| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Refresher Courses - Sunday June 10 |
From the Departments of Anesthesiology, Centre Hospitalier de l'Université de Montréal and Institut de Cardiologie de Montréal, Université de Montréal, Montreal, Quebec, Canada.
Address correspondence to: Dr. Jean-Francois Hardy, CHUM Hôtel Dieu de Montréal, Pavillon de Boullion, local 6-521, 3840 rue St-Urbain, Montréal, Quebec, Canada H2W 1T8. Phone: 514-8908000, ext. 15578; Fax: 514-412-7143; E-mail: jean-francois.hardy{at}umontreal.ca
MORE than 13 million units of red blood cells (RBC) are collected in the United States and over 11 million units are transfused to 3.4 million patients annually. It is estimated that 60–70% of these RBC units are transfused in the perioperative period. Such considerable numbers are in sharp contrast with the little knowledge we have on the true benefits of erythrocyte transfusions. In this refresher course lecture, we will discuss the risks of anemia and the benefits of RBC transfusions in the perioperative context. For a more detailed and formal review of the literature, the reader is referred to a recently published text by the same authors.1
Conceptually, the risk/benefit ratio of erythrocyte transfusions must take three factors into account: 1) the risks secondary to anemia which depend, in turn, on the patient's capacity to compensate for it; 2) the capacity of allogeneic RBC to correct these risks, a consideration which is all too often assumed but has not been well demonstrated; 3) the risks of transfusions themselves. While the infectious risks of blood products, specially HIV, have brought about a major reconsideration of the way we should be administering transfusions, the risks associated with the immunomodulating effects of blood products, as well as the "classic" complications (volume overload, ABO incompatibility, etc.) of RBC transfusions should be kept in mind. Given the inherent complexity of this risk/benefit analysis, it becomes rapidly apparent that any attempt to define a universal "transfusion trigger" is overly simplistic and unrealistic from a scientific standpoint.
The analogy with a "fever trigger" illustrates why the attempt to define a "transfusion trigger" is doomed to fail clinically. Attempting to define a "fever trigger", i.e., the fixed body temperature above which penicillin should be administered would, nowadays, be considered absurd. Yet, this is the approach adopted by several when transfusing RBC. Erythrocytes (penicillin) are administered for anemia (fever) without knowing if the underlying disease will respond to the elected therapeutic approach. Adopting a universal "fever trigger" would certainly cure a few individuals, but would needlessly expose a considerable number of patients to penicillin and result in a significant number of untoward events, allergic or other.
Erythrocyte transfusions: the arguments
In contrast with the paucity of objective data on the risks of anemia or the benefits of RBC transfusions, several arguments have been put forward either to support or to condone erythrocyte transfusions in the perioperative period.
Rationale in favour of RBC transfusions to maintain a high hemoglobin concentration
Classically, several arguments have supported the use of RBC transfusions to maintain a high hemoglobin concentration ([Hb]; above 100 g•L–1):
Rationale against RBC transfusions to maintain a high hemoglobin concentration
On the other hand, several arguments oppose the use of erythrocyte transfusions to maintain a high [Hb]:
Arguments are interesting and help formulate research hypotheses, but what data do we have? In fact, only nine randomized clinical trials on the benefits of transfusions have been published to this day (Table
).4–12 Thus, most of the evidence in favour or against erythrocyte transfusions is indirect. In the following sections, we will examine the effects of anemia and RBC transfusions, first on mortality and, subsequently, on morbidity.
|
Anemia and mortality: the Jehovah's Witnesses' experience
Ideally, randomized controlled trials comparing transfusion to no transfusion in various patient populations undergoing different medical and surgical procedures should be conducted to determine the risk/benefit ratios of erythrocyte transfusions. Such trials are extremely difficult to conduct. The best available surrogate trials are those reported on surgical outcomes of patients refusing transfusions for religious reasons.
