Disclosures: Jeffrey L Carson, MD Nothing to disclose. Steven Kleinman, MD Consultant/Advisory Boards: Creative Testing Solutions [blood donor laboratory testing]; Cerus [pathogen inactivation of blood components (platelet blood system, plasma blood system)]. Arthur J Silvergleid, MD Nothing to disclose. Jennifer S Tirnauer, MD Nothing to disclose.
Contributor disclosures are reviewed for conflicts of interest by the editorial group. When found, these are addressed by vetting through a multi-level review process, and through requirements for references to be provided to support the content. Appropriately referenced content is required of all authors and must conform to UpToDate standards of evidence.
INTRODUCTION — For many decades, the decision to transfuse red blood cells was based upon the "10/30 rule": transfusion was used to maintain a blood hemoglobin (Hgb) concentration above 10 g/dL (100 g/L) and a hematocrit above 30 percent . However, concern regarding transmission of blood-borne pathogens and efforts at cost containment caused a re-examination of transfusion practices in the 1980s. The 1988 National Institutes of Health Consensus Conference on Perioperative Red Blood Cell Transfusions suggested that no single criterion should be used as an indication for red cell component therapy, and that multiple factors related to the patient's clinical status and oxygen delivery needs should be considered . During the subsequent 25 years, a large body of clinical evidence was generated, resulting in the publication of many guidelines for red blood cell (RBC) transfusion in different settings [3-9]. A common theme of these guidelines is the need to balance the benefit of treating anemia with the desire to avoid unnecessary transfusion, with its associated costs and potential harms. This requires considerable diagnostic skill and acumen on the part of physicians ordering transfusions. At this time, laboratory and diagnostic tests have not demonstrated sufficient precision for providing data on which to base red cell transfusion decisions.
As blood transfusion practices are evaluated in randomized trials, we are increasingly able to emphasize clinical trial data, since these data provide the best evidence to guide transfusion decisions.
The indications and thresholds for RBC transfusion in adults will be reviewed here. General aspects of RBC collection, storage, safety, and administration, as well as practices for some special populations, are presented separately.
●(See "Massive blood transfusion".)
RATIONALE FOR TRANSFUSION
Role of blood in oxygen delivery — Blood delivers oxygen to the tissues, and the vast majority of oxygen delivered is bound to hemoglobin (Hgb) in red blood cells. Thus, anemia has the potential to reduce oxygen delivery. However, most patients are able to increase tissue oxygen delivery by increasing cardiac output over a range of Hgb concentrations. The major physiologic considerations relevant to anemic patients are the degree to which oxygen delivery to the tissues is adequate and whether compensatory mechanisms for maintaining oxygen delivery will become overwhelmed or deleterious . (See "Oxygen delivery and consumption".)
Oxygen delivery (DO2) is determined by the formula:
DO2 = cardiac output x arterial oxygen content
In healthy patients, DO2 can be raised by increasing cardiac output (increased heart rate in conscious patients and increased stroke volume in anesthetized patients). In critically ill patients, DO2 may become more dependent on arterial oxygen content, and oxygen utilization may become pathologically dependent upon DO2. This pathologic dependence may be due to elevated arterial lactate concentrations and a change in the slope of the oxygen extraction ratio. Determining what Hgb level is adequate in individual clinical scenarios has been the goal of a large number of clinical studies and randomized trials. (See 'Asymptomatic hospitalized patient' below.)
At rest, there is a large reserve in oxygen delivery, since the rate of delivery normally exceeds consumption by a factor of four. Thus, if intravascular volume is maintained during bleeding and cardiovascular status is not impaired, oxygen delivery theoretically will be adequate until the hematocrit falls below 10 percent because greater cardiac output, rightward shift of the oxygen-hemoglobin dissociation curve, and increased oxygen extraction can compensate for the decrease in arterial oxygen content (table 1).
These predictions were confirmed in a study in which healthy resting individuals underwent acute isovolemic reduction of their Hgb to 5 g/dL (equivalent to a hematocrit of approximately 15 percent) . Though some individuals did develop electrocardiogram (ECG) changes consistent with myocardial ischemia, there was little evidence of inadequate oxygen delivery, and the fall in Hgb was associated with progressive increases in stroke volume and heart rate (and therefore cardiac output), and a progressive reduction in the systemic vascular resistance. Heart rate was found to increase linearly in response to the acute isovolemic anemia . Of note, cognitive function measured by reaction time and immediate memory was impaired when the hemoglobin concentration was reduced to 5 to 6 g/dL .
The preceding considerations represent the optimal clinical response in healthy adults. However, blood transfusion is usually administered to patients who are ill with underlying comorbidities, and there is concern that compensatory mechanisms may be impaired in critically ill patients, particularly in patients with underlying cardiovascular disease. It has been argued in the past that this might justify prophylactic transfusion to maintain a Hgb of 10 g/dL. However, data in favor of this hemoglobin target level are sparse. To the contrary, multicenter randomized controlled trials indicate that compared with a target Hgb of 10 g/dL, target Hgb values of 7 to 8 g/dL are associated with equivalent or better outcomes in many patient populations. (See 'Transfusion thresholds' below.)
Impact of anemia on morbidity and mortality — While numerous observational studies have shown an association between anemia and increased mortality, it is not clear that correction of anemia will improve mortality. The following studies illustrate the deleterious effect of severe postoperative anemia:
●In a study of 1958 patients who declined blood transfusion for religious reasons, the odds of death rose as the preoperative hemoglobin (Hgb) fell, and the odds of death were much greater in patients with underlying cardiovascular disease .
●In a subset analysis of 300 postoperative patients, a Hgb between 7 and 8 g/dL appeared to have no immediate adverse effect on mortality, whereas there was a clear risk of postoperative death when the hemoglobin fell below 7 g/dL . In this study, 30-day in-hospital mortality for patients with various postoperative hemoglobin levels was as follows:
•Hgb 7.1 to 8.0 (n = 99) – Zero percent
•Hgb 5.1 to 7.0 (n = 110) – 9 percent
•Hgb 3.1 to 5.0 (n = 60) – 30 percent
•Hgb ≤3.0 (n = 31) – 64 percent
●A retrospective database review of 310,311 veterans >65 years of age undergoing non-cardiac surgery evaluated the association of preoperative anemia with mortality or cardiac events . The adjusted odds of death or cardiac events correlated inversely with the preoperative hematocrit. Even mild anemia (HCT 36.0 to 38.9) was associated with a 10 percent increase in events; this rose to a 52 percent increased risk with more severe anemia (HCT 18.0 to 20.9).
