Perioperative Blood Transfusion and Blood Conservation in ... · surgery, routine use of a cell-saving device, and implemen-tation of appropriate transfusion indications. An important

PCTCGTVECJ

TBCSUFKLCCHB

pbvAgp

ri

*idpeorgjpp

FwS

ATBf

©P

REPORT FROM THE STS WORKFORCE ON EVIDENCE BASED SURGERY

erioperative Blood Transfusion and Bloodonservation in Cardiac Surgery: The Society ofhoracic Surgeons and The Society ofardiovascular Anesthesiologists Clinical Practiceuideline*

he Society of Thoracic Surgeons Blood Conservation Guideline Task Force:ictor A. Ferraris, MD, PhD (Chair), Suellen P. Ferraris, PhD, Sibu P. Saha, MD,ugene A. Hessel II, MD, Constance K. Haan, MD, MS, B. David Royston, MD,harles R. Bridges, MD, ScD, Robert S. D. Higgins, MD, George Despotis, MD, and

eremiah R. Brown, PhD

he Society of Cardiovascular Anesthesiologists Special Task Force on Blood Transfusion:ruce D. Spiess, MD, FAHA (Chair), Linda Shore-Lesserson, MD, Mark Stafford-Smith, MD,. David Mazer, MD, Elliott Bennett-Guerrero, MD, Steven E. Hill, MD, andimon Body, MB, ChBniversity of Kentucky Chandler Medical Center, Lexington, Kentucky (VAF, SPF, SPS, EAH), University of Florida, Jacksonville,lorida (CKH), University of Pennsylvania Health System, Philadelphia, Pennsylvania (CRB), Harefield Hospital, London, Unitedingdom (BDR), Rush Presbyterian St. Lukes’ Medical Center, Chicago, Illinois (RSDH), Washington University Medical Center, St.ouis, Missouri (GD), Center for the Evaluative Clinical Sciences, Dartmouth Medical School, Lebanon, New Hampshire (JRB), Virginiaommonwealth University, Richmond, Virginia (BDS), Montefiore Medical Center, Bronx, New York (LS-L), Duke University Medicalenter, Durham, North Carolina (MS-S, EB-G, SEH), Keenan Research Center in the Li Ka Shing Knowledge Institute of St. Michael’sospital, University of Toronto, Toronto, Ontario, Canada (CDM), and Brigham and Women’s Hospital, Harvard Medical School,
oston, Massachusetts (SB)
tibleA

rtovot(vatrpptts

Background. A minority of patients having cardiacrocedures (15% to 20%) consume more than 80% of thelood products transfused at operation. Blood must beiewed as a scarce resource that carries risks and benefits.

careful review of available evidence can provideuidelines to allocate this valuable resource and improveatient outcomes.Methods. We reviewed all available published evidence

elated to blood conservation during cardiac operations,ncluding randomized controlled trials, published observa-

The Society of Thoracic Surgeons Clinical Practice Guidelines arentended to assist physicians and other health care providers in clinicalecision-making by describing a range of generally acceptable ap-roaches for the diagnosis, management, or prevention of specific dis-ases or conditions. These guidelines should not be considered inclusivef all proper methods of care or exclusive of other methods of careeasonably directed at obtaining the same results. Moreover, theseuidelines are subject to change over time, without notice. The ultimate

udgment regarding the care of a particular patient must be made by thehysician in light of the individual circumstances presented by theatient.

or the full text of this and other STS Practice Guidelines, visit http://ww.sts.org/sections/aboutthesociety/practiceguidelines at the officialTS Web site (www.sts.org).

ddress correspondence to Dr Ferraris, Division of Cardiovascular andhoracic Surgery, University of Kentucky Chandler Medical Center, CTW
ldg, Suite 320, 900 South Limestone, Lexington, KY 40536-0200; e-mail:
[email protected].

2007 by The Society of Thoracic Surgeonsublished by Elsevier Inc

ional information, and case reports. Conventional methodsdentified the level of evidence available for each of thelood conservation interventions. After considering the

evel of evidence, recommendations were made regardingach intervention using the American Heart Association/merican College of Cardiology classification scheme.Results. Review of published reports identified a high-

isk profile associated with increased postoperative bloodransfusion. Six variables stand out as important indicatorsf risk: (1) advanced age, (2) low preoperative red blood cellolume (preoperative anemia or small body size), (3) pre-perative antiplatelet or antithrombotic drugs, (4) reopera-ive or complex procedures, (5) emergency operations, and6) noncardiac patient comorbidities. Careful review re-ealed preoperative and perioperative interventions thatre likely to reduce bleeding and postoperative bloodransfusion. Preoperative interventions that are likely toeduce blood transfusion include identification of high-riskatients who should receive all available preoperative anderioperative blood conservation interventions and limita-

ion of antithrombotic drugs. Perioperative blood conserva-ion interventions include use of antifibrinolytic drugs,elective use of off-pump coronary artery bypass graft

See Appendix 1 for authors’ financial relationships
with industry.
Ann Thorac Surg 2007;83:S27–86 • 0003-4975/07/$32.00doi:10.1016/j.athoracsur.2007.02.099

http://www.sts.org/sections/aboutthesociety/practiceguidelines

http://www.sts.org/sections/aboutthesociety/practiceguidelines

http://www.sts.org

stisht

sbpbt

elbvsmcpmo

1

TotdfiFo[m

hamtw

S28 BLOOD TRANSFUSION/CONSERVATION IN CARDIAC SURGERY FERRARIS ET AL Ann Thorac SurgSTS AND SCA PRACTICE GUIDELINE 2007;83:S27–86

urgery, routine use of a cell-saving device, and implemen-ation of appropriate transfusion indications. An importantntervention is application of a multimodality blood con-ervation program that is institution based, accepted by allealth care providers, and that involves well thought out

ransfusion algorithms to guide transfusion decisions.Conclusions. Based on available evidence, institution-

pecific protocols should screen for high-risk patients, aslood conservation interventions are likely to be mostroductive for this high-risk subset. Available evidence-ased blood conservation techniques include (1) drugs

hat increase preoperative blood volume (eg, erythropoi-

mswvp

otoegtatb

(gctsdadkrs

2D

C

1doidc

bEfmp

tin) or decrease postoperative bleeding (eg, antifibrino-ytics), (2) devices that conserve blood (eg, intraoperativelood salvage and blood sparing interventions), (3) inter-entions that protect the patient’s own blood from thetress of operation (eg, autologous predonation and nor-ovolemic hemodilution), (4) consensus, institution-spe-

ific blood transfusion algorithms supplemented withoint-of-care testing, and most importantly, (5) a multi-odality approach to blood conservation combining all

f the above.(Ann Thorac Surg 2007;83:S27–86)

© 2007 by The Society of Thoracic Surgeons

) Methods Used in Developing Guidelines

able 1 describes the methods used to quantify the typesf evidence available to answer relevant questions andhe classification system used to summarize recommen-ations about clinically important questions. This classi-cation system is the same as that used by the Joint Taskorce for Guidelines of the American College of Cardi-logy (ACC) and the American Heart Association (AHAavailable at: http://circ.ahajournals.org/manual/

anual_IIstep6.shtml]).It is apparent that any medical or surgical intervention

as both systematic and random effects, some of whichre beneficial and some of which are not. The practice ofedicine often necessitates a probabilistic balancing of

hese conflicting effects, recognizing that there will al-ays be a residual level of uncertainty. Jenicek [1] sum-

Abbreviations and Acronyms

ACC � American College of CardiologyACT � activated clotting timeADP � adenosine diphosphateAHA � American Heart AssociationANH � acute normovolemic hemodilutionASA � American Society of AnesthesiologistsCABG � coronary artery bypass graft surgeryCPB � cardiopulmonary bypassCUF � conventional ultrafiltrationDDAVP � desmopressin acetateEACA � epsilon-aminocaproic acidEPO � erythropoietinGPI � glycoprotein inhibitorHIV � human immunodeficiency virusICU � intensive care unitLMWH � low-molecular-weight heparinMUF � modified ultrafiltrationNIH � National Institutes of HealthOPCAB � off-pump coronary artery bypassPEEP � positive end-expiratory pressureRCT � randomized controlled trialSCA � Society of Cardiovascular

AnesthesiologistsSTS � The Society of Thoracic SurgeonsTQM � total quality managementTXA � tranexamic acid

arizes this probabilistic approach as follows: “. . . thecience of medicine becomes a structured and organizeday of using probability, uncertainty, and facts in pre-

entive medicine and clinical care to best benefit theatient and the community.”Evidence-based guidelines are an attempt to reconcile

ften conflicting lines of evidence, giving greater weighto evidence derived from more methodologically rigor-us studies and those for which the overall weight ofvidence is most convincing. They must be viewed asuidelines and recommendations, not absolutes. Withhis in mind, the authors searched numerous sources forvailable evidence about specific questions relating tohe use of blood transfusions and blood conservationefore, during, and after cardiac operations.The committees from The Society of Thoracic Surgeons

STS) and the Society of Cardiovascular Anesthesiolo-ists (SCA) participated in guideline development. Theommittee members’ financial relationships with indus-ry are listed in Appendix 1. Appendix 2 outlines theteps employed in development of the final guidelineocument. Within these guidelines, cost analysis was notprimary consideration in formulating the recommen-

ations. Costs change constantly depending upon mar-ets and other forces. The science was the focus, and theisks and benefits from bleeding and blood transfusionhould not vary with societal economic forces.

) Risks and Benefits of Blood Transfusion—theilemma

lass IIa

. Given that the risk of transmission of known viraliseases with blood transfusion is currently rare, fearsf viral disease transmission should not limit admin-stration of indicated blood products. (This recommen-ation only applies to countries/blood banks whereareful blood screening exists.) (Level of evidence C)

Arguably, the practice of modern transfusion medicineegan with the discovery of blood groups by Landsteiner.ven though his work won the Nobel Prize in 1930, the

ull impact of his discovery lagged; and other develop-ents such as arterial anastomosis (Carrel), blood com-

onent therapy, refrigeration of blood components, or-

http://circ.ahajournals.org/manual/manual_IIstep6.shtml

http://circ.ahajournals.org/manual/manual_IIstep6.shtml

Table 1. Classification Scheme Used to Summarize Clinical Recommendationsa

Class I Class IIa Class IIb Class III

Estimate of certainty(precision) oftreatment effect

Benefit �� RiskProcedure/treatment SHOULD be

performed/administered.

Benefit �� Risk: additionalstudies with focused objec-tives needed.

IT IS REASONABLE to performprocedure/administer treatment.

Benefit � Risk: additional studies withbroad objectives needed; additionalregistry data would be helpful.

IT IS NOT UNREASONABLE to performprocedure/administer treatment.

Risk � Benefit: no additionalstudies needed.

Procedure/treatment shouldNOT be performed/administered AS IT IS NOTHELPFUL AND MAY BEHARMFUL.

Level AMultiple (3–5) popula-

tion risk strata eval-uated

General consistency ofdirection and magni-tude of effect

Recommendation that procedureor treatment is useful/effective

Sufficient evidence from multiplerandomized trials or meta-anal-yses

Recommendation in favor oftreatment or procedure beinguseful/effective

Some conflicting evidence frommultiple randomized trials ormeta-analyses

Recommendation’s usefulness/efficacyless well established

Greater conflicting evidence from multi-ple randomized trials or meta-analy-ses

Recommendation that procedureor treatment not useful/effec-tive and may be harmful

Sufficient evidence from multiplerandomized trials or meta-analyses

Level BLimited (2–3) popula-

tion risk strata eval-uated


Limited evidence from single ran-domized trial or nonrandomizedstudies


Some conflicting evidence fromsingle randomized trial ornonrandomized studies


Greater conflicting evidence from singleran domized trial or nonrandomizedstudies


Limited evidence from single ran-domized trial or nonrandom-ized studies

Level CVery limited (1–2) pop-

ulation risk strataevaluated


Only expert opinion, case studies,or standard-of-care


Only diverging expert opinion,case studies, or standard-of-care


Only expert opinion, case studies, orstandard-of-care


Only expert opinion, case studies,or standard-of-care

a Taken from the the AHA/ACC Manual for Guideline Writing Committees at http://circ.ahajournals.org/manual/manual_llstep6.shtml.

S29A

nn

Th

oracSu

rgB

LO

OD

TR

AN

SFUSIO

N/C

ON

SER

VA

TIO

NIN

CA

RD

IAC

SUR

GE

RY

FER

RA

RIS

ET

AL

2007;83:S27–86

STS

AN

DSC

AP

RA

CT

ICE

GU

IDE

LIN

E

http://circ.ahajournals.org/manual/manual_llstep6.shtml

gut

stpic(afbrbwpr

atrtdclrctt

b5drttiat

atatrspb[

mab1Labmt1abf

ooiAmcbgd[pweiaRhBicctst

T

I

I

T


anization of blood banks, use of anticoagulation, and thergency of treating war-injured patients were necessary

o bring transfusion into the modern era.As early as 1943, it was recognized that blood transfu-

ion could spread diseases, especially hepatitis [2]. Sincehat time, other problems such as risks associated withaid donors and concerns about disease transmission,

ncluding the current epidemic of human immunodefi-iency virus/acquired immunodeficiency syndromeHIV/AIDS) and transmission of hepatitis C, raisedwareness of the problems associated with blood trans-usion (Table 2), almost to the point that the benefits oflood transfusion are overlooked. It is important toealize that the viral and parasitic infectious risks oflood transfusion are dramatically increased in third-orld countries or in areas where modern blood bankingractices are not available (http://www.cochrane.org/eviews/en/ab002042.html).

Transfusion of red cells causes immunomodulation,lthough the extent and type of immune deficit is con-roversial. In some blood banks, leukocyte reduction ofed cells is used routinely. With non–leukocyte reducedransfusions in randomized trials, multiorgan failure andeath occur in as many as 10% of transfused intensiveare unit (ICU) patients, versus 5% in recipients ofeukocyte reduced transfusions [3–5]. Transfusion-elated acute lung injury may be the most commonomplication of transfusion, although this remains con-roversial. As many as 1 in 20 patients experienced a

able 2. Risks of Blood Transfusion

TypeOccurrence in Red Blood

Cell Units Transfused

nfectiousHuman immunodeficiency virus

[669, 670]1 in 1.4–2.4 � 106

Hepatitis B [669] 1 in 58,000–149,000Hepatitis C [670] 1 in 872,000–1.7 � 106

Bacterial infection [671] 1 in 2,000mmunologic reactions

Febrile nonhemolytic transfusionreactions [672–675]

1 in 100

Anaphylactic transfusionreactions [676]

1 in 20,000–50,000

ABO mismatch [677]Hemolysis 1 in 60,000Death 1 in 600,000Leukocyte-related target organ

injury [3–5, 678]1 in 20 to 1 in 50

Transfusion-related acute lunginjury [679]

1 in 2000

Post-transfusion purpura [680] Rareransfusion services errorDonor screening error [669]

(malaria, T cruzi, babesioses,Creutzfeld-Jakob disease)

1 in 4 � 106

Transfusion services error [681] 1 in 14,000

ransfusion-related death in some studies [3–5]. Blum- a

erg and coworkers [3] suggest that as many as 1 in 40 or0 patients receiving non–leukocyte reduced transfusionsied in excess of that seen in recipients of leukocyteeduced transfusions. The added cost of leukocyte reduc-ion may limit universal acceptance of this technique, buthere is some evidence that red cell transfusion-relatedmmunomodulation is a cause of significant morbidity,nd this morbidity may be reduced by leukocyte reduc-ion of red cells.

It is difficult to define the benefits of blood transfusion,s randomized trials to support the use of blood productso treat disease do not exist. Blood transfusion wasccepted long before the complications associated withransfusion could be documented. Many traumatic inju-ies (especially war-related injuries) were almost univer-ally fatal before the advent of blood transfusion. Theractice of blood transfusion saved countless lives longefore the complications of this therapy were recognized

6].Enhanced oxygen-carrying capacity [7], improved he-ostasis associated with blood component therapy [8, 9],

nd volume support of cardiac output are three acceptedenefits of blood transfusion. Adams and associates, in942 (through the pioneering blood banking work of Johnundy) [10, 11], on the basis of clinical observations [10]nd animal studies [11], introduced the “10/30” rule oflood transfusion. These authors suggested that theinimal ideal level of oxygen-carrying capacity is main-

ained by a hematocrit of around 30% and hemoglobin of0 g/dL. Because of the risks of transfusion with associ-ted costs, and lack of clear evidence regarding theenefit of blood transfusion, the 10/30 arbitrary rule has

allen into disfavor.There is lack of clear evidence regarding the benefit

f blood transfusion. However, clinical reports [12–14]f survival benefit support transfusion in certain clin-

cal situations. A task force of the American Society ofnesthesiologists (ASA) developed a consensus state-ent based mostly on level B and C evidence that

oncluded that “red blood cell transfusions should note dictated by a single hemoglobin ‘transfusion trig-er’ but instead should be based on the patient’s risk ofeveloping complications of inadequate oxygenation”

15]. They developed guidelines for transfusion ofacked red cells in adults that were accepted by othersithout much argument and with little high-level

vidence to support them. Their guidelines are listedn Table 3. The ASA guidelines do not specificallyddress the uniqueness of the cardiac surgery patient.evised ASA guidelines are available on line at:ttp://www.asahq.org/publicationsAndServices/CTGuidesFinal.pdf. Because of the lack of random-

zed trials to define the role of blood transfusion inardiac surgery and because of concerns about compli-ations of blood transfusion, it is reasonable to reviewhe available evidence supporting transfusion deci-ions for cardiac operations. The aim of this review iso provide clinically useful guidelines, based on avail-
ble evidence, to aid cardiothoracic surgeons and an-
http://www.cochrane.org/reviews/en/ab002042.html

http://www.cochrane.org/reviews/en/ab002042.html

http://www.asahq.org/publicationsAndServices/BCTGuidesFinal.pdf

http://www.asahq.org/publicationsAndServices/BCTGuidesFinal.pdf

ev

3O

a

Iasacsamdppfappdp

ttv

bAC

1(nuaipeumd

pducbtrltsiistltlrFy

TT

●

●

●

●

●

S31Ann Thorac Surg BLOOD TRANSFUSION/CONSERVATION IN CARDIAC SURGERY FERRARIS ET AL2007;83:S27–86 STS AND SCA PRACTICE GUIDELINE

sthesiologists in their decisions about blood conser-ation and transfusion.

) Causes of Blood Transfusion After Cardiacperations

) The “High-Risk” Patient

n formulating evidence-based guidelines, it is reason-ble to identify high-risk subsets of patients who con-ume most of the blood products during and after oper-tion and who are the most likely to benefit from bloodonservation interventions. A typical profile of transfu-ion for a cardiac practice is shown in Figure 1 [16]. Therere at least three remarkable features of Figure 1. First,ore than 50% of patients undergoing cardiac proce-

ures receive no allogeneic blood transfusion. Second,atients who receive more than 10 donor units of bloodroducts are in the 90th percentile of the patient trans-

usion profile. Third, 10% to 20% of patients consumebout 80% of the total blood product transfusions in thisopulation. Consistently, there is a high-risk subset ofatients who require large amounts of blood productsuring their cardiac procedures. By identifying high-riskatients and altering transfusion practices in this group,

able 3. American Society of Anesthesiology Guidelines forransfusion of Packed Red Cells in Adults [682]

Transfusion for patients on cardiopulmonary bypass withhemoglobin level � 6.0 g/dL is indicated [173, 174, 683, 684].Hemoglobin level � 7.0 g/dL in patients older than 65 yearsand patients with chronic cardiovascular or respiratorydiseases justifies transfusion [177, 685].For stable patients with hemoglobin level between 7 and 10g/dL, the benefit of transfusion is unclear [7, 174, 175].Transfusion is recommended for patients with acute bloodloss more than 1,500 mL or � 30% of blood volume.Evidence of rapid blood loss without immediate controlwarrants blood transfusion.

he variability in transfusion practices is expected torend toward standard levels and ultimately conservealuable resources.

) Multivariate Predictors and Observational Variablesssociated With Blood Transfusionlass I

. Preoperative identification of high-risk patientsadvanced age, preoperative anemia, small body size,on–coronary artery bypass graft surgery [CABG] orrgent operation, preoperative antithrombotic drugs,cquired or congenital coagulation/clotting abnormal-ties, and multiple patient comorbidities) should beerformed, and all available preoperative and periop-rative measures of blood conservation should bendertaken in this group as they account for theajority of blood products transfused. (Level of evi-

ence A)

Numerous reports identify multivariate predictors ofostoperative bleeding and blood transfusion after car-iac operations. Additionally, multiple publications doc-ment anecdotal or observational reports of acquired,ongenital, or process variables associated with increasedleeding. Table 4 is a summary of these publications with

he variables separated by patient-related, procedure-elated, and process-related factors. Very few studiesisted in Table 4 differentiate between red blood cellransfusion and non–red cell hemostatic factor transfu-ion. Wherever this distinction was made, it was takennto account in the table. A minority of the studies listedn the table provide odds ratios or p values to support thetrength of a particular risk factor as a predictor of bloodransfusion. As a consequence, the factors are simplyisted without an attempt to add a quantitative measureo each risk factor. Many of the risk factors in Table 4 areisted as a single factor, although it is realized that theisk may be a spectrum spanning low to high intensity.or example, the risk associated with age greater than 75ears is significantly greater than risk for a patient aged

Fig 1. Pareto diagram of the distribution ofnonautologous blood units transfused in 4,445cardiac surgical patients during a 4-year pe-riod. Eighty-six percent of the units transfusedare consumed by 20% of the patients who re-ceive more than 10 donor units per operation.(Bars � frequency; balls � cumulativepercent.)

TA

P

P

P

C


able 4. Multivariate and Observational Risk Factors Associated With Increased Bleeding or Blood Transfusion During orfter Cardiac Operations

Risk Factor

References SuggestingRisk Factor Is

Related to Transfusionof Red Blood Cells

References Suggesting RiskFactor Is AssociatedWith Transfusion ofNon–Red Blood Cell

Hemostatic Components

atient-related variablesAdvanced age or age more than 70 years 20, 26, 28, 143, 467,

686–692377, 687, 693

Preoperative anemia 7, 20, 26, 28, 49, 143,526, 686, 688, 689,691, 694, 695

7, 693, 695

Female gender 26, 28, 86, 143, 467,686, 688, 693, 696

76, 86, 377, 686, 688, 693,697

Body size or body surface area 20, 26, 28, 49, 86, 143,166, 467, 686, 688,691, 692

166, 691

Preoperative antithrombotic therapyHigh intensity (abciximab, clopidogrel, direct thrombin

inhibitors, low-molecular-weight heparin, long-actingdirect thrombin inhibitors, thrombolytic therapy)

54, 81, 103–107, 124,125, 698–712

54, 81, 104–107, 467, 698–701, 703–705, 707–709,712–715

Low intensity (aspirin, dipyridamole, eptifibatide, tirofiban) 17, 18, 87, 90, 467, 702,716–722

17, 18, 87, 89, 90, 377, 717,719, 720, 722

Preoperative coagulopathyHereditary coagulopathy or platelet defect (von

Willebrand’s disease, Hermansky-Pudlak, Bernard-Soulier, Scott, Werlhof, Glanzmann’s, hemophilia A or B,clotting factor deficiencies, etc)

22, 723–735 22, 723–735

Acquired coagulopathy or platelet abnormality (nonspecificplatelet defect measured by template bleeding time orPFA-100, chronic lymphocytic leukemia, cirrhosis, lupusanticoagulant, drug-related polycythmia vera,myelodysplastic syndrome, ITP, beta thalassemia, etc).

20, 21, 23, 25, 49, 736–741

21, 23, 25, 736–741

Cardiogenic shock, congestive heart failure, or poor leftventricular function

28, 143, 689, 691, 695 691

Renal insufficiency 28, 49, 691 691Insulin-dependent adult-onset diabetes mellitus 28, 742 28, 49, 76, 143, 697, 743Peripheral vascular disease 28Preoperative sepsis 744, 745 744, 745Liver failure or hypoalbuminemia 28, 746 746

rocedure-related variablesProlonged CPB time 45, 49, 377, 526, 686,

691–693, 695, 747377, 691, 693

Reoperation 28, 143, 377, 467, 689 377, 691Type of operation (other [aortic, complex, etc] � valve/CABG

� valve � CABG)377, 467, 695, 747 377

Increased protamine dose after CPB 377 377Increased cell-saving volume 377 377Intraoperative autologous donation 526, 693 693Need for transfusion while on CPB 377, 686, 687, 695 377, 686, 687, 695Use of polymerized starch for volume expansion 686, 748 686, 748

rocess-related variablesLack of transfusion algorithm with point-of-care testing 45, 276, 654–659, 749–

752276, 654–659, 751, 752

Use of internal mammary artery (either one or two) 747Reduced heparin dose 753 753Low body temperature in intensive care unit 377 377

ABG � coronary artery bypass graft surgery; CPB � cardiopulmonary bypass; ITP � idiopathic thrombocytopenic purpura.

5ttbttavd

tdaWraCibupomocra

apl(drn

cC

15opwacae

t[a2ieaavat

(hdfsf[w[t

p3ebqa

dTbwbcpifsatbawvfccpfiaa[

eIfccatn[Orcbp


5 years or younger, and it is quite likely that the risk ofransfusion associated with age is not a continuous func-ion. Likewise, the risk associated with some antithrom-otic drugs (eg, aspirin) is low on the risk spectrum while

hat of others (eg, clopidogrel) is much higher. No at-empt was made to measure the risk spectrum of vari-bles listed in Table 4 as the evidence base for theseariables as risk factors is limited and few quantitativeescriptors of risk are available.The studies listed in Table 4 are far too heterogeneous

o lend themselves to meta-analyses. The level of evi-ence in most cases is low, but consensus was used tossign importance to each of these predictor variables.

ithin each of the three major categories (patient-elated, procedure-related, or process-related), the vari-bles are arranged in descending order of importance.ertain of the variables stand out as high-risk predictors

ncluding advanced age, preoperative anemia or smallody size (ie, low red blood cell volume), the duration orrgency of operative intervention, the presence of anti-latelet or antithrombotic drugs taken shortly beforeperation, and the presence of multiple noncardiac co-orbidities (eg, renal failure, diabetes mellitus, acquired

r congenital coagulopathy, and so forth). Any of theseommon risk factors should alert the clinician to bleedingisk and should trigger blood conservation measuresimed at addressing the specific risks (see below).Special clinical situations call for special interventions

nd should be “red flags” for surgeons faced with theseroblems. Patients with acquired or congenital coagu-

opathies, patients scheduled for complex procedureseg, combined valve/coronary revascularization, aorticissection with deep hypothermic circulatory arrest),epeat cardiac procedures, sepsis with thrombocytope-ia, and Jehovah’s Witnesses fall into this category.

) Patient-Related Causes of Bleedinglass IIa

. Patients who have thrombocytopenia (less than0,000/mm2), who are hyperresponsive to aspirin orther antiplatelet drugs as manifested by abnormallatelet function tests or prolonged bleeding time, orho have known qualitative platelet defects representgroup at high risk for bleeding. Maximum blood

onservation interventions during cardiac proceduresre reasonable in these high-risk patients. (Level ofvidence B)

There is evidence that certain patients have an accen-uated response to the usual doses of antiplatelet drugs17]. Certain “hyperresponders” to average doses ofspirin exhibit very prolonged skin bleeding times [18–0]. This accentuated response to aspirin may result inncreased perioperative blood loss worsened by preop-rative antiplatelet therapy. The mechanisms of theseccentuated effects of aspirin and other antiplatelet drugsre undoubtedly related to individual pharmacogenomicariation and involve the antiplatelet, antiinflammatory,nticoagulant, and endothelial-protecting actions of
hese agents. m
Patients with thrombocytopenia from whatever causedefined as platelet count below 50,000) are at extremelyigh risk of excessive bleeding after CABG [21–25]. Ad-itionally, patients with preoperative anemia (eg, renal

ailure, repeated blood drawing during prolonged ICUtay, multiple recent percutaneous procedures, and soorth) exhibit increased perioperative blood transfusion20, 26–28]. One of the earliest observations about anemiaas that bleeding time was prolonged in anemic patients

29, 30]. Anemia-related bleeding abnormalities are likelyo be worsened by antiplatelet drugs [17].

Patients with other congenital or acquired qualitativelatelet defects are at increased bleeding risk [26–28,1–37]. Congenital defects include vonWillebrand’s dis-ase, Bernard-Soulier syndrome, Glanzmann’s throm-asthenia, storage-pool disease, and others. Acquiredualitative defects are seen in liver disease, renal diseasend drug induced qualitative platelet defects.

) Physician-Related Causes of Bleedinghere is no doubt that physician practices influenceleeding and blood transfusion. Surgical practices differidely and can dramatically influence perioperativeleeding and transfusion. Regardless of the patient’sondition at operation or immediately postoperatively,hysicians do not consistently apply the same triggers or

ndications for ordering blood transfusions. Use of trans-usions varies widely among centers [38–41]. Stover [2]howed the institution as well as the individual physicianre independent multivariate risk factors for allogeneicransfusion. Practice variability in the transfusion oflood products impacts resource utilization [42]. Johnsonnd associates [43] describe the comparison of outcomesith two different transfusion strategies (conservative

ersus liberal), and they concluded that no specific trans-usion trigger was justified but that the important indi-ator for transfusion should be patient symptoms andlinical condition. Variation in cardiopulmonary bypassractices, including time on bypass, influences platelet

unction and perioperative bleeding [44, 45]. Differencesn practice patterns relative to recognition, correction,nd exploration for excessive postoperative hemorrhagelso contribute to wide variability in transfusion practices46, 47].

) Procedure-Related Causes of Bleedingt is difficult to separate the procedure-related variablesrom those that are the result of physician practices, butertain procedure-related factors stand out. Repeat pro-edures have higher transfusion rates [48], and the typend urgency of operation are independent predictors forransfusion [49]. Hypothermia related to cardiopulmo-ary bypass influences platelet function and coagulation

50–52], and the effect can persist into the ICU [53].ff-pump cardiac surgery is associated with an overall

eduction in transfusion requirements [54–56]. Of theoronary artery bypass surgery patients, those who haveilateral internal mammary artery grafts suffer a greaterostoperative blood loss than saphenous vein or single
ammary artery graft patients [57]. Replacement of the

aiipHne6pa

fC

17abwwdatcab2tpos

C

1ltpgcuttdpima

C

1asid

pMi

apolpc

nptiHtwde2i

aab7amwG[pearmat

tra1ioautdteadea

tpacoa


ortic valve with a pulmonary autograft (Ross procedure)s a technically challenging operation, with up to 20%ncidence of postoperative bleeding necessitating reex-loration [58] and above average mortality rate [59].emostatic dysfunction and concomitant coagulation ab-ormalities are often seen in candidates for and recipi-nts of ventricular assist devices or artificial hearts [60–2]. These procedure-related variables influenceerioperative bleeding and should alert the surgeon andnesthesiologist to additional risk.

) Drug-Related Causes of Bleedinglass IIa

. It is reasonable to discontinue thienopyridines 5 todays before cardiac procedures to limit blood loss

nd transfusion. Failure to discontinue these drugsefore operation risks increased bleeding and possiblyorse outcome. Caution must be used with suddenithdrawal of antiplatelet therapy in the presence ofrug-eluting stents. That can lead to stent thrombosis,nd the surgical team should consider various alterna-ives to maintain stent patency. That could even in-lude hospitalization to convert thienopyridine ther-py to short-acting GP2b/3a inhibitors for a few daysefore operation. (Level of evidence B). It is reasonable to discontinue low-intensity an-iplatelet drugs (eg, aspirin) only in purely electiveatients without acute coronary syndromes beforeperation with the expectation that blood transfu-ion will be reduced. (Level of evidence A)

lass IIb

. Most high-intensity antithrombotic and antiplate-et drugs (including ADP-receptor inhibitors, directhrombin inhibitors, low molecular weight heparins,latelet glycoprotein inhibitors, tissue-type plasmino-en activator, streptokinase) are associated with in-reased bleeding after cardiac operations. It is notnreasonable to stop these medications before opera-

ion to decrease minor and major bleeding events. Theiming of discontinuation depends on the pharmaco-ynamic half-life for each agent as well as on theotential lack of reversibility. Unfractionated heparin

s a notable exception in that it is the only agent thatay be discontinued shortly before operation or not at

ll. (Level of evidence C)

lass III

. Dipyridamole is not indicated to reduce postoper-tive bleeding, is unnecessary to prevent graft occlu-ion after coronary artery bypass grafting, and mayncrease bleeding risk unnecessarily. (Level of evi-ence B)

Drug therapy has a growing role and influence in therevention and treatment of cardiovascular disease.any of the active cardiac drugs provide benefit by

nhibiting platelet function, or by causing clot lysis. Thus, o

ny surgical procedure required in the face of such drugsoses a greater than normal risk of perioperative hem-rrhage and transfusion requirement. A large body of

iterature defines the risk/benefit relationship of anti-latelet and anticoagulant drugs, and this subject isovered in detail below.

Stent implantation in coronary arteries and in saphe-ous vein grafts is accompanied by guideline-drivenharmacologic therapy to maintain patency and prevent

hrombo-occlusion [63–66], but with a relatively highncidence of bleeding and vascular complications [67, 68].

emorrhagic complications occur at reported rates be-ween 1.9% and 9.8% after placement of stents, withorse outcome among patients receiving transfusionsuring stent implantation [69–73]. Other studies showven higher rates of bleeding complications, at 21.8% to7.4%, with varying frequency of blood transfusion rang-ng between 0% and 9.5% [74, 75].

Preoperative antiplatelet and anticoagulant treatments prophylaxis for coronary occlusive disease is associ-ted with excessive intraoperative and postoperativeleeding as well as resultant transfusion, in most [20,6–81], but not all situations [82–86]. Therefore, thisspect of patients’ preoperative medication regimensust be managed for maximum cardioprotective benefit,hile minimizing risk of hemorrhagic complication.uidelines are available to aid in this decision process

87]. Patients with thrombocytopenia or with qualitativelatelet defects (eg, renal failure, vonWillebrand’s dis-ase, anemia, and so forth) are particularly sensitive tontiplatelet drugs and represent a high-risk subset thatequire multiple blood conservation interventions toinimize blood loss and transfusion. Discontinuation of

ntiplatelet and antithrombotic drugs before CABG inhese high-risk patients is prudent.

There is evidence that certain patients have an accen-uated response to the usual doses of preoperative aspi-in [17]. Certain “hyperresponders” to average doses ofspirin exhibit very prolonged skin bleeding times [18,9]. This accentuated response to aspirin may result inncreased perioperative blood loss [17]. The mechanismsf these effects of aspirin are undoubtedly multifactorialnd include the antiplatelet, antiinflammatory, anticoag-lant, and endothelial-protecting actions of aspirin. For

he vast majority of patients with acute coronary syn-romes who require urgent CABG, the best option seems

o be to continue aspirin until operation [88]. For thelective patient who requires CABG and does not haven acute coronary syndrome, it may be reasonable toiscontinue aspirin for a few days (2 to 3 days) with thexpectation that there will be less perioperative bleedingnd blood transfusion.Dipyridamole reduced postoperative blood loss and

ransfusions in one study, and was thought to preservelatelets during operative intervention [89]. Dipyrid-mole and aspirin in combination increased bleedingompared to placebo or aspirin alone [90 –93], and graftcclusion was reduced, but not eliminated [94]. Breyernd associates [95] demonstrated increased incidence
f delayed postoperative cardiac tamponade with rou-

trCai

rlotdftthiaameeaesff[dmdcApdowniTzpiwsaftpctiiwbo

flio

aitaoraoaitowoactna[

adwbb[auhdaffhCetroatLMabo

htahiaa(tPc


ine perioperative administration of aspirin and dipy-idamole. According to the Antithrombotic Trialists’ollaboration, the addition of dipyridamole to aspirindds no significant benefit over aspirin alone and mayncrease bleeding risk [96].

Thienopyridines are a class of antiplatelet drugs thateduce adenosine diphosphate (ADP)–mediated plate-et activation, with significant improvement of clinicalutcomes in many coronary and cardiovascular condi-ions. Ticlopidine was the first available thienopyri-ine, but initially, widespread use was limited by

requent side effects, as well as neutropenia andhrombotic thrombocytopenic purpura [97–99]. In con-rast, clopidogrel has a much better safety profile, andas become standard therapy after coronary stent

mplantation [97]. Because clopidogrel is so well toler-ted, and pretreatment before stent implantation isdvantageous to coronary artery patency, it is a com-on occurrence that patients undergoing urgent or

mergent coronary artery bypass surgery have recentxposure to dual antiplatelet therapy of clopidogrelnd aspirin. This dual therapy, although safe andffective during coronary intervention [100 –103], re-ults in higher postoperative bleeding, more trans-used blood products, and higher rate of reexplorationor mediastinal hemorrhage during emergency CABG104 –107]. To address this disturbing finding, Ley [108]escribed the implementation of a quality improve-ent initiative around preoperative exposure to clopi-

ogrel. The current ACC/AHA guidelines and theurrent STS guidelines recommend discontinuingDP-inhibitors 5 to 7 days before cardiac operations, ifossible, recognizing that operations sooner than 5ays in patients on ADP-inhibitors risk increased peri-perative bleeding and transfusions and possiblyorse long-term outcomes [87, 109]. However, with theew drug-eluting stents sudden withdrawal of platelet

nhibition appears to cause thrombotic risk [110, 111].he drug-eluting stents exhibit delayed endotheliali-ation compared with bare metal stents, and oncelatelet inhibition is removed, the foreign surface

nside the vessel wall creates a nidus for platelet andhite cell activation as well as a potential for hyper-

ensitivity reaction [112, 113]. Minimal evidence isvailable to guide therapy in this situation. Consensusavors some element of platelet inhibition in the pa-ient who requires CABG but has a coated stent inlace. Discussion with all members of the cardiovas-ular team including cardiologists, hematologists, andhe operating team is essential. Possible alternativesnclude shifting to a short half-life glycoprotein IIb/IIIanhibitor (GPI) or to a direct thrombin inhibitor whileaiting for the effects of thienopyridines to disperse,ut more evidence is required to determine the bestption.Platelet GPIs, introduced a decade ago, play a pro-

ound role in abolition of platelet aggregation and plate-et thrombus formation, and reduce the risk of acuteschemic complications after coronary occlusion and cor-
nary angioplasty. Coller and colleagues [114] reported h
n approximately twofold increased risk of major bleed-ng in patients treated with abciximab. Short-acting GPIs,irofiban and eptifibatide, in combination with aspirinnd heparin, are used during coronary intervention with-ut a significant increase in serious catheterization-elated bleeding events [115–117]. The longer acting GPI,bciximab, improves the clinical outcome of percutane-us coronary intervention, but causes thrombocytopeniand access site bleeding [118, 119]. Tirofiban and abcix-mab have comparable efficacy and bleeding complica-ions [120]. Limited evidence is available to guide the usef GPIs in patients with acute myocardial infarction orith unstable coronary syndromes who also need cardiacperations [87, 109]. These patients should be consideredt high risk for bleeding and blood transfusion, andlinical judgment regarding timing of intervention andransfusion needs is likely the most important determi-ant of outcomes. Again, guidelines based on consensusgreement are available to assist in this decision analysis87].

Heparin is an integral component of therapy forcute coronary events. Systematic review of six ran-omized controlled trials looking at heparin in patientsith acute myocardial infarction treated with throm-olytic therapy showed severe bleeding to be similaretween those receiving and not receiving heparin

121]. Low-molecular-weight heparins (LMWH) haven acceptable risk of bleeding in management ofnstable angina [122], and short-term unfractionatedeparin or LMWH is used in acute coronary syn-romes. Low-molecular-weight heparin after coronaryrtery stenting had a 10.5% incidence of hematomas oralse aneurysm, and a 3.25% incidence of blood trans-usion or surgical repair [123]. One study showedemorrhage and reexploration for bleeding afterABG to be significantly higher in patients receivingnoxaparin versus unfractionated heparin— but theiming of preoperative dosing is not described in thiseport [124]. Patients receiving LMWH within 12 hoursf cardiac surgery have significantly greater blood lossnd increased blood transfusion compared with pa-ients receiving intravenous heparin or a dose ofMWH more than 12 hours before operation [125].edalion and associates [126] showed that enoxaparin

dministered more than 8 hours before coronary arteryypass surgery is not associated with increased post-perative bleeding or transfusion requirement.Despite the almost universal use of unfractionated

eparin, especially during cardiopulmonary bypass,here remains concerns with this drug including hep-rin rebound, heparin resistance, protamine reaction,eparin-induced thrombocytopenia, and heparin-

nduced platelet dysfunction [127, 128]. Investigation oflternatives to unfractionated heparin resulted in thevailability of some direct-acting thrombin inhibitorseg, hirudin, bivalrudin, and argatroban) to limithromboses in patients with cardiovascular disease.otential advantages of direct thrombin inhibitors in-lude a more predictable anticoagulant response than
eparin. Based on randomized trials direct thrombin

itrbiabaxiilcpfp

iidnbdastwbBemaEhisppaiCswipdetidbOobrmbo

4T

aTstit1aetpbTasiaTcpii

bC

1cstn(2baam

C

1pcet

C

1ptd

ipa


nhibitors are superior to heparin in preventing arterialhromboses. For example, hirudin is superior to hepa-in in patients with acute myocardial infarction [129],ut safety concerns about increased bleeding and

nability to monitor anticoagulation limit use of thisgent. The shortest acting direct thrombin inhibitor,ivalirudin, has a half-life of 25 minutes, while that ofrgatroban is 45 minutes, hirudin 60 minutes, andimelagatran 3 hours. All of the direct thrombin inhib-tors are plagued by the lack of reversibility, difficultiesn monitoring the level of anticoagulation, and a pro-onged half-life compared with heparin. Argatroban isleared mainly by the liver, and its use is limited inatients with hepatic dysfunction, but may be pre-

erred for treatment of heparin-induced thrombocyto-enia in patients with renal dysfunction.Experience with direct thrombin inhibitors is limited

n patients having cardiac procedures after coronarynterventions [134]; however, a recent controlled ran-omized trial using bivalirudin during off-pump coro-ary artery bypass surgery (OPCAB) suggests thativalirudin is safe and effective as an anticoagulanturing OPCAB [135]. Likewise, argatroban was used asn anticoagulant during OPCAB with satisfactory re-ults [136]. Lack of a ready reversibility for directhrombin inhibitors is a major limitation, especiallyith longer acting agents, and can lead to catastrophicleeding when used in patients requiring CPB [137].ivalirudin is at least as effective as heparin in isch-mic heart disease, but with significant reduction inajor hemorrhage [130 –133]. The Randomized Evalu-

tion of PCI Linking Angiomax to Reduced Clinicalvents (REPLACE)-1 trial compared bivalirudin andeparin in patients undergoing percutaneous coronary

nterventions, and compiled a large prospective data-et of bivalirudin administered concomitantly withlanned GPIs. Occurrence of major bleeding was com-arable between bivalrudin and heparin in nonoper-ted patients, but experience with this agent is limitedn patients having cardiac procedures [134]. The Acuteatheterization and Urgent Intervention Triage

trategY (ACUITY) trial reported improved outcomeith decreased bleeding for percutaneous coronary

ntervention patients who received bivalirudin com-ared with unfractionated heparin. Trials of bivaliru-in for off pump and on pump cardiac surgery showedquivalent 30 day success rates compared with unfrac-ionated heparin [138 –140]. Bleeding was not dramat-cally different, although there appeared to be a ten-ency for early reexploration of the patients withivalirudin, as this was an unblinded series of studies.ne study from New Zealand showed that graft flow in

ff-pump CABG patients who received bivalirudin wasetter than in patients receiving unfractionated hepa-in for on-pump CABG [135]. Bivalirudin may proveost helpful in patients with heparin induced throm-

ocytopenia who require coronary revascularization or
ther cardiac procedures [141]. t
) Indications for Bloodransfusion—“Transfusion Trigger”

) General Considerationshe indications for blood product transfusion in cardiacurgery include treatment of coagulopathies and correc-ion of anemias with the ultimate goal of limiting bleed-ng and improving oxygen-carrying capacity. The indica-ions for transfusions evolved significantly since the early940s when subjective indications were developed toddress these theoretical considerations. In the modernra, the overall risk and benefit relationships of bloodransfusion therapy weigh the possible benefit of im-roved tissue oxygenation with the risk of transmittinglood-related diseases or inducing adverse reactions.he economic impact and cost of transfusion therapy inn environment in which blood can be a life-saving, yetcarce resource, mandates a careful understanding of thendications for blood transfusion and the development ofppropriate decision-making processes in this regard.he ultimate goal of any discussion regarding the indi-ations for blood component therapy is to maximizeatient benefit and limit risk when possible. A great deal

s known with regard to blood transfusion risks yet littles known of its benefits.

