- Email: [email protected]
Efforts to Change Transfusion Practice and Reduce Transfusion Rates Are Effective in Coronary Artery Bypass Surgery
ORIGINAL ARTICLES
Efforts to Change Transfusion Practice and Reduce Transfusion Rates Are Effective in Coronary Artery Bypass Surgery Jan Jesper Andreasen, MD, PhD,* Jesper Eske Sindby, MD,* Barbara Cristina Brocki,* Bodil Steen Rasmussen, MD, PhD,† and Claus Dethlefsen, MSc, PhD* Objectives: Efforts to decrease allogeneic blood transfusion and avoid unnecessary transfusions in cardiac surgery are important because transfusions are associated with increased postoperative morbidity and mortality. The purpose of the present study was to evaluate the long-term effects of multidisciplinary efforts to reduce allogeneic blood transfusion rates and avoid unnecessary red blood cell (RBC) transfusions in primary elective coronary artery bypass graft (CABG) surgery. Design: A retrospective observational study. Setting: A single center study in a university-affiliated hospital. Participants: A total of 450 patients undergoing primary elective CABG surgery during 2004, 2008, or 2010. Interventions: The application of systematic multimodal perioperative blood-sparing techniques and interventions directed to change transfusion behaviors.
Measurements and Main Results: The results from an audit on transfusion practices in 2004 were compared with similar audits performed in 2008 and 2010 using a before-and-after study design. The patient populations were comparable throughout the years. The median postoperative chest tube bleeding was decreased from 950 mL in 2004 to 750 mL in 2010. The proportion of patients transfused with allogeneic blood products was decreased from 64% to 47%. Overtransfusion with allogeneic RBCs defined as the proportion of patients transfused with RBCs discharged with hemoglobin >7 mmol/L (11.3 g/dL) was reduced from 36% to 16%. Conclusions: Multimodal efforts to change transfusion behaviors and decrease transfusion rates in CABG surgery have persistent effects for several years. © 2012 Elsevier Inc. All rights reserved.
A
risks of RBC transfusions) to reduce allogeneic blood transfusion rates and avoid unnecessary RBC transfusions in elective primary CABG surgery.
LLOGENEIC RED BLOOD CELL (RBC) transfusion has been associated with increased morbidity and mortality after cardiac surgery in several observational studies.1,2 Therefore, increasing attention is being paid toward the use of blood-sparing techniques and transfusion behaviors. Guidelines on transfusion practices in cardiothoracic surgery have been published,3,4 but there still is a considerable variation in transfusion practices nationally and internationally.5,6 A study of transfusion practices during elective coronary artery bypass graft (CABG) surgery previously was performed in Denmark to better understand the factors that determined the use of blood products during CABG surgery.6 This initial audit showed, for example, that several inappropriate RBC transfusions were given, and the audit created a basis for educational efforts to standardize transfusion practices. Several behavioral interventions appear effective in changing transfusion practices and reducing blood use.7 However, more information on the long-term effects of such interventions is needed because most previous studies only evaluated the early effects. The purpose of the present study was to evaluate the long-term effects of multidisciplinary efforts (eg, the routine perioperative use of tranexamic acid [TA] in all patients, the implementation of a guideline for the preoperative discontinuation of antiplatelet drugs, the implementation of a specific algorithm for the management of postoperative bleeding including transfusion, the use of thromboelastometry, the introduction of a new hospital form outlining general transfusion guidelines, educational efforts, and continued reminders regarding the effectiveness and
KEY WORDS: blood transfusion, coronary artery bypass graft surgery, red blood cells
METHODS This study was approved by the Danish Data Protection Agency (record number 1-16-02-1-08/017). A retrospective review of 150 consecutive medical records from each of the years 2008 and 2010 focusing on transfusion practices among patients undergoing primary elective CABG surgery was performed. Results were compared with the results from a similar audit also performed in the North Denmark Region in 2004 using a before/after study design.8 The patients were identified through the local hospital discharge register starting at the beginning of each year. In case of missing patient records, the record of the next patient in line was obtained.
From the Departments of *Cardiothoracic Surgery and †Anesthesiology, Center for Cardiovascular Research, Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark. Supported by a grant from Vestdansk Sundhedsvidenskabeligt Forskningsforums konsulenttjeneste. Presented in part at the Annual Meeting of the European Association of Cardiothoracic Anesthesiologists (EACTA) 2011, Vienna, Austria, June 1-4, 2011. Address reprint requests to Jan Jesper Andreasen, MD, PhD, Department of Cardiothoracic Surgery, Aalborg Hospital, Aarhus University Hospital, Hobrovej 18-22, DK-9100 Aalborg, Denmark. E-mail: [email protected] © 2012 Elsevier Inc. All rights reserved. 1053-0770/2604-0002$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.02.006
Journal of Cardiothoracic and Vascular Anesthesia, Vol 26, No 4 (August), 2012: pp 545-549
545
546
ANDREASEN ET AL
Haemoglobin threshold Low-risk patient <4,5 mmol/L
Postoperative bleeding e e
Cardiac tamponade
(<7.3 g/dL)
High-risk patient
4,5-6,0 mmol/L (7.3-9.7 g/dL)
Patients are pragmatically grouped as high- or low-risk patients based on the postoperative occurrence of concomitant diseases, symptoms and signs, eg, tachycardia, hypotension, renal function, pulmonary function, left ejection fraction, age.
Below threshold
ACT
Protamine
mL/h t > 500 Oplagt > 800 mL/2 h > 900 mL/3 h > 1200 mL/4 h
Suspicion of microvascular bleeding
Platelet count and function test
Unneutralized heparin
Red blood cells
“Surgical” bleeding?
Measurements of hemoglobin every hour
< 50 x 109 or reduced function
Platelets
Thromboelastometry
Decreased concentration of plasma clotting factors
Plasma
Fibrinolysis
Tranexamic acid
Normal coagulation
Decreased fibrinogen level
Fibrinogen
Re-exploration
Continuing bleeding and normal coagulation status Fig 1.
An algorithm for the postoperative management of bleeding in cardiac surgery.
