| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Providence Regional Medical Center Everett, Everett, Washington
b Advanced Perfusion Care, Inc, Pinehurst, North Carolina
c Everett Cardiovascular and Thoracic Surgical Associates, Everett, Washington
d Medical Data Research Center, Providence Health & Services, Portland, Oregon
Accepted for publication October 16, 2008.
* Address correspondence to Dr Brevig, 1330 Rockefeller Ave, Suite 400, Everett, WA 98206 (Email: james.brevig{at}providence.org).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Methods: A data driven, multidisciplinary effort to decrease allogeneic red blood cell transfusion was instituted in a community hospital. Numerous innovations in treatment protocols were implemented and evaluated. Clinical data from 2003 to 2007 will be presented. Yearly review of outcomes led to an evolving clinical practice and lowered transfusion rates.
Results: A total of 2,531 consecutive cardiac surgical procedures were performed during a five-year period. Using a multidisciplinary approach to quality improvement, and with the goal of using fewer blood products, our incidence of allogeneic red blood cell transfusion was decreased, from 43% in 2003 to 18% in 2007. Patient outcomes were not significantly changed.
Conclusions: Cardiac surgery in a community hospital can be performed safely with low utilization of allogeneic red blood cell transfusions. A multidisciplinary approach to blood conservation can result in lower transfusion rates and equivalent patient outcomes.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Historically, a hemoglobin of 10 g/dL and a hematocrit of 30% were common triggers for RBC transfusion. The Society of Thoracic Surgeon (STS) guidelines suggest for "patients after cardiac operations with hemoglobin levels below 6 g/dL, RBC transfusion is reasonable and can be life-saving. Transfusion of red cells is reasonable in most postoperative patients with hemoglobin levels of 7 g/dL or less, but no high-level evidence supports this recommendation" [3]. Kuduvalli and colleagues [4] suggest a decrease in mortality in patients not transfused. According to Scott [5], transfused cardiac surgery patients have twice the five-year mortality (15% vs 7%) of nontransfused patients even after correction for comorbidities and other factors. Engoren and colleagues [6] state the following: "After correction for comorbidities and other factors, transfusion was still associated with a 70% increase in mortality." Although transfusion seems to be associated with worse clinical outcomes in cardiac surgery, this association does not prove causation.
Carson and colleagues [7] found that anemia is a preoperative risk factor for increased postoperative surgical mortality and morbidity. Kulier and colleagues [8] evaluated the impact that preoperative anemia had when combined with other patient comorbidities. They showed preoperative anemia magnified the adverse effects of other comorbidities and the magnitude of this effect is dependent on the number of units of allogeneic blood products administered. Other database studies have examined outcomes and lowest hematocrit at various times during and after surgery. These lowest hematocrit studies have sent a message that a low hematocrit may be associated with bad outcomes, yet low hematocrit is a marker for transfusion and the authors did not examine transfusion in their analysis [9]. Despite the findings of DeFoe and colleagues [10], that a lowest hematocrit on bypass was significantly associated with in-hospital death, and Ranucci and colleagues [11], that a low hematocrit on bypass increases length of stay and renal failure, that has not been our experience.
There are several cardiac surgery blood conservation and management programs around the country that have been successful [12]. We will demonstrate that our multidisciplinary approach used to minimize transfusion is safe and associated with good patient outcomes.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Clinical data from all cardiac surgeries are submitted to the STS Adult Cardiac Database, the Washington State Clinical Outcomes and Assessment Program (COAP), and the Providence Health & Services (PH&S) Cardiovascular Study Group data set. Definitions of variables used for data collection align with those in the STS database [13]. We began collecting data on blood utilization in cardiac surgery in 1999. These data provided a baseline. Since then we have added metrics intended to help us understand our practices and processes. Interim data analysis was done as new products became available and new practices were identified.
A total of 2,531 consecutive cardiac surgeries from 2003 to 2007 were evaluated. Perioperative RBC transfusion rates (from arrival in the operative room to discharge) are reported by year. Blood products that were transfused after discharge were not tracked. Blood products used included RBCs, platelets, fresh frozen plasma, and cryoprecipitate. However, since 99% of patients who received any blood product received RBCs, the RBC transfusion rate is used throughout this report. Patient characteristics, type of procedure, incidence of RBC transfusion, units of RBCs transfused, and outcomes are reported by operation year.
Statistical Analysis
Dichotomous and categorical variables were presented as percentages and were tested by the 
2 test. Continuous variables were expressed as mean ± SD and one-way analysis of variance was performed to compare the means among groups.
Risk models were adopted for risk-adjusted comparison. Odds ratios [14] were compared among groups. Cumulative sum (CUSUM) analysis [15] was used to show the overall risk-adjusted results.
