Transfusion guidelines for cardiovascular surgery: Lessons learned from operations in Jehovah's Witnesses Richard K. Spence, M D , James B. Alexander, M D , A n t h o n y J. DelRossi, M D , Aurel D. Cernaianu, M D , Jonathan Cilley, Jr., M D , M a r k J. Pello, M D , U m u r Atabek, M D , R u d o l p h C. Camishion, M D , and Roger A. Vertrees, C C P ,
Camden, N.J. Patients undergoing cardiovascular surgery are among the top users of homologous blood transfusion (HBT). Awareness Of the risks of disease transmission and immune system modulation from HBT has prompted us to find alternatives such as autologous predonation (APD) and intraoperative autotransfusion (IAT). However, these latter options are not appropriate for all patients. We reviewed our experience with 59 Jehovah's Witness patients who underwent 63 elective cardiovascular procedures without either HBT or APD to determine the safety of operation without these modalities and to develop revised maximum surgical blood-ordering schedule guidelines for cardiovascular surgery. Estimated blood loss averaged 870 ml, but one third to one half of losses were replaced by IAT. IAT was not needed in lower extremity bypass operations in which the estimated blood loss was less than 150 ml. Three of 59 patients died (5.1%), but only one died of operative bleeding complications. We conclude that (1) elective cardiovascular operations can be done safely without the use of either HBT or APD, (2) HBT is not necessary in leg bypass procedures, and (3) maximum surgical blood-ordering schedule guidelines for HBT in major cardiovascular operations can be reduced to near zero by the use of intraoperative autotransfusion and acceptance of a postoperative hemoglobin nadir of 7.0 gm/dl. (J VAS¢ SURG 1992;16:825-31.)
If we are to eliminate the risks associated with homologous blood transfusion, our ultimate goal should be to perform cardiovascular operations with only the patient's own blood. Current maximum surgical blood-ordering schedules recommend typing and crossmatching 3 to 5 traits of homologous blood for major cardiovascular procedures. 13 Both autologous predonation of blood and perioperative autotransfusion have significantly reduced the need for homologous transfusion, but neither approach has completely eliminated its use. Our interest in bloodless cardiovascular surgery developed as part of a program designed to provide surgical care to Jehovah's Witness patients who refuse both homolFrom the Department of Surgery, Cooper Hospital/University Medical Center, Robert Wood Johnson Medical School in Camden, UMDNJ, Camden. Supported in part by National Heart, Lung, and Blood Institute grant R01 # K07HL02532-01A1. Presented at the Sixth Annual Meeting of the Eastern Vascular Society,New York, N.Y., April 30-May 3, 1992. Reprint requests: RichardK. Spence,MD, 3 CooperPlaza, Suite 411, Camden,NJ 08103. 24/6/40968
ogous and autologous predonated blood transfusion. Our success in performing major abdominal procedures without transfusion prompted us to examine our results in cardiovascular operations to determine (1) whether homologous blood use could be eliminated safely and (2) whether autologous predonation of blood was necessary.4 PATIENTS AND METHODS We examined the hospital course and outcome of 59 patients who underwent 63 elective cardiovascular procedures at our institution between the years 1984 and 1991 (Table I). There were 31 women and 28 men; the average age was 63 years (range 41 to 79 years). Elective major vascular procedures were defined as those involving the great vessels or for which the risk of bleeding was high and/or perioperative transfusion was considered likely according to prior experience and maximum surgical bloodordering schedules. We performed all procedures ourselves. Regional anesthesia was used where appropriate. All patients were Jehovah's Witnesses who refused homologous blood transfusion and autolo825
Journal of VASCULAR SURGERY
826 Spence et al.
Table I. Operative procedures Procedure
No.
CABG AVR AAA Leg bypass PC shunt Other
27 8 11 5 4 8
Preop. hemoglobin level in gm/dl (range) 12.8 12.1 I2.6 11.5 5.9 9.6
(9.4-i4.8) (8.4-14.5) (9.4-14.8) (9.7-12.7) (5.6-6.1) (5.2-12.8)
Postop. hemoglobin level in gm/dl (range) 7.3 7.0 8.9 10.5 5.0 8.4
(5.8-9.2) (6.3-8.4) (7.3-10.5) (9.1-12.2) (3.5-5.8) (2.7-12.2)
Blood loss (mO 800 1000 950 150 1600 475
CABG, Coronary artery bypass graft; AAA, aneurysm repair; PC shunt, portocaval shunt; Other, excision of infected graft (1), revision of femoropopliteal graft (3), and amputation (4).
gous predonation of blood on religious grounds. These patients accepted the use of a modified version of a washed-cell autotransfusion device (Electromedics, Englewood, Colo.) that was constructed to provide an unbroken circuit between the patient and the autotransfusion machine (Fig. 1). Intraoperative autotransfusion was not used in the 13 patients who underwent lower extremity bypass grafting, amputation, or infected graft excision. The preoperative hemoglobin level averaged 11.6 gm/dl and ranged between 5.2 and 14.8 gm/dl. The majority of our vascular patients (16 of 23, 70%) were anemic before operation with an average hemoglobin level of 10.1 gm/dl. Three patients with severe anemia (average hemoglobin level 6.3 gm/dl) were treated with a perioperative infusion of a perfluorocarbon compound as part of a clinical trial of this product's safety and efficacy in the treatment of anemia. Three other patients were treated with recombinant erythropoietin injections postoperatively as part of a clinical trial of this drug's role in the treatment of acute perioperative anemia. RESULTS Three (5.1%) of 59 patients died, two after aortic valve replacement and one after portocaval shunt. Only one patient, who underwent aortic valve replacement, died as a result of a combination of poor ventricular function and operative bleeding. The second patient died, on postoperative day 30, of an unexpected arrhythmia after aortic valve replacement. The third patient died 8 days after a portocaval shunt procedure when the shunt apparently occluded and he rapidly became exsanguinated from variceal bleeding. There was no cardiac morbidity in the remaining patients. Estimated blood loss (average 900 ml, range 250 to 2500 ml) was similar for both the cardiac and vascular patient groups (cardiac estimated blood loss 800 ml average, 300 to 2000 ml range; vascular estimated blood loss 955 ml average, 250 to 2500 ml
range). The average postoperative hemoglobin level was 7.4 gm/dl (range 2.7 to 14.5 gm/dl). The decrease in hemoglobin level was twice as great in the cardiac group (average hemoglobin drop 5 gm/dl) compared with that in the vascular group (average hemoglobin drop 2.2 gm/dl). Blood loss was substantially less in those patienst who underwent lower extremity bypass (150 ml average) when compared with that in the remaining vascular and cardiac patients. The average amount of reinfused autologous blood was 360 ml (range 250 to 1000 ml). DISCUSSION