A total of 2,340 patients underwent CABG surgery in Denmark without other concomitant cardiac surgery during 2004. Since 2004, the number of cases gradually has been decreasing to approximately 1,800 cases per year. The study was conducted within the population of the North Denmark Region, which comprises approximately 0.6 million persons or 11% of the total population in Denmark. All patients in the North Denmark Region are referred to Aalborg Hospital, Aarhus University Hospital, if they need CABG surgery. The following baseline patient characteristics were collected: sex, age, height, weight, a previous history of chronic obstructive lung disease, and left ventricular ejection fraction. Baseline laboratory characteristics were hemoglobin, creatinine, platelet count, and international normalized ratio (INR). Preoperative treatment with antiplatelet drugs such as acetylsalicylic acid, clopidogrel, and other nonsteroidal anti-inflammatory drugs also was recorded. Recorded perioperative data included the use of prophylactic antifibrinolytic drugs, the use of cardiopulmonary bypass (CPB), the aortic cross-clamp time, the CPB time, and the lowest hemoglobin during CPB. Recorded postoperative data included chest tube drainage 6 hours postoperatively and in total, the use of autotransfusion, the use of allogeneic blood products during the entire hospitalization, and the hemoglobin concentration at discharge. The hemoglobin concentration at discharge was defined as the last measurement of hemoglobin before discharge. The following severe postoperative complications also were recorded: 30-day mortality and reoperation because of bleeding or surgery for other major complications, such as mediastinitis and gastrointestinal complications. To restrict the analyses to patients who were transfused for reasons directly related to the CABG surgery in itself, patients who experienced severe complications were excluded because transfusions in these patients may not have been related directly to peri- and postoperative treatment of the microvascular bleeding after CABG surgery. Jehovah’s witnesses were not included. Surgical routines, anesthetic procedures, and standard normothermic or slightly hypothermic CPB procedures mainly were unchanged dur-
ing the study period. On-pump and off-pump (OPCAB) surgery were performed according to the preferences of the surgeons, who were senior surgeons as well as trainees. The left internal mammary artery (IMA) was used as a standard whenever the anterior descending artery was to be grafted. The right IMA, saphenous veins, and radial arteries were used as additional grafts according to the preferences of the surgeons. During surgery, blood from the operative field was returned to the CPB circuit. At the end of the surgery, the residual blood volume remaining in the CPB circuit was returned unprocessed to the patients. Postoperative autotransfusion of unwashed mediastinal blood was used routinely during 2004 but was abandoned in most patients after 2007 because of published recommendations.3 Preoperative autologous blood donation, acute normovolemic hemodilution, hemofiltration, and prophylactic transfusions of plasma and platelets were not used. Transfusions of allogeneic blood components (ie, RBC, fresh frozen plasma, and platelet concentrates) were given at the discretion of the attending anesthesiologist and/or surgeon. No specific transfusion algorithm was used during 2004. Indications for allogeneic transfusion were based on routine laboratory measurements of INR, the activated coagulation time, platelet counts, fibrinogen levels, hemoglobin, and hematocrit values in addition to the measurements of hemodynamic and physiologic data, the rate of blood loss, and the existence of concomitant diseases. A specific algorithm for the management of postoperative bleeding including transfusion was designed after 2004 (Fig 1). This algorithm included the use of thromboelastometry, which was introduced in the hospital during the study period. The algorithm was implemented for use in the intensive care unit together with educational efforts directed toward doctors and nurses regarding the potential value of thromboelastometry and the risks of allogeneic RBC transfusions. Thromboelastometry was performed in a laboratory 10 minutes away from the intensive care unit, and the results and suggestions for further treatment because of increased microvascular bleeding were given by doctors in the laboratory over the telephone.
TRANSFUSION PRACTICE BEHAVIOR
547
Table 1. The Unadjusted and Adjusted Relative Risk With 95% Confidence Intervals for Allogeneic Blood Transfusion According to the Year of Surgery Among Patients Undergoing Elective FirstTime CABG Surgery Year
Unadjusted RR (95% CI)
Adjusted RR (95% CI)
2004 2008 2010
1.0 (reference) 0.9 (0.7-1.0) 0.7 (0.6-0.9)
1.0 (reference) 0.7 (0.5-0.9) 0.6 (0.5-0.8)
NOTE. RRs were adjusted for sex, age, body mass index, preoperative hemoglobin and platelet count, INR, preoperative use of antiplatelet drugs, use of antifibrinolytic drugs, use of CPB, and time on CPB. Abbreviation: CI, confidence interval.
A guideline for the preoperative discontinuation of antiplatelet drugs based on the literature9,10 was implemented in 2008. At the time when the indication for CABG or OPCAB surgery in each patient was discussed by the local heart team consisting of both interventional cardiologists and cardiac surgeons, a decision was made whether the patient should continue or discontinue antiplatelet drugs. Furthermore, the prophylactic use of tranexamic acid in all patients became a routine. A total of 1.5 g of tranexamic acid (Cyklokapron; Pfizer Consumer Healthcare, Copenhagen, Denmark) were given as an intravenous bolus via a central venous catheter beginning at the induction of the anesthesia followed by a constant infusion of 200 mg/h until an additional 1.5 g was given. The dose regimen was unchanged throughout the study period. Educational efforts based on the results from the audit in 2004 were performed, and continuous reminders were given. Additionally, a new hospital form outlining general transfusion guidelines as recommended internationally2,3 and supported by the Danish National Board of Health was introduced in the hospital. The quality of the medical transfusion efforts was assessed primarily on the basis of the proportion of patients receiving transfusions and the proportion of overtransfusion defined as the proportion of patients transfused with RBCs and discharged with hemoglobin ⬎7 mmol/L (11.3 g/dL). If the hemoglobin concentrations in the blood after the last
