- Email: [email protected]
Reduction of blood utilization during myocardial revascularization
J
THoRAc CARDIOVASC SURG
1989;97:213-9
Reduction of blood utilization during myocardial revascularization The efficacy of four different blood conservation techniques in decreasing the homologous blood requirement in cardiac operations was studied prospectively in 100 patients undergoing myocardial revascularization. The patients were randomly assigned to four groups of 25 each as foUows: group I, retransfusion of oxygenator blood after termination of extracorporeal circulation; group II, processing of oxygenator content by means of a ceU separator; group ill, predonation of autologous blood and isovolumetric substitution of hydroxyethyl starch (10 ml/kg bodyweight)after the induction of anesthesia in addition to the use of a ceU separator; and group IV, predonation and the use of a ceU separator plus postoperative retransfusion of shed mediastinal blood. To form homologous groups, we accepted only male patients without impairment of left ventricular function for the study. In addition, patients with internal mammary artery grafts and a duration of extracorporeal circulation less than 45 minutes or more than 90 minutes were excluded. The bank blood requirement during hospitalization was 2132 ± 824 mI in group I, 1371 ± 928 mI in group 11,833 ± 599 mI in group III, and 408 ± 559 mI in group IV. The use of blood conservation techniques resulted in reductions of homologous blood requirements of 34 %, 60 %, and 80%, respectively, in groups II to IV as compared with the requirement in group I. There were no complications related to autologous blood transfusion. We conclude that the use of blood conservation techniquescan considerably reduce the homologous blood requirement in cardiac operations and therefore decrease transfusion-related risks.
Wulf Dietrich, MD, Andreas Barankay, MD, Gerd Dilthey, MD, Hans-Peter Mitto, MD, and Josef A. Richter, MD, Munich, Federal Republic of Germany
Both the risks associated with transfusions! and the limited availability of homologous blood? force us to reduce the homologous blood requirement in cardiac operations. In recent years, various methods have been developed to minimize homologous blood requirements. Techniques that lead to a 15% to 50% reduction in the homologous blood requirement are preoperative harvesting of autologous blood with separated storage of frozen plasma and erythrocytes," preoperative isovolumetric hernodilution.t' intraoperative hemoseparation with a cell separator.v? intraoperative hemofiltration," and postoperative retransfusion of shed mediastinal blood.I I. 12
From the Institute for Anesthesiology, German Heart Center Munich. Munich, Federal Republic of Germany. Received for publication Dec. 4, 1987. Accepted for publication Aug. 17, 1988. Address for reprints: Wulf Dietrich. MD, German Heart Center Munich, Institute for Anesthesiology, Lothstrasse 11, 0-8000 Munich 2.
However, these techniques have been applied to different patient groups, either alone or in combination, and the studied groups have not always been comparable. Therefore, it is often difficult to estimate the quantitative benefit of the method under investigation. In this study, homogenous patient groups were selected to assess the efficacy of combined autotransfusion methods on homologous blood requirements in coronary operations. Method After informed consent was obtained and with clinical investigation committee approval 100 male patients, operated on for coronary artery bypass grafting, were selected for this prospective study. Only patients with a left ventricular ejection fraction higher than 50% and a preoperative hemoglobin content exceeding 140 gm/L were included. Patients with extracorporeal circulation (ECC) times less than 45 minutes and longer than 90 minutes and patients receiving internal mammary artery implantation were excluded. Patients were randomly assigned to four groups. Group I. Unprocessed oxygenator blood was retransfused after termination of ECC. Group II. The blood remaining in the oxygenator after ECC
213
2 14
The Journal of Thoracic and Cardiovascular Surgery
Dietrich et al.
Table I. Patient data" Group I
Age (yr) Weight (kg) BSA (m') Op. time (min) ECC (min) Grafts (n) Days in ICU (n)
56.0 ± 73.9 ± 1.87 ± 216 ± 72 ± 2.9 ± 2.8 ±
Group II
6.6 7.1 0.1 26 II 0.6 0.9
54.1 ± 72.2 ± 1.84 ± 209 ± 66 ± 2.86 ± 2.9 ±
Group III
55.0 ± 75.3 ± 1.90 ± 203 ± 67 ± 2.9 ± 2.6 ±
6.8 8.5 0.1 32 12 0.5 1.4
Group IV
9.4 9.9 0.17 28 13 0.5l 1.0
55.7 ± 6.3 77.3 ± 9.4 1.93±0.15 208 ± 35 70 ± 17 2.7 ± 0.56 2.6 ± 0.96
*DitTerences between groups were not significant.
Table II. Homologous blood requirement (ml) Group I
Homologous Homologous Homologous Homologous Fresh frozen Homologous *p tp
blood intraoperatively blood 12 hr postop. blood 12-24 hr postop. blood during entire hospitalization plasma (Op. and ICU) packed cells
757 ± 1231 ± 34 ± 2066 ± 493 ± 70 ±
527 563 142 756 385 197
Group II
558 ± 686 ± 66 ± 1391 ± 552 ± 40 ±
547 634* 183 964* 368 200
Group III
10 ± 490 ± 146 ± 833 ± 291 ± 138 ±
50*t 522* 329* 599*t 381 192
Group IV
20 ± 283 ± 84 ± 409 ± 234 ± 90 ±
IOO*t 485* 244 559*t 303 189
< 0.01 versus group I. < 0.01 versus group II.