Techniques designed to limit the use of homologous blood transfusion in cardiovascular operations can be divided conveniently into preoperative, intraoperative, and postoperative measures, with the understanding that there is some overlap in each of these areas (Table II). Although autologous predonation is not an option for Jehovah's Wimesses, it can be of great value in others, reducing the need for homologous blood by as much as 73%. S-8 Success depends on a number of factors, including time available, hemoglobin level, patient disease, and cooperation, both from the patient and the physician. Predonation must start at least 1 month before the scheduled operation to provide 3 to 4 units of blood. Unfortunately, this is not always possible or advisable for some patients, especially those with symptomatic heart disease, large aneurysms, and recent variceal bleeding. Anemia (hemoglobin level < 11 gm/dl) eliminates 15% to 20% of patients from consideration. 9 Preoperative therapy with iron and erythropoietin has been shown to increase the number of units donated from 4.1 to 5.4 in a prospective, randomized study of patients scheduled for orthopedic operations, but this drug is not available for widespread use at present. 1° If the patient cannot or will not donate blood before hospitalization, the next best option is to collect and reinfuse the blood in the perioperative
Volume
16
Number
6
December
Bloodless cardiovascular surgery 827
1992
0 I
li
1,O00cc BAG RINGERS AND
0
° o e • • %ej m ~ . -~* e e e ' e
• o•
TRA
ER
EJ NO. 1
30,000u
i
HEPARIN
WASH SOLUTION
,\
/
"j /
O O O O D I Q O l I Q g O
--
~
~
WASTE • '''
'''
PROGRADE PRIME RETROGRADE PRIME
Fig. 1. Modified washed-cell autotransfusion circuit. period (i.e., to use autotransfusion). Two types of autotransfusion systems are in use: those that wash blood and those that do not. The former eliminate the risk ofreinfusion of free hemoglobin, coagulation byproducts, and contaminants contained in plasma, but they are expensive, time-consuming to use, and require technical expertise as compared with simpler, direct reinfusion devices. Both have been used successfully to reduce the need for homologous blood transfusion in major vascular operations) 1-~5 In our Jehovah's Witness patients, we were able to return more than 50% of shed blood with the use of a modified washer system with a continuous circuit to and from the patient (Fig. 1). Because estimated blood loss averaged only 150 ml in the patients undergoing lower extremity bypass, autotransfusion was not needed. We have used hemodilution techniques indirectly during heart operations in Jehovah's Witness patients via the heart-lung machine. In the patient who will allow preoperative removal and processing of blood, hemodilution offers the advantage over autotransfusion of a better blood product. Hemodilution does not activate plasma clotting factors or platelcts as does suctioning blood through a collection wand or circulating it through the heart-lung machine. Platelet sequestration (i.e., the immediate, preoperative separation of whole blood into packed cells and platelet-rich plasma component) can reduce blood
loss and homologous transfusion needs significantly. 16, 1 7 Blood loss can also be reduced by careful attention to operative detail. 18, 10 We prefer to use a midline approach for abdominal aortic operations because it is avascular and minimizes incisional bleeding, although blood loss is similar with the rerroperitoneal approach. 2° The key to preventing blood loss lies more in incorporating Halstedian principles into the approach most familiar to the surgeon, rather than in unquestioning adoption of new procedures. Dissection should be done along avascular planes with the use of cantery when possible. All potentially vascular structures should be clipped or tied before being cut. An intravenous infusion ofvasopressin started before making an incision in portocaval shunt procedures helps constrict dilated vessels. If prosthetic material is required for aortic replacement, woven Dacron fabric grafts result in less blood loss than more porous knitted grafts. 21 Gelatin-sealed and polytetrafluoroethylene materials may offer similar advantagesY Pharmacologic adjuncts hold promise for the future. Both desmopressin and aprotinin have been shown to reduce the amount of blood loss in cardiac operations and, consequently, the need for homologous transfusion. 2s, 24 Three of our severely anemic patients who received the perfluorocarbon compound had temporary increases in dissolved oxygen content, but the effect of the perfluorocarbon was
828
Spence et al.
Table II. Approach to bloodless surgery Preoperative measures Check hemoglobin and hematocrit, iron stores, and nutritional status Replace iron Erythropoietin Limit blood drawing Schedule autologous predonation Erythropoietin Intraoperative measures Hemodilution Autotransfusion Platelet sequestration Halstedian principles Blood substitutes/alternatives Pharmacologic support Heparin/protamine Desmopressin Aprotinin Fibrin glue Lower transfusion trigger Postoperative measures Antotransfusion Limit blood drawing Restore red cell mass Nutritional support Replace iron Erythropoietin Lower transfusion trigger
minimal. 25 Future formulations with higher concentrations and a longer half-life may be more useful. Erythropoietin holds much greater promise as an adjunct in bloodless surgery. Three of our patients who received erythropoietin had significantly higher hematocrits after 1 week compared with those in a group of similar, untreated patients. Postoperative blood conservation measures are primarily continuations of those steps taken preoperatively and intraoperatively. These include reinfusion of shed mediastinal blood, iron restoration, and use of erythropoietin. However, none of these adjuncts should substitute for early reexploration in a patient who is bleeding actively. Postoperative blood samples should be limited to essential studies. By using pediatric collection tubes and returning flush solutions from arterial and central venous lines, blood wastage can be eliminated. Now that postoperative hemoglobin levels of 7 to 9 gm/dl are routinely accepted, few patients can tolerate repeated, unnecessary blood sampling. An absolutely essential component of a bloodless surgery program is the willingness to accept a lower transfusion trigger (i.e., a perioperative hemoglobin level less than 10 gm/dl). Our previous work has shown that patients can safely undergo elective operations with preoperative hemoglobin levels as
Journal of VASCULAR SURGERY