) Transfusion Triggerslass IIa

. With hemoglobin levels below 6 g/dL, red bloodell transfusion is reasonable, as this can be life-aving. Transfusion is reasonable in most postopera-ive patients whose hemoglobin is less than 7 g/dL, buto high-level evidence supports this recommendation.

Level of evidence C). It is reasonable to transfuse non–red cell hemostaticlood products based on clinical evidence of bleedingnd preferabley guided by point-of-care tests thatssess hemostatic function in a timely and accurateanner. (Level of evidence C)

lass IIb

. It is not unreasonable to transfuse red cells in certainatients with critical noncardiac end-organ ischemia (eg,entral nervous system and gut) whose hemoglobin lev-ls are as high as 10 g/dL, but more evidence to supporthis recommendation is required. (Level of evidence C)

lass III

. Transfusion is unlikely to improve oxygen trans-ort when the hemoglobin concentration is greater

han 10 g/dL and is not recommended. (Level of evi-ence C)

Objective evidence to support transfusion decisions isncompletely developed. The hemoglobin, hematocrit, orlatelet count at which red cell or platelet transfusion isdministered is often referred to as the “transfusion
rigger.” While surgeons, anesthesiologists, and critical

cthadutismogt

pphhqfradtmspmoaaomcpceghqnu[

tdao5dd[wbaoiprt

capratsrpcfTcdteh[cbgoiW

ssoopotpJacroWei

tocoaTsBtuctAfsob


are specialists developed a long-standing tradition ofransfusion when the hemoglobin falls below 10 g/dL orematocrit below 30%, this practice no longer appearsppropriate since these recommendations were not evi-ence based. There is little scientific support for relyingpon a specific hemoglobin or hematocrit level as a

ransfusion trigger. Likewise, estimates of blood loss andntravascular blood volume are often inadequate to as-ess the needs for transfusion. More advanced measure-ents such as whole body oxygen-carrying capacity,

xygen consumption, oxygen extraction ratios, and oxy-en delivery provide more accurate means to estimatehe need for red blood cell transfusions [142].

During cardiac operations, cardiopulmonary bypassrovides oxygenation and systemic perfusion in a non-hysiologic manner. Utilizing systemic heparinization,emodilution and nonpulsatile blood flow in addition toypothermia, creates a number of physiologic conse-uences that affect systemic physiology and blood

ormed elements [143]. With respect to blood transfusionequirements, the use of balanced physiologic solutionss a pump prime, leads to normovolemic hemodilutionuring cardiopulmonary bypass. The putative advan-

ages of hemodilution during bypass include improve-ents in tissue perfusion by reducing blood viscosity. In

pite of reductions in hemoglobin-bound oxygen trans-ort to tissues, adequate oxygen delivery to tissues is wellaintained with hematocrits well below baseline physi-

logic levels. Hemodilution may contribute to perioper-tive coagulopathy by diluting the concentration of co-gulation factors and platelets. Anemia at the conclusionf cardiopulmonary bypass may theoretically be detri-ental to the recovering myocardium in the context of

oronary revascularization [144]. However, one studyerformed in patients immediately after CPB showed nohange in myocardial lactate extraction or productionven with elective hemodilution to hemoglobin of 5.0/dL [145]. Studies done in the early 1980s suggest thatemodilution to hemoglobin of 5.0 g/dL provides ade-uate oxygen delivery during CPB, but these studieseed confirmation in patients with compromised ventric-lar function and using modern perfusion techniques

146, 147].The physiologic threshold at which oxygen consump-

ion starts to decrease because of insufficient oxygenelivery in humans is a hemoglobin level of 3 to 4 g/dL,n oxygen extraction ratio of 0.44, a mixed venous oxygenf 34 mm Hg, and a mixed venous oxygen saturation of6% [148]. Of interest, the lowest limit of critical oxygenelivery of 333 mL O2/min/m2, below which oxygenelivery suffers, is the same for all mammalian species

149]. In well-controlled euvolemic hemodilution studies,hole body shift to anaerobic metabolism (ie, fallingelow critical oxygen delivery threshold) might be thebsolute indicator for transfusion. In the context of cor-nary artery disease, moderate hemodilution during clin-

cal cardiopulmonary bypass is well tolerated by cardiacatients [146]. However, the coronary artery blood floweserve and oxygen extraction may be exceeded by ex-
reme hemodilution to less than 6 g/dL [150]. Weisel and e
olleagues [148] found that patients at risk for perioper-tive ischemia, presumably most patients having cardiacrocedures, showed a delay in myocardial metabolicecovery with substantial hemodilution when postoper-tive hemoglobin levels fell below 6.0 g/dL. However,ransfusion of banked blood was not controlled for in thistudy. Increases in cardiac output, oxygen extractionatio, and red cell transit time in capillaries are therimary compensatory mechanisms for reduced oxygenarrying capacity. Patients with compromised ventricularunction may not be able to increase cardiac output.here are few data for cardiac surgery patients to addressritical end-organ oxygen delivery and utilization in theiseased state, although there are studies that report that

issue oxygenation is maintained in healthy, normovol-mic persons with a hemoglobin of 6 to 7 g/dL andematocrits of 15% to 20% [151, 152]. Fang and colleagues

153] could not find an increased mortality rate amongardiac surgical patients until the perioperative hemoglo-in was 5.0 or less. Observational studies provide generaluidelines to guide practice, although there are a numberf clinical series that provide conflicting data, particularly

n specialized patient populations such as the Jehovah’sitness patients.The original report by Ott and Cooley [154] of cardiac

urgery in Jehovah’s Witness children demonstrated thatuch high-risk surgery was possible with a mortality ratef 9.4%. This report helped to define the standards forpen heart surgery in this group of patients. In reviews ofublished series between 1983 and 1990 involving 1,404perations on Jehovah’s Witnesses, anemia and lack ofransfusion influenced operative mortality for only 20atients (1.4%) [155]. A synopsis of 61 reports of 4,722

ehovah’s Witnesses identified 23 deaths due to anemia,ll but 2 of which were patients with hemoglobin con-entrations less than 5 g/dL [156]. Based upon theseeports, major cardiovascular procedures including cor-nary bypass surgery, are performed safely in Jehovah’sitness patients; and these reports provide level B and C

vidence that acute and chronic anemia can be toleratedn certain clinical situations.

Because of the lack of high-level evidence to guideransfusion decisions, recommendations rely on expertpinions. In 1988, the National Institutes of Healthonvened a consensus conference and published rec-mmendations for blood transfusion, platelet therapy,nd administration of fresh frozen plasma [157].hrough these activities and a number of other con-ensus conferences by the American Association oflood Banks and the American College of Physicians in

he early 1990s, recommendations regarding bloodtilization and transfusion practices provided signifi-ant focus and attention on this issue [158, 159]. In 1990,he Transfusion Practices Committee of the Americanssociation of Blood Banks issued guidelines for trans-

usion of patients undergoing coronary artery bypassurgery [160]. In 1994, the ASA convened a task forcen blood component therapy to develop evidence-ased guidelines on the proper indication for periop-
rative and peripartum administration of red blood

ctaPdbtomiw(rrc

apohfnssssT

cB

C

1pgdvel2gcsofracspoa

C

1ig

tithtirclatcdttdsDcpVBw1ec[alnuo

ahanptstthplhocbn

d[socrr


ells, platelets, fresh frozen plasma, and cryoprecipi-ate [161]. This task force reviewed a total of 1,417rticles, of which 160 articles were considered relevant.ublished evidence was considered relevant if it ad-ressed the perioperative or peripartum use of thelood components and measured effectiveness in

erms of clinical outcomes. Identifying the primarybjective of red blood cell transfusion as the improve-ent of inadequate oxygen delivery, the task force

dentified two assumptions as the principles uponhich decisions to transfuse patients should be based:

1) surgical patients experience adverse outcomes as aesult of diminished oxygen carrying capacity, and (2)ed blood cell transfusions, by enhancing oxygen-arrying capacity, can prevent adverse outcomes.

After separating the effects of hypovolemia andnemia related to blood loss in this literature, it ap-ears that the absolute lowest threshold for hematocritr anemia in patients can not be established; however,emoglobin concentrations as low as 7 g/dL were

ound to be adequate to provide tissue oxygenation inormal, healthy, euvolemic persons. The NIH Consen-us Conference concluded that the evidence did notupport the use of a single criterion for transfusionuch as a transfusion trigger of 10 g/dL [157]. Theummation of consensus recommendations is shown inable 3.

) Indications for Transfusion on Cardiopulmonaryypass

lass IIa

. During cardiopulmonary bypass with moderate hy-othermia, transfusion of red cells for a hemoglobin of 6/dL or less is reasonable except in patients at risk forecreased cerebral oxygen delivery (ie, history of cerebro-ascular accident, diabetes mellitus, cerebrovascular dis-ase, carotid stenosis), in which case, higher hemoglobinevels may be justified. (Level of evidence C). In the setting of hemoglobin values exceeding 6/dL while on CPB, it is reasonable to transfuse redells based on the patient’s clinical situation, and thishould be considered the most important componentf the decision-making process. Indications for trans-usion of red blood cells in this setting are multifacto-ial and should be guided by patient-related factors (ie,ge, severity of illness, cardiac function, or risk forritical end-organ ischemia), the clinical setting (mas-ive or active blood loss), and laboratory or clinicalarameters (eg, hematocrit, SVO2, electrocardiographicr echocardiographic evidence of myocardial ischemia,nd so forth). (Level of evidence C)

lass IIb

. For patients on CPB with risk for critical end-organschemia/injury, it is not unreasonable to keep the hemo-
lobin level at 7 g/dL or more. (Level of evidence C) u
Much has been written regarding the need for improvedransfusion indicators in cardiac surgery, particularly dur-ng cardiopulmonary bypass [41]. Many reports have tiedhe decision to transfuse blood components (or the lowestematocrit on bypass) to postoperative outcome as a means

o either confirm the benefits of blood conservation ordentify predictors of poor outcome [162, 163]. Among theseeports, a number of prospective single-center and multi-enter reports offer insight to the management of hemodi-utional anemia on cardiopulmonary bypass. In 1997, Fangnd colleagues [153] reported the results of a single institu-ion in 2,738 coronary artery bypass graft operations, con-luding that after accounting for differences in patient andisease characteristics, a lowest hematocrit value of less

han or equal to 14% for low-risk patients and a value of lesshan or equal to 17% for high-risk patients was an indepen-ent risk factor for mortality. In a comprehensive databasetudy from the Northern New England Cardiovascularisease Study Group, the authors reported that of 6,980

onsecutive patients who underwent coronary artery by-ass at six medical centers in Maine, New Hampshire, andermont and at the Beth Israel Deaconess Hospital inoston between 1996 and 1998, those whose hematocrithile on cardiopulmonary bypass was allowed to fall below

9% had higher in hospital mortality rates, higher intraop-rative use of balloon support, and more frequent return toardiopulmonary bypass after initial attempts at separation164]. Women and patients with small body surface areand patients who were anemic before operation were mostikely to be severely anemic on cardiopulmonary bypass. Ofote, neither of these studies examined transfusion, thesual physician response to low hematocrit, as a covariater possible cause of the increased mortality.Several reports suggest worse outcome associated with

nemia during CPB. In a recent report evaluating lowematocrit during cardiopulmonary bypass, Karkoutind coworkers [165] evaluated the relationship betweenadir hematocrit during cardiopulmonary bypass anderioperative stroke while adjusting for variables known

o have an association with stroke and anemia. In pro-pectively evaluated patients (10,949 consecutive pa-ients) who underwent coronary artery bypass with ex-racorporeal circulation from 1999 to 2004, nadirematocrit during cardiopulmonary bypass was an inde-endent predictor of perioperative stroke. After control-

ing for confounding variables, each percent decrease inematocrit was associated with a 10% increase in thedds of suffering perioperative stroke. The authors con-luded that there is an independent direct associationetween degree of hemodilution during cardiopulmo-ary bypass and a risk of stroke.In a more recent review of the effects of low hematocrit

uring cardiopulmonary bypass, Habib and coworkers166] reported a retrospective analysis of operative re-ults and resource utilization in 5,000 consecutive cardiacperations with cardiopulmonary bypass. Stroke, myo-ardial infarction, low cardiac output, cardiac arrest,enal failure, prolonged ventilation, pulmonary edema,eoperation due to bleeding, sepsis, and multiorgan fail-
re were all significantly increased as lowest hematocrit

vssihhwaeseclptmrfipfcsb

seShfathfpwfitptlt(tsmdYaTaidp[cmfalu

st

tnptiowcar

dC

1brchei2batm

pcippcpetihnrl3

cjcdvpifa1rar


alue decreased below 22%. Consequently, hospitaltays, operative costs, and operative deaths were alsoignificantly greater as a function of hemodilution sever-ty. Long-term survival is improved among patients withigher hematocrits on CPB, suggesting that increasedemodilution severity during CPB is associated withorse perioperative outcomes. Like the reports by Fang

nd coworkers, Habib and associates did not review theffects of transfusion on outcome in their particulartudy. Two reports [167, 168] examined large databases tovaluate the effects of anemia on renal function duringardiac procedures. There was an association betweenowest hematocrit on CPB and the increased incidence ofostoperative renal dysfunction. However, the use of

ransfusion to treat low hemoglobin or to prevent he-odilutional anemia on bypass multiplied the risks of

enal dysfunction twofold to 3.5-fold. Because of thesendings, some authors suggest that transfusion of bloodroducts worsens outcomes during CPB [140], but trans-

usion may just be a marker of disease severity, not aause of poor outcome. More prospective randomizedtudies are required to amplify the relationship betweenlood transfusion and poor outcome after CPB.Because of the possible association of blood transfu-

ion on CPB with worse outcomes, several investigatorsvaluated this possibility with observational studies.urgenor and associates [169] showed that althoughemodilutional anemia increases the risk of low-output

ailure after cardiac procedures, an additional risk-djusted increase of 27% occurs with the transfusion of 1o 2 units of packed red blood cells regardless of nadirematocrit. Intraoperative packed red blood cells trans-

usion during CABG surgery seems to increase the risk ofostoperative low-output heart failure. Engoren and co-orkers [170] studied 1,915 patients who underwentrst-time isolated coronary artery bypass operations be-

ween 1994 and 1997 and found that 649 of the studyatients (34%) received a transfusion during their hospi-

alization. Transfused patients were older, smaller, moreikely to be female, and had more comorbidity. Theransfused patients also had twice the 5-year mortality15% versus 7%) of nontransfused patients. After correc-ion for comorbidities and other factors, transfusion wastill associated with a 70% increase in mortality. Byultivariate analysis, transfusion, peripheral vascular

isease, chronic obstructive pulmonary disease, Nework Heart Association cardiac functional class IV, andge were significant predictors of long-term mortality.he authors concluded that blood transfusions during orfter coronary bypass operations are associated withncreased long-term morbidity and mortality. Stroke andeath were increased in association with the utilization oflatelet transfusions in a study by Spiess and associates

167]. However, a more recent study by Karkouti andoworkers [171] did not find an excess morbidity orortality in CABG patients who received platelet trans-

usions. Patients who have lower hematocrits bleed morend otherwise receive more transfusions including plate-et transfusions. In a review of more than 15,000 patients
ndergoing CPB procedures at the Cleveland Clinic, a t
trong association was demonstrated between the use ofransfusions and postoperative infections [168].

All the data reviewed above are derived from observa-ional studies and therefore only document associations,ot cause and effect. Anemia may well drive a number ofhysiologic responses as well as physician behaviors (ie,

ransfusion). It is therefore unclear at this time howmportant any level of anemia is in creating organ failurer long-term adverse outcomes. Further, it is uncertainhat role blood transfusion plays in this complex pro-

ess. For these reasons, blood transfusions should bedministered with caution while adhering to guidelineecommendations whenever possible.

) Postoperative Indications for Transfusionlass IIa

. For patients after cardiac operations with hemoglo-in levels below 6 g/dL, red blood cell transfusion iseasonable and can be life-saving. Transfusion of redells is reasonable in most postoperative patients withemoglobin level of 7 g/dL or less, but no high-levelvidence supports this recommendation. (Level of ev-dence C). It is reasonable to transfuse non–red cell hemostaticlood products based on clinical evidence of bleedingnd preferably guided by specific point-of-care testshat assess hemostatic function in a timely and accurate

anner. (Level of evidence C)

Decisions about transfusion after operation in ICUatients are complex. It is obvious that the patient’slinical situation and disease status are important factorsn determining the need and indication for transfusion inatients undergoing coronary artery bypass surgery. Theatient’s volume status, pulmonary and cardiac status,erebrovascular status, the chronicity of anemia, theatient’s symptoms, the potential for blood loss, and thextent of surgery and risk of rebleeding all factor intohese decisions. Clinical indicators of hypovolemia andts secondary physiologic effects include tachycardia,ypotension, and oliguria. Physiologic indicators of sig-ificant impairments of the oxygen supply and demandatios include mixed venous oxygen saturation (SVO2)ess than 55%, or mixed venous oxygen tension less than0 mm Hg.Data to support postoperative transfusion decisions in

ardiothoracic practices are sparse. Utilizing clinicaludgment in the decision-making process, many clini-ians assume that anemia increases the risk of myocar-ial ischemia after CABG operations, although the pre-iously reviewed experience with Jehovah’s Witnessatients and other clinical reports offers a contrary opin-

on [155, 172]. Clinical indicators for postoperative trans-usion in coronary artery bypass graft patients taking intoccount their clinical and disease status were proposed in990, and updated in 1996, to provide guidance withespect to indications for transfusion [173, 174]. In thebsence of prospective randomized data, these expertecommendations provide a foundation upon which
ransfusion practices are based and still appear to be

g(

tHcsvfgsthgwgepat

a[icthp

dcsotCtlwtrpmro5awsbpcpggbbdb

5

aI

C

1nrher2ot(3(ttasw

cEPtd(CatisBpdiedfi

wsscsdbrfac

swfb


enerally applicable to modern cardiothoracic practiceTable 3).

More recent assessments of the impact of transfusionherapies are available. In a retrospective data analysis,

ébert and associates [175] found that patients withardiac disease admitted to the critical care unit had aignificantly higher risk of death with lower hemoglobinalues. To further elucidate the potential impact of trans-usions on mortality in critically ill patients admitted to aeneral ICU, the same investigators conducted a pro-pective, randomized, controlled trial [176] to determinehe outcome of a transfusion protocol that maintainedemoglobin level at 7 to 9 g/dL (strict protocol) or at 10/dL or more (liberal protocol). They reported that thereas no overall difference in mortality between the tworoups and that a restrictive protocol was at least asffective and possibly superior to the liberal transfusionrotocol. In a subset of patients who had known coronaryrtery disease, mortality was not different based uponransfusion behavior [177].

In a report from the Multicenter Study of Perioper-tive Ischemia Research Group, Spiess and coworkers178] reported that the risk of postoperative myocardialnfarction was highest in CABG patients with hemato-rit greater than 34% and with more severe left ven-ricular dysfunction, questioning the rationale for ad-ering to arbitrary transfusion thresholds in cardiacatients.Based upon nonrandomized observational studies,

escriptive case series, a few prospective randomizedlinical trials, and expert panel opinions, consensusuggests that red blood cell transfusion to improvexygen transport when a hemoglobin level is greaterhan 10 g/dL is almost never of benefit. The NIHonsensus Conference and the practice guidelines of

he ASA concluded that patients with hemoglobinevels greater than 10 g/dL did not require blood,hereas most patients with hemoglobin levels less

han 7 g/dL benefit from transfusion [157, 161]. In aecent prospective randomized trial of transfusion ofatients after CABG, in-dwelling tissue oxygen probeseasured the effect of transfusion with 1 or 2 units of

ed blood cells [179]. Transfusion did not improvexygen delivery to tissues, but increasing FiO2 from0% to 100% did improve tissue oxygen delivery. Othernimal work confirmed this finding [180, 181]. Suchork in hemorrhagic rat models did show that vital

igns (arterial pressure, central venous pressure) andlood gases (systemic arterial and mixed venous) im-roved by transfusion of banked blood. With standardritical monitoring, transfusion seems to improvehysiologic variables (blood pressure, arterial bloodases, mixed venous blood gases), but the tissue oxy-en delivery may actually be reduced by typicalanked blood [181]. Much more work needs to be doneefore the physiologic benefit of blood transfusion isetermined. Only then can realistic and evidence-
ased transfusion triggers be established. n
) Interventions to Limit Blood Transfusion

) Pharmacologic Agents) HEMOSTATIC DRUGS WITH ANTIFIBRINOLYTIC PROPERTIES

lass I

. High-dose aprotinin is indicated to reduce theumber of patients requiring blood transfusion, toeduce total blood loss, and to limit reexploration inigh-risk patients undergoing cardiac operations. Ben-fits of use should be balanced against the increasedisk of renal dysfunction. (Level of evidence A). Low-dose aprotinin is indicated to reduce the numberf patients requiring blood transfusion and to reduce theotal blood loss in patients having cardiac procedures.Level of evidence A). Lysine analogues like epsilon-aminocaproic acidEACA) and tranexamic acid (TXA) are indicated to reducehe number of patients who require blood transfusion, ando reduce total blood loss after cardiac operations. Thesegents are slightly less potent blood-sparing drugs and theafety profile of these drugs is less well studied comparedith aprotinin. (Level of evidence A)

Three drugs that play a role in perioperative bloodonservation and that limit fibrinolysis are in clinical use.psilon-aminocaproic acid (EACA [Amicar; Xanodyneharmaceuticals, Newport, KY]) inhibits fibrinolysis by

he inhibition of plasminogen inhibitors and to a lesseregree through antiplasmin activity. Tranexamic acid

TXA [Cyklokapron; Pharmacia & Upjohn, Somersetounty, NJ]) is similar in action to epsilon-aminocaproiccid but it is approximately 10 times more potent. Thesewo drugs are classified as lysine analogues because theynhibit plasminogen by binding to the lysine bindingites on the plasminogen molecule. Aprotinin (Trasylol;ayer Pharmaceuticals, West Haven, CT) is a bovinerotein that inhibits proteases with active serine resi-ues, especially plasmin, resulting in the attenuation of

nflammatory responses, fibrinolysis, and thrombin ger-nation. Aprotinin and the lysine analogues have veryifferent modes and scope of action but ultimately inhibitbrinolysis by limiting the action of plasmin.Since the early 1990s, antifibironlytic therapies have beenidely adopted for cardiac surgery. However with wide-

pread adoption, there is much concern regarding theaftety and efficacy of these drugs (http://www.fda.gov/der/drug/advisory/aprotinin.htm) [182, 183]. For this rea-on, the risks and benefits (ischemic complications versusecreased blood loss and transfusion) of these agents muste weighed and continually evaluated. Therefore, guidelineecommendations must not only summarize the evidenceor the use of the natural serine protease inhibitor aprotinin,nd of the lysine analogues, but must also identify safetyoncerns of these agents.

Heightened awareness of using drugs for blood con-ervation in cardiac operations surfaced in the late 1980sith the observation that aprotinin, given to humans in a

ar higher dose than previously used, was able to reduceleeding to the point where the majority of patients did
ot require transfusions. That was seen even in patients
http://www.fda.gov/cder/drug/advisory/aprotinin.htm


h[anrbbilr4od

bkpabpsatblrlp

p7iascdhfiohfl22stlfatvp

rTitmo1

CotTtcifbw(dtAt

Fcnltrtd


aving repeat surgery through a previous sternotomy184], a patient population regarded at that time as beingt a very high risk for transfusion. Subsequently, a largeumber of studies investigated the effects of various doseegimens of aprotinin, TXA, and EACA on postoperativeleeding and blood transfusion. Arguably, no otherlood conservation interventions have been as well stud-

ed and reported in the literature as have the antifibrino-ytic drugs, especially aprotinin. For example, at least 49andomized, controlled trials encompassing more than,000 subjects were identified that addressed the efficacyf aprotinin in limiting the number of patients transfuseduring cardiac procedures.A search of MEDLINE, EMBASE, and CINHAL data-

ases for the period January 1980 to July 2006 with theey words aprotinin, Trasylol, Iniprol, Midran, Cycloka-ron, tranexamic acid, Amicar, epsilon-aminocaproiccid, and EACA in combination with coronary-artery-ypass, myocardial revascularization, aortocoronary by-ass, valve, aortic or mitral surgery found more than 600tudies investigating the effect of lysine analogues andprotinin on postoperative bleeding. There are roughlyhree times as many articles published that address thelood conservation properties of aprotinin than of the

ysine analogues. At least four meta-analyses address theelative benefits of the three antifibrinolytic drugs inimiting blood loss and blood transfusion after cardiacrocedures [185–188].More than 70 trials compared aprotinin and placebo in

atients having cardiac procedures using CPB. Of these0 trials, 28 were randomized studies with adequatenformation to compare outcomes. Twenty-eight evalu-ble studies compared high-dose aprotinin to controlubjects in more than 4,000 patients having cardiac pro-edures. Meta-analysis showed a highly significant re-uction in blood transfusion in aprotinin-treated patientsaving CABG alone (Fig 2). Similar results were found inve studies [184, 189–192] of patients having redo CABGperations and in eight studies [193–200] of patientsaving valve or valve/CABG operations. The evidence

avoring reduced transfusion is somewhat less compel-ing, but still robust, for lower doses of aprotinin [201–11] or for dosing of aprotinin in the CPB circuit only [201,04, 207, 212–219]. The amount of blood transfused wastudied in 16 trials of patients receiving EACA, and in 35rials of patients receiving TXA. Similar benefits for theysine analogues were observed in the number of trans-usions and the number of patients transfused. For evalu-ble studies of the lysine analogues, significant reduc-ions in transfusions were seen in patients having aariety of cardiac procedures using CPB compared withlacebo-treated patients [215, 218, 220–226].Another important indicator of operative blood loss is

eturn to the operating room for control of hemorrhage.his outcome measure is an infrequent occurrence but

ndicates excessive bleeding. Exploration for postopera-ive hemorrhage is associated with serious postoperative

orbidity, and limiting this complication should improveperative mortality and morbidity [220, 227]. There were
7 high-dose aprotinin studies (14 CABG only and 3 e
ABG/valve) and 11 lysine analogue studies (6 CABGnly, and 5 CABG/valve) that reported the rate of returno the operating room for postoperative hemorrhage.able 5 summarizes the available evidence about return

o the operating room for postoperative hemorrhage afterardiac procedures. Aprotinin is associated with a signif-cant reduction in the rate of return to the operating roomor control of postoperative hemorrhage. No similarenefit in rate of return to the operating room is seenith the lysine analogues.

1) Meta-analyses of antifibrinolytic drug therapy. Indepen-ent randomized controlled trials (RCTs) demonstrate

he blood-saving benefits of antifibrinolytic therapies.lthough RCTs provided the initial evidence for use of

hese agents, meta-analyses are used to confirm their

ig 2. Relative risk ratios with 95% confidence intervals for primaryoronary artery bypass graft surgery (CABG) and high-dose aproti-in (840 mg). Forest plot of randomized trials (First authors are at

eft) comparing blood transfusion in high-dose aprotinin-treated pa-ients versus placebo-treated patients undergoing CABG only (noeoperations included). Studies to the left of midline favor aprotininherapy. There is a highly significant benefit from aprotinin in re-ucing the number of patients transfused after CABG.

fficacy and to assess safety. Four meta-analyses summa-

rl1sswddo[oatmq

cmwaspc

ilwTc(rIhLb

Rsb[l0L

fEwt

mfs3TfAmb

oaramfSrw[taabapatTafiea(ptfrtptartcwtStttad

TPA

AT

a

a

C


ize the primary efficacy of antifibrinolytic therapies inimiting blood transfusion during cardiac operations [47,86, 187, 228]. More importantly, these meta-analysesummarize the rare secondary outcomes (mortality,troke, myocardial infarction, and renal complications)here independent trials lack the statistical power toetect harm or benefit. For example, only one indepen-ent randomized trial reported a significant occurrencef renal dysfunction in patients treated with aprotinin229], whereas a recent meta-analysis and two largebservational studies suggest that high-dose aprotinin isssociated with an increased incidence of renal dysfunc-ion [182, 183, 230]. Despite the efforts by past RCTs and

eta-analyses to evaluate safety data of these agents,uestions remain about their safety and efficacy.The most recent meta-analysis is that of Sedrakyan and

oworkers [186]. These authors build on the previouseta-analyses [47, 187, 228] by updating previous trialsith more recent RCTs. All of the available meta-

nalyses analyzed trials, some of which were RCTs, toummarize the primary and secondary outcomes foratients treated with aprotinin, EACA, or TXA duringardiac operations using CPB.

The meta-analysis by Munoz and associates [47], donen 1999, showed that aprotinin and EACA significantlyimited postoperative blood transfusion and that EACAas a viable and cost-effective alternative to aprotinin.hese authors showed that high-dose aprotinin signifi-antly reduced packed red blood cell transfusion by 72%0.28%, 0.22% to 0.37%), whereas EACA reduced packeded blood cell transfusion by 68% (0.32%, 0.15% to 0.69%).n addition, Munoz and associates [47], showed thatigh-dose aprotinin reduced total blood loss by 53%.ow-dose aprotinin and EACA reduced total blood lossy 35%, significantly less than high-dose aprotinin.In 2001, Henry and associates [187] added current

CTs to those found by Munoz and associates [47] andhowed that aprotinin reduced the need for transfusiony 31% (relative risk [RR]: 0.69; 95% confidence interval

95%CI]: 0.63 to 0.76). This was true for pump prime dose,ow-dose, and high-dose aprotinin: RR 0.85 (0.73 to 0.99),.69 (0.59 to 0.80), and 0.68 (0.62 to 0.75), respectively.

able 5. Rates of Return to Operating Room forostoperative Hemorrhage in Patients Treated Withprotinin and Tranexamic Acida

TreatmentNo

Treatment

RR CI

Reexplore Reexplore

Yes No Yes No

protinin 18 1,099 56 1,043 0.316 0.188–0.53ranexamicacid

14 354 17 362 0.848 0.429–1.674

Results are pooled from 17 high-dose aprotinin studies and 11 tranex-mic acid studies.

I � confidence interval; RR � relative risk.

ikewise, TXA significantly reduced transfusion by 29% s

or all cardiac procedures. However, the evidence forACA was limited and was not significantly associatedith reduced transfusion (RR: 0.48; 95%CI: 0.19 to 1.19) in

his meta-analysis.In 2004, Sedrakyan and associates [186] reported aeta-analysis of patients having isolated CABG that

urther updated the RCTs that address the efficacy andafety of antifibrinolytic drugs. These authors examined5 RCTs encompassing 3,879 patients having CABG.hey found that aprotinin significantly reduced the need

or transfusion by 39% compared with control patients.protinin treatment had no effect on operative mortality,yocardial infarction, or renal failure but did have a

eneficial effect on limiting postoperative stroke.In these meta-analyses, high-dose aprotinin was the

nly agent shown to reduce the risk of reexploration forny cause (Table 5). Neither TXA nor EACA significantlyeduced the reexploration rate. In general, these meta-nalyses found that none of the agents reduced mortality,yocardial infarction, stroke, thrombosis, or renal dys-

unction. Two notable exceptions were the study byedrakyan and associates [186] that found a significanteduction in stroke for isolated CABG patients treatedith aprotinin, and the study by Munoz and associates

47] that found a trend toward increased renal dysfunc-ion in aprotinin-treated patients. Recently, Brown andssociates [230] updated the meta-analysis of Munoz andssociates [47] and found an excess of renal dysfunction,ut not renal failure, in aprotinin-treated patients. Brownnd coworkers [230] found that 5 of every 100 cardiacatients treated with aprotinin may have an increasebove control of postoperative serum creatinine of morehan 0.5 mg/dL during the early postoperative period.his increase in serum creatinine did not seem to bessociated with the development of renal failure. Thesendings support the earlier trial of D’Ambra and cowork-rs [229] and the observational studies of Mangano andssociates [182] and Karkouti and associates [183].2) Head-to-head comparisons of antifibrinolytic drugs. Aeculiarity of studies that compare antifibrinolytic agents

o placebo is that there is a wide variation in the trans-usion rates of control patients. For example, in oneecent study of 100 patients randomly assigned to apro-inin or placebo, 79% of control patients received bloodroducts compared with 53% of aprotinin-treated pa-

ients [231]. In another study of 75 patients randomlyssigned to EACA or placebo, 36% of control patientseceived blood products compared with 18% of EACA-reated patients [232]. Differences of this sort in theontrol transfusion rates suggest that these two studiesould not be suitable for a meta-analysis. In statistical

erms, this implies heterogeneity among the studies.tatistical tests are available to test the hypothesis that

he effect sizes among RCTs are equal. If the RCTs usedo perform a meta-analysis do not exhibit homogeneity,hat suggests a flawed meta-analysis [233]. None of thevailable meta-analyses that compare antifibrinolyticrugs to placebo report indices of homogeneity of effects
ize, and that is a shortcoming of these studies. This

smOmd

csapdowwatttAtaTatssticdlmc(rawtDtpr(TapadrlaKrfesrdo

redipidwathewas

nBrirthttTtpgal

haqnEmpTaPatecEhhc

mlpstirs


hortcoming makes it difficult if not impossible to deter-ine superiority of one antifibinolytic drug over another.ne way to address this shortcoming is to performeta-analysis on RCTs that compare one antifibrinolytic

rug directly to another in head-to-head comparisons.Unfortunately, the literature is sparse on head-to-head

omparisons of aprotinin with EACA or TXA. Fourteentudies, involving 1,057 patients, compared one or morentifibrinolytic drugs to another in head-to-head com-arisons among patients having various cardiac proce-ures [197, 198, 220, 222, 223, 234–243], and only the studyf Hardy and associates [220] compared EACA with TXA,hile all others compared aprotinin with TXA (9 studies),ith EACA (2 studies), or with both (3 studies). In nearly

ll of the studies that compared EACA with aprotinin,here was a suggestion that aprotinin resulted in lessransfusion and fewer numbers of patients transfusedhan either placebo or EACA [198, 223, 235, 240, 244].lthough the numbers are small, meta-analysis suggests

hat treatment with high-dose aprotinin is better thanny dose of EACA but is equivalent to treatment withXA (results not shown). Only 2 of 12 studies showed thatprotinin significantly reduced packed red blood cellsransfusion compared with TXA [197, 223]. None of the 13tudies that compared aprotinin with TXA or EACAuggested that either of the lysine analogues were betterhan aprotinin in reducing blood transfusion or in reduc-ng the number of patients transfused. Because of con-erns about blinding in these studies, because of wideose ranges in TXA-treated groups, because of apparent

ack of homogeneity, and because of small numbers,aking recommendations on the basis of these direct

omparisons is not justified.3) Safety of antifibrinolytic drugs in cardiac surgery. Twoecent observational studies [182, 183] and one meta-nalysis [185] suggest that aprotinin may be associatedith increased thrombotic risk, especially renal dysfunc-

ion, after cardiac procedures. That caused the Food andrug Administration to issue a safety alert suggesting

hat aprotinin should be used with caution and only foratients in whom the benefits of the drug outweighs theisks, mostly of renal dysfunction and of hypersensitivityhttp://www.fda.gov/cder/drug/advisory/aprotinin.htm).here has always been some concern about the effect ofprotinin on renal function because of its affinity for theroximal renal tubules [245, 246]. D’Ambra and associ-tes [229] observed a significantly higher rate of renalysfunction in patients undergoing valve surgery whoeceived aprotinin. This dysfunction appeared to be moreikely in patients with diabetes mellitus (16% versus 5%),

result supported by Mangano and associates [182].incaid and associates [247] in a retrospective logistic

egression analysis of 1,200 patients undergoing CABGound that the combination of angiotensin-convertingnzyme inhibitors plus aprotinin administration was as-ociated with an increase rate of acute renal failure (oddsatio 2.9, confidence interval: 1.4 to 5.8) while neitherrug alone was associated with this complication. Kark-
uti and associates [183], in a propensity score matched r
etrospective comparison of 898 patients who receivedither aprotinin or TXA, found that the incidence of renalysfunction and dialysis was higher among those receiv-

ng aprotinin (24% versus 17%, p � 0.01). It is likely thatatient comorbidities and drug-drug interactions involv-

ng aprotinin are associated with an excess of renalysfunction, suggesting that aprotinin should be usedith caution in patients with certain comorbidities such

s diabetes or with preexisting renal dysfunction. Pa-ients who receive certain preoperative medications thatave adverse renal effects such as angiotensin-convertingnzyme inhibitors may also be at risk for interactionsith aprotinin. Until further information becomes avail-

ble, the enhanced blood-sparing benefits of aprotininhould be applied with caution in certain patients.

Other safety issues affect the administration of aproti-in for blood conservation during cardiac procedures.ecause aprotinin is a bovine protein, hypersensitivityeactions upon administration to humans are a possibil-ty. Evidence suggests that a significant hypersensitivityeaction (50% of which are anaphylaxis) occurs in lesshan 0.1% of patients who have an initial exposure to theigh-dose regimen of aprotinin during cardiac opera-

ions. This rate rises to 5% if patients have a reexposureo aprotinin within 6 months of initial dosing [248, 249].he hypersensitivity reaction to initial exposure of apro-

inin is rare and is not greater than for other nonhumanrotein drug exposures. In some formulations of fibrinlue, aprotinin is used to prevent fibrinolysis. That can ben unrecognized cause of antibody formation and rarelyeads to anaphylactic reactions upon early reexposure.

Review of the safety issues of the antifibrinolytic drugsighlights an important point, namely most efficacy trialsre neither designed nor powered to uncover low fre-uency drug-related morbidity like renal failure, coro-ary graft occlusion, or other infrequent morbidities.ven trials designed to evaluate a specific drug-relatedorbidity may suffer from problems with statistical

ower. Few data exist to support the safety of EACA andXA, but no evidence suggests that excess complicationsre associated with use of these antifibrinolytic agents.hase IV studies are important to evaluate drug safetyfter regulatory approval [250, 251]. Additional phase IIIrials are needed to assess the safety of lysine analogues,specially with regard to renal dysfunction. There is aoncerning lack of reporting of renal complications forACA, TXA, and low-dose aprotinin. Likewise, moreigh-quality large phase III trials are needed to examineead-to-head comparisions of antifibrinolytic agents in-luding high- and low-dose aprotinin, EACA, and TXA.

Meta-analyses are effective at summarizing the treat-ent effects of reported outcomes of interest, but are

imited by the trial design and bias in recruitment ofatients. The meta-analyses reported to date do notupport the suggestions of Mangano and associates [182]hat aprotinin is associated with increased risk of mortal-ty, stroke, myocardial infarction, or dialysis-dependentenal failure. However, one meta-analysis supports theuggestions of these authors that there is an increased
isk of renal dysfunction, but not renal failure, associated

wct

TlOrrrlwadto

I

C

1agciba2amisd

C

1grpp

hrpcfmEcrww

[tpia

mqwcot

agMpcvmdbwTlpplotltbt

uadaLtanidlaaotir

I

C

1rfcat2u


ith the use of high-dose aprotinin during cardiac pro-edures [230]. Until more information is available, apro-inin should be used with caution.

In summary, high- and low-dose aprotinin, EACA, andXA are all effective at significantly reducing total blood

oss and the need for packed red blood cells transfusion.nly high-dose aprotinin has been shown to significantly

educe the risk for reexploration. None of the agentseduces mortality, myocardial infarction, thrombosis, orenal failure or renal dysfunction. Among only the iso-ated CABG patients, a significant reduction in strokeas observed in aprotinin treated patients. High-dose

protinin has been associated with increased risk of renalysfunction, but not renal failure. There is limited head-

o-head evidence to support the use of one agent over thether.

I) ERYTHROPOIETIN

lass IIa

. Recombinant human erythropoietin (EPO) is reason-ble to restore red blood cell volume in patients under-oing autologous preoperative blood donation beforeardiac procedures. However, no large-scale safety stud-es for use of this agent in cardiac surgical patients haveeen performed. Available evidence suggests an accept-ble safety profile. (Level of evidence A). Recombinant human erythropoietin is a reasonablelternative for anemic low-risk elective patients (he-oglobin < 13 g/dL) before cardiac procedures provid-

ng that EPO is given in conjunction with iron therapyeveral days or more before operation. (Level of evi-ence B)

lass IIb

. It is not unreasonable to use preoperative EPO,iven at least a few days before operation, to increaseed cell mass in elective patients who are at risk forostoperative anemia and depressed endogenous EPOroduction. (Level of evidence C)

Erythropoietin (EPO) is an endogenous glycoproteinormone that stimulates red blood cell production inesponse to hypoxia and anemia. Endogenous EPO isroduced by the kidney and its production is signifi-antly diminished in patients without adequate renalunction. Recombinant human EPO was developed in the

id 1980s and is commercially available in several forms.rythropoietin combined with oral iron therapy is indi-ated to treat anemia (hemoglobin levels � 13 g/dL) inenal failure, associated with chemotherapy or HIV, andhen given preoperatively, to reduce transfusions in aide range of operations [252].Abundant evidence, including four meta-analyses

253–256], exists to justify the preoperative administra-ion of EPO to reduce preoperative anemia especially inatients having autologous blood donation [257–260], and

n children [261, 262]. Erythropoietin appears to be safe
nd effective for the improvement of preoperative ane- s
ia, with the main side effect being hypertension. Ade-uate iron supplementation is required in conjunctionith EPO. A typical preoperative regimen of EPO is

ostly, and uncertainty exists about the cost effectivenessf EPO for patients undergoing autologous blood dona-ion before cardiac procedures.

Evidence supporting the preoperative use of EPO innemic patients (hemoglobin � 13 g/dL) without autolo-ous predonation is less compelling, but still supportive.ost of the literature in support of using EPO to reduce

reoperative anemia is anecdotal and relates successfulase reports in a handful of patients, especially in Jeho-ah’s Witnesses [253, 263–269]. Since preoperative ane-ia increases operative mortality and morbidity in car-

iac procedures [270], EPO can be expected to reduce thisy augmenting red cell mass in anemic patients treatedith iron, if given more than 1 week before operation.his recommendation is based on limited evidence and

ogical consensus. No large-scale safety studies exist inatients having cardiac procedures, and therefore if oneursues such use, it must be realized that this is “off-

abel” and not studied. For patients with unstable anginar even stable angina, there is limited support to pursuehe use of preoperative EPO because safety data areacking. Preoperative interventions using EPO seem jus-ified for elective patients with diminished blood volumeecause of the high risk of excessive blood transfusion in

his subset.Still fewer objective data are available regarding the

se of EPO to treat perioperative and postoperativenemia. Because the onset of action of the drug is 4 to 6ays, it is necessary to administer EPO a few days indvance of operation, a luxury that is not always possible.ogic supports the addition of EPO to patients expected

o have a large blood loss during operation or who arenemic preoperatively. Beta-blockers suppress endoge-ous EPO production [271]. Cytokines stimulated by the

nflammatory response associated with CPB limit pro-uction of EPO [272]. Perioperative renal ischemia may

imit the production of EPO. Likewise, careful perioper-tive management may improve tissue oxygen delivery,nd suppress endogenous EPO production despite post-perative anemia. All of these factors support the addi-ion of preoperative (a few days before operation) admin-stration of EPO, despite the lack of evidence, to treateduced red blood cell volume in selected patients.

