Gynecologic Oncology 132 (2014) 227–230
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Compliance rates and outcomes associated with a restrictive transfusion policy in gynecologic oncology patients Jonathan D. Boone a,⁎, Kenneth H. Kim b, Marisa Marques c, J. Michael Straughn a a b c
Department of Obstetrics and Gynecology, University of Alabama at Birmingham, USA Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, USA Department of Pathology, University of Alabama at Birmingham, USA
H I G H L I G H T S • Gynecologic oncologists are largely compliant with restrictive transfusion policies. • Morbidity and mortality are not increased with a restrictive transfusion policy. • Improved compliance should focus on limiting transfusions until hemoglobin is ≥7 g/dL and transfusing in single pRBCs unit increments.
a r t i c l e
i n f o
Article history: Received 11 August 2013 Accepted 19 October 2013 Available online 24 October 2013 Keywords: Transfusion Restrictive Compliance
a b s t r a c t Objectives. Blood products are scarce but essential medical resources. Initially transfusions showed increased perioperative complications, prolonged hospitalizations, and higher mortality. Recently developed restrictive transfusion policies have not shown those adverse affects in critically ill patients. Hospitals adopted these policies to guide blood product administration. The objective of this study is to determine compliance with a restrictive transfusion policy in gynecologic oncology patients. Methods. A retrospective chart review of gynecologic oncology patients undergoing transfusion with packed red blood cells (pRBCs) from 12/2008 to 9/2011 was performed. Cancer type and stage, surgical procedure, hemoglobin values, pRBC transfusions, intraoperative blood loss, and postoperative complications were collected. Each transfusion was classified as compliant or noncompliant. Results. A total of 582 patients requiring 2,276 blood transfusions were identified. The mean age was 55.9 years. Ovarian and endometrial cancers were the most common malignancies. Gynecologic oncologists were 81.1% compliant with the restrictive transfusion policy; 59.0% of transfusions were secondary to exceptions. Noncompliant transfusions were commonly given on the day of surgery when intraoperative blood loss was b 1500cc and for asymptomatic anemia. Only 64.7% of the transfusions were ordered in single unit increments. There was no significant difference in postoperative infections, thrombotic events, and mortality between compliant and noncompliant transfusions. Conclusion. The majority of gynecologic oncology patients receive transfusions compliant with the restrictive transfusion policy. Morbidity and mortality are not increased with a restrictive transfusion policy. Efforts to improve compliance should focus on limiting transfusions when the hemoglobin is ≥ 7 g/dL and transfusing in single pRBCs unit increments. © 2013 Elsevier Inc. All rights reserved.
Objectives Blood products are vital medical resources often found in short supply. There has been debate on the risks of anemia and benefits of transfusion specifically in critically ill patients. Early studies demonstrated poor surgical outcomes and decreased survival in patients with perioperative ⁎ Corresponding author at: Department of Obstetrics and Gynecology, University of Alabama at Birmingham, 1700 6th Avenue South, Birmingham, AL 35233, USA. Fax: 205975-6174. E-mail address: [email protected] (J.D. Boone). 0090-8258/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2013.10.019
anemia [1–3]. Some studies have shown that blood transfusions are associated with longer hospital stays and impaired recovery [4,5]. In 1999, the Transfusion Requirements in Critical Care (TRICC) trial demonstrated that a restrictive transfusion policy significantly decreased mortality during hospitalization in patients less acutely ill without active coronary ischemia [6]. Since the publication of that trial, Carless et al. confirmed that restricting transfusions does not increase the rates of cardiac events, myocardial infarction, stroke, pneumonia, thromboembolism, and mortality [7]. An additional study by Herbert et al. showed lower rates of worsening organ dysfunction when transfusions were restricted [8]. It is unclear if restrictive transfusion policies can be used
228
J.D. Boone et al. / Gynecologic Oncology 132 (2014) 227–230
in gynecologic oncology patients who have undergone complicated debulking surgeries or received myelosuppressive chemotherapeutic agents and radiation therapy. The TRICC trial showed that transfusions can be restricted in post-operative patients as 39% of these patients were included in the restrictive transfusion arm of the trial. Compliance with a restrictive transfusion policy and patient outcomes has not been studied in gynecologic oncology patients. In 2007, a restrictive transfusion policy was initiated at the University of Alabama at Birmingham (UAB). We sought to evaluate compliance and outcomes with a restrictive transfusion policy in gynecologic oncology patients. Methods After obtaining approval from the Institutional Review Board, a retrospective chart review of gynecologic oncology patients was performed. Patients who received packed red blood cell (pRBC) transfusions from 10/2008 to 9/2011 were included in the study. All hospitalized patients, both surgical and medical, were included. During this time period, a restrictive transfusion policy recommended no pRBC transfusions until hemoglobin concentrations dropped below 7 g/dL, and transfusions were given in one-unit increments. Several exceptions to this transfusion policy exist: intraoperative or day-of-surgery transfusion, symptomatic anemia, irreversible bone marrow suppression, active bleeding in absence of a surgical procedure, severe sepsis, active coronary ischemia, and transfusion on post-operative day one or later after a 1,500 