O R I G I N A L A R T I C L E

Patient blood transfusion management: dischargehemoglobin level as a surrogate marker for red blood cell

utilization appropriateness_3591 1..7

Jason Edwards, Chris Morrison, Maleeha Mohiuddin, Vladislav Tchatalbachev, Charmi Patel,

Vicki L. Schwickerath, Jay E. Menitove, and Gurmukh Singh

BACKGROUND: Blood transfusion management strate-gies minimize transfusion-associated risks, enhanceoutcomes, and reduce costs. We explored an associa-tion of discharge hemoglobin (Hb) with pretransfusionHb, transfusion indications, and red blood cell (RBC)transfusions. We stipulate that patients with dischargeHb concentrations greater than 10.0 g/dL, or even9.0 g/dL, received excessive RBC transfusions.STUDY DESIGN AND METHODS: We examinedaggregate data from five hospitals and for one of thehospitals, the focus hospital, we reviewed patientrecords for a period of 6 months. Data analysesincluded number of RBC units transfused and Hbvalues before transfusion, after transfusion, and atdischarge.RESULTS: In aggregate, 27% to 47% patients had dis-charge Hb levels greater than 10.0 g/dL. At the focushospital, 27% had a discharge Hb level greater than10 g/dL and 50.3% had a discharge Hb level greaterthan 9.0 g/dL. At the focus hospital, the mean Hbtrigger for transfusion was a Hb level of 7.3 g/dL; themean posttransfusion Hb level was 9.3 g/dL and meandischarge Hb level was 9.2 g/dL. Overall, 76% of thetransfusions were of an even number of RBC units.CONCLUSION: In aggregate, overutilization exceeded20%. At the focus hospital, approximately one-quarterof patients receiving transfusions had a Hb concentra-tion greater than 10.0 g/dL at discharge. Transfusedpatients’ discharge Hb concentration represents aneffective indicator for retrospective monitoring of trans-fusion appropriateness. In light of the large number ofpatients receiving even number transfusions, reviewingHb levels after transfusion of each RBC unit couldreduce unnecessary transfusions. Retrospective reviewof discharge Hb data focuses providers on transfusionoutcomes and affords an educational opportunity forblood utilization management.

Patient blood transfusion management (PBM)recognizes the complex interplay betweenimproving patient outcomes and reducingtransfusion costs that have outpaced inflation-

ary factors.1-6 PBM involves evidence-based practices formanaging anemia, hemostasis, and blood loss andimproving outcomes.7 Despite advances in proper bloodmanagement, nearly a quarter of the red blood cell (RBC)transfusions, in various studies, appear to be excessive.8-11

There is increasing recognition of adverse events associ-ated with blood transfusions.12-19

In this regard, we sought to determine communitypractices in a midwestern metropolitan area and investi-gate specific transfusion practices at an inner-city, safetynet teaching hospital at which many transfusion decisionsare made by physicians in training.

Traditionally, conservation of valuable blood andblood product resources is important to avoid or mini-mize periods of inadequate blood supply. From 1999 to2004, blood use in the United States increased by 16%,whereas blood use in the United Kingdom declined by8%.10,20 Current per-capita blood use in the United States is15% higher than in Europe and 44% higher than in Canadasuggesting possible overutilization within the UnitedStates.21 Interest in patient outcome results, currenthealth care reforms’ attention to outcome-based medical

ABBREVIATIONS: GI = gastrointestinal; PBM = patient blood

transfusion management.

From the Department of Pathology, Truman Medical Center,

University of Missouri Kansas City School of Medicine, and the

Community Blood Center of Kansas City, Kansas City, Missouri.

Address reprint requests to: Gurmukh Singh, Clinical

Pathology, Truman Medical Center, 2301 Holmes St, Kansas City,

MO 64108; e-mail: gurmukh@pol.net.

Received for publication September 26, 2011; revision

received January 12, 2012, and accepted January 21, 2012.

doi: 10.1111/j.1537-2995.2012.03591.x

TRANSFUSION **;**:**-**.

Volume **, ** ** TRANSFUSION 1

decision-making, adverse transfusion-associated events,and proper resource utilization highlight the importanceof evaluating proper transfusion practices.