II) DDAVP AND RECOMBINANT FACTOR VIIA

lass IIb

. Use of desmopressin acetate (DDAVP) is not un-easonable to attenuate excessive bleeding and trans-usion in certain patients with demonstrable and spe-ific platelet dysfunction known to respond to thisgent (eg, uremic or CPB-induced platelet dysfunction,ype I von Willebrand’s disease). (Level of evidence B). Use of recombinant factor VIIa concentrate is notnreasonable for the management of intractable non-
urgical bleeding that is unresponsive to routine he-

m(

C

1mc

amIWtorrab2DdpprgDtdpPwqfffamtwf

ee2acae[behmharct

irfaaawedssc

bI

E

C

1si

C

1o

elri1mCtrtnpfoPshpffutin

I

C

1mn


ostatic therapy after cardiac procedures using CPB.Level of evidence B)

lass III

. Routine prophylactic use of DDAVP is not recom-ended to reduce bleeding or blood transfusion after

ardiac operations using CPB. (Level of evidence A)

Desmopressin acetate (DDAVP [1-deamino-8-d-rginine vasopressin]) is a drug approved for replace-ent therapy in patients with pituitary hypofunction.

t releases endogenous factor VIII precursors, vonillebrand factor, and tissue-type plasminogen activa-

or from vascular endothelium. The endogenous storesf factor VIII precursor released by DDAVP are noteadily replaced and require as long as 2 weeks toeplenish. Numerous studies, including three meta-nalyses, addressed the efficacy of DDAVP in limitingleeding after cardiac operations using CPB [188, 228,73–288]. When administered prophylactically,DAVP does not reduce bleeding after cardiac proce-ures. This agent is not helpful intraoperatively forrophylactic purposes to reduce bleeding after cardiacrocedures. However, observational and prospective,andomized data suggest that certain patient sub-roups may benefit from off-label use of this agent.esmopressin acetate might be considered preopera-

ively in some patients with acquired or inheritedefects in primary hemostasis detected by abnormaloint-of-care tests that measure platelet function (eg,FA-100) [276, 289]. A vast majority (� 90%) of patientsho are identified preoperatively having either ac-uired or inherited (ie, von Willebrand’s disease) de-

ects in primary hemostasis by PFA-100 testing benefitrom use of DDAVP. Koscielny and coworkers [290]ound that DDAVP reversed more than 90% of PFA-100bnormalities of acquired or inherited defects in pri-ary hemostasis if given before operation. Reversal of

hese primary hemostatic abnormalities was associatedith significant decreases in bleeding and blood trans-

usion [290].Evidence exists that an important component of postop-

rative hemostasis is related to the factor VII/tissue factorxtrinsic pathway of thrombin and fibrin generation [291,92]. Hence, it is not surprising that when recombinant,ctivated factor VII (r-FVIIa) became available commer-ially, it was quickly used for intractable bleeding duringnd after cardiac procedures using CPB. Almost all of thexperience with this potent hemostatic drug is anectdotal293–304]. However, there seems to be a very consistentenefit from this drug in limiting life-threatening bleeding,specially when there is a marginal response to routineemostatic therapy, including patients on extracorporealembrane oxygenation [295], and children undergoing

igh-risk complex cardiac procedures [300]. Clinicians re-lized the uncontrolled nature of these observations and aandomized trial is under way [305]. Others have issuedautionary reports of using this agent off-label with respect
o a propensity toward thrombotic complications especially i
n patients at high risk [306–309]. Patients undergoingoutine cardiac surgery may be at a substantially higher riskor development of thrombotic complications with thisgent because of elevated systemic levels of tissue factornd thrombin that occur during CPB [308]. These concernsre mirrored by the FDA’s recent mandate to include aarning on the package insert of this agent. Until more

vidence is available about the risks and benefits of thisrug, it can only be recommended as a rescue therapy forevere intractable bleeding without an identifiable surgicalource that is unresponsive to routine approaches afterardiac procedures using CPB.

) Devices That Aid Blood Conservation) VENTILATOR-ASSISTED BLOOD CONSERVATION—POSITIVE END-

XPIRATORY PRESSURE

lass IIb

. A trial of therapeutic positive end-expiratory pres-ure (PEEP) to reduce excessive postoperative bleedings not unreasonable. (Level of evidence B)

lass III

. Use of prophylactic PEEP to reduce bleeding post-peratively is not effective. (Level Evidence B)

Increased end-expiratory airway pressure (PEEP) ex-rts mechanical pressure on the myocardium and mayimit microvascular bleeding after heart surgery. Severaleports address this hypothesis. Two uncontrolled stud-es found that application of escalating doses of PEEP (5,0, 15, 20 cm H2O) controlled excessive bleeding (� 200L/h) in 15 of 15 and 11 of 15 patients [310, 311].onversely, one small RCT in 35 patients who bled more

han 180 mL in the first hour after cardiac procedureseported no difference in subsequent bleeding amonghose who received PEEP at 10 cm H2O as compared witho PEEP [312]. Four small RCTs (80, 83, 85, and 129atients undergoing predominantly elective CABG)

ound no difference in chest drainage or return to theperating room for postoperative hemorrhage whenEEP was applied prophylactically (starting at time ofternal closure [311], on admission to ICU [313, 314], or 1our postoperatively [312]). Based on limited evidence,rophylactic PEEP is not useful, but PEEP administered

or unusually high chest tube output may be helpful in aew cases. When PEEP is effective, the benefits aresually apparent within an hour. More data are required

o assess the risk/benefit relationship of increasing PEEPn the postoperative setting in which cardiac hemody-amics may be compromised.

I) OXYGENATOR TYPES

lass IIb

. It is not unreasonable to use open venous reservoirembrane oxygenator systems during cardiopulmo-

ary bypass for reduction in blood utilization and
mproved safety. (Level of evidence C)

oaoeaibqsdnl

oeusetammmtcrepDomCaaboeaat

obtsbvstau

I

C

1aup

bbfpttgsosppiiecomipC

sppdiTsrfnbu

I

C

1hodrtB

coriam1wahr


Several oxygenator types are available for the conductf CPB, including bubble oxygenators as well as closednd open membrane oxygenators. Excellent summariesf the types of oxygenators are available [315]. A gradualvolution from film oxygenators to bubble oxygenatorsnd ultimately to membrane oxygenators occurred dur-ng the last 50 years. Bubble oxygenators require a directlood-gas interface whereas membrane oxygenators re-uire diffusion of gas through a permeable membraneeparating blood and oxygen. Film oxygenators haveisappeared from clinical use because of significant tech-ical problems, including large priming volumes and

abor intensive setup that limited clinical usefulness.Bubble oxygenators compare favorably to membrane

xygenators in most comparisons except for bypass runsxceeding 2 hours [316–323]. Initial objections to the routinese of membrane oxygenators included increased cost andetup time. Advantages of membrane oxygenators includelimination of the direct blood-gas interface, separate con-rol of oxygen and carbon dioxide, and elimination ofntifoaming agents in the prime. Responses of oxygenatoranufacturers reduced the cost premium associated withembrane oxygenators and, simultaneously, benefits ofembrane oxygenators appeared in the literature. Advan-

ages found with membrane oxygenators included fewererebral emboli [324], better biocompatibility [321, 323], andeduced blood utilization [325, 326]. Without much class Ividence base, membrane oxygenators have mostly re-laced bubble oxygenators in modern clinical practice.riving forces for this preference include decreasing costsf membrane oxygenators, easier setup of perfusion equip-ent, perceived benefit in complex cases with prolongedPB times, and improved safety with less chance of massiveir embolism. In the event of loss of venous return, largemounts of air can be transported across the oxygenator,ut only infrequently into the patient, with membranexygenators, whereas bubble oxygenators risk immediatembolization of air into the arterial circuit. That constitutesn important potential safety benefit of membrane oxygen-tors and consensus may favor membrane oxygenators forhis reason.

Membrane oxygenator systems can have either closedr open venous systems. No clear benefit in improvediocompatibility, lower embolic load, or reduced blood

ransfusion accrues to either open or closed systems. Oneerious drawback to the closed venous reservoir in mem-rane oxygenator systems is the handling of air in theenous reservoir. Either a venous air filter or an extraource of venous reservoir is required to remove air inhe closed membrane systems [327, 328], This constitutespotential perfusion difficulty that can be eliminated bysing the open membrane oxygenator systems.

II) PERFUSION BLOOD PUMPS

lass IIb

. All commercially available blood pumps providecceptable blood conservation during CPB. It is notnreasonable to prefer centrifugal pumps because of
erfusion safety features. (Level of evidence B) r
A critical component of the heart-lung machine is thelood pump. Current clinical use is about evenly dividedetween centrifugal and roller pumps. Fundamental dif-

erences exist between the two types, with centrifugalumps being nonocclusive and roller pumps being able

o generate unlimited pressure because of occlusion athe roller head. The nonocclusive nature of the centrifu-al pumps is one of the positive features of this pumpystem, especially for long-term support. Because theutput of the centrifugal pumps is related to inlet pres-ure and outlet pressure, small amounts of air in theump will “deprime” the centrifugal system, and theotential for massive air embolism is very much dimin-

shed. The nonocclusive feature of the centrifugal pumps also a potential draw-back as it allows for possiblexsanguination of the patient if the perfusionist forgets tolamp the arterial line at the end of perfusion. Suppliersf perfusion pumps are sensitive to the potential forishaps and include various alarms and safety measures

n the manufacture of these devices [329], so much so thaterfusion accidents related to the blood pumps duringPB are exceedingly rare.Perfusion blood pumps do not impact blood transfu-

ion in a significant way. Nine randomized trials com-ared blood loss in patients perfused with centrifugalumps versus roller pumps [330–338]. No significantifferences in bleeding or blood transfusion were found

n patients perfused with roller or centrifugal pumps.here are theoretical advantages in blood conservationeen with centrifugal pumps. These advantages includeeduced complement activation and preserved plateletunction [339–341]. So far, these theoretical benefits haveot translated into consistent reduction in bleeding orlood transfusion after cardiac procedures with centrif-gal pumps.

V) HEPARIN MANAGEMENT DURING CARDIOPULMONARY BYPASS

lass IIb

. For patients requiring longer CPB times (> 2 to 3ours), it is not unreasonable to consider maintenancef higher or patient-specific heparin concentrationsuring CPB to reduce hemostatic system activation,educe consumption of platelets and coagulation pro-eins, and reduce blood transfusion. (Level of evidence)

Anticoagulation is used during cardiac surgery to limitellular and coagulation factor activation and to preventvert thrombosis of the extracorporeal circuit. Heparin isoutinely used because it is effective, immediately revers-ble, generally well tolerated, and inexpensive. Unfraction-ted heparin is a polysaccharide mixture of low and high-olecular-weight fractions (ie, molecules ranging from

,000 to 50,000 Daltons) that differ functionally. Fractionsith minimum chain length of 18 oligosaccharide units andmolecular weight of approximately 4,500 Daltons or

igher preferentially inhibit thrombin [342]. Oligosaccha-ide chain length is important because thrombin inhibition
equires simultaneous binding of thrombin and antithrom-

btpttce

rdg(aht[spltcntr

p(tosgtm[sietbalob3

atsEsepthtwabcm

pi3�CCcth3ovldbchptarhipin[btpfisbauduaobtl

V

C

1thltm

m3pnA


in III by heparin, which acts as a template. Only one inhree unfractionated heparin molecules has the criticalentasaccharide sequence required for binding to anti-

hrombin III [343]. Although binding of heparin to anti-hrombin III inhibits thrombin and factor Xa [344], thisomplex also inhibits several other sites in the intrinsic andxtrinsic pathways [345, 346].

There is substantial variability of heparin anticoagulantesponsiveness, as illustrated by a wide range of heparinose-response curves in patients undergoing cardiac sur-ery with CPB [347, 348]. Impaired heparin responsivenessie, also termed heparin resistance) is often attributed tontithrombin III deficiency, but significant variability ineparin anticoagulant response may also result from pa-

ient differences in heparin binding to endothelial cells349], white cells [350], platelets [351, 352], or proteins [353]uch as vitronectin [354], vWF [355], or histidine-rich glyco-rotein [356]. The heparin tissue source (ie, intestinal versus

ung, porcine versus bovine), method of preparation, andhe molecular weight distribution of heparin used may alsoontribute to impaired responsiveness [357, 358]. Unfortu-ately, no currently available tests can help clinicians iden-

ify the specific cause of heparin resistance or variableesponse to heparin.

When compared with no monitoring, heparin/rotamine dosing guided by activated clotting time

ACT) did not consistently limit chest tube drainage andransfusion outcomes [359–369]. Of 11 studies, six dem-nstrated a reduction in chest tube drainage and fivehowed no difference. Only seven of the studies investi-ated transfusion outcomes, and five of these showedhat ACT-based protocols reduced transfusion require-

ents. Only two of these 11 studies were prospective360, 367], and neither involved a randomized, blindedtudy design, which may limit the accuracy and reliabil-ty of the conclusions. Likewise, when compared withither fixed dose [370] or ACT-based dosing [371–374],he impact of heparin concentration monitoring onleeding and blood conservation varied. Although someuthors suggest that excessive bleeding is related toarger doses of bovine heparin during CPB [374, 375],thers found no differences in blood loss when eitherovine [371, 376] or porcine heparin were used [372, 373,77, 378].One important randomized prospective trial evalu-

ted the effect of heparin and protamine administra-ion directed by a point-of-care whole blood hemosta-is system (Hepcon; Medtronic Blood Management,nglewood, Colorado) on bleeding and blood transfu-ion in 254 patients [373, 379]. When compared with anmpiric ACT-based dosing regimen for heparin androtamine, the intervention cohort using point-of-care

esting received 25% larger total doses of heparin andad 25% smaller protamine-to-heparin ratios. A pa-

ient-specific, reference heparin concentration curveas measured in each intervention patient before CPB,

nd ACT levels during CPB were maintained at 480 sased on the preoperative heparin concentrationurve. The protamine dose was calculated from the
easured, residual heparin concentration. When com- g
ared with control patients, fewer patients in thentervention cohort required platelets (22% versus4%, p � 0.03), fresh frozen plasma (11% versus 31%, p

0.001), and cryoprecipitate (0% versus 5%, p � 0.01).ontrol patients also had 10% longer operative post-PB closure times (p � 0.02), 15% more mediastinal

hest tube drainage (p � 0.05) in the first 4 postopera-ive hours, and twice as many control patients requiredemostatic blood product transfusions (17% versus3%, p � 0.005). This study suggests that maintenancef heparin concentrations that more effectively inacti-ate thrombin may preserve hemostasis during pro-onged CPB. The fact that generation of thrombinegradation products [374, 380] and inhibition of clot-ound thrombin [381] are inversely related to heparinoncentration supports this hypothesis. Higher, stableeparin concentrations during CPB can also preservelatelet function during prolonged CPB. In a recent

rial, less platelet activation (ie, lower platelet factor 4nd BTG levels) was demonstrated in patients whoeceived larger heparin doses [376]. Higher doses ofeparin limited blood transfusion in two clinical stud-

es that evaluated blood conservation in high-riskatients. The first study evaluated 31 patients requir-

ng repeat or combined cardiac procedures (ie, coro-ary revascularization plus valve repair/replacement)

377]. Maintenance of higher heparin concentrationsetter preserved consumable antithrombin III and fac-

ors I, V, and VIII, most likely related to better sup-ression of thrombin (65% reduction in FPA levels) andbrinolytic activity (50% reduction in D-dimers). Aecond study demonstrated that larger heparin dosesetter suppress thrombin (ie, lower TAT complexes)nd fibrinolytic activity (ie, lower D-dimers) in patientsndergoing deep hypothermic circulatory arrest thano lower doses of heparin, especially when aprotinin issed as a hemostatic agent [376]. Considering allvailable evidence, it is reasonable to use higher dosesf heparin with point-of-care patient-specific testing toetter inhibit thrombin generation and to limit blood

ransfusion in high-risk patients likely to require pro-onged CPB.

) PROTAMINE DOSING

lass IIb

. It is not unreasonable to use either protamineitration or empiric low-dose regimens (eg, 50% of totaleparin dose) to lower the total protamine dose and

ower the protamine-to-heparin ratio at the end of CPBo reduce bleeding and blood transfusion require-

ents. (Level of evidence B)

Monitoring protocols can markedly influence prota-ine doses used to neutralize heparin [370, 378, 382–

87]. Two recent studies compared the use of a newoint-of-care hemostasis system (RxDx System; Inter-ational Technidyne, Edison, New Jersey) to standardCT-based empiric regimens in adult patients under-
oing cardiac surgical procedures [378, 387]. The RxDx

ssmdccaaaw02abpRcmtnttttatpw[sp1thlCc

V

C

1abe2pchw

cbmcmdbu

etbThpptfTrccsi4h

tcti34trshtcst

cdrHAtp[temfedmeNhacttcpgi


ystem estimates patient-specific anticoagulant re-ponse to heparin (heparin response titration), deter-ines celite ACT values, and calculates protamine

ose by using ACT-based approximations of heparinoncentration (eg, protamine response test). Whenompared with control patients whose anticoagulationnd reversal were based on empiric regimens, Jobesnd associates [378] demonstrated that RxDx produced

50% reduction in the protamine dose (p � 0.01),hich resulted in 50% less chest tube drainage (p �

.01) and 80% fewer transfusions (p � 0.02) in the first4 postoperative hours. In contrast, Shore-Lessersonnd associates [387] did not observe a reduction inlood loss and transfusion requirements nor was therotamine dose substantially decreased by use of thexDx system. Of eight published studies that specifi-ally evaluated the clinical impact of reducing prota-ine dose, four showed either reduced blood loss or

ransfusion requirements [378, 382–384], and four didot [370, 385–387]. The discrepancy in outcomes be-

ween these studies probably relates to differences inhe relative extent of reduction in protamine dose andhe overall protamine-to-heparin ratio in the interven-ion cohort. Thus, in the four studies that demonstrated

favorable difference in outcome, better bleeding orransfusion outcomes occurred when the reduction inrotamine dose was approximately 50% or greater andhen the protamine-to-heparin ratio was less than 1

378, 382–384]. The four negative studies had eithermaller reductions in protamine dose (27% to 38%) orrotamine-to-heparin ratios consistently greater than, or both. It may be reasonable to use either protamineitration or empiric low-dose regimens (eg, 50% of totaleparin dose) to lower the total protamine dose and

ower the protamine-to-heparin ratio at the end ofPB, but more evidence is necessary before a higher

lass recommendation (class I or IIa) can be made.

I) HEPARIN-BONDED CIRCUITS AND LOW-DOSE HEPARIN FOR CPB

lass IIb

. Heparin-coated bypass circuits (either the oxygen-tor alone or the entire circuit) are not unreasonable forlood conservation in cardiac operations. (Level ofvidence B). Low doses of systemic heparinization (ACT ap-roximately 300 s) are not unreasonable for bloodonservation during CPB, but the possibility of under-eparinization and other safety concerns have not beenell studied. (Level of evidence B)

Activation of platelets and the blood coagulationascade occurs from interaction of the circulatinglood with the synthetic surfaces of the heart-lungachine. Coating of the oxygenator and of the bypass

ircuit minimizes this activation process and ultimatelyay improve blood conservation after cardiac proce-

ures. The most common synthetic coating used inypass circuits is heparin. Several commercial prod-
cts are available in the United States and Europe t
mploying different means of binding the heparin tohe synthetic surfaces. Heparin coating of the entireypass circuit or of just the oxygenator is possible.here is almost no evidence to suggest that one type ofeparin coating or coating the entire bypass circuitrovides superior benefit in blood transfusion or inostoperative bleeding despite numerous studies on

he subject [388 – 410]. The exact mechanism of benefitrom heparin-coated bypass circuits is uncertain.here are contradictory reports suggesting that hepa-in coating of the oxygenator or of the entire bypassircuit limits platelet activation [403, 411– 413], reducesomplement activation [414 – 418], alters cellular adhe-ion to the bypass tubing [419], or diminishes thenflammatory pulmonary injury seen after CPB [420,21]. No clear-cut preference to one particular type ofeparin coating arises from these touted advantages.Several studies have evaluated bleeding and blood

ransfusion in patients having CPB with heparin-oated circuits or oxygenators. Most reports suggesthat synthetic coating of the bypass circuit limits bleed-ng and blood transfusion after cardiac operations [392,95– 402], but other reports suggest otherwise [403–10]. The studies are too heterogeneous to be subjectedo meta-analyses, as many studies use different hepa-in coating, different heparin doses, or different perfu-ion techniques. There is no consistent evidence thateparin coating either of the entire bypass circuit or of

he oxygenator is preferred. Use of heparin bondedircuits appears to be safe and may provide someavings in blood use, but more information is requiredo justify a high-level recommendation for their use.

One proposed advantage of coating of the bypassircuit with heparin is the ability to reduce the heparinose given to patients during CPB and consequentlyeduce the protamine dose used to neutralize heparin.eparin effect is monitored during CPB with the ACT.n ACT value of greater than 400 s is used as the

raditional standard for safe CPB, based largely on arimate study done in 1978 using bubble oxygenators

422]. There is no single “best” ACT level or standardhat can be agreed upon from available literature. Forxtracorporeal circulation, heparin is necessary butay have harmful side effects, especially to platelet

unction [423]. Because of concerns about harmfulffects of heparin, several investigators proposed re-uced heparin to maintain an ACT level of approxi-ately 300 s as a safe alternative that limits heparin

xposure and protamine reversal [374, 398, 424, 425].umerous studies evaluated the impact of reducedeparin dose and heparin-coated circuits on postoper-tive bleeding and blood transfusion without a clearonsensus arising [391, 392, 395– 402, 426 – 436]. Again,he studies are too heterogeneous to lend themselveso meta-analysis, but no clear-cut advantage in bloodonservation accrues to low-dose heparin for patientserfused with heparin-coated circuits. One study sug-ests that lower doses of heparin are associated with

ncreased bleeding after operation [377]. Many inves-
igators in the field of blood conservation point out that

arsfbipsvcnhip

jpw4aaiiorptieotwhdpCcrtcdu

V

C

1cpd

C

1pdd

s

sidotacei[dilttrHpdqcimctpptfsmsfttt

uspbpisptnivo[

[patbol


multimodality approach that includes preoperativeisk reduction, perioperative blood salvage and bloodparing, and careful attention to postoperative trans-usion algorithms is the most successful strategy forlood conservation [49, 264, 437, 438]. With so many

nterventions playing a small part in a multimodalityrogram, it is unlikely that a single intervention willtand out unless there is unquestioned efficacy or aery large sample size is considered in a randomizedlinical trial. To date, a large enough clinical trial hasot been conducted to answer the question of whethereparin-coating with or without low-dose heparin lim-

ts blood transfusion, and high-level evidence to sup-ort this intervention does not exist.Three cautionary notes should be mentioned in con-

unction with low-dose heparinization during CPB. Re-orts suggest that increased thrombin may be generatedhen lower doses of heparin are used during CPB [431,

36]. Increased thrombin generation is almost certainlyssociated with increased platelet activation during CPBnd may result in increased blood loss [439]. Differencesn perioperative bleeding in patients with small increasesn thrombin generation may be hard to detect and maynly become apparent at longer or more difficult bypassuns. Second, because of important blood conservationroperties of antifibrinolytic agents like aprotinin, pa-

ients in whom low-dose heparin is used may be atncreased risk with certain of the antifibrinolytic agents,specially if circulatory arrest is a prominent part of theperation [440, 441]. There is limited support for usinghe combination of low doses of heparin in conjunctionith antifibrinolytic agents during CPB (with or withouteparin-coated bypass circuits). Third, preliminary evi-ence suggests that heparin coating induces gene ex-ression changes in the perfused cellular elements afterPB to a greater extent than protein (albumin) coated

ircuits [442]. Whether changes in gene expression play aole in blood conservation is uncertain, and only time willell what impact this has on bleeding. Because of theseoncerns and the lack of compelling safety data, lowoses of heparin during the conduct of CPB should besed with caution.

II) RED CELL AND PLATELET SAVING

lass I

. Routine use of red cell saving is helpful for bloodonservation in cardiac operations using CPB, except inatients with infection or malignancy. (Level of evi-ence A)

lass III

. Routine use of intraoperative platelet or plasma-heresis is not recommended for blood conservationuring cardiac operations using CPB. (Level of evi-ence A)

Intraoperative salvage of red blood cells, using a cell
aving apparatus, is an important part of blood con- r
ervation in cardiac operations using CPB. Beginningn the mid 1970s, commercially available cell-savingevices became available for routine use in cardiac andther high-risk operations. Several reports documenthe safety of red blood cell saving. Specifically, nodverse CNS events [443], no increased infectiousomplications [444], no increased hemolysis or fatmboli [445], reduced circulating markers of systemicnflammation with removal of most [446] but not all447] cytokines from suctioned blood, and overall re-uced complication rates [448] are associated with

ntraoperative use of red cell savers. At least 10 pub-ished reports, including some randomized controlledrials, support the routine use of red blood cell savingo reduce bleeding and blood transfusion, although theeduction in transfusions may be modest [449 – 458].owever, extensive use of cell salvage systems torocess contents of the extracorporeal circuit afteriscontinuation of CPB or with processing of extensiveuantities of cardiotomy-derived blood may lead to aritical loss of coagulation factors and platelets, result-ng in a bleeding diathesis. This is supported by a

ultivariate analysis that demonstrated that processedell salvage volumes were related to bleeding/ransfusion [377]. With usual red blood cell savingrotocols, about half of the blood suctioned from theatient is ultimately reconstituted. With careful atten-

ion to intraoperative hemostasis in patients at low riskor operative bleeding, it may be that use of a cell-aving device is not a cost-effective measure, althoughost assessments of cost favor routine use of cell-

aving devices [459 – 461]. Presence of malignancy, in-ection, or blood exposed to topical clotting agents (eg,hrombostat, fibrin glue, Tisseal, Gelfoam, or any otherhrombin-containing compound) are contraindicationso use of the cell-saving device.

Platelet-rich plasmapheresis is a continuous centrif-gation technique, using technology similar to red cellavers that allows selective removal of platelet richlasma from the circulating whole blood immediatelyefore the start of CPB. As platelet dysfunction is arominent part of CPB-induced coagulopathic bleed-

ng, it seemed reasonable to remove platelets from thetress of CPB and return these “spared platelets” to theatient at the end of operation. This technique has the

heoretical advantage that improvement in plateletumber and function might limit postoperative bleed-

ng without a decrease in oxygen carrying capacity. Byarying the centrifugation technique, it is possible tobtain fractions that are either rich or poor in platelets462, 463].

At least 18 studies [463– 480] and one meta-analysis481] evaluated the efficacy of platelet-rich plasma-heresis in limiting bleeding and blood transfusionfter cardiac operations. The highest quality studieshat involved blinding and randomization showed noenefit from this intervention [472– 474, 479], whereasther published reports showed some benefit of plate-

et-rich plasmapheresis. For several reasons, platelet-
ich plasmapheresis is not used clinically to any sig-

npasmtopbmptpmptttumis[imdrp

V

C

1dci

lTtlwopIfrtctCl4dBadf

cI

C

1mib

tmurcrwrresuatmOpfabac

I

O

C

1uimo

bogcibfagtkctmD


ificant extent. First, prophylactic transfusion of freshlatelets does not alter bleeding and blood transfusionfter operations [482, 483], so there is no reason touppose that platelets preserved by platelet-rich plas-apheresis will have a beneficial effect on postopera-

ive hemostasis. Further, there is evidence that intra-perative autologous donation of fresh whole bloodrovides the same or better platelet preservation andlood-sparing effect as perioperative platelet-rich plas-apheresis [484, 485]. Likewise, platelet-rich plasma-

heresis precludes the use of whole blood intraopera-ive autologous donation. The process of platelet-richlasmapheresis is labor intensive and any technicalistakes in harvesting results in lost blood volume and

latelet number, a potentially problematic occurrencehat alters blood conservation adversely. If for someechnical reason platelet-rich plasmapheresis is at-empted but the blood removed from the patient isnable to be reinfused, this constitutes an iatrogenicishap that is directly related to the intervention and

s harmful, potentially resulting in additional transfu-ion. This event may occur as often as 1 in 30 times471]. Given that platelet-rich plasmapheresis is laborntensive, costly, of marginal benefit, and possibly

ore risky than whole blood intraoperative autologousonation, this technique can not be recommended foroutine use in blood conservation during cardiacrocedures.

III) LEUKOCYTE FILTRATION

lass III

. Leukocyte filters on the CPB circuit for leukocyteepletion should not be used for perioperative bloodonservation and may actually activate leukocytes dur-ng CPB. (Level of evidence B)

Many of the harmful effects of CPB can be traced toeukocyte activation during extracorporeal perfusion.emporary removal of white blood cells around the

ime of CPB may limit some of this damage. Variouseukocyte filters are available for use in conjunctionith the perfusion apparatus. These filters are capablef depleting the systemic circulation of leukocytes andossibly improving organ function after operation.

mproved neutrophil adhesion [486], better endothelialunction [487] and lung function [488], limitation ofeperfusion injury [489], reduced rate of atrial fibrilla-ion [490], and improvement in brain function afterirculatory arrest [491] are putative benefits attributedo use of leukocyte depletion during various stages ofPB. None of these observed beneficial effects trans-

ates into consistent, significant clinical benefit [67,92– 495]. There is evidence that leukocyte depletionuring CPB may, in fact, activate white cells [496, 497].ecause of the lack of clinical benefit and the concernsbout worsening leukocyte function with leukocyteepletion, this methodology cannot be recommended

or blood conservation during CPB. o

) Perfusion Techniques and OPCAB) MINIMIZED EXTRACORPOREAL BYPASS CIRCUITS

lass IIb

. It is not unreasonable to use low prime and mini-ized extracorporeal bypass circuits to reduce the fall

n hematocrit during CPB as part of a multimodalitylood conservation program. (Level of evidence B)

Minimized, and sometimes integrated (combined cen-rifugal pump and oxygenator), low prime (450 to 800

L), reduced surface area extracorporeal circuits aresed clinically. Authors report less hemodilution andeduced inflammatory response with these minimizedircuits [498–503]. Abdel-Rahman and coworkers [503]eport reduced activation of coagulation and fibrinolysisith minimized bypass circuits and suggest a benefit in

educed transfusion associated with these circuits. Oneetrospective comparison of 970 patients undergoinglective CABG [499] and two prospective randomizedtudies by the same group [500, 501] of 400 patientsndergoing elective CABG and 100 patients undergoingortic valve replacement reported at least a 50% reduc-ion in red blood cell transfusion rates with use of

inimized circuits compared with conventional circuits.n the other hand, two smaller RCTs of 30 and 204atients undergoing CABG found no decrease in trans-

usion requirements with use of two commercially avail-ble minimized circuits [502, 504]. Possible transfusionenefits from minimized circuits must be balancedgainst possible risks including air embolism, and lack ofardiotomy suction associated with this technology [504].

I) ACUTE NORMOVOLEMIC HEMODILUTION (ALSO CALLED INTRA-

PERATIVE AUTOLOGOUS DONATION)

lass IIb

. Acute normovolemic hemodilution (ANH) is notnreasonable for blood conservation but its usefulness

s not well established. It could be utilized as part of aultipronged approach to blood conservation. (Level

f evidence B)

This technique involves removal of one to two units oflood immediately before surgery and is used in a varietyf fields of surgery including orthopedic surgery andeneral surgery. In the context of cardiac surgery, typi-ally one to two units of autologous blood are removedmmediately before the institution of cardiopulmonaryypass. Unstable patients are not appropriate candidates

or this technique, especially in the presence of evolvingcute myocardial infarction, unstable angina, or cardio-enic shock. Other typical exclusion criteria include pa-ients with significant preoperative anemia, sepsis ornown bacteremia. Relative contraindications may in-lude low ejection fraction (� 30%) [505, 506]. To main-ain circulating blood volume, the volume of blood re-

oved is usually replaced 1:1 with crystalloid or colloid.uring subsequent cardiopulmonary bypass, hematocrit
n bypass determines transfusion. Transfusion of the

sptmAotrbgivistidttia

cptbutpsvtcAo2sbcCAmtfgneArb

mtmttmpc(

AriddrmtumpT[spiwedcuci

mbcpstftiibttbIt

I

C

1efdd2bwds

odr


equestered autologous units is the first source of re-lacement, followed by transfusion of allogeneic blood if

he desired minimum hematocrit on bypass cannot beaintained with use of autologous blood. The use ofNH as a blood conservation method in surgery is basedn the principle that lowering the red blood cell concen-ration (hematocrit) during surgery will decrease theeduction in red cell mass lost for any given volume oflood lost. The expected reduction in the need for allo-eneic blood product administration is desirable to min-

mize the risk of transmission of hepatitis, HIV, otheriral and bacterial pathogens, and immunomodulationnduced by allogeneic red cell transfusion. In cardiacurgery, ANH has the additional theoretical advantagehat the blood removed from the bypass circuit before thenstitution of CPB is “protected” from the potentialeleterious effects of platelet activation and consump-

ion, hemolysis, complement activation, and the produc-ion of a variety of inflammatory cytokines includingnterleukin 6, interleukin 8, and tumor necrosis factorlpha associated with extracorporeal circulation [507].Several prospective randomized trials show a signifi-

ant decrease in the transfusion rate of allogeneic bloodroducts using this technique [508, 509] while others fail

o document any benefit [443, 505, 510]. In a recent studyy Hohn and associates [505], 80 adult patients whonderwent cardiac surgical procedures were randomized

o receive ANH or usual blood conservation care. Allatients in both groups underwent standard blood con-ervation procedures including use of a cell-saving de-ice, reinfusion of mediastinal shed blood, administra-ion of intravenous aprotinin, and reinfusion of bloodontained in the extracorporeal circuit after surgery.cute normovolemic hemodilution harvested an averagef 1,099 cc blood to achieve a calculated hematocrit of8% after the induction of anesthesia. There was notatistical difference in the number of allogeneic units oflood transfused or in the percentage of patients (29%,ontrol; and 33%, ANH) requiring blood transfusion.asati and associates [443], in a study of “low volume”NH randomly assigned 204 patients to ANH (5 to 8L/kg removed before heparinization therapy) or con-

rol (no ANH). These authors found no difference in therequency or number of transfusions between the tworoups [443]. Although both of these studies yieldedegative results, the study by Hohn was criticized forxcessive hemodilution with colloid and crystalloid afterNH such that more than 50% of the ANH patients

eceived all (33%) or a portion (22%) of the autologouslood back during CPB.The results of the study by Casati and associates [443]ay be of limited relevance as “low volume” ANH is not

he optimal strategy to use, based on mathematicalodels of ANH efficacy [511, 512], and since these au-

hors routinely use aprotinin, further limiting the poten-ial benefit of ANH in reducing transfusion require-

ents. Additionally, a recent randomized trial of ANH inatients undergoing liver resections showed a statisti-ally significant reduction in the frequency of transfusion
10% versus 36%, p � 0.05) in the patients who received d
NH [471], but a meta-analysis yielded inconclusiveesults [513]. Retrospective studies of the efficacy of ANHn adult cardiac surgery are equally divided among thoseemonstrating benefit [514–516], and those failing toocument a reduction in bleeding or in transfusionequirements [517, 518]. A cautionary note is worthentioning. If for some technical reason ANH is at-

empted but the blood removed from the patient isnable to be reinfused, this constitutes an iatrogenicishap that is directly related to the intervention and is

otentially harmful, resulting in additional transfusion.his event may occur as often as 1 in 30 times with ANH

471]. Furthermore, patients who most need blood con-ervation are least likely to be candidates for ANH, asreoperative anemia and low blood volume are contra-

ndications for this technique. For noncardiac surgery,ith very aggressive red cell harvest, ANH may be

fficacious and cost effective [519], but for cardiac proce-ures, the benefit of ANH is less well established, espe-ially with aggressive red cell harvest. Care is necessarysing ANH as it can be detrimental and is not likely toontribute significantly as the sole blood conservationntervention.

Combinations of ANH with other blood conservationsay be beneficial. For example, use of synthetic fluorocar-

ons or hemoglobin based oxygen carrier products inonjunction with ANH shows clinical promise in mostublished trials [520–524]. Additionally, several studiesuggest that a combination of blood conservation interven-ions is likely to provide the best reduction in blood trans-usion [469, 525–530]. It is entirely possible that the effect ofhe combinations of blood conservation interventions thatnclude ANH may be more beneficial than the sum of thendividual components. Any strategy for limiting excessivelood transfusion after cardiac procedures should identify

he institution-specific blood conservation interventionshat are efficacious and employ these interventions in com-ination, especially for the high-risk subset of patients.nstitution-specific risk/benefit profiles are essential to op-imize blood conservation.

II) PUMP SALVAGE AND INTRAOPERATIVE AUTOTRANSFUSION

lass IIb

. During CPB, it is not unreasonable to use intraop-rative autotransfusion, either with blood directlyrom cardiotomy suction or recycled using a cell-savingevice, to augment blood conservation. (Level of evi-ence C). Shortly after the completion of cardiopulmonaryypass, salvage of pump blood, either administeredithout washing or after washing with a cell-savingevice, is not unreasonable as a means of blood con-ervation. (Level of evidence C)

The autotransfusion of mediastinal blood from theperative field during CPB is routinely performeduring adult cardiac surgical procedures. In cases ofapid hemorrhage, reinfusion of blood using the car-
iotomy suction device can be lifesaving. However,

“ccptIsthaudeoectttm

aTaC[abavBfw

mTpsfstsppdcmbocaioc

I

C

1u

uctcflioo

C

1pa(

ati5a5Twewsmchoewmtpidcimb[pa5abssm

V

C

1bs


uncontrolled” cardiotomy suction during prolongedardiopulmonary bypass (longer than 3 hours) is asso-iated with an increase in postoperative bleeding com-ared with “controlled” cardiotomy suction designed

o minimize the associated aspiration of air [531].ntraoperative autotransfusion that employs a cell-aving device is capable of collecting blood both whenhe patient is heparinized and during periods where noeparin effect is present (eg, during internal thoracicrtery harvest) and the benefits of cell-saving devicese are discussed above. Suction of mediastinal blooduring CPB risks entrapment of particulate matter,specially fat, and some evidence exists that cardiot-my suction for autotransfusion is a source of fatmboli to the brain during CPB [532]. Reents andoworkers [507] suggest that contamination with bac-eria is more common with cell-saved autotransfusionhan with direct cardiotomy suction. Risks from auto-ransfusion during CPB are equal among the various

ethods used to recycle mediastinal blood.Surprisingly there is no evidence-based preference for

ny form of intraoperative autotransfusion during CPB.here does not seem to be a difference in blood transfusionfter autotransfusion using a cell-saving device only duringPB versus autotransfusion with direct cardiotomy suction

445, 533–537]. Two studies compared autotransfusion usingcell-saving device to discard of cardiotomy suctioned

lood during CPB [538, 539], with only the study of Natafnd coworkers [538] suggesting a benefit in blood conser-ation from cell-saving–aided autotransfusion during CPB.ased on available evidence, only a weak recommendation

or autotransfusion during CPB, either with cell saving orith cardiotomy suction, can be made.Salvage of pump blood at the completion of cardiopul-onary bypass is routinely performed in most centers.

here are limited data available on the impact of thisractice on the requirement for allogeneic blood transfu-ion. Six studies address the benefit of salvage of bloodrom the heart-lung machine at the end of CPB (pumpalvage) with mixed results [536, 540–544]. The studies areoo heterogeneous and underpowered to draw firm conclu-ions, but consensus favors the routine use of some form ofump blood salvage at the end of CPB. Discarding orurification (ie, use of cell-saving devices) of cardiotomy-erived mediastinal blood is not clearly beneficial for bloodonservation, but loss of coagulation factors and plateletsay be potentially detrimental with resultant excessive

leeding. Whether pump salvage using a cell-saving devicer using direct reinfusion, most surgeons agree that dis-arding the residual pump blood wastes a potentially valu-ble resource. No study suggests that pump salvage resultsn serious injury, so the risk/benefit ratio favors some formf reinfusion of residual blood from the heart-lung ma-hine, although with virtually no evidence.

V) POSTOPERATIVE SHED BLOOD REINFUSION

lass IIb

. Postoperative mediastinal shed blood reinfusion
sing washed mediastinal blood (cell saving) is not d
nreasonable for blood conservation when used inonjunction with other blood conservation interven-ions. Washing of shed mediastinal blood may de-rease lipid emboli, decrease the concentration of in-ammatory cytokines, and reinfusion of washed blood

s not unreasonable to limit blood transfusion as partf a multimodality blood conservation program. (Levelf evidence B)

lass III

. Direct reinfusion of shed mediastinal blood fromostoperative chest tube drainage is not recommendeds a means of blood conservation and may cause harm.Level of evidence B)

Studies using postoperative reinfusion of shed medi-stinal blood to limit blood transfusion first appeared inhe early 1970s. Since then, 17 reports provided conflict-ng results about the benefit of this intervention [543,45–561]. Nine of these 17 studies suggest that postoper-tive shed blood reinfusion limits blood transfusion [543,46, 547, 549, 551, 556, 557, 559, 561], and the rest do not.hese studies are heterogeneous as some comparedashed (cell saved) versus nonwashed (cardiotomy res-

rvoir) postoperative shed mediastinal blood reinfusionhile others compared either of these techniques to

imply discarding shed mediastinal blood. In addition,ediastinal blood contains fibrinolytics, inflammatory

ytokines, complement, endotoxin, tissue factor, and freeemoglobin [443, 507, 531, 547, 552, 554, 561–565]. Infusionf shed mediastinal blood results in activation of thextrinsic coagulation pathway [443, 554, 565], althoughashing of mediastinal blood may reduce these abnor-alities [507, 563]. Furthermore, mediastinal blood con-

ains lipid particles and thromboemboli that are incom-letely removed by existing blood filters, and these

mpurities are postulated to contribute to neurologicysfunction after cardiac surgery [291, 566, 567]. Use of aell-saving device to wash shed mediastinal blood resultsn a significant decrease in lipid emboli and in inflam-

atory cytokines in the reinfused blood compared withlood obtained directly using a cardiotomy reservoir

507, 563, 568]. Importantly, two studies suggest thatostoperative unwashed shed blood reinfusion is associ-ted with increased sternal or systemic infections [545,48]. Because of the risks associated with this techniquend the lack of clear cut benefit, direct reinfusion of shedlood can not be recommended for routine blood con-ervation; whereas, postoperative reinfusion of washedhed mediastinal blood may be helpful as part of aultimodality blood conservation program.

) PREOPERATIVE AUTOLOGOUS BLOOD DONATION

lass IIa

. Predonation of as much as two units of autologouslood is reasonable for blood conservation in carefullyelected (mostly elective) patients before routine car-
iac operation particularly when coupled with appro-

pi

pnmrginbopaslacchaonlsl

V

N

C

1ne

tontfalatgrobagcwIpoptd

V

C

1ftv(

C

1aa

uTtflccmrrwihsrt5tCd[a[btadSpdtmseatbtcswhtB


riately dosed erythropoietin (EPO) therapy and/orron therapy. (Level of evidence A)

Autologous blood donation a few days to a few weeksreoperatively is commonly practiced for a variety ofoncardiac surgical procedures to reduce the require-ent for allogeneic blood transfusion. This practice is not

outinely employed for patients undergoing cardiac sur-ery, owing in part to concerns about a possible increase

n the incidence of myocardial infarction or hemody-amic instability during the period after predonation butefore surgery. Nonetheless, there is an increasing bodyf literature [445, 569–577] documenting the safety of thisractice and a significant reduction in the transfusion ofllogeneic blood products associated with this practice inelected patients undergoing cardiac surgery, particu-arly when performed concomitantly with erythropoietinnd/or administration to allow for a compensatory in-rease in red cell production [572–574, 578]. However, theost effectiveness and benefit/risk ratio of this techniqueas been questioned [579, 580]. Further, the use of EPO tougment red cell production during preoperative autol-gous donation in patients having cardiac procedures isot recommended by the drug manufacturers because of

imited safety data. Hence, the use of EPO in this circum-tance has an unproven safety profile and is an “off-abel” application of this drug.

