Cases & Commentaries
- Understand that the traditional transfusion thresholds of hemoglobin below 10 g/dL and hematocrit below 30% are not supported by the evidence.
- List the common serious adverse effects of transfusion.
- Describe the evidence supporting a more restrictive transfusion strategy.
- State what transfusion threshold should be used in hospitalized patients with and without coronary artery disease.
Case & Commentary—Part 1:
A 77-year-old woman with chronic myelodysplastic syndrome (a disorder of the bone marrow leading to chronic anemia) underwent an uncomplicated coronary artery bypass grafting procedure for coronary artery disease. Two months later, she presented to the hospital with shortness of breath and was found to have mild congestive heart failure (CHF). An echocardiogram revealed mildly reduced left ventricular function. She was given diuretics and discharged to a skilled nursing facility (SNF) for rehabilitation.
At the SNF, a follow-up complete blood count showed a hemoglobin level of 7.1 g/dL (normal 12–16 g/dL). Of note, her baseline hemoglobin was between 8 g/dL and 10 g/dL. At the time, the patient remained weak but had no new symptoms. She also had no evidence of gastrointestinal or other bleeding.
Her oncologist (who was managing the myelodysplastic syndrome) received the report the following day and arranged for the patient to be transferred to the emergency department (ED) for evaluation and transfusion. On arriving at the ED, the patient had no symptoms and normal vital signs. Her physical examination was unchanged from prior examinations and revealed no source of bleeding. A repeat hemoglobin level in the ED was 8.1 g/dL.
This case highlights the dilemma facing many clinicians caring for patients with anemia. What is the optimal hemoglobin threshold for red cell transfusion? For many decades, clinicians followed the "10/30" rule—that is, red cell transfusion should be utilized in order to keep the hemoglobin at 10 g/dL and hematocrit at 30%. The "10/30" rule, first recommended by Adams and Lundy in 1942 to improve outcomes in postoperative patients, was based on anecdotal evidence. Despite the lack of experimental evidence to support it, the "10/30" rule remained the standard for transfusion practices for more than 4 decades.(1)
In the 1980s, with the recognition that human immunodeficiency virus (HIV) could be spread through the blood supply, concerns grew about the risk of blood-borne infections. In 1988, the National Institute of Health Consensus Conference on Perioperative Red Blood Cell Transfusions concluded that "available evidence does not support the use of a single criterion for transfusion such as a hemoglobin concentration of less than 100 g/L [10 g/dL]."(2) However, the "right" hemoglobin concentration remained undefined.
The elderly patient in this case was found to have a hemoglobin concentration of 7.1 g/dL and 8.1 g/dL on repeat, which is at her baseline. The difference was likely not representative of a change but within range of lab variability for the same patient. She did not appear to be actively bleeding, and the anemia was likely due to her underlying chronic hematologic disorder. She was weak but otherwise had no new symptoms. Although anemia is associated with fatigue, generalized weakness, dizziness, and impaired exercise capacity, this patient did not appear to be acutely symptomatic from the anemia. Patients with clear symptoms related to anemia may benefit from transfusion, but it is not clear that she would have. In addition, while transfusion might transiently improve the patient's fatigue, her underlying myelodysplasia will likely lead to anemia again, and frequent transfusions could be required to prevent anemia. Transfusion in patients with chronic anemia is best avoided because of risks of iron overload and transmission of blood-borne infections. A better solution would be to use erythropoietin, which is effective in some patients with this disorder.
Notably, because this patient also has underlying cardiovascular disease, she may be less tolerant of anemia. In a study of 1958 surgical patients who refused transfusions due to religious reasons, there was a significant increase in 30-day mortality in patients with underlying cardiovascular disease compared with patients without cardiovascular disease when the hemoglobin dropped below 10 g/dL.(3) What has not been proven, however, is whether red cell transfusion reduces this mortality. In fact, in a recent trial in 2000 elderly patients with underlying cardiovascular disease or risk factors, patients transfused for hemoglobin levels of 10 g/dL experienced no apparent benefit when compared with those transfused at 8 g/dL or for symptoms.(4) There are no randomized clinical trials in patients with congestive heart failure (CHF) or in patients with chronic hematological disorders like myelodysplasia.