Currently, the blood supply appears adequate withonly limited instances of spot shortages. In part, thisreflects the decline of elective surgical procedures duringthe Great Recession and, in part, reflects adaptation ofhospital-based blood conservation programs. However,this trend may reverse as the population ages.22

The PBM programs highlight multiple observational,cohort studies showing higher morbidity and mortalityin patients transfused at Hb levels greater than 8.0 to9.0 g/dL and two prospective, randomized controlledtrials documenting better outcomes in patients assignedto restrictive transfusion strategy groups, that is, transfu-sion threshold hemoglobin (Hb) concentrations less than7.0 to 8.0 g/dL.2,4,5,23,24

Moreover, wide variability in blood utilization forseemingly similar surgical procedures suggests a subjec-tive rather than evidence-based approach to blood trans-fusion ordering practices. For example, in a study byBennett-Guerrero and colleagues23 blood transfusionrates among hospitals ranged from 7.8% to 92.8% inpatients undergoing coronary artery bypass surgery. Casemix explained only 20.1% of the variation among hospitalsin RBC usage. Even though the investigators did notobserve differences in outcomes that could be attributedto variations in transfusion practices, it is obvious thathigher use of blood transfusions provides no additionalbenefit and contributes to higher costs and unnecessaryexposure to transfusions risks. In a randomized controlledtrial of transfusion requirements after cardiac surgery,Hajjar and colleagues4 demonstrated that transfusingeven a single RBC unit was associated with a significantlyincreased risk of serious postoperative morbidity andmortality. RBC transfusion was independently associatedwith a 1.2-fold increased risk of death at 30 days for eachunit transfused.

It is assumed that average posttransfusion Hbincrease is 1.0 g/dL per unit and that posttransfusionlevels greater than 10.0 g/dL suggests either that the pre-transfusion Hb level was above current threshold sug-gested by TRICC and FOCUS studies or that the patientwas transfused with an excessive number of RBC units.2,24

As such, a Hb level at discharge (discharge Hb) greaterthan 10.0 g/dL, or even 9.0 g/dL, in a transfused patientserves as a surrogate marker for excessive transfusion.

This study sought to determine the prevalence of dis-charge Hb of greater than 10.0 g/dL in patients receivingRBC transfusions at hospitals served by the CommunityBlood Center of Greater Kansas City (Kansas City, MO). Inaddition, a focused analysis of transfusion practices at aninner city, safety net, teaching hospital sought to investi-gate the characteristics of transfusion practices amongmedical specialty groups. Combined, we postulated that

we could predict the potential transfusion utilizationimpact of PBM programs and identify specific processimprovement opportunities.

MATERIALS AND METHODS

The community-based aspect of the study involved solici-tation to all hospitals served by the Community BloodCenter of Greater Kansas City, to provide the last Hb testresults, before discharge, for all patients receiving transfu-sion at their institution during a defined time period andfive hospitals responded, with variations in the extent ofdata provided. The transfusion services identified patientsby hospital service. Data analysis consisted of displayingRBC-transfused patients with “discharge” Hb levelsgreater than 10.0 g/dL as a percentage of RBC recipients.The patient-focused study occurred at a single tertiarycare, safety net, inner city, 283-bed teaching hospital witha Level I trauma center that uses only leukoreduced RBCs.

We evaluated patient records retrospectively and cat-egorized them into subspecialty groups based on patientdiagnoses and the specialty managing the patients’ care.Data obtained from patient records of those receiving ablood transfusion over a 6-month period included pre-transfusion Hb, number of units transfused, posttransfu-sion Hb, and discharge Hb (the last recorded Hb value).Patients were grouped by the medical specialty orderingthe RBC transfusions.

We performed statistical analyses using one-wayanalysis of variance with SNK post hoc analysis usingcomputer software (Prism, GraphPad, La Jolla, CA). Weconsidered values significant when p values were lessthan 0.05.

The suggested trigger point for RBC transfusion at thefocus hospital was a Hb level of 8.0 g/dL or lower. Exclu-sions included patients with active bleeding if the Hb wasexpected to drop below the suggested trigger point orpatients with comorbidities such as cardiovascular dis-eases, sepsis, and respiratory insufficiency. There are nocitywide guidelines for the use of RBCs although hospitalsutilize a Hb level of 8.0 g/dL as an acceptable trigger. Thepatient-focused study received approval from the TrumanMedical Center and University of Missouri–Kansas CityInstitutional Review Board.

RESULTS

The participating hospitals ranged in bed size from 146 to504 and included records from 250 to 1881 patients perhospital. Discharge Hb levels greater than 10.0 g/dLoccurred among 27% to 47% of transfused patients(Fig. 1). This occurred in 36% to 61% of patients treated onthe family practice service, 30% to 52% on the medicalservice, 38% to 52% of those treated by nephrologists, 16%to 50% of oncology inpatients, and 13% to 52% of ortho-pedic surgery patients.