transfusion was ⬎7.0 mmol/L (11.3 g/dL), there was probably at least 1 unit of RBCs given at a time when the hemoglobin concentration was ⬎6.5 mmol/L (10.5 g/dL) provided no major changes occurred in the fluid balance of the patient. Overtransfusion was defined as the proportion of patients transfused with RBCs discharged with hemoglobin ⬎7 mmol/L (11.3 g/dL) because there is no evidence supporting any benefit of RBC transfusions in patients without major bleeding if the hemoglobin concentration is ⬎6.5 mmol/L (10.5 g/dL). Descriptive statistics included the mean, standard deviation, median, interquartile range, and the percentage of patients when appropriate. Differences between years were tested for statistical significance by using chi-square analysis, analysis of variance, and Kruskal-Wallis tests in the case of nonnormality. A p value ⬍0.05 is considered significant. A sample size calculation was performed before the study to calculate how many patient records were needed to have an 80% chance of observing a 20% decrease in total chest tube drainage at a significance level of 0.05 compared with the results obtained in 2004. At least 106 patient records were needed. Missing data were excluded. To check whether missing data might play any major role, the authors repeated the analysis in Table 1 using multiple imputation of missing data. Altogether, 83 values were missing and were imputed 10 times. Using Rubin’s rules, the results were combined. Poisson regression with robust variance estimation was used to compare the use of allogeneic blood transfusion in different years. Specifically, the authors computed relative risks (RRs) with 95% confidence intervals for the requirement of ⱖ1 allogeneic blood transfusion. Poisson regression was used because, in contrast to logistic regression, it may provide more accurate estimates of adjusted RRs in cohorts with common outcomes.11 RRs were adjusted for age and sex of the patients, body mass index, preoperative hemoglobin and platelet count, INR, preoperative use of antiplatelet drugs, use of antifibrinolytic drugs, use of CPB, and CPB time. All analyses were performed out using Stata version 11.1 (Stata Corp; College Station, TX). RESULTS
Baseline characteristics of included patients undergoing isolated, elective, first-time CABG surgery during different years are shown in Table 2. Patients who died within 30 days and patients who underwent surgery because of complications,
Table 2. Baseline Characteristics of Patients Undergoing First-Time Coronary Artery Bypass Surgery in Different Years Variable
2004 (n ⫽ 139)
2008 (n ⫽ 147)
2010 (n ⫽ 145)
p Value
Male sex (%) Age (y) Height (cm) Weight (kg) Body mass index (kg/m2) Hemoglobin (mmol/L) Hemoglobin (g/dL) Creatinine (mmol/L) Platelet count (109/L) INR Ejection fraction (%) COPD (%) Preoperative use of acetylsalicylic acid (%)† Preoperative use of clopidogrel (%)† Patients without preoperative intake of antiplatelets (%)
81.3 65.8 ⫾ 8.6 172.2 ⫾ 8.8 80.3 ⫾ 14.9* 27.0 ⫾ 4.0* 8.5 ⫾ 1.0 13.7 ⫾ 1.6 99.2 ⫾ 62.3 253.7 ⫾ 80.8 1.0 ⫾ 0.2 57.5 ⫾ 15.4 5.8 54.0 27.1 46.0
76.2 67.1 ⫾ 9.4 171.8 ⫾ 8.8 81.7 ⫾ 14.4 27.7 ⫾ 4.2 8.3 ⫾ 0.9 13.4 ⫾ 1.5 93.5 ⫾ 39.4 287.7 ⫾ 89.9 1.0 ⫾ 0.1 50.3 ⫾ 12.3 8.8 63.3 26.5 36.7
78.2 67.3 ⫾ 9.5 172.1 ⫾ 8.5 79.8 ⫾ 14.8 26.8 ⫾ 3.8 8.4 ⫾ 0.9 13.5 ⫾ 1.5 79.4 ⫾ 20.3 273.5 ⫾ 90.8 1.0 ⫾ 0.2 50.2 ⫾ 12.7 11.7 57.9 13.8 42.1
0.54 0.36 0.93 0.50 0.17 0.38 0.38 ⬍0.01 ⬍0.01 0.69 ⬍0.01 0.21 0.28 0.01 0.28
NOTE. Data are presented as the mean ⫾ standard deviation or percentage of patients. Abbreviation: COPD, chronic obstructive pulmonary disease. *One missing. †Intake within 7 days before surgery.
548
ANDREASEN ET AL
Table 3. Operative Characteristics of Patients Undergoing Elective First-Time Coronary Artery Bypass Surgery in Different Years 2004 (n ⫽ 139)
2008 (n ⫽ 147)
2010 (n ⫽ 145)
p Value
54.0 47.8* 104.05 ⫾ 32.9 56.0 ⫾ 19.7 3.3 ⫾ 1.0 4.9 ⫾ 0.6 (mmol/L) 7.9 ⫾ 1.0 (g/dL)
96.6 99.3 76.0 ⫾ 27.5 40.8 ⫾ 15.6 3.2 ⫾ 0.8 4.8 ⫾ 0.8 (mmol/L) 7.7 ⫾ 1.3 (g/dL)
80.0 100.0 85.9 ⫾ 35.3 42.4 ⫾ 21.3 3.1 ⫾ 0.9 5.1 ⫾ 0.9 (mmol/L) 8.2 ⫾ 1.5 (g/dL)
⬍0.01 ⬍0.01 ⬍0.01 ⬍0.01 0.39 0.01 0.01
Variable
Percentage of patients operated with the use of CPB Percentage of patients treated with antifibrinolytic drugs CPB time (min) Cross-clamp time (min) Number of distal anastomoses Lowest preoperative hemoglobin on CPB
NOTE. Data are presented as the mean ⫾ standard deviation or as %. *Three missing.
which were mainly reoperation because of surgical bleeding (n ⫽ 3, 3, and 8 in the years 2004, 2008, and 2010, respectively), were excluded from the analyses leaving 139, 147, and 145 patients who underwent surgery in 2004, 2008, and 2010 for the study, respectively. In 2004, several patients were also reoperated on because of deep sternal wound infections (n ⫽ 8). In general, there were no differences between the study populations comparing baseline characteristics in different years. In contrast, there were differences regarding some periand postoperative characteristics (Tables 3 and 4). More patients underwent OPCAB in 2004 compared with the following years, and the percentage of patients treated with antifibrinolytic drugs in 2008 and 2010 was much higher than in 2004. Most of the patients treated with antifibrinolytic drugs in 2004 received TA; only a few received aprotinin. All patients treated with antifibrinolytic drugs in 2008 and 2010 received TA. The proportion of patients receiving unwashed autotransfusion was reduced from 72% in 2004 to below 1% in 2010. Median chest tube bleeding was decreased over the years, and the proportion of patients transfused with allogeneic blood products was decreased from 64% to 47%. Unadjusted and adjusted RRs with 95% confidence intervals for receiving ⱖ1 allogeneic blood transfusion according to the year of surgery are shown in Table 1. Overtransfusion with allogeneic RBCs, defined as the proportion of patients transfused with RBC discharged with hemoglobin ⬎7 mmol/L (11.3 g/dL), was reduced from 36% to 16%.
DISCUSSION
This study showed that efforts to change transfusion behaviors and decrease transfusion rates in CABG surgery have persistent effects for several years. However, it is important to recognize that even if increased focus was directed toward a reduction of transfusion rates and the elimination of unnecessary transfusions, transfusion rates are still high, and several unnecessary transfusion are given. Allogeneic RBC transfusion rates between 8% and 100% have been reported during CABG surgery in the past decade.5,12,13 These wide differences in transfusion rates may be explained by a variety of reasons, including different patient populations, differences in procedure-related factors, traditions, and norms. Therefore, individual centers should pay attention to their current transfusion practice. Several studies have shown that behavioral interventions appear effective in changing transfusion practices and reducing blood use,7 but most previous studies only evaluated the early effects. The positive long-term effects in the present study may be a result of continuous efforts to change transfusion practices even if most behavioral interventions took place between 2004 and 2008. RCTs have been performed studying the effect of interventions to reduce bleeding and blood product use.7 However, it was not possible to conduct an RCT to evaluate the effectiveness of combined behavioral interventions applied in the pres-
Table 4. Postoperative Blood Loss and Transfusion Requirements Among Patients Undergoing First-Time Coronary Artery Bypass Surgery in Different Years Variable
Chest tube drainage (mL) 6 hours Total chest tube drainage (mL) Percentage of patients receiving allogeneic transfusion Packed red blood cells (U/patient) Fresh frozen plasma (U/patient) Platelets (U/patient) Percentage of patients autotransfused Percentage of transfused patients (red blood cells), discharged with a hemoglobin concentration of ⬎7 mmo/L (11.3 g/dL)
2004 (n ⫽ 139)
2008 (n ⫽ 147)
2010 (n ⫽ 145)
p Value
560 (375-810)* 950 (690-1,295)* 64.2* 2 (2/3) (n ⫽ 83) 3 (2-4) (n ⫽ 39) 1 (1/2) (n ⫽ 28) 71.5* 35.7
460 (325-620) 720 (550-980) 55.8 2.5 (2/5) (n ⫽ 70) 4 (2/4) (n ⫽ 45) 2 (1/3) (n ⫽ 48) 6.1 21.4
440 (310-620) 750 (550-950) 46.9 2 (2/3) (n ⫽ 54) 2 (2/4) (n ⫽ 24) 2 (2/2) (n ⫽ 35) 0.7 16.1
⬍0.01 ⬍0.01 0.08 0.06 0.06 0.01 ⬍0.01 0.01
NOTE. Data are presented as the median (interquartile range) or as %. *Two missing.