Given that this patient had no evidence of active bleeding, no clear symptoms related to the anemia, and the repeat hemoglobin in the ED was within her usual baseline range, there was no indication for a red cell transfusion. It would be reasonable to have the patient return to the nursing facility, continue rehabilitation, and recheck the hemoglobin level in 1–2 weeks.
Case & Commentary—Part 2:
The ED provider contacted the oncologist with the results of the evaluation, and the oncologist requested the patient be admitted to the hospital and transfused with 2 units of blood. He knew about her heart failure and stated that diuretics should be given between each of the units.
The patient was given the blood and the diuretics. Thirty minutes after the second unit of blood finished transfusing, the patient became lethargic and was found to have new hypoxia with an oxygen saturation of 60% on room air. She rapidly worsened and had a respiratory arrest; a code blue was called. She was successfully resuscitated and intubated and transferred to the intensive care unit (ICU).
A chest radiograph revealed vascular congestion and edema. The clinicians caring for her believed that the patient developed hypoxic respiratory failure from acute volume overload in the setting of CHF or from transfusion-related acute lung injury (TRALI; an immune-related transfusion reaction that leads to pulmonary infiltrates and hypoxia 0–6 hours after transfusion). A repeat hemoglobin level at the time was 11.7 g/dL. Because a normal response to a unit of blood is an increase in hemoglobin of 1.0 g/dL, the hemoglobin level of 8.1 g/dL might have been inaccurate.
With supportive care and more diuretics, the patient's hypoxic respiratory failure improved and she was extubated a day later. She remained in the hospital for 2 more weeks recovering from the episode and was transferred back to the SNF.
"Over-transfusion" can occur in the clinical setting where there is good-quality, randomized evidence to justify withholding red blood cell transfusion and a transfusion is given anyway. In our opinion, over-transfusion is common. Determining the incidence of over-transfusion is difficult because there have been few studies examining this, and they were conducted before the most recent evidence was made available. However, several studies have estimated the proportion of inappropriate transfusion at 24%–75%.(5)
Transfusion is not without risks.(6) Clinicians and patients are often most concerned about infections such as HIV and hepatitis. However, two of the more common serious adverse effects of transfusion are transfusion-associated circulatory overload (TACO) and transfusion-related acute lung injury (TRALI). The patient in this case presumably developed either TACO or TRALI, which occurs in 2 to 3 per 100 (7) and 8.1 per 100,000 transfused, respectively.(8) This particular patient had several risk factors associated with TACO: advanced age, underlying cardiovascular disease, and left ventricular dysfunction. Other risk factors that have been identified include transfusion rate and cumulative transfusion volume. A positive fluid balance, which may have occurred here, also places patients at risk for TRALI. Interestingly, pre-treatment with a diuretic has not been shown to protect patients from developing either of these potentially fatal consequences.(7,8)
For unknown reasons in the past, physicians had been taught to give 2 units of blood rather than 1 unit at a time. If this patient had been given 1 unit and then reassessed for the desired clinical response, as is recommended by all modern guidelines (6,9), the clinicians might have avoided giving the second unit, which seems to have led to respiratory arrest. This reassessment would have been especially prudent given this patient's left ventricular dysfunction.
The optimal transfusion threshold continues to be a subject of much debate, although two trials provide important data. The first study is the Transfusion Requirements in Critical Care (TRICC) trial. In this trial, 838 normovolemic ICU patients were randomized to either a "restrictive" or a "liberal" transfusion strategy.(10) Patients in the restrictive group were transfused if the hemoglobin dropped below 7 g/dL, and those in the liberal group received red cell transfusions for hemoglobin levels less than 10 g/dL. There was a trend toward lower 30-day mortality in the restrictive group (18.7% vs. 23.3%), and all other negative outcomes occurred less frequently in restrictive group.