EDWARDS ET AL.

2 TRANSFUSION Volume **, ** **

For the single focus hospital subjected to detailedstudy, we obtained 475 patient records from adultpatients, 429 of which underwent analysis. We excludedrecords of 46 subjects that lacked complete information(predominately pretransfusion and/or discharge Hblevels). To determine transfusion trends throughout spe-cialties and patient groups, mean Hb value was deter-mined at several time points (Table 1). Although severalfactors influence a physician’s decision to initiate RBCtransfusion, pretransfusion Hb is the major quantifiablefactor. The mean pretransfusion Hb level was 7.3 g/dL forall groups combined. The distribution of pretransfusionHb at the focus hospital is given in Fig. 2. PretransfusionHb in trauma patients was significantly greater (p < 0.05)compared to all other specialties individually. Likewise,surgery patients had significantly (p < 0.05) greater pre-transfusion Hb levels when compared to inpatient inter-nal medicine, oncology, and obstetrics/gynecology.

A general trend between pretransfusion Hb and dis-charge Hb exists in all groups, with the exception ofsurgery and trauma (Fig. 3A). Unlike all other groups,trauma patients actually had slightly higher pretransfu-sion Hb values than discharge Hb values, presumablyreflecting the temporal relationship of injury and resusci-tation. Specifically a significant (p < 0.05) and strong cor-relation exists between pretransfusion Hb and dischargeHb (r2 = 0.72; Fig. 3B). We found no significant correlationbetween the number of units transfused and dischargeHb (data not shown). Furthermore, no significant differ-ence was observed when measuring units transfusedin patients grouped into low (<8.0 g/dL), intermediate(8.0-10.0 g/dL), or high (>10.0 g/dL) discharge Hb (datanot shown). Of note, a Hb concentration greater than

10.0 g/dL at discharge occurred in 27%of patients (Figs. 4 and 5). Discharge Hbwas greater than 9.0 g/dL in 50.3% ofpatients. The mean posttransfusion Hbwas 9.3 g/dL and the mean Hb at dis-charge was 9.2 g/dL. The distribution ofdischarge Hb is shown in Fig. 5. We didnot collect the data on time intervalbetween last transfusion and dischargebut there does not appear to be a mean-ingful change from posttransfusion todischarge Hb levels. In patients withpretransfusion Hb levels above 7.0 g/dL(within 1.0 g/dL of the trigger point) wefound that 82% were transfused with 2or more units of RBCs. In 76% of trans-fusions, the number of RBC units trans-fused comprised an even number(Fig. 6).

When comparing “elective” transfu-sions and those due to acute bleeding,the discharge Hb was above 10 g/dL in

27.8% of elective cases and in 32.7% of patients withacute bleeding. The elective cases included patients withunspecified chronic anemia, renal disease, and miscella-neous internal medicine patients. The group of patientswith acute bleeding included trauma, surgery, and gas-trointestinal (GI) bleed. It should be noted that if surgeryand trauma are examined separately, only 22.7% of thepatients had a discharge Hb level greater than 10 g/dL. Asshown in Fig. 4 nearly 50% of the GI bleed patients had adischarge Hb level of greater than 10.0 g/dL.

DISCUSSION

The notion of designating a discharge Hb level of greaterthan 10.0 g/dL, in a hospitalized patient who receivedtransfusions, as an indication of excessive transfusionseems to be a reasonable one, given that posttransfusionHb levels above 9.0 g/dL have not been shown to be asso-ciated with additional benefit. In response to a needexpressed in the literature, we are proposing that a dis-charge Hb level of greater than 10.0 g/dL should beincluded in the factors taken into account for a properPBM.8 One could propose an even more stringent targetby considering a discharge Hb level of greater than9.0 g/dL as an indication of overtransfusion.

We note that discharge Hb levels greater than10.0 g/dL ranged from 27% to 47% among transfusedpatients. This suggests at least one RBC transfusion wouldhave been avoided by following transfusion guidelinesconsistent with the TRICC and FOCUS studies.2,24

The mean pretransfusion Hb level was 7.3 g/dL forpatients receiving treatment at the focus hospital. Wefound a significant correlation between pretransfusion

Fig. 1. Percentage of transfused patients with “final” Hb level of greater than

10.0 g/dL by service. Final = latest Hb level before discharge; FP = family practice;

Med = medicine; Neph = nephrology; Onc = oncology; Ortho = orthopedic surgery.