was processed to packed cells with a cell separator* and retransfused until the end of the operation. Group III. After induction of anesthesia and before the start of the operation, isovolumetric hemodilution (harvesting of 10 ml/kg autologous blood) was performed under electrocardiographic and hemodynamic control. The blood loss was replaced with hydroxyethyl starch. After termination of ECC, the blood remaining in the oxygenator was processed by a cell separator. The preoperatively drawn blood and the packed cells were retransfused before the end of the operation. Group IV. Patients in group IV were managed as in group III. In addition, the shed mediastinal blood was retransfused on the intensive care unit (lCU). The cardiotomy reservoir of the heart-lung machine was used to collect this blood. 13. 14 The drained blood was retransfused intermittently according to the circulatory state of the patient and when at least 250 ml of blood had been collected in the reservoir. For reasons of sterility, the last retransfusion was performed 6 hours postoperatively. In all patients, a standard perfusion technique was performed with a bubble oxygenator primed with 1400 rnl of a crystalloid solution. Patients received 375 units of mucosaheparin per kilogram under control of the activated clotting time. After termination of ECC, heparin was antagonized by protamine chloride 4.5 mg/kg, Preoperative red cell volume was calculated by multiplying the hematocrit level by the blood volume, which was obtained by a nomogram of age, sex, and weight." The amounts of autologous and homologous blood, fresh frozen plasma, and crystalloid and colloid solutions required were recorded intra*Cell Saver, Haemonetics Corp., Braintree, Mass.
operatively and postoperatively. Blood loss was assessed 6, 12, and 24 hours postoperatively. Hemoglobin and hematocrit values were registered preoperatively, before ECC, 5 minutes after the start of ECC, at the end of ECC, 30 minutes after termination of ECC, on arrival in the ICU, 6 hours postoperatively, during the morning of the first postoperative day, on discharge from the ICU, and at hospital dismissal. The attending resident on the ICU was not advised as to which group each patient had been assigned. In all patients, signs of hypovolemia and hematocrit values below 30% were indications for homologous blood transfusions. The results were analyzed with the PMDP statistical software program." Analysis of variance for repeated measurements was performed for each group to determine whether significant differences were present. The Bonferroni modification of the t test for multiple comparisons was used to test the significance of the differences between the groups. A probability value less than 0.01 was considered statistically significant. Results were given as mean ± standard deviation.
Results Patient and operative data (Table I) show that the four groups of patients were comparable and that the surgical procedures were nearly identical. The requirements for homologous blood derivatives, infusion of crystalloid solutions, and homologous blood transfusions are shown in Tables II and III. If the total homologous blood requirement of group I is taken as 100%, groups II, III, and IV showed a reduction of 34%,
Volume 97
Reduction of blood utilization
Number 2 February 1989
2 15
Table m. Intraoperative infusion of crystalloids and colloids (ml) Crystalloids Colloids
Group I
Group II
Group III
Group IV
604 ± 407 582 ± 180
448 ± 174 709 ± 255
463 ± 268 1093 ± 330*t
479 ± 288 1076 ± 265*t
*p < 0.01 versus group I.
s» < 0.01
versus group II.
Table IV. Red cell volume (ml)* Preoperative At discharge from 1CU At dismissal from hospital
Group I
Group II
Group III
Group IV
2075 ± 215 1831 ± 192 1930 ± 217
2074 ± 293 1849 ± 221 1929 ± 249
2186 ± 288 1738 ± 223 1856 ± 241
2203 ± 250 1819 ± 228 1860 ± 199
'Differences between groups were not significant.
60%, and 80%, respectively. One patient in group I, 11 patients in group II, and 24 patients in both groups III and IV received no homologous blood transfusions during operation. No patient in group I, two patients (8%) in group II, three patients (12%) in group III, and eight patients (32%) in group IV completed hospitalization without having received homologous blood derivatives. The homologous blood requirements showed no significant correlations with preoperative patient data such as body surface area, body weight, preoperative hematocrit value, duration of ECC, or operation time in any group. However, we found a significant correlation (r = 0.46, P < 0.01) between preoperative red cell volume of all patients and total homologous blood requirement during the hospital stay. The preoperative red cell volume (Table IV) showed no significant difference among the four groups, though there was a tendency to higher values in groups III and IV. At discharge from the ICU and dismissal from the hospital, these values were within comparable ranges in all groups. Table V shows the amount of autologous blood received. Fig. I demonstrates that 5 minutes after heparinization the hematocrit value in the groups receiving preoperative hemodilution (III and IV) was only slightly lower (33.5% and 34.8%) than that in the groups receiving no hemodilution (I and II) (37.8% and 37.2%). While the patients were in the ICU, the mean hematocrit value was always above 30%. A comparison between group II (hemoseparation) and group I (retransfusion of unprocessed oxygenator blood) showed higher hematocrit values in the hemoseparation group than in group I
30 minutes after ECC (34% and 38%, respectively) and on arrival at the ICU (32.8% and 35.9%, respectively) (Fig. 2). Changes in hematocrit value in group III (preoperative hemodilution) compared with those in group II (no hemodilution) are shown in Fig. 3. Only group III (preoperative hemodilution) showed a significantly elevated blood loss 24 hours postoperatively (890 ± 359 ml) compared with group 1(649 ± 181 ml) and group II (664 ± 256 ml) (Table VI). No patient required reexploration and surgical hemostasis because of postoperative bleeding. One patient (group II) died 40 days postoperatively as a consequence of septic complications. However, there were no complications that might have been related to the blood saving methods used. Discussion
To investigate special blood saving methods, homogenous patient groups were studied. Although neither the ECC duration" nor the implantation of the mammary artery" is said to influence the blood requirement, only male patients being operated on for coronary artery disease for the first time, who were not receiving a mammary artery bypass graft and for whom the bypass time was between 45 and 90 minutes, were included in our study. The pivotal point, when the possibility of reducing homologous blood requirements is being considered, is the lowest accepted hematocrit level before, during, and after ECe. In healthy persons, Messmer and associates" showed that hematocrit values as low as 15% could be tolerated by the organism without any consequences and that nutritional blood flow increased with
The Journal of
2 16
T'toracic and Cardiovascular Surgery
Dietrich et al.
[%] 50
40
30
..... -G-
Group I
-o- Group II
E C C 20 2
3
)
4
5
6
7
8
Group III Group IV
9
10
Fig. l. Course of hematocrit value during hospitalization. 1. Preoperatively; 2. 5 minutes after heparin administration; 3. 5 minutes after pump start; 4. end of ECC; 5. 30 minutes after termination of ECC; 6. arrival at ICU; 7. 12 hours postoperatively; 8. postoperative day I; 9. discharge from ICU; 10. hospital dismissal. Despite reduction of blood requirements, there were no significant differences between groups. Postoperative normovolemic anemia was not accepted in these patients. During postoperative period, hematocrit value in all groups was always above 30%.
