ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Rakesh C. Arora
Jean-Francois Légaré
John A. Sullivan
Gregory M. Hirsch
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Arora, R. C.
Right arrow Articles by Hirsch, G. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arora, R. C.
Right arrow Articles by Hirsch, G. M.
Related Collections
Right arrow Coronary disease

Ann Thorac Surg 2004;78:1547-1554
© 2004 The Society of Thoracic Surgeons

Original article: cardiovascular

Identifying Patients at Risk of Intraoperative and Postoperative Transfusion in Isolated CABG: Toward Selective Conservation Strategies

Rakesh C. Arora, MD, PhDa, Jean-Francois Légaré, MDa, Karen J. Buth, MSa, John A. Sullivan, MDa, Gregory M. Hirsch, MDa,*

a Division of Cardiac Surgery, Dalhousie University, Halifax, Nova Scotia, Canada

Accepted for publication April 27, 2004.

* Address reprint requests to Dr Hirsch, New Halifax Infirmary, QEII HSC, Division of Cardiac Surgery, 1796 Summer Room 2006, Halifax, Nova Scotia, B3H 3A7, Canada
ghirsch{at}dal.ca


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results
 Comment
 References
 
BACKGROUND: Allogeneic blood product use during cardiac operation is often reported to exceed 40% despite published guidelines and costly blood conservation strategies. We developed a predictive model, based on eight preoperative risk factors, of allogeneic blood product transfusion rates in patients undergoing a cardiac procedure.

METHODS: All 3,046 consecutive, isolated coronary artery bypass graft (CABG) procedures at a university hospital from 1995 to 1998 were included. A logistic regression model was created to identify independent predictors of allogeneic blood product transfusion. This model was validated using a prospective patient sample.

RESULTS: Overall use of allogeneic blood products was 23% with a crude operative mortality of 2.1%. In isolated, elective, first-time CABG cases, 16.9% received allogeneic blood products. Independent predictors of blood product usage in CABG patients were preoperative hemoglobin 12.0 or less, emergent operation, renal failure, female sex, age 70 years or older, left ventricular ejection fraction 0.40 or less, redo procedure, and low body surface area. Prospective validation of this model on 2,117 consecutive isolated CABG patients demonstrated an observed-to-expected allogeneic blood product transfusion rate ratio of 1.06.

CONCLUSIONS: This internally validated logistic regression risk model is a sensitive and specific predictor of allogeneic blood product use in patients undergoing isolated CABG. Utilization of this model allows for preoperative risk stratification and may allow for more rational resource allocation of costly blood conservation strategies and blood bank resources.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results
 Comment
 References
 
The use of blood products after cardiac surgical interventions varies from 10% to 100% despite published transfusion guidelines [1–7]. Many factors have been proposed to explain this broad range of transfusion rates, including variation in individual physician practices, institutional policy, and patient comorbidities that will inform physicians about tolerable levels of anemia [6–8]. The substantial rate of allogeneic blood product utilization in cardiac operations, accounting for 10% of all blood products collected in the United States, raises concerns of associated morbidity such as hemolytic or allergic reactions, infections (human immunodeficiency virus, cytomegalovirus, hepatitis) [9], graft-versus-host disease [10] and an increased incidence of postoperative infections [3, 11] Taken together, these complications represent substantial human and financial costs [12–14].

In efforts to reduce perioperative blood product use, several effective blood conservation techniques have been developed. These include the use of nonhemic prime for the cardiopulmonary bypass (CPB) machine [15], salvage of blood from the surgical field with cardiotomy suction [16], hemodilution during CPB, retransfusion of all contents of the oxygenator at the end of CPB, acceptance of perioperative normovolemic anemia [17–19], and the routine use of antifibrinolytic medications. Additional techniques such as supplemental mechanical devices (ie, cell-saving systems), autotransfusion of shed mediastinal chest tube drainage, use of modified ultrafiltration, and autologous blood donation programs have also been demonstrated to reduce allogeneic blood product transfusion [2, 16, 20–22]. However, the cost effectiveness of the indiscriminate application of these expensive strategies for patients undergoing isolated coronary artery bypass grafting (CABG) procedures has been questioned [7, 16, 23]. To allow for appropriate allocation of blood bank resources, a preoperative determination of which patients are at high risk of transfusion is desirable, and by inference would merit the expense associated with these conservation strategies.

In the present study, we have retrospectively reviewed the perioperative use of blood products in all patients undergoing an isolated CABG procedure from 1995 through 1998 in a single academic institution. On analysis of this retrospective data, we sought to develop a preoperative risk model capable of predicting the perioperative use of allogeneic blood products and subsequently validate this model on a separate group of patients to guide strategies of blood conservation.