The study by Viele and Weiskopf identified 61 reports on 4722 Jehovah's Witnesses in which some or all patients presented a [Hb] 
80 g•L–1. All patients whose death was related to anemia died with a [Hb] below 50 g•L–1, except for three patients with cardiac disease.13 Also, deaths frequently occurred in the context of active bleeding. Conversely, 27 patients survived with a [Hb] 50 g•L–1, the lowest [Hb] reported for a non-transfused survivor being 14 g•L–1.14 While these data have significant limitations they suggest that, with the possible exception of cardiac surgery patients, most patients survive without transfusion at a [Hb] >50 g•L–1.
In 1996, Carson et al. reported the results of a retrospective cohort study performed in 1958 adult surgical patients declining blood transfusions for religious reasons. Risk of death increased from 1.3% when preoperative [Hb] was more than 120 g•L–1 to 33.3% when preoperative [Hb] was less than 60 g•L–1. Furthermore, risk of death was significantly more pronounced in patients with cardiovascular disease (history of angina, myocardial infarction, congestive heart failure or peripheral vascular disease).15
Mortality associated with anemia is associated not only with preoperative [Hb], but also with intraoperative blood losses (Figure 1). Risk of death is the highest in patients with cardiovascular disease and a decline of [Hb] $40 g•L–1.16 Previously, Spence et al. had reported a similar increase in mortality when blood losses exceeded 500 ml.17
|
Effect of allogeneic erythrocyte transfusion on short-term mortality
In one of the very few randomized controlled trials on the effect of allogeneic RBC transfusion, Hebert et al. demonstrated that maintaining [Hb] at a high, predetermined level in critically ill patients is associated with increased mortality.12 In patients with initial [Hb] <90 g•L–1 within 72 hr after admission in the intensive care unit, maintaining [Hb] between 70 and 90 g•L–1 (restrictive transfusion strategy) was associated with significantly lower mortality rates in patients with an APACHE II score #20 and in patients <55 yr of age, compared with maintaining [Hb] $100 g•L–1 (liberal transfusion strategy). Overall, in-hospital mortality rates were lower in the restrictive strategy group. Transfusion strategy did not influence 30-day mortality nor mortality in patients with significant cardiovascular disease. However, this last result may have been influenced by study design, attending physicians having declined enrollment in a greater proportion of such patients.12
In another attempt to define the [Hb] at which patients should receive a blood transfusion, Carson et al. reviewed a total of 8787 consecutive patients aged 60 or older undergoing surgical repair of a hip fracture. In this elderly population, perioperative transfusion in patients with a [Hb] $80 g•L–1 did not modify the risk of 30- or 90- day mortality. At lower [Hb], 90.5% of patients were transfused, preventing analysis of the association between transfusion and mortality.18
A pilot study of 84 hip fracture patients by the same authors appears to support the hypothesis that RBC transfusion for symptoms of anemia or a [Hb] <80 g•L–1 does not result in significant impairment of either rehabilitation, morbidity or mortality, compared to transfusion to maintain a threshold [Hb] of 100 g•L–1.11
Benefits of transfusion and HIV endemic
The vast majority of studies published on the risks and benefits of transfusion are based on North American or European information and data for the different complications associated with various therapeutic interventions, including transfusions. It must be realized that the environment leading to a transfusion decision may vary considerably in different parts of the world.19 Thus, in a context where healthy blood donors have an HIV positivity rate of 5–20% and supplies of test reagents to screen blood may not be maintained, the short-term benefit of transfusion on mortality is rapidly outweighed by the risk of HIV infection.
Effect of allogeneic erythrocyte transfusion on long-term mortality
Transfusion induced immunosuppression may be responsible for increased mortality after cancer surgery. Perioperative allogeneic RBC transfusion was associated with reduced survival for patients with stage III carcinoma of the esophagus in the study by Craig et al.20 Similarly, the meta-analysis by Chung et al. including 20 trials representing 5236 patients with colorectal carcinoma supported the hypothesis that perioperative blood transfusion increases the risk of disease recurrence, cancer death and death, with an estimated cumulative odds ratio of a negative outcome of 1.69.21
Influence of the age of transfused RBC
There is increasing suspicion in the medical community that the age of transfused erythrocytes may be, at least in part and for reasons that remain unclear, responsible for increased mortality in critically ill patients. In patients with severe sepsis, total number of RBC units transfused was not different in survivors and non-survivors while median age of RBC units administered during sepsis was lower in survivors compared to non-survivors (17 vs 25 days respectively).22 If these conclusions are confirmed by future studies and in other patient populations, blood transfusion practices will no doubt be affected radically.
