Jeffrey L. Carson, MD; Brenda J. Grossman, MD, MPH; Steven Kleinman, MD; Alan T. Tinmouth, MD; Marisa B. Marques, MD; Mark K. Fung, MD, PhD; John B. Holcomb, MD; Orieji Illoh, MD; Lewis J. Kaplan, MD; Louis M. Katz, MD; Sunil V. Rao, MD; John D. Roback, MD, PhD; Aryeh Shander, MD; Aaron A.R. Tobian, MD, PhD; Robert Weinstein, MD; Lisa Grace Swinton McLaughlin, MD; Benjamin Djulbegovic, MD, PhD; for the Clinical Transfusion Medicine Committee of the AABB
Note: Guidelines cannot account for individual variation among patients. They are not intended to supplant physician judgment with respect to specific patients or special clinical situations. Accordingly, the AABB considers adherence to these guidelines to be voluntary, with the ultimate determination regarding their application to be made by the physician in the light of each patient's unique circumstances.
Financial Support: Support for the development of this guideline was provided by the AABB, Bethesda, Maryland.
Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M11-3065.
Requests for Single Reprints: Jeffrey L. Carson, MD, Division of General Internal Medicine, UMDNJ–Robert Wood Johnson Medical School, 125 Paterson Street, New Brunswick, NJ 08903; e-mail, firstname.lastname@example.org.
Current Author Addresses: Dr. Carson: UMDNJ–Robert Wood Johnson Medical School, Division of General Internal Medicine, 125 Paterson Street, 2nd Floor, New Brunswick, NJ 08903-0019.
Dr. Grossman: Washington University, Department of Pathology and Immunology, Campus Box 8118, 660 South Euclid Avenue, Saint Louis, MO 63110.
Dr. Kleinman: University of British Columbia, 1281 Rockcrest Avenue, Victoria, British Columbia V9A 4W4, Canada.
Dr. Tinmouth: Ottawa Hospital Research Institute, 501 Smyth Road, Ottawa, Ontario K1H 8L6, Canada.
Dr. Marques: University of Alabama at Birmingham, 619 19th Street South, West Pavilion, P230G, Birmingham, AL 35249.
Dr. Fung: Fletcher Allen Health Care, 111 Colchester Avenue, Burlington, VT 05401.
Dr. Holcomb: University of Texas Medical Center at Houston, 6410 Fannin Street, UPB 1100, Houston, TX 77030.
Dr. Illoh: U.S. Food and Drug Administration, Center for Biologics Evaluation and Research, Division of Blood Applications, HFM-375, 1401 Rockville Pike, Rockville, MD 20852.
Dr. Kaplan: Yale University, 330 Cedar Street, BB-310, New Haven, CT 06520.
Dr. Katz: Mississippi Valley Regional Blood Center, 5500 Lakeview Parkway, Davenport, IA 52807.
Dr. Rao: Duke University, 508 Fulton Street (111A), Durham, NC 27705.
Dr. Roback: Emory University School of Medicine, EUH D-655, 1364 Clifton Road Northeast, Atlanta, GA 30322.
Dr. Shander: Englewood Hospital and Medical Center, 350 Engle Street, Englewood, NJ 07631.
Dr. Tobian: Johns Hopkins University, Carnegie 667, 600 North Wolfe Street, Baltimore, MD 21287.
Dr. Weinstein: University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655.
Dr. McLaughlin: American Red Cross, Biomedical Headquarters, Holland Laboratory, 15601 Crabbs Branch Way, E230, Rockville, MD 20855.
Dr. Djulbegovic: University of South Florida, 12901 Bruce B. Downs Boulevard, MDC02, Tampa, FL 33612.
Author Contributions: Conception and design: J.L. Carson, S. Kleinman, M.B. Marques, M.K. Fung, J.B. Holcomb, L.J. Kaplan, L.M. Katz, J.D. Roback, A. Shander, A.A.R. Tobian, B. Djulbegovic.
Analysis and interpretation of the data: J.L. Carson, B.J. Grossman, S. Kleinman, A.T. Tinmouth, M.B. Marques, M.K. Fung, L.J. Kaplan, L.M. Katz, S.V. Rao, J.D. Roback, A. Shander, A.A.R. Tobian, R. Weinstein, L.G.S. McLaughlin, B. Djulbegovic.