   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results
 Comment
 References
 
Data Source
The Maritime Heart Center Database is a prospectively collected database containing relevant patient demographic data, comorbidities, intraoperative variables, and postoperative outcomes at a large Canadian Centre. These data are captured for all cases undergoing a cardiac procedure at the QEII Health Sciences Centre and undergoes an internal audit annually. The data of 3,046 consecutive patients undergoing isolated CABG between March 1995 and December 1998 were analyzed retrospectively. Comorbid variables (diabetes mellitus, hypertension, hypercholesterolemia, preexisting renal failure), New York Heart Association (NYHA) class, ejection fraction (EF), redo operations, preoperative intravenous nitrate usage, and procedure urgency status were evaluated, and intraoperative and postoperative variables were collected for this study. The use of off-pump CABG (OPCAB) procedures was first initiated in October 1997. Subsequently, all OPCAB cases, from October 1997 to January 1999, were included in the analysis. The need for ethics approval was waived as only encrypted, anonymous patient identifiers were used in the data collection process.

Urgency status was determined using the Maritime Heart Centre Cardiac Surgery database definitions. An elective case was defined as any patient who remained clinically stable for more than 24 hours before the procedure. The term "in-house" elective referred to inpatients receiving medical treatment for more than 48 hours, pending surgery. Many of these patients were receiving intravenous heparin or nitrates until the time of operation. An urgent classification denoted patients who required a procedure to be performed within 24 hours to prevent further clinical deterioration, and an emergent/emergent salvage patient required an immediate operation. Body surface area (BSA) was calculated for each patient. Groupings for BSA were divided into approximate quartiles as follows: >2.04, >1.98 to ≤2.08, >1.86 to ≤1.90, and ≤1.80, as described previously [24]. All procedures were performed at a single tertiary care institution with university affiliation.

Blood Conservation Strategies
The techniques used at the study center included the use of nonhemic prime for the CPB machine [15], salvage of blood from the surgical field with cardiotomy suction [16], hemodilution during CPB, retransfusion of all contents of the oxygenator at the end of CPB, and acceptance of perioperative normovolemic anemia [17–19]. The use of antifibrinolytic medications, either epsilon aminocaproic acid or tranexamic acid, was routine in all cases except for patients with a known sensitivity to the agent. Aprotinin was used in less than 1% of cases, typically reserved for Jehovah Witness patients and/or selected cases considered at an increased risk of perioperative bleeding (e.g., complicated redo procedures and/or known coagulopathy not including patients on warfarin). The decision to use aprotinin was usually a joint determination between the individual surgeon and cardiac anesthetist. Before the discontinuation of epsilon aminocaproic acid we infused an initial 100 to 150 mg/kg bolus followed by continuous infusion of 10 mg/kg per hour until the termination of the case. For tranexamic acid, our usual dose was a bolus of 10 mg/kg and then an infusion of 1 mg/kg per hour. Aprotinin was given as a full Hammersmith dose with 2 million units on induction with an additional 2 million units in the CPB prime, followed by 500,000 U per hour until the administration of protamine. An intraoperative cell-saving device was used only occasionally, again, in complicated reoperative and in OPCAB procedures. Postoperatively nonhemic volume expanders were used routinely (ie, crystalloid and pentastarch solutions).

Transfusion Triggers
The need for perioperative blood product transfusion was determined on an individual, patient-by-patient basis by 1 of the 7 attending cardiovascular surgeons or supporting resident house staff. Overall transfusion rates of allogeneic blood products were captured for blood products given during and after the operation. The principal blood product used was packed red blood cells (pRBCs). The decision to use blood products other than pRBCs (ie, platelets, fresh frozen plasma, and cryoprecipitate) was made on a case-by-case basis for patients with evidence of ongoing blood loss or a demonstrable coagulopathy. Preoperative transfusions were not included in this analysis.

Although no rigid criterion or single absolute hemoglobin (Hb) value was used as a trigger for transfusion, in general, allogeneic blood product transfusion was not considered until the serum Hb was less than 7 g/dL, unless evidence suggested ongoing blood loss or the patient was clinically considered at risk for inadequate oxygen delivery. This determination included patients with signs of insufficient tissue perfusion (lactate levels > 2, mixed venous oxygen saturation < 65%, urine output < 0.5 mL · kg–1 · h–1), significant hemodynamic instability, the need for two or more inotropic agents, utilization of a intraaortic balloon pump (IABP), or organ dysfunction in two or more systems.