●In the FOCUS trial, which randomly assigned patients with anemia to liberal versus restrictive transfusion, overall mortality in patients requiring a transfusion was 42 percent (841 of 2016 patients) during approximately three years of observation; the median age of these patients was 82 years .
While these and other studies suggest that severe anemia is associated with poor outcome [17,18], data from randomized trials have shown that more aggressive correction of anemia does not necessarily improve these outcomes. Clinical trials are needed to establish whether anemia is merely a marker for more severe underlying disease or a direct cause of poor outcomes.
RISKS AND COMPLICATIONS OF TRANSFUSION — The risks and potential long term complications of red blood cell transfusion, and strategies to minimize these risks and complications, are discussed separately. These include the following:
●Infection is a risk of transfusion since transfusion-transmitted pathogens (eg, viruses, bacteria, and parasites) can be transmitted if they are present in donor blood and if they escape detection by screening assays. In addition, some studies have reported that transfusion-mediated immunosuppression may lead to increased risk of postoperative bacterial infection. (See "Transfusion transmitted bacterial infection" and "Transfusion transmitted HIV infection and AIDS" and "Epidemiology and transmission of hepatitis C virus infection", section on 'Blood transfusion' and "Epidemiology, transmission, and prevention of hepatitis B virus infection", section on 'Transfusion' and "Leukoreduction to prevent complications of blood transfusion", section on 'Potential adverse effects of transfused leukocytes'.)
●Allergic and immune transfusion reactions can occur in any patient, and are more common in multiply-transfused patients. (See "Immunologic blood transfusion reactions" and "Transfusion-associated immune and non immune-mediated hemolysis" and "Transfusion-related acute lung injury (TRALI)".)
●Volume overload is typically a concern in the elderly, small children, and those with compromised cardiac function. (See "Transfusion reactions caused by chemical and physical agents".)
●Hyperkalemia from potassium released from red blood cells during blood bank storage is primarily a concern in massive transfusion, impaired renal function, and infants/newborns. (See "Red blood cell transfusion in adults: Storage, specialized modifications, and infusion parameters", section on 'Potassium leakage'.)
●Iron overload becomes a concern after a large number of transfusions for chronic anemia . (See 'Ambulatory patient' below and "Iron overload syndromes other than hereditary hemochromatosis", section on 'Transfusional iron overload'.)
Society guidelines — Transfusion guidelines have been published by the following societies:
●American Society of Anesthesiology 
●British Committee for Standards in Hematology 
●Australian and New Zealand Society of Blood Transfusion 
●Eastern Association for Surgery of Trauma (EAST) and the American College of Critical Care Medicine of the Society of Critical Care Medicine (SCCM) 
●European Society of Cardiology (ESC) 
●Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists 
●AABB (formerly the American Association of Blood Banks) 
●American College of Physicians 
In general, the different guidelines have recommended that transfusion is not indicated for Hgb >10 g/dL, but the lower threshold varies from 6 g/dL to 8 g/dL. As an example, the AABB guidelines (which we co-authored) include the following recommendations for hemodynamically stable patients without active bleeding :
●Hgb <6 g/dL – Transfusion recommended except in exceptional circumstances
●Hgb 6 to 7 g/dL – Transfusion generally likely to be indicated
●Hgb 7 to 8 g/dL – Transfusion should be considered in postoperative surgical patients, including those with stable cardiovascular disease, after evaluating the patient’s clinical status
●Hgb 8 to 10 g/dL – Transfusion generally not indicated, but should be considered for some populations (eg, those with symptomatic anemia, ongoing bleeding, acute coronary syndrome with ischemia)
●Hgb >10 g/dL – Transfusion generally not indicated except in exceptional circumstances
The guidelines also emphasize that the decision to transfuse should not be based only on hemoglobin level but should incorporate individual patient characteristics and symptoms. Clinical judgment is critical in the decision to transfuse; therefore, transfusing RBCs above or below the specified hemoglobin threshold may be dictated by the clinical context. Similarly, the decision not to transfuse RBCs to a patient with a hemoglobin concentration below the recommended thresholds is also a matter of clinical judgment.
Overview of our approach — Optimal transfusion practice should provide enough RBCs to maximize clinical outcomes while avoiding unnecessary transfusions.
We consider many factors in deciding whether to transfuse patients with anemia, rather than basing the decision solely on the presence or absence of symptoms or on a specified Hgb level. The final decision to transfuse should incorporate the clinical status, co-morbidity, and the individual wishes of the patient. The Hgb level chosen is based on the results from clinical trials, but clinical judgment is required. It is also important to recognize that lower Hgb thresholds have not been tested in most clinical settings and may be tolerated by many patients. This approach is most consistent with the AABB Guidelines, which we coauthored .
For most patients, we prefer using a restrictive transfusion strategy (ie, giving less blood; transfusing at a lower Hgb level; and aiming for a lower target Hgb level) rather than a liberal transfusion strategy (ie, giving more blood; transfusing at a higher Hgb level). For most hemodynamically stable medical and surgical patients, we suggest considering transfusion at a Hgb of 7 to 8 g/dL, with the threshold based on the value established as safe in the clinical trial with the population that most closely resembles the patient. Some patients may tolerate a lower Hgb level (table 2).
Assessment of the post-transfusion hemoglobin level can be performed as early as 15 minutes following transfusion, as long as the patient is not actively bleeding. This practice is based on studies showing a high degree of concordance between values measured 15 minutes after completion of the transfusion versus longer intervals [21,22].
Major exceptions to the use of a threshold of 7 to 8 g/dL include the following:
●Symptomatic patients may be transfused at higher Hgb levels to treat symptoms. (See 'Symptomatic patient' below.)
●Patients with acute coronary syndromes have not been adequately evaluated in clinical trials and may require higher thresholds for transfusion. (See 'Acute coronary syndrome' below.)
●Threshold-based transfusion is not appropriate for patients requiring massive transfusion, such as in the setting of trauma, because it requires waiting for Hgb levels to be reported. (See "Massive blood transfusion" and "Initial evaluation and management of shock in adult trauma", section on 'Transfusion of red blood cells'.)
Our goal of avoiding unnecessary transfusion also guides our practice of transfusing one unit of red blood cells at a time, rather than requesting multiple units, for a hemodynamically stable patient who is not actively bleeding . Whenever possible, we also initiate or continue treatment of the underlying condition responsible for the anemia.