Table V. Autologous blood (ml) Packed cells Preoperatively drawn blood Shed mediastinal blood
Group I
Group II
Group III
Group IV
o o o
515 ± 197
501 ± 191 731 ± 201
626 ± i88 715 ± 142 378 ± 173
decreasing hematocrit value. However, these results cannot be generalized to patients having reduced organ blood supply as a result of atherosclerosis or to patients with impaired pump function of the heart. Unlike healthy persons, patients with coronary artery disease cannot increase coronary blood flow with decreasing hematocrit values, because further coronary artery vasodilation is impossible. In this case, hemoglobin content becomes the limiting factor in oxygen transport capacity. Weisel and colleagues" found a significant reduction in homologous blood requirements as a result of the postoperative administration of crystalloid solutions instead of plasma and packed red cells. However, this group observed that, despite operative revascularization, the acceptance of postoperative normovolemic anemia
o o
o
(hemoglobin level <70 gm/L) markedly prolonged the recovery time of the myocardial metabolism and that such anemia could even lead to anaerobic glycolysis under stress conditions. Animal models also showed a detrimental effect of hemodilution on myocardial contractility" or a-adrenergic stimulation under conditions of reduced coronary blood flow." On the other hand, other authors did not observe any electrocardiographic or enzymatic evidence of myocardial ischemia with acceptance of postoperative normovolemic anemia." 2J On the basis of these assumptions, we did not accept a postoperative normovolemic anemia. Consequently, the hematocrit value was kept above 30% postoperatively. In this study, preoperative isovolumetric hemodilution resulted in an additional reduction of blood requirement of 40% (group III versus group 11). This reduction can
Volume 97 Number 2 February 1989
Reduction of blood utilization
2 17
HEMATOCRIT
% 38
36 34 32 30 28
\
t
\\
\\ \\ \\
\\ \
.. Oxygenator retransfusion
Cellsaving
\
T_ _--'--_ _
• Cell Saver
. . . J . . . . - - _ - - - L_ _
ECC 5'
end of
ECC
30' after ECC
~
_
ICU
Fig. 2. The effect of hemoseparation. Retransfusion of autologous packed cells markedly increased hematocrit value shortly after termination of ECC. Unnecessary volume load was avoided during this period.
Table VI. Cumulative postoperative blood loss (ml) 6 hours postop. 12 hours postop. 24 hours postop.
Group I
Group II
Group III
Group IV
373 ± 102 476 ± 126 649 ± 181
394 ± 201 500 ± 230 664 ± 256
509 ± 240 664 ± 282* 890 ± 359*t
378 ± 173 533 ± 220 722 ± 233
*p < 0.01 versus group I. tp < 0.01 versus group II.
be explained by the retransfusion of autologous warm blood as volume substitution at the termination of operation. Only one patient in each of the groups receiving preoperative hemodilution (III and IV) received a homologous transfusion intraoperatively, whereas this was necessary for 21 and 16 patients in groups I and II, respectively. Despite this difference, the course of the hematocrit value after termination of ECC was parallel (Fig. 3). On the other hand, no measurable improvement in the coagulation system could be demonstrated with the transfusion of autologous warm blood; conversely, a significantly higher 24-hour blood loss in group III than in groups I and II was found. This might be explained by the higher amount of colloidal volume substitutes administered to patients in group III during the course of the operation. Hemoseparation with a cell separator allows a rapid elevation of the hematocrit value after termination of ECC without fluid overloading. Despite a lower intra-
operative homologous blood requirement, the hematocrit value of patients in group II (hemoseparation) was markedly higher than that in the group receiving no hemoseparation (group I) 30 minutes after ECC (Fig. 2). Only 15% to 20% of the circulating blood volume is processed with the oxygenator content, hence the protein loss, inevitable with hemoseparation, is not clinically relevant. Because the increase in the hematocrit value is the effect of hemoseparation, this method would be less effective when normovolemic anemia is accepted after termination of ECC. 22 Hemoseparation by means of a locally heparinized collecting system is not indicated in primary coronary operations," inasmuch as the amount of blood collected before and after ECC is usually too small for effective processing. Despite the low quality of postoperative drainage blood (consumed coagulation factors, enhanced fibrinolytic activity,":" low hemoglobin content (76 gm/L in
The Journal of Thoracic and Cardiovascular Surgery
2 1 8 Dietrich et al.
II II
RETRANSfUSION
preop
5' after heparin
ECC 5'
end of
30' after
ECC
ECC
ICU
12 h postop
Fig. 3. The effect of preoperative hemodilution. Only one patient in group III (preoperative hemodilution) received Y2 unit of banked blood intraoperatively, whereas this was necessary in 16 patients in group II (no hemodilution). Nevertheless, the difference in hematocrit value 30 minutes after ECC was negligible.