The addition of pRBCs to the CPB machine prime was considered (although not routinely given) if the calculated dilutional hematocrit was less than 0.21. The decision to add pRBCs to the prime was made in collaboration among the surgeon, anesthetist, and perfusionist at the time of the surgical procedure.

Statistics
The primary outcome was the transfusion of any allogeneic blood product (pRBCs, plasma, platelets, or cryoprecipitate) during the intraoperative or postoperative period. Statistical analysis was performed using SAS 8.2 (SAS Institute Inc, Cary, NC). Continuous and discrete variables underwent univariate analysis by t test, {chi}2, or Fisher exact test, as appropriate. Independent preoperative predictors of blood transfusion were identified by logistic regression, and variables were retained if their associated p < 0.05. A C-statistic was calculated as a measure of the sensitivity and specificity of the logistic regression model. The C-statistic is equivalent to the area under the receiver operating characteristic curve in which a C-value of 0.5 indicates no discriminatory power and a C-value of 1.0 indicates a perfect discrimination between patients requiring blood transfusion. Validation of this stepwise logistic model was performed using a bootstrapping procedure. C-statistic values were calculated for each of 200 bootstrap samples, and the 95% confidence interval (CI) was obtained from the 2.5 and 97.5 percentiles of the bootstrap distribution. In a second validation step, the original model (derivation group) was applied to a subsequent group of 2,117 isolated CABG cases (validation group) performed from January 1999 through September 2001.


   Results
Top
 Abstract
 Introduction
 Material and Methods
 Results
Comment
 References
 
Patient Population
Preoperative patient demographics are listed in Table 1 (derivation group). For the 3,046 isolated CABG patients, the mean age was 65.0 ± 10.6 years (range 26 to 93 years) and 35% (n = 1,054) were older than 70 years of age. In this group of patients approximately 73% were male and 93% were NYHA class III or IV. The most common form of anticoagulant used before operation was antiplatelet therapy (84% of patients); 19% (n = 565) were given intravenous nitrates.


View this table:
[in this window]
[in a new window]
  		Table 1. Patient Demographics

 
Reoperative procedures comprised 8% of all cases in the derivation group (Table 2). In the derivation group, the operation was classified as "in-house elective" for 33% (n = 1,011). In our current surgical practice, OPCAB is undertaken for approximately 10% to 15% of our isolated CABG patients; however, when the derivation group data were collected, OPCAB procedures comprised a smaller percentage (n = 83; 3% of derivation group). Therefore, we were unable to derive meaningful statistical analysis of the data as a group separate from the on-pump CABG cases in the derivation group. It is interesting to note, however, as displayed in Figure 1, that OPCAB patients had the lowest rates of transfusion among all groups of isolated CABG patients.


View this table:
[in this window]
[in a new window]
  		Table 2. Operative Variables

 


View larger version (14K):
[in this window]
[in a new window]
  		Fig 1. Rate of allogeneic blood product transfusion in CABG procedures. (CABG = coronary artery bypass grafting; OPCAB = off-pump CABG; redo = reoperative CABG procedure [includes patients who underwent a previous sternotomy for a non-CABG procedure]; 1st time = patient undergoing their first CABG procedure [no previous sternotomy]; urgent = transfusion rate for all patients needing a CABG procedure within 24 hours; emergent/emergent salvage = patients needing an immediate operation or undergoing cardiopulmonary resuscitation en route to the operating theater.)

 
Use of Blood Products
The overall allogeneic transfusion rate for all patients undergoing an isolated CABG procedure (irrespective of urgency) was 23%. This included 83 OPCAB cases, with a rate of blood product utilization of 3.6% (Fig 1). With respect to the frequency of the type of blood product used, 96.2% of all the patients transfused received packed red blood cells (pRBCs), while 15.3% received fresh frozen plasma, 10.2% received platelets, and 1% received cryoprecipitate. Some patients received more than one type of product. Of the patients transfused in the original derivation group, 3.8% received blood products that did not include pRBCs.

The in-hospital mortality for the derivation group was 2.1% (n = 63). Of these 63 patients, 85.7% received at least one allogeneic blood product transfusion versus 21.8% in the remaining patients (p < 0.001). Twenty-three patients (0.76%) required reexploration for postoperative bleeding, and 100% of these patients received a blood product.