low as 6 gm/dl. 4 The average preoperative hemoglobin level in this group of patients was 11.6 gm/dl with somewhat lower levels in the patients having vascular procedures (hemoglobin level 10.1 gm/dl). There were no intraoperative problems, deaths, or myocardial events from ischemia caused by a low hemoglobin level alone. Our results support the premise that survival in the anemic surgical patient depends more on the amount of blood lost during operation than on the starting hemoglobin level. By collecting and reinfusing shed blood, both absolute blood loss and mortality can be reduced. We realize that this group of Jehovah's Witness patients is unique in refusing to accept both homologous blood transfusion and autologous predonation and in the restrictions they place on the use of intraoperative autotransfusion. Our success, however, should not be attributed to the uniqueness inherent in the group, because these patients do not differ from those treated routinely by cardiovascular surgeons around the country. Our ability to perform major cardiovascular operations with the restrictions described herein is a result of the systematic application of blood-conservation techniques available to all. Our results show that major cardiovascular procedures can be done safely without the use of either homologous blood transfusion or autologons predonation. Not all patients will refuse homologous transfusion as do Jehovah's Witnesses. Nonetheless, the risks of disease transmission and immunomodulation offer clear evidence that we must find alternatives for all of our patients. We believe that bloodless cardiovascular surgery can be offered to the majority of patients if surgeons adopt the principles and use the techniques outlined herein. The willingness to accept a lower transfusion trigger by basing the decision to transfuse on the patient's overall clinical condition, not on a hemoglobin value alone, is essential. Coupling this acceptance with the use of multiple, effective blood conservation techniques tailored to the individual patient's needs will allow us to attain our goal of eliminating unnecessary homologous transfusion. REFERENCES 1. Lowery TA, Clark JA. Successful implementation of a maximum surgical blood ordering schedule. J Med Assoc Ga 1989;78(3): 155-8. 2. Axelrod FB, Pepkowitz SH, Goldfinger D. Establishment of a schedule of optimal preoperative collection of blood. Transfusion 1989;29:677-80. 3. Cosgrove DM, Loop FD, Lytle BN, et al. Determinants of blood utilization during myocadial revascularization. Ann Thorac Surg 1985;40:380-4. 4. Spence RK, Carson JA, Poses I~, et al. Elective surgery
Volume 16 Number 6 December 1992
without transfusion: influence of preoperative hemoglobin level and blood loss on mortality. Am J Surg 1990;159:320-4. 5. Owings DV, Kruskall MS, Thurer ILL, et al. Autologous blood donations prior to elective cardiac surgery. ~AMA 1989;262;1963-8. 6. Britto LW, Easflund DT, Dziuban SW, et al. Predonate autologous blood use in elective cardiac surgery. Ann Thorac Surg 1989;47:529-32. 7. Toy PTCY, Strauss RG, Stehling LG, et al. Predeposited autologous blood for elective surgery: a national multicenter study. N Engl J Med 1987;316:517-20. 8. Goodnough LT, Johnston MFM, Toy PTCA, et al. The variability of transfusion practice in coronary artery bypass surgery. JAMA 1991;265:86-90. 9. The National Blood Resource Education Program Expert Panel. The use of autologous blood. JAMA 1990;263:41447. 10. Goodnough LT. Erythropoietin as a pharmacologic altemafive to blood transfusion in the surgical patient. Transf Mcd Rev 1990;4:299-6. 11. Hailer JW Jr. Minimizing the use of homologous blood products during repair of abdominal aortic aneurysms. Surg Clin North Am 1989;69:817-26. 12. Brown G, Bookallil M, Herkes R. Use of the Cell Saver during elective abdominal aortic aneurysm surgery: influence on transfusion with blood bank- a retrospective survey. Anaesth Intensive Care 1991;19:546-50. 13. Boldt J, Kling D, von Bormann B, et al. Blood conservation in cardiac operations: cell separation versus hemofiltration. J Thorac Cardiovasc Surg 1989;97:832-40. 14. Dietrich W, Barankay A, Dikhey G, et al. Autotransfusion and hemoseparation in cardiac surgery: what can be saved in cardiac reoperations and operations of thoracic aortic aneurysms? Thorac Cardiovasc Surg 1989;37:84-8. 15. Pitrman RD, Inahra T. Eliminating hemologous blood
DISCUSSION Dr. Morris D. Kcrstein (Philadelphia, Pa.). Physicians face a very.special challenge in treating Jehovah's Witnesses. Members o f this faith have a deeply religious conviction, as you know, against accepting homologous or autologous whole blood, packed red cells, white cells, or platelets. However, many will accept the use of non-blood priming agents in heart-lung machines, dialysis machines, or similar equipment. In J A M A some years ago, it was stated that one need not be concerned about habllity, because Jehovah's Witnesses will take adequate legal steps to relieve liability as to their informed refusal o f blood. I applaud my colleagues who presented these data, because I think there are two lessons to be learned. One is that I believe it is appropriate to try to honor the wishes o f the Jehovah's Witnesses and not give them blood or blood products. The second issue is that the physician must knowingly enter into this contract ahead o f time and that the anesthesiologist as well must recognize it. A triangle o f partnership exists that must be addressed up front. I think that this is easy to say and hard to do.
Bloodless cardiovascular surgery 829
16. 17.
18.
19. 20.
21. 22. 23.
24. 25.
transfusions during abdominal aortic aneurysm repair. Am J Surg 1990;159:522-4. DelRossi AJ, Cernaianu AC, Vertrees RA, et al. Platelet-rich plasma reduces postoperative blood loss after cardiopuknonary bypass. J Thorac Cardiovasc Surg 1990;100:281-6. Giordano GF Sr, Girodano GF Jr, Rivers SL, et al. Determinants of homologous blood usage utilizing autologous platelet-rich plasma in cardiac operations. Ann Thorac Surg 1989;47:897-902. Pearlman NW, Stiegmann GV, Vance V, et al. A prospective study of incisional time, blood loss, pain, and healing with carbon dioxide laser, scalpel and electrosurgery. Arch Surg 1991; 126:1018-20. Spence RK. The status of bloodless surgery. Trans Med Rev 1991;4:274-86. Cambria RP, Brewster DC, Abbot WM, et al. Transperitoneal versus retroperitoneal approach to aortic reconstruction: a prospective study. ~ VASCSuRG 1990;11:314-25. Fisher JB, Dennis RC, Valeri CR, et al. Effect of graft material on loss of erythrocytes after aortic operations. SGO 1991; 173:131-6. Reid DB, Pollock JG. A prospective study of 100 gelatinsealed aortic grafts. Ann Vase Surg 1991;5:320-4. Salzman EW, Weinstein My, Wientranb RM, et al. Treatment with desmopressin acetate to reduce blood loss after cardiac surgery: a double blind randomized trial. N Engl J Med 1986;314:1402-6. D'Ambra MN, Risk SC. Aprotinin, erythropoietin, and blood substitutes. Int Anesthes Clin 1990;28:237-40. Spence RK, McCoy S, Costabile J, et al. Fluosol DA-20 in the treatment of severe anemia: randomized controlled study of 46 patients. Crit Care Med 1990;1227-30.
Submitted May 4, 1992; accepted July 13, 1992.
Understandably, not all o f us appreciate the preconceived ideas, desires, and plans o f a person who is a Jehovah's Witness. Though they firmly refuse blood transfusions, I am sure many of my colleagues could relate anecdotes about individuals who changed their minds and who subsequently agreed with the physician in charge to accept blood products. I note that in six patients here, recombinant erythropoietin was used, and in three patients a perfluorocarbon compound was used, so in fact there is treatment being used. Also, when each of us looks at amputation, lower extremity bypass, and removal of infected grafts, I think we can all comfortably say we would agree with the use of bloodless surgery. However, in aortic operations, I think the challenge becomes even more stressful to the surgeon and anesthesiologist. Death in elective operations seems to depend more on estimated blood loss than on preoperative hemoglobin levels. It is my opinion, therefore, that it is easy to approach the elective case; difficult and stressful to approach the
830
Spence et al.