TRANSFUSION PRACTICE BEHAVIOR
549
ent study, and, therefore, a before-and-after study design was used. There are several methodologic considerations of which to be aware.8 It was a strength of the present study that selection bias was reduced by studying consecutive and comparable patient populations identified through only one hospital administrative system. Comparable timeframes and time periods in the different years were chosen, and data were retrieved from the patient records using identical patient case report forms for every year although electronic data not entered in the database during 2004 were available regarding patients who underwent surgery in 2008 and 2010. Generalizability of the study was maximized because the authors selected a representative sample of the elective, first-time CABG population. It was also a strength of the study that the authors focused on long-term effects of behavioral interventions because initial improvement in performance caused by an increased focus of observing transfusion practices may bias early results. There were some methodologic limitations of the study, including all the risks of bias associated with the before-andafter study design. Although the authors stated that surgical routines, anesthetic procedures, and standard normothermic or slightly hypothermic CPB procedures mainly were unchanged during the study period, minor differences (eg, different staff members) may have had an impact on transfusion rates. Furthermore, residual confounding results may be present even if the authors adjusted for several well-known variables associated with bleeding and RBC transfusion during CABG surgery. Discontinuing antithrombotic and antiplatelet drugs in patients with stable angina 3 to 7 days before elective or urgent surgery generally is recommended to decrease the risk of minor and major bleeding events and transfusion.10,14 In the present study, the
authors recorded if the patients discontinued antiplatelet drugs 7 days before surgery, but patients discontinuing antiplatelet drugs 3 to 5 days before also may have benefited from this. There is no information about patient compliance in relation to a decision regarding the preoperative intake of antiplatelet drugs. Several techniques for blood conservation were used to optimize perioperative blood transfusion in the authors’ hospital during this study period, but several other techniques may also be used3 including the use of a protocol for intraoperative blood transfusion. Protamine administration also may be guided by the results from the thromboelastometry, but in the authors’ hospital activated coagulation time values were most often available before the results from thromboelastometry. During the study period, there was an increased use of platelet transfusions. There is no obvious reason for this, but it may be a result of an increased awareness of the risk of bleeding in patients taking antiplatelet drugs preoperatively. The authors are unable to evaluate the effectiveness of the different individual interventions used during the study period, and RCTs are needed to evaluate different strategies to improve transfusion practices. The authors did not have the opportunity to evaluate to which degree intraoperative blood loss led to requirements of RBC transfusion. In conclusion, efforts to change transfusion behaviors and decrease transfusion rates in CABG surgery have persistent effects for several years, but more still needs to be done because the transfusion rate is still high in the authors’ center and inappropriate allogeneic RBC transfusions still are common in different clinical scenarios.15
REFERENCES 1. Kuduvalli M, Oo AY, Newall N, et al. Effect of peri-operative red blood cell transfusion on 30-day and 1-year mortality following coronary artery bypass surgery. Eur J Cardiothorac Surg 27:592-598, 2005 2. Andreasen JJ, Dethlefsen C, Modrau IS, et al: Storage time of allogeneic red blood cells is associated with risk of severe postoperative infection after coronary artery bypass grafting. Eur J Cardiothorac Surg 39:329-334, 2011 3. Society of Thoracic Surgeons, Blood Conservation Guideline Task Force, Ferraris VA, Ferraris SP, et al: Perioperative blood transfusion and blood conservation in cardiac surgery: The Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists clinical practice guideline. Ann Thorac Surg 83:S27-S86, 2007 4. Society of Thoracic Surgeons, Blood Conservation Guideline Task Force, Ferraris VA, Brown JR, et al: Update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists blood conservation clinical practice guidelines. Ann Thorac Surg 91: 944-982, 2011 5. Rogers MA, Blumberg N, Saint S, et al: Hospital variation in transfusion and infection after cardiac surgery: A cohort study. BMC Med 7:37-37, 2009 6. Andreasen JJ, Westen M, Pallesen PA, et al: Transfusion practice in coronary artery bypass surgery in Denmark: A multicenter audit. Interact Cardiovasc Thorac Surg 6:623-627, 2007 7. Tinmouth A, Macdougall L, Fergusson D, et al: Reducing the amount of blood transfused: A systematic review of behavioral interventions to change physicians’ transfusion practices. Arch Intern Med 165:845-852, 2005
8. Fergusson D, Hebert P, Shapiro S: The before/after study design in transfusion medicine: Methodologic considerations. Transfus Med Rev 16:296-303, 2002 9. Cannon CP, Mehta SR, Aranki SF: Balancing the benefit and risk of oral antiplatelet agents in coronary artery bypass surgery. Ann Thorac Surg 80:768-779, 2005 10. Dunning J, Versteegh M, Fabbri A, et al: Guideline on antiplatelet and anticoagulation management in cardiac surgery. Eur J Cardiothorac Surg 34:73-92, 2008 11. McNutt LA, Wu C, Xue X, et al: Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 157:940-943, 2003 12. Snyder-Ramos SA, Möhnle P, Weng YS, et al: The ongoing variability in blood transfusion practices in cardiac surgery. Transfusion 48:1284-1299, 2008 13. Légaré JF, Buth KJ, King S, et al: Coronary bypass surgery performed off pump does not result in lower in-hospital morbidity than coronary artery bypass grafting performed on pump. Circulation 109: 887-892, 2004 14. Ferraris VA, Ferraris SP, Moliterno DJ, et al: Aspirin and other antiplatelet agents during operative coronary revascularization (executive summary). Ann Thorac Surg 79:1454-1461, 2005 15. Shander A, Fink A, Javidroozi M, et al: Appropriateness of allogeneic red blood cell transfusion: The international consensus conference on transfusion outcomes. Transfus Med Rev 25:232-246.e53, 2011
Efforts to Change Transfusion Practice and Reduce Transfusion Rates Are Effective in Coronary Artery Bypass Surgery Jan Jesper Andreasen, MD, PhD,* Jesper Eske Sindby, MD,* Barbara Cristina Brocki,* Bodil Steen Rasmussen, MD, PhD,† and Claus Dethlefsen, MSc, PhD* Objectives: Efforts to decrease allogeneic blood transfusion and avoid unnecessary transfusions in cardiac surgery are important because transfusions are associated with increased postoperative morbidity and mortality. The purpose of the present study was to evaluate the long-term effects of multidisciplinary efforts to reduce allogeneic blood transfusion rates and avoid unnecessary red blood cell (RBC) transfusions in primary elective coronary artery bypass graft (CABG) surgery. Design: A retrospective observational study. Setting: A single center study in a university-affiliated hospital. Participants: A total of 450 patients undergoing primary elective CABG surgery during 2004, 2008, or 2010. Interventions: The application of systematic multimodal perioperative blood-sparing techniques and interventions directed to change transfusion behaviors.