PBM AND DISCHARGE Hb

Volume **, ** ** TRANSFUSION 3

TAB

LE

1.D

ata

fro

mfo

cus

ho

spit

alo

nly

Spe

cial

tyA

dmit

Hb

leve

l(g/

dL)

Pre

tran

sfus

ion

Hb

leve

l(g/

dL)

RB

Cun

itstr

ansf

used

Pos

ttran

sfus

ion

Hb

leve

l(g/

dL)

Dis

char

geH

ble

vel(

g/dL

)F

emal

eM

ale

Age

(yea

rs)

Ren

al8.

7�

2.8

7.0

�1.

62.

6�

2.0

9.0

�1.

09.

1�

1.5

1713

53.2

Mis

cella

neou

s9.

6�

2.4

7.4

�1.

02.

1�

0.8

9.3

�1.

49.

7�

1.7

1112

57.2

GI

8.7

�2.

57.

1�

1.4

2.9

�1.

89.

5�

1.4

9.7

�1.

432

4156

.2S

ickl

ece

ll9.

8�

3.7

6.8

�0.

72.

8�

2.1

9.6

�1.

99.

5�

2.1

53

58.4

Sur

gery

10.5

�1.

98.

1�

1.6

3.1

�2.

39.

8�

1.2

9.4

�1.

326

1753

.4In

patie

ntin

tern

alm

edic

ine

8.3

�2.

56.

7�

1.4

2.4

�1.

29.

5�

1.4

9.4

�1.

346

4751

.0Tr

aum

a11

.8�

2.3

9.7

�2.

93.

3�

2.3

10.3

�2.

09.

3�

1.8

630

33.9

Obs

tetr

ics/

gyne

colo

gy8.

2�

3.0

6.9

�2.

22.

4�

1.3

9.0

�1.

28.

8�

1.1

520

36.6

Onc

olog

y7.

9�

2.2

6.9

�1.

62.

8�

2.7

8.8

�1.

38.

8�

1.1

1818

60.9

Sep

sis

9.7

�3.

76.

8�

0.7

2.8

�2.

19.

6�

1.9

9.5

�2.

05

258

.4

Tota

l8.

9�

2.7

7.3

�1.

92.

7�

2.7

9.3

�1.

59.

2�

1.5

232

197

49.0

Fig. 2. Distribution of pretransfusion Hb at the focus hospital.

A

B

Fig. 3. (A) Data from focus hospital only. Mean pretransfusion

Hb ( ) and discharge Hb (�) levels among patient groups.

The mean pretransfusion Hb level was greater than discharge

Hb level in trauma. (B) Data from focus hospital only. When

all patient groups are combined, a strong and significant

relationship exists between pretransfusion Hb level and

discharge Hb level. IM = inpatient internal medicine;

Misc. = miscellaneous; Ob/Gyn = obstetrics/gynecology.

EDWARDS ET AL.

4 TRANSFUSION Volume **, ** **

and discharge Hb (Fig. 3B) similar to a recently reportedstudy by others, except trauma (Fig. 3A).25 The guidelineswere followed and each patient received a mean of 2 to 3RBC units (Table 1). Interestingly, 82% of patients who had

pretransfusion Hb levels above 7.0 g/dL received 2 or moreunits of RBCs. We found that an even number of RBCswas transfused in 76% of the instances. Thus, 2 or 4 unitsare being ordered and/or transfused where 1 or 3 units,respectively, may have been sufficient. It bears pointing outthat not that many years ago, single-unit RBC transfusionswere considered inappropriate and triggered a review. Therationale, then, was that if a patient needed an RBC trans-fusion, at least 2 units should be administered to provide ameaningful increase in oxygen-carrying capacity.