Our approach of considering a threshold Hgb of 7 or 8 g/dL for most patients is supported by a Cochrane systematic review and metaanalysis of clinical trials of red cell transfusion. The Cochrane review identified 19 randomized clinical trials comparing higher versus lower transfusion thresholds in a total of 6264 medical and surgical patients (adults and children) [23,24]. Trials were included if transfusion was administered on the basis a transfusion trigger, defined as a hemoglobin or hematocrit level below which a blood transfusion was to be given. Most trials compared outcomes in patients transfused at Hgb thresholds between 7 and 10 g/dL; specific thresholds differed for each trial. Compared with liberal transfusion strategies (higher thresholds), restrictive strategies (lower thresholds) resulted in the following [9,23]:
●A 39 percent decrease in the probability of receiving a transfusion (46 versus 84 percent; relative risk 0.61; 95% CI 0.52-0.72)
●Fewer units (1.19) transfused per patient
●A trend towards a lower 30-day mortality (relative risk = 0.85; 95% CI 0.70-1.03) (figure 1)
●A trend towards a lower overall infection rate (relative risk = 0.81; 95% CI 0.66-1.00); however, there was no difference seen with pneumonia.
●No difference in functional recovery, or hospital or intensive care length of stay
●No increased risk of myocardial infarction (MI) when all trials were included (relative risk = 0.88; 95% CI 0.38-2.04).
However, the two largest trials found opposite effects of a restrictive strategy on the risk of MI.
•In the Transfusion Requirements In Critical Care (TRICC) trial of adult intensive care unit patients, a restrictive transfusion strategy was associated with lower risk of MI (0.7 versus 2.9 percent; relative risk = 0.25; 95% CI 0.07-0.88). (See 'Intensive care unit/septic shock' below.)
•In patients undergoing hip fracture repair in the Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS) trial, which was performed in patients with preexisting cardiovascular disease or risk factors, a restrictive transfusion strategy was associated with a non-statistically significant higher risk of MI (3.8 versus 2.3 percent; relative risk = 1.65; 95% CI 0.99-2.75). However, mortality was not adversely affected. (See 'Non-cardiac surgery' below.)
A second metaanalysis pooled the results of 17 trials (7456 hospitalized patients) [25,26]. There was no difference between restrictive and liberal transfusion strategies using the combined outcome of all serious infections and site-specific infections (risk ratio [RR] 0.92; 95% CI; 0.82-1.04). However, when the outcome measurement was restricted to all serious infections (eight trials, 6033 hospitalized patients), there was a lower risk with restrictive versus liberal transfusion strategy (RR 0.84; 95% CI 0.73-0.96). Also, in a subset analysis of seven trials in which all RBC units were leukoreduced, infections were less frequent with a restrictive than with a liberal transfusion strategy (RR 0.83; 95% CI 0.69-0.99).
Evidence from one real-world practice setting also shows that mortality is not adversely affected by the use of restrictive transfusion. An integrated health care system of 21 community hospitals conducted a review of electronic medical records in 218,056 patients with Hgb less than 10 g/dL who were hospitalized before or after institution of restrictive transfusion guidelines [27,28]. The 30-day mortality rate was unaffected by adoption of a restrictive practice (7.8 versus 7.8 percent) despite reduction in the number of units transfused (from 42 to 31 units per 100 patients).
Based on these results, we use a restrictive strategy with a threshold Hgb of 7 to 8 g/dL for most hemodynamically stable medical and surgical patients except those with acute coronary syndromes. (See 'Acute coronary syndrome' below.)
In deciding which restrictive threshold to use, we favor applying specific thresholds as closely as possible to the patient population in which they were established in randomized trials, rather than applying a single threshold to all patients (table 2) . This view is based on our recognition that different patient populations may have different clinical features that could potentially affect transfusion outcomes . As an example, a Hgb threshold of 7 g/dL may be safer for patients with gastrointestinal bleeding because it reduces portal pressure and decreases the chance of rebleeding, whereas a threshold of 8 g/dL for patients with pre-existing coronary disease may provide better oxygen delivery to a vulnerable myocardium. Similarly, the distinction between using a threshold of 7 g/dL for hemodynamically stable patients in the intensive care unit (ICU) and 8 g/dL for hemodynamically stable medical and surgical patients is based solely on the values used in randomized trials; we do not know if these two populations (ICU and medical/surgical) are biologically distinct and truly have different Hbg requirements.
Symptomatic patient — In some randomized trials of transfusion thresholds, symptoms of anemia were an indication for transfusion regardless of whether the Hgb was above the prescribed threshold [31,32]. We agree with the premise that symptomatic anemia should be treated with transfusion in all patients with Hgb <10 g/dL, regardless of the Hgb level, provided that the symptoms are severe enough and are clearly related to the anemia rather than the underlying condition.
Symptoms of anemia include symptoms of myocardial ischemia, orthostatic hypotension or tachycardia unresponsive to fluid replacement. While exertional symptoms can be helpful in alerting the clinician to the presence of anemia, they are generally not considered indications for red cell transfusion. (See "Approach to the adult patient with anemia", section on 'Symptoms'.)
Chronic anemia can present with symptoms such as irritability, weakness, and exercise intolerance. These symptoms of anemia are nonspecific and often not considered sufficient indications for transfusion. Decisions on whether to transfuse red blood cells to treat fatigue are covered separately. (See "Approach to the adult patient with anemia", section on 'Fatigue' and "Causes and diagnosis of iron deficiency anemia in the adult", section on 'Clinical manifestations'.)
Some patients will not manifest typical anemia symptoms for a variety of reasons (eg, altered mental status, diabetic neuropathy, analgesic therapy). Thus, surrogate measures (eg, ECG changes) may be useful in some situations. When transfusion is used in a symptomatic patient, it is important to determine whether symptoms have improved following the transfusion, because this may guide further decision-making. (See "Red blood cell transfusion in adults: Storage, specialized modifications, and infusion parameters".)
Acute coronary syndrome — The optimal transfusion threshold in the setting of acute coronary syndromes (ACS; ie, acute MI, unstable angina) remains unresolved . Our practice in patients with ACS is to transfuse when Hgb is <8 g/dL and to consider transfusion when the Hgb is between 8 and 10 g/dL. If the patient has ongoing ischemia or other symptoms, we maintain the Hgb ≥10 g/dL. In a stable, asymptomatic patient, it is unknown when to transfuse, although we tend to maintain a higher Hgb level using clinical judgment based on evaluating the patient's symptoms and underlying condition.