Measurements and Main Results: The results from an audit on transfusion practices in 2004 were compared with similar audits performed in 2008 and 2010 using a before-and-after study design. The patient populations were comparable throughout the years. The median postoperative chest tube bleeding was decreased from 950 mL in 2004 to 750 mL in 2010. The proportion of patients transfused with allogeneic blood products was decreased from 64% to 47%. Overtransfusion with allogeneic RBCs defined as the proportion of patients transfused with RBCs discharged with hemoglobin >7 mmol/L (11.3 g/dL) was reduced from 36% to 16%. Conclusions: Multimodal efforts to change transfusion behaviors and decrease transfusion rates in CABG surgery have persistent effects for several years. © 2012 Elsevier Inc. All rights reserved.
A
risks of RBC transfusions) to reduce allogeneic blood transfusion rates and avoid unnecessary RBC transfusions in elective primary CABG surgery.
LLOGENEIC RED BLOOD CELL (RBC) transfusion has been associated with increased morbidity and mortality after cardiac surgery in several observational studies.1,2 Therefore, increasing attention is being paid toward the use of blood-sparing techniques and transfusion behaviors. Guidelines on transfusion practices in cardiothoracic surgery have been published,3,4 but there still is a considerable variation in transfusion practices nationally and internationally.5,6 A study of transfusion practices during elective coronary artery bypass graft (CABG) surgery previously was performed in Denmark to better understand the factors that determined the use of blood products during CABG surgery.6 This initial audit showed, for example, that several inappropriate RBC transfusions were given, and the audit created a basis for educational efforts to standardize transfusion practices. Several behavioral interventions appear effective in changing transfusion practices and reducing blood use.7 However, more information on the long-term effects of such interventions is needed because most previous studies only evaluated the early effects. The purpose of the present study was to evaluate the long-term effects of multidisciplinary efforts (eg, the routine perioperative use of tranexamic acid [TA] in all patients, the implementation of a guideline for the preoperative discontinuation of antiplatelet drugs, the implementation of a specific algorithm for the management of postoperative bleeding including transfusion, the use of thromboelastometry, the introduction of a new hospital form outlining general transfusion guidelines, educational efforts, and continued reminders regarding the effectiveness and
KEY WORDS: blood transfusion, coronary artery bypass graft surgery, red blood cells
METHODS This study was approved by the Danish Data Protection Agency (record number 1-16-02-1-08/017). A retrospective review of 150 consecutive medical records from each of the years 2008 and 2010 focusing on transfusion practices among patients undergoing primary elective CABG surgery was performed. Results were compared with the results from a similar audit also performed in the North Denmark Region in 2004 using a before/after study design.8 The patients were identified through the local hospital discharge register starting at the beginning of each year. In case of missing patient records, the record of the next patient in line was obtained.
From the Departments of *Cardiothoracic Surgery and †Anesthesiology, Center for Cardiovascular Research, Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark. Supported by a grant from Vestdansk Sundhedsvidenskabeligt Forskningsforums konsulenttjeneste. Presented in part at the Annual Meeting of the European Association of Cardiothoracic Anesthesiologists (EACTA) 2011, Vienna, Austria, June 1-4, 2011. Address reprint requests to Jan Jesper Andreasen, MD, PhD, Department of Cardiothoracic Surgery, Aalborg Hospital, Aarhus University Hospital, Hobrovej 18-22, DK-9100 Aalborg, Denmark. E-mail: [email protected] © 2012 Elsevier Inc. All rights reserved. 1053-0770/2604-0002$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.02.006
Journal of Cardiothoracic and Vascular Anesthesia, Vol 26, No 4 (August), 2012: pp 545-549
545
546
ANDREASEN ET AL
Haemoglobin threshold Low-risk patient <4,5 mmol/L
Postoperative bleeding e e
Cardiac tamponade
(<7.3 g/dL)
High-risk patient
4,5-6,0 mmol/L (7.3-9.7 g/dL)
Patients are pragmatically grouped as high- or low-risk patients based on the postoperative occurrence of concomitant diseases, symptoms and signs, eg, tachycardia, hypotension, renal function, pulmonary function, left ejection fraction, age.
Below threshold
ACT
Protamine
mL/h t > 500 Oplagt > 800 mL/2 h > 900 mL/3 h > 1200 mL/4 h
Suspicion of microvascular bleeding
Platelet count and function test
Unneutralized heparin
Red blood cells
“Surgical” bleeding?
Measurements of hemoglobin every hour
< 50 x 109 or reduced function
Platelets
Thromboelastometry
Decreased concentration of plasma clotting factors
Plasma
Fibrinolysis
Tranexamic acid
Normal coagulation
Decreased fibrinogen level
Fibrinogen
Re-exploration
Continuing bleeding and normal coagulation status Fig 1.
An algorithm for the postoperative management of bleeding in cardiac surgery.
