homologous blood requirement per patient (2.7 units and 3.3 units in the control group) was much higher than in our study (0.5 unit and 1.4 units in the control group). Comparing our results with randomized studies [17-21] showing no reduction in the need of homologous blood transfusion, three main reasons could explain the differing results: the smaller number of patients in the F 0.4 studies [19, 20], higher or no criteria for transfusion [17, 18, 20, 21], and transfusion of smaller proportions of the shed mediastinal blood [17-21]. As there was no difference in postoperative hemoglobin or hematocrit levels during our study, it is evident that the transfusion practice was the same in both groups. We believe that because we could transfuse almost all the shed mediastinal blood over the first 18 postoperative hours and because the transfusion protocol was strictly respected, the patients receiving autotransfusion were at distinctly less risk for homologous blood transfusion. The autotransfusion of shed mediastinal blood with fixed criteria for blood transfusion reduced the average use of homologous blood by 1 unit per patient at a price of approximately $125 US (Danish price). The cost of the accessories to the Bard cardiotomy reservoir needed for autotransfusion is about $20 US. Like others [14, 16], we found no clinical or laboratory signs of coagulopathy. There were no septic reactions, and the number of sternal infections was similar in the two groups. The occurrence of perioperative acute myocardial infarctions was also similar. In conclusion, autotransfusion of shed mediastinal blood in patients undergoing elective, uncomplicated coronary artery bypass grafting halves the number of patients needing homologous blood and reduces the amount of homologous blood given. References 1. Roche JK, Stengle JM. Open-heart surgery and the demand for blood. JAMA 1973;225: Breyer RH, Engelman RM, Rousou JA, Lemeshow S. Blood conservation for myocardial revascularization. J Thorac Cardiovasc Surg 1987;93: Dahmen E, Malchus R, Hoppe I. Hepatitis after cardiosurgery: frequency and causes. Results of a prospective study with use of autologous blood. Thorac Cardiovasc Surg 1980; 28: Cohen ND, Munoz A, Reitz BA, et al. Transmission of retroviruses by transfusion of screened blood in patients undergoing cardiac surgery. N Engl J Med 1989;320: Kluge RM, Calla FM, McLaughlin JS, Hornick RB. Sources of contamination in open heart surgery. JAMA 1974;230: Peterman TA, Jaffa HW, Feorino PM, et al. Transfusionassociated acquired immunodeficiency syndrome in the United States. JAMA 1985;254: Goodnough LT, Johnston MFM, Ramsey G, et al. Guidelines for transfusion support in patients undergoing coronary artery bypass grafting. Ann Thorac Surg 1990;50: Britton LW, Eastlund DT, Dziuban SW, et al. Predonated autologous blood use in elective cardiac surgery. Ann Thorac Surg 1989;47: Cosgrove DM, Thurer RL, Lytle BW, Gill CG, Peter M, Loop FD. Blood conservation during myocardial revascularization. Ann Thorac Surg 1979;28: DelRossi AJ, Cernaianu AC, Vertrees RA, et al. Platelet-rich plasma reduces postoperative blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1990;100: Blundell J. Experiments on the transfusion of blood by the syringe. Lancet 1818;9: Duncan J. On reinfusion of blood in primary and other amputations. Br Med J 1886;1: Schaff HV, Hauer JM, Brawley RK. Autotransfusion in cardiac patients after operation. Surgery 1978;60: Schaff HV, Hauer JM, Bell WR, et al. Autotransfusion of shed

5 Ann Thorac Surg SCHMIDT ET AL ;61: AUTOTRANSFUSION IN CABG PATIENTS mediastinal blood after cardiac surgery: a prospective study. J Thorac Cardiovasc Surg 1978;75: Eng J, Kay PH, Murday AJ, et al. Postoperative autologous transfusion in cardiac surgery. A prospective, randomised study. Eur J Cardio-thorac Surg 1990;4: Lepore V, Radegran K. Autotransfusion of mediastinal blood in cardiac surgery. Scand J Thorac Cardiovasc Surg 1989;23: Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotransfusion following cardiac operations: a randomized, prospective study. Ann Thorac Surg 1979;27: Page R, Russell GN, Fox MA, Fabri BM, Lewis I, Williets T. Hard-shell cardiotomy reservoir for reinfusion of shed mediastinal blood. Ann Thorac Surg 1989;48: Adan A, Brutel de la Rivi6re A, Hass F, van Zalk A, de Nooij E. Autotransfusion of drained mediastinal blood after cardiac surgery: a reappraisal. Thorac Cardiovasc Surg 1988;36: Ward HB, Smith RR, Landis KP, Nemzek TG, Dalmasso AP, Swaim WR. Prospective, randomized trial of autotransfusion after routine cardiac operations. Ann Thorac Surg 1993;56: Bouboulis N, Kardara M, Kesteven PJ, Jayakrishnan AG. Autotransfusion after coronary artery bypass surgery: is there any benefit? J Card Surg 1994;9: REVIEW OF RECENT BOOKS Thoracic Transplantation Edited by Sara Shumway and Norman Shumway Cambridge, MA, Blackwell Scientific, I pp, illustrated, $ ISBN Reviewed by Lawrence H. Cohn, MD Norman Shumway is responsible for pioneering and sustaining cardiac transplantation. His original seminal work with Dr Richard Lower at Stanford Medical Center set in motion the longest and most successful clinical cardiac transplant program in the world. It is thus fitting that Dr Shumway co-edit with his cardiac surgeon daughter who, in her own right, is an excellent transplanter, a definitive book on thoracic organ transplantation. This book is remarkably thorough about all aspects of cardiac and pulmonary transplantation with a combination of medial, immunologic, surgical, and sociopolitical aspects of thoracic transplantation, which is totally complete in its scope. Most of the invited authors have been or are presently associated with the Stanford transplant program, testifying to the vast experiences of the writing faculty. There are also a number of other individuals from other prestigious transplant institutions such as Columbia, Duke, University of Minnesota, University of Pittsburgh, and Mayo Clinic. The book begins with one of the most humorous pieces of medical writing I have had the pleasure to read in a long time, the foreword by Richard Lower, Medical College of Virginia. Richard Lower was the first Shumway-trained resident and knows Norman Shumway better than anybody. He includes such quotes as "it was fun to work with Norm in the old days before he became a big shot" and several more colorful references to the true essence of the Shumway training era. Doctor Sara Shumway is the senior editor of the book, and she has invited a superb group of invited authors for the various chapters about the historical background, immunologic basis, organ preservation, donor procurement, pretransplantation recipient management, and operative techniques of heart, heartlung, and lung transplantations and important chapters on postoperative care (including one written by a United States Senator), the pathology of cardiac and pulmonary transplantation, and special considerations and results of single lobar and sequential lung transplantations. The final part discusses future prospects for transplantation, with a particularly strong case being made for the eventual use of xenografts once the immunologic basis of xenograft rejection is established. There is little to quarrel with in this extremely well organized book about thoracic transplantation. One might wonder about some of the newer technical modifications of cardiac and lung transplantation, which have not been encyclopedically set forth, but other than a few minor flourishes this really is a very easy to read, totally complete book on thoracic transplantation organized by a family of transplanters who go back to the beginnings of thoracic transplantation. Boston, Massachusetts 1996 by The Society of Thoracic Surgeons Ann Thorac Surg 1996;61: /96/$15.00 Published by Elsevier Science Inc PII S (96)