cc or greater intraoperative blood loss. Abstracted data included patient demographics, cancer type and stage, surgical procedure, pre-transfusion hemoglobin values, number of pRBC units transfused, estimated intraoperative blood loss, postoperative complications, and 30-day mortality. Statistical analyses were performed using a Fisher's exact test, and a p value of 0.05 or less was considered statistically significant. Results A total of 582 patients receiving 2,276 transfusions were included in this study. Patient demographics are outlined in Table 1. The mean age was 55.9 years and the majority of patients were Caucasian (57.5%). Ovarian cancer (44.8%) was the most common malignant diagnosis followed by endometrial (22.0%) and cervical cancer (19.8%). Eight-six (14.8%) patients with benign disease and 5 (0.8%) patients with premalignant disease were included in the analysis. The most common surgical procedures are listed in Table 2. The majority of the patients (83.4%) underwent an open abdominal procedure with the most common operation being a total abdominal hysterectomy (TAH) with or without bilateral salpingo-oopherectomy (BSO) and debulking. Transfusion events, number of units transfused, and compliance with the restrictive transfusion policy are detailed in Table 3. The mean pretransfusion hemoglobin was 7.8g/dL (range 3.6–14.6). The mean number of units transfused per person was 3.9 with the mean number of units transfused per transfusion event being 1.6 (range 1–13). Overall, 1,127 (81.1%) of the transfusion events were compliant with the restrictive transfusion policy. Three hundred seven (22.1%) of the transfusion events were issued when the hemoglobin concentration was b 7 g/dL; 820 (59.0%) transfusions were compliant with the restrictive transfusion policy based on the approved exceptions. The most common reasons given as an exception for transfusion were intraoperative or day-ofsurgery transfusion (36.7%), symptomatic anemia either postoperative or chemotherapy-induced (31.6%), and irreversible bone marrow suppression secondary to whole pelvic radiation therapy (14.0%). Two hundred sixty-three (18.9%) transfusion events resulting in 365 units of pRBCs were noncompliant with the restrictive transfusion policy. Reasons for violating the transfusion policy included post-operative hemoglobin concentrations greater than 7 g/dL with less than a 1,500 cc intraoperative blood loss, asymptomatic chemotherapy-induced anemia, anemia NOS, and preoperative anemia. Of the 90 patients receiving
Table 1 Patient demographics. n = 582 Age Mean (years) Range (years) Race Caucasian African–American Other Body mass index Mean (kg/m2) Range Malignancy Ovarian cancer Endometrial cancer Cervical cancer Gastrointestinal cancer Peritoneal cancer Vulvar cancer GTN Vaginal cancer Other Premalignant disease Borderline tumor Cervical dysplasia Benign disease Leiomyoma Ovarian cyst Tuboovarian abscess Endometriosis/adenomyosis Abdominal/vulvar Abscess Ovarian torsion Other
55.9 13–89 335 (57.5%) 235 (40.4%) 12 (2.1%) 29.2 12.1–65.2 491 (84.4%) 220 (44.8%) 108 (22.0) 97 (19.8%) 29 (5.9%) 14 (2.9%) 6 (1.2%) 6 (1.2%) 5 (1.0%) 6 (1.2%) 5 (0.8%) 3 (60.0%) 2 (40.0%) 86 (14.8%) 28 (32.6%) 25 (29.0%) 8 (9.3%) 7 (8.1%) 5 (5.8%) 4 (4.7%) 9 (10.5%)
GTN — gestational trophoblastic neoplasia.
noncompliant transfusions due to an estimated blood loss (EBL) of b1,500 cc, the mean intraoperative blood loss was 556 cc. The 243 surgeries requiring transfusion resulted in a total of 638 units being transfused with 2.6 units transfused per procedure. Patients undergoing an intraoperative transfusion had a slightly higher mean EBL of 1,260 cc compared to 1,020 cc for all patients undergoing a procedure. (Table 4) The most common surgical procedures requiring transfusion involved a TAH±BSO with debulking. Laparoscopic and robotic procedures were rarely associated with a transfusion as only 4 robotic-assisted (RA) laparoscopic procedures received pRBCs. The second part of the restrictive transfusion policy states that transfusions should be ordered in single unit increments unless there is active blood loss. We found that 900 (64.7%) transfusion events were compliant in transfusing one unit at a time followed by hemoglobin
Table 2 Surgical history.
TAH +/− BSO with debulking Bowel/mass resection TAH +/− BSO USO/BSO with debulking TAH +/− BSO with staging Cystoscopy EUA +/− biopsy Vaginectomy/vulvectomy USO/BSO Radical hysterectomy +/− BSO with staging Pelvic exenteration Robotic-assisted TLH/BSO+/−LND Laparoscopic BS0 Other
N = 404 123 62 59 40 38 10 8 6 5 5 5 3 1 39
TAH—total abdominal hysterectomy; BSO—bilateral salpingo-oophorectomy; USO—unilateral salpingo-oophorectomy; TLH—total laparoscopic hysterectomy; LND—lymph node dissection; EUA—exam under anesthesia.
J.D. Boone et al. / Gynecologic Oncology 132 (2014) 227–230 Table 3 Transfusions and compliance with restrictive transfusion policy. Mean hemoglobin (g/dL) Mean Range Transfusion events Total units transfused Compliant Noncompliant Mean units transfused per person Units per transfusion event Mean Range Compliance with restrictive transfusion policy (by event) Compliant (Hgb b7 g/dL) Compliant by exception Intraoperative transfusion on day of surgery Symptomatic anemia Irreversible bone marrow suppression Active bleeding Symptomatic chemotherapy-induced anemia Severe sepsis Transfusion POD1+ with N1500 cc blood loss Active coronary ischemia Noncompliant transfusions (by event) Reason for transfusion Post-operative with b1500 cc blood loss Chemotherapy-induced anemia without symptoms (n = 113) Anemia NOS Pre-operative anemia Compliant with transfusing 1 unit per transfusion event Not compliant with transfusing 1 unit per transfusion event Transfusion compliance by disease type Transfusion for benign disease Compliant Noncompliant Transfusion for malignant disease Compliant Noncompliant
Table 5 Patient complications.