Procedure Urgency, Redo Operations, and Intraaortic Balloon Pump Use
Procedure urgency was associated with increased rates of transfusion. Isolated elective ("true elective" and "in-house" elective) CABG procedures were associated with a transfusion rate of 18% versus 40% and 56% for urgent and emergent/emergent salvage procedures, respectively (p < 0.001) (Fig 1). Redo operations accounted for 7.7% of all isolated CABG cases, and elective redo operations (n = 162) had a higher transfusion rate (34%) than elective first-time CABG (17%) (p < 0.001). Approximately 10% of OPCAB procedures (n = 8) were done on an urgent basis, and 6% (n = 5) of OPCAB cases were redo operations.

Only 6.3% (n = 191) of patients required perioperative IABP assistance. Of these patients, 60.2% (n = 115) received an allogeneic transfusion compared with 20.6% transfusion rate in all other patients not requiring IABP use (p < 0.001). The timing of insertion of an IABP also appeared to be related to the risk of transfusion. Patients who required an IABP preoperatively (n = 153) had a transfusion rate of 54.2%, those receiving an IABP intraoperatively (n = 19) required a blood transfusion 73.7% of the time, and nearly all patients (94.7%) requiring a postoperative IABP (n = 19) received a blood transfusion.

Univariate Predictors
Several univariate risk factors were associated significantly with increased rate of blood transfusion; the strongest risk factor was the presence of a low preoperative Hb (≤ 12.0 g/dL) with a univariate odds ratio of 13.1 (Table 3). The mean ± SD preoperative Hb level of patients receiving blood transfusions was 12.0 ± 1.8 g/dL, compared with 14.0 ± 1.4 g/dL for patients not requiring blood transfusions (p < 0.001). In addition, low BSA (< 1.80), female sex, and renal failure were also associated with an increased rate of allogeneic transfusion. However, preoperative Hb was significantly lower in patients with preoperative renal failure (11.3 ± 1.8 g/dL versus 13.7 ± 1.6 g/dL with normal renal function, p < 0.001) and between sexes (12.5 ± 1.5 g/dL and 13.9 ± 1.6 g/dL in female versus male, p < 0.001). Additionally, a greater proportion of women had a preoperative Hb less than 12.0 g/dL (36.3% versus 12.9%, female versus male, p < 0.001). Furthermore, more women were in the lowest quartile of BSA (< 1.80) than men (71.5 versus 15.7 respectively; p < 0.001).


View this table:
[in this window]
[in a new window]
  		Table 3. Univariate Risk Factors for Isolated CABG Cases

 
Multivariable Analysis
The independent risk factors for increased rates of allogeneic blood product utilization are listed in Table 4. These risk factors are similar to predictors reported in previous studies of outcome after CABG operation [1, 2, 7, 25, 26].


View this table:
[in this window]
[in a new window]
  		Table 4. Multivariable Risk Factors for Isolated CABG

 
Predicted probability of blood transfusion was calculated using coefficients from the logistic regression model (Fig 2). The lowest risk factor, BSA < 1.80, carried a risk of 9.1% of transfusion whereas the strongest risk factor, preoperative Hb (< 12.0 g/dL), conferred a risk of 28.1%. The addition of further risk factors had an incremental risk of receiving allogeneic blood products, ultimately reaching 99.5% if all eight risk factors were present in a single patient.



View larger version (10K):
[in this window]
[in a new window]
  		Fig 2. Predictive model of risk of allogeneic blood transfusion. The predicted probability of transfusion was directly related to the number of preoperative risk factors. Curve shown indicates body surface area as a single risk factor with the cumulative probability of allogeneic transfusion with additional preoperative risk factors.

 
Model Validation
For the logistic regression model, the Hosmer-Lemeshow goodness of fit p value was 0.83 and the C-statistic was 0.84. The 95% CI of the C-statistic was calculated as 0.82 to 0.86 by bootstrap procedure. In the derivation group, the predicted allogeneic blood product transfusion rate was 22.8% while the actual observed transfusion rate was 23.1%, with an observed to expected (O:E) ratio of 1.01.

As a second validation step, the model was subsequently applied to a separate validation group of 2,117 consecutive patients (validation group) (January 1999 through September 2001). The validation group was similar to the derivation group except for a higher proportion of patients with diabetes mellitus, hypertension, and hypercholesterolemia (Table 1) and a lower proportion of patients undergoing a redo procedure or urgent status (Table 2). The predicted allogeneic blood product transfusion rate in the validation group was 22.2% while the actual observed transfusion rate was 23.6%, with an observed-to-expected ratio of 1.06. Using the coefficients from the original model, the C-statistic for the validation group was 0.79, with a 95% CI of 0.77 to 0.81 obtained by bootstrap procedure.

