The effect of intraoperative autotransfusion during coronary artery bypass grafting was studied in a randomized double-blind trial involving 38 patients. Nineteen patients had the collected RBCs washed and autotransfused (autotransfusion group), while the remaining patients had their washed cells discarded (control group). Postoperative hemoglobin and hematocrit values were similar. Exposure to banked blood was markedly decreased in the autotransfusion group compared with the control group. In addition, the mean volume of banked packed RBCs transfused per patient was significantly less in the autotransfusion group compared with the control group. Platelet utilization also was markedly decreased in the autotransfusion group. Cryoprecipitate and fresh frozen plasma utilization also was less in the autotransfusion group than in the control group, but this did not reach statistical significance. We conclude that the intraoperative use of autotransfusion decreases the volume of homologous blood products transfused, which results in reduced exposure of the patients to banked blood products.
Cardiac surgical procedures are a major source of the total amount of blood used for transfusion purposes in the United States.[1-5] In cardiac surgery, there is blood loss owing to the involvement of major vascular structures in the operation itself as well as to the disturbances in hemostasis which accompany the use of extracorporeal circulation techniques.[6-10] Loss and destruction of RBCs, platelets, and coagulation factors during cardiopulmonary bypass result in the increased need for blood and blood products with these procedures. Recently, there is increased interest among surgeons to minimize the exposure of patients to blood and blood products due to the multiple dangers associated with homologous blood transfusions. These dangers mandate that alternative strategies be utilized to decrease the need for homologous blood transfusions.
One method for minimizing homologous blood usage is the intraoperative salvage of shed blood from the operative field and, after processing, reinfusion of the RBCs. Several studies have demonstrated that the use of this technique is safe and can be effective in reducing homologous transfusion during open-heart surgery.[1,4] However, no study to date has evaluated intraoperative autotransfusion in a randomized, double-blind controlled study. In an attempt to eliminate the biases inherent in a nonblinded study, we evaluated the use of intraoperative autologous shed blood collection and reinfusion in a double-blind randomized trial.
MATERIALS AND METHODS
Fifty patients undergoing isolated primary coronary revascularization between July and December 1989 gave written informed consent to participate in this study, which was approved by the hospital's Institutional Review Board. These patients were randomized to have their collected intraoperative blood either washed and reinfused (autotransfusion group) or discarded (control group).
Patients were randomized by coded instruction packets which specified the processing and administration of the patient's salvaged intraoperative blood. Sealed instruction packets were randomized using a shuffle deck procedure, serially numbered, and assigned sequentially to patients in order of enrollment. The sealed instruction packets were sent with the patients to the operating room.
All operations were performed by three surgeons utilizing similar surgical technique, including cardiopulmonary bypass with moderate systemic hypothermia (28 [degrees] to 32 [degrees] C) and a membrane oxygenator. Systemic heparinization (3 mg/kg, Lyphomed, Deerfield, Ill) was achieved and the activated clotting time was maintained greater than 450 s. The priming solution for the cardiopulmonary bypass circuit consisted of 1,000 ml lactated Ringer's solution and 1,000 ml 5 percent human albumin.
Bypass was initiated at normothermic flow rates of 2.2 L/min/[m.sup.2] and decreased to 1.6 L/min/[m.sup.2] for core temperatures below 30 [degrees] C. An initial dose of 10 ml/kg cold cardioplegia was administered to achieve diastolic cardiac arrest and was supplemented with additional cardioplegia as needed to maintain myocardial septal temperature 15 [degrees] C or lower. The internal mammary artery was harvested and used as a bypass graft in all patients. The saphenous vein was used for the other bypass grafts. All distal anastomoses were performed during a single aortic cross-clamp period. Blood was scavenged from the surgical field in all cases using an autologous blood salvaging system (Cell Saver 4, Haemonetics Corp., Braintree, Mass).
