The risks associated with allogeneic blood transfusion are well known and have precipitated ongoing innovation in surgical blood conservation. Numerous alternatives to allogeneic blood transfusion are feasible for conserving blood in major surgery. The establishment of transfusion practice standards,1,2 advances in surgical techniques,3 perioperative blood recovery,4 and preoperative autologous donation (PAD),5 have led to improvements in the safety of major surgical procedures by reducing patient exposure to allogeneic blood transfusion. Because the patient's own blood is considered to be the safest blood,6 PAD has become a standard of care in certain major elective orthopedic surgical procedures.5,7 Numerous studies have demonstrated the effectiveness of PAD in reducing patient exposure to allogeneic blood.5,8-11
Despite the broad acceptance of PAD as a treatment option for patients undergoing major elective surgery, the high cost, logistic challenges, and suboptimal stimulation of endogenous erythropoiesis in response to serial phlebotomy have raised concerns.12 Further, PAD has been shown to compromise hemoglobin (Hb) levels, particularly in anemic patients who are already at higher risk for allogeneic blood transfusion.13 In orthopedic surgery patients, administration of recombinant human erythropoietin (Epoetin alfa) has been shown to increase reticulocyte count and Hb and hematocrit (HCT) levels, thereby reducing patient exposure to allogeneic blood transfusion in placebo-controlled studies.14,16 Another study demonstrated that a weekly regimen of 600 International Units (IU)/kg (approximately 40,000 IU in total) Epoetin alfa administered subcutaneously (SC) for a total of four doses was equally safe and effective as a daily regimen of 300 lU/kg (15 doses), and was less costly.14
Therefore, a multicenter, randomized, parallel-group study was conducted in anemic total joint arthroplasty patients comparing a weekly regimen of Epoetin alfa to participation in a PAD program.
MATERIALS AND METHODS
Study Design ana Patients. A prospective, randomized, open-label, parallel-group study was conducted at 58 sites in the United States. All investigators and centers participating in the study were monitored by The R. W. Johnson Pharmaceutical Research Institute (Raritan, NJ) throughout the duration of the study to ensure that the data were accurate, reliable, and verifiable. The study was approved by the institutional review board of each institution participating in the trial. A total of 490 out of a planned 1100 patients were enrolled at 58 medical centers in the United States between March 1996 and August 1997. The study was terminated early for administrative reasons.
Patient eligibility was determined during the screening visit (i.e., 2 weeks before the start of study treatment). Patients enrolled in the study were ≥ 18 years of age, were scheduled for total hip or knee replacement surgery, had a pretreatment Hb level ≥ 11 to ≤ 13 g/dL, were willing to participate in a PAD program, were in good general health, and had provided informed consent. Female patients were postmenopausal for > 1 year, surgically sterile, or using an acceptable form of birth control. Patients were excluded for clinically significant systemic disease or laboratory chemistry abnormalities, any primary hematologic disease, history of seizure disorder, uncontrolled hypertension (i.e., diastolic blood pressure ≥ 90 mm Hg), recent gastrointestinal or intracranial bleeding, or any contraindication to anticoagulant use.
Patients were centrally randomized and stratified by study center to one of two treatment groups: an Epoetin alfatreated group or a PAD group. In the Epoetin alfa arm, patients were administered Epoetin alfa (PROCRIT®, Ortho Biotech Inc., Raritan, NJ) 600 IU/kg weekly by SC injection beginning 21 days before surgery (Day -21), and on Days -14, and -7, and on the day of surgery (Day 0). For patients randomized to participate in PAD, blood donation was conducted according to each study center's standard of care and in accordance with the standards of the American Association of Blood Banks.
In addition, beginning with or before the first dose of Epoetin alfa or the first visit for autologous donation and continuing through the day of surgery, patients received a polysaccharide-iron complex (Niferex, Central Pharmaceuticals, Inc., Seymour, G?) that provided a minimum of 200 mg of elemental iron per day. Patients received anticoagulation therapy based on each investigator's practice standards.
Patients were screened initially at least 3 weeks before the scheduled surgery date, and screening procedures were performed within 2 weeks before the start of study medication or at the first PAD visit. A complete blood count was measured on the day of surgery, on postoperative Day 1, and at discharge. Transfusion data and other information on blood management were collected perioperatively and postoperatively. All investigators followed institutional standard operating procedures regarding surgical blood conservation practices wbere every effort was made to avoid transfusion unless warranted by clinical symptoms.
Statistical Methods, The sample size was calculated to detect a percentage of patients transfused that is 8% higher in the PAD group than in the Epoetin alfa group using the arcsin approximation to the normal distribution. Using data from previous studies, it was assumed that the percentage of patients transfused in the Epoetin alfa-treated group would be 15%. It was determined that a total sample size of 1000 (500 patients per study arm) could detect this 8% increase in transfusion rate with a onesided significance level of .025. To account for a possible 10% of patients not undergoing planned surgery, 550 patients would be required in each group.