this investigationj) shed mediastinal blood retransfusion can safely be used as an autologous volume substitute in the early postoperative period, which can postpone or even avoid homologous transfusions. Because of the low hematocrit value and the consumption of coagulation factors, the drainage fluid is solely a volume substitute; in addition, it neither improves the coagulation properties nor has a detrimental effect on coagulation. The shed mediastinal blood has a negligible effect on increasing the hematocrit value, at least with the amounts that on an average were retransfused. As in other studies,":" a reduction in homologous blood requirement with shed mediastinal blood retransfusion was found in this study (833 ± 598 ml in group III versus 408 ± 558 ml in group IV). Although this reduction was marked (50%), it was not considered statistically significant compared with the blood requirement of the group without retransfusion (group III) (p = 0.04). Cosgrove and associates '5 described the preoperative red cell volume as a predictor of the need for transfusion. We also found a significant correlation between red cell volume and blood requirement. The more pronounced decrease in this parameter in groups III and IV is not a sign of acceptance of lower hematocrit values in these groups. It is rather an
indication that in these groups the appropriate volume, that is, autologous blood, was available at the right time. As shown in Figs. 2 and 3, it was possible to increase the hematocrit value by either hemoseparation or preoperative hemodilution after ECC to an equal or even higher level compared with groups I and II, using less homologous blood. Though the preoperative red cell volume was nearly identical in groups I and II and in groups III and IV, respectively, the transfusion reduction between groups I and II was 40% and between groups III and IV was 50%. These findings cannot be explained by an unrecognized difference in red cell volume but only by the application of different blood saving methods. From this study, we conclude that the combination of different blood saving methoos leads to a significant reduction in autologous blood requirement during coronary operations. Intraoperative hemoseparation involves no risks but is not without a certain financial and technical expense. Prebypass isovolumetric hemodilution represents a safe and simple method when there are no contraindications and the patients are closely monitored. Postoperative retransfusion of shed blood (despite the low blood quality) allows an effective volume substitution especially in patients with unexpected large postop-
Volume 97 Number 2 February 1989
erative hemorrhage. All these methods have become standard procedures in our everyday practice. Statistical analysis was performed by Dr. W. Kopcke, Department of Medical Statistics and Data Processing, University Munich. REFERENCES I. Collins JA. Problems associated with the massive transfusion of stored blood. Surgery 1974;75:274-95. 2. Roche JK, Stengle JM. Open-heart surgery and the demand for blood. JAMA 1973;225:1516-21. 3. Schricker KT, Neidhardt B, van der Emde J. Die autologe Bluttransfusion tiefkiihlkonservierten Blutes in der Herzchirurgie. Dtsch Med Wochenschr 1981; 41:1333-7. 4. Toy PTCY, Strauss RG, Stehling LC, et al. Predeposited autologous blood for elective surgery. N Engl J Med 1987;316:517-20. 5. Love TR, Hendren WG, O'Keefe DD, Daggett WM. Transfusion of predonated autologous blood in elective cardiac surgery. Ann Thorac Surg 1987;43:508-12. 6. Lowenstein E. Blood conservation in open heart surgery. Cleve Clin Quart 1981;48:112-25. 7. Kaplan JA, Canarella C, Jones EL, et al. Autologous blood transfusion during cardiac surgery. J THORAC CARDIOVASC SURG 1977;74:4-10. 8. Moran JM, Babka R, Silberman S, et al. Immediate centrifugation of oxygenator contents after cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1978;76: 510-7. 9. Dietrich W, Gob E, Barankay A, Mitto HP, Richter JA. Reduzierung des Fremdblutverbrauches in der Koronarchirurgie durch Harnoseparation und isovolamische Hamodilution, Anaesthesist 1983;32:427-32. 10. Breyer RH, Engelman RM, Rousou JR, Lemeshow SA. A comparison of Cell Saver versus ultrafilter during coronary artery bypass operations. J THORAC CARDIOVASC SURG 1985;90:736-40. 11. Schaff HV, Hauer JM, Bell VR, et al. Autotransfusion of shed mediastinal blood after cardiac surgery: a prospective study. J THORAC CARDIOVASC SURG 1978;75:63241. 12. Breyer RH, Engelman RM, Rousou JA, Lemeshow S. Blood conservation for myocardial revascularization: is it cost-effective? J THORAC CARDIOVASC SURG 1987;93:51222. 13. Weniger J, Shanahan R. Reduction of blood requirements in cardiac surgery. Thorac Cardiovasc Surg 1982;30: 142-6.
Reduction of blood utilization
2 19
14. Cosgrove DM, Amiot DM, Meserko JJ. An improved technique for autotransfusion of shed mediastinal blood. Ann Thorac Surg 1985;40:519-20. 15. Cosgrove DM, Loop FD, Lytle BW, et al. Determinants of blood utilization during myocardial revascularization. Ann Thorac Surg 1985;40:380-4. 16. BMDP statistical software. Dixon WJ, ed. Berkeley, California: University of California Press, 1981. 17. Shevde K, Stigi S, Lee S. The effect of internal mammary artery grafting on the perioperative course. J Cardiothorae Anesth 1988;2:30-3. 18. Messmer K, Sunder-Plassmann L, Jesch F, et al. Oxygen supply to the tissues during limited normovolemic hemodilution. Res Exp Med (Berl) 1973;159:152-66. 19. Weisel RD, Charlesworth DC, Mickleborough LL, et al. Limitations of blood conservation. J THORAC CARDIOVASC SURG 1984;88:26-38. 20. Hagl S, Heimisch W, Meisner H, Mendler N. The effect of hemodilution on regional myocardial function in the presence of coronary stenosis. Basic Res Cardiol 1977;72:344-64. 21. Estafanous FG, Streng Z, Pedvinelli R, Azmy SS, Tarazi RC. Hemodilution affects the pressor response to norepinephrine. J Cardiothorae Anaesth 1987;1:36-41. 22. Cosgrove DM, Thurer RL, Lytle BW, et al. Blood conservation during myocardial revaseularization. Ann Thorae Surg 1979;28:184-9. 23. Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotransfusion following cardiac operations: a randomized prospective study. Ann Thorae Surg 1979;27:500-7. 24. Winton TL, Charrette EJP, Salerno T A. The cell saver during cardiac surgery: does it save? Ann Thorac Surg 1982;33:379-81. 25. Symbas PN. Extraoperative autotransfusion from hematothorax. Surgery 1978;84:722-7. 26. Adan A, Brutel de la Riviere A, Haas F, van Zalk A, de Nooij E. Autotransfusion of drained mediastinal blood after cardiac surgery: a reappraisal. Thorae Cardiovase Surg 1988;36:10-4. 27. Schaff HV, Hauer JM, Gardner TJ, et al. Routine use of autotransfusion following cardiac surgery: experience in 700 patients. Ann Thorae Surg 1979;27:493-9. 28. Cordell AR, Lavender SW. An appraisal of blood salvage techniques in vascular and cardiac operations. Ann Thorae Surg 1981;31:421-5. 29. Johnson RG, Rosenkrantz KR, Preston RA, et al. The efficacy of postoperative autotransfusion in patients undergoing cardiac operations. Ann Thorae Surg 1983;36:173-9.