I) RETROGRADE AUTOLOGOUS PRIMING OF THE CARDIOPULMO-

ARY BYPASS CIRCUIT

lass IIb

. Retrograde autologous priming of the CPB circuit isot unreasonable for blood conservation. (Level ofvidence B)

This method replaces the crystalloid prime volume ofhe cardiopulmonary bypass circuit with the patient’swn blood immediately before beginning cardiopulmo-ary bypass [458, 581–583]. Typically, the arterial limb of

he circuit is cleared retrograde through back bleedingrom the aortic cannula, and the venous limb is clearedntegrade using the blood pump. The displaced crystal-oid is collected and excluded from the circuit. A caution-ry note needs to be added. The connection of a patiento a vacuum assisted venous drainage system for retro-rade autologous priming (ie, with a closed hard shelleservoir) is a potentially dangerous situation. On at leastne occasion, this resulted in massive systemic air em-olism through a patent foramen ovale, and vacuum-ssisted venous drainage should not be used for retro-rade autologous priming. This technique maintainsolloid osmotic pressure and reduces extravascular lungater compared with standard priming techniques [458].

n two of three randomized trials, retrograde autologousriming caused a significant reduction in the frequencyf allogeneic blood transfusion compared with usualractice [458, 582, 583]. Two other studies showed a trend

oward lower usage of allogeneic blood transfusions but
id not reach statistical significance [458, 581]. l
II) HEMOFILTRATION

lass IIb

. Use of modified ultrafiltration is not unreasonableor blood conservation during CPB in pediatric pa-ients or in patients where the priming volume is aery significant fraction of the total blood volume.Level of evidence B)

lass III

. Routine use of ultrafiltration during or immedi-tely after CPB is not helpful for blood conservation indult cardiac operations. (Level of evidence B)

Both conventional ultrafiltration (CUF) and modifiedltrafiltration (MUF) are employed in cardiac surgery.he rationale behind the use of CUF is that this

echnique represents a method to remove the excessuid associated with the priming volume needed forardiopulmonary bypass. However, because additionalrystalloid or blood are often added to the circuit toaintain adequate circulating volume during CPB, the

esults of this technique are inconsistent with regard toeducing total body water but are generally associatedith a reduction in the systemic concentration of

nflammatory mediators [584]. Modified ultrafiltrationas been studied most extensively in pediatric cardiacurgical patients and results in statistically significanteductions in TBW, removal of inflammatory media-ors, and improvement in hemodynamic indices [584 –87]. The improved efficacy of MUF probably relates tohe fact that it is performed for a period of time afterPB is completed. There are several prospective ran-omized studies of CUF and MUF in adult patients

588 –593], including two studies that compared CUFlone [589, 593], two studies that compared MUF alone590, 591], and two studies that compared CUF com-ined with MUF to conventional CPB without ultrafil-

ration [588, 592]. In a study of 60 patients randomlyssigned to CUF versus standard CPB, there was noifference in allogeneic blood product usage [593].imilarly, in a randomized study of 97 patients com-aring MUF alone to conventional CPB, there was noifference in blood loss between the two groups, al-

hough there was a significant reduction in adhesionolecules and cytokines in the MUF group [591]. One

tudy demonstrated a significant reduction in postop-rative blood loss in the group that was randomlyssigned to CUF combined with MUF, and althoughhere was no difference in the amount of allogeneiclood transfused, there was a significant reduction in

he transfusion of fresh frozen plasma compared withontrol high-risk adult patients [588]. A retrospectivetudy of 61 patients who received CUF alone comparedith 57 patients who received conventional CPB forigh-risk adult cardiac procedures showed a trend

oward decreased bleeding in the CUF group [594].ecause of the lack of consistent benefit of CUF in

imiting blood transfusion in adults during CPB, this

tc

V

C

1reet

crssntcfdi[nosaaoIdbO2sdnaueoec

d

C

1cnsatt

C

1t

ie

fodbuettmtc

sa6cvbtUbttttttcmuoat

tp[epanidmw[

itFt6aam


echnique is not likely to be helpful for bloodonservation.

III) OFF-PUMP PROCEDURES FOR BLOOD CONSERVATION

lass IIa

. Off-pump coronary artery bypass (OPCAB) is aeasonable means of blood conservation, provided thatmergent conversion to on-pump bypass is unlikelyither based on surgeon experience or patient charac-eristics. (Level of evidence A)

Several randomized trials, and two meta-analysesompared the results of OPCAB to on-pump coronaryevasculalrization [56, 595–600]. In general, these studieshow a reduction in the use of allogeneic blood transfu-ions with OPCAB [56, 595–598, 600]. One study showedo difference in the total postoperative blood loss be-

ween the two techniques while documenting a signifi-ant decrease in the use of packed red blood cell trans-usions [595]. Another study showed no significantifference in the number of allogeneic units transfused or

n the postoperative blood loss between the two groups599]. Looking at the data more carefully shows that theumber of patients who crossed over from OPCAB ton-pump was not uniform among these reports. In thetudies by Puskas and associates [598], Khan and associ-tes [595], van Dijk and associates [596], and Angelini andssociates [597], crossovers to on-pump CABG were rare,ccurring in 1%, 4%, 7%, and 1% of patients, respectively.n contrast, in the one randomized trial that showed noifference in transfusion requirements or blood lossetween on-pump and OPCAB groups, crossovers fromPCAB to on-pump were more common, occurring in

2.5% of patients [599]. Also, many of the retrospectivetudies that have demonstrated decreased blood loss andecreased transfusion requirements with OPCAB haveot included cross-over patients in the off-pump group asn intention-to-treat analysis would require [601]. Everynfavorable outcome, including bleeding requiring reop-ration, was more common in the emergency OPCAB ton-pump conversion subgroup. This is distinctly differ-nt than patients in whom OPCAB without emergencyonversion was performed [601].

) Topical Agents/Tissue Glues

lass IIb

. Topical sealants used to assist in the repair ofomplex, high-risk cardiac and aortic procedures areot unreasonable to limit bleeding in certain keyituations (eg, left ventricular free wall rupture andortic dissection) but are associated with complicationshat may limit their usefulness in less high risk situa-ions. (Level of evidence C)

lass III

. Topical hemostatic agents that employ bovine
hrombin are not helpful for blood conservation dur- e
ng CPB and may be potentially harmful. (Level ofvidence B)

Numerous topical hemostatic agents are availableor use during cardiac operations. The wide variationf these topical agents, including some “home reme-ies,” makes an evidence-based comparison impossi-le. There is almost no evidence to support the routinese of these agents for blood conservation. What littlevidence exists is in the form of anecdotal or noncon-rolled reports with little indication of benefit, otherhan subjective appraisal [602, 603]. One randomized

ulticenter trial studied the use of fibrin sealant foropical hemostasis. This trial found no benefit in bloodonservation for fibrin sealant [604].

Bovine thrombin is a commonly used topical hemo-tatic agent in vascular and cardiac operations. Thisgent causes a profound immunologic response [605–08], especially in children, and is associated with worselinical outcome without much benefit in blood conser-ation in most cases [608, 609]. Antibodies generated toovine thrombin can be a problem if reexposure to

hrombin occurs soon after the initial exposure [608, 609].se of topical bovine thrombin cannot be recommendedecause of the potentially harmful immunologic reac-

ions that may occur with this drug. Likewise, some of theopical hemostatic agents contain aprotinin, especiallyhose manufactured in Europe. The use of aprotinin inopical hemostatic agents risks the potential of sensitiza-ion of patients to this agent and subsequent reexposureo aprotinin, especially within 6 months of initial dosing,arries a 5% incidence of some hypersensitivity reaction,ost commonly anaphylaxis [249, 610]. Because of the

nrecognized prior exposure to aprotinin, developmentf a careful drug history and other precautioinary stepsre essential before using aprotinin for blood conserva-ion [611].

Fibrin glue preparations, with or without the addi-ion of bovine thrombin, are used extensively in com-lex operations on the aorta, especially in Europe

612– 615]. A few cautionary notes surfaced after thisxtensive use. One report documents interference withrosthetic aortic valve function by fibrin glue used forortic repair [616]. Fibrin glue in the pericardium doesot reduce pericardial adhesions but has been sprayed

n the pericardium for this reason [617]. Two reportsocument the occurrence of aortic redissection uponid-term follow-up after resorcinol tissue sealantsere used in the repair of a type A aortic dissection

618, 619].Left ventricular free wall rupture after myocardial

nfarction or during mitral valve operations is a devas-ating problem with a high associated mortality [614, 620].ibrin sealants in conjunction with pericardial or syn-hetic covering are used to seal the rupture site [614, 621,22]. Not all fibrin glues provide mechanical strength inddition to hemostatic support and, in fact, most sealantsvailable in the United States do not provide enoughechanical strength for this purpose [623]. While the

xtensive use of these agents in these high-risk opera-

trp

6O

Motovdpsfi

aCrp1pwtssftci

bpaffdbrbtsIstvlcpacao

bIlcusaa8lwowd

isbhpbrfhgpcsa[aaabFc

TP

P

S

P

D

A

A


ions is unproven, the potential benefits outweigh theisks, and consensus favors their use only for theseotentially lethal conditions.

) Prophylaxis—Interventions Outside theperating Room

any interventions that influence blood conservationccur before operation. These interventions can occur inhe catheterization laboratory, in the ICU, or even in theutpatient setting. The team dedicated to blood conser-ation should hold discussions with cardiology to coor-inate an institution-wide approach for cardiovascularatients. Table 6 and the following discussion summarizeome of the important evidence-based perioperativendings that impact blood conservation.

) Catheterization Laboratory Interventionsatheterization access site complications are common,

anging from minor hematomas to major vascular com-romise [118, 624–629]. Access site complications occur% to 9% of the time depending on the complexity of theercutaneous procedure [628]. Five percent of patientsill require a transfusion after catheterization. One mul-

icenter randomized controlled trial [630] and one cohorttudy [627] looked at the method used to manage accessite control after catheterization, with no significant dif-erence in hematomas and bleeding complications iden-ified from any technique. A prospective multicenterlinical nursing trial of 4,010 patients revealed that nurs-

able 6. Summary of Recommendations for Perioperativerophylactic Measures for Blood Conservation

RecommendationLevel ofEvidence Class

reoperative screening of the intrinsiccoagulation system is not recommendedunless there is a clinical history ofbleeding diathesis.

B III

creening preoperative bleeding time is notunreasonable for high-risk patients,especially those who receive preoperativeantiplatelet drugs.

B IIb

reoperative hematocrit and platelet countare indicated for risk prediction, andabnormalities in these variables areamenable to intervention.

A I

evices aimed at obtaining directhemostasis at catheterization access sitesare not unreasonable for bloodconservation if operation is plannedwithin 24 hours.

C IIb

lternatives to laboratory blood sampling(eg, oximetry instead of arterial bloodgasses) are reasonable means of bloodconservation before operation.

C IIa

comprehensive, integrated, multimodalityblood conservation program in theintensive care unit is a reasonable meansto limit blood transfusion.

B IIa

ng interventions employed did not significantly decrease o

leeding at the groin access site [631]. Access site com-lications are more common if potent antiplatelet agentsre used during the procedure [118]. Given the low butairly constant rate of access site bleeding complicationsrom catheterization, consensus suggests that closureevices or collagen plug devices that directly controlleeding at the access site are preferred in patients whoequire operation within 24 hours of catheterization (Ta-le 6). This is a preoperative blood conservation adjunct

hat may prevent unrecognized blood loss at the accessite during CPB with high-dose systemic heparinization.n the case of CPB, high-dose heparin is a hematologictress equivalent to that of high-intensity antiplateletherapy, and consensus suggests that direct closure de-ices at the catheterization site are not unreasonable toimit bleeding and blood transfusion related to recentatheterization (within 24 hours) at the time of cardiacrocedures using CPB [118, 628]. Other interventionsimed at limiting blood loss around the time of cardiacatheterization will optimize preoperative red cell massnd ultimately decrease blood transfusion afterperation.

) Preoperative Laboratory Testingn the preoperative evaluation, thoughtful application ofaboratory tests and associated blood draws are likely toonserve blood. Scher [632] looked at 30 patients whonderwent reoperation for postoperative bleeding andhowed that bleeding occurred despite normal preoper-tive prothrombin and partial thromboplastin times anddequate platelet counts. A retrospective chart review of2 patients requiring exploration for bleeding showedimited correlation of preoperative coagulation testingith postoperative coagulopathy [633]. Routine screeningf the intrinsic coagulation system is likely to be unre-arding unless there is a clinical history of bleedingiathesis found on preoperative screening.The utility of preoperative bleeding times as a screen-

ng tool for blood conservation is controversial. Onetudy showed that an in vitro bleeding time device maye a convenient and useful tool to examine primaryemostasis or platelet function in some patients [634]. Arospective study in 219 liver biopsies concluded that theleeding time may give additional information on theisk of periprocedural bleeding in patients at high-riskor surgical bleeding. An observational study in patientsaving a broad spectrum of cardiac procedures sug-ested that bleeding time was an important multivariateredictor of excessive postoperative bleeding [20]. Inontrast, a retrospective review of 1,941 bleeding timeshowed that prolongation of the bleeding time may notlways be associated with excessive surgical blood loss635]. Four prospective investigations of bleeding timesnd bleeding, with study sizes of 118,636, 43,637, 167,638,nd 112,639, all showed that prolonged bleeding timesre not associated with clinically significant perioperativeleeding and are not a reliable routine screening test.urthermore, a position paper by the College of Ameri-an Pathologists and American Society of Clinical Pathol-
gists states that in the absence of a history of excessive

basppparIctb

piporlas

cMawt1ttIsdt[rmorsp(pgTpavne

mapbwrai[

slt

vmtptsboqmss

7A

C

1sftbbv

aaigmcstsdtmtmcrmnrcbmtsb

dtd


leeding or other high-risk features like preoperativentiplatelet therapy, the bleeding time fails as a routinecreening test [640]. Use of the bleeding time test inreoperative patients can only be recommended foratients at high risk for postoperative bleeding [17]. Aotential advantage of the bleeding time test is that anbnormally prolonged bleeding time may be one of theeversible causes of excessive postoperative hemorrhage.f a prolonged test is due to drug effect, it is possible toorrect the hemostatic defect by simply delaying opera-ive intervention until the drug effect wears off and theleeding time returns to the normal range.Preoperative hematocrit can be used as part of the

rediction of risk of blood transfusion, as demonstratedn a prospective study of 1,007 patients [26]. Furthermore,reoperative screening of 1,221 patients suggested thatne of the few preoperative risks that can be targeted toeduce postoperative blood transfusion was anemia andow red blood cell volume [641]. Screening for preoper-tive anemia is likely to suggest cost- and risk-reductiontrategies.

) Intensive Care Unit Processes and Practicesany ICU practices contribute to preoperative blood loss

nd ultimately to postoperative blood transfusion. Cor-in and associates [642] showed that phlebotomy con-

ributes significantly to a large number of transfusions in42 ICU patients with an ICU length of stay of longerhan 1 week. The equivalent of 30% of the total bloodransfused was phlebotomized from patients during theirCU stay. Blood conservation methods including use ofmall-volume tubes, elimination of arterial line bloodiscard, and elimination of standing orders for laboratory

ests decreased phlebotomy blood loss by 33% to 47%642]. Small volume (pediatric) phlebotomy tubes andeduced syringe volumes saved lost blood volume by asuch as 45% [643, 644]. In addition to smaller phlebot-

my volumes, the pattern of laboratory test use can beeduced to minimize the sampling. Drawing as few bloodpecimens as possible was part of a blood conservationrogram for a study of 50 Jehovah’s Witnesses [645]

referenced by Groom [437]). A prospective study of 190atients tested a multimodality blood conservation pro-ram against “standard” measures in a control group.he multimodality conservation program included use ofediatric blood tubes, returning all arterial line flushes,nd minimizing laboratory sampling. The blood conser-ation program significantly decreased bleeding andeed for allogeneic transfusion, in a safe and cost-ffective manner [530].Other blood conservation modalities in critical careonitoring include use of oximetry instead of intra-

rterial blood gas monitoring systems [646]. Threerospective studies looked at the efficacy of a newlood conserving intra-arterial blood gas system,hich gave reliable, accurate, and blood-conserving

esults. Zimmerman and Dellinger [647] studied 104rterial blood gas samples in 5 patients, whereas twonstitutions conducted larger prospective clinical trials
648, 649]. These studies suggest that an intra-arterial p
ystem that monitors oxygen saturation is beneficial inimiting blood loss from arterial blood gas determina-ions in the ICU.

A key aspect of ICU care and physician practice re-olves around comprehensive, optimized, integrated,ultimodality blood conservation programs. Several au-

hors identified a multimodality ICU blood conservationrogram as a significant factor in reducing blood utiliza-

ion [526, 530, 650, 651]. These studies emphasize theavings in blood transfusion that are caused by multiplelood conservation techniques (eg, pediatric blood tubes,ximetry instead of arterial blood gases, reduced fre-uency of blood testing, and so forth) implemented by aultidisciplinary team. These studies also demonstrate

ignificant cost savings associated with these blood con-ervation interventions.

) Transfusion and Blood Conservationlgorithms—the Multimodality Approach

lass I

. A multimodality approach involving multipletakeholders, institutional support, enforceable trans-usion algorithms supplemented with point-of-careesting, and all of the already mentioned efficaciouslood conservation interventions (Table 7) will limitlood transfusion and provide optimal blood conser-ation for cardiac operations. (Level of evidence A)

Guidelines in general, and in particular for the man-gement of bleeding during cardiac procedures, arevailable but are relatively unsuccessful in accomplish-ng change in practice, especially in limiting nonautolo-ous blood transfusion [41, 160, 652, 653]. There is stillarked variability in transfusion practices and blood

onservation interventions. There are at lease two rea-ons for this failure of guidelines. First, accurate andimely information about the platelet and coagulationtatus of patients in the operating room or in the ICU isifficult to obtain. Second, patient, physician, and insti-

utional process variability is difficult to control andanage. These shortcomings caused several investiga-

ors to evaluate transfusion algorithms in a prospectiveanner using institution-derived transfusion practices in

onjunction with accurate point-of-care testing to guideesponses to bleeding and to direct blood transfusion. Inost [530, 654–658] but not all [659] studies, the combi-

ation of point-of-care testing and transfusion algo-ithms used to direct valuable blood resources was bothost effective and efficacious in limiting nonautologouslood product transfusions and providing adequate he-ostasis. These studies include two randomized trials

hat support the use of point-of-care testing and transfu-ion algorithms to limit blood transfusion and enhancelood conservation [657, 658].Multiple point-of-care tests were used in the analyses

escribed above, and it is hard to recommend one overhe other. Further, it is not clear whether the algorithmseveloped for guiding transfusion and the multidisci-
linary approach are more important than the point-of-

T

P

H

L

L

R

P

A

D

T

R

U

R

L

D

R

T

P

G

P

I

I

W

I


able 7. Recommendations for a Multimodality Blood Conservation Program

Recommendation Class

reoperative identification of high-risk patients (advanced age, preoperative anemia, small body size, non–coronary arterybypass graft (CABG) or urgent operation, preoperative antithrombotic drugs, acquired or congenitalcoagulation/clotting abnormalities and multiple patient comorbidities) should be performed, and all availablepreoperative and perioperative measures of blood conservation should be undertaken in this group as they account forthe majority of blood products transfused. (Level of evidence A)

I

igh-dose aprotinin is indicated to reduce the number of patients requiring blood transfusion, to reduce total blood loss,and to limit reexploration in high-risk patients undergoing cardiac operations. Benefits of use should be balancedagainst the increased risk of renal dysfunction. (Level of evidence A)

I

ow-dose aprotinin is indicated to reduce the number of patients requiring blood transfusion and to reduce the totalblood loss in patients having cardiac procedures. (Level of evidence A)

I

ysine analogues like epsilon-aminocaproic acid (EACA) and tranexamic acid (TXA) are indicated to reduce the numberof patients who require blood transfusion, and to reduce total blood loss after cardiac operations. These agents areslightly less potent blood-sparing drugs and the safety profile of these drugs is less well studied compared withaprotinin. (Level of evidence A)

I

outine use of red-cell saving is helpful for blood conservation in cardiac operations using cardiopulmonary bypass(CPB), except in patients with infection or malignancy. (Level of evidence A)

I

reoperative hematocrit and platelet count are indicated for risk prediction, and abnormalities in these variables areamenable to intervention. (Level of evidence A)

I

multimodality approach involving multiple stakeholders, institutional support, enforceable transfusion algorithmssupplemented with point-of-care testing, and all of the already mentioned efficacious blood conservation interventionswill limit blood transfusion and provide optimal blood conservation for cardiac operations. (Level of evidence A)

I

ipyridamole is not indicated to reduce postoperative bleeding, is unnecessary to prevent graft occlusion after coronaryartery bypass grafting, and may increase bleeding risk unnecessarily. (Level of evidence B)

III

ransfusion is unlikely to improve oxygen transport when the hemoglobin concentration is greater than 10 g/dL and isnot recommended. (Level of evidence C)

III

outine prophylactic use of desmopressin acetate (DDAVP) is not recommended to reduce bleeding or blood transfusionafter cardiac operations using CPB. (Level of evidence A)

III

se of prophylactic positive end-expiratory pressure (PEEP) to reduce bleeding postoperatively is not effective. (Level ofevidence B)

III

outine use of intraoperative platelet or plasmapheresis is not recommended for blood conservation during cardiacoperations using CPB. (Level of evidence A)

III

eukocyte filters on the CPB circuit for leukocyte-depletion should not be used for perioperative blood conservation andmay actually activate leukocytes during CPB. (Level of evidence B)

III

irect reinfusion of shed mediastinal blood from postoperative chest tube drainage is not recommended as a means ofblood conservation and may cause harm. (Level of evidence B)

III

outine use of ultrafiltration during or immediately after CPB is not helpful for blood conservation in adult cardiacoperations. (Level of evidence B)

III

opical hemostatic agents that employ bovine thrombin are not helpful for blood conservation during CPB and may bepotentially harmful. (Level of evidence B)

III

reoperative screening of the intrinsic coagulation system is not recommended unless there is a clinical history ofbleeding diathesis. (Level of evidence B)

III

iven that the risk of transmission of known viral diseases with blood transfusion is currently rare, fears of viral diseasetransmission should not limit administration of INDICATED blood products. (This recommendation only applies tocountries/blood banks where careful blood screening exists.) (Level of evidence C)

IIa

atients who have thrombocytopenia (less than 50,000/mm2), who are hyperresponsive to aspirin or other antiplateletdrugs as manifested by abnormal platelet function tests or prolonged bleeding time, or who have known qualitativeplatelet defects represent a high-risk group for bleeding. Maximum blood conservation interventions—during cardiacprocedures are reasonable for these high-risk patients. (Level of evidence B)

IIa

t is reasonable to discontinue thienopyridines 5 to 7 days before cardiac procedures to limit blood loss and transfusion.Failure to discontinue these drugs before operation risks increased bleeding and possibly worse outcome. Caution mustbe used with sudden withdrawal of antiplatelet therapy in the presence of drug-eluting stents. This can lead to stentthrombosis, and the surgical team should consider various alternatives to maintain stent patency. That could eveninclude hospitalization to convert thienopyridine therapy to short-acting GP2b/3a inhibitors for a few days beforeoperation. (Level of evidence B)

IIa

t is reasonable to discontinue low-intensity antiplatelet drugs (eg, aspirin) only in purely elective patients without acutecoronary syndromes before operation with the expectation that blood transfusion will be reduced. (Level of evidence A)

IIa

ith hemoglobin levels below 6 g/dL, red blood cell transfusion is reasonable as this can be life-saving. Transfusion isreasonable in most postoperative patients whose hemoglobin is less than 7 g/dL but no high-level evidence supportsthis recommendation. (Level of evidence C)

IIa

t is reasonable to transfuse non–red cell hemostatic blood products based on clinical evidence of bleeding and preferably IIa
guided by point-of-care tests that assess hemostatic function in a timely and accurate manner. (Level of evidence C)

T

D

I

I

I

R

R

P

O

A

A

T

M

I

I

I

U

U

AI

A


able 7. Recommendations for a Multimodality Blood Conservation Program (Continued)

Recommendation Class

uring cardiopulmonary bypass with moderate hypothermia, transfusion of red cells for a hemoglobin level � 6 g/dL isreasonable except in patients at risk for decreased cerebral oxygen delivery (ie, history of cerebrovascular accident,diabetes mellitus, cerebrovascular disease, carotid stenosis) where higher hemoglobin levels may be justified. (Level ofevidence C)

IIa

n the setting of hemoglobin values exceeding 6 g/dL while on CPB, it is reasonable to transfuse red cells based on thepatient’s clinical situation, and this should be considered as the most important component of the decision-makingprocess. Indications for transfusion of red blood cells in this setting are multifactorial and should be guided by patient-related factors (ie, age, severity of illness, cardiac function, or risk for critical end-organ ischemia), the clinical setting(massive or active blood loss), and laboratory or clinical variables (eg, hematocrit, SVO2, electrocardiographic orechocardiographic evidence of myocardial ischemia, etc). (Level of evidence C)

IIa

n patients after cardiac operations with hemoglobin levels below 6 g/dL, red blood cell transfusion is reasonable and canbe life-saving. Transfusion of red cells is reasonable in most postoperative patients with hemoglobin level of or less, 7g/dL but no high-level evidence supports this recommendation. (Level of evidence C)

IIa

t is reasonable to transfuse non–red cell hemostatic blood products based on clinical evidence of bleeding and preferablyguided by specific point-of-care tests that assess hemostatic function in a timely and accurate manner. (Level ofevidence C)

IIa

ecombinant human erythropoietin (EPO) is reasonable to restore red blood cell volume in patients undergoingautologous preoperative blood donation before cardiac procedures. However, no large-scale safety studies for use ofthis agent in cardiac surgical patients have been performed. Available evidence suggests an acceptable safety profile.(Level of evidence A)

IIa

ecombinant human EPO is a reasonable alternative for anemic low-risk elective patients (hemoglobin � 13 g/dL) beforecardiac procedures provided that EPO is given in conjunction with iron therapy several days or more before operation.(Level of evidence B)

IIa

redonation of as much as two units of autologous blood is reasonable for blood conservation in carefully selected(mostly elective) patients before routine cardiac operation, particularly when coupled with appropriately dosed EPOtherapy and/or iron therapy. (Level of evidence A)

IIa

ff-pump coronary artery bypass (OPCAB) is a reasonable means of blood conservation, provided that emergentconversion to on-pump bypass is unlikely either based on surgeon experience or patient characteristics. (Level ofevidence A)

IIa

lternatives to laboratory blood sampling (eg, oximetry instead of arterial blood gasses) are reasonable means of bloodconservation before operation. (Level of evidence B)

IIa

comprehensive, integrated, multimodality blood conservation program in the intensive care unit is a reasonable meansto limit blood transfusion. (Level of evidence B)

IIa

otal quality management (TQM), including continuous measurement and analysis of blood conservation interventions aswell as assessment of new blood conservation techniques, is reasonable to implement a complete blood conservationprogram. (Level of evidence B)

IIa

ost high-intensity antithrombotic and antiplatelet drugs (including adenosine diphosphate–receptor inhibitors, directthrombin inhibitors, low-molecular-weight heparins, platelet glycoprotein inhibitors, tissue-type plasminogenactivation, streptokinase) are associated with increased bleeding after cardiac operations. It is not unreasonable to stopthese medications before operation to decrease minor and major bleeding events. The timing of discontinuationdepends on the pharmacodynamic half-life for each agent as well as the potential lack of reversibility. Unfractionatedheparin is a notable exception in that it is the only agent that may be discontinued shortly before operation or not atall. (Level of evidence C)

IIb

t is not unreasonable to transfuse red cells in certain patients with critical noncardiac end-organ ischemia (eg, centralnervous system and gut) whose hemoglobin levels are as high as 10 g/dL, but more evidence to support thisrecommendation is required. (Level of evidence C)

IIb

n patients on CPB with risk for critical end-organ ischemia/injury, it is not unreasonable to keep the hemoglobin level at� 7 g/dL or higher. (Level of evidence C)

IIb

t is not unreasonable to use preoperative EPO, given at least a few days before operation, to increase red cell mass inelective patients who are at risk for postoperative anemia and depressed endogenous EPO production. (Level ofevidence C)

IIb

se of DDAVP is not unreasonable to attenuate excessive bleeding and transfusion in certain patients with demonstrableand specific platelet dysfunction known to respond to this agent (eg, uremic or CPB-induced platelet dysfunction, type Ivon Willebrand’s disease). (Level of evidence B)

IIb

se of recombinant factor VIIa concentrate is not unreasonable for the management of intractable nonsurgical bleedingthat is unresponsive to routine hemostatic therapy after cardiac procedures using CPB. (Level of evidence B)

IIb

trial of therapeutic PEEP to reduce excessive postoperative bleeding is not unreasonable. (Level of evidence B) IIbt is not unreasonable to use open venous reservoir membrane oxygenator systems during CPB for reduction in blood

utilization and improved safety. (Level of evidence C)IIb

ll commercially available blood pumps provide acceptable blood conservation during CPB. It is not unreasonable to IIb
prefer centrifugal pumps because of perfusion safety features. (Level of evidence B)

ctr[ttvrgepe

8A

C

1ttccd

mtpasmtcHiminsd

bmTpa

T

F

I

H

L

I

A

D

S

P

RU

T

S

D


are testing used. At least one study strongly suggestshat it is the multimodality approach that is importantather than the individual components of the process530]. Table 7 lists the evidence-based recommendationshat make up a multimodal approach to blood conserva-ion. It is impossible to assign importance to each indi-idual variable in Table 7, and it is possible that the endesult of these interventions on blood conservation isreater than the sum of the parts. Nontheless, availablevidence supports the use of transfusion algorithms andoint-of-care tests to guide transfusion behavior andnhance blood conservation.

) Summary Treatment Strategy—the TQMpproach

lass IIa

. Total quality management (TQM), including con-inuous measurement and analysis of blood conserva-ion interventions as well as assessment of new bloodonservation techniques, is reasonable to implement aomplete blood conservation program. (Level of evi-

able 7. Recommendations for a Multimodality Blood Conser

Recommen

or patients requiring longer CPB times (more than 2–3 hours),or patient-specific heparin concentrations during CPB to reduplatelets and coagulation proteins, and reduce blood transfus

t is not unreasonable to use either protamine titration or empilower the total protamine dose and lower the protamine/hepatransfusion requirements. (Level of evidence B)eparin-coated bypass circuits (either the oxygenator alone orconservation in cardiac operations. (Level of evidence B)

ow doses of systemic heparinization (activated clotting time �CPB, but the possibility of underheparinization and other safevidence B)

t is not unreasonable to use low prime and minimized extracorduring CPB as part of a multimodality blood conservation procute normovolemic hemodilution is not unreasonable for blooIt could be utilized as part of a multipronged approach to blouring CPB, it is not unreasonable to use intraoperative autotrasuction or recycled using a cell-saving device, to augment blo

hortly after the completion of cardiopulmonary bypass, salvagafter washing with a cell-saving device, is not unreasonable a

ostoperative mediastinal shed blood reinfusion using washedblood conservation when used in conjunction with other blooblood may decrease lipid emboli and decrease the concentratblood may be reasonable to limit blood transfusion as part ofevidence B)

etrograde autologous priming of the CPB circuit is not unreasse of modified ultrafiltration is not unreasonable for blood confor whom the priming volume is a very significant fraction of

opical sealants used to assist in the repair of complex high-rislimit bleeding in certain key situations (eg, left ventricular frewith complications that may limit their usefulness in less hig

creening preoperative bleeding time is not unreasonable in hiantiplatelet drugs. (Level of evidence B)evices aimed at obtaining direct hemostasis at catheterizationoperation is planned within 24 hours. (Level of evidence C)

ence B) f

Several threads of evidence support the notion that aultifaceted approach to blood conservation produces

he best results. This approach combines key features ofreoperative assessment to identify high-risk patientslong with perioperative modifications and the totalupport of participating health care providers with com-on goals of limiting postoperative bleeding and blood

ransfusion. Assuming that a multimodality approachan be achieved in a given institution, is that enough?ow does one assess the adequacy of interventions? An

mportant part of blood conservation is ongoing assess-ent of various interventions that are used at a given

nstitution. This is important because of the random andonrandom variations that may result from blood con-ervation interventions resulting in different outcomes atifferent institutions.Continuous assessment of quality improvement has

een termed “total quality management” (TQM). Blu-enthal and Scheck [660] list four major elements of a

QM project: (1) cohort of similar patients, (2) a stablerocess of care, (3) a set of prioritized outcomes, and (4)n integrated measurement and analysis system. It is the

n Program (Continued)

n Class

not unreasonable to consider maintenance of higheremostatic system activation, reduce consumption ofLevel of evidence B)

IIb

w-dose regimens (eg, 50% of total heparin dose) toatio at the end of CPB to reduce bleeding and blood

IIb

ntire circuit) are not unreasonable for blood IIb

s) are not unreasonable for blood conservation duringoncerns have not been well studied. (Level of

IIb

al bypass circuits to reduce the fall in hematocrit. (Level of evidence B)

IIb

servation, but its usefulness is not well established.onservation. (Level of evidence B)

IIb

sion, either with blood directly from cardiotomyonservation. (Level of evidence C)

IIb

ump blood, either administered without washing oreans of blood conservation. (Level of evidence C)

IIb

astinal blood (cell saving) is not unreasonable fornservation interventions. Washing of shed mediastinalf inflammatory cytokines, and reinfusion of washed

ultimodality blood conservation program. (Level of

IIb

le for blood conservation. (Level of evidence B) IIbation during CPB in pediatric patients or in patientstotal blood volume. (Level of evidence B)

IIb

diac and aortic procedures are not unreasonable toll rupture and aortic dissection) but are associated

situations. (Level of evidence C)

IIb

sk patients, especially those who receive preoperative IIb

ss sites are not unreasonable for blood conservation if IIb

vatio

datio

it isce hion. (ric lorin r

the e

300ety c

poregram

d conod cnsfuod ce of ps a mmedid coion oa m

onabservthe

k care wa

h riskgh-ri

acce

ourth part of these criteria that fails in most TQM

ebcMebrspodabvssoiicuirc

sCimieStTppvapaltwufoPpcvmohhgmiduH

buobwc

9

Bcsicpcdstoeboaast

R


ndeavors. The evidence is mostly available to limitlood transfusion after cardiac procedures (ie, there is aohort of similar patients with prioritized outcomes).

ost institutions have already implemented many of thevidence-based blood conservation mechanisms (ie, sta-le process of care). The part that is lacking in ultimatelyeducing bleeding and blood transfusion is the “mea-urement and analysis” of which interventions are im-ortant at a given institution. A complete data system isften lacking in blood conservation. Collecting outcomesata on admission and discharge hematocrit, blood lossnd blood product transfusion, and reoperation forleeding is essential. Accurate risk adjustment usingarious risk factors (eg, surgeon, operation, age, sex, bodyurface area, and so forth) complements this data analy-is. Quality improvement employs the triad of accurateutcomes measurement, thorough risk adjustment, and

dentifying the impact of the various blood conservationnterventions on risk-adjusted outcomes. The TQM pro-ess must be viewed as an ongoing process that contin-ally adjusts interventions consistent with improvement

n blood transfusion practices. Most importantly, thisequires active participation of all members of the healthare team [661].

As an example, one successful set of innovative TQMtudies was carried out by the Northern New Englandardiovascular Disease Study Group [662–665]. These

nvestigators used a risk-adjustment scheme to predictortality in patients undergoing CABG at five different

nstitutions. After risk stratification, significant variabilityxisted among the different institutions and providers.tatistical methods suggested that the variation in mor-

ality rate was nonrandom (ie, “special variability” in theQM vernacular). A peer-based, confidential TQMroject addressed this variability and attempted to im-rove outcomes in the region. To study this nonrandomariability, representatives from each institution visitedll institutions and reviewed the processes involved inerforming CABG. Surgical technique, communicationmong providers, leadership, decision making, trainingevels, and environment were assessed at each institu-ion. Significant variation among many of the processesas observed, and attempts to correct deficiencies werendertaken at each institution. Subsequent publications

rom these authors suggest that this approach improvesutcomes for all providers at all institutions [666, 667].reliminary evidence suggests that this collaborative,eer-based, confidential TQM approach results in a de-rease in operative mortality and, importantly, that pro-iders who were outliers for risk-adjusted operativeortality were able to correct deficiencies and improve

perative outcomes. Each of the practitioners in the fiveospitals had implemented what they thought wereigh-quality interventions in CABG patients before be-inning the TQM intervention. It was not until actualeasurement of the process of care occurred that signif-

cant variability was noted and steps were taken toecrease this variability. A similar TQM process wasndertaken for blood conservation at the New York
ospital with similar results [438]. Likewise, a compara-
le TQM process improved the quality of care in patientsndergoing percutaneous coronary interventions at an-ther institution [668]. This type of TQM approach tolood conservation is viewed as the final common path-ay for applying evidence-based information to blood

onservation.

) Summary

ased on available evidence, institution-specific proto-ols should screen for high-risk patients, as blood con-ervation interventions are likely to be most productiven this high-risk subset. Available evidence-based bloodonservation techniques include (1) drugs that increasereoperative blood volume (eg, erythropoietin) or de-rease postoperative bleeding (eg, antifibrinolytics), (2)evices that conserve blood (eg, intraoperative bloodalvage and blood-sparing interventions), (3) interven-ions that protect the patient’s own blood from the stressf operation (eg, autologous predonation and normovol-mic hemodilution), (4) consensus, institution-specificlood transfusion algorithms supplemented with point-f-care testing, and most importantly (5) a multimodalitypproach to blood conservation combining all of thebove supplemented with a TQM approach to the mea-urement and analysis of all blood conservation interven-ions used.

eferences

1. Jenicek M. Foundations of evidence-based medicine. NewYork: Parthenon, 2002.

2. Stover EP, Siegel LC, Parks R, et al. Variability in transfu-sion practice for coronary artery bypass surgery persistsdespite national consensus guidelines: a 24-institutionstudy. Institutions of the Multicenter Study of PerioperativeIschemia Research Group. Anesthesiology 1998;88:327–33.

3. Blumberg N, Heal JM, Cowles JW, et al. Leukocyte-reducedtransfusions in cardiac surgery: results of an implementa-tion trial. Am J Clin Pathol 2002;118:376–81.

4. van de Watering LM, Hermans J, Houbiers JG, et al.Beneficial effects of leukocyte depletion of transfused bloodon postoperative complications in patients undergoing car-diac surgery: a randomized clinical trial. Circulation 1998;97:562–8.

5. Bilgin YM, van de Watering LM, Eijsman L, et al. Double-blind, randomized controlled trial on the effect of leuko-cyte-depleted erythrocyte transfusions in cardiac valve sur-gery. Circulation 2004;109:2755–60.

6. Lewisohn R. Blood transfusion: 50 years ago and today.Surg Gynecol Obstet 1955;101:362–8.

7. Weiskopf RB, Viele MK, Feiner J, Kelley S, et al. Humancardiovascular and metabolic response to acute, severe iso-volemic anemia. JAMA 1998;279:217–21.

8. Livio M, Gotti E, Marchesi D, Mecca G, Remuzzi G, deGaetano G. Uraemic bleeding: role of anaemia and benefi-cial effect of red cell transfusions. Lancet 1982;2:1013–5.

9. Boneu B, Fernandez F. The role of the hematocrit inbleeding. Transfus Med Rev 1987;1:182–5.

10. Adams RC, Lundy JS. Anesthesia in cases of poor surgicalrisk: some suggestions for decreasing the risk. Surg Gy-necol Obstet 1942;74:10–1.

11. Chapler CK, Cain SM. The physiologic reserve in oxygencarrying capacity: studies in experimental hemodilution.
Can J Physiol Pharmacol 1986;64:7–12.


12. Goodnough LT, Despotis GJ. Transfusion medicine: sup-port of patients undergoing cardiac surgery. Am J Cardio-vasc Drugs 2001;1:337–51.

13. Carson JL, Noveck H, Berlin JA, Gould SA. Mortality andmorbidity in patients with very low postoperative Hb levelswho decline blood transfusion. Transfusion 2002;42:812–8.

14. Carson JL, Duff A, Poses RM, Berlin JA, et al. Effect ofanaemia and cardiovascular disease on surgical mortalityand morbidity. Lancet 1996;348:1055–60.

15. American Society of Anesthesiologists. Practice guidelinesfor blood component therapy: a report by the AmericanSociety of Anesthesiologists Task Force on Blood Compo-nent Therapy. Anesthesiology 1996;84:732–47.

16. Ferraris V, Ferraris S. Limiting excessive postoperativeblood transfusion after cardiac procedures: a review. TexHeart Inst J 1995:22:216–30.

17. Ferraris VA, Ferraris SP, Joseph O, Wehner P, Mentzer RMJr. Aspirin and postoperative bleeding after coronary arterybypass grafting. Ann Surg 2002;235:820–7.

18. Ferraris VA, Ferraris SP, Lough FC, Berry WR. Preoperativeaspirin ingestion increases operative blood loss after coro-nary artery bypass grafting. Ann Thorac Surg 1988;45:71–4.

19. Fiore LD, Brophy MT, Lopez A, Janson P, Deykin D. Thebleeding time response to aspirin. Identifying the hyperre-sponder. Am J Clin Pathol 1990;94:292–6.

20. Ferraris VA, Gildengorin V. Predictors of excessive blooduse after coronary artery bypass grafting. A multivariateanalysis. J Thorac Cardiovasc Surg 1989;98:492–7.

21. Miyagi Y, Yamauchi S, Suzuki S, et al. Investigation ofcoronary artery bypass grafting for a patient with myelo-dysplastic syndrome. Ann Thorac Cardiovasc Surg 2001;7:250–3.

22. Christiansen S, Rotker J, Roeder N, et al. Are patients withWerlhof’s disease at increased risk for bleeding complica-tions when undergoing cardiac surgery? Eur J CardiothoracSurg 2000;18:353–6.

23. Mathew TC, Vasudevan R, Leb L, Pezzella SM, Pezzella AT.Coronary artery bypass grafting in immune thrombocyto-penic purpura. Ann Thorac Surg 1997;64:1059–62.

24. Boccara F, Michel M, Hermine O, Kaplan C, Aumont MC.[Thrombocytopenia after cardiac surgery due to anti-platelet allo-immunization. Apropos of a case.] Arch MalCoeur Vaiss 1997;90:107–10.

25. Jubelirer SJ. Coronary artery bypass in two patients withimmune thrombocytopenic purpura without preoperativesplenectomy. WV Med J 1992;88:510–1.

26. Karkouti K, Cohen MM, McCluskey SA, Sher GD. A mul-tivariable model for predicting the need for blood transfu-sion in patients undergoing first-time elective coronarybypass graft surgery. Transfusion 2001;41:1193–203.

27. Tempe D, Bajwa R, Cooper A, et al. Blood conservation insmall adults undergoing valve surgery. J Cardiothorac VascAnesth 1996;10:502–6.

28. Magovern JA, Sakert T, Benckart DH, et al. A model forpredicting transfusion after coronary artery bypass graft-ing. Ann Thorac Surg 1996;61:27–32.

29. Kabakus N, Yilmaz B, Caliskan U. Investigation of plateletaggregation by impedance and optic methods in childrenwith iron deficiency anaemia. Haematologia (Budapest)2000;30:107–15.

30. Nilsson IM, Magnusson S, Borchgrevink C. The Duke andIvy methods for determination of the bleeding time.Thromb Diath Haemorrh 1963;10:223–34.

31. Wisloff F, Godal HC. Prolonged bleeding time with ade-quate platelet count in hospital patients. Scand J Haematol1981;27:45–50.

32. Castaman G, Federici AB, Rodeghiero F, Mannucci PM.Von Willebrand’s disease in the year 2003: towards thecomplete identification of gene defects for correct diagnosisand treatment. Haematologica 2003;88:94–108.

33. Bellucci S, Caen J. Molecular basis of Glanzmann’s throm-
basthenia and current strategies in treatment. Blood Rev2002;16:193–202.
34. Badhe BA, Jayanthi S, Datta T. Clinical spectrum of Glanz-mann’s thrombasthenia. Indian J Pathol Microbiol 2000;43:297–302.

35. Rand ML, Dean JA. Platelet function in autoimmune (idio-pathic) thrombocytopenic purpura. Acta Paediatr 1998;424(Suppl):57–60.

36. Kenny D, Morateck PA, Gill JC, Montgomery RR. Thecritical interaction of glycoprotein (GP) IBbeta withGPIX—a genetic cause of Bernard-Soulier syndrome. Blood1999;93:2968–75.

37. Jeter EK, Scott A, Kizer J, Lazarchick J. Impaired plateletfunction associated with parenteral nafcillin. Ann Clin LabSci 1990;20:79–84.

38. Goodnough LT, Johnston MF, Toy PT. The variability oftransfusion practice in coronary artery bypass surgery.Transfusion medicine academic award group. JAMA 1991;265:86–90.

39. Belisle S, Hardy JF. Hemorrhage and the use of bloodproducts after adult cardiac operations: myths and realities.Ann Thorac Surg 1996;62:1908–17.

40. Goodnough LT, Soegiarso RW, Birkmeyer JD, Welch HG.Economic impact of inappropriate blood transfusions incoronary artery bypass graft surgery. Am J Med 1993;94:509–14.

41. Goodnough LT, Despotis GJ, Hogue CW Jr, Ferguson TB Jr.On the need for improved transfusion indicators in cardiacsurgery. Ann Thorac Surg 1995;60:473–80.

42. Savino JS, Ley C, Boisvert D, et al. Practice pattern variabil-ity for myocardial revascularization: impact on resource useacross 24 centers. J Cardiothorac Vasc Anesth 2002;16:149–56.