A total of 2,340 patients underwent CABG surgery in Denmark without other concomitant cardiac surgery during 2004. Since 2004, the number of cases gradually has been decreasing to approximately 1,800 cases per year. The study was conducted within the population of the North Denmark Region, which comprises approximately 0.6 million persons or 11% of the total population in Denmark. All patients in the North Denmark Region are referred to Aalborg Hospital, Aarhus University Hospital, if they need CABG surgery. The following baseline patient characteristics were collected: sex, age, height, weight, a previous history of chronic obstructive lung disease, and left ventricular ejection fraction. Baseline laboratory characteristics were hemoglobin, creatinine, platelet count, and international normalized ratio (INR). Preoperative treatment with antiplatelet drugs such as acetylsalicylic acid, clopidogrel, and other nonsteroidal anti-inflammatory drugs also was recorded. Recorded perioperative data included the use of prophylactic antifibrinolytic drugs, the use of cardiopulmonary bypass (CPB), the aortic cross-clamp time, the CPB time, and the lowest hemoglobin during CPB. Recorded postoperative data included chest tube drainage 6 hours postoperatively and in total, the use of autotransfusion, the use of allogeneic blood products during the entire hospitalization, and the hemoglobin concentration at discharge. The hemoglobin concentration at discharge was defined as the last measurement of hemoglobin before discharge. The following severe postoperative complications also were recorded: 30-day mortality and reoperation because of bleeding or surgery for other major complications, such as mediastinitis and gastrointestinal complications. To restrict the analyses to patients who were transfused for reasons directly related to the CABG surgery in itself, patients who experienced severe complications were excluded because transfusions in these patients may not have been related directly to peri- and postoperative treatment of the microvascular bleeding after CABG surgery. Jehovah’s witnesses were not included. Surgical routines, anesthetic procedures, and standard normothermic or slightly hypothermic CPB procedures mainly were unchanged dur-
ing the study period. On-pump and off-pump (OPCAB) surgery were performed according to the preferences of the surgeons, who were senior surgeons as well as trainees. The left internal mammary artery (IMA) was used as a standard whenever the anterior descending artery was to be grafted. The right IMA, saphenous veins, and radial arteries were used as additional grafts according to the preferences of the surgeons. During surgery, blood from the operative field was returned to the CPB circuit. At the end of the surgery, the residual blood volume remaining in the CPB circuit was returned unprocessed to the patients. Postoperative autotransfusion of unwashed mediastinal blood was used routinely during 2004 but was abandoned in most patients after 2007 because of published recommendations.3 Preoperative autologous blood donation, acute normovolemic hemodilution, hemofiltration, and prophylactic transfusions of plasma and platelets were not used. Transfusions of allogeneic blood components (ie, RBC, fresh frozen plasma, and platelet concentrates) were given at the discretion of the attending anesthesiologist and/or surgeon. No specific transfusion algorithm was used during 2004. Indications for allogeneic transfusion were based on routine laboratory measurements of INR, the activated coagulation time, platelet counts, fibrinogen levels, hemoglobin, and hematocrit values in addition to the measurements of hemodynamic and physiologic data, the rate of blood loss, and the existence of concomitant diseases. A specific algorithm for the management of postoperative bleeding including transfusion was designed after 2004 (Fig 1). This algorithm included the use of thromboelastometry, which was introduced in the hospital during the study period. The algorithm was implemented for use in the intensive care unit together with educational efforts directed toward doctors and nurses regarding the potential value of thromboelastometry and the risks of allogeneic RBC transfusions. Thromboelastometry was performed in a laboratory 10 minutes away from the intensive care unit, and the results and suggestions for further treatment because of increased microvascular bleeding were given by doctors in the laboratory over the telephone.
TRANSFUSION PRACTICE BEHAVIOR
547
Table 1. The Unadjusted and Adjusted Relative Risk With 95% Confidence Intervals for Allogeneic Blood Transfusion According to the Year of Surgery Among Patients Undergoing Elective FirstTime CABG Surgery Year
Unadjusted RR (95% CI)
Adjusted RR (95% CI)
2004 2008 2010
1.0 (reference) 0.9 (0.7-1.0) 0.7 (0.6-0.9)
1.0 (reference) 0.7 (0.5-0.9) 0.6 (0.5-0.8)
NOTE. RRs were adjusted for sex, age, body mass index, preoperative hemoglobin and platelet count, INR, preoperative use of antiplatelet drugs, use of antifibrinolytic drugs, use of CPB, and time on CPB. Abbreviation: CI, confidence interval.
A guideline for the preoperative discontinuation of antiplatelet drugs based on the literature9,10 was implemented in 2008. At the time when the indication for CABG or OPCAB surgery in each patient was discussed by the local heart team consisting of both interventional cardiologists and cardiac surgeons, a decision was made whether the patient should continue or discontinue antiplatelet drugs. Furthermore, the prophylactic use of tranexamic acid in all patients became a routine. A total of 1.5 g of tranexamic acid (Cyklokapron; Pfizer Consumer Healthcare, Copenhagen, Denmark) were given as an intravenous bolus via a central venous catheter beginning at the induction of the anesthesia followed by a constant infusion of 200 mg/h until an additional 1.5 g was given. The dose regimen was unchanged throughout the study period. Educational efforts based on the results from the audit in 2004 were performed, and continuous reminders were given. Additionally, a new hospital form outlining general transfusion guidelines as recommended internationally2,3 and supported by the Danish National Board of Health was introduced in the hospital. The quality of the medical transfusion efforts was assessed primarily on the basis of the proportion of patients receiving transfusions and the proportion of overtransfusion defined as the proportion of patients transfused with RBCs and discharged with hemoglobin ⬎7 mmol/L (11.3 g/dL). If the hemoglobin concentrations in the blood after the last