Although several factors, including comorbidities,length of stay, treatments given, transfusion, and underly-ing morbidities, contribute to a patient’s Hb at discharge,we attribute a discharge Hb level of greater than 10.0 g/dLas being indicative, in part, of overly aggressive transfu-sion practices. Interestingly, our finding that 27% to 47% oftransfused patients had discharge Hb levels greater than10.0 g/dL is relatively similar to the 34% reported inpatients undergoing elective orthopedic surgery.26 Amongthe group of actively bleeding patients (trauma, surgery,and GI), all had high pretransfusion Hb levels but only GIpatients had discharge Hb levels of greater than 10.0 g/dLin appreciably large numbers. It may be that in traumaand surgery patients, the treating physician could morereadily observe the extent of bleeding and blood loss,whereas in GI bleeds the blood loss being hidden fromobservation may have prompted the clinician to err on theside of “safety” by overestimating the amount of blood lossand thus “overtransfusing.” We noted that the number ofRBC units transfused did not correlate with high dischargeHb; however, the relationship between pretransfusion Hblevels and discharge Hb levels suggests that providers maybe initiating transfusions when the Hb level has not fallenbelow the optimum trigger point. Thus, “overtransfusion”may be more accurately called premature initiation oftransfusion in patients who may not have needed trans-fusion. Providers generally did not check Hb and/or Hctafter each transfusion and usually ordered RBC units ineven numbers, which probably resulted in more RBCtransfusions than may have been warranted.

The rate of overtransfusion, that is, discharge Hb levelof greater than 10 g/dL in patients with elective transfu-sions was 27.8%. This rate is not meaningfully differentfrom the overall overtransfusion rate of 27% at the focushospital. If GI bleed patients are excluded, the rate of over-transfusion in acutely bleeding patients due to traumaand surgery was appreciably lower at 22.7%. The over-transfusion in elective cases, that is, transfusion forchronic anemia, may be due to the desire to top off thepatient and prolong the interval between transfusions.

Our study was not designed to correlate Hb level toclinical outcomes; rather, it was designed to identifyopportunities for improvement and gain an understand-ing of current practices. Data addressing acute versuschronic transfusion situations were not collected.

Fig. 4. Data from focus hospital only. Group data indicating

total cases, total cases with discharge Hb levels of greater than

10.0 g/dL (�), and percent discharged with Hb levels greater

than 10.0 g/dL ( ). GI, Misc., and sepsis had the three highest

proportions of cases with discharge Hb levels above 10.0 g/dL.

IM = inpatient internal medicine; Misc. = miscellaneous;

Ob/Gyn = obstetrics/gynecology.

Fig. 5. Distribution of discharge Hb level at the focus hospital.

Fig. 6. Distribution of the number of RBC units transfused at

the focus hospital.

PBM AND DISCHARGE Hb

Volume **, ** ** TRANSFUSION 5

We infer that implementation of RBC transfusionguidelines based on a Hb “trigger” of 7.0 g/dL will result insignificant transfusion utilization declines. Current RBCutilization excess may approach 10% to 20% (midpointrange of 27% to 47%, 50% effectiveness). We suggest thatimplementing strategies correlating pretransfusion Hbdata with transfusion orders, coupled with increasing thefrequency of single-unit transfusion episodes will achievethe projected 10% to 20% RBC utilization savings.

ACKNOWLEDGMENTS

The authors are grateful to Ms Shay Weber for her kind coopera-

tion in data collection. We are also grateful for the support from

Sarah Morrison Fellowship, University of Missouri–Kansas City

School of Medicine.

REFERENCES

1. Marik PE, Corwin HL. Efficacy of red blood cell transfusion

in the critically ill: a systematic review of the literature. Crit

Care Med 2008;36:2667-74.

2. Hebert PC, Wells G, Blajchman MA, Marshall MJ, Martin C,

Pagliarello G, Tweeddale M, Schweitzer I, Yetisir E. A multi-

center, 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;340:409-17.

3. 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;42:812-8.

4. Hajjar LA, Vincent J, Galas FR, Nakamura RE, Silva CM,

Santos MH, Fukushima J, Kalil Filho R, Sierra DB, Lopes

NH, Mauad T, Roquim AC, Sundin MR, Leão WC, Almeida

JP, Pomerantzeff PM, Dallan LO, Jatene FB, Stolf NA, Auler

JO Jr. Transfusion requirements after cardiac surgery. The

TRACS Randomized Controlled Trial. JAMA 2010;304:1559-

67.

5. Bracey AW, Radovancevic R, Riggs SA, Houston S, Cozart

H, Vaughn WK, Radovancevic B, McAllister HA Jr, Cooley

DA. Lowering the hemoglobin threshold for transfusion in

coronary artery bypass procedures: effect on patient

outcome. Transfusion 1999;39:1070-7.

6. Walters JH, Ness PM. Patient blood management: a

growing challenge and opportunity. Transfusion 2011;51:

902-3.