The primary measure of efficacy was the proportion of patients in each treatment group requiring perioperative allogeneic blood transfusion. Secondary end points included comparison of mean number of units transfused and Hb levels, measured preoperatively, postoperatively, and at discharge. The intent-totreat population included patients who were randomly assigned to a treatment group. The modified intent-to-treat population included patients who were randomized to a treatment group and who underwent surgery as planned.
Analyses of treatment efficacy were performed on a modified intent-to-treat population. The proportion of patients in each treatment group requiring allogeneic blood transfusion was compared using the Pearson chi-square test. The number of units transfused was evaluated on a per-patient basis and compared using two-sample t tests. Between-group differences in mean Hb levels were assessed using two-sample t tests.
Safety was evaluated by recording the incidence and severity of adverse events and by evaluating clinical laboratory tests and vital signs. Adverse events were monitored continuously from the onset of treatment until hospital discbarge. All safety analyses were conducted on the intent-to-treat population, which included all treated patients. Adverse events were classified according to World Health Organization definitions. Adverse events were categorized by body system, preferred term, severity, and relationship to study drug. Serious adverse events and deaths were recorded.
Patient Demographics and Outcome Summary. Two hundred forty-one patients in the Epoetin alfa-treated group and 249 patients in the PAD group were enrolled in the study. Overall, 87% of patients in each treatment group completed the study. Sixty-two patients (32 patients in the Epoetin alfa group and 30 patients in the PAD group) withdrew from the study before going to surgery. Of the 241 patients randomized to Epoetin alfa, 204 (85%) received all four doses of study medication, 227 patients (94%) received at least one dose of Epoetin alfa, and 14 patients withdrew from the study before receiving any study medication. Demographic and baseline characteristics of patients in the two treatment groups who went to surgery (modified intent-to-treat population) were similar and are shown in Table 1.
The majority of surgical procedures were primary knee and hip arthroplasty (Table 2). The estimated intraoperative blood loss was 391 ± 427 mL in the Epoetin alfa-treated group and 302 ± 267 mL in the PAD group. Blood recovery was performed in a small percentage of the patients in each group. The volume of blood collected by postoperative cell salvage and the volume of recovered blood transfused were similar in the two treatment groups.
Transfusion Data. Transfusion requirements for the two groups are illustrated in Figure 1. Only 27 (12.9%) of the Epoetin alfa-treated patients were transfused with allogeneic blood whereas 42 (19.2%) of the PAD patients (P = .078) were transfused with allogeneic blood. In addition, 156 (71.2%) of the PAD patients received autologous blood. Therefore, 163 (74.4%) of PAD patients and 27 (12.9%) of Epoetin alfa-treated patients required transfusions.
Patient demographic and baseline characteristics*
Surgical procedures, estimated blood loss, and blood salvage*
The mean and total number of units of blood transfused are presented in Table 3. A mean of 0.36 unit of allogeneic blood per patient was transfused in the PAD group compared with 0.25 unit in the Epoetin alfatreated group; this difference was not statistically significant. However, PAD patients received a total of 325 units of blood (79 units allogeneic and 246 units autologous), whereas Epoetin alfa-treated patients received only 54 units of blood (53 units allogeneic and 1 unit autologous - one patient randomized to Epoetin alfa group mistakenly donated 1 unit of autologous blood). Mean values for pretransfusion Hb levels ("transfusion triggers") for patients receiving allogeneic transfusions were similar between groups (Epoetin alfa-treated patients, 8.4 ± 1.05 g/dL; PAD patients, 8.3 ± 0.84 g/dL). The mean pretransfusion Hb values for allogeneic (8.3 ± 0.84 g/dL) versus autologous (8.5 ± 0.94 g/dL) transfusion in PAD patients were also similar (Table 4). The reasons for transfusions were similar for the PAD group and the Epoetin alfa group. The most common reason for transfusion was anemia, which was attributed to perioperative blood loss (Table 5).
Fig 1 : Percent of patients transfused with allogeneic blood or autologous blood, and percent of patients who received any transfusion (total transfusion). Patients were treated weekly with an average of 40,000 International Units of recombinant human erythropoietin (Epoetin alfa) on Days -21, -14, -7, and on the day of surgery, or participated in preoperative autologous donation (PAD). *P= .078 between treatment group difference.