43. Johnson RG, Thurer RJ, Kruskall M, et al. Comparison oftwo transfusion strategies after elective operations for myo-cardial revascularization. J Thorac Cardiovasc Surg 1992;104:307–14.

44. Fischlein T. Coagulation defects in cardiac surgery. Hamo-staseologie 2000;20:110–6.

45. Herwaldt LA, Swartzendruber SK, Zimmerman MB, ScholzDA, Franklin JA, Caldarone CA. Hemorrhage after coro-nary artery bypass graft procedures. Infect Control HospEpidemiol 2003;24:44–50.

46. Michelson EL, Torosian M, Morganroth J, MacVaugh H.Early recognition of surgically correctable causes of exces-sive mediastinal bleeding after coronary artery bypass graftsurgery. Am J Surg 1980;139:313–7.

47. Munoz JJ, Birkmeyer NJ, Dacey LJ, et al. Trends in rates ofreexploration for hemorrhage after coronary artery bypasssurgery. Northern New England Cardiovascular DiseaseStudy Group. Ann Thorac Surg 1999;68:1321–5.

48. Bracey AW, Radovancevic R, Radovancevic B, McAllisterHA Jr, Vaughn WK, Cooley DA. Blood use in patientsundergoing repeat coronary artery bypass graft proce-dures: multivariate analysis. Transfusion 1995;35:850–4.

49. Moskowitz DM, Klein JJ, Shander A, et al. Predictors oftransfusion requirements for cardiac surgical procedures ata blood conservation center. Ann Thorac Surg 2004;77:626–34.

50. Valeri CR, Khabbaz K, Khuri SF, et al. Effect of skintemperature on platelet function in patients undergoingextracorporeal bypass. J Thorac Cardiovasc Surg 1992;104:108–16.

51. Schimmer A, Mody M, Sager M, Garvey MB, Hogarth M,Freedman J. Platelet cold agglutinins: a flow cytometricanalysis. Transfus Sci 1998;19:217–24.

52. Michelson AD, Barnard MR, Khuri SF, Rohrer MJ, MacGre-gor H, Valeri CR. The effects of aspirin and hypothermia onplatelet function in vivo. Br J Haematol 1999;104:64–8.

53. Despotis GJ, Filos KS, Zoys TN, Hogue CW Jr, Spitznagel E,Lappas DG. Factors associated with excessive postoperativeblood loss and hemostatic transfusion requirements: amultivariate analysis in cardiac surgical patients. Anesth
Analg 1996;82:13–21.


54. Nuttall GA, Erchul DT, Haight TJ, et al. A comparison ofbleeding and transfusion in patients who undergo coronaryartery bypass grafting via sternotomy with and withoutcardiopulmonary bypass. J Cardiothorac Vasc Anesth 2003;17:447–51.

55. Sohrabi F, Mispireta LA, Fiocco M, Dibos LA, Buescher PC,Sloane PJ. Effects of off-pump coronary artery bypass graft-ing on patient outcome. J Investig Med 2003;51:27–31.

56. Wijeysundera DN, Beattie WS, Djaiani G, et al. Off-pumpcoronary artery surgery for reducing mortality and morbid-ity: meta-analysis of randomized and observational studies.J Am Coll Cardiol 2005;46:872–82.

57. Davies M-J, Picken J, Buxton B-F, Fuller J-A. Blood-conservation techniques for coronary-artery bypass sur-gery at a private hospital. Med J Aust 1988;149:517–49.

58. Santini F, Luciani GB, Pessotto R, et al. [Replacement of theaortic valve with a pulmonary autograft: experience at theUniversity of Verona.] G Ital Cardiol 1997;27:141–5.

59. Stelzer P, Jones DJ, Elkins RC. Aortic root replacement withpulmonary autograft. Circulation 1989;80:III209–213.

60. Green K, Liska J, Egberg N, et al. Hemostatic disturbancesassociated with implantation of an artificial heart. ThrombRes 1987;48:349–62.

61. Wang IW, Kottke-Marchant K, Vargo RL, McCarthy PM.Hemostatic profiles of HeartMate ventricular assist devicerecipients. ASAIO J 1995;41:M782–7.

62. Schmid C, Weyand M, Hammel D, Deng MC, Nabavi D,Scheld HH. Effect of platelet inhibitors on thromboembo-lism after implantation of a Novacor N100—preliminaryresults. Thorac Cardiovasc Surg 1998;46:260–2.

63. Meyer BJ. [Current approach in antithrombotic treatmentduring coronary interventions: specific thrombocyte aggre-gation inhibitors and direct thrombin antagonists in high-risk patients.] Schweiz Med Wochenschr 1996;126:1961–9.

64. Machraoui A, Germing A, von Dryander S, et al. Highpressure coronary stenting: efficacy and safety of aspirinversus coumadin plus aspirin—preliminary results of arandomized study. J Invasive Cardiol 1997;9:171–6.

65. Wheeldon N, Cumberland D. Pharmacologic prevention ofacute ischemic complications of coronary angioplasty.Cathet Cardiovasc Diagn 1997;42:249–56.

66. Morice MC, Aubry P, Benveniste E, Bourdonnec C, Com-meau P. The MUST trial: acute results and six-monthclinical follow-up. J Invas Cardiol 1998;10:457–63.

67. Lust RM, Bode AP, Yang L, Hodges W, Chitwood WR Jr.In-line leukocyte filtration during bypass. Clinical resultsfrom a randomized prospective trial. ASAIO J 1996;42:M819–22.

68. Hashimoto H, Bohmer RM, Harrell LC, Palacios IF. Con-tinuous quality improvement decreases length of stay andadverse events: a case study in an interventional cardiologyprogram. Am J Manag Care 1997;3:1141–50.

69. Karrillon GJ, Morice MC, Benveniste E, et al. Intracoronarystent implantation without ultrasound guidance and withreplacement of conventional anticoagulation by antiplatelettherapy. 30-day clinical outcome of the French MulticenterRegistry. Circulation 1996;94:1519–27.

70. Schomig A, Neumann FJ, Kastrati A, et al. A randomizedcomparison of antiplatelet and anticoagulant therapy afterthe placement of coronary-artery stents. N Engl J Med1996;334:1084–9.

71. Reffelmann T, vom Dahl J, Hendriks F, Klues HG, HanrathP. [Reduction of in-hospital complications after electivepercutaneous transluminal coronary angioplasty using acombined pretreatment with ticlopidine and aspirin.] ZKardiol 1998;87:872–80.

72. Scavetta K, Oh C, Caldron R, et al. Results of saphenousvein graft stent implantation: single center results from useof oversized balloon catheters. Angiology 1999;50:891–9.

73. Yang X, Alexander KP, Chen AY, et al. The implications ofblood transfusions for patients with non-ST-segment ele-
vation acute coronary syndromes: results from the CRU-
SADE national quality improvement initiative. J Am CollCardiol 2005;46:1490–5.

74. Iturbe T, Sanchez C, Moreno JA, et al. [Comparative studyof 2 antithrombotic regimens in 75 patients with coronaryendoprostheses.] Sangre (Barcelona) 1997;42:387–90.

75. Yip HK, Chang HW, Wu CJ, et al. A safe and effectiveregimen without heparin therapy after successful primarycoronary stenting in patients with acute myocardial infarc-tion. Jpn Heart J 2000;41:697–711.

76. Giordano GF Sr, Giordano GF Jr, Rivers SL, et al. Determi-nants of homologous blood usage utilizing autologousplatelet-rich plasma in cardiac operations. Ann ThoracSurg 1989;47:897–902.

77. Pfisterer M, Burkart F, Jockers G, et al. Trial of low-doseaspirin plus dipyridamole versus anticoagulants for pre-vention of aortocoronary vein graft occlusion. Lancet 1989;2:1–7.

78. Inada E. Blood coagulation and autologous blood transfu-sion in cardiac surgery. J Clin Anesth 1990;2:393–406.

79. van der Meer J, Hillege HL, Kootstra GJ, et al. Prevention ofone-year vein-graft occlusion after aortocoronary-bypasssurgery: a comparison of low-dose aspirin, low-dose aspirinplus dipyridamole, and oral anticoagulants. The CABADASresearch group of the Interuniversity Cardiology Instituteof The Netherlands. Lancet 1993;342:257–64.

80. Loewen P, Sunderji R, Gin K. The efficacy and safety ofcombination warfarin and ASA therapy: a systematic re-view of the literature and update of guidelines. Can J Car-diol 1998;14:717–26.

81. Pothula S, Sanchala VT, Nagappala B, Inchiosa MA Jr. Theeffect of preoperative antiplatelet/anticoagulant prophy-laxis on postoperative blood loss in cardiac surgery. AnesthAnalg 2004;98:4–10.

82. van der Meer J, Brutel de la Riviere A, van Gilst WH, et al.Effects of low-dose aspirin (50 mg/day), low-dose aspirinplus dipyridamole, and oral anticoagulant agents afterinternal mammary artery bypass grafting: patency andclinical outcome at 1 year. CABADAS research group of theInteruniversity Cardiology Institute of The Netherlands.Prevention of coronary artery bypass graft occlusion byaspirin, dipyridamole and acenocoumarol/phenprocoumonstudy. J Am Coll Cardiol 1994;24:1181–8.

83. Chauhan S, Ghosh T, Srivastava S, et al. Heparin dosingand postoperative blood loss in patients taking aspirin.Asian Cardiovasc Thorac Ann 1998;6:270–2.

84. Jakics J, Lee J, Ikeda S. Preoperative aspirin and heparintherapy does not increase perioperative blood loss andblood product requirements in coronary artery bypass graftsurgery. J Anesth 1999;13:8–13.

85. Grubitzsch H, Wollert HG, Eckel L. Emergency coronaryartery bypass grafting: does excessive preoperative antico-agulation increase bleeding complications and transfusionrequirements? Cardiovasc Surg 2001;9:510–6.

86. Bjessmo S, Ivert T. Blood loss after coronary artery bypasssurgery: relations to patient variables and antithrombotictreatment. Scand Cardiovasc J 2000;34:438–45.

87. Ferraris VA, Ferraris SP, Moliterno DJ, et al. The Society ofThoracic Surgeons practice guideline series: aspirin andother antiplatelet agents during operative coronary revas-cularization (executive summary). Ann Thorac Surg 2005;79:1454–61.

88. Mangano DT. Aspirin and mortality from coronary bypasssurgery. N Engl J Med 2002;347:1309–17.

89. Teoh KH, Christakis GT, Weisel RD, et al. Dipyridamolepreserved platelets and reduced blood loss after cardiopul-monary bypass. J Thorac Cardiovasc Surg 1988;96:332–41.

90. Goldman S, Copeland J, Moritz T, et al. Improvement inearly saphenous vein graft patency after coronary arterybypass surgery with antiplatelet therapy: results of a Vet-erans Administration Cooperative Study. Circulation 1988;77:1324–32.

91. Hays LJ, Beller GA, Moore CA, et al. Short-term infarctvessel patency with aspirin and dipyridamole started 24 to

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1


36 hours after intravenous streptokinase. Am Heart J1988;115:717–21.

92. Sanz G, Pajaron A, Alegria E, et al. Prevention of earlyaortocoronary bypass occlusion by low-dose aspirin anddipyridamole. Grupo Espanol para el Seguimiento delInjerto Coronario (GESIC). Circulation 1990;82:765–73.

93. Stein PD, Dalen JE, Goldman S, Schwartz L, Theroux P,Turpie AG. Antithrombotic therapy in patients with saphe-nous vein and internal mammary artery bypass grafts.Chest 1995;108(Suppl):424–30.

94. Chesebro JH. Effect of dipyridamole and aspirin on veingraft patency after coronary bypass operations. ThrombRes 1990;12(Suppl):5–10.

95. Breyer RH, Rousou JA, Engelman RM, Lemeshow S. Latepostoperative tamponade following coronary artery bypassgrafting in patients on antiplatelet therapy. Ann ThoracSurg 1985;39:27–9.

96. Antithrombotic Trialists’ Collaboration. Collaborativemeta-analysis of randomised trials of antiplatelet therapyfor prevention of death, myocardial infarction, and strokein high risk patients. Br Med J 2002;324:71–86.

97. Lubbe DF, Berger PB. The thienopyridines. J Interv Cardiol2002;15:85–93.

98. Yoon Y, Shim WH, Lee DH, et al. Usefulness of cilostazolversus ticlopidine in coronary artery stenting. Am J Cardiol1999;84:1375–80.

99. Casella G, Ottani F, Pavesi PC, et al. Safety and efficacyevaluation of clopidogrel compared with ticlopidine afterstent implantation: an updated meta-analysis. Ital Heart J2003;4:677–84.

00. Muller C, Buttner HJ, Petersen J, Roskamm H. A random-ized comparison of clopidogrel and aspirin versus ticlopi-dine and aspirin after the placement of coronary-arterystents. Circulation 2000;101:590–3.

01. Jarvis B, Simpson K. Clopidogrel: a review of its use in theprevention of atherothrombosis. Drugs 2000;60:347–77.

02. Hankey GJ, Sudlow CL, Dunbabin DW. Thienopyridines oraspirin to prevent stroke and other serious vascular eventsin patients at high risk of vascular disease? A systematicreview of the evidence from randomized trials. Stroke2000;31:1779–84.

03. Peters RJ, Mehta SR, Fox KA, et al. Effects of aspirin dosewhen used alone or in combination with clopidogrel inpatients with acute coronary syndromes: observations fromthe Clopidogrel in Unstable angina to prevent RecurrentEvents (CURE) study. Circulation 2003;108:1682–7.

04. Yende S, Wunderink RG. Effect of clopidogrel on bleedingafter coronary artery bypass surgery. Crit Care Med 2001;29:2271–5.

05. Hongo RH, Ley J, Dick SE, Yee RR. The effect of clopidogrelin combination with aspirin when given before coronaryartery bypass grafting. J Am Coll Cardiol 2002;40:231–7.

06. Gansera B, Schmidtler F, Spiliopoulos K, Angelis I, Neu-maier-Prauser P, Kemkes BM. Urgent or emergent coro-nary revascularization using bilateral internal thoracic ar-tery after previous clopidogrel antiplatelet therapy. ThoracCardiovasc Surg 2003;51:185–9.

07. Ray JG, Deniz S, Olivieri A, et al. Increased blood productuse among coronary artery bypass patients prescribedpreoperative aspirin and clopidogrel. BMC Cardiovasc Dis-ord 2003;3:3.

08. Ley SJ. Quality care outcomes in cardiac surgery: the role ofevidence-based practice. AACN clinical issues. Adv PractAcute Crit Care 2001;12:606–17.

09. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA2002 guideline update for the management of patients withunstable angina and non-ST-segment elevation myocardialinfarction—summary article: a report of the American Col-lege of Cardiology/American Heart Association task forceon practice guidelines (committee on the management ofpatients with unstable angina). J Am Coll Cardiol 2002;40:
1366–74.
10. Iakovou I, Schmidt T, Bonizzoni E, et al. Incidence, predic-tors, and outcome of thrombosis after successful implanta-tion of drug-eluting stents. JAMA 2005;293:2126–30.

11. Jeremias A, Sylvia B, Bridges J, et al. Stent thrombosis aftersuccessful sirolimus-eluting stent implantation. Circulation2004;109:1930–2.

12. Joner M, Finn AV, Farb A, et al. Pathology of drug-elutingstents in humans: delayed healing and late thrombotic risk.J Am Coll Cardiol 2006;48:193–202.

13. Nebeker JR, Virmani R, Bennett CL, et al. Hypersensitivitycases associated with drug-eluting coronary stents: a re-view of available cases from the Research on Adverse DrugEvents and Reports (RADAR) project. J Am Coll Cardiol2006;47:175–81.

14. Coller BS, Anderson K, Weisman HF. New antiplateletagents: platelet GPIIb/IIIa antagonists. Thromb Haemost1995;74:302–8.

15. Harrington RA, Kleiman NS, Kottke-Marchant K, et al.Immediate and reversible platelet inhibition after intrave-nous administration of a peptide glycoprotein IIb/IIIa in-hibitor during percutaneous coronary intervention. Am JCardiol 1995;76:1222–7.

16. RESTORE Investigators. Effects of platelet glycoproteinIIb/IIIa blockade with tirofiban on adverse cardiac events inpatients with unstable angina or acute myocardial infarc-tion undergoing coronary angioplasty. The RESTORE In-vestigators. Randomized Efficacy Study of Tirofiban forOutcomes and REstenosis. Circulation 1997;96:1445–53.

17. Goa KL, Noble S. Eptifibatide: a review of its use in patientswith acute coronary syndromes and/or undergoing percu-taneous coronary intervention. Drugs 1999;57:439–62.

18. Choussat R, Black A, Bossi I, Fajadet J, Marco J. Vascularcomplications and clinical outcome after coronary angio-plasty with platelet IIb/IIIa receptor blockade. Comparisonof transradial versus transfemoral arterial access. EurHeart J 2000;21:662–7.

19. Tcheng JE, Kereiakes DJ, Lincoff AM, et al. Abciximabreadministration: results of the ReoPro ReadministrationRegistry. Circulation 2001;104:870–5.

20. Gowda MS, Vacek JL, Lakkireddy DJ, Brosnahan K,Beauchamp GD. Differential benefits and outcomes oftirofiban versus abciximab for acute coronary syndromes incurrent clinical practice. Angiology 2003;54:211–8.

21. Mahaffey KW, Granger CB, Collins R, et al. Overview ofrandomized trials of intravenous heparin in patients withacute myocardial infarction treated with thrombolytic ther-apy. Am J Cardiol 1996;77:551–6.

22. Agnelli G, Sonaglia F. Low-molecular-weight heparins inthe management of unstable angina. Haemostasis 2000;30(Suppl 2):158–67; discussion 146–7.

23. Zemour G, Morice MC, Benveniste E, et al. [Coronaryartery stenting without antivitamin K. Results after amonth]. Arch Mal Coeur Vaiss 1996;89:291–7.

24. Jones HU, Muhlestein JB, Jones KW, et al. Preoperative useof enoxaparin compared with unfractionated heparin in-creases the incidence of re-exploration for postoperativebleeding after open-heart surgery in patients who presentwith an acute coronary syndrome: clinical investigation andreports. Circulation 24 2002;106(Suppl 1):19–22.

25. Kincaid EH, Monroe ML, Saliba DL, Kon ND, Byerly WG,Reichert MG. Effects of preoperative enoxaparin versusunfractionated heparin on bleeding indices in patientsundergoing coronary artery bypass grafting. Ann ThoracSurg 2003;76:124–8.

26. Medalion B, Frenkel G, Patachenko P, Hauptman E, SassonL, Schachner A. Preoperative use of enoxaparin is not a riskfactor for postoperative bleeding after coronary artery by-pass surgery. J Thorac Cardiovasc Surg 2003;126:1875–9.

27. Pifarre R, Walenga JM, Fareed J. Clinical disadvantageswith the use of heparin as an anticoagulant. In: Pifarre R,ed. New anticoagulants for the cardiovascular patient.
Philadelphia: Hanley and Belfus1997:1–7.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1


28. Weitz JI, Crowther M. Direct thrombin inhibitors. ThrombRes 2002;106:V275–84.

29. Organization to Assess Strategies for Ischemic Syndromes(OASIS) Investigators. Comparison of the effects of twodoses of recombinant hirudin compared with heparin inpatients with acute myocardial ischemia without ST eleva-tion: a pilot study. Circulation 1997;96:769–77.

30. Bittl JA, Chaitman BR, Feit F, Kimball W, Topol EJ. Biva-lirudin versus heparin during coronary angioplasty forunstable or postinfarction angina: final report reanalysis ofthe Bivalirudin Angioplasty Study. Am Heart J 2001;142:952–9.

31. Kong DF, Topol EJ, Bittl JA, et al. Clinical outcomes ofbivalirudin for ischemic heart disease. Circulation 1999;100:2049–53.

32. Seybert AL. Bivalirudin administration during percutane-ous coronary intervention: emphasis on high-risk patients.Pharmacotherapy 2002;22(Suppl):112–8.

33. Mahaffey KW, Lewis BE, Wildermann NM, et al. Theanticoagulant therapy with bivalirudin to assist in theperformance of percutaneous coronary intervention in pa-tients with heparin-induced thrombocytopenia (ATBAT)study: main results. J Invas Cardiol 2003;15:611–6.

34. Lincoff AM, Bittl JA, Kleiman NS, et al. Comparison ofbivalirudin versus heparin during percutaneous coronaryintervention (the Randomized Evaluation of PCI LinkingAngiomax to Reduced Clinical Events [REPLACE]-1 trial).Am J Cardiol 2004;93:1092–6.

35. Merry AF, Raudkivi PJ, Middleton NG, et al. Bivalirudinversus heparin and protamine in off-pump coronary arterybypass surgery. Ann Thorac Surg 2004;77:925–31.

36. Green JA, Cooper CL, Falcucci OA, Safwat A, Slaughter T,Spiess BD. Argatroban for off-pump coronary artery bypasssurgery. Anesth Analg 2003;97:1201 [author reply 1201-2].

37. McDonald SB, Kattapurum BM, Saleem R, Joist HJ, AvidanM, Despotis GJ. Monitoring hirudin anticoagulation in twopatients undergoing cardiac surgery with a plasma-modified act method. Anesthesiology 2002;97:509–12.

38. Koster A, Yeter R, Buz S, et al. Assessment of hemostaticactivation during cardiopulmonary bypass for coronaryartery bypass grafting with bivalirudin: results of a pilotstudy. J Thorac Cardiovasc Surg 2005;129:1391–4.

39. Koster A, Spiess B, Chew DP, et al. Effectiveness of biva-lirudin as a replacement for heparin during cardiopulmo-nary bypass in patients undergoing coronary artery bypassgrafting. Am J Cardiol 1 2004;93:356–9.

40. Spiess BD. Choose one: damned if you do/damned if youdon’t! Crit Care Med 2005;33:1871–4.

41. Mann MJ, Tseng E, Ratcliffe M, et al. Use of bivalirudin, adirect thrombin inhibitor, and its reversal with modifiedultrafiltration during heart transplantation in a patient withheparin-induced thrombocytopenia. J Heart Lung Trans-plant 2005;24:222–5.

42. Jan KM, Chien S. Effect of hematocrit variations on coro-nary hemodynamics and oxygen utilization. Am J Physiol1977;233:H106–13.

43. Utley JR. Pathophysiology of cardiopulmonary bypass: cur-rent issues. J Card Surg 1990;5:177–89.

44. Utley JR. Historical perspectives and basic pathophysiol-ogy. Semin Thorac Cardiovasc Surg 1990;2:292–9.

45. Mathru M, Kleinman B, Blakeman B, Dries D, Zecca A, RaoT. Cardiovascular adjustments and gas exchange duringextreme hemodilution in humans. Crit Care Med 1991;19:700–4.

46. Cosgrove DM, Loop FD, Lytle BW, et al. Determinants ofblood utilization during myocardial revascularization. AnnThorac Surg 1985;40:380–4.

47. Lowenstein E. Blood conservation in open heart surgery.Cleve Clin Q 1981;48:112–5.

48. Weisel RD, Charlesworth DC, Mickleborough LL, et al.Limitations of blood conservation. J Thorac Cardiovasc
Surg 1984;88:26–38.
49. Torres Filho IP, Spiess BD, Pittman RN, Barbee RW, WardKR. Experimental analysis of critical oxygen delivery. Am JPhysiol Heart Circ Physiol 2005;288:H1071–9.

50. van Woerkens EC, Trouwborst A, van Lanschot JJ. Pro-found hemodilution: what is the critical level of hemodilu-tion at which oxygen delivery-dependent oxygen consump-tion starts in an anesthetized human? Anesth Analg 1992;75:818–21.

51. Doak GJ, Hall RI. Does hemoglobin concentration affectperioperative myocardial lactate flux in patients undergo-ing coronary artery bypass surgery? Anesth Analg 1995;80:910–6.

52. Spahn DR, Smith LR, Veronee CD, et al. Acute isovolemichemodilution and blood transfusion. Effects on regionalfunction and metabolism in myocardium with compro-mised coronary blood flow. J Thorac Cardiovasc Surg1993;105:694–704.

53. Fang WC, Helm RE, Krieger KH, et al. Impact of minimumhematocrit during cardiopulmonary bypass on mortality inpatients undergoing coronary artery surgery. Circulation1997;96(Suppl 2):194–9.

54. Ott DA, Cooley DA. Cardiovascular surgery in Jehovah’sWitnesses. Report of 542 operations without blood transfu-sion. JAMA 1977;238:1256–8.

55. Spence RK, Alexander JB, DelRossi AJ, et al. Transfusionguidelines for cardiovascular surgery: lessons learned fromoperations in Jehovah’s Witnesses. J Vasc Surg 1992;16:825–31.

56. Viele MK, Weiskopf RB. What can we learn about the needfor transfusion from patients who refuse blood? The expe-rience with Jehovah’s Witnesses. Transfusion 1994;34:396–401.

57. NIH Consensus Conference. Perioperative red blood celltransfusion. JAMA 1988;260:2700–3.

58. American College of Physicians. Practice strategies forelective red blood cell transfusion. Ann Intern Med 1992;116:403–6.

59. Stehling L, Luban NL, Anderson KC, et al. Guidelines forblood utilization review. Transfusion 1994;34:438–48.

60. Goodnough LT, Johnston MF, Ramsey G, et al. Guidelinesfor transfusion support in patients undergoing coronaryartery bypass grafting. Transfusion Practices Committee ofthe American Association of Blood Banks. Ann Thorac Surg1990;50:675–83.

61. American Society of Anesthesiologists Task Force on BloodComponent Therapy. Practice guidelines for blood compo-nent therapy: A report by the American Society of Anes-thesiologists task force on blood component therapy. An-esthesiology 1996;84:732–47.

62. Spence R, Carson J, Poses R, et al. Elective surgery withouttransfusion: influence of preoperative hemoglobin leveland blood loss on mortality. Am J Surg 1990;159:320–4.

63. Bracey AW, Radovancevic R, Riggs SA, et al. Lowering thehemoglobin threshold for transfusion in coronary arterybypass procedures: effect on patient outcome. Transfusion1999;39:1070–7.

64. DeFoe GR, Ross CS, Olmstead EM, et al. Lowest hematocriton bypass and adverse outcomes associated with coronaryartery bypass grafting. Northern New England Cardiovas-cular Disease Study Group. Ann Thorac Surg 2001;71:769–76.

65. Karkouti K, Djaiani G, Borger MA, et al. Low hematocritduring cardiopulmonary bypass is associated with in-creased risk of perioperative stroke in cardiac surgery. AnnThorac Surg 2005;80:1381–7.

66. Habib RH, Zacharias A, Schwann TA, Riordan CJ, DurhamSJ, Shah A. Adverse effects of low hematocrit duringcardiopulmonary bypass in the adult: should current prac-tice be changed? J Thorac Cardiovasc Surg 2003;125:1438–50.

67. Spiess BD, Royston D, Levy JH, et al. Platelet transfusions
during coronary artery bypass graft surgery are associatedwith serious adverse outcomes. Transfusion 2004;44:1143–8.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2


68. Banbury MK, Brizzio ME, Rajeswaran J, Lytle BW, Black-stone EH. Transfusion increases the risk of postoperativeinfection after cardiovascular surgery. J Am Coll Surg2006;202:131–8.

69. Surgenor SD, DeFoe GR, Fillinger MP, et al. Intraoperativered blood cell transfusion during coronary artery bypassgraft surgery increases the risk of postoperative low-outputheart failure. Circulation 2006;114(Suppl 1):43–8.

70. Engoren MC, Habib RH, Zacharias A, Schwann TA, Rior-dan CJ, Durham SJ. Effect of blood transfusion on long-term survival after cardiac operation. Ann Thorac Surg2002;74:1180–6.

71. Karkouti K, Wijeysundera DN, Yau TM, et al. Platelettransfusions are not associated with increased morbidity ormortality in cardiac surgery. Can J Anaesth 2006;53:279–87.

72. Cohen E, Neustein SM, Silvay G. Profound anemia follow-ing cardiac surgery. J Cardiothorac Vasc Anesth 1993;7:721–3.

73. Robertie PG, Gravlee GP. Safe limits of isovolemic hemodi-lution and recommendations for erythrocyte transfusion.Int Anesthesiol Clin 1990;28:197–204.

74. American Society of Anesthesiologists. Practice guidelinesfor blood component therapy: a report by the AmericanSociety of Anesthesiologists Task Force on Blood Compo-nent Therapy. Anesthesiology 1996;84:732–47.

75. Hebert PC, Wells G, Tweeddale M, et al. Does transfusionpractice affect mortality in critically ill patients? TransfusionRequirements in Critical Care (TRICC) Investigators andthe Canadian Critical Care Trials Group. Am J Respir CritCare Med 1997;155:1618–23.

76. Hebert PC, Wells G, Blajchman MA, et al. A multicenter,randomized, controlled clinical trial of transfusion require-ments in critical care. Transfusion Requirements in CriticalCare Investigators, Canadian Critical Care Trials Group.N Engl J Med 1999;340:409–17.

77. Hebert PC, Yetisir E, Martin C, et al. Is a low transfusionthreshold safe in critically ill patients with cardiovasculardiseases? Crit Care Med 2001;29:227–34.

78. Spiess BD, Ley C, Body SC, et al. Hematocrit value onintensive care unit entry influences the frequency of Q-wave myocardial infarction after coronary artery bypassgrafting. The Institutions of the Multicenter Study of Peri-operative Ischemia (McSPI) Research Group. J Thorac Car-diovasc Surg 1998;116:460–7.

79. Suttner S, Piper SN, Kumle B, et al. The influence ofallogeneic red blood cell transfusion compared with 100%oxygen ventilation on systemic oxygen transport and skel-etal muscle oxygen tension after cardiac surgery. AnesthAnalg 2004;99:2–11.

80. van Bommel J, de Korte D, Lind A, et al. The effect of thetransfusion of stored RBCs on intestinal microvascularoxygenation in the rat. Transfusion 2001;41:1515–23.

81. Tsai AG, Cabrales P, Intaglietta M. Microvascular perfusionupon exchange transfusion with stored red blood cells innormovolemic anemic conditions. Transfusion 2004;44:1626–34.

82. Mangano DT, Tudor IC, Dietzel C. The risk associated withaprotinin in cardiac surgery. N Engl J Med 2006;354:353–65.

83. Karkouti K, Beattie WS, Dattilo KM, et al. A propensityscore case-control comparison of aprotinin and tranexamicacid in high-transfusion-risk cardiac surgery. Transfusion2006;46:327–38.

84. Royston D, Bidstrup BP, Taylor KM, Sapsford RN. Effect ofaprotinin on need for blood transfusion after repeat open-heart surgery. Lancet 1987;2:1289–91.

85. Brown JR, Birkmeyer NJ, O’Connor GT. Aprotinin in car-diac surgery [Letter]. N Engl J Med 2006;354:1954–5.

86. Sedrakyan A, Treasure T, Elefteriades JA. Effect of aprotininon clinical outcomes in coronary artery bypass graft sur-gery: a systematic review and meta-analysis of randomized
clinical trials. J Thorac Cardiovasc Surg 2004;128:442–8.
87. Henry DA, Moxey AJ, Carless PA, et al. Anti-fibrinolytic usefor minimising perioperative allogeneic blood transfusion.Cochrane Database Syst Rev 2001:CD001886.

88. Laupacis A, Fergusson D. Drugs to minimize perioperativeblood loss in cardiac surgery: meta-analyses using periop-erative blood transfusion as the outcome. The InternationalStudy of Perioperative Transfusion (ISPOT) Investigators.Anesth Analg 1997;85:1258–67.

89. Levy JH, Pifarre R, Schaff HV, et al. A multicenter, double-blind, placebo-controlled trial of aprotinin for reducingblood loss and the requirement for donor-blood transfu-sion in patients undergoing repeat coronary artery bypassgrafting. Circulation 1995;92:2236–44.

90. Lemmer JH Jr, Stanford W, Bonney SL, et al. Aprotinin forcoronary bypass operations: efficacy, safety, and influenceon early saphenous vein graft patency. A multicenter,randomized, double-blind, placebo-controlled study.J Thorac Cardiovasc Surg 1994;107:543–51.

91. Cosgrove DM III, Heric B, Lytle BW, et al. Aprotinintherapy for reoperative myocardial revascularization: aplacebo-controlled study. Ann Thorac Surg 1992;54:1031–8.

92. Minami K, Notohamiprodjo G, Buschler H, Prohaska W,Reichelt W, Korfer R. Alpha-2 plasmin inhibitor-plasmincomplex and postoperative blood loss: double-blind studywith aprotinin in reoperation for myocardial revasculariza-tion. J Thorac Cardiovasc Surg 1993;106:934–6.

93. Swart MJ, Gordon PC, Hayse-Gregson PB, et al. High-doseaprotinin in cardiac surgery—a prospective, randomizedstudy. Anaesth Intens Care 1994;22:529–33.

94. Carrera A, Martinez MV, Garcia-Guiral M, Herrero E, PeralA, Planas A. [Use of high doses of aprotinin in cardiacsurgery.] Rev Esp Anestesiol Reanim 1994;41:13–9.

95. Rocha E, Hidalgo F, Llorens R, Melero JM, Arroyo JL,Paramo JA. Randomized study of aprotinin and DDAVP toreduce postoperative bleeding after cardiopulmonary by-pass surgery. Circulation 1994;90:921–7.

96. Maccario M, Fumagalli C, Deangelis R, et al. [Comparisonbetween low and high doses of aprotinin in heart surgery.]Minerva Anestesiol 1994;60:315–20.

97. Diprose P, Herbertson MJ, O’Shaughnessy D, Deakin CD,Gill RS. Reducing allogeneic transfusion in cardiac surgery:a randomized double-blind placebo-controlled trial of an-tifibrinolytic therapies used in addition to intra-operativecell salvage. Br J Anaesth 2005;94:271–8.

98. Penta de Peppo A, Pierri MD, Scafuri A, et al. Intraopera-tive antifibrinolysis and blood-saving techniques in cardiacsurgery. Prospective trial of 3 antifibrinolytic drugs. TexHeart Inst J 1995;22:231–6.

99. Corbeau JJ, Monrigal JP, Jacob JP, et al. [Comparison ofeffects of aprotinin and tranexamic acid on blood loss inheart surgery.] Ann Fr Anesth Reanim 1995;14:154–61.

00. Casas JI, Zuazu-Jausoro I, Mateo J, et al. Aprotinin versusdesmopressin for patients undergoing operations with car-diopulmonary bypass. A double-blind placebo-controlledstudy. J Thorac Cardiovasc Surg 1995;110:1107–7.

01. Lemmer JH Jr, Dilling EW, Morton JR, et al. Aprotinin forprimary coronary artery bypass grafting: a multicenter trialof three dose regimens. Ann Thorac Surg 1996;62:1659–68.

02. Hardy JF, Desroches J, Belisle S, Perrault J, Carrier M,Robitaille D. Low-dose aprotinin infusion is not clinicallyuseful to reduce bleeding and transfusion of homologousblood products in high-risk cardiac surgical patients. Can JAnaesth 1993;40:625–31.

03. Liu B, Belboul A, Radberg G, et al. Effect of reducedaprotinin dosage on blood loss and use of blood products inpatients undergoing cardiopulmonary bypass. Scand J Tho-rac Cardiovasc Surg 1993;27:149–55.

04. Hayashida N, Isomura T, Sato T, Maruyama H, Kosuga K,Aoyagi S. Effects of minimal-dose aprotinin on coronaryartery bypass grafting. J Thorac Cardiovasc Surg 1997;114:
261–9.

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2


05. Alvarez JM, Quiney NF, McMillan D, et al. The use ofultra-low-dose aprotinin to reduce blood loss in cardiacsurgery. J Cardiothorac Vasc Anesth 1995;9:29–33.

06. Pugh SC, Wielogorski AK. A comparison of the effects oftranexamic acid and low-dose aprotinin on blood loss andhomologous blood usage in patients undergoing cardiacsurgery. J Cardiothorac Vasc Anesth 1995;9:240–4.

07. Bailey CR, Kelleher AA, Wielogorski AK. Randomizedplacebo-controlled double-blind study of three aprotininregimens in primary cardiac surgery. Br J Surg 1994;81:969–73.

08. Bertrand P, Mazzucotelli JP, Loisance D, Elsayed A, De-leuze P, Cachera JP. [Aprotinin in cardiac surgery in pa-tients with platelet anti-aggregant treatment.] Arch MalCoeur Vaiss 1993;86:1471–4.

09. Harder MP, Eijsman L, Roozendaal KJ, van Oeveren W,Wildevuur CR. Aprotinin reduces intraoperative and post-operative blood loss in membrane oxygenator cardiopul-monary bypass. Ann Thorac Surg 1991;51:936–41.

10. Locatelli A, Bertollo D, Bianchi T, et al. [Aprotinin incardiosurgery: a randomized prospective study with differ-ent protocols for use.] Minerva Anestesiol 1990;56:973–5.

11. Alajmo F, Calamai G, Perna AM, et al. High-dose aprotinin:hemostatic effects in open heart operations. Ann ThoracSurg 1989;48:536–9.

12. Hardy JF, Belisle S, Couturier A, Robitaille D. Randomized,placebo-controlled, double-blind study of an ultra-low-dose aprotinin regimen in reoperative and/or complexcardiac operations. J Card Surg 1997;12:15–22.

13. Rossi M, Storti S, Martinelli L, et al. A pump-prime apro-tinin dose in cardiac surgery: appraisal of its effects on thehemostatic system. J Cardiothorac Vasc Anesth 1997;11:835–9.

14. Speekenbrink RG, Wildevuur CR, Sturk A, Eijsman L.Low-dose and high-dose aprotinin improve hemostasis incoronary operations. J Thorac Cardiovasc Surg 1996;112:523–30.

15. Speekenbrink RG, Vonk AB, Wildevuur CR, Eijsman L.Hemostatic efficacy of dipyridamole, tranexamic acid, andaprotinin in coronary bypass grafting. Ann Thorac Surg1995;59:438–42.

16. Dignan RJ, Law DW, Seah PW, et al. Ultra-low doseaprotinin decreases transfusion requirements and is costeffective in coronary operations. Ann Thorac Surg 2001;71:158–64.

17. Santamaria A, Mateo J, Oliver A, et al. The effect of twodifferent doses of aprotinin on hemostasis in cardiopulmo-nary bypass surgery: similar transfusion requirements andblood loss. Haematologica 2000;85:1277–84.

18. Misfeld M, Dubbert S, Eleftheriadis S, Siemens HJ, WagnerT, Sievers HH. Fibrinolysis-adjusted perioperative low-dose aprotinin reduces blood loss in bypass operations.Ann Thorac Surg 1998;66:792–9.

19. Tabuchi N, De Haan J, Boonstra PW, Huet RC, van OeverenW. Aprotinin effect on platelet function and clotting duringcardiopulmonary bypass. Eur J Cardiothorac Surg 1994;8:87–90.

20. Hardy JF, Belisle S, Dupont C, et al. Prophylactic tranex-amic acid and epsilon-aminocaproic acid for primary myo-cardial revascularization. Ann Thorac Surg 1998;65:371–6.

21. Brown RS, Thwaites BK, Mongan PD. Tranexamic acid iseffective in decreasing postoperative bleeding and transfu-sions in primary coronary artery bypass operations: adouble-blind, randomized, placebo-controlled trial. AnesthAnalg 1997;85:963–70.

22. Blauhut B, Harringer W, Bettelheim P, Doran JE, Spath P,Lundsgaard-Hansen P. Comparison of the effects of apro-tinin and tranexamic acid on blood loss and related vari-ables after cardiopulmonary bypass. J Thorac CardiovascSurg 1994;108:1083–91.

23. Menichetti A, Tritapepe L, Ruvolo G, et al. Changes in
coagulation patterns, blood loss and blood use after cardio-pulmonary bypass: aprotinin versus tranexamic acid versus
2

epsilon aminocaproic acid. J Cardiovasc Surg (Torino)1996;37:401–7.

24. Zabeeda D, Medalion B, Sverdlov M, et al. Tranexamic acidreduces bleeding and the need for blood transfusion inprimary myocardial revascularization. Ann Thorac Surg2002;74:733–8.

25. Andreasen JJ, Nielsen C. Prophylactic tranexamic acid inelective, primary coronary artery bypass surgery usingcardiopulmonary bypass. Eur J Cardiothorac Surg 2004;26:311–7.

26. Pleym H, Stenseth R, Wahba A, Bjella L, Karevold A, DaleO. Single-dose tranexamic acid reduces postoperativebleeding after coronary surgery in patients treated withaspirin until surgery. Anesth Analg 2003;96:923–8.

27. Moulton M, Creswell L, Mackey M, Cox J, Rosenbloom M.Reexploration for bleeding is a risk factor for adverseoutcomes after cardiac operations. J Thorac CardiovascSurg 1996:111:1037–46.

28. Fremes SE, Wong BI, Lee E, et al. Metaanalysis of prophy-lactic drug treatment in the prevention of postoperativebleeding. Ann Thorac Surg 1994;58:1580–8.

29. D’Ambra MN, Akins CW, Blackstone EH, et al. Aprotinin inprimary valve replacement and reconstruction: a multi-center, double-blind, placebo-controlled trial. J Thorac Car-diovasc Surg 1996;112:1081–9.

30. Brown JR, Birkmeyer NJ, O’Connor GT. Aprotinin in car-diac surgery. N Engl J Med 2006;354:1953–7 [author reply1953-7].

31. van der Linden J, Lindvall G, Sartipy U. Aprotinin de-creases postoperative bleeding and number of transfusionsin patients on clopidogrel undergoing coronary artery by-pass graft surgery: a double-blind, placebo-controlled, ran-domized clinical trial. Circulation 2005;112(Suppl 1):276–80.

32. Kreisler KR, Vance RA, Cruzzavala J, Mahnken JD. Hepa-rin-bonded cardiopulmonary bypass circuits reduce therate of red blood cell transfusion during elective coronaryartery bypass surgery. J Cardiothorac Vasc Anesth 2005;19:608–11.

33. Petitti DB. Meta-analysis, decision analysis, and cost-effectiveness analysis: methods for quantitative synthesis inmedicine. 2nd ed. New York: Oxford University Press,2000:115–6.

34. Raivio P, Suojaranta-Ylinen R, Kuitunen AH. Recombinantfactor VIIa in the treatment of postoperative hemorrhageafter cardiac surgery. Ann Thorac Surg 2005;80:66–71.

35. Greilich PE, Okada K, Latham P, Kumar RR, Jessen ME.Aprotinin but not epsilon-aminocaproic acid decreasesinterleukin-10 after cardiac surgery with extracorporealcirculation: randomized, double-blind, placebo-controlledstudy in patients receiving aprotinin and epsilon-aminocaproic acid. Circulation 2001;104(Suppl 1):265–9.

36. Casati V, Guzzon D, Oppizzi M, et al. Tranexamic acidcompared with high-dose aprotinin in primary electiveheart operations: effects on perioperative bleeding andallogeneic transfusions. J Thorac Cardiovasc Surg 2000;120:520–7.

37. Wong BI, McLean RF, Fremes SE, et al. Aprotinin andtranexamic acid for high transfusion risk cardiac surgery.Ann Thorac Surg 2000;69:808–16.

38. Nuttall GA, Oliver WC, Ereth MH, et al. Comparison ofblood-conservation strategies in cardiac surgery patients athigh risk for bleeding. Anesthesiology 2000;92:674–82.

39. Bernet F, Carrel T, Marbet G, Skarvan K, Stulz P. Reductionof blood loss and transfusion requirements after coronaryartery bypass grafting: similar efficacy of tranexamic acidand aprotinin in aspirin-treated patients. J Card Surg1999:14:92–7.

40. Casati V, Guzzon D, Oppizzi M, et al. Hemostatic effects ofaprotinin, tranexamic acid and epsilon-aminocaproic acidin primary cardiac surgery. Ann Thorac Surg 1999;68:2252–7.

41. Mongan PD, Brown RS, Thwaites BK. Tranexamic acid andaprotinin reduce postoperative bleeding and transfusions

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2


during primary coronary revascularization. Anesth Analg1998;87:258–65.

42. Dryden PJ, O’Connor JP, Jamieson WR, et al. Tranexamicacid reduces blood loss and transfusion in reoperativecardiac surgery [see comments]. Can J Anaesth 1997;44:934–41.