7.8 3.6–14.6 1,390 2,276 1,911 365 3.9 1.6 1–13 1,127 (81.1%) 307 (22.1%) 820 (59.0%) 301 (36.7%) 198 (24.2%) 115 (14.0%) 65 (7.9%) 61 (7.4%) 56 (6.8%) 17 (2.1%) 7 (0.9%) 263 (18.9%) 90 (34.2%) 78 (29.7%) 73 (27.8%) 22 (8.3%) 900 (64.7%) 490 (35.3%) p = 0.83 162 133 (82.1%) 29 (17.9%) 1,228 994 (80.9%) 234 (19.1%)
concentration monitoring then ordering additional transfusions as needed. Four hundred ninety (35.3%) transfusion events were noncompliant and occurred when two or more units were given before a repeat hemoglobin was checked. We compared transfusions for both benign and malignant disease to determine compliance for each disease state. Compliance rates for both benign and malignant disease were similar; roughly 81% of transfusions for both disease states were compliant with the restrictive transfusion policy. The complications analyzed in this study were postoperative infection, thrombotic events, and 30-day mortality (Table 5). Postoperative infection consisted of urinary tract infection, pneumonia, wound infection, abscess, and sepsis. The thrombotic events recorded were either deep vein thrombosis (DVT) or pulmonary thromboembolism (PTE). Only
Table 4 Intraoperative hemodynamic status and transfusion. Total patients requiring transfusion Total surgeries requiring transfusion Total units transfused intraoperative Mean units transfused per procedure EBL per procedure (cc) Mean Range EBL per procedure requiring transfusion (cc) Mean Range EBL intraoperatively with b 1500 cc blood loss (cc) Mean Range
229
233 243 638 2.6 1,020 0–6,000 1,260 50–6,000 556 25–1,400
Postoperative complications per patient Infection Compliant with policy Noncompliant with policy Thrombotic event Compliant with policy Noncompliant with policy 30-day Mortality Compliant with policy Noncompliant with policy Transfusion reaction
n = 582 66 (11.3%) p = 0.122 51 15 29 (5%) p = 0.222 23 6 32 (5.5%) p = 0.335 29 3 5 (0.22%)
five (0.22%) transfusion reactions occurred in the 582 patients. There were 66 patients (11.3%) who developed a postoperative infection, 29 (5%) who were diagnosed with a DVT or PTE, and 32 (5.5%) who died within 30 days of being discharged from the hospital. There was no difference in complication rates in the compliant versus noncompliant transfusions (Table 5).
Discussion Approximately 24million blood products and 10million units of blood are transfused each year, and this resource is becoming increasingly more expensive and less available [9–11]. In the early 1990s, red blood cell transfusions were given to maintain a hemoglobin concentration of 10 g/dL or greater to increase oxygen delivery with the intent of avoiding the detrimental effects of oxygen debt seen in anemic patients [12–14]. Several studies evaluated the safety of blood product administration. Heier et al. showed that if the hemoglobin concentration is greater than 8 g/dL, lung function and volume status are more important to clinical outcome than the hemoglobin concentration [15]. This same study noted that there were no “measurable changes of human cognitive function” after an acute drop in the hemoglobin concentration to 7 g/dL in normovolemic patients [15]. Numerous studies have shown that blood transfusions were associated with increased nosocomial infection rates, multisystem organ failure, longer hospital stays, and mortality [15–18]. The TRICC trial performed by Hebert et al. marked a major change in transfusion medicine because it demonstrated that a restrictive transfusion policy is at least as effective as a liberal transfusion strategy [6]. It is still argued that no single hemoglobin value can serve as a transfusion trigger without accounting for patient and clinical variables [19]. However, in support of the TRICC trial, a systematic review of a few randomized controlled trials on blood transfusion practices concluded that a single laboratory value can serve as a restrictive transfusion indicator in patients without cardiac disease [5]. In 2007, the blood bank at UAB instituted a restrictive transfusion policy allowing transfusion with pRBCs only when the hemoglobin concentration falls below 7 g/dL unless one of the established exceptions is met. Exceptions included intraoperative or day-of-surgery transfusion in the setting of active bleeding, symptomatic anemia (postoperative or chemotherapy-induced), irreversible bone marrow suppression (as in the case of whole pelvic radiation therapy), severe sepsis, active coronary ischemia, and transfusion on postoperative day one or later when the intraoperative blood loss was at least 1,500 cc. Active bleeding such from cervical and vaginal lesions or from the rectum as a result of radiation proctitis was also an exception. For the purpose of this study, patients receiving chemotherapy were considered to have reversible bone marrow suppression; however, the argument can be made for transfusing patients with a low hemoglobin in order to avoid the hemoglobin falling below 7g/dL when myelosuppressive chemotherapeutic agents are used. In that case, an additional 113 units of pRBCs given would have been compliant.
230
J.D. Boone et al. / Gynecologic Oncology 132 (2014) 227–230
In this current study, 81% of pRBC transfusions were compliant with a restrictive transfusion policy in gynecologic oncology patients. The majority of these transfusions were compliant by exception, most commonly because of intraoperative transfusion. Transfusion compliance was similar whether the underlying pathology was malignant or benign demonstrating that surgeons were not more likely to transfuse a patient just because they had a confirmed malignancy. An additional component of the UAB restrictive transfusion policy states that pRBC transfusions are to be given one unit at a time followed by rechecking the patient's hemoglobin concentration before an additional unit is given. Importantly, only 65% of the transfusion events in this study were given in single unit increments. This certainly highlights the tendency of gynecologic oncologists to transfuse at least two or more units of blood without rechecking hemoglobin concentration. This issue is an area for potential improvement in the use of blood products. Six hundred thirty-eight intraoperative transfusions were given during surgery and were considered exceptions to the restrictive transfusion policy given active intraoperative bleeding. More transfusions were administered for malignant disease than benign disease which is expected given the vascular nature of malignant tumors and the more extensive surgery involved with cancer resections. The majority of the surgical cases for malignant disease involved removal of the uterus and/ or bilateral fallopian tubes and ovaries, omentum, and a bowel resection. Mean blood loss during surgery resulting in an intraoperative or same day pRBC transfusion was 1,260cc. Transfusions given in the immediate postoperative period that were not compliant with the restrictive transfusion policy had an average intraoperative blood loss of 556 cc and comprised 34.2% of the noncompliant transfusions. These 90 transfusions indicate a possible area for cost savings and quality improvement. Four patients undergoing robotic procedures received pRBCs. There