Our opinion is based on a lack of clinical trial data that support a lower threshold, and a suggestion from a pilot trial of 110 patients with ACS that a threshold of 10 g/dL is safer in patients with ACS . This trial found that compared with transfusion for a Hgb <8 g/dL (restrictive strategy), transfusion to raise the Hgb ≥10 (liberal strategy) was associated with greater survival at 30 days (98 versus 87 percent). Other experts, including other authors for UpToDate, prefer to give transfusions only when the Hgb is at a slightly lower threshold than is used for some patients in our practice . (See "Overview of the non-acute management of unstable angina and non-ST elevation myocardial infarction", section on 'Red cell transfusion' and "Overview of the non-acute management of ST elevation myocardial infarction", section on 'Red cell transfusion'.)
There is variation in transfusion practice in patients undergoing percutaneous coronary intervention (PCI) ; a discussion of transfusion in patients undergoing PCI complicated by periprocedural bleeding is presented separately. (See "Periprocedural bleeding in patients undergoing percutaneous coronary intervention", section on 'Blood transfusion'.)
Asymptomatic hospitalized patient — As described above, for most hemodynamically stable medical and surgical patients, we suggest considering transfusion at a Hgb of 7 to 8 g/dL. Some patients will remain asymptomatic from anemia at lower Hgb levels; conversely transfusion at higher Hgb levels is often appropriate for symptomatic patients and in the setting of an acute coronary syndrome. In addition, threshold-based transfusion may not be appropriate for patients requiring massive transfusion. Discussions surrounding specific clinical scenarios are presented separately. The final decision regarding transfusion must take into account the patient’s wishes and clinical status.
Cardiovascular disease — The decision of whether to transfuse patients with cardiovascular disease should consider the nature of the cardiovascular disorder. As an example, it is possible that patients with acute coronary syndromes require different thresholds for transfusion than do patients with stable coronary artery disease or patients with congestive heart failure.
Preexisting coronary artery disease — A transfusion threshold of 8 g/dL is supported by subgroup analysis of two randomized transfusion trials that included patients with coronary artery disease.
●Patients with preexisting coronary artery disease were included in the Transfusion Trigger Trial for Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS) trial, which found that compared with a threshold of 10 g/dL, a restrictive transfusion strategy (transfusion at a threshold of 8 g/dL or for symptoms) was not associated with worse outcomes, with the exception of an increase in MI that was marginally statistically significant. (See 'Non-cardiac surgery' below.)
●Patients with coronary artery disease were also included in the Transfusion Requirements in Critical Care (TRICC) trial, which found that compared with a threshold of 10 g/dL, a restrictive strategy (transfusion at a threshold of 7 g/dL) was associated with lower mortality. (See 'Intensive care unit/septic shock' below.)
Based on these data, we consider the threshold of 8 g/dL safe for asymptomatic medical patients with stable coronary artery disease .
Transfusion of symptomatic patients with coronary artery disease is guided by the symptoms and clinical judgment, and the optimal threshold for patients with acute coronary syndromes is unresolved. (See 'Symptomatic patient' above and 'Acute coronary syndrome' above and "Overview of the non-acute management of ST elevation myocardial infarction", section on 'Red cell transfusion'.)
Heart failure — Anemia and heart failure (HF) often coexist for a variety of reasons (eg, cytokine changes, dilutional anemia, medical therapy for HF). Many experts consider anemia to be a surrogate marker for poor prognosis in individuals with HF, rather than a therapeutic target. This idea was supported by a randomized, controlled trial of 2278 patients with systolic heart failure and anemia, in which increasing the Hgb concentration from 9-12 g/dL to 13 g/dL using erythropoietin did not improve outcomes .
The approach to transfusion (including more restrictive thresholds for asymptomatic individuals and transfusion for symptoms if Hgb is <10 g/dL) and other management strategies in patients with heart failure (eg, attention to the volume load from the transfusion) is discussed in detail separately. (See "Approach to anemia in adults with heart failure", section on 'Transfusion'.)
Trauma/massive transfusion — Use of massive transfusion in the critically ill, hemodynamically unstable patient cannot be guided by Hgb levels alone and often cannot await interval measurements of Hgb. This issue is discussed separately. (See "Massive blood transfusion" and "Initial evaluation and management of shock in adult trauma", section on 'Transfusion of red blood cells'.)
Intensive care unit/septic shock — Restrictive transfusion appears to be safe in medical patients in an intensive care unit (ICU), with the possible exception of patients with ischemic heart disease/acute coronary syndrome.
The use of a threshold of 7 g/dL in hemodynamically stable patients in the ICU is supported by data from the Transfusion Requirements in Critical Care (TRICC) trial . This trial randomly assigned 838 critically ill, euvolemic patients with a hemoglobin less than 9 g/dL within 72 hours of admission to an intensive care unit to a restrictive transfusion strategy (red cells transfused for Hgb concentration <7 g/dL and Hgb maintained at 7 to 9 g/dL) or a liberal strategy (red cells transfused for Hgb <10 g/dL and Hgb maintained at 10 to 12 g/dL). The mean age was 58, and 82 percent were on mechanical ventilation.
Compared with liberal transfusion, 30-day mortality favored the restrictive strategy but was not statistically significant (23 percent in the liberal group versus 19 percent in the restrictive group). However, 30-day mortality rates were lower with the restrictive strategy in two predefined subgroups:
●Patients who were less severely ill (APACHE II score ≤20; mortality 9 versus 16 percent)
●Patients <55 years of age (mortality 6 versus 13 percent)
In contrast, in patients with ischemic heart disease, there was a reversal in the trend in 30-day mortality, with 30-day mortality in the restrictive strategy arm slightly higher than in the liberal strategy group (26 versus 21 percent) .
Important morbidities were also lower in the restrictive transfusion strategy group as a whole. As examples, rates of myocardial infarction and pulmonary edema were lower in the restrictive group than the liberal strategy group (0.7 versus 2.9 percent and 5.3 versus 10.7 percent, respectively).
The use of a threshold of 7 g/dL was also shown to be safe in patients with septic shock. The Transfusion Requirements in Septic Shock (TRISS) trial randomly assigned 998 patients with septic shock and a hemoglobin level less than 9 g/dL to a restrictive or a liberal transfusion strategy (transfusion at a hemoglobin ≤7g/dL or ≤9 g/dL, respectively) . Consensus criteria for sepsis were used (eg, infection, systemic inflammatory response, hypotension). Transfusions were given as single units of prestorage leukoreduced RBCs. Mortality at 90 days was similar in those transfused with the restrictive and the liberal strategy (43 versus 45 percent; relative risk, 0.94, 95% CI 0.78-1.09). Other outcomes (eg, ischemic events, transfusion reactions, use of vasopressor or inotropic therapy, need for mechanical ventilation) were also similar between the two groups. (See "Evaluation and management of severe sepsis and septic shock in adults", section on 'Additional therapies'.)