The effects of anemia and transfusions on morbidity
Anemia and morbidity: the Jehovah's Witnesses' experience
A number of cases of extreme hemodilution, usually following surgical hemorrhage, have been reported. The lowest reported [Hb] associated with a successful outcome in the absence of transfusion appears to be 14 g•L–1, but the patient required intubation and intensive care for 14 days before being transferred to the ward.14 On the other hand, transfusion of allogeneic RBC allowed recovery from acidosis and extubation as early as ten hours postoperatively despite a nadir [Hb] of 11 g•L–1.23 Thus, transfusion of allogeneic RBC allows a much more rapid discontinuation of measures initiated to sustain life during severe anemia (mechanical ventilation of the lungs, high inspired O2 concentration, sedation and muscle paralysis, hypothermia, etc.).
Anemia, transfusion and cardiovascular morbidity
In the presence of anemia, clinicians are very concerned with the possibility of cardiovascular morbidity. Controlled and progressive anemia occurs in patients with cardiac disease undergoing phlebotomy to donate autologous blood prior to elective open heart operations. While this situation poses a theoretical risk of aggravating the underlying cardiac disease, results of most clinical investigations conclude that this type of anemia is usually well tolerated clinically.24 The absence of significant untoward effects of anemia in cardiac patients admitted to predonate blood may be explained by the careful evaluation and selection of populations and by the limited extent of anemia, most programs excluding patients when the hematocrit drops below 32%.
Hemodilution has been studied quite extensively in humans. Weiskopf et al. induced acute isovolemic reduction of [Hb] to 50 g•L–1 in healthy patients prior to anesthesia and in volunteers not undergoing surgery and concluded that this degree of anemia did not result in inadequate O2 transport.25 Myocardial ischemia occurred infrequently and could not be related to [Hb]. Fontana et al. studied profound hemodilution (average [Hb] 30 g•L–1) in children. ST segment depression was observed in one patient ([Hb] 21 g•L–1) and resolved immediately upon reinfusion of autologous blood.26
In elderly patients without known cardiac disease, isovolemic hemodilution to an average [Hb] of 88 g•L–1 is well tolerated, without electrocardiographic signs of myocardial ischemia.27 In patients with coronary artery disease who are receiving chronic beta blocker therapy, isovolemic hemodilution to an average [Hb] of 99 g•L–1 is well tolerated also, irrespective of age (35–81 yr) or left ventricular ejection fraction (26–83%).28
The effect of anemia on cardiovascular morbidity after cardiac surgery is somewhat controversial. A hematocrit <28% was associated with myocardial ischemia and morbid cardiac events in the observational study by Nelson et al.29 Of note, only 27 patients were included and all patients with morbid cardiac events had also been transfused, rendering the differentiation between the effects of anemia and those of transfusion impossible. Spiess et al. concluded that a high hematocrit value (34%) upon entering the intensive care unit after myocardial revascularization was associated with a higher incidence of Q-wave myocardial infarction and was an independent predictor of infarction.30 Thus, it may be prudent to transfuse patients with cardiovascular disease when the hematocrit falls below 28%, but achieving normal (greater than 40%) hematocrit values should not be attempted.31 Clinicians must realize that the cardiovascular benefit of prophylactic RBC transfusion has never been demonstrated.
Alternatively, since there appears to be some (clinical and experimental) evidence that transfusion of RBC (in small amounts) can revert ischemic changes and restore normal myocardial function, we suggest that, in patients with suspected or proven cardiovascular disease, myocardial ischemia be monitored closely and incremental (unit by unit) transfusion of RBC be performed until cardiac function is normalized, rather than adopt a fixed transfusion threshold for all such patients.