THoRAc CARDIOVASC SURG
1989;97:213-9
Reduction of blood utilization during myocardial revascularization The efficacy of four different blood conservation techniques in decreasing the homologous blood requirement in cardiac operations was studied prospectively in 100 patients undergoing myocardial revascularization. The patients were randomly assigned to four groups of 25 each as foUows: group I, retransfusion of oxygenator blood after termination of extracorporeal circulation; group II, processing of oxygenator content by means of a ceU separator; group ill, predonation of autologous blood and isovolumetric substitution of hydroxyethyl starch (10 ml/kg bodyweight)after the induction of anesthesia in addition to the use of a ceU separator; and group IV, predonation and the use of a ceU separator plus postoperative retransfusion of shed mediastinal blood. To form homologous groups, we accepted only male patients without impairment of left ventricular function for the study. In addition, patients with internal mammary artery grafts and a duration of extracorporeal circulation less than 45 minutes or more than 90 minutes were excluded. The bank blood requirement during hospitalization was 2132 ± 824 mI in group I, 1371 ± 928 mI in group 11,833 ± 599 mI in group III, and 408 ± 559 mI in group IV. The use of blood conservation techniques resulted in reductions of homologous blood requirements of 34 %, 60 %, and 80%, respectively, in groups II to IV as compared with the requirement in group I. There were no complications related to autologous blood transfusion. We conclude that the use of blood conservation techniquescan considerably reduce the homologous blood requirement in cardiac operations and therefore decrease transfusion-related risks.
Wulf Dietrich, MD, Andreas Barankay, MD, Gerd Dilthey, MD, Hans-Peter Mitto, MD, and Josef A. Richter, MD, Munich, Federal Republic of Germany
Both the risks associated with transfusions! and the limited availability of homologous blood? force us to reduce the homologous blood requirement in cardiac operations. In recent years, various methods have been developed to minimize homologous blood requirements. Techniques that lead to a 15% to 50% reduction in the homologous blood requirement are preoperative harvesting of autologous blood with separated storage of frozen plasma and erythrocytes," preoperative isovolumetric hernodilution.t' intraoperative hemoseparation with a cell separator.v? intraoperative hemofiltration," and postoperative retransfusion of shed mediastinal blood.I I. 12
From the Institute for Anesthesiology, German Heart Center Munich. Munich, Federal Republic of Germany. Received for publication Dec. 4, 1987. Accepted for publication Aug. 17, 1988. Address for reprints: Wulf Dietrich. MD, German Heart Center Munich, Institute for Anesthesiology, Lothstrasse 11, 0-8000 Munich 2.
However, these techniques have been applied to different patient groups, either alone or in combination, and the studied groups have not always been comparable. Therefore, it is often difficult to estimate the quantitative benefit of the method under investigation. In this study, homogenous patient groups were selected to assess the efficacy of combined autotransfusion methods on homologous blood requirements in coronary operations. Method After informed consent was obtained and with clinical investigation committee approval 100 male patients, operated on for coronary artery bypass grafting, were selected for this prospective study. Only patients with a left ventricular ejection fraction higher than 50% and a preoperative hemoglobin content exceeding 140 gm/L were included. Patients with extracorporeal circulation (ECC) times less than 45 minutes and longer than 90 minutes and patients receiving internal mammary artery implantation were excluded. Patients were randomly assigned to four groups. Group I. Unprocessed oxygenator blood was retransfused after termination of ECC. Group II. The blood remaining in the oxygenator after ECC
213
2 14
The Journal of Thoracic and Cardiovascular Surgery
Dietrich et al.
Table I. Patient data" Group I
Age (yr) Weight (kg) BSA (m') Op. time (min) ECC (min) Grafts (n) Days in ICU (n)
56.0 ± 73.9 ± 1.87 ± 216 ± 72 ± 2.9 ± 2.8 ±
Group II
6.6 7.1 0.1 26 II 0.6 0.9
54.1 ± 72.2 ± 1.84 ± 209 ± 66 ± 2.86 ± 2.9 ±
Group III
55.0 ± 75.3 ± 1.90 ± 203 ± 67 ± 2.9 ± 2.6 ±
6.8 8.5 0.1 32 12 0.5 1.4
Group IV
9.4 9.9 0.17 28 13 0.5l 1.0
55.7 ± 6.3 77.3 ± 9.4 1.93±0.15 208 ± 35 70 ± 17 2.7 ± 0.56 2.6 ± 0.96
*DitTerences between groups were not significant.
Table II. Homologous blood requirement (ml) Group I
Homologous Homologous Homologous Homologous Fresh frozen Homologous *p tp
blood intraoperatively blood 12 hr postop. blood 12-24 hr postop. blood during entire hospitalization plasma (Op. and ICU) packed cells
757 ± 1231 ± 34 ± 2066 ± 493 ± 70 ±
527 563 142 756 385 197
Group II
558 ± 686 ± 66 ± 1391 ± 552 ± 40 ±
547 634* 183 964* 368 200
Group III
10 ± 490 ± 146 ± 833 ± 291 ± 138 ±
50*t 522* 329* 599*t 381 192
Group IV
20 ± 283 ± 84 ± 409 ± 234 ± 90 ±
IOO*t 485* 244 559*t 303 189
< 0.01 versus group I. < 0.01 versus group II.
