Anemia is a potent risk factor for mortality and morbidity in surgical patients, and its management has begun to shift away from allogeneic blood transfusion in recent years. This article reviews the clinical importance of perioperative anemia, the role and shortcomings of blood transfusion, and the pros and cons of alternative approaches to managing perioperative anemia. I conclude with an overview of a program for perioperative blood product use at my institution, Cleveland Clinic.
SIGNIFICANCE OF PERIOPERATIVE ANEMIA
Prevalence depends on many factors
The reported prevalence of anemia in surgical patients varies widely—from 5% to 76%1—and depends on the patient’s disease and comorbidities, the surgical procedure and associated blood loss, and the definition of anemia used. The prevalence of preoperative anemia increases with patient age and is higher in women than in men.2
A multiplier of risk
Anemia is an important multiplier of mortality risk. For example, the presence of anemia raises the relative risk of 2-year mortality from 2.05 to 3.37 in patients with chronic kidney disease, from 2.86 to 3.78 in patients with heart failure, and from 4.86 to 6.07 in patients with concomitant heart failure and chronic kidney disease.3
Adverse effects of anemia have been demonstrated specifically in the perioperative setting as well. A large retrospective cohort study showed that a preoperative hemoglobin concentration of less than 6 g/dL increases the risk of death 30 days after surgery by a factor of 26 relative to a concentration of 12 g/dL or greater in surgical patients who declined blood transfusion for religious reasons.4 The anemia-associated mortality risk was especially pronounced among patients with cardiovascular disease.4 Other studies have demonstrated perioperative anemia to be associated with increases in the risk of death,5 cardiac events,6 pneumonia,7 and postoperative delirium.8
IS BLOOD TRANSFUSION THE ANSWER
The use of allogeneic blood transfusion to manage anemia and blood loss is a concept that originated several centuries ago and has changed little over the years.
Blood supply challenges
Blood collection has historically lagged demand, resulting in a blood supply insufficient to meet transfusion needs. According to the federal government’s 2007 National Blood Collection and Utilization Survey Report, 6.89% of US hospitals reported that they cancelled elective surgery on 1 or more days in the prior year because of a lack of blood availability, and 13.5% experienced at least 1 day in which nonsurgical blood needs could not be met.9 Unless practices are changed to increase blood donation, these unmet tranfusion needs may grow.
Joint Commission set to measure blood management
In response to this challenge, an advisory panel formed by the Joint Commission has identified 17 performance measures related to blood conservation and appropriate transfusion.10 These measures are currently in development, and we expect to see some types of metrics in the near future. Such metrics are likely to further prioritize blood management for US hospitals.
Safety of the blood supply: Viral transmission down, TRALI risk persists
The safety of the blood supply has improved markedly. Sophisticated testing and public demand have led to a dramatic decline in the risk of transfusion-related transmission of HIV, hepatitis C virus, and hepatitis B virus.11
Despite this progress, the risk of transfusion-related acute lung injury (TRALI) has persisted in recent years. TRALI is characterized by acute onset of noncardiogenic pulmonary edema within 6 hours of blood product transfusion. Believed to be immune-mediated, TRALI is thought to occur as antibodies to human leukocyte antigens develop, inducing capillary leak syndrome.12 The patients most commonly affected are those who receive plasma from multiparous female donors. A recent evaluation of transfusion-related fatalities reported to the US Food and Drug Administration (FDA) revealed a continual rise in fatal TRALI cases in the United States from 2001 to 2006.13–15 TRALI was implicated in more than half of all transfusion-related fatalities reported to the FDA in 2006, a higher number than for any other single cause.13
At the same time, there is evidence that hemovigilance can reduce TRALI risk. In the United Kingdom, the Serious Hazards of Transfusion Steering Group introduced in late 2003 a policy of using plasma from male donors as much as possible, in view of the association of TRALI with plasma from multiparous female donors. The effort appeared to pay off: whereas TRALI accounted for 6.8% of all transfusion-related adverse events reported in the United Kingdom during the period 1996–2003,16 this proportion declined to just 1.9% in 2006.17
Finally, despite the progress in screening blood for more established infections like HIV and the hepatitis viruses, some additional infections now must be considered when assessing blood supply safety. These include diseases newly recognized as being transmissible by blood, or for which blood donor screening is not currently available, or that are newly emergent infections for which the potential for spread by transfusion is unknown. For such diseases—which include malaria and West Nile virus—the risk of transmission through transfusion is low, as they are much more likely to be acquired by other means.
Transfusion and outcomes: Not a strong record
Transfusion has never undergone safety and efficacy evaluation by the FDA. Given the challenges of conducting a randomized study of transfusion in the perioperative setting, we may never have high-quality data to assess transfusion in this setting.
A few studies merit mention, however. The Transfusion Requirement in Critical Care (TRICC) trial was conducted in 838 critically ill patients in the intensive care setting.18 Patients were randomized to a strategy of either liberal transfusion (begun when hemoglobin fell below 10 g/dL) or restrictive transfusion (begun when hemoglobin fell below 7 g/dL). Thirty-day mortality was similar between patients in the two strategy groups, but the restrictive strategy was associated with significantly lower mortality in at least two subgroups: patients with myocardial infarction and patients with pulmonary edema. Further subgroup analysis found no benefit of early or aggressive transfusion in patients with coronary artery disease or in those requiring mechanical ventilation.
Rao et al performed a meta-analysis of three large international trials of patients with acute coronary syndromes to determine whether blood transfusion to correct anemia in this setting was associated with improved survival.19 They found significantly higher mortality among patients who underwent transfusion compared with those who did not, prompting them to urge caution in the use of transfusion to maintain arbitrary hematocrit levels in stable patients with ischemic heart disease.
Similarly, a risk-adjusted, propensity-matched analysis of 6,301 patients undergoing noncardiac surgery found that receipt of 4 U of blood or more was a predictor of greater mortality, higher risk of infection, and longer hospital stay.20 Moreover, in an observational cohort study of 11,963 patients who underwent isolated coronary artery bypass graft surgery, each unit of red blood cells transfused was associated with an incrementally increased risk of adverse outcome (eg, mortality, renal injury, need for ventilator support, lengthened hospital stay, infection).21 The latter study found that transfusion was the single factor most reliably associated with increased risk of postoperative morbidity.
Additional studies have echoed these findings—ie, that perioperative blood transfusion has been associated with a host of adverse outcomes, including increased morbidity and length of stay, increased rates of postoperative infection, as well as immunosuppression, viral transmission, and acute transfusion reactions.5,22,23