These results demonstrate that a restrictive strategy of red cell transfusion is at least as effective as a liberal transfusion strategy in critically ill patients in the ICU, with the possible exception of patients with underlying ischemic heart disease . (See 'Acute coronary syndrome' above and "Use of blood products in the critically ill", section on 'RBC indications'.)
A trial that randomly assigned 198 patients in the ICU following major abdominal surgery for cancer to transfusion at a hemoglobin threshold of 7 versus 9 g/dL favored the higher threshold . A composite outcome at 30 days composed of all-cause mortality, stroke, myocardial infarction, pulmonary embolism, heart failure, cardiac arrest, septic shock, acute kidney injury, respiratory distress, mesenteric ischemia, and septic shock was less frequent in the liberal group (19 versus 36 patients; 20 versus 36 percent). However, the larger TRICC and TRISS trials in patients in the ICU, and the FOCUS trial in patients undergoing non-cardiac surgery, all favor restrictive transfusion. (See 'Non-cardiac surgery' below.)
Acute bleeding — Acute bleeding is an especially challenging clinical setting in which to evaluate red cell transfusion thresholds. For patients with massive bleeding or who are hemodynamically unstable, transfusion should be guided by the pace of the bleeding and the ability to stop the bleeding, rather than by the Hgb. Therefore, the use of transfusion in acutely hemorrhaging patients cannot be based on thresholds. (See "Use of blood products in the critically ill", section on 'Red blood cells' and "Massive blood transfusion" and "Initial evaluation and management of shock in adult trauma", section on 'Transfusion of red blood cells'.)
For patients who are bleeding but hemodynamically stable, some guidance is provided by a randomized trial that suggested a restrictive transfusion strategy was safe in the setting of acute gastrointestinal bleeding when there was access to rapid endoscopic treatment . (See "Approach to acute upper gastrointestinal bleeding in adults", section on 'Blood transfusions'.)
This single center trial randomized 921 patients with acute upper gastrointestinal bleeding to a restrictive or a liberal transfusion strategy (transfusion threshold of 7 g/dL versus 9 g/dL) and determined all cause mortality at 45 days . Patients with massive bleeding, acute coronary syndrome, history of peripheral vascular disease or stroke, and hemoglobin >12 g/dL were excluded. All patients underwent emergent upper endoscopy within six hours and were treated with endoscopic therapy as needed. When compared with the liberal transfusion threshold, the restrictive transfusion threshold in these bleeding patients resulted in the following:
●A lower percent of patients undergoing transfusion (49 versus 86 percent) and fewer transfusions (mean 1.5 versus 3.7 units)
●Fewer complications (40 versus 48 percent)
●Less subsequent (further) bleeding (10 versus 16 percent; hazard ratio 0.62; 95% CI 0.43-0.91)
●Fewer deaths due to uncontrolled bleeding (0.7 versus 3.1 percent)
●Fewer deaths from any cause (5 versus 9 percent; hazard ratio 0.55; 95% CI 0.33-0.92)
This study raises the possibility that in patients with bleeding from other sites (eg, gynecologic, trauma) who are hemodynamically stable; who are not at increased risk for complications (eg, from unstable coronary artery disease); and who have access to rapid surgical intervention, a restrictive transfusion strategy may be safe and might be associated with improved outcomes. Randomized trials to guide transfusion practice in patients with bleeding from other sites are awaited. (See "Management of hemorrhage in gynecologic surgery".)
Non-cardiac surgery — Results of a randomized trial in patients undergoing hip surgery suggest that it is reasonable to use a lower threshold restrictive strategy of 8 g/dL for patients who have undergone surgery, in the absence of symptoms of anemia, even in elderly patients with underlying cardiovascular disease or cardiovascular risk factors.
The optimal transfusion threshold for perioperative transfusion was examined in the Transfusion Trigger Trial for Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS) trial . This trial randomized 2016 patients with preexisting cardiovascular disease or cardiovascular risk factors, to liberal versus restrictive postoperative transfusion after hip repair surgery. All patients were ≥50 years (mean, 82 years) with a postoperative Hgb <10 g/dL. The liberal transfusion group received immediate transfusion of one unit of packed RBCs plus subsequent transfusions to raise the Hgb level to >10 g/dL whenever it fell below this level. The restrictive transfusion group received single unit transfusions only if they developed symptoms of anemia (defined as chest pain, orthostatic hypotension, tachycardia unresponsive to fluid resuscitation, or congestive heart failure) or, in the absence of symptoms, when the Hgb level fell below 8 g/dL.
The primary outcome of the study was death or an inability to walk 10 feet or across a room without assistance at the 60-day evaluation. Secondary outcomes included a combined outcome of in-hospital myocardial infarction, unstable angina, or death; and later death for any reason. Results included the following [16,31]:
●The liberal and restrictive groups had similar rates of death or inability to walk 10-feet unassisted at the 60-day evaluation (35.2 versus 34.7 percent, respectively; OR 1.01; 95% CI 0.84-1.22). Similar results were found at 30-day follow-up.
●The liberal and restrictive groups had similar rates of the composite endpoint of in-hospital acute coronary syndrome or death (4.3 versus 5.2 percent, respectively; OR 0.82; 99% CI 0.48-1.42). Separately, the endpoint of MI was lower in the liberal group (2.3 versus 3.8 percent, OR 0.60; 99% CI 0.30-1.19), while the endpoint of in-hospital death was higher in the liberal group (2.0 versus 1.4 percent, OR 1.44; 99% CI 0.58-3.56).
●Mortality was similar for the liberal and restrictive groups.
•At 60 days, rates of death were 7.6 and 6.6 percent, respectively (OR 1.17; 99% CI 0.75-1.83).
•At approximately three years, rates of death were 43 and 41 percent, respectively (HR 1.09; 95% CI 0.75-1.25).
The causes of death, and the proportion of deaths due to cardiovascular disease, cancer, and infection were comparable between the liberal and restrictive groups.
Major intraabdominal cancer surgery resulting in a requirement for care in the intensive care unit is a possible exception; however, we do not use liberal transfusion in these patients and await additional trials. (See 'Intensive care unit/septic shock' above.)
Issues related to autologous blood salvage and reinfusion during surgery are discussed separately. (See "Surgical blood conservation: Blood salvage".)