emergency case. Elective operations can be done safely, it is said, in patients with a preoperative hemoglobin level as low as 6 gm/dl. I have the following questions: What happens when the patient has a hemoglobin level less than 6 gm/dl? What happens when the patient has a ruptured aneurysm? What happens even when the patient has other general surgical problems with gastrointestinal bleeding? And finally, what happens in your heart if you see the patient's status worsening during operation? Are you then going to give blood or blood products? Dr. Brian L. Thiele (Hershey, Pa.). I would like to congratulate Dr. Spence and his group for bringing this issue to the fore. I agree wholeheartedly with him that although we are colored by the fact that blood transfusions are lifesaving in life-threatening situations, in large measure we have come to take transfusions a little bit too much for granted so that we just use blood whenever we think we have to. And therein lies one of the important issues that Dr. Spence alluded to, and that is the issue of the traditional transfusion trigger. I will echo the words of numerous people in the literature who say there is absolutely no scientific evidence at all to support the view that when a patient's hemoglobin level is 10 gm/dl or the hematocrit level is 30%, organ function is improved by blood transfusion. In fact, there is a wealth of experimental evidence to say that organ function can continue fairly well unabated down to hematocrit levels in the region of 20% and even as low as 15%. I believe we have become aware of the fact that blood transfusion is not an innocuous problem. And I think we fail to ask ourselves whether there are complications that occur because we give these people blood transfusions. Are there postoperative infections that are related to the administration of blood? I think we should heighten our sensitivity about the use of blood, particularly in this age of AIDS, but also in the bigger picture and the broad issue of what the deleterious effects really are. I would like to make a plea for people to be scrupulous about the surgical technique they use and also to be much more critical about when they use blood. We have had a fairly long experience now at our institution with routine hemodilution at the time of aortic operation, and have observed these patients in the postoperative period, many of them with coronary artery disease, down to hematocrit levels in the region of 20% without serious cardiac sequelae. So there is no question that it can be done, and I applaud Dr. Spence for bringing this important issue to our notice. Dr. Joseph A. Buda (New York, N.Y.). Would you specifically state that transfusions should not be given for a hematocrit level of 30% or above? Dr. Thiele. That is my practice, and my residents know enough not to order a transfusion for a patient with a hematocrit level of 30%. Dr. Buda. Thank you. Dr. William B. Iams (Camp Hill, Pa.). I would like to
Journal of VASCULAR SURGERY
ask Dr. Spence what his lower transfusion trigger is. Does he transfuse at hematocrits of 24%, 22% or 20%? And how does he vary this with the age of the patient, for example, in the patient over 70, or for the patient with chronic obstructive pulmonary disease? Dr. Frederic Jarrett (Pittsburgh, Pa.). I have a comment. I agree absolutely with what Dr. Thiele has said. However, I think the problem very often in the postoperative period is not that the surgeon really believes that the patient with a hematocrit of 28% needs 2 units of packed red cells but rather that another individual caring for the patient orders 2 units of blood and before you can stop it, they have been given. As has been pointed out, blood transfusion is not innocuous treatment and we should be fine-tuned to its complications and circumspect in its use when it is not absolutely necessary. Dr. David Schechter (New York, N.Y.). Because of the large population of Jehovah's Witnesses in New York City, we encounter this problem fairly often. It used to be that the minimum hemoglobin level thought safe was 10 gm/dl. Then 8 became acceptable. N o w we have heard that it is 6 gm/dl! Actually, the ability of some patients to withstand very severe degrees of anemia is astonishing. Nevertheless, when it comes to children I would urge against testing such tolerance and that, instead, needed blood be transfused without the slightest hesitation. Children are considered wards of the state. Regardless of the parents' religious beliefs or desires in the matter, therefore, one should obtain a court order and proceed with blood administration. In emergencies the judge's order may even be given retroactively. I am not aware of any lawsuits arising from such an approach (Schechter DC. Problems relevant to major surgical operations in Jehovah's Witnesses. Am J Surg 1968;116:73-80). Dr. Buda. Is that as true these days, David, as it has been in getting a surrogate judge to give these orders? I do not know if it is for operations or emergency life-saving procedures in New York City. I was not aware that these procedures included transfusion of blood. Dr. Schechter. Yes, it is, barring'some very recent legislation to the contrary. By the way, the largest series of cardiovascular surgical procedures done without blood transfusion were those of Charles Bailey and Denton Cooley. I well recall Dr. Bailey operating in the Bronx on many Jehovah's Witnesses during the 1960s and 1970s. He certainly was very meticulous regarding hemostasis and did not have the benefit of devices such as autotransfusion apparatus or the argon-beam coagulator. There were patients who bled a great deal nevertheless. Where children were concerned there was never any delay in seeking and procuring legal mandate for blood transfusion, and every single time the judge gave a "go ahead." Dr. Buda. Thank you very much. Dr. Richard K. Spence. Dr. Thiele supported our message that this is not primarily a presentation about surgery in Jehovah's Witnesses. The techniques and approaches described are applicable to all patients. Perhaps
Volume 16 Number 6 December 1992
the hardest concept to deal with is the transfusion trigger. As Dr. Thiele told you, the idea of 10 gm/dl for hemoglobin level as being the absolute nadir for transfusion is a myth dating to 1941. It was propounded by Dr. Lundy, an anesthesiologist at the Mayo Clinic, who had little scientific evidence to support his statement. In fact, there was evidence to the contrary that patients did quite well with hemoglobin levels lower than 10 gm/dl. I implore you not to make decisions to transfuse based on a trigger alone. Look at t h e individual patient and the specific setting. We are currently analyzing preoperative measures such as age, cardiac disease, and hemoglobin level to determine their impact on outcome in the patient with anemia who undergoes surgery. I do not have final information for you, but I can tell you that patients who have normal cardiac function appear to tolerate hemodilutions down to a hematocrit o f 20%. Also, we have been using intraoperative transesophageal echocardiography to measure volume status and wall-motion abnormalities. This information may be helpful in determining the need for transfusion. Measuring and calculating oxygen variables in the intensive care unit is useful. Should we not transfuse for a hematocrit of 30%? I think the answer was given, that is, base your decision on the patient's condition. Transfusion is indeed not innocuous. There is clear evidence from the laboratory and from the clinical arena that transfusion produces immunomodulation, leading to an increased risk o f infection. What do we do if the hemoglobin level is less than 6 gm/dl? What do we do about the emergency patient? In emergency cases our approach is to stabilize the patient and move to the operating room as quickly and as safely as possible. You must remember that with a Witness who is bleeding you have no time to sit back, transfuse, and wait to make a decision another day. Early surgery can be lifesaving. Do I give blood to a Witness who is going to die? No. I have come to grips with that decision. If the patient comes to me and says, "I can't, I won't accept blood," and ifI make that contract, then I will not transfuse. I f that means the patient dies, then the patient dies. Fortunately, in my experience I have had only two or three patients who have died who might have been saved by transfusion. When
Bloodless cardiovascular surgery 831
dealing with a Jehovah's Witness, you have to make a moral decision ahead of time as to whether you will agree to and accept the position that the Jehovah's Witness has presented to you, and that is, no blood transfusion. Remember that if you cannot transfuse, you may be able to transfer. I f there are patients with whom you are uncomfortable dealing, send them to someone who will treat them. I am not making a pitch for more patients; what I am saying is, if you cannot handle the situation, send the patient to someone who can. The courts today will allow you to give a transfusion to a minor child when his or her life is endangered. You do not have to wait for a court order, because the law says you can act in an emergency. After the fact, when the child is alive, you can go back and get the approval. However, the courts are getting a little stricter. Most judges will not give you a blanket approval or total control of the child. You will be limited to transfusion at one time for one specific reason, so make sure you have a good one. Dr. Buda. I have one other question for my own education. H o w have you managed to educate your anesthesiologists to work with you with individuals who have a hemoglobin level less than 10 gm/dl or a hematocrit level less than 30%? Dr. Spence. It has been a gradual process. Early on, we got the help o f a couple of people who were brave and believed in the need to limit transfusion. Now, as Dr. Thiele said, your house staff, colleagues, and the like will not let you transfuse at 10 gm/dl without good cause, so it is somewhat easier. Dr. Buda. Well, I would like to send some o f our anesthesiologists down to you for training if you do not mind. Dr. Iams. A Food and Drug Administration circular that came out about a year ago stated that transfusion for a hematocrit of 27% or greater was contraindicated. So, you just refer your anesthesiologist to the F o o d and Drug Administration; if he wants to fight that, that is fine. Dr. Buda. Well, I was thinking in terms o f starting elective or semielective cases at that level and not so much emergency work. Thank you.