Risk Stratification
In using the generated risk model, we were able to subdivide the original cohort (derivation group) of patients into four groups based on the predicted risk of allogeneic transfusion: very low (≤ 10% risk), low (> 10% to 20% risk), intermediate (> 20% to 50% risk), and high (> 50% risk) (Fig 3). When using these cut-off points, most patients in our original series of patients were allocated to the very low risk (39%) or low-risk group (29%), followed by an equal distribution in both the intermediate and high-risk groups (16% in each group). A small percentage of allogeneic transfusion occurred in the very low and low-risk groups, whereas substantially larger percentages of patients were transfused in both the remaining two groups (p < 0.0001).



View larger version (9K):
[in this window]
[in a new window]
  		Fig 3. Risk stratification of patients based on preoperative risk factors and the rates of transfusion for each group. Very low, low, intermediate, and high-risk ({square}) groups refer to the percentage of patients in the original cohort who had a ≤ 10%, > 10% to 20%, > 20% to 50%, and > 50% preoperative risk of transfusion. The black bars ({blacksquare}) refer to distribution of patients who received an allogeneic transfusion. (y-axis = percentage of patients in derivation group.)

 

   Comment
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
References
 
Nearly 10% of all collected pRBCs are used in patients undergoing cardiac procedures [23, 27, 28]. This relatively high rate of transfusion is costly, and exposes patients undergoing cardiac operation to significant risk of complications related to allogeneic blood products [10, 29, 30]. Many centers continue to report allogeneic blood transfusion rates of more than 50% despite pressures to minimize use of blood products [9, 27, 31]. In our patient population, which is representative of most large cardiac surgical centers in Canada [32, 33], the overall observed transfusion rate of 23% in the derivation group of patients (n = 3046) with an overall mortality of 2.0% compares favorably with other reports in the literature [2, 7, 26, 31, 34, 35]. A significant proportion of these patients were older than 70 years (34.6%), 7.7% were reoperative procedures, and 92.6% were in NYHA functional class III or IV.

The lowest transfusion rates were achieved for elective, first-time isolated CABG procedures, with a rate of 17.3% when using CPB and 2.7% for OPCAB cases (Fig 1) Costly supplemental methods of blood conservation, such as the use of preoperative donation of autologous blood and postoperative autotransfusion of shed mediastinal blood [21], were not used at our institution and therefore not included in this study. These data suggest that indiscriminate implementation of these conservation techniques is not necessary to achieve relatively low overall transfusion rates.

A risk model generated from a retrospective analysis of 3,046 consecutive, isolated CABG patients reliably predicted allogeneic blood product transfusion rates based on eight, easily identifiable, preoperative risk factors. Independent risk factors identified for blood transfusion in isolated CABG patients in this study included preoperative Hb less than 12.07 g/dL, emergent operation, renal failure, female sex, age older than 70 years, EF less than 0.40, redo procedure, and low BSA, are similar to those reported by others [1]. Validation of this model in a separate group of 2,117 CABG patients confirmed its predictive value.

Practical Implementation of the Predictive Model
The safety and cost efficacy of autologous blood donation programs remain controversial [36]. Data derived from a prospective study examining a preoperative autologous donation program for CABG patients demonstrated significant associated cost implications in the order of $508,000 to $909,000 (US dollars) per quality-adjusted year of life saved [31]. Similarly, whereas autotransfusion of shed mediastinal blood may reduce the risk of infection and administrative errors associated with stored blood, several studies, including two prospective, randomized trials, failed to show any significant reduction in the amount of allogeneic transfusion requirements in postoperative cardiac patients [22, 37, 38].

One conclusion to be drawn from these studies is that the cost of implementing these techniques is not justified. An alternate conclusion may be that widespread utilization of these strategies may not be rational; however, selected use in patients most likely to benefit may demonstrate a cost-effective savings. For efficacious delivery, patients who are most likely to benefit from these interventions would need to be identified preoperatively.

Our findings suggest that, based on easily identifiable preoperative patient characteristics, subgroups of patients have a very low, low, intermediate, or high risk of receiving allogeneic blood products (Figs 2, 3). Utilizing the predictive model in this manner, the efficacy of costly blood conservation strategies could be evaluated in a subset of patients, and cost-effectiveness determined.