Cardiac surgery — Transfusion thresholds in cardiac surgery have been evaluated in several randomized trials. Together, data from these trials suggest that a restrictive transfusion strategy with a Hgb threshold of 7.5 to 8 g/dL appears to be reasonable in patients undergoing cardiac surgery with cardiopulmonary bypass.
●The use of a restrictive versus liberal transfusion threshold was evaluated in the Transfusion Indication Threshold Reduction (TITRe2) trial, which randomly assigned 2007 patients undergoing cardiac surgery to postoperative transfusion for a hemoglobin threshold of <7.5 g/dL versus <9 g/dL . Approximately two-thirds of the patients had underlying coronary artery disease, and the procedures were distributed between coronary artery bypass grafting (CABG), valve surgery, or both (41, 30, and 20 percent, respectively). The incidence of a composite endpoint of infection or an ischemic event was similar in the restrictive and liberal groups (35 versus 33 percent). In post-hoc secondary analyses, there were more deaths at 90 days in the restrictive group (4.2 versus 2.6 percent); however, the 30 day mortality was similar (2.6 versus 1.9 percent) as was the incidence of pulmonary complications (13 versus 12 percent). Transfusions were significantly reduced with the restrictive strategy (median 1 versus 2 units per patient; transfusion avoided in 36 versus 5 percent); and transfusion costs were lower with restrictive transfusion (median savings, approximately £140 [USD $200]).
●Another trial randomly assigned 428 consecutive patients undergoing CABG to postoperative transfusion at a Hgb <8 g/dL versus <9 g/dL . There was no difference in morbidity, mortality, or self-assessment for fatigue or anemia between the two groups. Postoperative transfusion rates were significantly lower for the group with the lower transfusion threshold (0.9 versus 1.4 RBC units/patient), amounting to a savings of 500 RBC units per 1000 CABG.
●A third trial randomly assigned 502 consecutive patients who underwent cardiac surgery with cardiopulmonary bypass to a liberal or restrictive transfusion strategy (to maintain hematocrit at 30 or 24 percent respectively) throughout surgery and the postoperative period (Transfusion Requirements After Cardiac Surgery; TRACS) . The primary outcome was a composite endpoint of 30-day all cause mortality, cardiogenic shock, acute respiratory distress syndrome, or acute renal injury requiring dialysis or hemofiltration. There was no difference in this composite endpoint between the groups (10 percent liberal versus 11 percent restrictive). Independent of transfusion strategy, the number of transfusions correlated with clinical complications and death (HR 1.2 for each unit transfused).
Based on these trials, a restrictive transfusion threshold (ie, to maintain the Hgb above 7.5 or 8 g/dL or the hematocrit above 21 or 24 percent) appears to be reasonable in this population. This issue is discussed in more detail separately. (See "Early noncardiac complications of coronary artery bypass graft surgery", section on 'Blood transfusion'.)
Chronic kidney disease — Management of anemia in patients with chronic kidney disease is complex. Discussion of transfusion and alternatives to transfusion (eg, erythropoietin, iron) in this setting are presented separately. (See "Anemia of chronic kidney disease: Target hemoglobin/hematocrit for patients treated with erythropoietic agents" and "Anemia and the renal transplant recipient", section on 'Therapy' and "Iron balance in nondialysis, peritoneal dialysis, and home hemodialysis patients".)
Ambulatory patient — Symptoms from chronic anemia in ambulatory (ie, non-hospitalized) patients differ from those caused by acute decreases in Hgb concentration in hospitalized patients, because there is time for compensatory mechanisms to occur. The optimal transfusion threshold in ambulatory patients has not been studied.
Some patients with chronic anemia (eg, from bone marrow failure syndromes) may be dependent upon red cell replacement over a period of months or years, which can lead to iron overload. Approximately 200 mg of iron are delivered per unit of RBC; this iron is released when hemoglobin from the transfused RBCs is metabolized after red cell death. Given the progressive loss of red cell viability that occurs during storage, the strategy of selecting the "freshest-available" units in order to maximize post-transfusion survival may be optimal but is dependent on blood center and hospital logistics. This strategy will reduce the total number of units the patient will need to receive, and therefore reduce the total amount of iron delivered to the patient. Chelating therapy is recommended after transfer of approximately 10 to 20 red cell units in patients who are anticipated to require ongoing red cell transfusion support . (See "Iron overload syndromes other than hereditary hemochromatosis", section on 'Transfusional iron overload'.)
Red cell transfusion in patients with acquired or congenital hemolytic anemia is more complex, because transfusion also suppresses erythropoiesis. This issue is discussed separately. (See "Treatment of beta thalassemia", section on 'Chronic hypertransfusion therapy' and "Treatment of autoimmune hemolytic anemia: Warm agglutinins", section on 'Red blood cell transfusion' and "Red blood cell transfusion in sickle cell disease".)
Erythropoietin treatment may be an alternative to chronic transfusion for some ambulatory patients. (See "Role of erythropoiesis-stimulating agents in the treatment of anemia in patients with cancer" and "Erythropoietin for the anemia of chronic kidney disease among predialysis and peritoneal dialysis patients" and "Erythropoietin for the anemia of chronic kidney disease in hemodialysis patients" and "Surgical blood conservation: Preoperative autologous blood donation", section on 'Role of erythropoietin'.)
Oncology patient — There are two major groups of oncology patients for whom transfusion may be indicated:
●Patients undergoing myelosuppressive chemotherapy
●Patients with terminal cancer receiving palliative care
The approach to blood transfusion may differ for these groups depending on the goals of therapy.
In treatment — Patients undergoing cancer therapy with curative intent should be transfused similarly to other medical patients, with transfusion for symptoms and consideration of a threshold of 7 to 8 g/dL in the absence of symptoms. (See 'Asymptomatic hospitalized patient' above and 'Ambulatory patient' above.)
Palliative care — Small observational studies have shown that transfusion offers symptom relief to patients with advanced cancer [46,47]. A Cochrane Database review found no randomized trials of transfusion in patients with advanced cancer, and 12 observational studies that included 653 patients . These studies showed that anemic individuals had subjective responses in symptoms that ranged from 31 to 70 percent. However, receiving unnecessary transfusions takes time away from other activities. Thus, we believe the use of transfusion in oncology patients should be made on a case-by-case basis. (See "Overview of managing common non-pain symptoms in palliative care", section on 'Fatigue'.)