was processed to packed cells with a cell separator* and retransfused until the end of the operation. Group III. After induction of anesthesia and before the start of the operation, isovolumetric hemodilution (harvesting of 10 ml/kg autologous blood) was performed under electrocardiographic and hemodynamic control. The blood loss was replaced with hydroxyethyl starch. After termination of ECC, the blood remaining in the oxygenator was processed by a cell separator. The preoperatively drawn blood and the packed cells were retransfused before the end of the operation. Group IV. Patients in group IV were managed as in group III. In addition, the shed mediastinal blood was retransfused on the intensive care unit (lCU). The cardiotomy reservoir of the heart-lung machine was used to collect this blood. 13. 14 The drained blood was retransfused intermittently according to the circulatory state of the patient and when at least 250 ml of blood had been collected in the reservoir. For reasons of sterility, the last retransfusion was performed 6 hours postoperatively. In all patients, a standard perfusion technique was performed with a bubble oxygenator primed with 1400 rnl of a crystalloid solution. Patients received 375 units of mucosaheparin per kilogram under control of the activated clotting time. After termination of ECC, heparin was antagonized by protamine chloride 4.5 mg/kg, Preoperative red cell volume was calculated by multiplying the hematocrit level by the blood volume, which was obtained by a nomogram of age, sex, and weight." The amounts of autologous and homologous blood, fresh frozen plasma, and crystalloid and colloid solutions required were recorded intra*Cell Saver, Haemonetics Corp., Braintree, Mass.
operatively and postoperatively. Blood loss was assessed 6, 12, and 24 hours postoperatively. Hemoglobin and hematocrit values were registered preoperatively, before ECC, 5 minutes after the start of ECC, at the end of ECC, 30 minutes after termination of ECC, on arrival in the ICU, 6 hours postoperatively, during the morning of the first postoperative day, on discharge from the ICU, and at hospital dismissal. The attending resident on the ICU was not advised as to which group each patient had been assigned. In all patients, signs of hypovolemia and hematocrit values below 30% were indications for homologous blood transfusions. The results were analyzed with the PMDP statistical software program." Analysis of variance for repeated measurements was performed for each group to determine whether significant differences were present. The Bonferroni modification of the t test for multiple comparisons was used to test the significance of the differences between the groups. A probability value less than 0.01 was considered statistically significant. Results were given as mean ± standard deviation.
Results Patient and operative data (Table I) show that the four groups of patients were comparable and that the surgical procedures were nearly identical. The requirements for homologous blood derivatives, infusion of crystalloid solutions, and homologous blood transfusions are shown in Tables II and III. If the total homologous blood requirement of group I is taken as 100%, groups II, III, and IV showed a reduction of 34%,
Volume 97
Reduction of blood utilization
Number 2 February 1989
2 15
Table m. Intraoperative infusion of crystalloids and colloids (ml) Crystalloids Colloids
Group I
Group II
Group III
Group IV
604 ± 407 582 ± 180
448 ± 174 709 ± 255
463 ± 268 1093 ± 330*t
479 ± 288 1076 ± 265*t
*p < 0.01 versus group I.
s» < 0.01
versus group II.
Table IV. Red cell volume (ml)* Preoperative At discharge from 1CU At dismissal from hospital
Group I
Group II
Group III
Group IV
2075 ± 215 1831 ± 192 1930 ± 217
2074 ± 293 1849 ± 221 1929 ± 249
2186 ± 288 1738 ± 223 1856 ± 241
2203 ± 250 1819 ± 228 1860 ± 199
'Differences between groups were not significant.
60%, and 80%, respectively. One patient in group I, 11 patients in group II, and 24 patients in both groups III and IV received no homologous blood transfusions during operation. No patient in group I, two patients (8%) in group II, three patients (12%) in group III, and eight patients (32%) in group IV completed hospitalization without having received homologous blood derivatives. The homologous blood requirements showed no significant correlations with preoperative patient data such as body surface area, body weight, preoperative hematocrit value, duration of ECC, or operation time in any group. However, we found a significant correlation (r = 0.46, P < 0.01) between preoperative red cell volume of all patients and total homologous blood requirement during the hospital stay. The preoperative red cell volume (Table IV) showed no significant difference among the four groups, though there was a tendency to higher values in groups III and IV. At discharge from the ICU and dismissal from the hospital, these values were within comparable ranges in all groups. Table V shows the amount of autologous blood received. Fig. I demonstrates that 5 minutes after heparinization the hematocrit value in the groups receiving preoperative hemodilution (III and IV) was only slightly lower (33.5% and 34.8%) than that in the groups receiving no hemodilution (I and II) (37.8% and 37.2%). While the patients were in the ICU, the mean hematocrit value was always above 30%. A comparison between group II (hemoseparation) and group I (retransfusion of unprocessed oxygenator blood) showed higher hematocrit values in the hemoseparation group than in group I
30 minutes after ECC (34% and 38%, respectively) and on arrival at the ICU (32.8% and 35.9%, respectively) (Fig. 2). Changes in hematocrit value in group III (preoperative hemodilution) compared with those in group II (no hemodilution) are shown in Fig. 3. Only group III (preoperative hemodilution) showed a significantly elevated blood loss 24 hours postoperatively (890 ± 359 ml) compared with group 1(649 ± 181 ml) and group II (664 ± 256 ml) (Table VI). No patient required reexploration and surgical hemostasis because of postoperative bleeding. One patient (group II) died 40 days postoperatively as a consequence of septic complications. However, there were no complications that might have been related to the blood saving methods used. Discussion
To investigate special blood saving methods, homogenous patient groups were studied. Although neither the ECC duration" nor the implantation of the mammary artery" is said to influence the blood requirement, only male patients being operated on for coronary artery disease for the first time, who were not receiving a mammary artery bypass graft and for whom the bypass time was between 45 and 90 minutes, were included in our study. The pivotal point, when the possibility of reducing homologous blood requirements is being considered, is the lowest accepted hematocrit level before, during, and after ECe. In healthy persons, Messmer and associates" showed that hematocrit values as low as 15% could be tolerated by the organism without any consequences and that nutritional blood flow increased with
The Journal of
2 16
T'toracic and Cardiovascular Surgery
Dietrich et al.
[%] 50
40
30
..... -G-
Group I
-o- Group II
E C C 20 2
3
)
4
5
6
7
8
Group III Group IV
9
10
Fig. l. Course of hematocrit value during hospitalization. 1. Preoperatively; 2. 5 minutes after heparin administration; 3. 5 minutes after pump start; 4. end of ECC; 5. 30 minutes after termination of ECC; 6. arrival at ICU; 7. 12 hours postoperatively; 8. postoperative day I; 9. discharge from ICU; 10. hospital dismissal. Despite reduction of blood requirements, there were no significant differences between groups. Postoperative normovolemic anemia was not accepted in these patients. During postoperative period, hematocrit value in all groups was always above 30%.