An example, proposed by others in both limited cardiac and noncardiac patient populations, are potential cost savings by eliminating the need for an obligatory preoperative allogeneic blood cross-match (at an additional cost of Canadian $72.00/patient at our center) in patients identified as "low risk" for transfusion. If we set the criteria for a "low-risk patient" at a predicted risk of transfusion of 20% or less (as suggested by others in noncardiac operation patients [39] (n = 2,058 or 68% of isolated CABG cases in our center), the cost saving per annum of performing a group and screen only (ABO typing and basic antibodies) versus a full cross-match is approximately $148,000 ($Cdn) at our institution. We have, since the close of this study, initiated a "no cross-match" policy in a limited manner for elective low risk patients at our institution. At 6 months (n = 300 patients), we have experienced no morbidity or mortality related to this procedural change.

This model, in contrast to previous studies [1, 34, 35], examined the transfusion rates of all postoperative blood products for a large group of consecutive patients and included both on-pump and off-pump cases, first-time and redo CABG, as well as elective, urgent, and emergent cases. In addition, this robust predictive model was further validated on a separate large group of consecutive patients. Others [25] have suggested that, due to baseline differences, ubiquitously applying a predictive model to all isolated CABG patients may not be appropriate and use of the model should therefore be limited to elective, first-time CABG cases. In contrast, our model included 1,011 patients (33% of the derivation group) who were, as stated "in-house" elective receiving medical treatment (including intravenous nitrates) for more than 48 hours before operation. Inclusion of these patients and subsequent application of our model to a large validation group, using only "basic" blood conservation strategies, demonstrated its generalizability to a more diverse group than purely elective, first-time isolated CABG cases. These results broaden the application of such data to a more representative group of patients in most surgeons' practices. An example of an easy-to-use preoperative "risk-score" card based on our model that could be used to predict the risk of transfusion is illustrated in Figure 4.



View larger version (32K):
[in this window]
[in a new window]
  		Fig 4. Sample score card for predicting the risk of receiving allogeneic blood products to determine if a patient receives a group and screen versus full cross-match and type blood analysis. (BSA = body surface area [m2]; EF = left ventricular ejection fraction; renal failure = serum creatinine > 2 mg/dL.)

 
Limitations
This study, while an assessment of transfusion rates at a single center, provided important information on transfusion practices. We showed that despite having many levels of decision makers (house staff, cardiac surgeons, anesthetists) and a lack of rigid transfusion criterion, the overall transfusion rate was maintained at less than 25% with low mortality in a population with a high premorbid risk [32, 33]. Additionally, the model was validated internally on a large group of consecutive, isolated CABG patients to ensure generalizability to a diverse (both elective and nonelective) group of patients. Validation of our model in another center with similar transfusion and perioperative blood conservation strategies would add further strength to the predictive power of this model as site specificity may also play a role in allogeneic transfusion rates.

Although utilizing the predictive model allows for stratification of risk based on preoperative risk factors, an overlap exists in groups of patients with two or more risk factors (Fig 2). Nevertheless, patients at either extreme of the model have predictable rates of transfusion and therefore are amenable to both further study and cost-saving strategies.

In summary, we evaluated blood transfusion practices after CABG operation. The patient population was representative of a tertiary referral center and the findings are likely applicable to other similar populations found in North American centers [32, 33]. A stable predictive model was generated from a large CABG population in which we could predict preoperatively, on an individual basis, the risk of receiving any blood products after operation. The strategy of introducing a no-cross-match policy is one example of the utility of the model presented in this study. The model provides the framework for a rational approach of appropriately targeting more costly strategies such as autologous blood donation, cell-saving systems, and the use of more expensive fibrinolytics (ie, aprotinin). By eliminating unnecessary expense among low-risk patients and identifying high-risk patients in whom more costly blood conservation strategies could be effective, this predictive model could have a substantial effect on both quality and cost of care.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
 References
 