The second, and most recent, trial is the Transfusion Trigger Trial for Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS).(4) This study randomized 2016 patients undergoing surgical repair for hip fracture with a history of cardiovascular disease, or with risk factors of cardiovascular disease, to a restrictive or liberal transfusion strategy. Those in the restrictive group received red cell transfusion only if the hemoglobin concentration was less than 8 g/dL, or if they developed symptoms of anemia. Patients in the liberal group were transfused to maintain a hemoglobin level greater than 10 g/dL. The primary outcome (being able to walk 10 feet or across the room at 60 days) was similar in both the liberal group (35.2%) and restrictive group (34.7%). The rates of in-hospital death or acute coronary syndrome were also not statistically significant between the liberal group (4.3%) and restrictive group (5.2%).
A recent meta-analysis examined 19 randomized, controlled studies that evaluated transfusion thresholds utilizing a restrictive or liberal transfusion strategy.(11) Using a restrictive transfusion strategy saved an average of 1.19 units of blood per patient transfused. Hemoglobin concentration was 1.5% lower in the patients allocated to a restrictive strategy. There was no difference in 30-day mortality or any other clinical outcomes between restrictive and liberal transfusion strategies.
Incorporating recent, emerging evidence and this systematic review, the AABB (formerly, the American Association of Blood Banks) published new guidelines on red blood cell transfusions in 2012.(6) In stable, hospitalized patients, these guidelines recommend adhering to a restrictive transfusion strategy, with a transfusion threshold at a hemoglobin level of 7 g/dL (in ICU patients) and 8 g/dL (in surgical and medical patients). In patients with a prior history of cardiovascular disease, the AABB also recommends a restrictive transfusion strategy and states that transfusion should be considered at a hemoglobin level of 8 g/dL or less, or symptoms of anemia related to coronary artery disease (e.g., chest pain thought to be cardiac in origin, CHF symptoms, or unexplained tachycardia or hypotension unresponsive to fluid replacement). Due to a lack of evidence, the AABB could not recommend for or against a liberal or restrictive transfusion strategy in patients with the acute coronary syndromes (myocardial infarctions or unstable angina). We advise carefully assessing each patient for the need of transfusion. Finally, the AABB suggests that symptoms of anemia, as well as hemoglobin level, should also be factored in the decision to transfuse. It is important to emphasize that the hemoglobin concentration of 7–8 g/dL is recommended to consider transfusion. If the patient is stable, blood may not be needed.
We recommend using the AABB guidelines, which integrate the latest and best-quality evidence. Disseminating the guidelines requires an intense educational campaign to include all those who would be involved in making the decision to transfuse a patient—clinicians, nursing staff, blood bank personnel, and housestaff. A systematic review evaluated 19 studies that examined various behavioral interventions aimed at reducing blood product use.(12) The different interventions studied included individual/group education, guidelines, audit with feedback, audit with approval, or the implementation of a new form with criteria for transfusion. All of these methods were equally as effective in reducing either the amount of blood products transfused per patient (range, 9%–77%) or the proportion of patients transfused (range, 17%–79%). Similarly, a more recent study found that targeted education combined with a decision support intervention with computerized physician order entry decreased the percentage of inappropriate transfusions by 8.8%–13%.(13) In the end, specific interventions must be tailored to each institution's unique characteristics and provider and patient profile.
Ultimately, the decision of whether or not to transfuse a particular patient is complex. Each patient's unique clinical course and situation must be considered, and the risks versus the benefits of a red blood cell transfusion must be weighed. The evidence suggests that a restrictive transfusion strategy, utilizing a hemoglobin concentration of 7–8 g/dL (or symptoms of anemia) as a transfusion threshold, has been shown to be safe in most stable hospitalized, perioperative, and ICU patients, including those with underlying cardiovascular disease. The transfusion threshold for patients with acute coronary syndrome is unknown, and this area needs further research.