Drafting of the article: J.L. Carson, B.J. Grossman, S. Kleinman, A.T. Tinmouth, M.K. Fung, J.B. Holcomb, O. Illoh, J.D. Roback, A. Shander, A.A.R. Tobian, B. Djulbegovic.
Critical revision of the article for important intellectual content: J.L. Carson, B.J. Grossman, S. Kleinman, A.T. Tinmouth, M.B. Marques, M.K. Fung, J.B. Holcomb, O. Illoh, L.J. Kaplan, L.M. Katz, S.V. Rao, J.D. Roback, A. Shander, A.A.R. Tobian, R. Weinstein, B. Djulbegovic.
Final approval of the article: J.L. Carson, B.J. Grossman, A.T. Tinmouth, M.B. Marques, M.K. Fung, J.B. Holcomb, O. Illoh, L.J. Kaplan, L.M. Katz, S.V. Rao, J.D. Roback, A. Shander, A.A.R. Tobian, R. Weinstein, L.G.S. McLaughlin, B. Djulbegovic.
Provision of study materials or patients: J.L. Carson.
Statistical expertise: J.L. Carson, B. Djulbegovic.
Administrative, technical, or logistic support: M.K. Fung, B. Djulbegovic.
Collection and assembly of data: J.L. Carson, M.B. Marques, L.J. Kaplan, L.M. Katz, A.A.R. Tobian, L.G.S. McLaughlin.
Although approximately 85 million units of red blood cells (RBCs) are transfused annually worldwide, transfusion practices vary widely. The AABB (formerly, the American Association of Blood Banks) developed this guideline to provide clinical recommendations about hemoglobin concentration thresholds and other clinical variables that trigger RBC transfusions in hemodynamically stable adults and children.
These guidelines are based on a systematic review of randomized clinical trials evaluating transfusion thresholds. We performed a literature search from 1950 to February 2011 with no language restrictions. We examined the proportion of patients who received any RBC transfusion and the number of RBC units transfused to describe the effect of restrictive transfusion strategies on RBC use. To determine the clinical consequences of restrictive transfusion strategies, we examined overall mortality, nonfatal myocardial infarction, cardiac events, pulmonary edema, stroke, thromboembolism, renal failure, infection, hemorrhage, mental confusion, functional recovery, and length of hospital stay.
The AABB recommends adhering to a restrictive transfusion strategy (7 to 8 g/dL) in hospitalized, stable patients (Grade: strong recommendation; high-quality evidence).
The AABB suggests adhering to a restrictive strategy in hospitalized patients with preexisting cardiovascular disease and considering transfusion for patients with symptoms or a hemoglobin level of 8 g/dL or less (Grade: weak recommendation; moderate-quality evidence).
The AABB cannot recommend for or against a liberal or restrictive transfusion threshold for hospitalized, hemodynamically stable patients with the acute coronary syndrome (Grade: uncertain recommendation; very low-quality evidence).
The AABB suggests that transfusion decisions be influenced by symptoms as well as hemoglobin concentration (Grade: weak recommendation; low-quality evidence).
Adverse effects of RBC transfusion contrasted with other risks.