Anemia, transfusion and non-cardiovascular morbidity
A number of clinical trials have suggested that oxygen delivery should be maintained at high levels to minimize tissue hypoxia and improve survival in critically ill patients. Consequently, it has become customary to maintain a [Hb] of at least 100 g•L–1 in these patients. While intuitively attractive, this assumption is not supported by solid evidence. The meta analysis by Heyland et al. suggests that, in general, interventions destined to achieve supraphysiologic goals of cardiac index, DO2 and VO2 are ineffective, with the possible exception of patients in whom the therapy is initiated preoperatively.32
In the landmark trial of transfusion requirements in critical care presented earlier, multiple organ dysfunction score was marginally improved and number of organs failing and length of stay in the intensive care unit or in the hospital were unaffected by maintaining [Hb] between 70 and 90 g•L–1 compared to maintaining [Hb] >100 g•L–1 as is the usual practice in most institutions.12 Thus, in light of the available evidence, transfusion of allogeneic RBC to maintain a [Hb] >100 g•L–1 is no longer acceptable and may, in fact, be deleterious in critically ill patients.
Transfusion of allogeneic RBC and hemostasis
An often ignored effect of RBC transfusion is the improvement of hemostatic function. Transfusion of RBC normalizes the bleeding time in anemic thrombocytopenic patients despite persistent thrombocytopenia. Similarly, RBC shorten the bleeding time and control the hemorrhagic diathesis of uremic patients.
More recently, erythrocytes have been shown to modulate biochemical and functional responsiveness of activated platelets, suggesting that erythrocytes contribute to thrombosis and hemostasis and supporting the concept that thrombus formation is a multicellular event. Another mechanism by which erythrocytes modulate hemostasis is the rheological effect of red cells on the margination of platelets (Figure 2), thereby optimizing their interaction with the injured endothelium. During profound normovolemic hemodilution, abnormal hemostasis develops prior to compromise of global tissue oxygenation33 suggesting that hemodilution may be limited more by preservation of normal coagulation.
|
Benefits of transfusion in various patient populations
In ten anemic patients with chronic obstructive pulmonary disease (COPD), transfusion of allogeneic RBC to a target [Hb] of 110–120 g•L–1 reduced minute ventilation and the work of breathing in the study by Schönhofer et al.34 Confirmation of a beneficial effect of RBC transfusion in patients with severe COPD remains to be confirmed in a large scale study before recommending high target [Hb] in these patients.
In patients with acute severe gastrointestinal hemorrhage, early transfusion of at least two units of allogeneic RBC above a [Hb] of 80 g•L–1 increased rebleeding and the need for surgery in the study by Blair et al.5
Perioperative blood transfusion increased the risk of disease recurrence after resection for colorectal carcinoma in the meta-analysis by Chung et al.21 Blood sparing strategies that avoid the use of blood products altogether should probably be recommended in these patients.
Transfusion requirements for RBC were associated with decreased graft and patient survival in adult patients undergoing hepatic transplantation.35
Requirement for Cell-saver (>five units) and for more than five units of RBC were associated with an increased risk of acute postoperative renal failure in thoracic and thoracoabdominal aortic surgery in the study by Godet et al.36
Finally, in patients with sickle cell disease, a conservative regimen designed to increase [Hb] >100 g•L–1 was as effective as the aggressive transfusion regimen designed to decrease the hemoglobin S level to less than 30%.9 Patients managed conservatively were less transfused (2.5 units vs 5.0 units of RBC on average) and the incidence of transfusion related complications (mainly alloimmunization and hemolytic reactions) was halved. This is one of the very few randomized clinical trials contrasting transfusion strategies with sufficient power to rule out clinically important differences in its primary outcome, perioperative sickle cell crises in this case.
Summary and conclusions
Most contemporary transfusion practices are not based on solid evidence. Overall, anemia increases the risk of mortality only when [Hb] falls below 50 g•L–1. This effect is magnified by blood losses and cardiovascular disease, indirectly suggesting that transfusions are life saving in this context. No evidence supports a beneficial effect of transfusions in less sick patients and those without cardiovascular disease when [Hb] is above 70 g•L–1.
Specially in areas of the world where blood transmissible diseases are endemic, the considerable risks of HIV infection nearly always outweigh the benefits of transfusion, even in severely anemic patients.