were three RA-TLH with either unilateral salpingo-oopherectomy (USO) or BSO and one RA-BSO with pelvic lymphadenectomy (LND). Two of the RA-TLH were done for leiomyomatous disease and one for stage 1 endometrial cancer. The RA-BSO with LND was an aborted radical hysterectomy for cervical cancer after advanced stage disease was discovered. In total, three units of pRBCs given after the laparoscopic cases were not compliant with the restrictive transfusion policy having an EBL of less than 100 cc while the other three units given were compliant because of symptomatic anemia and active bleeding from the cervix. We compared complications associated with compliant transfusions versus those that were noncompliant. Bernard et al. showed that an intraoperative transfusion with a single unit of pRBCs was associated with more pneumonia, sepsis, surgical site infection, and increased mortality with transfusion of a second unit of pRBCs [1]. Our analysis included commonly reported complications including postoperative infection, thrombotic events, and 30-day mortality [1,16,18,20]. There were similar postoperative infections, thrombotic events, and 30-day mortality events in the compliant versus noncompliant patients (Table 5). We also noted five transfusion reactions, 0.22%, which is consistent with National Blood Collection and Utilization Survey's adverse transfusion reaction rate of 0.25% [11]. The delivery of cost-effective healthcare is a central focus of medicine today. Our blood bank estimates $220 per unit of leukoreduced red blood cells, and the 263 transfusions not compliant with the restrictive transfusion policy resulted in 365 pRBC unit transfusions totaling a potential $80,300 in savings. As with any retrospective study, one of the major weaknesses is the accuracy of the retrospective review of the medical record. It is difficult to determine if patient's symptoms related to her anemia were always recorded accurately. The noncompliant transfusions could have been compliant because complaints of lightheadedness or dizziness were not documented in the medical record. Physicians are historically inaccurate in estimating intraoperative blood loss [21,22]. This makes it difficult to evaluate an accurate estimated blood loss. Also, it is difficult to control for confounders that may have contributed to the rates of postoperative infections, thrombotic events, and 30-day mortality. Overall,
the gynecologic oncology division at UAB is largely compliant with the hospital's restrictive transfusion policy. It is apparent that more emphasis should be placed on transfusing by single-unit increments to allow time for the patient to respond to the first transfusion. A second important area for improvement includes the 263 transfusions administered when the hemoglobin was greater than or equal to 7 g/dL, and the cost savings associated with appropriate use of that resource. Further research and quality review is needed to ensure compliance with restrictive transfusion policies in gynecologic oncology, as previous retrospective reviews of blood product use have not improved transfusion practices [23,24]. Conflict of interest statement The authors declare that there are no conflicts of interest.
References [1] Bernard AC, Davenport DL, Chang PK, Vaughn TB, Zwischenberger JB. Intraoperative transfusion of 1 U to 2 U packed red blood cells is associated with increased 30-day mortality. J Am Coll Surg 2009;208(5):931–7 [937 e1-2; discussion 938-9]. [2] Oliver E, Carrio ML, Rodriguez-Castro D, Javierre C, Farrero E, Torrado H, et al. Relationships among haemoglobin level, packed red cell transfusion and clinical outcomes in patients after cardiac surgery. Intensive Care Med 2009;35(9):1548–55. [3] Beattie WS, Karkouti K, Wijeysundera DN, Tait G. Risk associated with preoperative anemia in noncardiac surgery: a single-center cohort study. Anesthesiology 2009;110(3):574–81. [4] Vamvakas EC, Blajchman MA. Blood still kills: six strategies to further reduce allogeneic blood transfusion-related mortality. Transfus Med Rev 2010;24(2):77–124. [5] Carson JL, Hill S, Carless P, Herbert P, Henry D. Transfusion triggers: a systematic review of the literature. Transfus Med Rev 2002;16(3):187–99. [6] Herbert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340(6):409–17. [7] Carless PA, Henry DA, Carson JL, Herbert PP, McClelland B, Ker K. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev 2010;10:CD002042. [8] Herbert PC, Yetisir E, Martin C, Blajchman MA, Wells G, Marshall J, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med 2001;29(2):227–34. [9] Kanavos P, Yfantopoulos J, Vandoros C, Politis C. The economics of blood: gift of life or a commodity? Int J Technol Assess Health Care 2006;22(3):338–43. [10] Goodnough LT, Levy JH, Murphy MF. Concepts of blood transfusion in adults. Lancet 2013;381(9880):1845–54. [11] Whitacker BI, Henry RA, Schlumpf K, Schulman J, Green J. Report of the US Department of Health and Human Services. The 2009 national blood collection and utilization survey report. Washington, DC: US Department of Health and Human Services, Office of the Assistant Secretary for Health; 2011 1–58. [12] Russell JA, Phang PT. The oxygen delivery/consumption controversy. Approaches to management of the critically ill. Am J Respir Crit Care Med 1994;149(2 Pt 1):533–7. [13] Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA 1993;270(22):2699–707. [14] Gattinoni L, Brazzi L, Pelosi P, Latini R, Tognoni G, Pesenti A, et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 1995;333(16):1025–32. [15] Heier HE, Bugge W, Hjelmeland K, Soreide E, Sorlie D, Haheim LL. Transfusion vs. alternative treatment modalities in acute bleeding: a systematic review. Acta Anaesthesiol Scand 2006;50(8):920–31. [16] Bochicchio GV, Napolitano L, Joshi M, Bochicchio K, Meyer W, Scalea TM. Outcome analysis of blood product transfusion in trauma patients: a prospective, risk-adjusted study. World J Surg 2008;32(10):2185–9. [17] Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM. Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma 2003;54(5):898–905 [discussion 905–7]. [18] Zilberberg MD, Shorr AF. Effect of a restrictive transfusion strategy on transfusionattributable severe acute complications and costs in the US ICUs: a model simulation. BMC Health Serv Res 2007;7:138. [19] Goodnough LT, Despotis JG, Hogue CW, Ferguson TB. On the need for improved transfusion indicators in cardiac surgery. Ann Thorac Surg 1995;60(2):473–80. [20] Abu-Rustum NR, Richard S, Wilton A, Lev G, Sonoda Y, Hensley ML, et al. Transfusion utilization during adnexal or peritoneal cancer surgery: effects on symptomatic venous thromboembolism and survival. Gynecol Oncol 2005;99(2):320–6. [21] Ashburn JC, Harrison T, Ham JJ, Strote J. Emergency physician estimation of blood loss. West J Emerg Med 2012;13(4):376–9. [22] Santoso JT, Dinh TA, Omar S, Hannigan EV. Surgical blood loss in abdominal hysterectomy. Gynecol Oncol 2001;82(2):364–6. [23] Goodnough LT, Audet AM. Utilization review for red cell transfusions. Are we just going through the motions? Arch Pathol Lab Med 1996;120(9):802–3. [24] Haspel RL, Uhl L. How do I audit hospital blood product utilization? Transfusion 2012;52(2):227–30.