Alghamdi and colleagues' Transfusion Risk Understanding Scoring Tool [16] was chosen to calculate the predicted risk of transfusion. The predictors included age, gender, weight, preoperative hemoglobin and creatinine, reoperation, urgency of operation, and type of operation. Because only the preoperative hematocrit was available in our data, the formula of hematocrit (%) = 2.953 x hemoglobin (g/dL) [17] was used to calculate hemoglobin values.
The data sharing collaborative of the PH&S Cardiovascular Study Group produced a risk model for coronary artery bypass (CABG) operative mortality, based on 12,845 isolated CABGs from 1997 through June 2002 [18]. This model did not contain blood transfusion as a risk factor. Recently, the PH&S risk model was recalibrated using 12,962 additional CABG surgeries performed from 2002 to 2007 (results not shown). The recalibrated PH&S model expected mortality was used to risk-adjust the CABG outcome results.
Strategies of Blood Conservation: Preoperative Evaluation and Management
Our intention is to minimize our patients' operative risk by optimizing their preoperative condition. Preoperative anemia, especially when combined with advanced age or chronic renal insufficiency is a common indication for a Blood Conservation Coordinator (BCC) consultation. Forty-three percent of our CABG patients are admitted with acute coronary syndromes and have surgery before discharge, while 4% of our patients are emergent or salvage.
Although there are no agreed upon indications, each surgeon decides which of their patients will benefit from a BCC referral. Currently 22% of our patients are evaluated by the BCC. These evaluations include a fasting ferritin, transferrin, serum iron, and complete blood count, in order to assure that adequate iron stores are available for RBC production. Seventy percent of these patients receive preoperative intravenous iron. Our preoperative goal for patients (without underlying inflammatory processes) is to have a ferritin of 200 to 300 ng/mL. Because of concerns about safety and efficacy, only about 2% of patients are treated with erythropoietic stimulating agents preoperatively. The cost of these interventions is mitigated by a diagnosis of latent iron deficiency, iron deficiency anemia, and(or) renal insufficiency that require treatment. In addition to the clinical role, the BCC acts as a driver of blood conservation and transfusion reduction efforts in order to help keep the team on track.
Surgical Technique
Anesthesia techniques and fluid administration are not standardized among the cardiac anesthesiologists. Typical anesthetic agents used are midazolam, fentanyl citrate, isoflurane (all from Abbott Laboratories, North Chicago, IL), propofol (Astra Zeneca s.p.a., Caponago, Italy), and neuromuscular blockade. During the study period, aprotinin (Bayer Healthcare, Wayne, NJ) use was low and those patients who did not receive aprotinin received aminocaproic acid (Hospira Inc, Lake Forest, IL), with dosage ranging from 7.5 to 10.0 grams, both as a loading dose and in the pump prime, at the discretion of the anesthesiologist. Intravenous fluid administration is judicious. Prior to cardiopulmonary bypass (CPB), blood pressure is maintained with phenylephrine (Baxter, Irvine, CA) or norepinephrine bitrate (Bedford Labs, Bedford, OH) rather than fluid administration as long as monitoring parameters do not indicate a deleterious fluid deficit.
The same four surgeons performed cardiac surgery during the baseline and project periods. The most recent surgeon addition was in 1997. A senior surgeon retired after an injury in September 2006. All have made concerted efforts to standardize surgical techniques and follow agreed-upon protocols. The CPB is employed in greater than 95% of patients. A median sternotomy is routinely used. Meticulous attention is paid to surgical hemostasis throughout the procedure. Saphenous vein and radial artery donor sites are closed immediately after harvest to minimize blood loss. Irrigation of the surgeons' hands and of the field is minimized to avoid introducing crystalloid into the CPB circuit. A single cross-clamp technique is used to perform proximal anastomoses in CABG patients. When irrigation is used in the field for valves, cardiotomy suction and vents are temporarily discontinued. Shed blood is routinely returned to the perfusion circuit, except for one surgeon who only uses cardiotomy suction on CABG patients if unusual bleeding is noted. Wall suction is used for any small blood collections. Shed blood during these cases is rarely greater than 200 mL.
Previous small-scale studies in our institution have shown no significant decrease in blood loss or transfusion rate when a cell saver is used. Therefore, only patients who have a high risk of needing a blood transfusion, or who do not accept blood products, routinely have a cell saver used. When a cell saver is used, a portion of the remaining blood in the CPB circuit is processed through it. Otherwise all blood in the circuit is returned to the patient.
Perfusion
The techniques used on every case are vacuum-assisted venous drainage with dry 3/8'' tubing, 10 foot arterial-venous loop, retrograde autologous prime, and the drainage of saline prime from the cardioplegia circuit. We use a polymer-coated perfusion circuit (Terumo, Ann Arbor, MI). Our priming method involves removal of normal saline (Hospira Inc) from the circuit and replacement with mannitol, sodium bicarbonate (both from Hospira Inc), heparin sodium (APP Pharmaceuticals, LLC, Schaumburg, IL), and albumin (Grifols Biologicals, Inc, Los Angeles, CA). Normal saline prime volume ranges between 175 mL and 425 mL depending on whether albumin is used. Tubing lengths and prime volume have decreased through optimization of the perfusion circuit and have remained stable since 2003.
