espite decades of effort, red blood cell (RBC) transfusion practice remains suboptimal. Large variations in the indications for and timing of RBC transfusion have been documented among coronary artery bypass graft (CABG) surgery patients that are not explained by patient or surgical variables, but rather by differences in provider and institutional preferences.¹

This variation persists despite the availability of practice guidelines. One of the oldest transfusion triggers is the “10/30” rule, which originated from comments made by Adams and Lundy in 1942.² Several transfusion guidelines have been published more recently based on the best available evidence. While medical guidelines are believed to be an efficacious method to improve patient care, they have been ineffective in reducing unwarranted transfusions for three reasons.

First, a prescribed hemoglobin trigger is not appropriate for all patients and clinical settings because a consistent physiologic deterioration is not observed among all patients as the hemoglobin falls. Second, many physicians remain unaware of these transfusion guidelines. Finally, there really has not been a clear understanding of the risks of anemia relative to the risks and potential benefit of RBC transfusion.

Risks of Anemia
There are numerous reports of severe anemia being well-tolerated in healthy subjects. Acute normovolemic hemodilutional anemia has been safely performed with animal models in dogs and baboons as well as with human subjects with and without surgery. Data from patients who decline RBC transfusion for religious reasons suggest that mortality is more related to substantial blood loss than a low preoperative hematocrit. The effect was significantly more pronounced among patients with cardiovascular disease.³

Studies from several prospective observational cardiac surgical databases have reported the association of hemodilutional anemia during cardiopulmonary bypass (CPB) and an increased risk of renal failure, stroke and mortality during CABG surgery. Plausible explanations for these observations include injury as a result of exposure to hemodilutional anemia or to intraoperative RBC transfusions administered as treatment for anemia. A recent report by the Northern New England Cardiovascular Disease Study Group observed that among patients managed without intraoperative RBC transfusion, exposure to hemodilutional anemia during CPB was associated with increased need for prolonged inotropes, post-CPB intra-aortic balloon pumps and return to CPB after initial separation.⁴ These observations support the concept that intraoperative anemia reduces the oxygen supply available to the tissues to adequately meet demand, leading to ischemic tissue injury and subsequent adverse outcomes.

Risks of RBC Transfusion
During the 1990s, the risks of RBC transfusion seemed to be well-characterized. For example viruses such as cytomegalovirus, hepatitis C, hepatitis B, HIV and HTLV can be transmitted by RBC transfusions. Evidence has been accumulating more recently, however, that RBC transfusions are complex biologic products capable of initiating a systemic inflammatory response, inducing nonspecific immunosuppression and perhaps occluding local microvasculature, causing local tissue hypoxemia. Observational evidence to support immunomodulation by RBC transfusions includes: 1) improved renal transplant outcome; 2) increased risk of cancer recurrence and postoperative infection; and 3) increased risk of acute respiratory distress syndrome and multiorgan failure among patients previously exposed to RBC transfusions.

Benefits of Transfusion
Several studies evaluating transfusion in adults with critical illness sepsis and acute coronary syndromes have been published and will be briefly reviewed.

The first large, prospective, randomized trial of transfusion therapy in critically ill patients without active bleeding was published seven years ago.⁵ The Transfusion Requirements in Critical Care, or TRICC, trial evaluated a restrictive transfusion strategy maintaining hemoglobin between 7 and 9 gm/dL versus a liberal strategy maintaining hemoglobin between 10 and 12 gm/dL. Inclusion criteria included anemic euvolemic patients who were not actively bleeding. Patients with chronic anemia or following cardiac surgery were excluded, and a large number of patients with significant coronary artery disease were not enrolled in the study at the discretion of the attending physician.

This study showed that the restrictive strategy was “at least as effective as and possibly superior to a liberal transfusion strategy.” Furthermore, subgroup analysis showed an association of improved 30-day survival in patients younger than 55 years old or those with APACHE II scores lower than 20 managed with the restrictive strategy.

Another subgroup analysis of 357 patients with cardiovascular disease showed no difference in mortality rates between the restrictive and liberal strategies for this subgroup.⁶ A trend for decreased survival was observed, however, for patients in the restrictive group with the diagnosis of acute coronary syndromes (ACS) [ACS, acute myocardial infarction (AMI) or unstable angina]. Because of these findings, the authors stated that a restrictive transfusion strategy “appears to be safe in most critically ill patients with cardiovascular disease, with the possible exception of patients with AMI and unstable angina.”