Table V. Autologous blood (ml) Packed cells Preoperatively drawn blood Shed mediastinal blood
Group I
Group II
Group III
Group IV
o o o
515 ± 197
501 ± 191 731 ± 201
626 ± i88 715 ± 142 378 ± 173
decreasing hematocrit value. However, these results cannot be generalized to patients having reduced organ blood supply as a result of atherosclerosis or to patients with impaired pump function of the heart. Unlike healthy persons, patients with coronary artery disease cannot increase coronary blood flow with decreasing hematocrit values, because further coronary artery vasodilation is impossible. In this case, hemoglobin content becomes the limiting factor in oxygen transport capacity. Weisel and colleagues" found a significant reduction in homologous blood requirements as a result of the postoperative administration of crystalloid solutions instead of plasma and packed red cells. However, this group observed that, despite operative revascularization, the acceptance of postoperative normovolemic anemia
o o
o
(hemoglobin level <70 gm/L) markedly prolonged the recovery time of the myocardial metabolism and that such anemia could even lead to anaerobic glycolysis under stress conditions. Animal models also showed a detrimental effect of hemodilution on myocardial contractility" or a-adrenergic stimulation under conditions of reduced coronary blood flow." On the other hand, other authors did not observe any electrocardiographic or enzymatic evidence of myocardial ischemia with acceptance of postoperative normovolemic anemia." 2J On the basis of these assumptions, we did not accept a postoperative normovolemic anemia. Consequently, the hematocrit value was kept above 30% postoperatively. In this study, preoperative isovolumetric hemodilution resulted in an additional reduction of blood requirement of 40% (group III versus group 11). This reduction can
Volume 97 Number 2 February 1989
Reduction of blood utilization
2 17
HEMATOCRIT
% 38
36 34 32 30 28
\
t
\\
\\ \\ \\
\\ \
.. Oxygenator retransfusion
Cellsaving
\
T_ _--'--_ _
• Cell Saver
. . . J . . . . - - _ - - - L_ _
ECC 5'
end of
ECC
30' after ECC
~
_
ICU
Fig. 2. The effect of hemoseparation. Retransfusion of autologous packed cells markedly increased hematocrit value shortly after termination of ECC. Unnecessary volume load was avoided during this period.
Table VI. Cumulative postoperative blood loss (ml) 6 hours postop. 12 hours postop. 24 hours postop.
Group I
Group II
Group III
Group IV
373 ± 102 476 ± 126 649 ± 181
394 ± 201 500 ± 230 664 ± 256
509 ± 240 664 ± 282* 890 ± 359*t
378 ± 173 533 ± 220 722 ± 233
*p < 0.01 versus group I. tp < 0.01 versus group II.
be explained by the retransfusion of autologous warm blood as volume substitution at the termination of operation. Only one patient in each of the groups receiving preoperative hemodilution (III and IV) received a homologous transfusion intraoperatively, whereas this was necessary for 21 and 16 patients in groups I and II, respectively. Despite this difference, the course of the hematocrit value after termination of ECC was parallel (Fig. 3). On the other hand, no measurable improvement in the coagulation system could be demonstrated with the transfusion of autologous warm blood; conversely, a significantly higher 24-hour blood loss in group III than in groups I and II was found. This might be explained by the higher amount of colloidal volume substitutes administered to patients in group III during the course of the operation. Hemoseparation with a cell separator allows a rapid elevation of the hematocrit value after termination of ECC without fluid overloading. Despite a lower intra-
operative homologous blood requirement, the hematocrit value of patients in group II (hemoseparation) was markedly higher than that in the group receiving no hemoseparation (group I) 30 minutes after ECC (Fig. 2). Only 15% to 20% of the circulating blood volume is processed with the oxygenator content, hence the protein loss, inevitable with hemoseparation, is not clinically relevant. Because the increase in the hematocrit value is the effect of hemoseparation, this method would be less effective when normovolemic anemia is accepted after termination of ECC. 22 Hemoseparation by means of a locally heparinized collecting system is not indicated in primary coronary operations," inasmuch as the amount of blood collected before and after ECC is usually too small for effective processing. Despite the low quality of postoperative drainage blood (consumed coagulation factors, enhanced fibrinolytic activity,":" low hemoglobin content (76 gm/L in
The Journal of Thoracic and Cardiovascular Surgery
2 1 8 Dietrich et al.
II II
RETRANSfUSION
preop
5' after heparin
ECC 5'
end of
30' after
ECC
ECC
ICU
12 h postop
Fig. 3. The effect of preoperative hemodilution. Only one patient in group III (preoperative hemodilution) received Y2 unit of banked blood intraoperatively, whereas this was necessary in 16 patients in group II (no hemodilution). Nevertheless, the difference in hematocrit value 30 minutes after ECC was negligible.