Cold 4:1 blood cardioplegia is used, with high dose potassium, to limit crystalloid administration. Normothermia is maintained except in cases of circulatory arrest or specific surgeon request. The use of a hemoconcentrator is determined on a case by case basis.
Postoperative Period
Postoperative care is provided by the surgeons and physician assistants. Other subspecialists are consulted by the surgeon as needed. We do not have a standard hematocrit transfusion trigger or specific treatment protocols. Our major strategy for limiting the use of RBCs postoperatively is permissive anemia. Our indications for transfusion include persistent hypotension, orthostatic hypotension, tachycardia, and inability to participate in rehabilitation. The nursing staff is encouraged to restrict crystalloid administration, but may use amounts up to 2 L prior to involving the surgeon in further decision making. Five percent albumin is commonly used for additional volume resuscitation. Judicious use of pressors is encouraged to further minimize hemodilution. Routine lab draws currently include the following: three hematocrits, three potassium levels, two renal panels, one magnesium level, and two arterial blood gases during the first 72 hours. We do not use pediatric tubes for blood draws. Our extubation protocol has been developed in concert with respiratory therapy and minimizes the use of arterial blood gases. The policies regarding frequency and timing of these lab draws are reviewed annually. Consultants are strongly discouraged from administering blood products without prior discussion with the operating surgeon. The majority of our patients are treated with an anemia management protocol, which consists of supplementation with oral iron, intramuscular B-12, vitamin C, and folate. This is often started on postoperative day 2 for patients with a hematocrit of less than 35%.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
|
|
|
|
|
|
Our isolated CABG patient population (n = 1,617) from 2003 to 2007 was used to compare specific patient characteristics and risk-adjusted outcomes as the RBC utilization rate decreased (Table 2). The CABG patient demographics, medical history, comorbidities, and procedure status were similar over time. The predicted risk of transfusion varied between 46% and 49% and was not significantly different over time (p = 0.415). The odds ratio for transfusion decreased from 0.6 to 0.1. Figure 6 compares our blood utilization to state and national benchmark trends that use COAP and STS data. Our hospital is one of 18 open-heart surgery programs in Washington State and accounts for about 10% of the surgical volume. The blood product utilization in our facility is lower than these benchmarks.
|
|
2 test (p = 0.452). In 2007 there were seven CABG deaths. Causes of death included: pulmonary embolism, malignant ventricular arrhythmia after left ventricular assist device explantation, incomplete revascularization in the setting of a fresh myocardial infarction and cardiogenic shock, coronary embolization, necrotic bowel, and two patients with multiorgan failure. Five of the seven received RBCs. All of them were administered intraoperatively or early in the postoperative period and the average ante-mortem hematocrit was 28.7%. This increase is responsible for the rise in mortality for all cardiac procedures from 3.0% to 3.5%. The in-hospital mortality for procedures other than isolated CABG actually trended downward (8.5%, 4.9%, 7.2%, 5.3%, 5.1%) for years 2003 to 2007, respectively. CUSUM analysis reveals that the apparent rise in mortality for all CABG patients is confined to those patients who received RBCs. Figure 7 shows the risk-adjusted CABG mortality using CUSUM analysis. These curves were computed separately for patients who did not receive (upper panel) or did receive (lower panel) RBC transfusions. The transfused group has a CUSUM that ends above zero, indicating about five more deaths than expected, but the curve remains within the prediction limits. For the non-transfused group, the CUSUM remains very close to the lower prediction limit indicating about five fewer deaths than expected.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Transfusion practices in cardiac surgery are the subject of increasing interest. We began our project without a clear endpoint in mind. We were unable to predict how much reduction in blood utilization could be achieved without impairing patient safety. Each time the team looked at data, ideas for further lowering transfusion rates were discussed. Our success in reducing blood utilization has come about because of a collaborative multidisciplinary approach. All members of the team changed their practices based on outcomes data. Interestingly, the surgeons were not the main impediment to improvement. As an example, one of the first areas examined was reoperation for bleeding after cardiac surgery. Surgeon data were discussed in an unblinded format. The surgeons with higher reoperation for bleeding rates then discussed with their colleagues what techniques were used to minimize bleeding. We also measured total drainage for the first 24-hour period to make sure we were not simply raising our threshold for reoperation. This resulted in a best practice being identified and adopted by the surgeons. This same improvement methodology is used throughout the cardiac surgery program.