There are three observational studies that provide some further insight of treatment of anemia among patients with acute coronary syndromes. Wu et al. retrospectively analyzed 78,974 Medicare beneficiaries hospitalized with AMI.⁷ Anemia on admission was associated with increased 30-day mortality, and transfusion of patients with hematocrit less than 30 percent was associated with improved survival. Rao et al. found different results when studying 24,112 patients with ACS who were prospectively enrolled in three trials (GUSTO IIb, PURSUIT and PARAGON B).⁸

This retrospective analysis of prospectively collected data showed an association between increased 30-day mortality and transfusion, which was significant for nadir hematocrit as low as 25 percent. This suggests that a nadir hematocrit as low as 25 may be tolerated in otherwise stable patients with AMI. The authors, however, caution that this data should not be used to change practice due to its retrospective nature. Finally Yang et al. retrospectively evaluated the effect of transfusion among 74,241 patients with ACS and also showed that patients who were transfused were associated with a higher risk of death or reinfarction.⁹

Together these observations provide conflicting results; therefore a prospective trial of transfusion among patients presenting with acute coronary syndromes needs to be done. Until then variation in RBC transfusion practice among this important population will most likely persist.

There is one other randomized trial that provides some evidence regarding the role of RBC transfusion as part of early goal-directed therapy for the treatment of sepsis or septic shock. Rivers et al. randomized septic patients to either standard resuscitation or an explicit goal-directed protocol.¹⁰ RBC transfusions were indicated in the goal-directed protocol to maintain central venous oxygen saturation (ScvO₂) > 70 percent, if the hematocrit was < 30 percent. Patients in the early goal-directed therapy group required significantly more fluid, transfusions and inotropic therapy and had higher hematocrit than the standard therapy group. Patients in the early goal-directed group experienced superior hospital and 28-day and 60-day mortality compared to those patients managed with standard resuscitation. Because there were multiple interventions used in this protocol, it is not possible to separate the relative importance of RBC transfusion to the survival benefit.

How to Improve?
Since transfusion is not without risk and the “triggers” remain controversial, every effort should be made to minimize blood loss (use of blood conservation techniques) and optimize patients prior to and following surgery (use of erythropoietin, iron, etc.). Even taking this approach, though, transfusion may be needed. Unfortunately a single “transfusion trigger” cannot be applied to all patients. Instead the decision to transfuse needs to be based on several factors, including rate and amount of ongoing bleeding, acute versus chronic anemia and possibly physiologic triggers.

In acute hemorrhagic shock transfusion, decisions should be based on the rate and amount of hemorrhage. For euvolemic patients who are not actively bleeding, maintaining the hemoglobin between 7-9 g/dL is as safe as hemoglobin between 10-12 g/dL and, in fact, may be superior among patients younger than 55 years old or with APACHE II scores less than 20.5 For patients in the early resuscitation phase of sepsis or septic shock, maintaining the hematocrit greater than 30 percent is reasonable if the response to fluids and inotropic therapy is not adequate.

For patients with significant cardiovascular disease, transfusion strategy is more controversial. In patients with a history of cardiovascular disease, but without an acute coronary syndrome, maintaining the hemoglobin between 7 to 9 g/dL appears to be safe. The management of anemia in patients with acute coronary syndromes, however, remains confusing at best, and firm recommendations will have to await prospective randomized trials.

Clearly these trials and observations among critically ill patients have advanced our knowledge regarding the transfusion management of specific populations of patients, many of whom frequent both operating rooms and critical care units. Transfusion, though, is controversial in large patient populations such as cardiac surgery and acute coronary syndromes. In these patients, transfusion decisions based on the risks of anemia versus the risks and benefits of transfusion will be made at the bedside and, for now, remain part of “the art of medicine.”

References:
1. Surgenor DM, Churcill EL, Wallace WH, et al. Determinants of red cell, platelet, plasma and cryoprecipitate transfusions during coronary artery bypass graft surgery: The Collaborative Hospital Transfusion Study. Transfusion. 1996; 36:521-532.
2. Adams RC, Lundy JS. Anesthesia in cases of poor risk. Some suggestions for decreasing the risk. Surg Gynecol Obstet. 1942; 74:1011-1101.
3. Carson JL, Spence RK, Poses RM. Severity of anaemia and operative mortality and morbidity. Lancet. 1988; 1:727-729.
4. Surgenor SD, DeFoe GR, Fillinger Likosky DS, et al. Intraoperative red blood cell transfusion during CABG surgery increases the risk of post-operative low output heart failure. Circulation. 2005. In Press.
i Hebert PC, Wellis G, Blajchman MA, et al. A multicentered, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999; 340:409-417.
ii Hebert PC, Yetisir E, Martin C, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med. 2001; 29:227-234.
iii Wu W, Rathore SS, Wang Y, et al. Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med. 201; 345:1230-1236.
iv Rao SV, Jollis JG, Harrington RA. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA. 2004; 292:1555-1562.
v Yang X, Alexander KP, Chen AY, et al. The implications of blood transfusions for patients with non-ST-segment elevation acute coronary syndromes. J Am Coll Cardiol. 2005; 46:1490-1495.
vi Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of sever sepsis and septic shock. N Engl J Med. 2001;

Stephen D. Surgenor, M.D., M.S., is Associate Professor and Chief of Critical Care Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire.

Michael H. Wall, M.D., F.C.C.M., is Associate Professor, Vice-Chair for Clinical Affairs and Director of Cardiothoracic Anesthesiology, University of Texas Southwestern Medical Center, Dallas, Texas.