Definitive evidence that minimal [Hb] greater than 90–100 g•L–1 in patients with cardiovascular disease prevents cardiac morbidity is lacking. Therefore, symptomatic transfusion therapy is suggested. Transfusions to improve oxygen transport are definitely not indicated in critically ill adults. Transfusions may be deleterious in patients with several, non-cardiovascular diagnoses. Conversely, in anemic shock, transfusions allow early withdrawal of supportive therapy. Erythrocyte transfusions improve hemostasis. They reduce perioperative complications in patients with sickle cell disease.
Transfusion therapy must be goal oriented and individualized to achieve, in each patient under our care, the optimal balance between the benefits and the risks of allogeneic erythrocyte administration.
References
1 Hardy JF, Bélisle S. The benefits of allogeneic erythrocyte transfusions: what evidence do we have? In: NATA, (Ed.). Transfusion Medicine and Alternatives to Blood Transfusion. Paris: R and J - Editions Médicales, 2000: 48-59.
2 Contreras M, Barbara JA. Infections related to red cell transfusions including variant Creutzfeldt-Jakob disease. Transfusion Alternatives in Transfusion Medicine (TATM) 2000; 2: 5–12.
3 Heiss MM. Immunosuppression and transfusion. Transfusion Alternatives in Transfusion Medicine (TATM) 2000; 2: 13–5.
4 Weisel RD, Charlesworth DC, Mickleborough LL, et al. Limitations of blood conservation. J Thorac Cardiovasc Surg 1984; 88: 26–38.[Abstract]
5 Blair SD, Janvrin SB, McCollum CN, Greenhalgh RM. Effect of early blood transfusion on gastrointestinal haemorrhage. Br J Surg 1986; 73: 783–5.[Medline]
6 Fortune JB, Feustel PJ, Saifi J, Stratton HH, Newell JC, Shah DM. Influence of hematocrit on cardiopulmonary function after acute hemorrhage. J Trauma 1987; 27: 243–9.[Medline]
7 Johnson RG, Thurer RL, Kruskall MS, et al. Comparison of two transfusion strategies after elective operations for myocardial revascularization. J Thorac Cardiovasc Surg 1992; 104: 307–14.[Abstract]
8 Hebert PC, Wells G, Marshall J, et al. Transfusion requirements in critical care. A pilot study. Canadian Critical Care Trials Group. JAMA 1995; 273: 1439–44.[Abstract]
9
Vichinsky EP, Haberkern CM, Neumayr L, et al. A comparison of conservative and aggressive transfusion regimens in the perioperative management of sickle cell disease. The Preoperative Transfusion in Sickle Cell Disease Study Group. N Engl J Med 1995; 333: 206–13.
10 Bush RL, Pevec WC, Holcroft JW. A prospective, randomized trial limiting perioperative red blood cell transfusions in vascular patients. Am J Surg 1997; 174: 143–8.[Medline]
11 Carson JL, Terrin ML, Barton FB, et al. A pilot randomized trial comparing symptomatic vs. hemoglobin- level-driven red blood cell transfusions following hip fracture. Transfusion 1998; 38: 522–9.[Medline]
12
Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999; 340: 409–17.
13 Viele MK, Weiskopf RB. What can we learn about the need for transfusion from patients who refuse blood? The experience with Jehovah's Witnesses. Transfusion 1994; 34: 396–401.[Medline]
14 Brimacombe J, Skippen P, Talbutt P. Acute anaemia to a haemoglobin of 14 g.l-1 with survival. Anaesth Intensive Care 1991; 19: 581–3.[Medline]
15 Carson JL, Duff A, Poses RM, et al. Effect of anaemia and cardiovascular disease on surgical mortality and morbidity. Lancet 1996; 348: 1055–60.[Medline]
16 Carson JL, Poses RM, Spence RK, Bonavita G. Severity of anaemia and operative mortality and morbidity. Lancet 1988; 1: 727–9.[Medline]
17 Spence RK, Carson JA, Poses R, et al. Elective surgery without transfusion: influence of preoperative hemoglobin level and blood loss on mortality. Am J Surg 1990; 159: 320–4.[Medline]
18
Carson JL, Duff A, Berlin JA, et al. Perioperative blood transfusion and postoperative mortality. JAMA 1998; 279: 199–205.