Risk is depicted on a logarithmic scale. Shaded bars represent the risk per RBC unit transfused, and unshaded bars represent the risk for fatality per person per year for various life events. During 2007 through 2008, HIV incidence in blood donors was 3.1 per 100 000 person-years. Residual risk was estimated as 1:1 467 000 transfused blood components or 6.8 per 10 million transfused components (10). During 2007 through 2008, HCV incidence in blood donors was 5.1 per 100 000 person-years with residual risk estimate of 0.87 per million transfused blood components (1:1 149 000) or 8.7 per 10 million transfused components (10). For 2006 to 2008, HBV incidence in blood donors was 3.41 to 3.43 per 100 000 person-years. The estimated residual risk for HBV was 1 in 282 000 to 1 in 357 000 transfused blood components (11) or 2.8 per million to 3.6 per million transfused blood components. In a recently published, large, prospective study with active recipient surveillance, the rate of TRALI occurrence in 2009 was 0.81 (95% CI, 0.44 to 1.49) per 10 000 transfused blood components or 8.1 per 100 000 transfused blood components (12). Although the literature has a wide range of TRALI risk estimates (1, 13–16), we have selected the rate reported in this recent prospective study. Three studies of TACO have produced similar results. In a study of 901 intensive care unit patients, 6% of patients who received transfusions developed TACO. Median units transfused were 2 RBCs and 3 overall (including plasma and platelets) (17). The rate per transfused RBC unit was 2 to 3 per 100. In 382 patients undergoing hip and knee replacement, 1% developed TACO after surgery (18). In a study of patients having total hip and knee arthroplasty, 8% developed fluid overload necessitating diuretic use and 4% of patients who did not receive transfusions developed fluid overload, leading to a TACO estimate of 4% (19). In published studies from the late 1990s, the risk for fatal hemolysis was estimated to range from 1.3 to 1.7 per million (5.9 to 7.7 per 10 million) transfused RBC units in one report (20) and 1 per 1 800 000 or 8.5 per 10 million in a second report (21). More recently, transfusion-related fatalities due to hemolysis reported to the U.S. Food and Drug Administration averaged 12.5 deaths per year from 2005 to 2010 (22). With 15 million RBC units transfused per year, the estimated risk for death due to hemolysis is 1:1 250 000 or 8 per 10 million RBC units. Fever (febrile nonhemolytic transfusion reactions) was estimated to be 1.1% with prestorage leukoreduction and 2.15% with poststorage leukoreduction (23). Death from medical error as reported by the Institute of Medicine was 1.3 to 2.9 per 1000 hospital admissions (24). Life-threatening transfusion reactions, defined as reactions requiring major medical intervention (for example, vasopressors, intubation, or transfer to an intensive care unit), occurred in 1:139 908 transfusions or 7.1 per million transfusions (1). Fatal motor vehicle accidents were estimated at 13.1 per 100 000 persons in 2008 or 1.3 per 10 000 persons (25). The rate of firearm homicide (which excludes suicide) was 4 per 100 000 persons in 2008 (25). Fatal falls were estimated at 8.2 deaths per 100 000 persons in 2008 (25). Lightning fatalities ranged from 0.02 per million (2 per 100 million) persons in California and Massachusetts to 2.0 per million persons in Wyoming (0 risk in Hawaii, Rhode Island, and Alaska) (26). The odds of being killed on a single airline flight on the 30 airlines with the best accident rates were 1 per 29.4 million. Among the 25 airlines with the worst accident records, rates were 1.7 per million per flight (27). Modified from Dzik (2002) (28). HBV = hepatitis B virus; HCV = hepatitis C virus; RBC = red blood cell; TACO = transfusion-associated circulatory overload; TRALI = transfusion-related acute lung injury.
Evidence Tables for Transfusion Outcomes
Evidence Tables for Clinical Outcomes
The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.
Shamsul A, Bhuiyan, Hospitalist, Dr Karim Badar (PGY2)
Queens Hospital Center
March 29, 2012
TRANSFUSION TRIGGERS point is how much effective?
Thanks to all authors and AARB for such a beautiful updated recommendation. Since beginning of my hospitalist carrier, every time I have been thinking are we over utilizing our resources regarding blood transfusion. From observation I have seen physician transfusing blood at number he/or she does not feel comfortable. Well known terms as "transfusion triggers differed among physician". I have been encouraging my resident group that don't learn what your attending wants, try to look for evidence behind it. I guess this new update will be great resource for me to teach my resident group. Again I would love to see more study regarding cost-effectiveness too. Since we are in the era of going through medical reforming, cost cutting, preventing readmission to save money. Can we save really reasonable amount of money avoiding aggressive blood transfusion. Now a days lot of talk going on regarding , cardiovascular disease, MI, CHF, infection (pneumonia), readmission, cutting payment by insurance providers. Surprisingly all these medical conditions was evaluated in different Blood transfusion study, such as FOCUS trial, TRICC study. Wondering, are we contributing or provocation some way by doing blood transfusion in this group of pt , when they come to hospital with borderline hb of 7-8, and receive blood transfusion along with Rx for admitted condition (PNA/MI/CHF etc). Hope to see in near future readmission group of pt with blood transfusion association from any trials for further evidence that will help to make clear decision. Not sure current recommendation with "TRANSFUSION TRIGGERS" point is how much effective. Rewarding Physician/hospital for less blood transfusion by insurance provider also might be very encouraging.