After weaning from cardiopulmonary bypass, the heparin was reversed with protamine sulfate (Eli Lilly, Indianapolis) in dosages sufficient to return the activated clotting time to baseline. Satisfactory hemostasis was achieved with electrocoagulation and, when needed, topical application of thrombin (Park-Davis, Liditz, Pa) and Gelfoam (UpJohn, Washington, D.C.). Mediastinal and pleural drainage tubes were placed prior to sternal closure and connected to a hard-shell cardiotomy reservoir.
At the conclusion of the operative procedure, the assigned instruction packet was opened by the perfusionist and anesthesiologist to reveal the randomization group. If the patient was randomized to receive the blood, the collected shed blood was washed and reinfused. Otherwise, the scavenged blood was discarded. In cases in which the patient was assigned to the autotransfusion group, the blood in the cardiopulmonary bypass circuit was washed and reinfused. In the control group, the pump blood was reinfused without being washed. The protocol is summarized in Table 1. The surgeon was uniformed as to the handling of the shed blood and pump blood and remained blinded to the group assignment until the completion of the study.
Decisions regarding postoperative transfusion of RBCs and blood components were made by the anesthesiologist and surgeon based strictly on the following protocol: (1) Packed RBC transfusions were given if the patient's hemoglobin was less than 7 g/dl or if the patient was hemodynamically unstable due to volume loss. (2) Fresh-frozen plasma transfusions were given if the patients were bleeding and the prothrombin time was greater than 15 s. (3) Cryoprecipitate was given to patients who were bleeding and had serum fibrinogen levels less than 150 mg/dl. (4) Patients received platelet transfusions if their platelet counts were below 90,000/[mm.sup.3] and they were bleeding, or if their platelet counts were below 50,000/[mm.sup.3]. Amicar (70 mg/kg; maximum, 5 g; Winthrope-Breon Laboratories, New York) and vasopressin (0.3 [micro]g/kg given intravenously, maximum, 20 [micro]g; Rhone-Poulene Rorer Pharmaceuticals, Ft. Washington, Pa) were given only when deemed necessary and agreed upon by the anesthesiologist and surgeon.
All patients received routine postoperative care as per the patient management protocols of the surgical ICU. Patients were weaned from mechanical ventilation and extubated within 24 h postoperatively. After extubation, patients were transferred to the ward for convalescence until ready for discharge.
Variables with dichotomous outcomes were analyzed by means of Fisher's exact test. Due to the nonnormality in the continuous variables, the Wilcoxon rank sum test was used to downweight the outlying values in comparing the study and control groups. Values are expressed as mean [+ or -] SEM, with probability values 0.05 or less considered significant.
Of 50 patients initially enrolled in the study, 38 completed the study per protocol and formed the basis of this report. Twelve patients were excluded because of protocol violations: six did not have their intraoperative blood processed per protocol and six did not have sufficient data recorded for analysis. Table 2 lists the demographic description of the patient groups. The mean age of the patients in the study was 65 years, with a range from 42 to 79 years. Both groups were similar with respect to age, gender, history of smoking, preoperative use of aspirin, coumadin, and persantine and incidence of diabetes and hypertension. The preoperative hematology and coagulation test profiles of the two groups are summarized in Table 3. The two groups had similar preoperative hemoglobin and hematocrit values, as well as partial thromboplastin times and platelet counts. The prothrombin time was significantly longer in the autotransfusion group compared with that in the control group. However, both values were well within the normal range. In addition, no patient had an abnormal prothrombin time in the study. Preoperative activated clotting times were also similar between the two groups.
[TABULAR DATA OMITTED]
Intraoperative parameters are shown in Table 4. Aortic cross-clamp time, cardiopulmonary bypass time, internal mammary artery usuage, and number of bypasses per patient were similar between the two groups. Heparin dosages were also similar in both groups. Mean intraoperative hemoglobin values during the operative procedures were similar. Total postoperative blood loss was not significantly different between the groups.