Camden, N.J. Patients undergoing cardiovascular surgery are among the top users of homologous blood transfusion (HBT). Awareness Of the risks of disease transmission and immune system modulation from HBT has prompted us to find alternatives such as autologous predonation (APD) and intraoperative autotransfusion (IAT). However, these latter options are not appropriate for all patients. We reviewed our experience with 59 Jehovah's Witness patients who underwent 63 elective cardiovascular procedures without either HBT or APD to determine the safety of operation without these modalities and to develop revised maximum surgical blood-ordering schedule guidelines for cardiovascular surgery. Estimated blood loss averaged 870 ml, but one third to one half of losses were replaced by IAT. IAT was not needed in lower extremity bypass operations in which the estimated blood loss was less than 150 ml. Three of 59 patients died (5.1%), but only one died of operative bleeding complications. We conclude that (1) elective cardiovascular operations can be done safely without the use of either HBT or APD, (2) HBT is not necessary in leg bypass procedures, and (3) maximum surgical blood-ordering schedule guidelines for HBT in major cardiovascular operations can be reduced to near zero by the use of intraoperative autotransfusion and acceptance of a postoperative hemoglobin nadir of 7.0 gm/dl. (J VAS¢ SURG 1992;16:825-31.)
If we are to eliminate the risks associated with homologous blood transfusion, our ultimate goal should be to perform cardiovascular operations with only the patient's own blood. Current maximum surgical blood-ordering schedules recommend typing and crossmatching 3 to 5 traits of homologous blood for major cardiovascular procedures. 13 Both autologous predonation of blood and perioperative autotransfusion have significantly reduced the need for homologous transfusion, but neither approach has completely eliminated its use. Our interest in bloodless cardiovascular surgery developed as part of a program designed to provide surgical care to Jehovah's Witness patients who refuse both homolFrom the Department of Surgery, Cooper Hospital/University Medical Center, Robert Wood Johnson Medical School in Camden, UMDNJ, Camden. Supported in part by National Heart, Lung, and Blood Institute grant R01 # K07HL02532-01A1. Presented at the Sixth Annual Meeting of the Eastern Vascular Society,New York, N.Y., April 30-May 3, 1992. Reprint requests: RichardK. Spence,MD, 3 CooperPlaza, Suite 411, Camden,NJ 08103. 24/6/40968
ogous and autologous predonated blood transfusion. Our success in performing major abdominal procedures without transfusion prompted us to examine our results in cardiovascular operations to determine (1) whether homologous blood use could be eliminated safely and (2) whether autologous predonation of blood was necessary.4 PATIENTS AND METHODS We examined the hospital course and outcome of 59 patients who underwent 63 elective cardiovascular procedures at our institution between the years 1984 and 1991 (Table I). There were 31 women and 28 men; the average age was 63 years (range 41 to 79 years). Elective major vascular procedures were defined as those involving the great vessels or for which the risk of bleeding was high and/or perioperative transfusion was considered likely according to prior experience and maximum surgical bloodordering schedules. We performed all procedures ourselves. Regional anesthesia was used where appropriate. All patients were Jehovah's Witnesses who refused homologous blood transfusion and autolo825
Journal of VASCULAR SURGERY
826 Spence et al.
Table I. Operative procedures Procedure
No.
CABG AVR AAA Leg bypass PC shunt Other
27 8 11 5 4 8
Preop. hemoglobin level in gm/dl (range) 12.8 12.1 I2.6 11.5 5.9 9.6
(9.4-i4.8) (8.4-14.5) (9.4-14.8) (9.7-12.7) (5.6-6.1) (5.2-12.8)
Postop. hemoglobin level in gm/dl (range) 7.3 7.0 8.9 10.5 5.0 8.4
(5.8-9.2) (6.3-8.4) (7.3-10.5) (9.1-12.2) (3.5-5.8) (2.7-12.2)
Blood loss (mO 800 1000 950 150 1600 475
CABG, Coronary artery bypass graft; AAA, aneurysm repair; PC shunt, portocaval shunt; Other, excision of infected graft (1), revision of femoropopliteal graft (3), and amputation (4).
gous predonation of blood on religious grounds. These patients accepted the use of a modified version of a washed-cell autotransfusion device (Electromedics, Englewood, Colo.) that was constructed to provide an unbroken circuit between the patient and the autotransfusion machine (Fig. 1). Intraoperative autotransfusion was not used in the 13 patients who underwent lower extremity bypass grafting, amputation, or infected graft excision. The preoperative hemoglobin level averaged 11.6 gm/dl and ranged between 5.2 and 14.8 gm/dl. The majority of our vascular patients (16 of 23, 70%) were anemic before operation with an average hemoglobin level of 10.1 gm/dl. Three patients with severe anemia (average hemoglobin level 6.3 gm/dl) were treated with a perioperative infusion of a perfluorocarbon compound as part of a clinical trial of this product's safety and efficacy in the treatment of anemia. Three other patients were treated with recombinant erythropoietin injections postoperatively as part of a clinical trial of this drug's role in the treatment of acute perioperative anemia. RESULTS Three (5.1%) of 59 patients died, two after aortic valve replacement and one after portocaval shunt. Only one patient, who underwent aortic valve replacement, died as a result of a combination of poor ventricular function and operative bleeding. The second patient died, on postoperative day 30, of an unexpected arrhythmia after aortic valve replacement. The third patient died 8 days after a portocaval shunt procedure when the shunt apparently occluded and he rapidly became exsanguinated from variceal bleeding. There was no cardiac morbidity in the remaining patients. Estimated blood loss (average 900 ml, range 250 to 2500 ml) was similar for both the cardiac and vascular patient groups (cardiac estimated blood loss 800 ml average, 300 to 2000 ml range; vascular estimated blood loss 955 ml average, 250 to 2500 ml
range). The average postoperative hemoglobin level was 7.4 gm/dl (range 2.7 to 14.5 gm/dl). The decrease in hemoglobin level was twice as great in the cardiac group (average hemoglobin drop 5 gm/dl) compared with that in the vascular group (average hemoglobin drop 2.2 gm/dl). Blood loss was substantially less in those patienst who underwent lower extremity bypass (150 ml average) when compared with that in the remaining vascular and cardiac patients. The average amount of reinfused autologous blood was 360 ml (range 250 to 1000 ml). DISCUSSION