  1. Cosgrove DM, Loop FD, Lytle BW, et al. Determinants of blood utilization during myocardial revascularization. Ann Thorac Surg. 1985;40:380–384[Abstract]
  2. Breyer RH, Engleman RM, Rousou JA, Lemeshow S. Blood conservation for myocardial revascularization. J Thorac Cardiovasc Surg. 1987;93:512–522[Abstract]
  3. LoCicero J, Massad M, Gandy K, et al. Aggressive blood conservation in coronary artery surgery: impact on patient care. J Cardiovasc Surg. 1990;31:559–563[Medline]
  4. Giordano GF, Rivers SL, Chung GKT, et al. Autologous platelet-rich plasma in cardiac surgery: effect on intraoperative and postoperative transfusion requirements. Ann Thorac Surg. 1988;46:416–419[Abstract]
  5. Hasley PB, Lave JR, Hanusa BH, et al. Variation in the use of red blood cell transfusions. A study of four common medical and surgical conditions. Med Care. 1995;33:1145–1160[Medline]
  6. Welch HG, Meehan KR, Goodnough LT. Prudent strategies for elective red blood cell transfusion. Ann Intern Med. 1992;116:393–402[Medline]
  7. Goodnough LT, Despotis GJ, Hogue CW, Ferguson TB. On the need for improved transfusion indicators in cardiac surgery. Ann Thorac Surg. 1995;60:473–480[Abstract/Free Full Text]
  8. Goodnough LT, Marilyn FM, Shah T, Chernosky A. A two-institution study of transfusion practice in 78 consecutive adult elective open-heart procedures. Am J Clin Pathol. 1989;91:468–472[Medline]
  9. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP. Transfusion medicine. Blood transfusion (first of two parts). N Engl J Med. 1999;340:438–447[Free Full Text]
  10. Thaler M, Shamiss A, Orgad S, et al. The role of blood from HLA-homozygous donors in fatal transfusion-associated graft versus host disease after open heart surgery. N Engl J Med. 1989;321:25–28[Medline]
  11. Chelemer SB, Prato S, Cox PM Jr, O'Connor GT, Morton JR. Association of bacterial infection and red blood cell transfusion after coronary artery bypass surgery. Ann Thorac Surg. 2002;73:138–142[Abstract/Free Full Text]
  12. Goodnough LT, Shuck JM. Review of risk, options, and informed consent for blood transfusion in elective surgery. Am J Surg. 1990;159:602–609[Medline]
  13. Goodnough LT, Soegiarso W, Birkmeyer JD, Welch HG. Economic impact of inappropriate blood transfusion in coronary artery bypass graft surgery. Am J Med. 1993;94:509–513[Medline]
  14. Hebert PC, Wells G, Blajchman M, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999;340:409–417[Abstract/Free Full Text]
  15. Cooley DA, Beall AC Jr, Grondin P. Open-heart operations with disposable oxygenators, 5 per cent dextrose prime, and normothermia. Surgery. 1962;52:713–719[Medline]
  16. Winton TL, Charrette EJ, Salerno TA. The cell saver during cardiac surgery: does it save? Ann Thorac Surg. 1982;33:379–381[Abstract]
  17. Hallowell P, Bland JH, Dalton BC, et al. The effect of hemodilution with albumin or Ringer's lactate on water balance and blood use in open-heart surgery. Ann Thorac Surg. 1978;25:22–29[Abstract]
  18. Diehl JT, Lester JL 3rd, Cosgrove DM. Clinical comparison of hetastarch and albumin in postoperative cardiac patients. Ann Thorac Surg. 1982;34:674–679[Abstract]
  19. Kirklin JK, Lell WA, Kouchoukos NT. Hydroxyethyl starch versus albumin for colloid infusion following cardiopulmonary bypass in patients undergoing myocardial revascularization. Ann Thorac Surg. 1984;37:40–46[Abstract]
  20. Koziol DE, Holland PV, Alling DW, et al. Antibody to hepatitis B core antigen as a paradoxical marker for non-A, non-B hepatitis agents in donated blood. Ann Intern Med. 1986;104:488–495[Abstract/Free Full Text]
  21. Schaff HV, Hauer JM, Bell WR, et al. Autotransfusion of shed mediastinal blood after cardiac surgery, a prospective study. J Thorac Cardiovasc Surg. 1978;75:632–641[Abstract]
  22. Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotransfusion following cardiac operations: a randomized, prospective study. Ann Thorac Surg. 1979;27:500–507[Abstract]
  23. Cosgrove DM, Thurer RL, Lytle BW, Gill CC, Loop FD. Blood conservation during myocardial revascularization. Ann Thorac Surg. 1979;28:184–189[Abstract]
  24. O'Connor GT, Plume SK, Olmstead EM, et al. Multivariate prediction of in-hospital mortality associated with coronary artery bypass graft surgery. Circulation. 1992;85:2110–2118[Abstract/Free Full Text]
  25. Karkouti K, Cohen MM, McCluskey SA, Sher GD. A multivariable model for predicting the need for blood transfusion in patients undergoing first-time elective coronary bypass graft surgery. Transfusion. 2001;41:1193–1203[Medline]
  26. Goodnough LT, Soegiarso W, Geha AS. Blood lost and blood transfused in coronary artery bypass graft operation as implications for blood transfusion and blood conservation strategies. Surg Gynecol Obstet. 1993;177:345–351[Medline]
  27. Goodnough LT, Johnston MF, Toy PT. The variability of transfusion practice in coronary artery bypass surgery. JAMA. 1991;265:86–90[Abstract/Free Full Text]
  28. Scott WJ, Kessler R, Wernly JA. Blood conservation in cardiac surgery. Ann Thorac Surg. 1990;50:541–546
  29. Stevens CE, Aach RD, Hollinger FB, et al. Hepatitis B virus antibody in blood donors and the occurrence of non-A, non-B hepatitis in transfusion recipients. An analysis of the transfusion-transmitted viruses study. Ann Intern Med. 1984;101:733–738[Medline]
  30. Cohen NA, Munoz A, Reitz BA, et al. Transmission of retroviruses by transfusion of screened blood in patients undergoing cardiopulmonary bypass. N Engl J Med. 1989;320:1172–1176[Abstract]
  31. Birkmeyer JD, AuBuchon JP, Littenberg B, et al. Cost-effectiveness of preoperative autologous donation in coronary artery bypass grafting. Ann Thorac Surg. 1994;57:161–169[Abstract]
  32. Ghali WA, Quan H, Brant R. Coronary artery bypass grafting in Canada: national and provincial mortality trends, 1992–1995. CMAJ. 1998;159:25–31[Abstract]
  33. Ghali WA, Quan H, Brant R. CABG in Canada. CMAJ. 1999;161:941[Free Full Text]
  34. McGill N, O'Shaughnessy D, Pickering R, Herbertson M, Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial. BMJ. 2002;324:1–7[Abstract/Free Full Text]
  35. Balachandran S, Cross MH, Karthikeyan S, Mulpur A, Hansbro SD, Hobson P. Retrograde autologous priming of the cardiopulmonary bypass circuit reduces the blood transfusion after coronary artery bypass. Ann Thorac Surg. 2002;73:1912–1918[Abstract/Free Full Text]
  36. Popovsky MA, Whitaker B, Arnold NL. Severe outcomes of allogeneic and autologous blood donation: frequency and characterization. Transfusion. 1995;35:734–737[Medline]
  37. Ward HB, Smith RRA, Landis KP, Nemzek TG, Dalmasso AP, Swaim WR. Prospective, randomized trail of autotransfusion after routine cardiac operations. Ann Thorac Surg. 1993;56:137–141[Abstract]
  38. Roberts S, Early GL, Brown B, Hannah H, MacDonald HL. Autotransfusion of unwashed mediastinal shed blood fails to decrease banked blood requirements in patients undergoing aortocoronary surgery. Am J Surg. 1991;162:477–480[Medline]
  39. van Klei WA, Moons KG, Leyssius AT, Knape JT, Rutten CL, Grobbee DE. A reduction in type and screen: preoperative prediction of RBC transfusions in surgery procedures with intermediate transfusion risks. Br J Anaesth. 2001;87:250–257[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
A. Szekely, Z. Cserep, E. Sapi, T. Breuer, C. A. Nagy, P. Vargha, I. Hartyanszky, A. Szatmari, and A. Treszl
Risks and Predictors of Blood Transfusion in Pediatric Patients Undergoing Open Heart Operations
Ann. Thorac. Surg., January 1, 2009; 87(1): 187 - 197.
[Abstract] [Full Text] [PDF]