Harvey G. Klein MD, Charles Natanson MD
Clinical Center, NIH
July 14, 2012
The AABB’s recommendation, adhering to a “restrictive” transfusion threshold that relies on a hemoglobin determination of 7-8 gm/dL or less for hemodynamically stable patients, is well-intentioned, but problematic.(1) We are not surprised that the editorial accompanying this proposed new guideline arrives at a different conclusion.(2) We concur. We too believe that transfusion thresholds should be titrated based on many important individual patient laboratory and physiological variables. Why the disparity after considering the same decade of clinical experience and trial data? The weaknesses of the most influential trial are well described in the accompanying editorial. Since this and other trials upon which the AABB guideline committee based their recommendations never tested routine titrated care, the standard of care at the time, the trials could not determine whether either treatment – two arbitrarily selected hemoglobin determinations defined as “restrictive” or “liberal”- improved outcome or saved blood.(3) This trial design allows one to conclude only which of these two strategies should not be used. The less harmful treatment could still be worse than usual titrated care. The patients in this trial had a mortality risk of approximately 2/1000 or 0.2% per unit of transfused red blood cells, a death rate from transfusion some two orders of magnitude greater than that estimated at the time by the General Accounting Office for usual transfusion practice (0.001% or one death for every 10,000 units of transfused blood).(4) Differences in patient mix and follow-up seem unlikely to explain this enormous discrepancy. There is a real possibility that the new AABB guidelines have come to the wrong conclusion. We recognize that Carson et al. have softened their recommendations by acknowledging that “clinical judgment is critical in the decision to transfuse … transfusing RBCs above or below the specified hemoglobin threshold may be dictated by the clinical context.” We hope that this sentence will not be overlooked by those who seek a simple protocol-driven care approach. We do not believe that their systematic review provides sufficient direct evidence to change the previous recommendations by AABB and others regarding titrated transfusion practice. We concur with the authors’ observation that “clinical trials are needed in other patient populations.” However randomized, controlled trials with the addition of a titrated care arm in each of the conditions they describe are neither practical nor affordable. Since expert clinicians now appear to recognize more than one “standard of care” for red cell transfusion, the time is ripe to undertake comparative effectiveness trials using real world data and large populations to help resolve this conundrum. (5)
(1) Carson JL, Grossman BJ, Kleinman S, Tinmouth AT, Marques MB, Fung MK, Holcomb JB, Illoh O, Kaplan LJ, Katz LM, Rao SV, Roback JD, Shander A, Tobian AA, Weinstein R, Swinton McLaughlin LG, Djulbegovic B; for the Clinical Transfusion Medicine Committee of the AABB. Red Blood Cell Transfusion: A Clinical Practice Guideline From the AABB Ann Intern Med. 2012;157(1):49-58
(2) Vincent JL. Indications for blood transfusions: too complex to base on a single number? Ann Intern Med. 2012;157(1):71-2
(3) Deans KJ, Minneci PC, Suffredini AF, Danner RL, Hoffman WD, Ciu X, Klein HG, Schechter AN, Banks SM, Eichacker PQ, Natanson C. Randomization in clinical trials of titrated therapies: unintended consequences of using fixed treatment protocols. Crit Care Med. 2007;35(6):1509-16
(4) United States General Accounting Office. Blood Supply: Transfusion- Associated Risks. 1997, Washington DC, GAO/PEMD-97-2
(5) Klein HG. Comparative effectiveness research: welcome to the real world. Transfusion. 2012; 52(6):1162-4
Jeffrey L Carson, MD, Sunil V Rao, MD, Louis M Katz, MD
UMDNJ-Robert Wood Johnson Medical School; Duke Univ School of Medicine, Durham, NC; Mississippi Valley Regional Blood Center and Carver College of Medicine, Univ of Iowa, Davenport, IA
August 21, 2012