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Compliance rates and outcomes associated with a restrictive transfusion policy in gynecologic oncology patients Jonathan D. Boone a,⁎, Kenneth H. Kim b, Marisa Marques c, J. Michael Straughn a a b c
Department of Obstetrics and Gynecology, University of Alabama at Birmingham, USA Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, USA Department of Pathology, University of Alabama at Birmingham, USA
H I G H L I G H T S • Gynecologic oncologists are largely compliant with restrictive transfusion policies. • Morbidity and mortality are not increased with a restrictive transfusion policy. • Improved compliance should focus on limiting transfusions until hemoglobin is ≥7 g/dL and transfusing in single pRBCs unit increments.
a r t i c l e
i n f o
Article history: Received 11 August 2013 Accepted 19 October 2013 Available online 24 October 2013 Keywords: Transfusion Restrictive Compliance
a b s t r a c t Objectives. Blood products are scarce but essential medical resources. Initially transfusions showed increased perioperative complications, prolonged hospitalizations, and higher mortality. Recently developed restrictive transfusion policies have not shown those adverse affects in critically ill patients. Hospitals adopted these policies to guide blood product administration. The objective of this study is to determine compliance with a restrictive transfusion policy in gynecologic oncology patients. Methods. A retrospective chart review of gynecologic oncology patients undergoing transfusion with packed red blood cells (pRBCs) from 12/2008 to 9/2011 was performed. Cancer type and stage, surgical procedure, hemoglobin values, pRBC transfusions, intraoperative blood loss, and postoperative complications were collected. Each transfusion was classified as compliant or noncompliant. Results. A total of 582 patients requiring 2,276 blood transfusions were identified. The mean age was 55.9 years. Ovarian and endometrial cancers were the most common malignancies. Gynecologic oncologists were 81.1% compliant with the restrictive transfusion policy; 59.0% of transfusions were secondary to exceptions. Noncompliant transfusions were commonly given on the day of surgery when intraoperative blood loss was b 1500cc and for asymptomatic anemia. Only 64.7% of the transfusions were ordered in single unit increments. There was no significant difference in postoperative infections, thrombotic events, and mortality between compliant and noncompliant transfusions. Conclusion. The majority of gynecologic oncology patients receive transfusions compliant with the restrictive transfusion policy. Morbidity and mortality are not increased with a restrictive transfusion policy. Efforts to improve compliance should focus on limiting transfusions when the hemoglobin is ≥ 7 g/dL and transfusing in single pRBCs unit increments. © 2013 Elsevier Inc. All rights reserved.
Objectives Blood products are vital medical resources often found in short supply. There has been debate on the risks of anemia and benefits of transfusion specifically in critically ill patients. Early studies demonstrated poor surgical outcomes and decreased survival in patients with perioperative ⁎ Corresponding author at: Department of Obstetrics and Gynecology, University of Alabama at Birmingham, 1700 6th Avenue South, Birmingham, AL 35233, USA. Fax: 205975-6174. E-mail address: [email protected] (J.D. Boone). 0090-8258/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2013.10.019
anemia [1–3]. Some studies have shown that blood transfusions are associated with longer hospital stays and impaired recovery [4,5]. In 1999, the Transfusion Requirements in Critical Care (TRICC) trial demonstrated that a restrictive transfusion policy significantly decreased mortality during hospitalization in patients less acutely ill without active coronary ischemia [6]. Since the publication of that trial, Carless et al. confirmed that restricting transfusions does not increase the rates of cardiac events, myocardial infarction, stroke, pneumonia, thromboembolism, and mortality [7]. An additional study by Herbert et al. showed lower rates of worsening organ dysfunction when transfusions were restricted [8]. It is unclear if restrictive transfusion policies can be used
228
J.D. Boone et al. / Gynecologic Oncology 132 (2014) 227–230
in gynecologic oncology patients who have undergone complicated debulking surgeries or received myelosuppressive chemotherapeutic agents and radiation therapy. The TRICC trial showed that transfusions can be restricted in post-operative patients as 39% of these patients were included in the restrictive transfusion arm of the trial. Compliance with a restrictive transfusion policy and patient outcomes has not been studied in gynecologic oncology patients. In 2007, a restrictive transfusion policy was initiated at the University of Alabama at Birmingham (UAB). We sought to evaluate compliance and outcomes with a restrictive transfusion policy in gynecologic oncology patients. Methods After obtaining approval from the Institutional Review Board, a retrospective chart review of gynecologic oncology patients was performed. Patients who received packed red blood cell (pRBC) transfusions from 10/2008 to 9/2011 were included in the study. All hospitalized patients, both surgical and medical, were included. During this time period, a restrictive transfusion policy recommended no pRBC transfusions until hemoglobin concentrations dropped below 7 g/dL, and transfusions were given in one-unit increments. Several exceptions to this transfusion policy exist: intraoperative or day-of-surgery transfusion, symptomatic anemia, irreversible bone marrow suppression, active bleeding in absence of a surgical procedure, severe sepsis, active coronary ischemia, and transfusion on post-operative day one or later after a 1,500 cc or greater