Techniques designed to limit the use of homologous blood transfusion in cardiovascular operations can be divided conveniently into preoperative, intraoperative, and postoperative measures, with the understanding that there is some overlap in each of these areas (Table II). Although autologous predonation is not an option for Jehovah's Wimesses, it can be of great value in others, reducing the need for homologous blood by as much as 73%. S-8 Success depends on a number of factors, including time available, hemoglobin level, patient disease, and cooperation, both from the patient and the physician. Predonation must start at least 1 month before the scheduled operation to provide 3 to 4 units of blood. Unfortunately, this is not always possible or advisable for some patients, especially those with symptomatic heart disease, large aneurysms, and recent variceal bleeding. Anemia (hemoglobin level < 11 gm/dl) eliminates 15% to 20% of patients from consideration. 9 Preoperative therapy with iron and erythropoietin has been shown to increase the number of units donated from 4.1 to 5.4 in a prospective, randomized study of patients scheduled for orthopedic operations, but this drug is not available for widespread use at present. 1° If the patient cannot or will not donate blood before hospitalization, the next best option is to collect and reinfuse the blood in the perioperative
Volume
16
Number
6
December
Bloodless cardiovascular surgery 827
1992
0 I
li
1,O00cc BAG RINGERS AND
0
° o e • • %ej m ~ . -~* e e e ' e
• o•
TRA
ER
EJ NO. 1
30,000u
i
HEPARIN
WASH SOLUTION
,\
/
"j /
O O O O D I Q O l I Q g O
--
~
~
WASTE • '''
'''
PROGRADE PRIME RETROGRADE PRIME
Fig. 1. Modified washed-cell autotransfusion circuit. period (i.e., to use autotransfusion). Two types of autotransfusion systems are in use: those that wash blood and those that do not. The former eliminate the risk ofreinfusion of free hemoglobin, coagulation byproducts, and contaminants contained in plasma, but they are expensive, time-consuming to use, and require technical expertise as compared with simpler, direct reinfusion devices. Both have been used successfully to reduce the need for homologous blood transfusion in major vascular operations) 1-~5 In our Jehovah's Witness patients, we were able to return more than 50% of shed blood with the use of a modified washer system with a continuous circuit to and from the patient (Fig. 1). Because estimated blood loss averaged only 150 ml in the patients undergoing lower extremity bypass, autotransfusion was not needed. We have used hemodilution techniques indirectly during heart operations in Jehovah's Witness patients via the heart-lung machine. In the patient who will allow preoperative removal and processing of blood, hemodilution offers the advantage over autotransfusion of a better blood product. Hemodilution does not activate plasma clotting factors or platelcts as does suctioning blood through a collection wand or circulating it through the heart-lung machine. Platelet sequestration (i.e., the immediate, preoperative separation of whole blood into packed cells and platelet-rich plasma component) can reduce blood
loss and homologous transfusion needs significantly. 16, 1 7 Blood loss can also be reduced by careful attention to operative detail. 18, 10 We prefer to use a midline approach for abdominal aortic operations because it is avascular and minimizes incisional bleeding, although blood loss is similar with the rerroperitoneal approach. 2° The key to preventing blood loss lies more in incorporating Halstedian principles into the approach most familiar to the surgeon, rather than in unquestioning adoption of new procedures. Dissection should be done along avascular planes with the use of cantery when possible. All potentially vascular structures should be clipped or tied before being cut. An intravenous infusion ofvasopressin started before making an incision in portocaval shunt procedures helps constrict dilated vessels. If prosthetic material is required for aortic replacement, woven Dacron fabric grafts result in less blood loss than more porous knitted grafts. 21 Gelatin-sealed and polytetrafluoroethylene materials may offer similar advantagesY Pharmacologic adjuncts hold promise for the future. Both desmopressin and aprotinin have been shown to reduce the amount of blood loss in cardiac operations and, consequently, the need for homologous transfusion. 2s, 24 Three of our severely anemic patients who received the perfluorocarbon compound had temporary increases in dissolved oxygen content, but the effect of the perfluorocarbon was
828
Spence et al.
Table II. Approach to bloodless surgery Preoperative measures Check hemoglobin and hematocrit, iron stores, and nutritional status Replace iron Erythropoietin Limit blood drawing Schedule autologous predonation Erythropoietin Intraoperative measures Hemodilution Autotransfusion Platelet sequestration Halstedian principles Blood substitutes/alternatives Pharmacologic support Heparin/protamine Desmopressin Aprotinin Fibrin glue Lower transfusion trigger Postoperative measures Antotransfusion Limit blood drawing Restore red cell mass Nutritional support Replace iron Erythropoietin Lower transfusion trigger
minimal. 25 Future formulations with higher concentrations and a longer half-life may be more useful. Erythropoietin holds much greater promise as an adjunct in bloodless surgery. Three of our patients who received erythropoietin had significantly higher hematocrits after 1 week compared with those in a group of similar, untreated patients. Postoperative blood conservation measures are primarily continuations of those steps taken preoperatively and intraoperatively. These include reinfusion of shed mediastinal blood, iron restoration, and use of erythropoietin. However, none of these adjuncts should substitute for early reexploration in a patient who is bleeding actively. Postoperative blood samples should be limited to essential studies. By using pediatric collection tubes and returning flush solutions from arterial and central venous lines, blood wastage can be eliminated. Now that postoperative hemoglobin levels of 7 to 9 gm/dl are routinely accepted, few patients can tolerate repeated, unnecessary blood sampling. An absolutely essential component of a bloodless surgery program is the willingness to accept a lower transfusion trigger (i.e., a perioperative hemoglobin level less than 10 gm/dl). Our previous work has shown that patients can safely undergo elective operations with preoperative hemoglobin levels as
Journal of VASCULAR SURGERY
low as 6 gm/dl. 4 The average preoperative hemoglobin level in this group of patients was 11.6 gm/dl with somewhat lower levels in the patients having vascular procedures (hemoglobin level 10.1 gm/dl). There were no intraoperative problems, deaths, or myocardial events from ischemia caused by a low hemoglobin level alone. Our results support the premise that survival in the anemic surgical patient depends more on the amount of blood lost during operation than on the starting hemoglobin level. By collecting and reinfusing shed blood, both absolute blood loss and mortality can be reduced. We realize that this group of Jehovah's Witness patients is unique in refusing to accept both homologous blood transfusion and autologous predonation and in the restrictions they place on the use of intraoperative autotransfusion. Our success, however, should not be attributed to the uniqueness inherent in the group, because these patients do not differ from those treated routinely by cardiovascular surgeons around the country. Our ability to perform major cardiovascular operations with the restrictions described herein