We appreciate the opportunity to reiterate that our guidelines state that transfusion be CONSIDERED in specific patient subgroups when the nadir hemoglobin reaches 7-8 g/dL.(1) In addition, the decision to transfuse should be influenced by signs and symptoms. We based our recommendations on the best available evidence: 19 randomized clinical trials in 6,264 patients. However, clinical trials provide an average effect in the population studied. Thus, it is likely that some patients need more or less blood to improve outcomes. The “conundrum” is what clinical factors should influence the “routine titrated” transfusion decision. The largest trial evaluating transfusion thresholds used pre-specified symptoms (e.g., chest pain, orthostatic hypotension or tachycardia unresponsive to fluid resuscitation),(2) which we included in our guidelines. While this approach is based on a randomized trial, these symptoms will not be applicable to all clinical settings. We agree with the editorial accompanying the guidelines that other parameters such as fatigue, dyspnea, mechanical ventilation, or the use of low SvO2 (venous oxygen saturation) are unproven.(3)Unfortunately, there is no evidence to support other “important individual patient laboratory and physiological variables” as the basis for transfusion decisions. Drs. Klein and Natanson suggest that clinical trials evaluating transfusion thresholds should include an arm where patients are managed using the standard of care, which, in their opinion, is individually titrated transfusions. While it is true that this approach is often recommended, the evidence strongly suggests that clinicians do not follow this recommendation and have not adopted this approach as standard of care. For example, a recent study in orthopedic surgery documents that transfusion decisions were mostly based on hemoglobin levels.(4) Similarly, a study conducted in patients undergoing percutaneous coronary intervention showed that most transfusions were given when the hemoglobin dropped to a certain threshold, rather than for individual clinical circumstances. (5) This is also consistent with our anecdotal observations at our own hospitals. Given that most clinicians do not individually titrate transfusions, the proposed design would be very difficult to implement and would not reflect “real-world” practice. A more clinically applicable trial design incorporates pre-specified symptoms for transfusion along with hemoglobin thresholds. We stand by our recommendations: in pre-specified groups of patients, randomized clinical trials have demonstrated that restrictive transfusion using 7-8 g/dL threshold reduces blood use without harm. When this hemoglobin level is reached, we recommend considering transfusion, but only after evaluating the patient’s clinical status. Jeffrey L Carson, MDSunil V Rao, MDLouis M Katz, MD
1. Carson JL, Grossman BJ, Kleinman S, Tinmouth AT, Marques MB, Fung MK, et al. Red Blood Cell Transfusion: A Clinical Practice Guideline From the AABB*. Annals of internal medicine. 2012;157(1):49-58.
2. Carson JL, Terrin ML, Noveck H, Sanders DW, Chaitman BR, Rhoads GG, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. The New England journal of medicine. 2011;365(26):2453-62.
3. Vincent JL. Indications for blood transfusions: too complex to base on a single number? Annals of internal medicine. 2012;157(1):71-2.
4. Vuille-Lessard E, Boudreault D, Girard F, Ruel M, Chagnon M, Hardy JF. Red blood cell transfusion practice in elective orthopedic surgery: a multicenter cohort study. Transfusion. 2010;50(10):2117-24.
5. Moscucci M, Ricciardi M, Eagle KA, Kline E, Bates ER, Werns SW, et al. Frequency, predictors, and appropriateness of blood transfusion after percutaneous coronary interventions. The American journal of cardiology. 1998;81(6):702-7.
Carson JL, Grossman BJ, Kleinman S, Tinmouth AT, Marques MB, Fung MK, et al. Red Blood Cell Transfusion: A Clinical Practice Guideline From the AABB*. Ann Intern Med. 2012;157:49–58. doi: 10.7326/0003-4819-157-1-201206190-00429
Download citation file:
Published: Ann Intern Med. 2012;157(1):49-58.
Acute Coronary Syndromes, Cardiology, Coronary Heart Disease, Emergency Medicine, Guidelines.
Copyright © 2019 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use