this investigationj) shed mediastinal blood retransfusion can safely be used as an autologous volume substitute in the early postoperative period, which can postpone or even avoid homologous transfusions. Because of the low hematocrit value and the consumption of coagulation factors, the drainage fluid is solely a volume substitute; in addition, it neither improves the coagulation properties nor has a detrimental effect on coagulation. The shed mediastinal blood has a negligible effect on increasing the hematocrit value, at least with the amounts that on an average were retransfused. As in other studies,":" a reduction in homologous blood requirement with shed mediastinal blood retransfusion was found in this study (833 ± 598 ml in group III versus 408 ± 558 ml in group IV). Although this reduction was marked (50%), it was not considered statistically significant compared with the blood requirement of the group without retransfusion (group III) (p = 0.04). Cosgrove and associates '5 described the preoperative red cell volume as a predictor of the need for transfusion. We also found a significant correlation between red cell volume and blood requirement. The more pronounced decrease in this parameter in groups III and IV is not a sign of acceptance of lower hematocrit values in these groups. It is rather an
indication that in these groups the appropriate volume, that is, autologous blood, was available at the right time. As shown in Figs. 2 and 3, it was possible to increase the hematocrit value by either hemoseparation or preoperative hemodilution after ECC to an equal or even higher level compared with groups I and II, using less homologous blood. Though the preoperative red cell volume was nearly identical in groups I and II and in groups III and IV, respectively, the transfusion reduction between groups I and II was 40% and between groups III and IV was 50%. These findings cannot be explained by an unrecognized difference in red cell volume but only by the application of different blood saving methods. From this study, we conclude that the combination of different blood saving methoos leads to a significant reduction in autologous blood requirement during coronary operations. Intraoperative hemoseparation involves no risks but is not without a certain financial and technical expense. Prebypass isovolumetric hemodilution represents a safe and simple method when there are no contraindications and the patients are closely monitored. Postoperative retransfusion of shed blood (despite the low blood quality) allows an effective volume substitution especially in patients with unexpected large postop-
Volume 97 Number 2 February 1989
erative hemorrhage. All these methods have become standard procedures in our everyday practice. Statistical analysis was performed by Dr. W. Kopcke, Department of Medical Statistics and Data Processing, University Munich. REFERENCES I. Collins JA. Problems associated with the massive transfusion of stored blood. Surgery 1974;75:274-95. 2. Roche JK, Stengle JM. Open-heart surgery and the demand for blood. JAMA 1973;225:1516-21. 3. Schricker KT, Neidhardt B, van der Emde J. Die autologe Bluttransfusion tiefkiihlkonservierten Blutes in der Herzchirurgie. Dtsch Med Wochenschr 1981; 41:1333-7. 4. Toy PTCY, Strauss RG, Stehling LC, et al. Predeposited autologous blood for elective surgery. N Engl J Med 1987;316:517-20. 5. Love TR, Hendren WG, O'Keefe DD, Daggett WM. Transfusion of predonated autologous blood in elective cardiac surgery. Ann Thorac Surg 1987;43:508-12. 6. Lowenstein E. Blood conservation in open heart surgery. Cleve Clin Quart 1981;48:112-25. 7. Kaplan JA, Canarella C, Jones EL, et al. Autologous blood transfusion during cardiac surgery. J THORAC CARDIOVASC SURG 1977;74:4-10. 8. Moran JM, Babka R, Silberman S, et al. Immediate centrifugation of oxygenator contents after cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1978;76: 510-7. 9. Dietrich W, Gob E, Barankay A, Mitto HP, Richter JA. Reduzierung des Fremdblutverbrauches in der Koronarchirurgie durch Harnoseparation und isovolamische Hamodilution, Anaesthesist 1983;32:427-32. 10. Breyer RH, Engelman RM, Rousou JR, Lemeshow SA. A comparison of Cell Saver versus ultrafilter during coronary artery bypass operations. J THORAC CARDIOVASC SURG 1985;90:736-40. 11. Schaff HV, Hauer JM, Bell VR, et al. Autotransfusion of shed mediastinal blood after cardiac surgery: a prospective study. J THORAC CARDIOVASC SURG 1978;75:63241. 12. Breyer RH, Engelman RM, Rousou JA, Lemeshow S. Blood conservation for myocardial revascularization: is it cost-effective? J THORAC CARDIOVASC SURG 1987;93:51222. 13. Weniger J, Shanahan R. Reduction of blood requirements in cardiac surgery. Thorac Cardiovasc Surg 1982;30: 142-6.
Reduction of blood utilization
2 19
14. Cosgrove DM, Amiot DM, Meserko JJ. An improved technique for autotransfusion of shed mediastinal blood. Ann Thorac Surg 1985;40:519-20. 15. Cosgrove DM, Loop FD, Lytle BW, et al. Determinants of blood utilization during myocardial revascularization. Ann Thorac Surg 1985;40:380-4. 16. BMDP statistical software. Dixon WJ, ed. Berkeley, California: University of California Press, 1981. 17. Shevde K, Stigi S, Lee S. The effect of internal mammary artery grafting on the perioperative course. J Cardiothorae Anesth 1988;2:30-3. 18. Messmer K, Sunder-Plassmann L, Jesch F, et al. Oxygen supply to the tissues during limited normovolemic hemodilution. Res Exp Med (Berl) 1973;159:152-66. 19. Weisel RD, Charlesworth DC, Mickleborough LL, et al. Limitations of blood conservation. J THORAC CARDIOVASC SURG 1984;88:26-38. 20. Hagl S, Heimisch W, Meisner H, Mendler N. The effect of hemodilution on regional myocardial function in the presence of coronary stenosis. Basic Res Cardiol 1977;72:344-64. 21. Estafanous FG, Streng Z, Pedvinelli R, Azmy SS, Tarazi RC. Hemodilution affects the pressor response to norepinephrine. J Cardiothorae Anaesth 1987;1:36-41. 22. Cosgrove DM, Thurer RL, Lytle BW, et al. Blood conservation during myocardial revaseularization. Ann Thorae Surg 1979;28:184-9. 23. Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotransfusion following cardiac operations: a randomized prospective study. Ann Thorae Surg 1979;27:500-7. 24. Winton TL, Charrette EJP, Salerno T A. The cell saver during cardiac surgery: does it save? Ann Thorac Surg 1982;33:379-81. 25. Symbas PN. Extraoperative autotransfusion from hematothorax. Surgery 1978;84:722-7. 26. Adan A, Brutel de la Riviere A, Haas F, van Zalk A, de Nooij E. Autotransfusion of drained mediastinal blood after cardiac surgery: a reappraisal. Thorae Cardiovase Surg 1988;36:10-4. 27. Schaff HV, Hauer JM, Gardner TJ, et al. Routine use of autotransfusion following cardiac surgery: experience in 700 patients. Ann Thorae Surg 1979;27:493-9. 28. Cordell AR, Lavender SW. An appraisal of blood salvage techniques in vascular and cardiac operations. Ann Thorae Surg 1981;31:421-5. 29. Johnson RG, Rosenkrantz KR, Preston RA, et al. The efficacy of postoperative autotransfusion in patients undergoing cardiac operations. Ann Thorae Surg 1983;36:173-9.