- Although the blood supply in the United States has been shown to be safe, transfusion is not without its risks.
- In stable, hospitalized patients, a restrictive transfusion strategy, using a hemoglobin concentration 7–8 g/dL, or the presence of symptoms of anemia as transfusion thresholds, has been shown to be safe.
- A restrictive transfusion strategy may also be utilized in stable hospitalized patients with existing cardiovascular disease.
- Transfusion should be considered at hemoglobin transfusion threshold but not necessarily administered. The decision to transfuse should be based on a thorough assessment of each patient's individual clinical circumstances and symptoms.
- The transfusion threshold in patients with the acute coronary syndrome is unknown.
Manish S. Patel, MD
Assistant Professor of Medicine
Division of General Internal Medicine
UMDNJ-Robert Wood Johnson Medical School
New Brunswick, New Jersey
Jeffrey L. Carson, MD
Richard C. Reynolds Professor of Medicine
Chief, Division of General Internal Medicine
UMDNJ-Robert Wood Johnson Medical School
New Brunswick, New Jersey
Faculty Disclosure: Drs. Patel and Carson have declared that neither they, nor any immediate member of their families, have a financial arrangement nor other relationship with the manufacturers of any commercial products discussed in this continuing medical education activity. The commentary does not include information regarding investigational or off-label use of pharmaceutical products or medical devices.
1. Adams RC, Lundy JS. Anesthesia in cases of poor surgical risk: some suggestions for decreasing the risk. Surg Gynecol Obstet. 1942;74:1011-1019.
2. Consensus conference. Perioperative red blood cell transfusion. JAMA. 1988;260:2700-2703. [go to PubMed] 3. Carson JL, Duff A, Poses RM, et al. Effect of anaemia and cardiovascular disease on surgical mortality and morbidity. Lancet. 1996;348:1055-1060. [go to PubMed] 4. Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;365:2453-2462. [go to PubMed] 5. Barr PJ, Donnelly M, Cardwell CR, Parker M, Morris K, Bailie KE. The appropriateness of red blood cell use and the extent of overtransfusion: right decision? Right amount? Transfusion. 2011;51:1684-1694. [go to PubMed]
3. Carson JL, Duff A, Poses RM, et al. Effect of anaemia and cardiovascular disease on surgical mortality and morbidity. Lancet. 1996;348:1055-1060. [go to PubMed]
4. Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011;365:2453-2462. [go to PubMed]
5. Barr PJ, Donnelly M, Cardwell CR, Parker M, Morris K, Bailie KE. The appropriateness of red blood cell use and the extent of overtransfusion: right decision? Right amount? Transfusion. 2011;51:1684-1694. [go to PubMed]
6. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice guideline from the AABB. Ann Intern Med. 2012;157:49-58. [go to PubMed]
7. Li G, Rachmale S, Kojicic M, et al. Incidence and transfusion risk factors for transfusion-associated circulatory overload among medical intensive care unit patients. Transfusion. 2011;51:338-343. [go to PubMed]
8. Toy P, Gajic O, Bacchetti P, et al. Transfusion related acute lung injury: incidence and risk factors. Blood. 2012;119:1757-1767. [go to PubMed]
9. 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;37:3124-3157. [go to PubMed]
10. 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;340:409-417. [go to PubMed]
11. Carson JL, Carless PA, Hébert PC. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev. 2012;4:CD002042. [go to PubMed]
12. Tinmouth A, Macdougall L, Fergusson D, et al. Reducing the amount of blood transfused: a systematic review of behavioral interventions to change physicians' transfusion practices. Arch Intern Med. 2005;165:845-852. [go to PubMed]
13. Rothschild JM, McGurk S, Honour M, et al. Assessment of education and computerized decision support interventions for improving transfusion practice. Transfusion. 2007;47:228-239. [go to PubMed]