[TABULAR DATA OMITTED]
Intraoperative and postoperative utilization of banked blood products is summarized in Table 5. Patients in the autotransfusion group received a mean of 0.7 units (192.1 [+ or -] 79.6 ml) of packed RBCs in comparison with the control group which received, ie, a mean of 1.4 units (389.5 [+ or -] 100.1 ml) per patient (p [less than] 0.05). Platelet utilization was also higher in the control group, with the autotransfusion group receiving a mean of 0.6 units (31.6 [+ or -] 21.7 ml) and patients in the control group receiving a mean of 5.2 units (256.6 [+ or -] 77.2 ml [p [less than] 0.05]). The use of fresh-frozen plasma was not significantly different between the two groups. The use of cryoprecipate did not reach statistical significance between the two groups; however, no patient in the autotransfusion group received cryoprecipitate. Total blood products administered for the autotransfusion group was a mean of 371.6 [+ or -] 135.5 ml per patient in the autotransfusion group compared with 793.4 [+ or -] 181.6 ml for the control group (p [less than] 0.05).
The overall risk of a patient in this study being exposed to any banked blood products is shown in Table 6. In the autotransfusion group only 5 of the 19 patients received transfusions of packed RBCs compared with 13 of the 19 patients in the control group (p [less than] 0.05). Platelet utilization was higher in the control group, with 9 of the 19 patients receiving platelet transfusions compared with 2 of the 19 patients in the autotransfusion group (p [less than] 0.05). The number of patients in each group receiving fresh-frozen plasma was the same, and two patients in the control group receiving cryoprecipitate. In considering the overall exposure to banked blood products, 37 percent of the patients in the autotransfusion group were exposed to one or more transfusions of banked blood products, compared with 84 percent of the patients in the control group (p [less than] 0.05).
[TABULAR DATA OMITTED]
Table 7 lists the hematologic and coagulation profiles of the patients for the immediate postoperative period and on postoperative day 5. In comparing the two groups, there are no significant differences throughout the postoperative course.
Recently, public attention focused on the risks of blood transfusion have heightened public interest in reducing transfusion requirements during all surgical procedures. Cardiac surgical procedures have traditionally required blood transfusions, making blood conservation even more important.[1-4] Techniques of blood conservation used during open-heart surgery have been shown to reduce blood and blood product transfusion requirements.[11-19] While in some of these studies, the patients were randomly assigned to treatment modalities, in none of these studies was the surgeon blinded to the treatment group. The heightened awareness to blood loss which comes from employing blood conservation techniques could influence the individual surgeon's attention to meticulous hemostasis. In our study, we attempted to eliminate this bias by blinding the surgeon to the treatment group.
[TABULAR DATA OMITTED]
Intraoperative autotransfusion of washed scavenged RBCs is an attractive technique because it allows a high proportion of the shed RBCs to be reinfused without the hemodilution or activation of clotting factors seen with scavenged blood returned through the cardiotomy suction.[5,8,9] We were able to confirm the efficacy of washed shed blood reinfusion, since the transfusion requirements were significantly less for this group despite similar preoperative and postoperative hematocrit levels. One theoretical disadvantage of using reinfusion of washed cells is that it removes the plasma and its coagulation factors. We were not able, however, to demonstrate any deleterious effect on the coagulation system by utilizing the reinfusion of washed cells, nor were we able to demonstrate any increase in postoperative bleeding.
The total exposure of patients to banked blood products was significantly less for the autotransfusion group. The autotransfusion group patients required only 13 units of packed RBCs and 12 units of platelets, compared with 27 units of packed RBCs and 99 units of platelets for the control group patients. The increased use of platelets in the control group may represent a more aggressive use of blood components in the treatment of low hemoglobin or bleeding in these patients during the first 24 h after surgery.
The cost of adding this technology to the operation was $157.00 per case for the disposable supplies. The average cost for banked blood products in the control group was $500.72 compared with $132.02 in the autotransfusion group, a savings of $368.70 per patient (based on American Red Cross charges, October 1992). This cost analysis tends to underestimate the value of this technique as any complication of blood transfusion can cause tremendous morbidity and mortality, and avoidable transfusions should be eliminated. We were unable to demonstrate any increase in complexity of the operative procedure. Management of the autotransfusion device was accomplished by the perfusionist and required no additional personnel during the operative procedure. In addition, we were unable to demonstrate any detrimental effects of the technique on postoperative hematologic or coagulation profiles.