intraoperative blood loss. Abstracted data included patient demographics, cancer type and stage, surgical procedure, pre-transfusion hemoglobin values, number of pRBC units transfused, estimated intraoperative blood loss, postoperative complications, and 30-day mortality. Statistical analyses were performed using a Fisher's exact test, and a p value of 0.05 or less was considered statistically significant. Results A total of 582 patients receiving 2,276 transfusions were included in this study. Patient demographics are outlined in Table 1. The mean age was 55.9 years and the majority of patients were Caucasian (57.5%). Ovarian cancer (44.8%) was the most common malignant diagnosis followed by endometrial (22.0%) and cervical cancer (19.8%). Eight-six (14.8%) patients with benign disease and 5 (0.8%) patients with premalignant disease were included in the analysis. The most common surgical procedures are listed in Table 2. The majority of the patients (83.4%) underwent an open abdominal procedure with the most common operation being a total abdominal hysterectomy (TAH) with or without bilateral salpingo-oopherectomy (BSO) and debulking. Transfusion events, number of units transfused, and compliance with the restrictive transfusion policy are detailed in Table 3. The mean pretransfusion hemoglobin was 7.8g/dL (range 3.6–14.6). The mean number of units transfused per person was 3.9 with the mean number of units transfused per transfusion event being 1.6 (range 1–13). Overall, 1,127 (81.1%) of the transfusion events were compliant with the restrictive transfusion policy. Three hundred seven (22.1%) of the transfusion events were issued when the hemoglobin concentration was b 7 g/dL; 820 (59.0%) transfusions were compliant with the restrictive transfusion policy based on the approved exceptions. The most common reasons given as an exception for transfusion were intraoperative or day-ofsurgery transfusion (36.7%), symptomatic anemia either postoperative or chemotherapy-induced (31.6%), and irreversible bone marrow suppression secondary to whole pelvic radiation therapy (14.0%). Two hundred sixty-three (18.9%) transfusion events resulting in 365 units of pRBCs were noncompliant with the restrictive transfusion policy. Reasons for violating the transfusion policy included post-operative hemoglobin concentrations greater than 7 g/dL with less than a 1,500 cc intraoperative blood loss, asymptomatic chemotherapy-induced anemia, anemia NOS, and preoperative anemia. Of the 90 patients receiving
Table 1 Patient demographics. n = 582 Age Mean (years) Range (years) Race Caucasian African–American Other Body mass index Mean (kg/m2) Range Malignancy Ovarian cancer Endometrial cancer Cervical cancer Gastrointestinal cancer Peritoneal cancer Vulvar cancer GTN Vaginal cancer Other Premalignant disease Borderline tumor Cervical dysplasia Benign disease Leiomyoma Ovarian cyst Tuboovarian abscess Endometriosis/adenomyosis Abdominal/vulvar Abscess Ovarian torsion Other
55.9 13–89 335 (57.5%) 235 (40.4%) 12 (2.1%) 29.2 12.1–65.2 491 (84.4%) 220 (44.8%) 108 (22.0) 97 (19.8%) 29 (5.9%) 14 (2.9%) 6 (1.2%) 6 (1.2%) 5 (1.0%) 6 (1.2%) 5 (0.8%) 3 (60.0%) 2 (40.0%) 86 (14.8%) 28 (32.6%) 25 (29.0%) 8 (9.3%) 7 (8.1%) 5 (5.8%) 4 (4.7%) 9 (10.5%)
GTN — gestational trophoblastic neoplasia.
noncompliant transfusions due to an estimated blood loss (EBL) of b1,500 cc, the mean intraoperative blood loss was 556 cc. The 243 surgeries requiring transfusion resulted in a total of 638 units being transfused with 2.6 units transfused per procedure. Patients undergoing an intraoperative transfusion had a slightly higher mean EBL of 1,260 cc compared to 1,020 cc for all patients undergoing a procedure. (Table 4) The most common surgical procedures requiring transfusion involved a TAH±BSO with debulking. Laparoscopic and robotic procedures were rarely associated with a transfusion as only 4 robotic-assisted (RA) laparoscopic procedures received pRBCs. The second part of the restrictive transfusion policy states that transfusions should be ordered in single unit increments unless there is active blood loss. We found that 900 (64.7%) transfusion events were compliant in transfusing one unit at a time followed by hemoglobin
Table 2 Surgical history.
TAH +/− BSO with debulking Bowel/mass resection TAH +/− BSO USO/BSO with debulking TAH +/− BSO with staging Cystoscopy EUA +/− biopsy Vaginectomy/vulvectomy USO/BSO Radical hysterectomy +/− BSO with staging Pelvic exenteration Robotic-assisted TLH/BSO+/−LND Laparoscopic BS0 Other
N = 404 123 62 59 40 38 10 8 6 5 5 5 3 1 39
TAH—total abdominal hysterectomy; BSO—bilateral salpingo-oophorectomy; USO—unilateral salpingo-oophorectomy; TLH—total laparoscopic hysterectomy; LND—lymph node dissection; EUA—exam under anesthesia.
J.D. Boone et al. / Gynecologic Oncology 132 (2014) 227–230 Table 3 Transfusions and compliance with restrictive transfusion policy. Mean hemoglobin (g/dL) Mean Range Transfusion events Total units transfused Compliant Noncompliant Mean units transfused per person Units per transfusion event Mean Range Compliance with restrictive transfusion policy (by event) Compliant (Hgb b7 g/dL) Compliant by exception Intraoperative transfusion on day of surgery Symptomatic anemia Irreversible bone marrow suppression Active bleeding Symptomatic chemotherapy-induced anemia Severe sepsis Transfusion POD1+ with N1500 cc blood loss Active coronary ischemia Noncompliant transfusions (by event) Reason for transfusion Post-operative with b1500 cc blood loss Chemotherapy-induced anemia without symptoms (n = 113) Anemia NOS Pre-operative anemia Compliant with transfusing 1 unit per transfusion event Not compliant with transfusing 1 unit per transfusion event Transfusion compliance by disease type Transfusion for benign disease Compliant Noncompliant Transfusion for malignant disease Compliant Noncompliant
Table 5 Patient complications.