is a result of the systematic application of blood-conservation techniques available to all. Our results show that major cardiovascular procedures can be done safely without the use of either homologous blood transfusion or autologons predonation. Not all patients will refuse homologous transfusion as do Jehovah's Witnesses. Nonetheless, the risks of disease transmission and immunomodulation offer clear evidence that we must find alternatives for all of our patients. We believe that bloodless cardiovascular surgery can be offered to the majority of patients if surgeons adopt the principles and use the techniques outlined herein. The willingness to accept a lower transfusion trigger by basing the decision to transfuse on the patient's overall clinical condition, not on a hemoglobin value alone, is essential. Coupling this acceptance with the use of multiple, effective blood conservation techniques tailored to the individual patient's needs will allow us to attain our goal of eliminating unnecessary homologous transfusion. REFERENCES 1. Lowery TA, Clark JA. Successful implementation of a maximum surgical blood ordering schedule. J Med Assoc Ga 1989;78(3): 155-8. 2. Axelrod FB, Pepkowitz SH, Goldfinger D. Establishment of a schedule of optimal preoperative collection of blood. Transfusion 1989;29:677-80. 3. Cosgrove DM, Loop FD, Lytle BN, et al. Determinants of blood utilization during myocadial revascularization. Ann Thorac Surg 1985;40:380-4. 4. Spence RK, Carson JA, Poses I~, et al. Elective surgery
Volume 16 Number 6 December 1992
without transfusion: influence of preoperative hemoglobin level and blood loss on mortality. Am J Surg 1990;159:320-4. 5. Owings DV, Kruskall MS, Thurer ILL, et al. Autologous blood donations prior to elective cardiac surgery. ~AMA 1989;262;1963-8. 6. Britto LW, Easflund DT, Dziuban SW, et al. Predonate autologous blood use in elective cardiac surgery. Ann Thorac Surg 1989;47:529-32. 7. Toy PTCY, Strauss RG, Stehling LG, et al. Predeposited autologous blood for elective surgery: a national multicenter study. N Engl J Med 1987;316:517-20. 8. Goodnough LT, Johnston MFM, Toy PTCA, et al. The variability of transfusion practice in coronary artery bypass surgery. JAMA 1991;265:86-90. 9. The National Blood Resource Education Program Expert Panel. The use of autologous blood. JAMA 1990;263:41447. 10. Goodnough LT. Erythropoietin as a pharmacologic altemafive to blood transfusion in the surgical patient. Transf Mcd Rev 1990;4:299-6. 11. Hailer JW Jr. Minimizing the use of homologous blood products during repair of abdominal aortic aneurysms. Surg Clin North Am 1989;69:817-26. 12. Brown G, Bookallil M, Herkes R. Use of the Cell Saver during elective abdominal aortic aneurysm surgery: influence on transfusion with blood bank- a retrospective survey. Anaesth Intensive Care 1991;19:546-50. 13. Boldt J, Kling D, von Bormann B, et al. Blood conservation in cardiac operations: cell separation versus hemofiltration. J Thorac Cardiovasc Surg 1989;97:832-40. 14. Dietrich W, Barankay A, Dikhey G, et al. Autotransfusion and hemoseparation in cardiac surgery: what can be saved in cardiac reoperations and operations of thoracic aortic aneurysms? Thorac Cardiovasc Surg 1989;37:84-8. 15. Pitrman RD, Inahra T. Eliminating hemologous blood
DISCUSSION Dr. Morris D. Kcrstein (Philadelphia, Pa.). Physicians face a very.special challenge in treating Jehovah's Witnesses. Members o f this faith have a deeply religious conviction, as you know, against accepting homologous or autologous whole blood, packed red cells, white cells, or platelets. However, many will accept the use of non-blood priming agents in heart-lung machines, dialysis machines, or similar equipment. In J A M A some years ago, it was stated that one need not be concerned about habllity, because Jehovah's Witnesses will take adequate legal steps to relieve liability as to their informed refusal o f blood. I applaud my colleagues who presented these data, because I think there are two lessons to be learned. One is that I believe it is appropriate to try to honor the wishes o f the Jehovah's Witnesses and not give them blood or blood products. The second issue is that the physician must knowingly enter into this contract ahead o f time and that the anesthesiologist as well must recognize it. A triangle o f partnership exists that must be addressed up front. I think that this is easy to say and hard to do.
Bloodless cardiovascular surgery 829
16. 17.
18.
19. 20.
21. 22. 23.
24. 25.
transfusions during abdominal aortic aneurysm repair. Am J Surg 1990;159:522-4. DelRossi AJ, Cernaianu AC, Vertrees RA, et al. Platelet-rich plasma reduces postoperative blood loss after cardiopuknonary bypass. J Thorac Cardiovasc Surg 1990;100:281-6. Giordano GF Sr, Girodano GF Jr, Rivers SL, et al. Determinants of homologous blood usage utilizing autologous platelet-rich plasma in cardiac operations. Ann Thorac Surg 1989;47:897-902. Pearlman NW, Stiegmann GV, Vance V, et al. A prospective study of incisional time, blood loss, pain, and healing with carbon dioxide laser, scalpel and electrosurgery. Arch Surg 1991; 126:1018-20. Spence RK. The status of bloodless surgery. Trans Med Rev 1991;4:274-86. Cambria RP, Brewster DC, Abbot WM, et al. Transperitoneal versus retroperitoneal approach to aortic reconstruction: a prospective study. ~ VASCSuRG 1990;11:314-25. Fisher JB, Dennis RC, Valeri CR, et al. Effect of graft material on loss of erythrocytes after aortic operations. SGO 1991; 173:131-6. Reid DB, Pollock JG. A prospective study of 100 gelatinsealed aortic grafts. Ann Vase Surg 1991;5:320-4. Salzman EW, Weinstein My, Wientranb RM, et al. Treatment with desmopressin acetate to reduce blood loss after cardiac surgery: a double blind randomized trial. N Engl J Med 1986;314:1402-6. D'Ambra MN, Risk SC. Aprotinin, erythropoietin, and blood substitutes. Int Anesthes Clin 1990;28:237-40. Spence RK, McCoy S, Costabile J, et al. Fluosol DA-20 in the treatment of severe anemia: randomized controlled study of 46 patients. Crit Care Med 1990;1227-30.
Submitted May 4, 1992; accepted July 13, 1992.
Understandably, not all o f us appreciate the preconceived ideas, desires, and plans o f a person who is a Jehovah's Witness. Though they firmly refuse blood transfusions, I am sure many of my colleagues could relate anecdotes about individuals who changed their minds and who subsequently agreed with the physician in charge to accept blood products. I note that in six patients here, recombinant erythropoietin was used, and in three patients a perfluorocarbon compound was used, so in fact there is treatment being used. Also, when each of us looks at amputation, lower extremity bypass, and removal of infected grafts, I think we can all comfortably say we would agree with the use of bloodless surgery. However, in aortic operations, I think the challenge becomes even more stressful to the surgeon and anesthesiologist. Death in elective operations seems to depend more on estimated blood loss than on preoperative hemoglobin levels. It is my opinion, therefore, that it is easy to approach the elective case; difficult and stressful to approach the
830
Spence et al.