43. Bennett-Guerrero E, Sorohan JG, Canada AT, et al. Epsilon-aminocaproic acid plasma levels during cardiopulmonarybypass. Anesth Analg 1997;85:248–51.

44. Bennett-Guerrero E, Sorohan JG, Gurevich ML, et al. Cost-benefit and efficacy of aprotinin compared with epsilon-aminocaproic acid in patients having repeated cardiacoperations: a randomized, blinded clinical trial. Anesthesi-ology 1997;87:1373–80.

45. Lemmer JH Jr, Stanford W, Bonney SL, et al. Aprotinin forcoronary artery bypass grafting: effect on postoperativerenal function. Ann Thorac Surg 1995;59:132–6.

46. Feindt PR, Walcher S, Volkmer I, et al. Effects of high-doseaprotinin on renal function in aortocoronary bypass graft-ing. Ann Thorac Surg 1995;60:1076–80.

47. Kincaid EH, Ashburn DA, Hoyle JR, Reichert MG, Ham-mon JW, Kon ND. Does the combination of aprotinin andangiotensin-converting enzyme inhibitor cause renal fail-ure after cardiac surgery? Ann Thorac Surg 2005;80:1388–93.

48. Royston D, Chhatwani A. Safety aspects of aprotinin ther-apy in cardiac surgery patients. Expert Opin Drug Saf2006;5:539–52.

49. Beierlein W, Scheule AM, Dietrich W, Ziemer G. Fortyyears of clinical aprotinin use: a review of 124 hypersensi-tivity reactions. Ann Thorac Surg 2005;79:741–8.

50. Vlahakes GJ. The value of phase 4 clinical testing. N EnglJ Med 2006;354:413–5.

51. Hunter D. First, gather the data. N Engl J Med 2006;354:329–31.

52. Helm R, Krieger KH. Erythropoietin in cardiac surgery.New York: Springer, 1998.

53. Monk TG. Preoperative recombinant human erythropoietinin anemic surgical patients. Crit Care 2004;8(Suppl 2):45–8.

54. Goldberg MA. Perioperative epoetin alfa increases redblood cell mass and reduces exposure to transfusions:results of randomized clinical trials. Semin Hematol 1997;34(Suppl 2):41–7.

55. Hardy JF. Pharmacological strategies for blood conserva-tion in cardiac surgery: erythropoietin and antifibrinolytics.Can J Anaesth 2001;48(Suppl):24–31.

56. Alghamdi AA, Albanna MJ, Guru V, Brister SJ. Does the useof erythropoietin reduce the risk of exposure to allogeneicblood transfusion in cardiac surgery? A systematic reviewand meta-analysis. J Card Surg 2006;21:320–6.

57. Madi-Jebara SN, Sleilaty GS, Achouh PE, et al. Postopera-tive intravenous iron used alone or in combination withlow-dose erythropoietin is not effective for correction ofanemia after cardiac surgery. J Cardiothorac Vasc Anesth2004;18:59–63.

58. Osaka M, Fukuda I, Ohuchi H. [Aprotinin and recombinanthuman erythropoietin reduce the need for homologousblood transfusion in cardiac surgery.] Jpn J Thorac Cardio-vasc Surg 1998;46:846–53.

59. Yoshikawa Y, Niwaya K, Hasegawa J, et al. [Effect of bloodconservation in open-heart surgery: a comparison of 3different methods.] Kyobu Geka 1994;47:1059–62.

60. Price S, Pepper JR, Jaggar SI. Recombinant human eryth-ropoietin use in a critically ill Jehovah’s witness aftercardiac surgery. Anesth Analg 2005;101:325–7.

61. Sonzogni V, Crupi G, Poma R, et al. Erythropoietin therapyand preoperative autologous blood donation in childrenundergoing open heart surgery. Br J Anaesth 2001;87:429–34.

62. Shimpo H, Mizumoto T, Onoda K, Yuasa H, Yada I.Erythropoietin in pediatric cardiac surgery: clinical efficacy
and effective dose. Chest 1997;111:1565–70.
63. Sowade O, Warnke H, Scigalla P, et al. Avoidance ofallogeneic blood transfusions by treatment with epoetinbeta (recombinant human erythropoietin) in patients un-dergoing open-heart surgery. Blood 1997;89:411–8.

64. Rosengart TK, Helm RE, Klemperer J, Krieger KH, IsomOW. Combined aprotinin and erythropoietin use for bloodconservation: results with Jehovah’s Witnesses. Ann ThoracSurg 1994;58:1397–403.

65. Fullerton DA, Campbell DN, Whitman GJ. Use of humanrecombinant erythropoietin to correct severe preoperativeanemia. Ann Thorac Surg 1991;51:825–6.

66. Gaudiani VA, Mason HD. Preoperative erythropoietin inJehovah’s Witnesses who require cardiac procedures. AnnThorac Surg 1991;51:823–4.

67. Hoynck van Papendrecht MA, Jeekel H, Busch OR, Mar-quet RL. Efficacy of recombinant erythropoietin for stimu-lating erythropoiesis after blood loss and surgery. Anexperimental study in rats. Eur J Surg 1992;158:83–7.

68. Beholz S, Liu J, Thoelke R, Spiess C, Konertz W. Use ofdesmopressin and erythropoietin in an anaemic Jehovah’sWitness patient with severely impaired coagulation capac-ity undergoing stentless aortic valve replacement. Perfu-sion 2001;16:485–9.

69. Yamagishi I, Sakurada T, Abe T. Cardiac surgery using onlyautologous blood for a patient with hereditary spherocyto-sis: a case report. Ann Thorac Cardiovasc Surg 1998;4:294–7.

70. Spiess BD. Transfusion of blood products affects outcomein cardiac surgery. Semin Cardiothorac Vasc Anesth 2004;8:267–81.

71. Fisher JW. Pharmacologic modulation of erythropoietinproduction. Annu Rev Pharmacol Toxicol 1988;28:101–22.

72. Means RT Jr. Clinical application of recombinant erythro-poietin in the anemia of chronic disease. Hematol OncolClin North Am 1994;8:933–44.

73. Pleym H, Stenseth R, Wahba A, et al. Prophylactic treat-ment with desmopressin does not reduce postoperativebleeding after coronary surgery in patients treated withaspirin before surgery. Anesth Analg 2004;98:578–84; tableof contents.

74. Ozkisacik E, Islamoglu F, Posacioglu H, et al. Desmopressinusage in elective cardiac surgery. J Cardiovasc Surg(Torino) 2001;42:741–7.

75. Oliver WC Jr, Santrach PJ, Danielson GK, Nuttall GA,Schroeder DR, Ereth MH. Desmopressin does not reducebleeding and transfusion requirements in congenital heartoperations. Ann Thorac Surg 2000;70:1923–30.

76. Despotis GJ, Levine V, Saleem R, Spitznagel E, Joist JH. Useof point-of-care test in identification of patients who canbenefit from desmopressin during cardiac surgery: a ran-domised controlled trial. Lancet 1999;354:106–10.

77. Cattaneo M, Harris AS, Stromberg U, Mannucci PM. Theeffect of desmopressin on reducing blood loss in cardiacsurgery—a meta-analysis of double-blind, placebo-controlled trials. Thromb Haemost 1995;74:1064–70.

78. Horrow JC, Van Riper DF, Strong MD, Brodsky I, Parmet JL.Hemostatic effects of tranexamic acid and desmopressinduring cardiac surgery. Circulation 1991;84:2063–70.

79. Temeck BK, Bachenheimer LC, Katz NM, Coughlin SS,Wallace RB. Desmopressin acetate in cardiac surgery: adouble-blind, randomized study. South Med J 1994;87:611–5.

80. Reynolds LM, Nicolson SC, Jobes DR, et al. Desmopressindoes not decrease bleeding after cardiac operation in youngchildren. J Thorac Cardiovasc Surg 1993;106:954–8.

81. Sheridan DP, Card RT, Pinilla JC, et al. Use of desmopres-sin acetate to reduce blood transfusion requirements dur-ing cardiac surgery in patients with acetylsalicylic-acid-induced platelet dysfunction. Can J Surg 1994;37:33–6.

82. Dilthey G, Dietrich W, Spannagl M, Richter JA. Influence ofdesmopressin acetate on homologous blood requirements
in cardiac surgical patients pretreated with aspirin. J Car-diothorac Vasc Anesth 1993 1993;7:425–30.

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3


83. Marquez J, Koehler S, Strelec SR, et al. Repeated doseadministration of desmopressin acetate in uncomplicatedcardiac surgery: a prospective, blinded, randomized study.J Cardiothorac Vasc Anesth 1992;6:674–6.

84. Ansell J, Klassen V, Lew R, et al. Does desmopressin acetateprophylaxis reduce blood loss after valvular heart opera-tions? A randomized, double-blind study. J Thorac Cardio-vasc Surg 1992;104:117–23.

85. Kuitunen AH. Haemostatic responses to desmopressin ac-etate after primary coronary artery bypass surgery. AnnChir Gynaecol 1992;81:11–8.

86. Lazenby W, Russo I, Zadeh B, et al. Treatment with des-mopressin acetate in routine coronary artery bypass sur-gery to improve postoperative hemostasis. Circulation 1990;82(Suppl 4):413–9.

87. Salmenpera M, Kuitunen A, Hynynen M, Heinonen J.Hemodynamic responses to desmopressin acetate afterCABG: a double-blind trial. J Cardiothorac Vasc Anesth1991;5:146–9.

88. Reich DL, Hammerschlag BC, Rand JH, et al. Desmopressinacetate is a mild vasodilator that does not reduce blood lossin uncomplicated cardiac surgical procedures. J Cardiotho-rac Vasc Anesth 1991;5:142–5.

89. Mongan PD, Hosking MP. The role of desmopressin acetatein patients undergoing coronary artery bypass surgery. Acontrolled clinical trial with thromboelastographic riskstratification. Anesthesiology 1992;77:38–46.

90. Koscielny J, von Tempelhoff GF, Ziemer S, et al. A practicalconcept for preoperative management of patients withimpaired primary hemostasis. Clin Appl Thromb Hemost2004;10:155–66.

91. Chung JH, Gikakis N, Rao AK, Drake TA, Colman RW,Edmunds LH Jr. Pericardial blood activates the extrinsiccoagulation pathway during clinical cardiopulmonary by-pass. Circulation 1996;93:2014–8.

92. Hattori T, Khan MM, Colman RW, Edmunds LH Jr. Plasmatissue factor plus activated peripheral mononuclear cellsactivate factors VII and X in cardiac surgical wounds. J AmColl Cardiol 2005;46:707–13.

93. von Heymann C, Redlich U, Jain U, et al. Recombinantactivated factor VII for refractory bleeding after cardiacsurgery—a retrospective analysis of safety and efficacy. CritCare Med 2005;33:2241–6.

94. Diprose P, Herbertson MJ, O’Shaughnessy D, Gill RS.Activated recombinant factor VII after cardiopulmonarybypass reduces allogeneic transfusion in complex non-coronary cardiac surgery: randomized double-blind pla-cebo-controlled pilot study. Br J Anaesth 2005;95:596–602.

95. Wittenstein B, Ng C, Ravn H, Goldman A. Recombinantfactor VII for severe bleeding during extracorporeal mem-brane oxygenation following open heart surgery. PediatrCrit Care Med 2005;6:473–6.

96. Hyllner M, Houltz E, Jeppsson A. Recombinant activatedfactor VII in the management of life-threatening bleedingin cardiac surgery. Eur J Cardiothorac Surg 2005;28:254–8.

97. Kogan A, Berman M, Stein M, Vidne BA, Raanani E.Recombinant factor VIIa use in cardiac surgery—expanding the arsenal therapy for intractable bleeding?J Cardiovasc Surg (Torino) 2004;45:569–71.

98. Vanek T, Straka Z, Hrabak J, Jares M, Brucek PJ, Votava J.Use of recombinant activated factor VII in cardiac surgeryfor an effective treatment of severe intractable bleeding. JpnHeart J 2004;45:855–60.

99. Aggarwal A, Malkovska V, Catlett JP, Alcorn K. Recombi-nant activated factor VII (rFVIIa) as salvage treatment forintractable hemorrhage. Thromb J 2004;2:9.

00. Tobias JD, Simsic JM, Weinstein S, Schechter W, Kartha V,Michler R. Recombinant factor VIIa to control excessivebleeding following surgery for congenital heart disease inpediatric patients. J Intens Care Med 2004;19:270–3.

01. Pychynska-Pokorska M, Moll JJ, Krajewski W, Jarosik P.
The use of recombinant coagulation factor VIIa in uncon-trolled postoperative bleeding in children undergoing car-
diac surgery with cardiopulmonary bypass. Pediatr CritCare Med 2004;5:246–50.

02. Egan JR, Lammi A, Schell DN, Gillis J, Nunn GR. Recom-binant activated factor VII in paediatric cardiac surgery.Intens Care Med 2004;30:682–5.

03. Stratmann G, Russell IA, Merrick SH. Use of recombinantfactor VIIa as a rescue treatment for intractable bleedingfollowing repeat aortic arch repair. Ann Thorac Surg 2003;76:2094–7.

04. Al Douri M, Shafi T, Al Khudairi D, et al. Effect of theadministration of recombinant activated factor VII (rFVIIa;NovoSeven) in the management of severe uncontrolledbleeding in patients undergoing heart valve replacementsurgery. Blood Coagul Fibrinolysis 2000;11(Suppl 1):121–7.

05. Herbertson M. Recombinant activated factor VII in cardiacsurgery. Blood Coagul Fibrinolysis 2004;15(Suppl 1):31–2.

06. Dietrich W, Spannagl M. Caveat against the use of acti-vated recombinant factor VII for intractable bleeding incardiac surgery. Anesth Analg 2002;94:1369–70 [author re-ply 1370-1].

07. von Heymann C, Ziemer S, Kox WJ, Spies C. Caveat againstthe use of feiba in combination with recombinant factorVIIa. J Thorac Cardiovasc Surg 2003;126:1667–8.

08. Despotis G, Avidan M, Lublin DM. Off-label use of recom-binant factor VIIA concentrates after cardiac surgery. AnnThorac Surg 2005;80:3–5.

09. Goodnough LT. Experiences with recombinant human fac-tor VIIa in patients with thrombocytopenia. Semin Hematol2004;41(Suppl 1):25–9.

10. Ilabaca PA, Ochsner JL, Mills NL. Positive end-expiratorypressure in the management of the patient with a postop-erative bleeding heart. Ann Thorac Surg 1980;30:281–4.

11. Hoffman WS, Tomasello DN, MacVaugh H. Control ofpostcardiotomy bleeding with PEEP. Ann Thorac Surg1982;34:71–3.

12. Murphy DA, Finlayson DC, Craver JM, et al. Effect ofpositive end-expiratory pressure on excessive mediastinalbleeding after cardiac operations. A controlled study. J Tho-rac Cardiovasc Surg 1983;85:864–9.

13. Zurick AM, Urzua J, Ghattas M, Cosgrove DM, EstafanousFG, Greenstreet R. Failure of positive end-expiratory pres-sure to decrease postoperative bleeding after cardiac sur-gery. Ann Thorac Surg 1982;34:608–11.

14. Collier B, Kolff J, Devineni R, Gonzalez LS III. Prophylacticpositive end-expiratory pressure and reduction of postop-erative blood loss in open-heart surgery. Ann Thorac Surg2002;74:1191–4.

15. DeBois W, Krieger KH. The influence of oxygenator typeand priming volume on blood requirements. In: KriegerKH, Isom OW, eds. Blood conservation in cardiac surgery.New York: Springer, 1998:327–54.

16. Liddicoat JE, Bekassy SM, Beall AC Jr, Glaeser DH, De-Bakey ME. Membrane versus bubble oxygenator: clinicalcomparison. Ann Surg 1975;181:747–53.

17. Clark RE, Beauchamp RA, Magrath RA, Brooks JD, Fergu-son TB, Weldon CS. Comparison of bubble and membraneoxygenators in short and long perfusions. J Thorac Cardio-vasc Surg 1979;78:655–66.

18. Hicks GL, Zwart HH, DeWall RA. Membrane versus bub-ble oxygenators: a prospective study of 52 patients. ArchSurg 1979;114:1285–7.

19. Hessel EA II, Johnson DD, Ivey TD, Miller DW Jr. Mem-brane versus bubble oxygenator for cardiac operations. Aprospective randomized study. J Thorac Cardiovasc Surg1980;80:111–22.

20. Sade RM, Bartles DM, Dearing JP, Campbell LJ, LoadholtCB. A prospective randomized study of membrane versusbubble oxygenators in children. Ann Thorac Surg 1980;29:502–11.

21. Huber R, Mennicke C, Heitmann KU, Helling HJ, HuberPM. Membrane versus bubble oxygenator in hyperthermic
regional perfusion: a prospective randomized clinicalstudy. Eur Surg Res 1994;26:210–20.

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

33

3

3

3

3


22. Tabak C, Eugene J, Stemmer EA. Erythrocyte survivalfollowing extracorporeal circulation. A question of mem-brane versus bubble oxygenator. J Thorac Cardiovasc Surg1981;81:30–3.

23. Subramanian VA, Berger RL. Comparative evaluation of anew disposable rotating membrane oxygenator with bub-ble oxygenator. Ann Thorac Surg 1976;21:48–54.

24. Blauth CI, Smith PL, Arnold JV, Jagoe JR, Wootton R, TaylorKM. Influence of oxygenator type on the prevalence andextent of microembolic retinal ischemia during cardiopul-monary bypass. Assessment by digital image analysis.J Thorac Cardiovasc Surg 1990;99:61–9.

25. Zadeh BJ, Holazo R, Conlon CE, et al. A clinical evaluationof three modern blood oxygenators. Perfusion 1987;2:263–70.

26. Parker JL, Hackett JE, Clark D, Crane TN, Reed CC.Membrane versus bubble oxygenators: a clinical compari-son of postoperative blood loss. Cardiovasc Dis 1979;6:78–84.

27. Mitsumaru A, Yozu R, Matayoshi T, et al. Efficiency of an airfilter at the drainage site in a closed circuit with a centrif-ugal blood pump: an in vitro study. ASAIO J 2001;47:692–5.

28. Mehra AP, Akins A, Maisuria A, Glenville BE. Air handlingcharacteristics of five membrane oxygenators. Perfusion1994;9:357–62.

29. Pedersen TH, Videm V, Svennevig JL, et al. Extracorporealmembrane oxygenation using a centrifugal pump and aservo regulator to prevent negative inlet pressure. AnnThorac Surg 1997;63:1333–9.

30. Wahba A, Philip A, Bauer MF, Kaiser M, Aebert H, Birn-baum DE. The blood saving potential of vortex versus rollerpump with and without aprotinin. Perfusion 1995;10:333–41.

31. Scott DA, Silbert BS, Blyth C, O’Brien J, Santamaria J. Bloodloss in elective coronary artery surgery: a comparison ofcentrifugal versus roller pump heads during cardiopulmo-nary bypass. J Cardiothorac Vasc Anesth 2001;15:322–5.

32. Wheeldon DR, Bethune DW, Gill RD. Vortex pumping forroutine cardiac surgery: a comparative study. Perfusion1990;5:135–43.

33. Jakob H, Hafner G, Iversen S, et al. Reoperation and thecentrifugal pump? Eur J Cardiothorac Surg 1992;(Suppl1):59–63.

34. Onoda K, Kondo C, Mizumoto T, et al. Clinical experiencewith Nikkiso centrifugal pumps for extracorporeal circula-tion. Artif Organs 1994;18:706–10.

35. Yoshikai M, Hamada M, Takarabe K, Okazaki Y, Ito T.Clinical use of centrifugal pumps and the roller pump inopen heart surgery: a comparative evaluation. Artif Organs1996;20:704–6.

36. Takarabe K, Yoshikai M, Murayama J, Hamada M, Ito T.Clinical evaluation of the centrifugal pump in open heartsurgery: a comparative study of different pumps. ArtifOrgans 1997;21:760–2.

37. Misoph M, Babin-Ebell J, Schwender S. A comparativeevaluation of the effect of pump type and heparin-coatedsurfaces on platelets during cardiopulmonary bypass. Tho-rac Cardiovasc Surg 1997;45:302–6.

38. Morgan IS, Codispoti M, Sanger K, Mankad PS. Superiorityof centrifugal pump over roller pump in paediatric cardiacsurgery: prospective randomised trial. Eur J CardiothoracSurg 1998;13:526–32.

39. Salama A, Hugo F, Heinrich D, et al. Deposition of terminalC5b-9 complement complexes on erythrocytes and leuko-cytes during cardiopulmonary bypass. N Engl J Med 1988;318:408–14.

40. Moen O, Fosse E, Braten J, et al. Differences in bloodactivation related to roller/centrifugal pumps and heparin-coated/uncoated surfaces in a cardiopulmonary bypassmodel circuit. Perfusion 1996;11:113–23.

41. Linneweber J, Chow TW, Kawamura M, Moake JL, Nose Y.
In vitro comparison of blood pump induced platelet micro-
aggregates between a centrifugal and roller pump duringcardiopulmonary bypass. Int J Artif Organs 2002;25:549–55.

42. Bray B, Lane DA, Freyssinet JM, Pejler G, Lindahl U.Anti-thrombin activities of heparin. Effect of saccharidechain length on thrombin inhibition by heparin cofactor IIand by antithrombin. Biochem J 1989;262:225–32.

43. Choay J, Petitou M, Lormeau JC, Sinay P, Casu B, Gatti G.Structure-activity relationship in heparin: a synthetic pen-tasaccharide with high affinity for antithrombin III andeliciting high anti-factor Xa activity. Biochem Biophys ResCommun 1983;116:492–9.

44. Hirsh J, Dalen JE, Fuster V, Harker LB, Patrono C, Roth G.Aspirin and other platelet-active drugs. The relationshipamong dose, effectiveness, and side effects. Chest 1995;108(Suppl):247–57.

45. Olds RJ, Lane DA, Beresford CH, Abildgaard U, HughesPM, Thein SL. A recurrent deletion in the antithrombingene, AT106-108(-6 bp), identified by DNA heteroduplexdetection. Genomics 1993;16:298–9.

46. Lindahl AK, Abildgaard U, Staalesen R. The anticoagulanteffect in heparinized blood and plasma resulting frominteractions with extrinsic pathway inhibitor. Thromb Res1991;64:155–68.

47. Despotis GJ, Hogue CW Jr, Santoro SA, Joist JH, BarnesPW, Lappas DG. Effect of heparin on whole blood activatedpartial thromboplastin time using a portable, whole bloodcoagulation monitor. Crit Care Med 1995;23:1674–9.

48. Despotis GJ, Joist JH, Goodnough LT. Monitoring of hemo-stasis in cardiac surgical patients: impact of point-of-caretesting on blood loss and transfusion outcomes. Clin Chem1997;43:1684–96.

49. Barzu T, van Rijn JL, Petitou M, Tobelem G, Caen JP.Heparin degradation in the endothelial cells. Thromb Res1987;47:601–9.

50. Wu HF, Lundblad RL, Church FC. Neutralization of hepa-rin activity by neutrophil lactoferrin. Blood 1995;85:421–8.

51. Lane DA, Pejler G, Flynn AM, Thompson EA, Lindahl U.Neutralization of heparin-related saccharides by histidine-rich glycoprotein and platelet factor 4. J Biol Chem 251986;261:3980–6.

52. Kelton JG, Hirsh J. Bleeding associated with antithrombotictherapy. Semin Hematol 1980;17:259–91.

53. Teoh KH, Young E, Bradley CA, Hirsh J. Heparin bindingproteins. Contribution to heparin rebound after cardiopul-monary bypass. Circulation 1993;88:II420–425.

54. Preissner KT, Wassmuth R, Muller-Berghaus G. Physico-chemical characterization of human S-protein and its func-tion in the blood coagulation system. Biochem J 1985;231:349–55.

55. Sobel M, McNeill PM, Carlson PL, et al. Heparin inhibitionof von Willebrand factor-dependent platelet function invitro and in vivo. J Clin Invest 1991;87:1787–93.

56. Hull RD, Raskob GE, Hirsh J, et al. Continuous intravenousheparin compared with intermittent subcutaneous heparinin the initial treatment of proximal-vein thrombosis. N EnglJ Med 1986;315:1109–14.

57. Hirsh J. Heparin. N Engl J Med 1991;324:1565–74.58. Thomas DP, Barrowcliffe TW, Johnson EA. The influence of

tissue source, salt and molecular weight and heparin activ-ity. Scand J Haematol 1980;36(Suppl):40–9.

59. Verska JJ. Control of heparinization by activated clottingtime during bypass with improved postoperative hemosta-sis. Ann Thorac Surg 1977;24:170–3.

60. Babka R, Colby C, El-Etr A, Pifarre R. Monitoring ofintraoperative heparinization and blood loss following car-diopulmonary bypass surgery. J Thorac Cardiovasc Surg1977;73:780–2.

61. Roth JA, Cukingnan RA, Scott CR. Use of activated coagu-lation time to monitor heparin during cardiac surgery. AnnThorac Surg 1979;28:69–72.

62. Papaconstantinou C, Radegran K. Use of the activated
coagulation time in cardiac surgery. Effects on heparin-

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3


protamine dosages and bleeding. Scand J ThoracCardiovasc Surg 1981;15:213–5.

63. Jumean HG, Sudah F. Monitoring of anticoagulant therapyduring open-heart surgery in children with congenitalheart disease. Acta Haematol 1983;70:392–5.

64. Niinikoski J, Laato M, Laaksonen V, Jalonen J, Inberg MV.Use of activated clotting time to monitor anticoagulationduring cardiac surgery. Scand J Thorac Cardiovasc Surg1984;18:57–61.

65. Lefemine AA, Lewis M. Activated clotting time for controlof anticoagulation during surgery. Am Surg 1985;51:274–8.

66. Preiss DU, Schmidt-Bleibtreu H, Berguson P, Metz G.Blood transfusion requirements in coronary artery surgerywith and without the activated clotting time (ACT) tech-nique. Klin Wochenschr 1985;63:252–6.

67. Metz S, Keats AS. Low activated coagulation time duringcardiopulmonary bypass does not increase postoperativebleeding. Ann Thorac Surg 1990;49:440–4.

68. Akl BF, Vargas GM, Neal J, Robillard J, Kelly P. Clinicalexperience with the activated clotting time for the control ofheparin and protamine therapy during cardiopulmonarybypass. J Thorac Cardiovasc Surg 1980;79:97–102.

69. Dearing JP, Bartles DM, Stroud MR, et al. Activated clottingtimes versus protocol anticoagulation management. J ExpCorp Tech 1998;15:17–9.

70. Jobes DR, Schwartz AJ, Ellison N, Andrews R, Ruffini RA,Ruffini JJ. Monitoring heparin anticoagulation and its neu-tralization. Ann Thorac Surg 1981;31:161–6.

71. Gravlee GP, Rogers AT, Dudas LM, et al. Heparin manage-ment protocol for cardiopulmonary bypass influences post-operative heparin rebound but not bleeding. Anesthesiol-ogy 1992;76:393–401.

72. Sakurada T, Kikuchi Y, Koushima R, Nakashima S, HachiroY, Kagaya H. [Clinical evaluation of heparin concentrationand activated clotting time monitoring (HEPCON HMS)system.] Nippon Kyobu Geka Gakkai Zasshi 1997;45:836–40.

73. Despotis GJ, Hogue CW Jr, Santoro SA, Joist JH, BarnesPW, Lappas DG. Effect of heparin on whole blood activatedpartial thromboplastin time using a portable, whole bloodcoagulation monitor. Crit Care Med 1995;23:1674–9.

74. Gravlee GP, Haddon WS, Rothberger HK, et al. Heparindosing and monitoring for cardiopulmonary bypass. Acomparison of techniques with measurement of subclinicalplasma coagulation. J Thorac Cardiovasc Surg 1990;99:518–27.

75. Boldt J, Schindler E, Osmer C, Wittstock M, Stertmann WA,Hempelmann G. Influence of different anticoagulation reg-imens on platelet function during cardiac surgery. Br JAnaesth 1994;73:639–44.

76. Okita Y, Takamoto S, Ando M, et al. Coagulation andfibrinolysis system in aortic surgery under deep hypother-mic circulatory arrest with aprotinin: the importance ofadequate heparinization. Circulation 1997;96(Suppl 2):376–81.

77. Despotis GJ, Filos KS, Zoys TN, Hogue CW Jr, Spitznagel E,Lappas DG. Factors associated with excessive postoperativeblood loss and hemostatic transfusion requirements: amultivariate analysis in cardiac surgical patients. AnesthAnalg 1996;82:13–21.

78. Jobes DR, Aitken GL, Shaffer GW. Increased accuracy andprecision of heparin and protamine dosing reduces bloodloss and transfusion in patients undergoing primary car-diac operations. J Thorac Cardiovasc Surg 1995;110:36–45.

79. Despotis GJ, Joist JH, Hogue CW Jr, et al. The impact ofheparin concentration and activated clotting time monitor-ing on blood conservation. A prospective, randomizedevaluation in patients undergoing cardiac operation. J Tho-rac Cardiovasc Surg 1995;110:46–54.

80. Yasuda M, Hasunuma Y, Adachi H, et al. Expression andfunction of fibronectin binding integrins on rat mast cells.
Int Immunol 1995;7:251–8.
81. Weitz JI, Hudoba M, Massel D, Maraganore J, Hirsh J.Clot-bound thrombin is protected from inhibition by hep-arin-antithrombin III but is susceptible to inactivation byantithrombin III-independent inhibitors. J Clin Invest 1990;86:385–91.

82. Berger RL, Ramaswamy K, Ryan TJ. Reduced protaminedosage for heparin neutralization in open-heart operations.Circulation 1968;37(Suppl 2):154–7.

83. Guffin AV, Dunbar RW, Kaplan JA, Bland JW Jr. Successfuluse of a reduced dose of protamine after cardiopulmonarybypass. Anesth Analg 1976;55:110–3.

84. Moriau M, Masure R, Hurlet A, et al. Haemostasis disor-ders in open heart surgery with extracorporeal circulation.Importance of the platelet function and the heparin neu-tralization. Vox Sang 1977;32:41–51.

85. Ottesen S, Stormorken H, Hatteland K. The value of acti-vated coagulation time in monitoring heparin therapy dur-ing extracorporeal circulation. Scand J Thorac CardiovascSurg 1984;18:123–8.

86. Keeler JF, Shah MV, Hansbro SD. Protamine—the need todetermine the dose. Comparison of a simple protaminetitration method with an empirical dose regimen for rever-sal of heparinisation following cardiopulmonary bypass.Anaesthesia 1991;46:925–8.

87. Shore-Lesserson L, Reich DL, DePerio M. Heparin andprotamine titration do not improve haemostasis in cardiacsurgical patients. Can J Anaesth 1998;45:10–8.

88. Moen O, Fosse E, Brockmeier V, et al. Disparity in bloodactivation by two different heparin-coated cardiopulmo-nary bypass systems. Ann Thorac Surg 1995;60:1317–23.

89. Fukutomi M, Kobayashi S, Niwaya K, Hamada Y, KitamuraS. Changes in platelet, granulocyte, and complement acti-vation during cardiopulmonary bypass using heparin-coated equipment. Artif Organs 1996;20:767–76.

90. Baksaas ST, Videm V, Pedersen T, et al. Comparison ofthree oxygenator-coated and one total-circuit-coated extra-corporeal devices. Perfusion 1999;14:119–27.

91. Ovrum E, Tangen G, Oystese R, Ringdal MA, Istad R.Comparison of two heparin-coated extracorporeal circuitswith reduced systemic anticoagulation in routine coronaryartery bypass operations. J Thorac Cardiovasc Surg 2001;121:324–30.

92. Svenmarker S, Sandstrom E, Karlsson T, et al. Neurologicaland general outcome in low-risk coronary artery bypasspatients using heparin coated circuits. Eur J CardiothoracSurg 2001;19:47–53.

93. Flom-Halvorsen HI, Ovrum E, Brosstad F, Tangen G, Ring-dal M, Oystese R. Effects of two differently heparin-coatedextracorporeal circuits on markers for brain and myocardialdysfunction. Perfusion 2002;17:339–45.

94. Palatianos GM, Foroulis CN, Vassili MI, et al. A prospec-tive, double-blind study on the efficacy of the biolinesurface-heparinized extracorporeal perfusion circuit. AnnThorac Surg 2003;76:129–35.

95. Sinci V, Kalaycioglu S, Gunaydin S, et al. Evaluation ofheparin-coated circuits with full heparin dose strategy. AnnThorac Cardiovasc Surg 1999;5:156–63.

96. Aldea GS, Doursounian M, O’Gara P, et al. Heparin-bonded circuits with a reduced anticoagulation protocol inprimary CABG: a prospective, randomized study. AnnThorac Surg 1996;62:410–8.

97. Kuitunen AH, Heikkila LJ, Salmenpera MT. Cardiopulmo-nary bypass with heparin-coated circuits and reduced sys-temic anticoagulation. Ann Thorac Surg 1997;63:438–44.

98. von Segesser LK, Weiss BM, Pasic M, Garcia E, Turina MI.Risk and benefit of low systemic heparinization duringopen heart operations. Ann Thorac Surg 1994;58:391–8.

99. von Segesser LK, Weiss BM, Garcia E, von Felten A, TurinaMI. Reduction and elimination of systemic heparinizationduring cardiopulmonary bypass. J Thorac Cardiovasc Surg
1992;103:790–9.

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4


00. Mahoney CB, Lemole GM. Transfusion after coronary ar-tery bypass surgery: the impact of heparin-bonded circuits.Eur J Cardiothorac Surg 1999;16:206–10.

01. Weiss BM, von Segesser LK, Turina MI, Seifert B, Pasch T.Perioperative course and recovery after heparin-coatedcardiopulmonary bypass: low-dose versus high-dose hep-arin management. J Cardiothorac Vasc Anesth 1996;10:464–70.

02. McCarthy PM, Yared JP, Foster RC, Ogella DA, Borsh JA,Cosgrove DM III. A prospective randomized trial of DurafloII heparin-coated circuits in cardiac reoperations. AnnThorac Surg 1999;67:1268–73.

03. Boonstra PW, Gu YJ, Akkerman C, Haan J, Huyzen R, vanOeveren W. Heparin coating of an extracorporeal circuitpartly improves hemostasis after cardiopulmonary bypass.J Thorac Cardiovasc Surg 1994;107:289–92.

04. Saenz A, Larranaga G, Alvarez L, et al. Heparin-coatedcircuit in coronary surgery. A clinical study. Eur J Cardio-thorac Surg 1996;10:48–53.

05. Wagner WR, Johnson PC, Thompson KA, Marrone GC.Heparin-coated cardiopulmonary bypass circuits: hemo-static alterations and postoperative blood loss. Ann ThoracSurg 1994;58:734–41.

06. Belboul A, Akbar O, Lofgren C, Jungbeck M, Storm C,Roberts A. Improved blood cellular biocompatibility withheparin coated circuits during cardiopulmonary bypass.J Cardiovasc Surg (Torino) 2000;41:357–62.

07. Baufreton C, Le Besnerais P, Jansen P, Mazzucotelli JP,Wildevuur CR, Loisance DY. Clinical outcome after coro-nary surgery with heparin-coated extracorporeal circuitsfor cardiopulmonary bypass. Perfusion 1996;11:437–43.

08. Ereth MH, Nuttall GA, Clarke SH, et al. Biocompatibility oftrillium biopassive surface-coated oxygenator versus un-coated oxygenator during cardiopulmonary bypass. J Car-diothorac Vasc Anesth 2001;15:545–50;discussion 539-41.

09. Stammers AH, Christensen KA, Lynch J, Zavadil DP,Deptula JJ, Sydzyik RT. Quantitative evaluation of heparin-coated versus non-heparin-coated bypass circuits duringcardiopulmonary bypass. J Extra Corpor Technol 1999;31:135–41.

10. Svenmarker S, Haggmark S, Jansson E, et al. Use ofheparin-bonded circuits in cardiopulmonary bypass im-proves clinical outcome. Scand Cardiovasc J 2002;36:241–6.

11. van der Kamp KW, van Oeveren W. Contact, coagulationand platelet interaction with heparin treated equipmentduring heart surgery. Int J Artif Organs 1993;16:836–42.

12. Thelin S, Bagge L, Hultman J, Borowiec J, Nilsson L,Thorelius J. Heparin-coated cardiopulmonary bypass cir-cuits reduce blood cell trauma. Experiments in the pig. EurJ Cardiothorac Surg 1991;5:486–91.

13. Palatianos GM, Dewanjee MK, Smith W, et al. Plateletpreservation during cardiopulmonary bypass with iloprostand Duraflo-II heparin-coated surfaces. ASAIO Trans 1991;37:620–2.

14. Svennevig JL, Geiran OR, Karlsen H, et al. Complementactivation during extracorporeal circulation. In vitro com-parison of Duraflo II heparin-coated and uncoated oxygen-ator circuits. J Thorac Cardiovasc Surg 1993;106:466–72.

15. Mollnes TE, Videm V, Gotze O, Harboe M, Oppermann M.Formation of C5a during cardiopulmonary bypass: inhibi-tion by precoating with heparin. Ann Thorac Surg 1991;52:92–7.

16. Videm V, Svennevig JL, Fosse E, Semb G, Osterud A,Mollnes TE. Reduced complement activation with heparin-coated oxygenator and tubings in coronary bypass opera-tions. J Thorac Cardiovasc Surg 1992;103:806–13.

17. Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, HuyzenRJ, Wildevuur CR. Heparin-coated circuits reduce the in-flammatory response to cardiopulmonary bypass. AnnThorac Surg 1993;55:917–22.

18. Fosse E, Thelin S, Svennevig JL, et al. Duraflo II coating of
cardiopulmonary bypass circuits reduces complement ac-tivation, but does not affect the release of granulocyte
enzymes: a European multicentre study. Eur J CardiothoracSurg 1997;11:320–7.

19. Borowiec JW, Bylock A, van der Linden J, Thelin S. Heparincoating reduces blood cell adhesion to arterial filters duringcoronary bypass: a clinical study. Ann Thorac Surg 1993;55:1540–5.

20. Ranucci M, Cirri S, Conti D, et al. Beneficial effects ofDuraflo II heparin-coated circuits on postperfusion lungdysfunction. Ann Thorac Surg 1996;61:76–81.

21. Redmond JM, Gillinov AM, Stuart RS, et al. Heparin-coatedbypass circuits reduce pulmonary injury. Ann Thorac Surg1993;56:474–9.

22. Young JA, Kisker CT, Doty DB. Adequate anticoagulationduring cardiopulmonary bypass determined by activatedclotting time and the appearance of fibrin monomer. AnnThorac Surg 1978;26:231–40.

23. Kestin AS, Valeri CR, Khuri SF, et al. The platelet functiondefect of cardiopulmonary bypass. Blood 1993;82:107–17.

24. Cardoso PF, Yamazaki F, Keshavjee S, et al. A reevaluationof heparin requirements for cardiopulmonary bypass.J Thorac Cardiovasc Surg 1991;101:153–60.

25. Eisses MJ, Velan T, Aldea GS, Chandler WL. Strategies toreduce hemostatic activation during cardiopulmonary by-pass. Thromb Res 2006;117:689–703.

26. Olsson C, Siegbahn A, Halden E, Nilsson B, Venge P,Thelin S. No benefit of reduced heparinization in thoracicaortic operation with heparin-coated bypass circuits. AnnThorac Surg 2000;69:743–9.

27. O’Gara PJ, Treanor PR, Lilly KJ, et al. Technique for routineuse of heparin bonded circuits with a reduced anticoagu-lation protocol. J Extra Corpor Technol 2000;32:207–13.

28. Lilly KJ, O’Gara PJ, Treanor PR, et al. Heparin-bondedcircuits without a cardiotomy: a description of a minimallyinvasive technique of cardiopulmonary bypass. Perfusion2002;17:95–7.

29. Paparella D, Al Radi OO, Meng QH, Venner T, Teoh K,Young E. The effects of high-dose heparin on inflammatoryand coagulation parameters following cardiopulmonarybypass. Blood Coagul Fibrinolysis 2005;16:323–8.

30. Mullen JC, Bentley MJ, Gelfand ET, et al. Coronary arterybypass surgery with heparin-coated perfusion circuits andlow-dose heparinization. Can J Surg 2002;45:166–72.

31. Mirow N, Brinkmann T, Minami K, Tenderich G, KleesiekK, Korfer R. Heparin-coated extracorporeal circulation withfull and low dose heparinization: comparison of thrombinrelated coagulatory effects. Artif Organs 2001;25:480–5.

32. Suehiro S, Shibata T, Sasaki Y, et al. Use of heparin-coatedcardiopulmonary bypass circuit with low-dose heparin re-duces postoperative bleeding. Ann Thorac Cardiovasc Surg1999;5:225–9.

33. Mirow N, Brinkmann T, Minami K, et al. Low dose systemicheparinization combined with heparin-coated extracorpo-real circulation. Effects related to platelets. J CardiovascSurg (Torino) 2001;42:579–85.

34. Nakajima T, Kawazoe K, Ishibashi K, et al. Reduction ofheparin dose is not beneficial to platelet function. AnnThorac Surg 2000;70:186–90.

35. Turner-Gomes SO, Nitschmann EP, Norman GR, AndrewME, Williams WG. Effect of heparin loading during con-genital heart operation on thrombin generation and bloodloss. Ann Thorac Surg 1997;63:482–8.

36. Aldea GS, O’Gara P, Shapira OM, et al. Effect of anticoag-ulation protocol on outcome in patients undergoing CABGwith heparin-bonded cardiopulmonary bypass circuits.Ann Thorac Surg 1998;65:425–33.

37. Groom RC. High or low hematocrits during cardiopulmo-nary bypass for patients undergoing coronary artery bypassgraft surgery? An evidence-based approach to the question.Perfusion 2002;17:99–102.

38. Helm R, Krieger KH. Development and application of amultimodality blood conservation program. In: Krieger KH,
Isom OW, eds. Blood conservation in cardiac surgery. NewYork: Springer, 1998:631–70.

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4


39. Ferraris VA, Ferraris SP, Singh A, et al. The platelet throm-bin receptor and postoperative bleeding. Ann Thorac Surg1998;65:352–8.

40. von Segesser LK, Garcia E, Turina MI. Low-dose heparinversus full-dose heparin with high-dose aprotinin duringcardiopulmonary bypass. A preliminary report. Tex HeartInst J 1993;20:28–32.

41. Okita Y, Takamoto S, Ando M, Morota T, Yamaki F,Kawashima Y. Is use of aprotinin safe with deep hypother-mic circulatory arrest in aortic surgery? Investigations onblood coagulation. Circulation 1996;94(Suppl 2):177–81.

42. Seeburger J, Hoffmann J, Wendel HP, Ziemer G, Aebert H.Gene expression changes in leukocytes during cardiopul-monary bypass are dependent on circuit coating. Circula-tion 2005;112(Suppl 1):224–8.

43. Svenmarker S, Engstrom KG. The inflammatory responseto recycled pericardial suction blood and the influence ofcell-saving. Scand Cardiovasc J 2003;37:158–64.

44. Ishida T, Nakano K, Nakatani H, et al. [Relation betweenintraoperative salvaged blood transfusion and postopera-tive infection after cardiac surgery.] Kyobu Geka 2002;55:763–7.

45. Jewell AE, Akowuah EF, Suvarna SK, Braidley P, Hopkin-son D, Cooper G. A prospective randomised comparison ofcardiotomy suction and cell saver for recycling shed bloodduring cardiac surgery. Eur J Cardiothorac Surg 2003;23:633–6.

46. Perttila J, Leino L, Poyhonen M, Salo M. Leucocyte contentin blood processed by autotransfusion devices duringopen-heart surgery. Acta Anaesthesiol Scand 1995;39:445–8.

47. Amand T, Pincemail J, Blaffart F, Larbuisson R, Limet R,Defraigne JO. Levels of inflammatory markers in the bloodprocessed by autotransfusion devices during cardiac sur-gery associated with cardiopulmonary bypass circuit. Per-fusion 2002;17:117–23.

48. Deleuze P, Intrator L, Liou A, Contremoulins I, Cachera JP,Loisance DY. Complement activation and use of a cell saverin cardiopulmonary bypass. ASAIO Trans 1990;36:M179–81.

49. Moran JM, Babka R, Silberman S, et al. Immediate centrif-ugation of oxygenator contents after cardiopulmonary by-pass. Role in maximum blood conservation. J Thorac Car-diovasc Surg 1978;76:510–7.