7.8 3.6–14.6 1,390 2,276 1,911 365 3.9 1.6 1–13 1,127 (81.1%) 307 (22.1%) 820 (59.0%) 301 (36.7%) 198 (24.2%) 115 (14.0%) 65 (7.9%) 61 (7.4%) 56 (6.8%) 17 (2.1%) 7 (0.9%) 263 (18.9%) 90 (34.2%) 78 (29.7%) 73 (27.8%) 22 (8.3%) 900 (64.7%) 490 (35.3%) p = 0.83 162 133 (82.1%) 29 (17.9%) 1,228 994 (80.9%) 234 (19.1%)
concentration monitoring then ordering additional transfusions as needed. Four hundred ninety (35.3%) transfusion events were noncompliant and occurred when two or more units were given before a repeat hemoglobin was checked. We compared transfusions for both benign and malignant disease to determine compliance for each disease state. Compliance rates for both benign and malignant disease were similar; roughly 81% of transfusions for both disease states were compliant with the restrictive transfusion policy. The complications analyzed in this study were postoperative infection, thrombotic events, and 30-day mortality (Table 5). Postoperative infection consisted of urinary tract infection, pneumonia, wound infection, abscess, and sepsis. The thrombotic events recorded were either deep vein thrombosis (DVT) or pulmonary thromboembolism (PTE). Only
Table 4 Intraoperative hemodynamic status and transfusion. Total patients requiring transfusion Total surgeries requiring transfusion Total units transfused intraoperative Mean units transfused per procedure EBL per procedure (cc) Mean Range EBL per procedure requiring transfusion (cc) Mean Range EBL intraoperatively with b 1500 cc blood loss (cc) Mean Range
229
233 243 638 2.6 1,020 0–6,000 1,260 50–6,000 556 25–1,400
Postoperative complications per patient Infection Compliant with policy Noncompliant with policy Thrombotic event Compliant with policy Noncompliant with policy 30-day Mortality Compliant with policy Noncompliant with policy Transfusion reaction
n = 582 66 (11.3%) p = 0.122 51 15 29 (5%) p = 0.222 23 6 32 (5.5%) p = 0.335 29 3 5 (0.22%)
five (0.22%) transfusion reactions occurred in the 582 patients. There were 66 patients (11.3%) who developed a postoperative infection, 29 (5%) who were diagnosed with a DVT or PTE, and 32 (5.5%) who died within 30 days of being discharged from the hospital. There was no difference in complication rates in the compliant versus noncompliant transfusions (Table 5).
Discussion Approximately 24million blood products and 10million units of blood are transfused each year, and this resource is becoming increasingly more expensive and less available [9–11]. In the early 1990s, red blood cell transfusions were given to maintain a hemoglobin concentration of 10 g/dL or greater to increase oxygen delivery with the intent of avoiding the detrimental effects of oxygen debt seen in anemic patients [12–14]. Several studies evaluated the safety of blood product administration. Heier et al. showed that if the hemoglobin concentration is greater than 8 g/dL, lung function and volume status are more important to clinical outcome than the hemoglobin concentration [15]. This same study noted that there were no “measurable changes of human cognitive function” after an acute drop in the hemoglobin concentration to 7 g/dL in normovolemic patients [15]. Numerous studies have shown that blood transfusions were associated with increased nosocomial infection rates, multisystem organ failure, longer hospital stays, and mortality [15–18]. The TRICC trial performed by Hebert et al. marked a major change in transfusion medicine because it demonstrated that a restrictive transfusion policy is at least as effective as a liberal transfusion strategy [6]. It is still argued that no single hemoglobin value can serve as a transfusion trigger without accounting for patient and clinical variables [19]. However, in support of the TRICC trial, a systematic review of a few randomized controlled trials on blood transfusion practices concluded that a single laboratory value can serve as a restrictive transfusion indicator in patients without cardiac disease [5]. In 2007, the blood bank at UAB instituted a restrictive transfusion policy allowing transfusion with pRBCs only when the hemoglobin concentration falls below 7 g/dL unless one of the established exceptions is met. Exceptions included intraoperative or day-of-surgery transfusion in the setting of active bleeding, symptomatic anemia (postoperative or chemotherapy-induced), irreversible bone marrow suppression (as in the case of whole pelvic radiation therapy), severe sepsis, active coronary ischemia, and transfusion on postoperative day one or later when the intraoperative blood loss was at least 1,500 cc. Active bleeding such from cervical and vaginal lesions or from the rectum as a result of radiation proctitis was also an exception. For the purpose of this study, patients receiving chemotherapy were considered to have reversible bone marrow suppression; however, the argument can be made for transfusing patients with a low hemoglobin in order to avoid the hemoglobin falling below 7g/dL when myelosuppressive chemotherapeutic agents are used. In that case, an additional 113 units of pRBCs given would have been compliant.
230
J.D. Boone et al. / Gynecologic Oncology 132 (2014) 227–230
In this current study, 81% of pRBC transfusions were compliant with a restrictive transfusion policy in gynecologic oncology patients. The majority of these transfusions were compliant by exception, most commonly because of intraoperative transfusion. Transfusion compliance was similar whether the underlying pathology was malignant or benign demonstrating that surgeons were not more likely to transfuse a patient just because they had a confirmed malignancy. An additional component of the UAB restrictive transfusion policy states that pRBC transfusions are to be given one unit at a time followed by rechecking the patient's hemoglobin concentration before an additional unit is given. Importantly, only 65% of the transfusion events in this study were given in single unit increments. This certainly highlights the tendency of gynecologic oncologists to transfuse at least two or more units of blood without rechecking hemoglobin concentration. This issue is an area for potential improvement in the use of blood products. Six hundred thirty-eight intraoperative transfusions were given during surgery and were considered exceptions to the restrictive transfusion policy given active intraoperative bleeding. More transfusions were administered for malignant disease than benign disease which is expected given the vascular nature of malignant tumors and the more extensive surgery involved with cancer resections. The majority of the surgical cases for malignant disease involved removal of the uterus and/ or bilateral fallopian tubes and ovaries, omentum, and a bowel resection. Mean blood loss during surgery resulting in an intraoperative or same day pRBC transfusion was 1,260cc. Transfusions given in the immediate postoperative period that were not compliant with the restrictive transfusion policy had an average intraoperative blood loss of 556 cc and comprised 34.2% of the noncompliant transfusions. These 90 transfusions indicate a possible area for cost savings and quality improvement. Four patients undergoing robotic procedures received pRBCs. There were three RA-TLH with either unilateral salpingo-oopherectomy (USO) or BSO and one RA-BSO with pelvic lymphadenectomy (LND). Two of the RA-TLH were done for leiomyomatous disease and one for stage 1 endometrial cancer. The RA-BSO with LND was an aborted radical hysterectomy for cervical cancer after advanced stage disease was discovered. In total, three units of pRBCs given after the laparoscopic cases were not compliant with the restrictive transfusion policy having an EBL of less than 100 cc while the other three units given were compliant because of symptomatic anemia and active bleeding from the cervix. We compared complications associated with compliant transfusions versus those that were noncompliant. Bernard et al. showed that an intraoperative transfusion with a single unit of pRBCs was associated with more pneumonia, sepsis, surgical site infection, and increased mortality with transfusion of a second unit of pRBCs [1]. Our analysis included commonly reported complications including postoperative infection, thrombotic events, and 30-day mortality [1,16,18,20]. There were similar postoperative infections, thrombotic events, and 30-day mortality events in the compliant versus noncompliant patients (Table 5). We also noted five transfusion reactions, 0.22%, which is consistent with National Blood Collection and Utilization Survey's adverse transfusion reaction rate of 0.25% [11]. The delivery of cost-effective healthcare is a central focus of medicine today. Our blood bank estimates $220 per unit of leukoreduced red blood cells, and the 263 transfusions not compliant with the restrictive transfusion policy resulted in 365 pRBC unit transfusions totaling a potential $80,300 in savings. As with any retrospective study, one of the major weaknesses is the accuracy of the retrospective review of the medical record. It is difficult to determine if patient's symptoms related to her anemia were always recorded accurately. The noncompliant transfusions could have been compliant because complaints of lightheadedness or dizziness were not documented in the medical record. Physicians are historically inaccurate in estimating intraoperative blood loss [21,22]. This makes it difficult to evaluate an accurate estimated blood loss. Also, it is difficult to control for confounders that may have contributed to the rates of postoperative infections, thrombotic events, and 30-day mortality. Overall,
the gynecologic oncology division at UAB is largely compliant with the hospital's restrictive transfusion policy. It is apparent that more emphasis should be placed on transfusing by single-unit increments to allow time for the patient to respond to the first transfusion. A second important area for improvement includes the 263 transfusions administered when the hemoglobin was greater than or equal to 7 g/dL, and the cost savings associated with appropriate use of that resource. Further research and quality review is needed to ensure compliance with restrictive transfusion policies in gynecologic oncology, as previous retrospective reviews of blood product use have not improved transfusion practices [23,24]. Conflict of interest statement The authors declare that there are no conflicts of interest.