transfusion was ⬎7.0 mmol/L (11.3 g/dL), there was probably at least 1 unit of RBCs given at a time when the hemoglobin concentration was ⬎6.5 mmol/L (10.5 g/dL) provided no major changes occurred in the fluid balance of the patient. Overtransfusion was defined as the proportion of patients transfused with RBCs discharged with hemoglobin ⬎7 mmol/L (11.3 g/dL) because there is no evidence supporting any benefit of RBC transfusions in patients without major bleeding if the hemoglobin concentration is ⬎6.5 mmol/L (10.5 g/dL). Descriptive statistics included the mean, standard deviation, median, interquartile range, and the percentage of patients when appropriate. Differences between years were tested for statistical significance by using chi-square analysis, analysis of variance, and Kruskal-Wallis tests in the case of nonnormality. A p value ⬍0.05 is considered significant. A sample size calculation was performed before the study to calculate how many patient records were needed to have an 80% chance of observing a 20% decrease in total chest tube drainage at a significance level of 0.05 compared with the results obtained in 2004. At least 106 patient records were needed. Missing data were excluded. To check whether missing data might play any major role, the authors repeated the analysis in Table 1 using multiple imputation of missing data. Altogether, 83 values were missing and were imputed 10 times. Using Rubin’s rules, the results were combined. Poisson regression with robust variance estimation was used to compare the use of allogeneic blood transfusion in different years. Specifically, the authors computed relative risks (RRs) with 95% confidence intervals for the requirement of ⱖ1 allogeneic blood transfusion. Poisson regression was used because, in contrast to logistic regression, it may provide more accurate estimates of adjusted RRs in cohorts with common outcomes.11 RRs were adjusted for age and sex of the patients, body mass index, preoperative hemoglobin and platelet count, INR, preoperative use of antiplatelet drugs, use of antifibrinolytic drugs, use of CPB, and CPB time. All analyses were performed out using Stata version 11.1 (Stata Corp; College Station, TX). RESULTS
Baseline characteristics of included patients undergoing isolated, elective, first-time CABG surgery during different years are shown in Table 2. Patients who died within 30 days and patients who underwent surgery because of complications,
Table 2. Baseline Characteristics of Patients Undergoing First-Time Coronary Artery Bypass Surgery in Different Years Variable
2004 (n ⫽ 139)
2008 (n ⫽ 147)
2010 (n ⫽ 145)
p Value
Male sex (%) Age (y) Height (cm) Weight (kg) Body mass index (kg/m2) Hemoglobin (mmol/L) Hemoglobin (g/dL) Creatinine (mmol/L) Platelet count (109/L) INR Ejection fraction (%) COPD (%) Preoperative use of acetylsalicylic acid (%)† Preoperative use of clopidogrel (%)† Patients without preoperative intake of antiplatelets (%)
81.3 65.8 ⫾ 8.6 172.2 ⫾ 8.8 80.3 ⫾ 14.9* 27.0 ⫾ 4.0* 8.5 ⫾ 1.0 13.7 ⫾ 1.6 99.2 ⫾ 62.3 253.7 ⫾ 80.8 1.0 ⫾ 0.2 57.5 ⫾ 15.4 5.8 54.0 27.1 46.0
76.2 67.1 ⫾ 9.4 171.8 ⫾ 8.8 81.7 ⫾ 14.4 27.7 ⫾ 4.2 8.3 ⫾ 0.9 13.4 ⫾ 1.5 93.5 ⫾ 39.4 287.7 ⫾ 89.9 1.0 ⫾ 0.1 50.3 ⫾ 12.3 8.8 63.3 26.5 36.7
78.2 67.3 ⫾ 9.5 172.1 ⫾ 8.5 79.8 ⫾ 14.8 26.8 ⫾ 3.8 8.4 ⫾ 0.9 13.5 ⫾ 1.5 79.4 ⫾ 20.3 273.5 ⫾ 90.8 1.0 ⫾ 0.2 50.2 ⫾ 12.7 11.7 57.9 13.8 42.1
0.54 0.36 0.93 0.50 0.17 0.38 0.38 ⬍0.01 ⬍0.01 0.69 ⬍0.01 0.21 0.28 0.01 0.28
NOTE. Data are presented as the mean ⫾ standard deviation or percentage of patients. Abbreviation: COPD, chronic obstructive pulmonary disease. *One missing. †Intake within 7 days before surgery.
548
ANDREASEN ET AL
Table 3. Operative Characteristics of Patients Undergoing Elective First-Time Coronary Artery Bypass Surgery in Different Years 2004 (n ⫽ 139)
2008 (n ⫽ 147)
2010 (n ⫽ 145)
p Value
54.0 47.8* 104.05 ⫾ 32.9 56.0 ⫾ 19.7 3.3 ⫾ 1.0 4.9 ⫾ 0.6 (mmol/L) 7.9 ⫾ 1.0 (g/dL)
96.6 99.3 76.0 ⫾ 27.5 40.8 ⫾ 15.6 3.2 ⫾ 0.8 4.8 ⫾ 0.8 (mmol/L) 7.7 ⫾ 1.3 (g/dL)
80.0 100.0 85.9 ⫾ 35.3 42.4 ⫾ 21.3 3.1 ⫾ 0.9 5.1 ⫾ 0.9 (mmol/L) 8.2 ⫾ 1.5 (g/dL)
⬍0.01 ⬍0.01 ⬍0.01 ⬍0.01 0.39 0.01 0.01
Variable
Percentage of patients operated with the use of CPB Percentage of patients treated with antifibrinolytic drugs CPB time (min) Cross-clamp time (min) Number of distal anastomoses Lowest preoperative hemoglobin on CPB
NOTE. Data are presented as the mean ⫾ standard deviation or as %. *Three missing.
which were mainly reoperation because of surgical bleeding (n ⫽ 3, 3, and 8 in the years 2004, 2008, and 2010, respectively), were excluded from the analyses leaving 139, 147, and 145 patients who underwent surgery in 2004, 2008, and 2010 for the study, respectively. In 2004, several patients were also reoperated on because of deep sternal wound infections (n ⫽ 8). In general, there were no differences between the study populations comparing baseline characteristics in different years. In contrast, there were differences regarding some periand postoperative characteristics (Tables 3 and 4). More patients underwent OPCAB in 2004 compared with the following years, and the percentage of patients treated with antifibrinolytic drugs in 2008 and 2010 was much higher than in 2004. Most of the patients treated with antifibrinolytic drugs in 2004 received TA; only a few received aprotinin. All patients treated with antifibrinolytic drugs in 2008 and 2010 received TA. The proportion of patients receiving unwashed autotransfusion was reduced from 72% in 2004 to below 1% in 2010. Median chest tube bleeding was decreased over the years, and the proportion of patients transfused with allogeneic blood products was decreased from 64% to 47%. Unadjusted and adjusted RRs with 95% confidence intervals for receiving ⱖ1 allogeneic blood transfusion according to the year of surgery are shown in Table 1. Overtransfusion with allogeneic RBCs, defined as the proportion of patients transfused with RBC discharged with hemoglobin ⬎7 mmol/L (11.3 g/dL), was reduced from 36% to 16%.
DISCUSSION
This study showed that efforts to change transfusion behaviors and decrease transfusion rates in CABG surgery have persistent effects for several years. However, it is important to recognize that even if increased focus was directed toward a reduction of transfusion rates and the elimination of unnecessary transfusions, transfusion rates are still high, and several unnecessary transfusion are given. Allogeneic RBC transfusion rates between 8% and 100% have been reported during CABG surgery in the past decade.5,12,13 These wide differences in transfusion rates may be explained by a variety of reasons, including different patient populations, differences in procedure-related factors, traditions, and norms. Therefore, individual centers should pay attention to their current transfusion practice. Several studies have shown that behavioral interventions appear effective in changing transfusion practices and reducing blood use,7 but most previous studies only evaluated the early effects. The positive long-term effects in the present study may be a result of continuous efforts to change transfusion practices even if most behavioral interventions took place between 2004 and 2008. RCTs have been performed studying the effect of interventions to reduce bleeding and blood product use.7 However, it was not possible to conduct an RCT to evaluate the effectiveness of combined behavioral interventions applied in the pres-
Table 4. Postoperative Blood Loss and Transfusion Requirements Among Patients Undergoing First-Time Coronary Artery Bypass Surgery in Different Years Variable
Chest tube drainage (mL) 6 hours Total chest tube drainage (mL) Percentage of patients receiving allogeneic transfusion Packed red blood cells (U/patient) Fresh frozen plasma (U/patient) Platelets (U/patient) Percentage of patients autotransfused Percentage of transfused patients (red blood cells), discharged with a hemoglobin concentration of ⬎7 mmo/L (11.3 g/dL)
2004 (n ⫽ 139)
2008 (n ⫽ 147)
2010 (n ⫽ 145)
p Value
560 (375-810)* 950 (690-1,295)* 64.2* 2 (2/3) (n ⫽ 83) 3 (2-4) (n ⫽ 39) 1 (1/2) (n ⫽ 28) 71.5* 35.7
460 (325-620) 720 (550-980) 55.8 2.5 (2/5) (n ⫽ 70) 4 (2/4) (n ⫽ 45) 2 (1/3) (n ⫽ 48) 6.1 21.4
440 (310-620) 750 (550-950) 46.9 2 (2/3) (n ⫽ 54) 2 (2/4) (n ⫽ 24) 2 (2/2) (n ⫽ 35) 0.7 16.1
⬍0.01 ⬍0.01 0.08 0.06 0.06 0.01 ⬍0.01 0.01
NOTE. Data are presented as the median (interquartile range) or as %. *Two missing.