7. Klein H. WHA Resolution 63.12–Overview of WHO Secre-

tariat Report and WHA Resolution. Presentation to Fortieth

Meeting Advisory Committee on Blood Safety and Avail-

ability, Department of Health and Human Services. June 7,

2011. Bethesda, Maryland.

8. Barr PJ, Donnelly M, Cardwell CR, Parker M, Morris K,

Bailie KEM. The appropriateness of red blood cell use and

the extent of overtransfusion: right decision? Right

amount? Transfusion 2011;51:1684-94.

9. O’Brien KL, Champeaux AL, Sundell ZE, Short MW, Roth

BJ. Transfusion medicine knowledge in Postgraduate Year 1

residents. Transfusion 2010;50:1649-53.

10. Wallis JP, Wells AW, Chapman CE. Changing indications

for red cell transfusion from 2000 to 2004 in the North of

England. Transfus Med 2006;16:411-7.

11. Drüeke TB, Locatelli F, Clyne N, Eckardt KU, Macdougall

IC, Tsakiris D, Burger HU, Scherhag A; CREATE Investiga-

tors. Normalization of hemoglobin level in patients with

chronic kidney disease and anemia. N Engl J Med 2006;

355:2071-84.

12. FDA/CBER. Fatalities reported to FDA following blood

collection and transfusion. Annual summary for fiscal

year 2007. 2008. [cited 2012 Feb 29]. Available from:

URL: http://www.fda.gov/BiologicsBloodVaccines/

SafetyAvailability/ReportaProblem/TransfusionDonation

Fatalities/ucm118316.htm

13. Toy P, Popovsky MA, Abraham E, Ambruso DR, Holness

LG, Kopko PM, McFarland JG, Nathens AB, Silliman CC,

Stroncek D; National Heart, Lung and Blood Institute

Working Group on TRALI. Transfusion-related acute

lung injury: definition and review. Crit Care Med 2005;33:

721-6.

14. Silliman CC, Ambruso DR, Boshkov LK. Transfusion-

related acute lung injury. Blood 2005;105:2266-73.

15. Vamvakas EC, Blajchman MA. Transfusion-related

mortality: the ongoing risks of allogeneic blood transfusion

and the available strategies for their prevention. Blood

2009;113:3406-17.

16. Heddle NM, Klama LN, Griffith L, Roberts R, Shukla G,

Kelton JG. A prospective study to identify the risk factors

associated with acute reactions to platelet and red cell

transfusions. Transfusion 1993;33:794-7.

17. Linden JV, Pisciotto PT. Transfusion-associated graft-

versus-host disease and blood irradiation. Transfus Med

Rev 1992;6:116.

18. Hillyer CD, Blumberg N, Glynn SA, Ness PM; NHLBI

Working Group in Transfusion Recipient Epidemiology and

Outcomes Research. Transfusion recipient epidemiology

and outcomes research: possibilities for the future. Trans-

fusion 2008;48:1530-7.

19. Hendrickson JE, Hillyer CD. Noninfectious serious hazards

of transfusion. Anesth Analg 2009;108:759-69.

20. Yazer M, Triulzi D. Messages from national blood data col-

lection reports. Transfusion 2007;47:366-8.

21. MacPherson J, Mahoney CB, Katz L, Haarmann J, Bianco

C. Contribution of blood to hospital revenue in the United

States. Transfusion 2007;47:114S-6S.

22. Drackley A, Newbold KB, Paez A, Heddle N. Forecasting

Ontario’s blood supply and demand. Transfusion. 2012;52:

366-74.

23. Bennett-Guerrero E, Zhao Y, O’Brien SM, Ferguson TB Jr,

Peterson ED, Gammie JS, Song HK. Variation in use of

blood transfusion in coronary artery bypass graft surgery.

JAMA 2010;304:1568-75.

EDWARDS ET AL.

6 TRANSFUSION Volume **, ** **

24. Carson JL, Terrin ML, Magaziner J, Noveck H, Sanders DW,

Chaitman BR, Rhoads GG, Nemo G, Dragert K, Beaupre L,

Hildebrand K, Macaulay W, Lewis C, Cook DR, Dobbin G,

Zakriya KJ, Apple FS, Horney RA, Magaziner J; FOCUS

Investigators. Liberal or restrictive transfusion in high-risk

patients after hip surgery. N Engl J Med 2011;365:2453-62.

25. Boucher BA, Hannon TJ. Blood management: a primer for

clinicians. Pharmacotherapy. 2007;27:1394-411.

26. 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. Transfu-

sion 2010;50:2117-24.

PBM AND DISCHARGE Hb

Volume **, ** ** TRANSFUSION 7