19 Lackritz EM, Ruebush TKd, Zucker JR, Adungosi JE, Were JB, Campbell CC. Blood transfusion practices and blood-banking services in a Kenyan hospital. Aids 1993; 7: 995–9.[Medline]
20
Craig SR, Adam DJ, Yap PL, et al. Effect of blood transfusion on survival after esophagogastrectomy for carcinoma. Ann Thorac Surg 1998; 66: 356–61.
21 Chung M, Steinmetz OK, Gordon PH. Perioperative blood transfusion and outcome after resection for colorectal carcinoma. Br J Surg 1993; 80: 427–32.[Medline]
22
Purdy FR, Tweeddale MG, Merrick PM. Association of mortality with age of blood transfused in septic ICU patients. Can J Anaesth 1997; 44: 1256–61.
23 Zollinger A, Hager P, Singer T, Friedl HP, Pasch T, Spahn DR. Extreme hemodilution due to massive blood loss in tumor surgery. Anesthesiology 1997; 87: 985–7.[Medline]
24
Hardy JF, Bélisle S, Décary F. Cardiac surgical patients must not be denied the benefits of autologous blood predonation. Can J Anaesth 1994; 41: 1021–6.
25
Weiskopf RB, Viele MK, Feiner J, et al. Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA 1998; 279: 217–21.
26 Fontana JL, Welborn L, Mongan PD, Sturm P, Martin G, Bünger R. Oxygen consumption and cardiovascular function in children during profound intraoperative normovolemic hemodilution. Anesth Analg 1995; 80: 219–25.[Abstract]
27 Spahn DR, Zollinger A, Schlumpf RB, et al. Hemodilution tolerance in elderly patients without known cardiac disease. Anesth Analg 1996; 82: 681–6.[Abstract]
28 Spahn DR, Schmid ER, Seifert B, Pasch T. Hemodilution tolerance in patients with coronary artery disease who are receiving chronic ß-adrenergic blocker therapy. Anesth Analg 1996; 82: 687–94.[Abstract]
29 Nelson AH, Fleisher LA, Rosenbaum SH. Relationship between postoperative anemia and cardiac morbidity in high-risk vascular patients in the intensive care unit. Crit Care Med 1993; 21: 860–6.[Medline]
30
Spiess BD, Ley C, Body SC, et al. Hematocrit value on intensive care unit entry influences the frequency of Q-wave myocardial infarction after coronary artery bypass grafting. The Institutions of the Multicenter Study of Perioperative Ischemia (McSPI) Research Group. J Thorac Cardiovasc Surg 1998; 116: 460–7.
31 Wahr JA. Risks associated with under-transfusion. Transfusion Alternatives in Transfusion Medicine (TATM) 2000; 2: 21–6.
32 Heyland DK, Cook DJ, King D, Kernerman P, Brun-Buisson C. Maximizing oxygen delivery in critically ill patients: a methodologic appraisal of the evidence. Crit Care Med 1996; 24: 517–24.[Medline]
33 McLoughlin TM, Fontana JL, Alving B, Mongan PD, Bunger R. Profound normovolemic hemodilution: hemostatic effects in patients and in a porcine model. Anesth Analg 1996; 83: 459–65.[Abstract]
34 Schönhofer B, Wenzel M, Geibel M, Kohler D. Blood transfusion and lung function in chronically anemic patients with severe chronic obstructive pulmonary disease. Crit Care Med 1998; 26: 1824–8.[Medline]
35
Cacciarelli TV, Keeffe EB, Moore DH, et al. Effect of intraoperative blood transfusion on patient outcome in hepatic transplantation. Arch Surg 1999; 134: 25–9.
36 Godet G, Fleron MH, Vicaut E, et al. Risk factors for acute postoperative renal failure in thoracic or thoracoabdominal aortic surgery: a prospective study. Anesth Analg 1997; 85: 1227–32.[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