We found the major barrier to improvement was changing the culture of the entire patient care team, many of whom were convinced that historic patterns of a liberal use of blood products constituted the best patient care. In order to communicate our program goals to caregivers we present our data on a yearly basis. Regular education regarding patient outcomes is crucial to motivate team members to continue to focus on transfusion reduction. This has helped overcome reluctance to change longstanding transfusion practices.
Over time, our transfusions have moved out of the operating room and into the postoperative setting. Transfusion in cardiac surgery is rarely an emergency, and usually does not need to be administered in the operating room or on an urgent basis. Postoperatively, the surgeon can assess each patient's hemodynamic status and tolerance of anemia, and decide whether transfusion is warranted. This allows a more systematic and methodical evaluation than is possible in the operating room, or by phone conversation between nursing staff and an on-call physician at an inopportune hour.
Many patients who previously would have received RBCs now avoid transfusion with equivalent if not improved outcomes. Identification of those patients who benefit from transfusion is more difficult. Our patients' mean preoperative hematocrit is 39%. We make a concerted effort to perform cardiac surgery on an elective basis and optimize RBC mass. Our mean "lowest on bypass" hematocrit in 2007 was 26%. Our average discharge hematocrit is 28% and patients are commonly sent home with a hematocrit in the low 20s. It has become apparent to us that in order to further reduce our blood utilization we need to be more effective in educating patients, cardiologists, and primary care physicians regarding postoperative anemia.
During our five-year experience, we implemented many strategies in an effort to decrease our blood utilization. Not all interventions had the desired result. The essential components of our program that led to an exceptionally low transfusion rate are the following: surgeon attention to hemostasis, creative perfusion strategies to reduce hemodilution, an effective BCC, a treatment protocol for anemia management without transfusion, the leadership of a physician champion, the willingness of all caregivers to change practice based on ongoing education and data review regarding blood utilization, institutional and multidisciplinary commitment to reduce blood utilization, an organizational commitment to collect and review detailed, and provider-specific blood utilization data.
Cardiac surgery can be performed safely with a low utilization of blood products and with good patient outcomes. Performance improvement is heavily reliant on accurate, relevant data that serve to inform the change process. Regular reporting of data is critical to keeping team members informed and engaged. Reducing the number of patients who receive blood or the amount of blood given to patients who need blood requires the attention and involvement of all members of the patient care team.
|
Footnotes |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  D. Moskowitz and A. Shander Reply Ann. Thorac. Surg., November 1, 2011; 92(5): 1932 - 1933. [Full Text] [PDF]
|
|
|
|
 |
|
 F. Masud, K. Larson-Pollock, C. Leveque, and D. Vykoukal Establishing a Culture of Blood Management Through Education: A Quality Initiative Study of Postoperative Blood Use in CABG Patients at Methodist DeBakey Heart & Vascular Center American Journal of Medical Quality, September 1, 2011; 26(5): 349 - 356. [Abstract] [PDF]
|
|
|
|
 |
|
 M. Ranucci, S. Aronson, W. Dietrich, C. M. Dyke, A. Hofmann, K. Karkouti, M. Levi, G. J. Murphy, F. W. Sellke, L. Shore-Lesserson, et al. Patient blood management during cardiac surgery: Do we have enough evidence for clinical practice? J. Thorac. Cardiovasc. Surg., August 1, 2011; 142(2): 249.e1 - 249.e32. [Full Text] [PDF]
|
|
|
|
|
|
V. A. Ferraris, J. R. Brown, G. J. Despotis, J. W. Hammon, T. B. Reece, S. P. Saha, H. K. Song, E. R. Clough, L. J. Shore-Lesserson, L. T. Goodnough, et al. 2011 Update to The Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists Blood Conservation Clinical Practice Guidelines Ann. Thorac. Surg., March 1, 2011; 91(3): 944 - 982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Reyes, M. Prieto, P. Alvarez, M. Orts, J. Bustamante, G. Santos, A. Sarraj, and A. Planas Cell saving systems do not reduce the need of transfusion in low-risk patients undergoing cardiac surgery Interact CardioVasc Thorac Surg, February 1, 2011; 12(2): 189 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Moskowitz, J. N. McCullough, A. Shander, J. J. Klein, C. A. Bodian, R. S. Goldweit, and M. A. Ergin The Impact of Blood Conservation on Outcomes in Cardiac Surgery: Is It Safe and Effective? Ann. Thorac. Surg., August 1, 2010; 90(2): 451 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Engoren Invited Commentary Ann. Thorac. Surg., August 1, 2010; 90(2): 458 - 459. [Full Text] [PDF]
|
|
|
|
|
|
G. L. Grunkemeier, R. Jin, and Y. Wu Cumulative Sum Curves and Their Prediction Limits Ann. Thorac. Surg., February 1, 2009; 87(2): 361 - 364. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