Patients under hospice care continue to receive treatments that improve their comfort and quality of life. Contrary to popular misconceptions, hospice care does not exclude the use of blood transfusion to alleviate symptoms. However, the hospice model of care in the United States addresses the costs associated with transfusion differently from other benefits. Thus, patients who are benefitting from transfusion near the end of life should determine which specific benefits are available from their hospice provider. (See "Hospice: Philosophy of care and appropriate utilization in the United States", section on 'Common questions and misperceptions about hospice' and "Hospice: Philosophy of care and appropriate utilization in the United States", section on 'Limitations'.)
HOSPITAL-WIDE OVERSIGHT PROGRAMS — Many hospitals have developed general guidelines for the appropriate use of blood transfusion. A patient blood management program uses "an evidence-based multidisciplinary approach to optimizing the care of patients who might need transfusion." Patient blood management programs "include interventions taken early in the preparation of medical and surgical patients for treatment, as well as techniques and strategies in the preoperative, operative, and postoperative periods or completion of treatment" . Three pillars of this type of program include optimizing hematopoiesis, minimizing blood loss and bleeding, and harnessing and optimizing tolerance of anemia .
Two components of patient blood management offer the greatest opportunity to reduce blood use:
●Since preoperative anemia is strongly associated with increased risk of transfusion in surgical patients , it is important to screen for anemia early enough prior to surgery to have time to evaluate the cause of anemia and treat it if possible.
●Use of a restrictive transfusion approach reduces blood transfusion for those who do not need it.
In addition to these approaches, some other techniques that may reduce blood use include stopping drugs that impair hemostasis (ie, aspirin) when possible, and using meticulous surgical technique.
We are in favor of such programs, because they attempt to reduce unnecessary transfusion. However, such programs and broad guidelines should not supersede clinical judgment in decisions regarding transfusion, especially by clinicians who are familiar with the individual patient. As an example, if a patient is experiencing symptoms that are known to reflect cardiac ischemia in that individual, transfusion may be appropriate. Alternatively, if a patient is known to tolerate a lower Hgb than that specified in the guideline, then it may be possible for that patient to avoid transfusion.
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topics (see "Patient information: Blood donation and transfusion (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Anemia is associated with adverse clinical outcomes. However, randomized clinical trials are required to establish if transfusion is beneficial or harmful in anemic patients. (See 'Rationale for transfusion' above.)
●There is excellent clinical trial evidence that suggests that a restrictive policy of transfusion at a hemoglobin (Hgb) concentration of 7 to 8 g/dL should guide transfusion decisions in most patients. The use of transfusion thresholds that restrict transfusion to this Hgb concentration are safe in most patient populations, may improve clinical outcomes, and will reduce unnecessary transfusion (table 2). (See 'Rationale for transfusion' above and 'Overview of our approach' above and 'Society guidelines' above.)
●All patients should be assessed clinically when transfusion is considered. If the patient is stable, transfusion may not be needed even when the Hgb level is 7 to 8 g/dL. (See 'Symptomatic patient' above.)
●For patients in the intensive care unit, including those with septic shock, we recommend transfusion to maintain the Hgb at >7 g/dL rather than a higher threshold (Grade 1B); however there may be cases in which the patient is asymptomatic at a Hgb of 7 g/dL, and clinician judgment may support not administering a transfusion. (See 'Overview of our approach' above and 'Intensive care unit/septic shock' above.)
●For most medical and surgical patients who are hemodynamically stable, as well as ambulatory patients, we recommend blood transfusion to maintain the hemoglobin at ≥7 to 8 g/dL rather than 10 g/dL, with the threshold based on the value established as safe in the clinical trial that most closely resembles the patient (Grade 1B); however, there may be cases in which the patient is asymptomatic at a Hgb <8 g/dL, and clinician judgment may support not administering a transfusion. (See 'Overview of our approach' above and 'Asymptomatic hospitalized patient' above and 'Ambulatory patient' above.)
Exceptions include the following:
•Symptomatic patients with Hgb <10 g/dL should be transfused to improve hemodynamic instability and symptoms of myocardial ischemia. (See 'Symptomatic patient' above.)
•For patients with acute coronary syndromes, we use an individualized approach. We transfuse when the Hgb is <8 g/dL; we consider transfusion when the Hgb is between 8 and 10 g/dL; and we maintain the Hgb ≥10 g/dL in the patient with symptoms or ongoing ischemia. In a stable, asymptomatic patient, it is unknown when to transfuse, although we tend to maintain a higher Hgb level based on evaluating the patient's symptoms and underlying condition. Other experts, including other authors for UpToDate, prefer a slightly lower Hgb threshold for transfusion in this population. (See 'Acute coronary syndrome' above and "Overview of the non-acute management of ST elevation myocardial infarction", section on 'Red cell transfusion' and "Overview of the non-acute management of unstable angina and non-ST elevation myocardial infarction", section on 'Red cell transfusion'.)
•Patients requiring massive transfusion (eg, from trauma or ongoing bleeding) often cannot be managed using Hgb thresholds. This issue is discussed separately. (See "Massive blood transfusion" and "Initial evaluation and management of shock in adult trauma", section on 'Transfusion of red blood cells'.)
●Transfusion may be appropriate in the palliative setting. Some hospice programs provide blood transfusion for comfort and symptom relief. (See 'Palliative care' above.)
●Transfusion of one unit of blood at a time is reasonable for hemodynamically stable patients, with assessment of symptoms immediately after transfusion and post-transfusion Hgb levels, which can be done as early as 15 minutes and as late as 24 hours after transfusion. (See 'Overview of our approach' above.)
●Hospital-wide patient blood management programs may be helpful in guiding transfusion practices and reducing unnecessary transfusions, but they should not supersede clinical judgment. (See 'Hospital-wide oversight programs' above.)
●Risks and complications of transfusion are discussed separately. (See 'Risks and complications of transfusion' above.)
- Wang JK, Klein HG. Red blood cell transfusion in the treatment and management of anaemia: the search for the elusive transfusion trigger. Vox Sang 2010; 98:2.
- Consensus conference. Perioperative red blood cell transfusion. JAMA 1988; 260:2700.
- Practice Guidelines for blood component therapy: A report by the American Society of Anesthesiologists Task Force on Blood Component Therapy. Anesthesiology 1996; 84:732.
- Murphy MF, Wallington TB, Kelsey P, et al. Guidelines for the clinical use of red cell transfusions. Br J Haematol 2001; 113:24.
- Clinical Practice Guidelines: Appropriate Use of Red Blood Cells. 2001. http://www.anzsbt.org.au/publications/documents/UseRedBlood_001.pdf (Accessed on September 08, 2011).
- Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med 2009; 37:3124.