50. Keeling MM, Gray LA Jr, Brink MA, Hillerich VK, Bland KI.Intraoperative autotransfusion. Experience in 725 consecu-tive cases. Ann Surg 1983;197:536–41.

51. Tempe DK, Banerjee A, Virmani S, et al. Comparison of theeffects of a cell saver and low-dose aprotinin on blood lossand homologous blood usage in patients undergoing valvesurgery. J Cardiothorac Vasc Anesth 2001;15:326–30.

52. Lee J, Ikeda S, Johnston MF. Efficacy of intraoperative bloodsalvage during coronary artery bypass grafting. MinervaCardioangiol 1997;45:395–400.

53. McCarthy PM, Popovsky MA, Schaff HV, et al. Effect ofblood conservation efforts in cardiac operations at the MayoClinic. Mayo Clin Proc 1988;63:225–9.

54. Cordell AR, Lavender SW. An appraisal of blood salvagetechniques in vascular and cardiac operations. Ann ThoracSurg 1981;31:421–5.

55. Winton TL, Charrette EJ, Salerno TA. The cell saver duringcardiac surgery: does it save? Ann Thorac Surg 1982;33:379–81.

56. Mayer ED, Welsch M, Tanzeem A, et al. Reduction ofpostoperative donor blood requirement by use of the cellseparator. Scand J Thorac Cardiovasc Surg 1985;19:165–71.

57. Hall RI, Schweiger IM, Finlayson DC. The benefit of theHemonetics cell saver apparatus during cardiac surgery.Can J Anaesth 1990;37:618–23.

58. Parrot D, Lancon JP, Merle JP, et al. Blood salvage in cardiacsurgery. J Cardiothorac Vasc Anesth 1991;5:454–6.

59. Popovsky MA, Devine PA, Taswell HF. Intraoperative au-tologous transfusion. Mayo Clin Proc 1985;60:125–34.

60. Solomon MD, Rutledge ML, Kane LE, Yawn DH. Costcomparison of intraoperative autologous versus homolo-gous transfusion. Transfusion 1988;28:379–82.

61. Breyer RH, Engelman RM, Rousou JA, Lemeshow S. Bloodconservation for myocardial revascularization. Is it costeffective? J Thorac Cardiovasc Surg 1987;93:512–22.

62. Boldt J, von Bormann B, Kling D, Jacobi M, Moosdorf R,Hempelmann G. Preoperative plasmapheresis in patientsundergoing cardiac surgery procedures. Anesthesiology1990;72:282–8.

63. Boldt J, Zickmann B, Ballesteros M, Oehmke S, StertmannF, Hempelmann G. Influence of acute preoperative plasma-pheresis on platelet function in cardiac surgery. J Cardio-thorac Vasc Anesth 1993;7:4–9.

64. Harke H, Tanger D, Furst-Denzer S, Paoachrysanthou C,Bernhard A. [Effect of a preoperative separation of plateletson the postoperative blood loss subsequent to extracorpo-real circulation in open heart surgery (author’s transla-tion).] Anaesthesist 1977;26:64–71.

65. Ferrari M, Zia S, Valbonesi M, et al. A new technique forhemodilution, preparation of autologous platelet-richplasma and intraoperative blood salvage in cardiac surgery.Int J Artif Organs 1987;10:47–50.

66. Giordano GF, Rivers SL, Chung GK, et al. Autologousplatelet-rich plasma in cardiac surgery: effect on intraoper-ative and postoperative transfusion requirements. AnnThorac Surg 1988;46:416–9.

67. Giordano GF Sr, Giordano GF Jr, Rivers SL, et al. Determi-nants of homologous blood usage utilizing autologousplatelet-rich plasma in cardiac operations. Ann ThoracSurg 1989;47:897–902.

68. Jones JW, McCoy TA, Rawitscher RE, Lindsley DA. Effectsof intraoperative plasmapheresis on blood loss in cardiacsurgery. Ann Thorac Surg 1990;49:585–90.

69. Jones JW, Rawitscher RE, McLean TR, Beall AC Jr, ThornbyJI. Benefit from combining blood conservation measures incardiac operations. Ann Thorac Surg 1991;51:541–6.

70. Davis PH, Hachinski VC. The cardiac factor in stroke. CurrOpin Neurol Neurosurg 1992;5:39–43.

71. Ferraris VA, Berry WR, Klingman RR. Comparison of bloodreinfusion techniques used during coronary artery bypassgrafting. Ann Thorac Surg 1993;56:433–40.

72. Shore-Lesserson L, Reich DL, DePerio M, Silvay G. Autol-ogous platelet-rich plasmapheresis: risk versus benefit inrepeat cardiac operations. Anesth Analg 1995;81:229–35.

73. Tobe CE, Vocelka C, Sepulvada R, et al. Infusion of autol-ogous platelet rich plasma does not reduce blood loss andproduct use after coronary artery bypass. A prospective,randomized, blinded study. J Thorac Cardiovasc Surg 1993;105:1007–14.

74. Ereth MH, Oliver WC Jr, Beynen FM, et al. Autologousplatelet-rich plasma does not reduce transfusion of homol-ogous blood products in patients undergoing repeat valvu-lar surgery. Anesthesiology 1993;79:540–547; discussion527A.

75. Wong CA, Franklin ML, Wade LD. Coagulation tests, bloodloss, and transfusion requirements in platelet-rich plasma-pheresed versus nonpheresed cardiac surgery patients.Anesth Analg 1994;78:29–36.

76. Menges T, Welters I, Wagner RM, Boldt J, Dapper F,Hempelmann G. The influence of acute preoperative plas-mapheresis on coagulation tests, fibrinolysis, blood lossand transfusion requirements in cardiac surgery. Eur J Car-diothorac Surg 1997;11:557–63.

77. Armellin G, Sorbara C, Bonato R, Pittarello D, Dal Cero P,Giron G. Intraoperative plasmapheresis in cardiac surgery.J Cardiothorac Vasc Anesth 1997;11:13–7.

78. Menges T, Wagner RM, Welters I, Ruwoldt R, Boldt J,Hempelmann G. The role of the prote in C-thrombomodulin system and fibrinolysis during cardiovas-cular surgery: influence of acute preoperative plasma-
pheresis. J Cardiothorac Vasc Anesth 1996;10:482–9.

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

4

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5


79. Wajon P, Gibson J, Calcroft R, Hughes C, Thrift B. Intraop-erative plateletpheresis and autologous platelet gel do notreduce chest tube drainage or allogeneic blood transfusionafter reoperative coronary artery bypass graft. AnesthAnalg 2001;93:536–42.

80. Boey SK, Ong BC, Dhara SS. Preoperative plateletpheresisdoes not reduce blood loss during cardiac surgery. Can JAnaesth 1993;40:844–50.

81. Rubens FD, Fergusson D, Wells PS, Huang M, McGowanJL, Laupacis A. Platelet-rich plasmapheresis in cardiacsurgery: a meta-analysis of the effect on transfusion re-quirements. J Thorac Cardiovasc Surg 1998;116:641–7.

82. Harding SA, Shakoor MA, Grindon AJ. Platelet support forcardiopulmonary bypass surgery. J Thorac Cardiovasc Surg1975;70:350–3.

83. Simon TL, Akl BF, Murphy W. Controlled trial of routineadministration of platelet concentrates in cardiopulmonarybypass surgery. Ann Thorac Surg 1984;37:359–64.

84. Mohr R, Martinowitz U, Lavee J, Amroch D, Ramot B, GoorDA. The hemostatic effect of transfusing fresh whole bloodversus platelet concentrates after cardiac operations. J Tho-rac Cardiovasc Surg 1988;96:530–4.

85. Lavee J, Martinowitz U, Mohr R, et al. The effect oftransfusion of fresh whole blood versus platelet concen-trates after cardiac operations. A scanning electron micro-scope study of platelet aggregation on extracellular matrix.J Thorac Cardiovasc Surg 1989;97:204–12.

86. Chen YF, Tsai WC, Lin CC, et al. Leukocyte depletionattenuates expression of neutrophil adhesion moleculesduring cardiopulmonary bypass in human beings. J ThoracCardiovasc Surg 2002;123:218–24.

87. Chen YF, Tsai WC, Lin CC, et al. Effect of leukocytedepletion on endothelial cell activation and transendothe-lial migration of leukocytes during cardiopulmonary by-pass. Ann Thorac Surg 2004;78:634–43.

88. Karaiskos TE, Palatianos GM, Triantafillou CD, et al. Clin-ical effectiveness of leukocyte filtration during cardiopul-monary bypass in patients with chronic obstructive pulmo-nary disease. Ann Thorac Surg 2004;78:1339–44.

89. Palatianos GM, Balentine G, Papadakis EG, et al. Neutro-phil depletion reduces myocardial reperfusion morbidity.Ann Thorac Surg 2004;77:956–61.

90. Olivencia-Yurvati AH, Wallace WE, Wallace N, et al. Intra-operative treatment strategy to reduce the incidence ofpostcardiopulmonary bypass atrial fibrillation. Perfusion2002;17(Suppl):35–9.

91. Rimpilainen J, Pokela M, Kiviluoma K, et al. Leukocytefiltration improves brain protection after a prolonged pe-riod of hypothermic circulatory arrest: a study in a chronicporcine model. J Thorac Cardiovasc Surg 2000;120:1131–41.

92. Efstathiou A, Vlachveis M, Tsonis G, Asteri T, Psarakis A,Fessatidis IT. Does leukodepletion during elective cardiacsurgery really influence the overall clinical outcome? J Car-diovasc Surg (Torino) 2003;44:197–204.

93. Martens S, Matheis G, Wimmer-Greinecker G, et al. Hep-arin coating of the extracorporeal circuit combined withleukocyte filtration reduces coagulation activity, blood lossand blood product substitution. Int J Artif Organs 2001;24:484–8.

94. Stefanou DC, Gourlay T, Asimakopoulos G, Taylor KM.Leucodepletion during cardiopulmonary bypass reducesblood transfusion and crystalloid requirements. Perfusion2001;16:51–8.

95. Gott JP, Cooper WA, Schmidt FE Jr, et al. Modifying risk forextracorporeal circulation: trial of four antiinflammatorystrategies. Ann Thorac Surg 1998;66:747–54.

96. Ilmakunnas M, Pesonen EJ, Ahonen J, Ramo J, Siitonen S,Repo H. Activation of neutrophils and monocytes by aleukocyte-depleting filter used throughout cardiopulmo-nary bypass. J Thorac Cardiovasc Surg 2005;129:851–9.

97. Scholz M, Simon A, Matheis G, et al. Leukocyte filtration
fails to limit functional neutrophil activity during cardiacsurgery. Inflamm Res 2002;51:363–8.
98. Fromes Y, Gaillard D, Ponzio O, et al. Reduction of theinflammatory response following coronary bypass graftingwith total minimal extracorporeal circulation. Eur J Cardio-thorac Surg 2002;22:527–33.

99. Wiesenack C, Liebold A, Philipp A, et al. Four years’experience with a miniaturized extracorporeal circulationsystem and its influence on clinical outcome. Artif Organs2004;28:1082–8.

00. Remadi JP, Rakotoarivello Z, Marticho P, et al. Aortic valvereplacement with the minimal extracorporeal circulation(Jostra MECC System) versus standard cardiopulmonarybypass: a randomized prospective trial. J Thorac CardiovascSurg 2004;128:436–41.

01. Remadi JP, Rakotoarivelo Z, Marticho P, Benamar A. Pro-spective randomized study comparing coronary artery by-pass grafting with the new mini-extracorporeal circulationJostra system or with a standard cardiopulmonary bypass.Am Heart J 2006;151:198.

02. Abdel-Rahman U, Ozaslan F, Risteski PS, et al. Initialexperience with a minimized extracorporeal bypass system:is there a clinical benefit? Ann Thorac Surg 2005;80:238–43.

03. Abdel-Rahman U, Martens S, Risteski P,et al. The use ofminimized extracorporeal circulation system has a benefi-cial effect on hemostasis—a randomized clinical study.Heart Surg Forum 2006;9:E543–8.

04. Nollert G, Schwabenland I, Maktav D, et al. Miniaturizedcardiopulmonary bypass in coronary artery bypass surgery:marginal impact on inflammation and coagulation but lossof safety margins. Ann Thorac Surg 2005;80:2326–32.

05. Hohn L, Schweizer A, Licker M, Morel DR. Absence ofbeneficial effect of acute normovolemic hemodilution com-bined with aprotinin on allogeneic blood transfusion re-quirements in cardiac surgery. Anesthesiology 2002;96:276–82.

06. Jamnicki M, Kocian R, van der Linden P, Zaugg M, SpahnDR. Acute normovolemic hemodilution: physiology, limi-tations, and clinical use. J Cardiothorac Vasc Anesth 2003;17:747–54.

07. Reents W, Babin-Ebell J, Misoph MR, Schwarzkopf A, ElertO. Influence of different autotransfusion devices on thequality of salvaged blood. Ann Thorac Surg 1999;68:58–62.

08. Helm RE, Klemperer JD, Rosengart TK, et al. Intraoperativeautologous blood donation preserves red cell mass but doesnot decrease postoperative bleeding. Ann Thorac Surg1996;62:1431–41.

09. Kochamba GS, Pfeffer TA, Sintek CF, Khonsari S. Intraop-erative autotransfusion reduces blood loss after cardiopul-monary bypass. Ann Thorac Surg 1996;61:900–3.

10. Kahraman S, Altunkaya H, Celebioglu B, Kanbak M, Pasao-glu I, Erdem K. The effect of acute normovolemic hemodi-lution on homologous blood requirements and total esti-mated red blood cell volume lost. Acta Anaesthesiol Scand1997;41:614–7.

11. Ramnath AN, Naber HR, de Boer A, Leusink JA. No benefitof intraoperative whole blood sequestration and autotrans-fusion during coronary artery bypass grafting: results of arandomized clinical trial. J Thorac Cardiovasc Surg 2003;125:1432–7.

12. Casati V, Speziali G, D’Alessandro C, et al. Intraoperativelow-volume acute normovolemic hemodilution in adultopen-heart surgery. Anesthesiology 2002;97:367–73.

13. Weiskopf RB. Hemodilution and candles. Anesthesiology2002;97:773–5.

14. Weiskopf RB. Efficacy of acute normovolemic hemodilutionassessed as a function of fraction of blood volume lost.Anesthesiology 2001;94:439–46.

15. Matot I, Scheinin O, Jurim O, Eid A. Effectiveness of acutenormovolemic hemodilution to minimize allogeneic bloodtransfusion in major liver resections. Anesthesiology 2002;97:794–800.

16. Bryson GL, Laupacis A, Wells GA. Does acute normovol-
emic hemodilution reduce perioperative allogeneic trans-

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5


fusion? A meta-analysis. The International Study ofPerioperative Transfusion. Anesth Analg 1998;86:9–15.

17. Or TH, Yang MW, Fan WL, Chan KH, Lee TY. Acutenormovolemic hemodilution in coronary artery bypassgraft surgery. Ma Zui Xue Za Zhi 1991;29:586–91.

18. Petry AF, Jost J, Sievers H. Reduction of homologous bloodrequirements by blood-pooling at the onset of cardiopul-monary bypass. J Thorac Cardiovasc Surg 1994;107:1210–4.

19. Monk TG, Goodnough LT, Brecher ME, Colberg JW, And-riole GL, Catalona WJ. A prospective randomized compar-ison of three blood conservation strategies for radicalprostatectomy. Anesthesiology 1999;91:24–33.

20. Spahn DR, Waschke KF, Standl T, et al. Use of perflubronemulsion to decrease allogeneic blood transfusion in high-blood-loss non-cardiac surgery: results of a Europeanphase 3 study. Anesthesiology 2002;97:1338–49.

21. Levy JH, Goodnough LT, Greilich PE, et al. Polymerizedbovine hemoglobin solution as a replacement for allogeneicred blood cell transfusion after cardiac surgery: results of arandomized, double-blind trial. J Thorac Cardiovasc Surg2002;124:35–42.

22. Cheng DC, Mazer CD, Martineau R, et al. A phase IIdose-response study of hemoglobin raffimer (Hemolink) inelective coronary artery bypass surgery. J Thorac Cardio-vasc Surg 2004;127:79–86.

23. Hill SE, Gottschalk LI, Grichnik K. Safety and preliminaryefficacy of hemoglobin raffimer for patients undergoingcoronary artery bypass surgery. J Cardiothorac Vasc Anesth2002;16:695–702.

24. Carmichael FJ. Recent developments in hemoglobin-basedoxygen carriers—an update on clinical trials. TransfusApher Sci 2001;24:17–21.

25. Tyson GS, Sladen RN, Spainhour V, Savitt MA, FergusonTB Jr, Wolfe WG. Blood conservation in cardiac surgery.Preliminary results with an institutional commitment. AnnSurg 1989;209:736–42.

26. Scott WJ, Rode R, Castlemain B, et al. Efficacy, complica-tions, and cost of a comprehensive blood conservationprogram for cardiac operations. J Thorac Cardiovasc Surg1992;103:1001–7.

27. Monk TG, Goodnough LT. Blood conservation strategies tominimize allogeneic blood use in urologic surgery. Am JSurg 1995;170(Suppl):69–73.

28. Scott WJ, Kessler R, Wernly JA. Blood conservation incardiac surgery. Ann Thorac Surg 1990;50:843–51.

29. Goodnough LT, Soegiarso RW, Geha AS. Blood lost andblood transfused in coronary artery bypass graft operationas implications for blood transfusion and blood conserva-tion strategies. Surg Gynecol Obstet 1993;177:345–51.

30. Helm RE, Rosengart TK, Gomez M, et al. Comprehensivemultimodality blood conservation: 100 consecutive CABGoperations without transfusion. Ann Thorac Surg 1998;65:125–36.

31. Boonstra PW, van Imhoff GW, Eysman L, et al. Reducedplatelet activation and improved hemostasis after con-trolled cardiotomy suction during clinical membrane oxy-genator perfusions. J Thorac Cardiovasc Surg 1985;89:900–6.

32. Brooker RF, Brown WR, Moody DM, et al. Cardiotomysuction: a major source of brain lipid emboli during cardio-pulmonary bypass. Ann Thorac Surg 1998;65:1651–5.

33. Sakert T, Gil W, Rosenberg I, et al. Cell saver efficacy forroutine coronary artery bypass surgery. Perfusion 1996;11:71–7.

34. Eichert I, Isgro F, Kiessling AH, Saggau W. Cell saver,ultrafiltration and direct transfusion: comparative study ofthree blood processing techniques. Thorac Cardiovasc Surg2001;49:149–52.

35. Kiessling AH, Khalil M, Assaf O, Isgro F, Kretz KU, SaggauW. Blood-air interface during cardiopulmonary bypass.Asian Cardiovasc Thorac Ann 2004;12:198–201.

36. Merville C, Charlet P, Zerr C, Bricard H. [Effectiveness ofthe Cell Saver and salvage from ultrafiltered extracorporeal

circulation in heart surgery.] Ann Fr Anesth Reanim1991;10:548–53.

37. Schonberger JP, van Oeveren W, Bredee JJ, Everts PA, deHaan J, Wildevuur CR. Systemic blood activation duringand after autotransfusion. Ann Thorac Surg 1994;57:1256–62.

38. Nataf P, Jault F, Vaissier E, et al. [Systematic peroperativeuse of an autotransfusion system in heart surgery. Aprospective and comparative study in 283 patients.] ArchMal Coeur Vaiss 1989;82:1869–73.

39. Westerberg M, Bengtsson A, Jeppsson A. Coronary surgerywithout cardiotomy suction and autotransfusion reducesthe postoperative systemic inflammatory response. AnnThorac Surg 2004;78:54–9.

40. de Haan J, Boonstra PW, Monnink SH, Ebels T, vanOeveren W. Retransfusion of suctioned blood during car-diopulmonary bypass impairs hemostasis. Ann ThoracSurg 1995;59:901–7.

41. Walpoth BH, Volken U, Pfaffli T, et al. [Quantitative andqualitative analysis of various blood-preserving measuresin heart surgery.] Schweiz Med Wochenschr 1990;120:817–22.

42. Okies JE, Goodnight SH, Litchford B, Connell RS, Starr A.Effects of infusion of cardiotomy suction blood duringextracorporeal circulation for coronary artery bypass sur-gery. J Thorac Cardiovasc Surg 1977;74:440–4.

43. Dietrich W, Barankay A, Dilthey G, Mitto HP, Richter JA.Reduction of blood utilization during myocardial revascu-larization. J Thorac Cardiovasc Surg 1989;97:213–9.

44. Daane CR, Golab HD, Meeder JH, Wijers MJ, Bogers AJ.Processing and transfusion of residual cardiopulmonarybypass volume: effects on haemostasis, complement activa-tion, postoperative blood loss and transfusion volume.Perfusion 2003;18:115–21.

45. Dial S, Nguyen D, Menzies D. Autotransfusion of shedmediastinal blood: a risk factor for mediastinitis after car-diac surgery? Results of a cluster investigation. Chest2003;124:1847–51.

46. Martin J, Robitaille D, Perrault LP, et al. Reinfusion ofmediastinal blood after heart surgery. J Thorac CardiovascSurg 2000;120:499–504.

47. Dalrymple-Hay MJ, Pack L, Deakin CD, et al. Autotransfu-sion of washed shed mediastinal fluid decreases the re-quirement for autologous blood transfusion following car-diac surgery: a prospective randomized trial. EurJ Cardiothorac Surg 1999;15:830–4.

48. Body SC, Birmingham J, Parks R, et al. Safety and efficacy ofshed mediastinal blood transfusion after cardiac surgery: amulticenter observational study. Multicenter Study of Peri-operative Ischemia Research Group. J Cardiothorac VascAnesth 1999;13:410–6.

49. Komiya T, Ban K, Yamazaki K, Date O, Nakamura T,Kanzaki Y. [Blood conservation effect and safety of shedmediastinal blood autotransfusion after cardiac surgery.]Jpn J Thorac Cardiovasc Surg 1998;46:961–5.

50. Vertrees RA, Conti VR, Lick SD, Zwischenberger JB, Mc-Daniel LB, Shulman G. Adverse effects of postoperativeinfusion of shed mediastinal blood. Ann Thorac Surg 1996;62:717–23.

51. de Varennes B, Nguyen D, Denis F, Ergina P, Latter D,Morin JE. Reinfusion of mediastinal blood in CABG pa-tients: impact on homologous transfusions and rate ofre-exploration. J Card Surg 1996;11:387–95.

52. Morris JJ, Tan YS. Autotransfusion: is there a benefit in acurrent practice of aggressive blood conservation? AnnThorac Surg 1994;58:502–8.

53. Ward HB, Smith RR, Landis KP, Nemzek TG, Dalmasso AP,Swaim WR. Prospective, randomized trial of autotransfu-sion after routine cardiac operations. Ann Thorac Surg1993;56:137–41.

54. Roberts SR, Early GL, Brown B, Hannah H III, McDonald
HL. Autotransfusion of unwashed mediastinal shed bloodfails to decrease banked blood requirements in patients

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5

5


undergoing aortocoronary bypass surgery. Am J Surg1991;162:477–80.

55. Ovrum E, Holen EA, Lindstein Ringdal MA. Elective coro-nary artery bypass surgery without homologous bloodtransfusion. Early results with an inexpensive blood con-servation program. Scand J Thorac Cardiovasc Surg 1991;25:13–8.

56. Eng J, Kay PH, Murday AJ, et al. Postoperative autologoustransfusion in cardiac surgery. A prospective, randomisedstudy. Eur J Cardiothorac Surg 1990;4:595–600.

57. Page R, Russell GN, Fox MA, Fabri BM, Lewis I, Williets T.Hard-shell cardiotomy reservoir for reinfusion of shedmediastinal blood. Ann Thorac Surg 1989;48:514–7.

58. Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotrans-fusion following cardiac operations: a randomized, pro-spective study. Ann Thorac Surg 1979;27:500–7.

59. Schaff HV, Hauer JM, Bell WR, et al. Autotransfusion ofshed mediastinal blood after cardiac surgery: a prospectivestudy. J Thorac Cardiovasc Surg 1978;75:632–41.

60. Szczerkowski P. [The effect of reinfusion of mediastinalblood drained after coronary surgery on coagulation, fibri-nolysis and volume of blood products transfused.] AnnAcad Med Stetin 2003;49:247–58.

61. Schmidt H, Mortensen PE, Folsgaard SL, Jensen EA. Auto-transfusion after coronary artery bypass grafting halves thenumber of patients needing blood transfusion. Ann ThoracSurg 1996;61:1177–81.

62. Sherman MM, Dobnik DB, Dennis RC, Berger RL. Autolo-gous blood transfusion during cardiopulmonary bypass.Chest 1976;70:592–5.

63. Laub GW, Dharan M, Riebman JB, et al. The impact ofintraoperative autotransfusion on cardiac surgery. A pro-spective randomized double-blind study. Chest 1993;104:686–9.

64. Wahl GW, Feins RH, Alfieres G, Bixby K. Reinfusion ofshed blood after coronary operation causes elevation ofcardiac enzyme levels. Ann Thorac Surg 1992;53:625–7.

65. Spanier T, Tector K, Schwartz G, et al. Endotoxin in pooledpericardial blood contributes to the systemic inflammatoryresponse during cardiac surgery. Perfusion 2000;15:427–31.

66. Philippou H, Adami A, Davidson SJ, Pepper JR, Burman JF,Lane DA. Tissue factor is rapidly elevated in plasma col-lected from the pericardial cavity during cardiopulmonarybypass. Thromb Haemost 2000;84:124–8.

67. Tabuchi N, de Haan J, Boonstra PW, van Oeveren W.Activation of fibrinolysis in the pericardial cavity duringcardiopulmonary bypass. J Thorac Cardiovasc Surg 1993;106:828–33.

68. Appelblad M, Engstrom G. Fat contamination of pericardialsuction blood and its influence on in vitro capillary-poreflow properties in patients undergoing routine coronaryartery bypass grafting. J Thorac Cardiovasc Surg 2002;124:377–86.

69. Moody DM, Brown WR, Challa VR, Stump DA, ReboussinDM, Legault C. Brain microemboli associated with cardio-pulmonary bypass: a histologic and magnetic resonanceimaging study. Ann Thorac Surg 1995;59:1304–7.

70. Borowiec JW, Bozdayi M, Jaramillo A, Nilsson L, Venge P,Henze A. Influence of two blood conservation techniques(cardiotomy reservoir versus cell-saver) on biocompatibil-ity of the heparin coated cardiopulmonary bypass circuitduring coronary revascularization surgery. J Card Surg1997;12:190–7.

71. Yoda M, Nonoyama M, Shimakura T. [Safety and efficacy ofautologous blood donation before elective off-pump coro-nary artery bypass grafting.] Kyobu Geka 2004;57:367–9.

72. Yoda M, Nonoyama M, Shimakura T. Autologous blooddonation before elective off-pump coronary artery bypassgrafting. Surg Today 2004;34:21–3.

73. Jovin IS, Stelzig G, Strelitz JC, et al. Post-operative course ofcoronary artery bypass surgery patients who pre-donate
autologous blood. Int J Cardiol 2003;92:235–9.
74. Maze Y, Yada M, Katayama Y, Kimura M, Shomura S.[Preoperative autologous blood donation in patients under-going open heart surgery.] Kyobu Geka 2003;56:765–7.

75. Kiyama H, Ohshima N, Imazeki T, Yamada T. Autologousblood donation with recombinant human erythropoietin inanemic patients. Ann Thorac Surg 1999;68:1652–6.

76. Lewis CE, Hiratzka LF, Woods SE, Hendy MP, Engel AM.Autologous blood transfusion in elective cardiac valveoperations. J Card Surg 2005;20:513–8.

77. Dietrich W, Thuermel K, Heyde S, Busley R, Berger K.Autologous blood donation in cardiac surgery: reduction ofallogeneic blood transfusion and cost-effectiveness. J Car-diothorac Vasc Anesth 2005;19:589–96.

78. Kiyama H, Ohshima N, Imazeki T, Yamada T. [Autologousblood donation with recombinant human erythropoietinprior to elective cardiac surgery in anemic patients.] KyobuGeka 1998;51:741–4.

79. Kulier AH, Gombotz H, Fuchs G, Vuckovic U, Metzler H.Subcutaneous recombinant human erythropoietin and au-tologous blood donation before coronary artery bypasssurgery. Anesth Analg 1993;76:102–6.

80. Schved JF. Preoperative autologous blood donation: a ther-apy that needs to be scientifically evaluated. Transfus ClinBiol 2005;12:365–9.

81. Walpoth BH, Eggensperger N, Hauser SP, et al. Effects ofunprocessed and processed cardiopulmonary bypass bloodretransfused into patients after cardiac surgery. Int J ArtifOrgans 1999;22:210–6.

82. Dalrymple-Hay MJ, Dawkins S, Pack L, et al. Autotransfu-sion decreases blood usage following cardiac surgery—aprospective randomized trial. Cardiovasc Surg 2001;9:184–7.

83. Murphy GS, Szokol JW, Nitsun M, et al. The failure ofretrograde autologous priming of the cardiopulmonarybypass circuit to reduce blood use after cardiac surgicalprocedures. Anesth Analg 2004;98:1201–7.

84. Montenegro LM, Greeley WJ. Pro: the use of modifiedultrafiltration during pediatric cardiac surgery is a benefit.J Cardiothorac Vasc Anesth 1998;12:480–2.

85. Gaynor JW. The effect of modified ultrafiltration on thepostoperative course in patients with congenital heart dis-ease. Semin Thorac Cardiovasc Surg Pediatr Card SurgAnnu 2003;6:128–39.

86. Huang H, Yao T, Wang W, et al. Continuous ultrafiltrationattenuates the pulmonary injury that follows open heartsurgery with cardiopulmonary bypass. Ann Thorac Surg2003;76:136–40.

87. Berdat PA, Eichenberger E, Ebell J, et al. Elimination ofproinflammatory cytokines in pediatric cardiac surgery:analysis of ultrafiltration method and filter type. J ThoracCardiovasc Surg 2004;127:1688–96.

88. Kiziltepe U, Uysalel A, Corapcioglu T, Dalva K, Akan H,Akalin H. Effects of combined conventional and modifiedultrafiltration in adult patients. Ann Thorac Surg 2001;71:684–93.

89. Blanchard N, Toque Y, Trojette F, Quintard JM, BenammarA, Montravers P. Hemodynamic and echocardiographiceffects of hemofiltration performed during cardiopulmo-nary bypass. J Cardiothorac Vasc Anesth 2000;14:393–8.

90. Boga M, Islamoglu F, Badak I, et al. The effects of modifiedhemofiltration on inflammatory mediators and cardiac per-formance in coronary artery bypass grafting. Perfusion2000;15:143–50.

91. Grunenfelder J, Zund G, Schoeberlein A, et al. Modifiedultrafiltration lowers adhesion molecule and cytokine lev-els after cardiopulmonary bypass without clinical relevancein adults. Eur J Cardiothorac Surg 2000;17:77–83.

92. Tassani P, Richter JA, Eising GP, et al. Influence of com-bined zero-balanced and modified ultrafiltration on thesystemic inflammatory response during coronary artery
bypass grafting. J Cardiothorac Vasc Anesth 1999;13:285–91.

5

5

5

5

5

5

5

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6


93. Babka RM, Petress J, Briggs R, Helsal R, Mack J. Conven-tional haemofiltration during routine coronary bypass sur-gery. Perfusion 1997;12:187–92.

94. Raman JS, Hata M, Bellomo R, Kohchi K, Cheung HL,Buxton BF. Hemofiltration during cardiopulmonary bypassfor high risk adult cardiac surgery. Int J Artif Organs2003;26:753–7.

95. Khan NE, De Souza A, Mister R, et al. A randomizedcomparison of off-pump and on-pump multivessel coro-nary-artery bypass surgery. N Engl J Med 2004;350:21–8.

96. van Dijk D, Nierich AP, Jansen EW, et al. Early outcomeafter off-pump versus on-pump coronary bypass surgery:results from a randomized study. Circulation 2001;104:1761–6.

97. Angelini GD, Taylor FC, Reeves BC, Ascione R. Early andmidterm outcome after off-pump and on-pump surgery inBeating Heart Against Cardioplegic Arrest Studies (BHA-CAS 1 and 2): a pooled analysis of two randomised con-trolled trials. Lancet 2002;359:1194–9.

98. Puskas JD, Williams WH, Duke PG, et al. Off-pump coro-nary artery bypass grafting provides complete revascular-ization with reduced myocardial injury, transfusion re-quirements, and length of stay: a prospective randomizedcomparison of two hundred unselected patients undergo-ing off-pump versus conventional coronary artery bypassgrafting. J Thorac Cardiovasc Surg 2003;125:797–808.

99. Czerny M, Baumer H, Kilo J, et al. Complete revasculariza-tion in coronary artery bypass grafting with and withoutcardiopulmonary bypass. Ann Thorac Surg 2001;71:165–9.

00. Cheng DC, Bainbridge D, Martin JE, Novick RJ. Doesoff-pump coronary artery bypass reduce mortality, morbid-ity, and resource utilization when compared with conven-tional coronary artery bypass? A meta-analysis of random-ized trials. Anesthesiology 2005;102:188–203.

01. Jones RH. Intraoperative crossover: the well-kept surgicalsecret to apparent surgical success. J Am Coll Cardiol2005;45:1529–31.

02. Porte RJ, Leebeek FW. Pharmacological strategies to de-crease transfusion requirements in patients undergoingsurgery. Drugs 2002;62:2193–211.

03. Jakob H, Campbell CD, Qiu ZK, Pick R, Replogle RL.Evaluation of fibrin sealing for cardiovascular surgery.Circulation 1984;70:I138–46.

04. Rousou J, Levitsky S, Gonzalez-Lavin L, et al. Randomizedclinical trial of fibrin sealant in patients undergoing rester-notomy or reoperation after cardiac operations. A multi-center study. J Thorac Cardiovasc Surg 1989;97:194–203.

05. Israels SJ, Israels ED. Development of antibodies to bovineand human factor V in two children after exposure totopical bovine thrombin. Am J Pediatr Hematol Oncol1994;16:249–54.

06. Banninger H, Hardegger T, Tobler A, et al. Fibrin glue insurgery: frequent development of inhibitors of bovinethrombin and human factor V. Br J Haematol 1993;85:528–32.

07. Berruyer M, Amiral J, Ffrench P, et al. Immunization bybovine thrombin used with fibrin glue during cardiovascu-lar operations. Development of thrombin and factor Vinhibitors. J Thorac Cardiovasc Surg 1993;105:892–7.

08. Muntean W, Zenz W, Finding K, Zobel G, Beitzke A.Inhibitor to factor V after exposure to fibrin sealant duringcardiac surgery in a two-year-old child. Acta Paediatr1994;83:84–7.

09. Ortel TL, Mercer MC, Thames EH, Moore KD, Lawson JH.Immunologic impact and clinical outcomes after surgicalexposure to bovine thrombin. Ann Surg 2001;233:88–96.

10. Schlag G, Seifert J. Fibrin sealant, aprotinin, and immuneresponse in children undergoing operations for congenitalheart disease. J Thorac Cardiovasc Surg 1998;116:1082–3.

11. Dietrich W, Spath P, Ebell A, Richter JA. Prevalence ofanaphylactic reactions to aprotinin: analysis of two hun-
dred forty-eight reexposures to aprotinin in heart opera-tions. J Thorac Cardiovasc Surg 1997;113:194–201.
12. Bachet J, Guilmet D. The use of biological glue in aorticsurgery. Cardiol Clin 1999;17:779–96; ix-x.

13. Seguin JR, Picard E, Frapier JM, Chaptal PA. Aortic valverepair with fibrin glue for type A acute aortic dissection.Ann Thorac Surg 1994;58:304–7.

14. Zogno M, La Canna G, Ceconi C, et al. Postinfarction leftventricular free wall rupture: original management andsurgical technique. J Card Surg 1991;6:396–9.

15. Nakajima T, Kawazoe K, Izumoto H, Kataoka T, Kazui T.Effective use of fibrin glue for acute aortic dissection. AnnThorac Surg 2005;79:1793–4.

16. Birmingham B. TEE diagnosis of mechanical AVR dysfunc-tion associated with biological glue. Anesth Analg 2001;93:1627–8.

17. Boris WJ, Gu J, McGrath LB. Effectiveness of fibrin glue inthe reduction of postoperative intrapericardial adhesions.J Invest Surg 1996;9:327–33.

18. Bingley JA, Gardner MA, Stafford EG, et al. Late complica-tions of tissue glues in aortic surgery. Ann Thorac Surg2000;69:1764–8.

19. Kazui T, Washiyama N, Bashar AH, et al. Role of biologicglue repair of proximal aortic dissection in the develop-ment of early and midterm redissection of the aortic root.Ann Thorac Surg 2001;72:509–14.

20. Hvass U, Chatel D, Frikha I, Pansard Y, Depoix JP, JulliardJM. Left ventricular free wall rupture. Long-term resultswith a pericardial patch and fibrin glue repair. Eur J Car-diothorac Surg 1995;9:75–6.

21. Coletti G, Torracca L, Zogno M, et al. Surgical managementof left ventricular free wall rupture after acute myocardialinfarction. Cardiovasc Surg 1995;3:181–6.

22. Otaki M, Kitamura N. Left ventricular rupture followingmitral valve replacement. Chest 1993;104:1431–5.

23. Basu S, Marini CP, Bauman FG, et al. Comparative study ofbiological glues: cryoprecipitate glue, two-component fi-brin sealant, and “French” glue. Ann Thorac Surg 1995;60:1255–62.

24. Nikolsky E, Mehran R, Halkin A, et al. Vascular complica-tions associated with arteriotomy closure devices in pa-tients undergoing percutaneous coronary procedures: ameta-analysis. J Am Coll Cardiol 15 2004;44:1200–9.

25. Tron C, Koning R, Eltchaninoff H, et al. A randomizedcomparison of a percutaneous suture device versus manualcompression for femoral artery hemostasis after PTCA.J Interv Cardiol 2003;16:217–21.

26. Crocker CH, Cragun KT, Timimi FK, et al. Immediateambulation following diagnostic coronary angiographyprocedures utilizing a vascular closure device (The Closer).J Invas Cardiol 2002;14:728–32.

27. Chamberlin JR, Lardi AB, McKeever LS, et al. Use ofvascular sealing devices (VasoSeal and Perclose) versusassisted manual compression (Femostop) in transcathetercoronary interventions requiring abciximab (ReoPro).Catheter Cardiovasc Interv 1999;47:143–8.

28. Muller DW, Shamir KJ, Ellis SG, Topol EJ. Peripheralvascular complications after conventional and complexpercutaneous coronary interventional procedures. Am JCardiol 1992;69:63–8.

29. Eggebrecht H, Haude M, Woertgen U, et al. Systematic useof a collagen-based vascular closure device immediatelyafter cardiac catheterization procedures in 1,317 consecu-tive patients. Catheter Cardiovasc Interv 2002;57:486–95.

30. Sanborn TA, Gibbs HH, Brinker JA, Knopf WD, KosinskiEJ, Roubin GS. A multicenter randomized trial comparing apercutaneous collagen hemostasis device with conven-tional manual compression after diagnostic angiographyand angioplasty. J Am Coll Cardiol 1993;22:1273–9.

31. Juran NB, Rouse CL, Smith DD, O’Brien MA, DeLuca SA,Sigmon K. Nursing interventions to decrease bleeding atthe femoral access site after percutaneous coronary inter-vention. SANDBAG Nursing Coordinators. Standards of
Angioplasty Nursing Techniques to Diminish BleedingAround the Groin. Am J Crit Care 1999;8:303–13.

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6


32. Scher KS. Unplanned reoperation for bleeding. Am Surg1996;62:52–5.

33. Hall TS, Sines JC, Spotnitz AJ. Hemorrhage related reex-ploration following open heart surgery: the impact of pre-operative and post-operative coagulation testing. Cardio-vasc Surg 2002;10:146–53.

34. Tsujinaka T, Itoh T, Uemura Y, Sakon M, Kambayashi J,Mori T. Clinical application of a new in vitro bleeding timedevice on surgical patients. Jpn J Surg 1988;18:430–7.

35. Barber A, Green D, Galluzzo T, Ts’ao CH. The bleedingtime as a preoperative screening test. Am J Med 1985;78:761–4.

36. De Caterina R, Lanza M, Manca G, Strata GB, Maffei S,Salvatore L. Bleeding time and bleeding: an analysis of therelationship of the bleeding time test with parameters ofsurgical bleeding. Blood 1994;84:3363–70.

37. Burns ER, Billett HH, Frater RW, Sisto DA. The preopera-tive bleeding time as a predictor of postoperative hemor-rhage after cardiopulmonary bypass. J Thorac CardiovascSurg 1986;92:310–2.

38. Gewirtz AS, Miller ML, Keys TF. The clinical usefulness ofthe preoperative bleeding time. Arch Pathol Lab Med1996;120:353–6.

39. Stiles KP, Hill C, LeBrun CJ, Reinmuth B, Yuan CM, AbbottKC. The impact of bleeding times on major complicationrates after percutaneous real-time ultrasound-guided renalbiopsies. J Nephrol 2001;14:275–9.

40. Peterson P, Hayes TE, Arkin CF, et al. The preoperativebleeding time test lacks clinical benefit: College of Ameri-can Pathologists’ and American Society of Clinical Pathol-ogists’ position article. Arch Surg 1998;133:134–9.

41. Ferraris VA, Ferraris SP, Singh A. Operative outcome andhospital cost. J Thorac Cardiovasc Surg 1998;115:593–602;discussion 602–3.

42. Corwin H, Parsonnet K, Gettinger A. RBC transfusion inthe ICU. Is there a reason? Chest 1995;108:767–71.

43. Foulke G, Harlow D. Effective measures for reducing bloodloss from diagnostic laboratory tests in intensive care unitpatients. Crit Care Med 1989;17:1143–5.

44. Smoller B, Kruskall M, Horowitz G. Reducing adult phle-botomy blood loss with the use of pediatric-sized bloodcollection tubes. Am J Clin Pathol 1989;91:701–3.

45. Rosengart TK, Helm RE, DeBois WJ, Garcia N, Krieger KH,Isom OW. Open heart operations without transfusion usinga multimodality blood conservation strategy in 50 Jehovah’sWitness patients: implications for a “bloodless” surgicaltechnique. J Am Coll Surg 1997;184:618–29.

46. Hay WJ, Brockway J, Eyzaguirre M. Neonatal pulse oxim-etry: accuracy and reliability. Pediatrics 1989;83:717–22.

47. Zimmerman J, Dellinger R. Initial evaluation of a newintra-arterial blood gas system in humans. Crit Care Med1993;21:495–500.

48. Silver M, Jubran H, Stein S, McSweeney T, Jubran F.Evaluation of a new blood-conserving arterial line systemfor patients in intensive care units. Crit Care Med 1993;21:507–11.

49. Peruzzi W, Parker M, Lichtenthal P, Cochran-Zul C, Toth B,Blake M. A clinical evaluation of a blood conservationdevice in medical intensive care unit patients. Crit CareMed 1993;21:501-6.

50. Civetta J, Hudson-Civetta J. Maintaining quality of carewhile reducing charges in the ICU. Ten ways. Ann Surg1985;202:524–32.

51. Cannon MA, Beattie C, Speroff T, France D, Mistak B,Drinkwater D. The economic benefit of organizational re-structuring of the cardiothoracic intensive care unit. J Car-diothorac Vasc Anesth 2003;17:565–70.

52. Goodnough LT, Despotis GJ. Establishing practice guide-lines for surgical blood management. Am J Surg 1995;170(Suppl):16–20.

53. McDonnell Norms Group. Enhancing the use of clinical
guidelines: a social norms perspective. J Am Coll Surg2006;202:826–36.
54. Nuttall GA, Oliver WC, Santrach PJ, et al. Efficacy of asimple intraoperative transfusion algorithm for noneryth-rocyte component utilization after cardiopulmonary by-pass. Anesthesiology 2001;94:773–81; discussion 775A-6A.

55. Royston D, von Kier S. Reduced haemostatic factor trans-fusion using heparinase-modified thrombelastographyduring cardiopulmonary bypass. Br J Anaesth 2001;86:575–8.

56. Avidan MS, Alcock EL, Da Fonseca J, et al. Comparison ofstructured use of routine laboratory tests or near-patientassessment with clinical judgement in the management ofbleeding after cardiac surgery. Br J Anaesth 2004;92:178–86.