TRANSFUSION PRACTICE BEHAVIOR
549
ent study, and, therefore, a before-and-after study design was used. There are several methodologic considerations of which to be aware.8 It was a strength of the present study that selection bias was reduced by studying consecutive and comparable patient populations identified through only one hospital administrative system. Comparable timeframes and time periods in the different years were chosen, and data were retrieved from the patient records using identical patient case report forms for every year although electronic data not entered in the database during 2004 were available regarding patients who underwent surgery in 2008 and 2010. Generalizability of the study was maximized because the authors selected a representative sample of the elective, first-time CABG population. It was also a strength of the study that the authors focused on long-term effects of behavioral interventions because initial improvement in performance caused by an increased focus of observing transfusion practices may bias early results. There were some methodologic limitations of the study, including all the risks of bias associated with the before-andafter study design. Although the authors stated that surgical routines, anesthetic procedures, and standard normothermic or slightly hypothermic CPB procedures mainly were unchanged during the study period, minor differences (eg, different staff members) may have had an impact on transfusion rates. Furthermore, residual confounding results may be present even if the authors adjusted for several well-known variables associated with bleeding and RBC transfusion during CABG surgery. Discontinuing antithrombotic and antiplatelet drugs in patients with stable angina 3 to 7 days before elective or urgent surgery generally is recommended to decrease the risk of minor and major bleeding events and transfusion.10,14 In the present study, the
authors recorded if the patients discontinued antiplatelet drugs 7 days before surgery, but patients discontinuing antiplatelet drugs 3 to 5 days before also may have benefited from this. There is no information about patient compliance in relation to a decision regarding the preoperative intake of antiplatelet drugs. Several techniques for blood conservation were used to optimize perioperative blood transfusion in the authors’ hospital during this study period, but several other techniques may also be used3 including the use of a protocol for intraoperative blood transfusion. Protamine administration also may be guided by the results from the thromboelastometry, but in the authors’ hospital activated coagulation time values were most often available before the results from thromboelastometry. During the study period, there was an increased use of platelet transfusions. There is no obvious reason for this, but it may be a result of an increased awareness of the risk of bleeding in patients taking antiplatelet drugs preoperatively. The authors are unable to evaluate the effectiveness of the different individual interventions used during the study period, and RCTs are needed to evaluate different strategies to improve transfusion practices. The authors did not have the opportunity to evaluate to which degree intraoperative blood loss led to requirements of RBC transfusion. In conclusion, efforts to change transfusion behaviors and decrease transfusion rates in CABG surgery have persistent effects for several years, but more still needs to be done because the transfusion rate is still high in the authors’ center and inappropriate allogeneic RBC transfusions still are common in different clinical scenarios.15
REFERENCES 1. Kuduvalli M, Oo AY, Newall N, et al. Effect of peri-operative red blood cell transfusion on 30-day and 1-year mortality following coronary artery bypass surgery. Eur J Cardiothorac Surg 27:592-598, 2005 2. Andreasen JJ, Dethlefsen C, Modrau IS, et al: Storage time of allogeneic red blood cells is associated with risk of severe postoperative infection after coronary artery bypass grafting. Eur J Cardiothorac Surg 39:329-334, 2011 3. Society of Thoracic Surgeons, Blood Conservation Guideline Task Force, Ferraris VA, Ferraris SP, et al: Perioperative blood transfusion and blood conservation in cardiac surgery: The Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists clinical practice guideline. Ann Thorac Surg 83:S27-S86, 2007 4. Society of Thoracic Surgeons, Blood Conservation Guideline Task Force, Ferraris VA, Brown JR, et al: Update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists blood conservation clinical practice guidelines. Ann Thorac Surg 91: 944-982, 2011 5. Rogers MA, Blumberg N, Saint S, et al: Hospital variation in transfusion and infection after cardiac surgery: A cohort study. BMC Med 7:37-37, 2009 6. Andreasen JJ, Westen M, Pallesen PA, et al: Transfusion practice in coronary artery bypass surgery in Denmark: A multicenter audit. Interact Cardiovasc Thorac Surg 6:623-627, 2007 7. Tinmouth A, Macdougall L, Fergusson D, et al: Reducing the amount of blood transfused: A systematic review of behavioral interventions to change physicians’ transfusion practices. Arch Intern Med 165:845-852, 2005
8. Fergusson D, Hebert P, Shapiro S: The before/after study design in transfusion medicine: Methodologic considerations. Transfus Med Rev 16:296-303, 2002 9. Cannon CP, Mehta SR, Aranki SF: Balancing the benefit and risk of oral antiplatelet agents in coronary artery bypass surgery. Ann Thorac Surg 80:768-779, 2005 10. Dunning J, Versteegh M, Fabbri A, et al: Guideline on antiplatelet and anticoagulation management in cardiac surgery. Eur J Cardiothorac Surg 34:73-92, 2008 11. McNutt LA, Wu C, Xue X, et al: Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 157:940-943, 2003 12. Snyder-Ramos SA, Möhnle P, Weng YS, et al: The ongoing variability in blood transfusion practices in cardiac surgery. Transfusion 48:1284-1299, 2008 13. Légaré JF, Buth KJ, King S, et al: Coronary bypass surgery performed off pump does not result in lower in-hospital morbidity than coronary artery bypass grafting performed on pump. Circulation 109: 887-892, 2004 14. Ferraris VA, Ferraris SP, Moliterno DJ, et al: Aspirin and other antiplatelet agents during operative coronary revascularization (executive summary). Ann Thorac Surg 79:1454-1461, 2005 15. Shander A, Fink A, Javidroozi M, et al: Appropriateness of allogeneic red blood cell transfusion: The international consensus conference on transfusion outcomes. Transfus Med Rev 25:232-246.e53, 2011