- Hamm CW, Bassand JP, Agewall S, et al. ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2011; 32:2999.
- Society of Thoracic Surgeons Blood Conservation Guideline Task Force, Ferraris VA, Brown JR, et al. 2011 update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists blood conservation clinical practice guidelines. Ann Thorac Surg 2011; 91:944.
- Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB*. Ann Intern Med 2012; 157:49.
- Weiskopf RB, Viele MK, Feiner J, et al. Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA 1998; 279:217.
- Weiskopf RB, Feiner J, Hopf H, et al. Heart rate increases linearly in response to acute isovolemic anemia. Transfusion 2003; 43:235.
- Weiskopf RB, Kramer JH, Viele M, et al. Acute severe isovolemic anemia impairs cognitive function and memory in humans. Anesthesiology 2000; 92:1646.
- Carson JL, Duff A, Poses RM, et al. Effect of anaemia and cardiovascular disease on surgical mortality and morbidity. Lancet 1996; 348:1055.
- Carson JL, Noveck H, Berlin JA, Gould SA. Mortality and morbidity in patients with very low postoperative Hb levels who decline blood transfusion. Transfusion 2002; 42:812.
- Wu WC, Schifftner TL, Henderson WG, et al. Preoperative hematocrit levels and postoperative outcomes in older patients undergoing noncardiac surgery. JAMA 2007; 297:2481.
- Carson JL, Sieber F, Cook DR, et al. Liberal versus restrictive blood transfusion strategy: 3-year survival and cause of death results from the FOCUS randomised controlled trial. Lancet 2014.
- Hanna EB, Alexander KP, Chen AY, et al. Characteristics and in-hospital outcomes of patients with non-ST-segment elevation myocardial infarction undergoing an invasive strategy according to hemoglobin levels. Am J Cardiol 2013; 111:1099.
- Shander A, Javidroozi M, Naqvi S, et al. An update on mortality and morbidity in patients with very low postoperative hemoglobin levels who decline blood transfusion (CME). Transfusion 2014; 54:2688.
- Brittenham GM. Iron-chelating therapy for transfusional iron overload. N Engl J Med 2011; 364:146.
- Qaseem A, Humphrey LL, Fitterman N, et al. Treatment of anemia in patients with heart disease: a clinical practice guideline from the American College of Physicians. Ann Intern Med 2013; 159:770.
- Elizalde JI, Clemente J, Marín JL, et al. Early changes in hemoglobin and hematocrit levels after packed red cell transfusion in patients with acute anemia. Transfusion 1997; 37:573.
- Wiesen AR, Hospenthal DR, Byrd JC, et al. Equilibration of hemoglobin concentration after transfusion in medical inpatients not actively bleeding. Ann Intern Med 1994; 121:278.
- Carson JL, Carless PA, Hebert PC. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev 2012; 4:CD002042.
- Carson JL, Carless PA, Hébert PC. Outcomes using lower vs higher hemoglobin thresholds for red blood cell transfusion. JAMA 2013; 309:83.
- Rohde JM, Dimcheff DE, Blumberg N, et al. Health care-associated infection after red blood cell transfusion: a systematic review and meta-analysis. JAMA 2014; 311:1317.
- Rogers MA. Red blood cell transfusion strategies and health care-associated infection--reply. JAMA 2014; 312:2042.
- Roubinian NH, Escobar GJ, Liu V, et al. Decreased red blood cell use and mortality in hospitalized patients. JAMA Intern Med 2014; 174:1405.
- Roubinian NH, Escobar GJ, Liu V, et al. Trends in red blood cell transfusion and 30-day mortality among hospitalized patients. Transfusion 2014; 54:2678.
- Salpeter SR, Buckley JS, Chatterjee S. Impact of more restrictive blood transfusion strategies on clinical outcomes: a meta-analysis and systematic review. Am J Med 2014; 127:124.
- Carson JL, Hebert PC. Should we universally adopt a restrictive approach to blood transfusion? It's all about the number. Am J Med 2014; 127:103.
- Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med 2011; 365:2453.
- Carson JL, Brooks MM, Abbott JD, et al. Liberal versus restrictive transfusion thresholds for patients with symptomatic coronary artery disease. Am Heart J 2013; 165:964.
- Kansagara D, Dyer E, Englander H, et al. Treatment of anemia in patients with heart disease: a systematic review. Ann Intern Med 2013; 159:746.
- Rao SV, Jollis JG, Harrington RA, et al. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004; 292:1555.
- Sherwood MW, Wang Y, Curtis JP, et al. Patterns and outcomes of red blood cell transfusion in patients undergoing percutaneous coronary intervention. JAMA 2014; 311:836.
- Swedberg K, Young JB, Anand IS, et al. Treatment of anemia with darbepoetin alfa in systolic heart failure. N Engl J Med 2013; 368:1210.
- Hébert 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.
- Hébert 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.
- Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med 2014; 371:1381.
- Hébert PC, Carson JL. Transfusion threshold of 7 g per deciliter--the new normal. N Engl J Med 2014; 371:1459.
- de Almeida JP, Vincent JL, Galas FR, et al. Transfusion requirements in surgical oncology patients: a prospective, randomized controlled trial. Anesthesiology 2015; 122:29.
- Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med 2013; 368:11.
- Murphy GJ, Pike K, Rogers CA, et al. Liberal or restrictive transfusion after cardiac surgery. N Engl J Med 2015; 372:997.
- Bracey AW, Radovancevic R, Riggs SA, et al. Lowering the hemoglobin threshold for transfusion in coronary artery bypass procedures: effect on patient outcome. Transfusion 1999; 39:1070.
- Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA 2010; 304:1559.
- Gleeson C, Spencer D. Blood transfusion and its benefits in palliative care. Palliat Med 1995; 9:307.
- Sciortino AD, Carlton DC, Axelrod A, et al. The efficacy of administering blood transfusions at home to terminally ill cancer patients. J Palliat Care 1993; 9:14.
- Preston NJ, Hurlow A, Brine J, Bennett MI. Blood transfusions for anaemia in patients with advanced cancer. Cochrane Database Syst Rev 2012; 2:CD009007.
- www.aabb.org/resources/bct/pbm/Documents/best-practices-pbm.pdf (Accessed on February 19, 2013).
- Shander A, Van Aken H, Colomina MJ, et al. Patient blood management in Europe. Br J Anaesth 2012; 109:55.
- Gombotz H, Rehak PH, Shander A, Hofmann A. Blood use in elective surgery: the Austrian benchmark study. Transfusion 2007; 47:1468.