57. Despotis GJ, Grishaber JE, Goodnough LT. The effect of anintraoperative treatment algorithm on physicians’ transfu-sion practice in cardiac surgery. Transfusion 1994;34:290–6.

58. Shore-Lesserson L, Manspeizer HE, DePerio M, Francis S,Vela-Cantos F, Ergin MA. Thromboelastography-guidedtransfusion algorithm reduces transfusions in complex car-diac surgery. Anesth Analg 1999;88:312–9.

59. Capraro L, Kuitunen A, Salmenpera M, Kekomaki R. On-site coagulation monitoring does not affect hemostatic out-come after cardiac surgery. Acta Anaesthesiol Scand 2001;45:200–6.

60. Blumenthal D, McAlearney AS. Improving clinical practice:total quality management and the physician. 1st ed. SanFrancisco: Jossey-Bass Publishers, 1995.

61. Cohn LH, Rosborough D, Fernandez J. Reducing costs andlength of stay and improving efficiency and quality of carein cardiac surgery. Ann Thorac Surg 1997;64(Suppl):58–60;discussion 80–2.

62. O’Connor GT, Plume SK, Olmstead EM, et al. A regionalprospective study of in-hospital mortality associated withcoronary artery bypass grafting. The Northern New En-gland Cardiovascular Disease Study Group. JAMA 14 1991;266:803–9.

63. O’Connor GT, Plume SK, Olmstead EM, et al. Multivariateprediction of in-hospital mortality associated with coronaryartery bypass graft surgery. Northern New England Car-diovascular Disease Study Group. Circulation 1992;85:2110–8.

64. Kasper JF, Plume SK, O’Connor GT. A methodology for QIin the coronary artery bypass grafting procedure involvingcomparative process analysis. QRB Qual Rev Bull 1992;18:129–33.

65. Nugent WC, Schults WC. Playing by the numbers: howcollecting outcomes data changed by life. Ann Thorac Surg1994;58:1866–70.

66. Nugent WC. Benchmarking cardiac surgery: a regionalinitiative to improve outcomes. Interview by Reggi Veatch.Qual Lett Healthcare Lead 1994;6:11–3.

67. O’Connor GT, Plume SK, Olmstead EM, et al. A regionalintervention to improve the hospital mortality associatedwith coronary artery bypass graft surgery. The NorthernNew England Cardiovascular Disease Study Group. JAMA1996;275:841–6.

68. Caputo RP, Ho KK, Stoler RC, et al. Effect of continuousquality improvement analysis on the delivery of primarypercutaneous transluminal coronary angioplasty for acutemyocardial infarction. Am J Cardiol 1997;79:1159–64.

69. Goodnough LT, Shander A, Brecher ME. Transfusion med-icine: looking to the future. Lancet 2003;361:161–9.

70. AuBuchon JP, Birkmeyer JD, Busch MP. Safety of the bloodsupply in the United States: opportunities and controver-sies. Ann Intern Med 1997;127:904–9.

71. Yomtovian R, Lazarus HM, Goodnough LT, Hirschler NV,Morrissey AM, Jacobs MR. A prospective microbiologicsurveillance program to detect and prevent the transfusionof bacterially contaminated platelets. Transfusion 1993;33:902–9.

72. 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 celltransfusions. Transfusion 1993;33:794–7.

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

7

7

7

7

7

7

7

7

7

7

7

7

7

7


73. Yazer MH, Podlosky L, Clarke G, Nahirniak SM. The effectof prestorage WBC reduction on the rates of febrile non-hemolytic transfusion reactions to platelet concentrates andRBC. Transfusion 2004;44:10–5.

74. Paglino JC, Pomper GJ, Fisch GS, Champion MH, SnyderEL. Reduction of febrile but not allergic reactions to RBCsand platelets after conversion to universal prestorage leu-koreduction. Transfusion 2004;44:16–24.

75. King KE, Shirey RS, Thoman SK, Bensen-Kennedy D, TanzWS, Ness PM. Universal leukoreduction decreases theincidence of febrile nonhemolytic transfusion reactions toRBCs. Transfusion 2004;44:25–9.

76. Pineda AA, Taswell HF. Transfusion reactions associatedwith anti-IgA antibodies: report of four cases and review ofthe literature. Transfusion 1975;15:10–5.

77. Linden JV, Paul B, Dressler KP. A report of 104 transfusionerrors in New York State. Transfusion 1992;32:601–6.

78. Wallis JP, Chapman CE, Orr KE, Clark SC, Forty JR. Effectof WBC reduction of transfused RBCs on postoperativeinfection rates in cardiac surgery. Transfusion 2002;42:1127–34.

79. Silliman CC, Paterson AJ, Dickey WO, et al. The associationof biologically active lipids with the development of trans-fusion-related acute lung injury: a retrospective study.Transfusion 1997;37:719–26.

80. Mueller-Eckhardt C. Post-transfusion purpura. Br JHaematol 1986;64:419–24.

81. Linden JV, Wagner K, Voytovich AE, Sheehan J. Transfu-sion errors in New York State: an analysis of 10 years’experience. Transfusion 2000;40:1207–13.

82. Relman AS. Assessment and accountability: the third rev-olution in medical care. N Engl J Med 1988;319:1220–2.

83. Hill SE, van Wermeskerken GK, Lardenoye JW, et al.Intraoperative physiologic variables and outcome in car-diac surgery: part I. In-hospital mortality. Ann Thorac Surg2000;69:1070–6.

84. Stehling L, Zauder HL. Acute normovolemic hemodilution.Transfusion 1991;31:857–68.

85. Wu WC, Rathore SS, Wang Y, Radford MJ, Krumholz HM.Blood transfusion in elderly patients with acute myocardialinfarction. N Engl J Med 2001;345:1230–6.

86. Dial S, Delabays E, Albert M, et al. Hemodilution andsurgical hemostasis contribute significantly to transfusionrequirements in patients undergoing coronary artery by-pass. J Thorac Cardiovasc Surg 2005;130:654–61.

87. Herwaldt LA, Swartzendruber SK, Zimmerman B, ScholzDA, Franklin J, Caldarone CA. Hemorrhage after coronaryartery bypass graft procedures. Infect Control Hosp Epide-miol 2003;24:44–50.

88. Khanna MP, Hebert PC, Fergusson DA. Review of theclinical practice literature on patient characteristics associ-ated with perioperative allogeneic red blood cell transfu-sion. Transfus Med Rev 2003;17:110–9.

89. Litmathe J, Boeken U, Feindt P, Gams E. Predictors ofhomologous blood transfusion for patients undergoingopen heart surgery. Thorac Cardiovasc Surg 2003;51:17–21.

90. Parolari A, Antona C, Gerometta P, et al. The effect of “highdose” aprotinin and other factors on bleeding and revisionsfor bleeding in adult coronary and valve operations: ananalysis of 2190 patients during a five-year period (1987-1991). Eur J Cardiothorac Surg 1995;9:77–82.

91. Parr KG, Patel MA, Dekker R, et al. Multivariate predictorsof blood product use in cardiac surgery. J CardiothoracVasc Anesth 2003;17:176–81.

92. Schwann TA, Habib RH, Zacharias A, et al. Effects of bodysize on operative, intermediate, and long-term outcomesafter coronary artery bypass operation. Ann Thorac Surg2001;71:521–30; discussion 530–1.

93. Parolari A, Antona C, Rona P, et al. The effect of multipleblood conservation techniques on donor blood exposure inadult coronary and valve surgery performed with a mem-
brane oxygenator: a multivariate analysis on 1310 patients.J Card Surg 1995;10:227–35.
94. Carson JL, Noveck H, Berlin JA, Gould SA. Mortality andmorbidity in patients with very low postoperative Hb levelswho decline blood transfusion. Transfusion 2002;42:812–8.

95. Liu B, Belboul A, Larsson S, Roberts D. Factors influencinghaemostasis and blood transfusion in cardiac surgery. Per-fusion 1996;11:131–43.

96. Surgenor DM, Wallace EL, Churchill WH, Hao SH, Chap-man RH, Collins JJ Jr. Red cell transfusions in coronaryartery bypass surgery (DRGs 106 and 107). Transfusion1992;32:458–64.

97. Yamak B, Iscan Z, Mavitas B, et al. Low-dose oral antico-agulation and antiplatelet therapy with St. Jude Medicalheart valve prosthesis. J Heart Valve Dis 1999;8:665–73.

98. McDonald SB, Renna M, Spitznagel EL, et al. Preoperativeuse of enoxaparin increases the risk of postoperative bleed-ing and re-exploration in cardiac surgery patients. J Car-diothorac Vasc Anesth 2005;19:4–10.

99. Ascione R, Ghosh A, Rogers CA, Cohen A, Monk C,Angelini GD. In-hospital patients exposed to clopidogrelbefore coronary artery bypass graft surgery: a word ofcaution. Ann Thorac Surg 2005;79:1210–6.

00. Pivalizza EG, Warters RD, Gottschalk LI, Luehr SL, Hart-well EA. Clopidogrel and platelet transfusion in patientsundergoing coronary artery bypass graft surgery. Anaes-thesia 2003;58:603–4.

01. Silvestry SC, Smith PK. Current status of cardiac surgery inthe abciximab-treated patient. Ann Thorac Surg 2000;70(Suppl):12–9.

02. Dyke C, Bhatia D. Inhibitors of the platelet receptor glyco-protein IIb-IIIa and complications during percutaneouscoronary revascularization. Management strategies for thecardiac surgeon. J Cardiovasc Surg (Torino) 1999;40:505–16.

03. Dyke CM. Safety of glycoprotein IIb-IIIa inhibitors: a heartsurgeon’s perspective. Am Heart J 1999;138:307–16.

04. Clark SC, Vitale N, Zacharias J, Forty J. Effect of lowmolecular weight heparin (fragmin) on bleeding after car-diac surgery. Ann Thorac Surg 2000;69:762–5.

05. Myhre U, Stenseth R, Karevold A, et al. Bleeding followingcoronary surgery after preoperative low-molecular-weightheparin. Asian Cardiovasc Thorac Ann 2004;12:3–6.

06. Lemmer JH Jr. Clinical experience in coronary bypasssurgery for abciximab-treated patients. Ann Thorac Surg2000;70(Suppl):33–7.

07. Hein OV, von Heymann C, Morgera S, Konertz W, ZiemerS, Spies C. Protracted bleeding after hirudin anticoagula-tion for cardiac surgery in a patient with HIT II and chronicrenal failure. Artif Organs 2005;29:507–10.

08. Shah AC, Genoni M, Niederhauser U, Maloigne M, TurinaM. [R-hirudin (lepirudin, refludan) as an alternative anti-coagulant in heparin-induced thrombocytopenia duringcardiopulmonary bypass connection.] Schweiz MedWochenschr 2000;130:896–9.

09. Kwapisz MM, Schindler E, Muller M, Akinturk H. Pro-longed bleeding after cardiopulmonary bypass with recom-binant hirudin in heart transplantation. Eur J CardiothoracSurg 1999;16:256–7.

10. Skubas NJ, Despotis GJ, Vlasnic JJ, Moon MR. Preoperativeuse of enoxaparin and tirofiban: possible association withincreased bleeding postbypass. Anesthesiology 1999;91:869–72.

11. Berkowitz SD, Stinnett S, Cohen M, Fromell GJ, Bigonzi F.Prospective comparison of hemorrhagic complications aftertreatment with enoxaparin versus unfractionated heparinfor unstable angina pectoris or non-ST-segment elevationacute myocardial infarction. Am J Cardiol 2001;88:1230–34.

12. Ng HJ, Koh LP, Lee LH. Successful control of postsurgicalbleeding by recombinant factor VIIa in a renal failurepatient given low molecular weight heparin and aspirin.Ann Hematol 2003;82:257–8.

13. Lee LY, DeBois W, Krieger KH, et al. The effects of plateletinhibitors on blood use in cardiac surgery. Perfusion 2002;
17:33–7.

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7


14. Dyke CM, Bhatia D, Lorenz TJ, et al. Immediate coronaryartery bypass surgery after platelet inhibition with eptifi-batide: results from PURSUIT. Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using In-tegrelin Therapy. Ann Thorac Surg 2000;70:866 –71;discussion 871–2.

15. Lemmer JH Jr, Metzdorff MT, Krause AH Jr, Martin MA,Okies JE, Hill JG. Emergency coronary artery bypass graftsurgery in abciximab-treated patients. Ann Thorac Surg2000;69:90–5.

16. Marso SP, Bhatt DL, Roe MT, et al. Enhanced efficacy ofeptifibatide administration in patients with acute coronarysyndrome requiring in-hospital coronary artery bypassgrafting. PURSUIT Investigators. Circulation 2000;102:2952–8.

17. Sethi GK, Copeland JG, Goldman S, Moritz T, Zadina K,Henderson WG. Implications of preoperative administra-tion of aspirin in patients undergoing coronary arterybypass grafting. Department of Veterans Affairs Coopera-tive Study on Antiplatelet Therapy. J Am Coll Cardiol1990;15:15–20.

18. Kallis P, Tooze JA, Talbot S, Cowans D, Bevan DH, TreasureT. Pre-operative aspirin decreases platelet aggregation andincreases post-operative blood loss—a prospective, ran-domised, placebo controlled, double-blind clinical trial in100 patients with chronic stable angina. Eur J CardiothoracSurg 1994;8:404–9.

19. Taggart DP, Siddiqui A, Wheatley DJ. Low-dose preopera-tive aspirin therapy, postoperative blood loss, and transfu-sion requirements. Ann Thorac Surg 1990;50:424–8.

20. Pang JT, Fort S, Della Siega A, Cohen EA. Emergencycoronary artery bypass surgery in the era of glycoproteinIIb/IIIa receptor antagonist use. J Card Surg 2002;17:425–31.

21. Goldman S, Copeland J, Moritz T, et al. Starting aspirintherapy after operation. Effects on early graft patency.Department of Veterans Affairs Cooperative Study Group.Circulation 1991;84:520–6.

22. Bashein G, Nessly ML, Rice AL, Counts RB, Misbach GA.Preoperative aspirin therapy and reoperation for bleedingafter coronary artery bypass surgery. Arch Intern Med1991;151:89–93.

23. Aris A, Pisciotta AV, Hussey CV, Gale H, Lepley D. Open-heart surgery in von Willebrand’s disease. J Thorac Cardio-vasc Surg 1975;69:183–7.

24. Welsby IJ, Monroe DM, Lawson JH, Hoffmann M. Recom-binant activated factor VII and the anaesthetist. Anaesthe-sia 2005;60:1203–12.

25. Nakagawa M, Okuno M, Okamoto N, Fujino H, Kato H.Bernard-Soulier syndrome associated with 22q11.2 mi-crodeletion. Am J Med Genet 1 2001;99:286–8.

26. Gerritsen SW, Akkerman JW, Sixma JJ. Correction of thebleeding time in patients with storage pool deficiency byinfusion of cryoprecipitate. Br J Haematol 1978;40:153–60.

27. Lederer DJ, Kawut SM, Sonett JR, et al. Successful bilaterallung transplantation for pulmonary fibrosis associated withthe Hermansky-Pudlak syndrome. J Heart Lung Transplant2005;24:1697–9.

28. Demirkiran O, Utku T, Urkmez S, Dikmen Y. Chediak-Higashi syndrome in the intensive care unit. PaediatrAnaesth 2004;14:685–8.

29. Miyatake T, Matsui Y, Koyama M, Watanabe T, Yamada S,Yasuda K. Cardiac surgery for a patient with familiallecithin: cholesterol acyltransferase deficiency. Jpn J ThoracCardiovasc Surg 2005;53:317–9.

30. DeBois W, Liu J, Lee L, et al. Cardiopulmonary bypass inpatients with pre-existing coagulopathy. J Extra CorporTechnol 2005;37:15–22.

31. Palanzo DA, Sadr FS. Coronary artery bypass grafting in a
patient with haemophilia B. Perfusion 1995;10:265–70.
32. Cartmill TB, Castaldi PA, Halliday EJ. Cardiac-valve sur-gery in factor-VII deficiency. Lancet 1968;1:752–3.

33. Hilgartner M, Engle AM, Redo SF. Cardiac surgery inpatients with plasma thromboplastin antecedent (P.T.a.)deficiency. J Thorac Cardiovasc Surg 1965;49:974–81.

34. Brunken R, Follette D, Wittig J. Coronary artery bypass inhereditary factor XI deficiency. Ann Thorac Surg 1984;38:406–8.

35. MacKinlay N, Taper J, Renisson F, Rickard K. Cardiacsurgery and catheterization in patients with haemophilia.Haemophilia 2000;6:84–8.

36. Cmolik BL, Spero JA, Magovern GJ, Clark RE. Redo cardiacsurgery: late bleeding complications from topical throm-bin-induced factor V deficiency. J Thorac Cardiovasc Surg1993;105:222–8.

37. Mohr R, Golan M, Martinowitz U, Rosner E, Goor DA,Ramot B. Effect of cardiac operation on platelets. J ThoracCardiovasc Surg 1986;92:434–41.

38. Marumoto A, Ashida Y, Kuroda H, et al. Mitral valve repairfor mitral insufficiency due to infective endocarditis in apatient with idiopathic thrombocytopenic purpura. AnnThorac Cardiovasc Surg 2005;11:48–50.

39. Oba J, Aoki H, Yoshida T, Kanaoka T, Oe K. Mitral valvereplacement in a patient with idiopathic thrombocytopenicpurpura. Jpn J Thorac Cardiovasc Surg 2000;48:129–31.

40. Richards KM, Ferraris VA. Mitral valve replacement in apatient with idiopathic thrombocytopenic purpura: preop-erative treatment with danazol. J Cardiovasc Surg (Torino)1991;32:840–2.

41. Bennett J, Kolodziej M. Disorders of platelet function. DisMonth 1992;38:581–631.

42. Ferraris VA, Ferraris SP. Risk factors for postoperativemorbidity. J Thorac Cardiovasc Surg 1996;111:731–41.

43. Sellman M, Intonti MA, Ivert T. Reoperations for bleedingafter coronary artery bypass procedures during 25 years.Eur J Cardiothorac Surg 1997;11:521–7.

44. Mathew J, Abreo G, Namburi K, Narra L, Franklin C.Results of surgical treatment for infective endocarditis inintravenous drug users. Chest 1995;108:73–7.

45. Ralph-Edwards A, David TE, Bos J. Infective endocarditis inpatients who had replacement of the aortic root. AnnThorac Surg 1994;58:429–33.

46. Kaplan M, Cimen S, Kut MS, Demirtas MM. Cardiacoperations for patients with chronic liver disease. HeartSurg Forum 2002;5:60–5.

47. Baele PL, Ruiz-Gomez J, Londot C, Sauvage M, Van DyckMJ, Robert A. Systematic use of aprotinin in cardiac sur-gery: influence on total homologous exposure and hospitalcost. Acta Anaesthesiol Belg 1992;43:103–12.

48. Knutson JE, Deering JA, Hall FW, et al. Does intraoperativehetastarch administration increase blood loss and transfu-sion requirements after cardiac surgery? Anesth Analg2000;90:801–7.

49. Sandrelli L, Pardini A, Lorusso R, Sala ML, Licenziati M,Alfieri O. Impact of autologous blood predonation on acomprehensive blood conservation program. Ann ThoracSurg 1995;59:730–5.

50. Shore-Lesserson L. Monitoring anticoagulation and hemo-stasis in cardiac surgery. Anesthesiol Clin North Am 2003;21:511–26.

51. Goodnough LT, Despotis GJ. Future directions in utiliza-tion review: the role of transfusion algorithms. Transfus Sci1998;19:97–105.

52. Spiess BD, Gillies BS, Chandler W, Verrier E. Changesin transfusion therapy and reexploration rate after insti-tution of a blood management program in cardiac sur-gical patients. J Cardiothorac Vasc Anesth 1995;9:168 –73.

53. Despotis GJ, Joist JH, Hogue CW Jr, et al. More effectivesuppression of hemostatic system activation in patientsundergoing cardiac surgery by heparin dosing based on
heparin blood concentrations rather than ACT. ThrombHaemost 1996;76:902–8.

7

7

7

7

A

A

V

SSECB

CRGJBLMC

ESS

A

M

1T

a

Twotrtaaf


54. Blackstone EH. Breaking down barriers: helpful break-through statistical methods you need to understand better.J Thorac Cardiovasc Surg 2001;122:430–9.

55. Rosenbaum PR, Rubin DB. The central role of the propen-sity score in observatinoal studies for causal effects. Bi-

. Body No

or their interests in blood conservation as well as their willing-

ntwaotaE

b

ACtSoarCi

56. Benson K, Hartz AJ. A comparison of observational studiesand randomized, controlled trials. N Engl J Med 2000;342:1878–86.

57. Concato J, Shah N, Horwitz RI. Randomized, controlledtrials, observational studies, and the hierarchy of research

ometrika 1983;70:41–55. designs. N Engl J Med 2000;342:1887–92.

ppendix 1

uthors’ Financial Relationships With Industry

Author Disclosure Lecture Fees, Consultant, Paid Work Research Grant Support

.A. Ferraris Yes AstraZeneca, Aventis, Bayer, Networkfor Advancement of TransfusionAlternatives (NATA), and TheMedicines Company

American Heart Association, Aventis,Bayer, BioMarin Pharma, Guilford,Medtronic, National Heart, Lungand Blood Institute, and TheMedicines Company

.P. Ferraris No

.P. Saha No

.A. Hessel No.K. Haan No.D. Royston Yes Azko Nobel, Bayer, GSK, Johnson &

Johnson PharmaceuticalsBayer, Department of Health (UK),

Fresenius, Inotherapeutics,National Blood Service (UK)

.R. Bridges Yes Bayer

.S.D. Higgins No.J. Despotis Yes Bayer, Dyax, Inspire, Medtronic Medtronic

.R. Brown No.D. Spiess Yes Bayer, The Medicines Company Bayer, Synthetic Blood International. Shore-Lesserson Yes Bayer, NovoNordisk Bayer. Stafford-Smith No.D. Mazer Yes Bayer, NovoNordisk Physician Services Inc (PSI) and

Canadian Institutes of HealthResearch (CIHR)

. Bennett-Guerrero No

.E. Hill Yes Bayer, HemoConcepts

ppendix 2

ethods Used in Guideline Development

. Appointment of Blood Conservation Guidelineask Force Members

. The Guideline Writing Group

he Chairman of the Blood Conservation Writing Group (VAF)as appointed by the Evidence-based Workforce of The Societyf Thoracic Surgeons. The Writing Group Chairman appointedhe Writing Group from among volunteers, some of whom wereecruited from members of the Workforce. Other members ofhe Writing Group were recruited from a pool of surgeons andnesthesiologists with a known interest in blood conservationnd blood transfusion. Writing Group members were recruited

ess to devote long hours to research and synthesis of informa-ion on various blood conservation topics. The Writing Groupas charged by the STS Evidence-based Workforce with gener-

ting a complete document on blood conservation for review bythers. Oversight of the Writing Group was the responsibility ofhe Operating Board of the STS Council on Quality, Researchnd Patient Safety (Richard P. Anderson, Chair) and by the STSxecutive Committee.

. The Guideline Review Group

fter a draft of the complete guideline was created, the Society ofardiovascular Anesthesia (SCA) was approached about participa-

ion in the guideline review process. A lead anesthesiologist of theCA review group (BDS) was chosen by the Executive Committeef the SCA, and the lead reviewer picked acknowledged expertnesthesiologists in the field of blood conservation to help with theeview of the rough draft of the blood conservation guidelines.omments and critiques from the Review Group of the SCA are

ncluded in the final blood conservation document.

2D

Ttofassii(ciSsGrm

saldoreecacs

watiiui(s(PfAcip

tRmnstsmi

spgmbgm

3S

AawefdfitbwtTpca

4t

Eecsmt


. Conduct of the Literature Search and Documentevelopment

he lead author of the guideline (VAF) developed an outline forhe document and assigned topics in the outline to each memberf the STS Writing Group. Each Writing Group member per-ormed an exhaustive search of available literature on theirssigned topic. The Writing Group members used the followingtrategy for accumulating available evidence. First, a completeearch of the published literature was essential. Often thisnvolved searching many public access information sources,ncluding MEDLINE (www.ncbi.nlm.nih.gov/entrez), EMBASEwww.embase.com), CINAHL (www.cinahl.com), and the Co-hrane Collaboration (www.cochrane.org) for reports publishedn the peer-reviewed and non–peer-reviewed medical literature.earch of these databases using keywords relating to the as-igned topic generated a list of relevant references. Each Writingroup member synthesized an evidence-based narrative and

ecommendation based on these references and on other infor-ation sources.Second, all types of information sources were considered in

ummarizing available evidence. One has to realize that onlybout one third of the world’s medical literature appears onarge computer databases such as MEDLINE. Computerizedatabases such as MEDLINE are incomplete, especially in areasf subspecialization like cardiothoracic surgery. For completeeview of the medical literature on a particular assigned subject,ach Writing Group author was encouraged to conduct anxhaustive search of all available literature, not only Web-basedomputer databases like MEDLINE, but also unpublished trialsnd so-called “fugitive literature” (government reports, pro-eedings of conferences, published PhD theses, industry spon-ored studies, and so forth).

Traditionally, there is a hierarchy among information sourcesith more weight given to more rigorous studies such as RCTs

nd systematic reviews (eg, meta-analyses) than to observa-ional studies or to case reports. Because of this hierarchy, manynvestigators ignore other sources that may provide usefulnformation. Each Writing Group member was charged withnderstanding the following hierarchy of information sources,

n order of descending importance: randomized controlled trialsRCTs), systematic reviews (eg, meta-analyses), observationaltudies (eg, case-control or cohort studies), fugitive literaturegovernment reports, proceedings of conferences, publishedhD theses, unpublished negative randomized trials, and so

orth), case reports, expert consensus, and internet sources.lthough this hierarchy is pseudoscientific, RCTs are usually

onsidered the only means of proving cause and effect of a givenntervention, but a propensity adjusted observational study may
rovide similar information [764, 755]. There is some evidence b
hat well-done observational studies give results comparable toCTs dealing with similar outcomes [756, 757]. Likewise, infor-ation retrieved from the “fugitive” literature (eg, unpublished

egative randomized trials) may provide useful information forummary analysis that might otherwise be ignored in a litera-ure search. Each Writing Group author was encouraged toearch all of the information sources in the hierarchy, as eachight contribute important information and none should be

gnored.Each member of the STS Writing Group was encouraged to

ummarize numerous RCTs using meta-analysis where appro-riate. If several RCTs with similar endpoints and with homo-eneous research groups appeared in the literature search, theneta-analysis was performed. Two of the Writing Group mem-

ers with biostatistical experience (VAF and BDR) reviewed alluideline content for statistical rigor, especially theeta-analyses.

. Peer Review and Oversight by Professionalocieties

fter completion of the final draft by the STS Writing Group andfter review by the SCA Review Group, the revised documentas disseminated to all members of the STS and the SCA

lectronically. The lead author (VAF) compiled the commentsrom this peer-based review and added them to the finalocument where appropriate. After this peer-based review, thenal document was again circulated to the Writing Group and

he Review Group for final approval. After majority concurrencey the Writing Group and the Review Group, the final documentas reviewed by the Executive Committee of the STS. Only after

his final review was the manuscript submitted to The Annals ofhoracic Surgery. The manuscript was then subjected to the usualeer-review process of the The Annals, and the reviewers’omments were added to the revised manuscript whereppropriate.

. Recommendations of the Writing Group andhe Review Group

ach member of the Writing Group and the Review Groupvaluated the final document with all recommendations andlassifications of level of evidence, using the classificationcheme in Table 1. Each Writing Group and Review Groupember voted on the final recommendations and the final vote

alley appears in the following spreadsheet (Appendix 2, Ta-
le 1).

A

P

H

L

L

R

P

A

D

T

R

U

R

L

D

R

T

P

G


ppendix 2, Table 1. Final Voting on Recommendations by Writing Group and Review Group Members (n � 17)

Recommendation Class Yes NoReason(s) fordisagreement

reoperative identification of high-risk patients (advanced age, preoperative anemia,small body size, non–coronary artery bypass graft surgery (CABG) or urgentoperation, preoperative antithrombotic drugs, acquired or congenital coagulation/clotting abnormalities and multiple patient comorbidities) should be performed,and all available preoperative and perioperative measures of blood conservationshould be undertaken in this group because they account for the majority ofblood products transfused. (Level of evidence A)

I 17 0

igh-dose aprotinin is indicated to reduce the number of patients requiring bloodtransfusion, to reduce total blood loss, and to limit reexploration in high-riskpatients undergoing cardiac operations. Benefits of use should be balancedagainst the increased risk of renal dysfunction. (Level of evidence A)

I 15 2 Concerned aboutrenal safetyissues—shouldadd qualifyingstatement.

ow-dose aprotinin is indicated to reduce the number of patients requiring bloodtransfusion and to reduce the total blood loss in patients having cardiacprocedures. (Level of evidence A)

I 15 2 No safety data.

ysine analogues like epsilon-aminocaproic acid (EACA) and tranexamic acid (TXA)are indicated to reduce the number of patients who require blood transfusion,and to reduce total blood loss after cardiac operations. These agents are slightlyless potent blood-sparing drugs and the safety profile of these drugs is less wellstudied compared with aprotinin. (Level of evidence A)

I 16 1 Should be class IIbbecause of limitedsafety data.

outine use of red-cell saving is helpful for blood conservation in cardiac operationsusing cardiopulmonary bypass (CPB), except in patients with infection ormalignancy. (Level of evidence A)

I 17 0

reoperative hematocrit and platelet count are indicated for risk prediction, andabnormalities in these variables are amenable to intervention. (Level of evidenceA)

I 17 0

multimodality approach involving multiple stakeholders, institutional support,enforceable transfusion algorithms supplemented with point-of-care testing, andall of the already mentioned efficacious blood conservation interventions will limitblood transfusion and provide optimal blood conservation for cardiac operations.(Level of evidence A)

I 17 0

ipyridamole is not indicated to reduce postoperative bleeding, is unnecessary toprevent graft occlusion after coronary artery bypass grafting, and may increasebleeding risk unnecessarily. (Level of evidence B)

III 17 0

ransfusion is unlikely to improve oxygen transport when the hemoglobinconcentration is greater than 10 g/dL and is not recommended. (Level of evidenceC)

III 17 0

outine prophylactic use of desmopressin acetate (DDAVP) is not recommended toreduce bleeding or blood transfusion after cardiac operations using CPB. (Level ofevidence A)

III 17 0

se of prophylactic positive end-expiratory pressure (PEEP) to reduce bleedingpostoperatively is not effective. (Level evidence B)

III 17 0

outine use of intraoperative platelet or plasmapheresis is not recommended forblood conservation during cardiac operations using CPB. (Level of evidence A)

III 17 0

eukocyte filters on the CPB circuit for leukocyte-depletion should not be used forperioperative blood conservation and may actually activate leukocytes duringCPB. (Level of evidence B)

III 16 1 Leukoreduction maybe helpful forother thingsbesides bloodconservation.

irect reinfusion of shed mediastinal blood from postoperative chest tube drainageis not recommended as a means of blood conservation and may cause harm.(Level of evidence B)

III 16 1 May replace red cellmass in excessivebleeding.

outine use of ultrafiltration during or immediately after CPB is not helpful forblood conservation in adult cardiac operations. (Level of evidence B)

III 15 2 May have small butsignificant effect.

opical hemostatic agents that employ bovine thrombin are not helpful for bloodconservation during CPB and may be potentially harmful. (Level of evidence B)

III 17 0

reoperative screening of the intrinsic coagulation system is not recommendedunless there is a clinical history of bleeding diathesis. (Level of evidence B)

III 16 1 Useful but low yield.

iven that the risk of transmission of known viral diseases with blood transfusion iscurrently rare, fears of viral disease transmission should not limit administrationof INDICATED blood products. (This recommendation only applies to countries/blood banks where careful blood screening exists.) (Level of evidence C)

IIa 15 2 Concerned that newviral or otherinfectious agentswill change this
recommendation.

A

P

I

I

W

I

D

I

I

I

R


ppendix 2, Table 1. Final Voting on Recommendations by Writing Group and Review Group Members (n � 17) (Continued)


atients who have thrombocytopenia (less than 50,000/mm2), who arehyperresponsive to aspirin or other anti-platelet drugs as manifested by abnormalplatelet function tests or prolonged bleeding time, or who have known qualitativeplatelet defects represent a high-risk group for bleeding. Maximum bloodconservation interventions during cardiac procedures are reasonable for thesehigh-risk patients. (Level of evidence B)

IIa 17 0

t is reasonable to discontinue thienopyridines 5 to 7 days before cardiac proceduresto limit blood loss and transfusion. Failure to discontinue these drugs beforeoperation risks increased bleeding and possibly worse outcome. Caution must beused with sudden withdrawal of antiplatelet therapy in the presence of drug-eluting stents. This can lead to stent thrombosis and the surgical team shouldconsider various alternatives to maintain stent patency. That could even includehospitalization to convert thienopyridine therapy to short-acting GP2b/3ainhibitors for a few days before operation. (Level of evidence B)

IIa 17 0

t is reasonable to discontinue low-intensity antiplatelet drugs (eg, aspirin) only inpurely elective patients without acute coronary syndromes before operation withthe expectation that blood transfusion will be reduced. (Level of evidence A)

IIa 15 2 Should be class IIbbecause of toomany otherconflicting dataand possibilitythat preoperativeaspirin isprotective.

ith hemoglobin levels below 6 g/dL, red blood cell transfusion is reasonable, asthis can be life-saving. Transfusion is reasonable in most postoperative patientswhose hemoglobin is less than 7 g/dL but no high-level evidence supports thisrecommendation. (Level of evidence C)

IIa 17 0

t is reasonable to transfuse non–red cell hemostatic blood products based onclinical evidence of bleeding and preferably guided by point-of-care tests thatassess hemostatic function in a timely and accurate manner. (Level of evidence C)

IIa 17 0

uring cardiopulmonary bypass with moderate hypothermia, transfusion of redcells for a hemoglobin level � 6 g/dL is reasonable except in patients at risk fordecreased cerebral oxygen delivery (ie, history of cerebrovascular accident,diabetes mellitus, cerebrovascular disease, carotid stenosis) where higherhemoglobin levels may be justified. (Level of evidence C)

IIa 17 0

n the setting of hemoglobin values exceeding 6 g/dL while on CPB, it is reasonableto transfuse red cells based on the patient’s clinical situation, and this should beconsidered as the most important component of the decision-making process.Indications for transfusion of red blood cells in this setting are multifactorial andshould be guided by patient-related factors (ie, age, severity of illness, cardiacfunction, or risk for critical end-organ ischemia), the clinical setting (massive oractive blood loss), and laboratory or clinical variables (eg, hematocrit, SVO2,electrocardiographic or echocardiographic evidence of myocardial ischemia, etc).(Level of evidence C)

IIa 17 0

n patients after cardiac operations with hemoglobin levels below 6 g/dL, red bloodcell transfusion is reasonable and can be life-saving. Transfusion of red cells isreasonable in most postoperative patients with hemoglobin � 7 g/dL, but nohigh-level evidence supports this recommendation. (Level of evidence C)

IIa 17 0

t is reasonable to transfuse non–red cell hemostatic blood products based onclinical evidence of bleeding and preferably guided by specific point-of-care teststhat assess hemostatic function in a timely and accurate manner. (Level ofevidence C)

IIa 17 0

ecombinant human erythropoietin (EPO) is reasonable to restore red blood cellvolume in patients undergoing autologous preoperative blood donation beforecardiac procedures. However, no large-scale safety studies for use of this agent incardiac surgical patients have been performed. Available evidence suggests anacceptable safety profile. (Level of evidence A)

IIa 13 4 Should be class IIbbecause of limitedsafety data.Predonation isfrequentlyimpractical andhas low efficacy.The use of a high-cost drug thatdoes not havegood evidence toaugment a low-efficacy procedureis unwarranted
(class III).

A

R

P

O

A

A

T

M

I

I

I

U




ecombinant human EPO is a reasonable alternative for anemic low-risk electivepatients (hemoglobin �13 g/dL) before cardiac procedures, provided that EPO isgiven in conjunction with iron therapy several days or more before operation.(Level of evidence B)

IIa 16 1 Needs severalweeks, not days(not practical).

redonation of as much as two units of autologous blood is reasonable for bloodconservation in carefully selected (mostly elective) patients before routine cardiacoperation, particularly when coupled with appropriately dosed EPO therapyand/or iron therapy. (Level of evidence A)

IIa 14 3 Should be class IIbbecause of limitedsafety data.Predonation isfrequentlyimpractical andhas low efficacy.The use of a high-cost drug thatdoes not havegood evidence toaugment a low-efficacy procedureis unwarranted(class III).

ff-pump coronary artery bypass (OPCAB) is a reasonable means of bloodconservation, provided that emergent conversion to on-pump bypass is unlikelyeither based on surgeon experience or patient characteristics. (Level of evidenceA)

IIa 16 1 Onlyrecommendationthat is dependenton technical skill.Not arecommendationfor every program,only those withappropriatetechnical skills.

lternatives to laboratory blood sampling (eg, oximetry instead of arterial bloodgasses) are reasonable means of blood conservation before operation. (Level ofevidence B)

IIa 16 1 Alternatives are notreliable enough.

comprehensive, integrated, multimodality blood conservation program in theintensive care unit is a reasonable means to limit blood transfusion. (Level ofevidence B)

IIa 17 0

otal quality management (TQM), including continuous measurement and analysisof blood conservation interventions as well as assessment of new bloodconservation techniques, is reasonable to implement a complete bloodconservation program. (Level of evidence B)

IIa 17 0

ost high-intensity antithrombotic and antiplatelet drugs (including adenosinediphosphate–receptor inhibitors, direct thrombin inhibitors, low-molecular-weightheparins, platelet glycoprotein inhibitors, tissue-type plasminogen activators,streptokinase) are associated with increased bleeding after cardiac operations. It isnot unreasonable to stop these medications before operation to decrease minorand major bleeding events. The timing of discontinuation depends on thepharmacodynamic half-life for each agent as well as the potential lack ofreversibility. Unfractionated heparin is a notable exception in that it is the onlyagent that may be discontinued shortly before operation or not at all. (Level ofevidence C)

IIb 16 1 Too broad of acategory.

t is not unreasonable to transfuse red cells in certain patients with criticalnoncardiac end-organ ischemia (eg, central nervous system and gut) whosehemoglobin levels are as high as 10 g/dL, but more evidence to support thisrecommendation is required. (Level of evidence C)

IIb 16 1 Not enough data.

n patients on CPB with risk for critical end-organ ischemia/injury, it is notunreasonable to keep the hemoglobin level �7 g/dL. (Level of evidence C)

IIb 16 1 Evidence does notjustify thisrecommendation.

t is not unreasonable to use preoperative EPO, given at least a few days beforeoperation, to increase red cell mass in elective patients who are at risk forpostoperative anemia and depressed endogenous EPO production. (Level ofevidence C)

IIb 17 0

se of DDAVP is not unreasonable to attenuate excessive bleeding and transfusionin certain patients with demonstrable and specific platelet dysfunction known torespond to this agent (eg, uremic or CPB-induced platelet dysfunction, type I von

IIb 17 0

Willebrand’s disease). (Level of evidence B)

A

U

A

I

A

F

I

H

L

I

A

D

S

P




se of recombinant factor VIIa concentrate is not unreasonable for the managementof intractable nonsurgical bleeding that is unresponsive to routine hemostatictherapy after cardiac procedures using CPB. (Level of evidence B)

IIb 16 1 Should be class IIIuntil more safetyand efficacy dataare available.

trial of therapeutic PEEP to reduce excessive postoperative bleeding is notunreasonable. (Level of evidence B)

IIb 15 2 Does not believethat evidence iscompellingenough.

t is not unreasonable to use open venous reservoir membrane oxygenator systemsduring cardiopulmonary bypass for reduction in blood utilization and improvedsafety. (Level of evidence C)

IIb 14 3 Closed venoussystem is saferand better forblood conservationbased on personalobservation. Needto separate type ofreservoir fromtype ofoxygenator.

ll commercially available blood pumps provide acceptable blood conservationduring CPB. It is not unreasonable to prefer centrifugal pumps because ofperfusion safety features. (Level of evidence B)

IIb 16 1 Centrifugal pumpsare standard ofcare.

or patients requiring longer CPB times (more than 2–3 hours), it is notunreasonable to consider maintenance of higher or patient-specific heparinconcentrations during CPB to reduce hemostatic system activation, reduceconsumption of platelets and coagulation proteins, and reduce blood transfusion.(Level of evidence B)

IIb 16 1 Not necessary.

t is not unreasonable to use either protamine titration or empiric low-doseregimens (eg, 50% of total heparin dose) to lower the total protamine dose andlower the protamine/heparin ratio at the end of CPB to reduce bleeding andblood transfusion requirements. (Level of evidence B)

IIb 16 1 Not enough data.

eparin-coated bypass circuits (either the oxygenator alone or the entire circuit) arenot unreasonable for blood conservation in cardiac operations. (Level of evidenceB)

IIb 17 0

ow doses of systemic heparinization (activated blood clotting time �300 s) are notunreasonable for blood conservation during CPB, but the possibility ofunderheparinization and other safety concerns have not been well studied. (Levelof evidence B)

IIb 17 0

t is not unreasonable to use low prime and minimized extracorporeal bypasscircuits to reduce the fall in hematocrit during CPB as part of a multimodalityblood conservation program. (Level of evidence B)

IIb 16 1 May cause too muchhemoconcentrationwith resultantrenal impairmentfrom nonpulsatileflow.

cute normovolemic hemodilution is not unreasonable for blood conservation butits usefulness is not well established. It could be utilized as part of amultipronged approach to blood conservation. (Level of evidence B)

IIb 16 1 It is a very lowefficacy procedurewith nosupportingevidence.

uring CPB, it is not unreasonable to use intraoperative autotransfusion, either withblood directly from cardiotomy suction or recycled using a cell-saving device, toaugment blood conservation. (Level of evidence C)

IIb 16 1 May have someharmful CNSeffects.

hortly after the completion of cardiopulmonary bypass, salvage of pump blood,either administered without washing or after washing with a cell-saving device, isnot unreasonable as a means of blood conservation. (Level of evidence C)

IIb 16 1 Discard of availablefresh autologouswhole blood isintuitively wrongas part of amultimodalityapproach.

ostoperative mediastinal shed blood reinfusion using washed mediastinal blood(cell saving) is not unreasonable for blood conservation when used in conjunctionwith other blood conservation interventions. Washing of shed mediastinal bloodmay decrease lipid emboli and decrease the concentration of inflammatorycytokines, and reinfusion of washed blood may be reasonable to limit bloodtransfusion as part of a multimodality blood conservation program. (Level of

IIb 15 2 May be helpful forblood conservationbut harmful forproinflammatoryeffects. Notenough safety

evidence B) data.

A

R

U

T

S

D




etrograde autologous priming of the CPB circuit is not unreasonable for bloodconservation. (Level of evidence B)

IIb 15 2 Should be class IIaaccording toavailable data ascited. No safetydata, and maycause harm (classIIIrecommendation).

se of modified ultrafiltration is not unreasonable for blood conservation duringCPB in pediatric patients or in patients for whom the priming volume is a verysignificant fraction of the total blood volume. (Level of evidence B)

IIb 16 1 Do not feel that datasupports thisrecommendation.

opical sealants used to assist in the repair of complex high-risk cardiac and aorticprocedures are not unreasonable to limit bleeding in certain key situations (eg,left ventricular free wall rupture and aortic dissection) but are associated withcomplications that may limit their usefulness in less high risk situations. (Level ofevidence C)

IIb 16 1 Should not replacegood surgicaltechnique.

creening preoperative bleeding time is not unreasonable in high-risk patients,especially those who receive preoperative antiplatelet drugs. (Level of evidence B)

IIb 14 3 Bleeding time is notavailable and isnot recommendedfor screeningpurposes by otherprofessionalorganizations.Identification of ahigh-risk patient isadequate.Bleeding time willnot usefullysubdivide thishigh-risk groupinto either a lowerrisk group, or intoan even higherrisk group.

evices aimed at obtaining direct hemostasis at catheterization access sites are notunreasonable for blood conservation if operation is planned within 24 hours.

IIb 16 1 Not enough data tosupport

(Level of evidence C) recommendation.

Documents

Perioperative Blood Transfusion and Blood Conservation in ... · surgery, routine use of a cell-saving device, and implemen-tation of appropriate transfusion indications. An important