We conclude that reinfusion of washed, scavenged blood reduces blood transfusion requirements during cardiac surgery. This finding supports the recommendation that this technique be used routinely during cardiac procedures.
 Scott WJ, Rode R, Castlemain B, Kessler R, Follis F, Pett SB, et al. Efficacy, complications, and cost of a comprehensive blood conservation program for cardiac operations. J Thorac Cardiovasc Surg 1992; 5:1001-07
 Cosgrove DM, Thurer RL, Lytle BW, Gill CG, Peter M, Loop FD. Blood conservation during myocardial revascularization. Ann Thorac Surg 1979; 28:184-90
 Schaff HV, Hauer J, Gardener TJ, Donahoo JS, Walkins L Jr, Gott VL, et al. Routine use of autotransfusion following cardiac surgery: experience in 700 patients. Ann Thorac Surg 1979; 27:493-99
 Ovrum E, Einfrid AH, Abdelnoor M, Oystese R. Conventional blood conservation techniques in 500 consecutive coronary artery bypass operations. Ann Thorac Surg 1991; 52:500-05
 Keeling MM, Gray LA Jr, Brink MA, Hillerich VK, Bland KI. Intraoperative autotransfusion: experience in 725 consecutive cases. Ann Surg 1983; 197:536-41
 Tector AJ, Gabriel RP, Mateicka WE, Oparah SS, Flemma RJ, Lepley D. Reduction of blood usage on open heart surgery. Chest 1976; 70:454-57
 Giordana GF, Goldman DS, Mammana RB, Marco JD, Nestor ND, Raczkowski AR, et al. Intraoperative autotransfusion in cardiac operations. J Thorac Cardiovasc Surg 1988; 96:382-86
 Ulmas J, O'Neill TP, Heparin removal in an autotransfusion device. Transfusion 1981; 21:70-3
 Council of Scientific Affairs. Autologous blood transfusions. JAMA 1986; 256:2378-80
 Utley JR, Moores WY, Stephens DB. Blood conservation techniques. Ann Thorac Surg 1981; 31:482-90
 Fleming AW, Green DC, Radcliffe JH, St. James OM, Fleming EW. Development of a practical autologous blood transfusion program. Am Surg 1977; 43:794-801
 Cove H, Marloff J, Sacks HJ, Sherbecoe R, Goldfinger D. Autologous blood transfusion in coronary artery bypass surgery. Transfusion 1977; 16:245-48
 Pilam MV, McGoon DC, Tarhan S. Failure of transfusion of autologous whole blood to reduce banked-blood requirements in open-heart surgical patients. J Thorac Cardiovasc Surg 1975; 70:338-43
 Sherman MM, Dobnik DB, Dennis RC, Berger RI. Autologous blood transfusion during cardiopulmonary bypass. Chest 1976; 70:592-95
 Kaplan JA, Cannarella C, Jones EL, Kutner MH, Hatcher CR, Dunbar RW. Autologous blood transfusion during cardiac surgery. J Thorac Cardiovasc Surg 1977; 74:4-10
 Moran JM, Babka R, Silberman S, Rice LP, Pifarre R, Sullivan JH, et al. Immediate centrifugation of oxygenator contents after cardiopulmonary bypass: role in maximum blood conservation. J Thorac Cardiovasc Surg 1978; 76:510-17
 Mattox KL. Comparison of techniques of autotransfusion. Surgery 1978; 84:700-02
 Schaff HB, Hauer JM, Brawley RK. Autotransfusion in cardiac surgical patients after operation. Surgery 1978; 84:713-18
 Schaff HB, Hauer JM, Bell WR, Gardner TJ, Donahoo JS, Gott VL, et al. Routine use of autotransfusion following cardiac surgery: experience in 700 patients. Ann Thorac Surg 1979; 27:493-99
COPYRIGHT 1993 American College of Chest Physicians
COPYRIGHT 2004 Gale Group