References [1] Bernard AC, Davenport DL, Chang PK, Vaughn TB, Zwischenberger JB. Intraoperative transfusion of 1 U to 2 U packed red blood cells is associated with increased 30-day mortality. J Am Coll Surg 2009;208(5):931–7 [937 e1-2; discussion 938-9]. [2] Oliver E, Carrio ML, Rodriguez-Castro D, Javierre C, Farrero E, Torrado H, et al. Relationships among haemoglobin level, packed red cell transfusion and clinical outcomes in patients after cardiac surgery. Intensive Care Med 2009;35(9):1548–55. [3] Beattie WS, Karkouti K, Wijeysundera DN, Tait G. Risk associated with preoperative anemia in noncardiac surgery: a single-center cohort study. Anesthesiology 2009;110(3):574–81. [4] Vamvakas EC, Blajchman MA. Blood still kills: six strategies to further reduce allogeneic blood transfusion-related mortality. Transfus Med Rev 2010;24(2):77–124. [5] Carson JL, Hill S, Carless P, Herbert P, Henry D. Transfusion triggers: a systematic review of the literature. Transfus Med Rev 2002;16(3):187–99. [6] Herbert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340(6):409–17. [7] Carless PA, Henry DA, Carson JL, Herbert PP, McClelland B, Ker K. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev 2010;10:CD002042. [8] Herbert PC, Yetisir E, Martin C, Blajchman MA, Wells G, Marshall J, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med 2001;29(2):227–34. [9] Kanavos P, Yfantopoulos J, Vandoros C, Politis C. The economics of blood: gift of life or a commodity? Int J Technol Assess Health Care 2006;22(3):338–43. [10] Goodnough LT, Levy JH, Murphy MF. Concepts of blood transfusion in adults. Lancet 2013;381(9880):1845–54. [11] Whitacker BI, Henry RA, Schlumpf K, Schulman J, Green J. Report of the US Department of Health and Human Services. The 2009 national blood collection and utilization survey report. Washington, DC: US Department of Health and Human Services, Office of the Assistant Secretary for Health; 2011 1–58. [12] Russell JA, Phang PT. The oxygen delivery/consumption controversy. Approaches to management of the critically ill. Am J Respir Crit Care Med 1994;149(2 Pt 1):533–7. [13] Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA 1993;270(22):2699–707. [14] Gattinoni L, Brazzi L, Pelosi P, Latini R, Tognoni G, Pesenti A, et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 1995;333(16):1025–32. [15] Heier HE, Bugge W, Hjelmeland K, Soreide E, Sorlie D, Haheim LL. Transfusion vs. alternative treatment modalities in acute bleeding: a systematic review. Acta Anaesthesiol Scand 2006;50(8):920–31. [16] Bochicchio GV, Napolitano L, Joshi M, Bochicchio K, Meyer W, Scalea TM. Outcome analysis of blood product transfusion in trauma patients: a prospective, risk-adjusted study. World J Surg 2008;32(10):2185–9. [17] Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM. Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma 2003;54(5):898–905 [discussion 905–7]. [18] Zilberberg MD, Shorr AF. Effect of a restrictive transfusion strategy on transfusionattributable severe acute complications and costs in the US ICUs: a model simulation. BMC Health Serv Res 2007;7:138. [19] Goodnough LT, Despotis JG, Hogue CW, Ferguson TB. On the need for improved transfusion indicators in cardiac surgery. Ann Thorac Surg 1995;60(2):473–80. [20] Abu-Rustum NR, Richard S, Wilton A, Lev G, Sonoda Y, Hensley ML, et al. Transfusion utilization during adnexal or peritoneal cancer surgery: effects on symptomatic venous thromboembolism and survival. Gynecol Oncol 2005;99(2):320–6. [21] Ashburn JC, Harrison T, Ham JJ, Strote J. Emergency physician estimation of blood loss. West J Emerg Med 2012;13(4):376–9. [22] Santoso JT, Dinh TA, Omar S, Hannigan EV. Surgical blood loss in abdominal hysterectomy. Gynecol Oncol 2001;82(2):364–6. [23] Goodnough LT, Audet AM. Utilization review for red cell transfusions. Are we just going through the motions? Arch Pathol Lab Med 1996;120(9):802–3. [24] Haspel RL, Uhl L. How do I audit hospital blood product utilization? Transfusion 2012;52(2):227–30.