Home page
 SEMIN CARDIOTHORAC VASC ANESTHHome page
F. E. Ralley
Programmatic Blood Conservation in Cardiac Surgery
Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2007; 11(4): 242 - 246.
[Abstract] [PDF]

Home page
 CirculationHome page
G. J. Murphy, B. C. Reeves, C. A. Rogers, S. I.A. Rizvi, L. Culliford, and G. D. Angelini
Increased Mortality, Postoperative Morbidity, and Cost After Red Blood Cell Transfusion in Patients Having Cardiac Surgery
Circulation, November 27, 2007; 116(22): 2544 - 2552.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
A. J. Rodrigues, P. R. B. Evora, S. Bassetto, P. M. Luciano, L. Alves Jr, A. S. Filho, and W. V. de Andrade Vicente
Efficacy and Safety of Aprotinin Use for Reoperative Valvular Surgery
Ann. Thorac. Surg., June 1, 2007; 83(6): 2060 - 2065.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Rakesh C. Arora
Jean-Francois Légaré
John A. Sullivan
Gregory M. Hirsch
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Arora, R. C.
Right arrow Articles by Hirsch, G. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arora, R. C.
Right arrow Articles by Hirsch, G. M.
Related Collections
Right arrow Coronary disease

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
ANN THORAC SURG ASIAN CARDIOVASC THORAC ANN EUR J CARDIOTHORAC SURG
J THORAC CARDIOVASC SURG ICVTS ALL CTSNet JOURNALS