emergency case. Elective operations can be done safely, it is said, in patients with a preoperative hemoglobin level as low as 6 gm/dl. I have the following questions: What happens when the patient has a hemoglobin level less than 6 gm/dl? What happens when the patient has a ruptured aneurysm? What happens even when the patient has other general surgical problems with gastrointestinal bleeding? And finally, what happens in your heart if you see the patient's status worsening during operation? Are you then going to give blood or blood products? Dr. Brian L. Thiele (Hershey, Pa.). I would like to congratulate Dr. Spence and his group for bringing this issue to the fore. I agree wholeheartedly with him that although we are colored by the fact that blood transfusions are lifesaving in life-threatening situations, in large measure we have come to take transfusions a little bit too much for granted so that we just use blood whenever we think we have to. And therein lies one of the important issues that Dr. Spence alluded to, and that is the issue of the traditional transfusion trigger. I will echo the words of numerous people in the literature who say there is absolutely no scientific evidence at all to support the view that when a patient's hemoglobin level is 10 gm/dl or the hematocrit level is 30%, organ function is improved by blood transfusion. In fact, there is a wealth of experimental evidence to say that organ function can continue fairly well unabated down to hematocrit levels in the region of 20% and even as low as 15%. I believe we have become aware of the fact that blood transfusion is not an innocuous problem. And I think we fail to ask ourselves whether there are complications that occur because we give these people blood transfusions. Are there postoperative infections that are related to the administration of blood? I think we should heighten our sensitivity about the use of blood, particularly in this age of AIDS, but also in the bigger picture and the broad issue of what the deleterious effects really are. I would like to make a plea for people to be scrupulous about the surgical technique they use and also to be much more critical about when they use blood. We have had a fairly long experience now at our institution with routine hemodilution at the time of aortic operation, and have observed these patients in the postoperative period, many of them with coronary artery disease, down to hematocrit levels in the region of 20% without serious cardiac sequelae. So there is no question that it can be done, and I applaud Dr. Spence for bringing this important issue to our notice. Dr. Joseph A. Buda (New York, N.Y.). Would you specifically state that transfusions should not be given for a hematocrit level of 30% or above? Dr. Thiele. That is my practice, and my residents know enough not to order a transfusion for a patient with a hematocrit level of 30%. Dr. Buda. Thank you. Dr. William B. Iams (Camp Hill, Pa.). I would like to
Journal of VASCULAR SURGERY
ask Dr. Spence what his lower transfusion trigger is. Does he transfuse at hematocrits of 24%, 22% or 20%? And how does he vary this with the age of the patient, for example, in the patient over 70, or for the patient with chronic obstructive pulmonary disease? Dr. Frederic Jarrett (Pittsburgh, Pa.). I have a comment. I agree absolutely with what Dr. Thiele has said. However, I think the problem very often in the postoperative period is not that the surgeon really believes that the patient with a hematocrit of 28% needs 2 units of packed red cells but rather that another individual caring for the patient orders 2 units of blood and before you can stop it, they have been given. As has been pointed out, blood transfusion is not innocuous treatment and we should be fine-tuned to its complications and circumspect in its use when it is not absolutely necessary. Dr. David Schechter (New York, N.Y.). Because of the large population of Jehovah's Witnesses in New York City, we encounter this problem fairly often. It used to be that the minimum hemoglobin level thought safe was 10 gm/dl. Then 8 became acceptable. N o w we have heard that it is 6 gm/dl! Actually, the ability of some patients to withstand very severe degrees of anemia is astonishing. Nevertheless, when it comes to children I would urge against testing such tolerance and that, instead, needed blood be transfused without the slightest hesitation. Children are considered wards of the state. Regardless of the parents' religious beliefs or desires in the matter, therefore, one should obtain a court order and proceed with blood administration. In emergencies the judge's order may even be given retroactively. I am not aware of any lawsuits arising from such an approach (Schechter DC. Problems relevant to major surgical operations in Jehovah's Witnesses. Am J Surg 1968;116:73-80). Dr. Buda. Is that as true these days, David, as it has been in getting a surrogate judge to give these orders? I do not know if it is for operations or emergency life-saving procedures in New York City. I was not aware that these procedures included transfusion of blood. Dr. Schechter. Yes, it is, barring'some very recent legislation to the contrary. By the way, the largest series of cardiovascular surgical procedures done without blood transfusion were those of Charles Bailey and Denton Cooley. I well recall Dr. Bailey operating in the Bronx on many Jehovah's Witnesses during the 1960s and 1970s. He certainly was very meticulous regarding hemostasis and did not have the benefit of devices such as autotransfusion apparatus or the argon-beam coagulator. There were patients who bled a great deal nevertheless. Where children were concerned there was never any delay in seeking and procuring legal mandate for blood transfusion, and every single time the judge gave a "go ahead." Dr. Buda. Thank you very much. Dr. Richard K. Spence. Dr. Thiele supported our message that this is not primarily a presentation about surgery in Jehovah's Witnesses. The techniques and approaches described are applicable to all patients. Perhaps
Volume 16 Number 6 December 1992
the hardest concept to deal with is the transfusion trigger. As Dr. Thiele told you, the idea of 10 gm/dl for hemoglobin level as being the absolute nadir for transfusion is a myth dating to 1941. It was propounded by Dr. Lundy, an anesthesiologist at the Mayo Clinic, who had little scientific evidence to support his statement. In fact, there was evidence to the contrary that patients did quite well with hemoglobin levels lower than 10 gm/dl. I implore you not to make decisions to transfuse based on a trigger alone. Look at t h e individual patient and the specific setting. We are currently analyzing preoperative measures such as age, cardiac disease, and hemoglobin level to determine their impact on outcome in the patient with anemia who undergoes surgery. I do not have final information for you, but I can tell you that patients who have normal cardiac function appear to tolerate hemodilutions down to a hematocrit o f 20%. Also, we have been using intraoperative transesophageal echocardiography to measure volume status and wall-motion abnormalities. This information may be helpful in determining the need for transfusion. Measuring and calculating oxygen variables in the intensive care unit is useful. Should we not transfuse for a hematocrit of 30%? I think the answer was given, that is, base your decision on the patient's condition. Transfusion is indeed not innocuous. There is clear evidence from the laboratory and from the clinical arena that transfusion produces immunomodulation, leading to an increased risk o f infection. What do we do if the hemoglobin level is less than 6 gm/dl? What do we do about the emergency patient? In emergency cases our approach is to stabilize the patient and move to the operating room as quickly and as safely as possible. You must remember that with a Witness who is bleeding you have no time to sit back, transfuse, and wait to make a decision another day. Early surgery can be lifesaving. Do I give blood to a Witness who is going to die? No. I have come to grips with that decision. If the patient comes to me and says, "I can't, I won't accept blood," and ifI make that contract, then I will not transfuse. I f that means the patient dies, then the patient dies. Fortunately, in my experience I have had only two or three patients who have died who might have been saved by transfusion. When
Bloodless cardiovascular surgery 831
dealing with a Jehovah's Witness, you have to make a moral decision ahead of time as to whether you will agree to and accept the position that the Jehovah's Witness has presented to you, and that is, no blood transfusion. Remember that if you cannot transfuse, you may be able to transfer. I f there are patients with whom you are uncomfortable dealing, send them to someone who will treat them. I am not making a pitch for more patients; what I am saying is, if you cannot handle the situation, send the patient to someone who can. The courts today will allow you to give a transfusion to a minor child when his or her life is endangered. You do not have to wait for a court order, because the law says you can act in an emergency. After the fact, when the child is alive, you can go back and get the approval. However, the courts are getting a little stricter. Most judges will not give you a blanket approval or total control of the child. You will be limited to transfusion at one time for one specific reason, so make sure you have a good one. Dr. Buda. I have one other question for my own education. H o w have you managed to educate your anesthesiologists to work with you with individuals who have a hemoglobin level less than 10 gm/dl or a hematocrit level less than 30%? Dr. Spence. It has been a gradual process. Early on, we got the help o f a couple of people who were brave and believed in the need to limit transfusion. Now, as Dr. Thiele said, your house staff, colleagues, and the like will not let you transfuse at 10 gm/dl without good cause, so it is somewhat easier. Dr. Buda. Well, I would like to send some o f our anesthesiologists down to you for training if you do not mind. Dr. Iams. A Food and Drug Administration circular that came out about a year ago stated that transfusion for a hematocrit of 27% or greater was contraindicated. So, you just refer your anesthesiologist to the F o o d and Drug Administration; if he wants to fight that, that is fine. Dr. Buda. Well, I was thinking in terms o f starting elective or semielective cases at that level and not so much emergency work. Thank you.