Summary of units transfused per patient*
Pretransfusion hemoglobin ("transfusion trigger")
Perioperative Hemoglobin Levels. At baseline, mean Hb values in both treatment groups were 12.3 ± 0.6 g/dL, Before surgery, the mean Hh value increased to 13.8 ± 1.2 g/dL in the Epoetin alfa group but decreased to 11.1 ± 1.0 g/dL in the PAD group which was a statistically significant difference (2.7 g/dL difference between treatment groups; P < .0001) (Fig 2). Postoperatively, the mean Hb value in Epoetin alfa-treated patients, 11.0 ± 1.4 g/dL, remained significantly higher compared with the mean Hb value in PAD patients, 9.2 ± 1.1 g/dL (1.8 g/dL difference between treatment groups; P < .0001). Moreover, despite a greater volume of perioperative blood transfusion in the PAD group (325 units in PAD group versus 54 units in Epoetin alfa group), mean Hb values at discharge were approximately 1 g/dL higher in the Epoetin alfa group than values in the PAD group (Epoetin alfa, 10.5 ± 1.3 g/dL versus PAD, 9.5 ± 1.1 g/dL, P < .0001 between treatment group difference).
Safety Results. Both treatments were safe and well tolerated. The overall incidence of adverse events was similar between treatment groups: 149 (62%) adverse events were reported in the Epoetin alfa-treated group compared with 131 (53%) adverse events in the PAD group. A summary of perioperative adverse events that occurred in at least 2.5% of patients in any treatment group is provided in Table 6. The most common adverse events reported in both groups were nausea, constipation, and fever. These adverse events are commonly reported in the perioperative setting and are generally related to surgery, anesthesia, and analgesia. Dizziness was reported more commonly in the PAD group (9%) compared with the Epoetin alfa group (2%). Most adverse events were considered mild or moderate in severity.
Thirty-one (6.3%) adverse events were considered serious or unexpected: 19 (7.9%) in the Epoetin alfa group and 12 (4.8%) in the PAD group. Of these events, 11 (2.2%) were thrombotic/vascular in nature (6 were in the Epoetin alfa-treated group and 5 were in the PAD group). Two patients in the PAD group developed pulmonary emboli on the day following surgery and died (the only deaths in the study). An additional PAD patient developed pulmonary emboli 3 days after surgery. None of the Epoetin alfa-treated patients had pulmonary emboli. Three patients in the Epoetin alfa-treated group and 2 patients in the PAD group had deep vein thrombosis (DVT). One report of DVT, considered by the investigator to be unrelated to treatment, occurred 5 days after surgery in an Epoetin alfa-treated patient who refused anticoagulation therapy, but did not result in study discontinuation. The two additional reports of DVT for patients treated with Epoetin alfa were serious, and were considered possibly related to treatment. The DVTs occurring in both patients in the PAD group were reported to be serious; 1 of these patients had also developed pulmonary emboli. Both patients went on to complete the study. One patient in the PAD group developed angina pectoris of marked severity and was withdrawn from the study. In the Epoetin alfa group, there were 2 patients with cerebrovascular accident and 1 patient with myocardial infarction. These events were considered to be unrelated to treatment, and these patients went on to complete the study.
In placebo-controlled studies, Epoetin alfa has been shown to reduce the need for transfusion in anemic patients undergoing orthopedic surgery by increasing the preoperative Hb and the rate of postoperative erythropoietic recovery.14"17 This study sought to compare, for the first time, Epoetin alfa to the standard of care in total joint arthroplasty, PAD.
Three weeks prior to surgery mean Hb levels were the same for both patient groups. However, Epoetin alfa administered in 4 doses (600 lU/kg on Days -21,-14, and -7 prior to surgery, and on the day of surgery) increased the mean preoperative Hb concentration by approximately 1.5 g/dL, whereas the PAD group exhibited a decrease in mean preoperative Hb concentration of approximately 1.2 g/dL. This finding is consistent with other investigations demonstrating that PAD reduces preoperative Hb and HCT.SJ8-20 The current study also confirms that the endogenous erythropoietic response to phlebotomy was insufficient to compensate for the RBC mass lost during predonation. In patients with low baseline Hb, PAD can exacerbate anemia to clinically undesirable levels. In contrast, Beris has shown that Epoetin alfa can reverse the reduction in Hb levels observed during serial phlebotomy in PAD patients.21 In addition to significantly higher mean preoperative Hb levels, the mean postoperative and discharge Hb levels were significantly higher in the Epoetin alfa-treated patients compared with PAD patients (P < .0001).
Fig 2: Hemoglobin levels from baseline through discharge in recombinant human erythropoietin (Epoetin alfa) and preoperative autologous donation (PAD) patient groups. Patients were treated weekly with approximately 40,000 International Units of Epoetin alfa on Days -21, -14, -7, and on the day of surgery, or participated in PAD. BL = Baseline; Preop = Preoperative; Postop = Postoperative. All data are means ± standard deviation. Difference between treatment groups in mean Hb values, BL = 0; Preop = 2.7 g/dL; Postop = 1 .8 g/dL; Discharge = 1 .0 g/dL. »Statistically significant difference between treatment groups using a 2sample t test (P < .0001).
Reasons for transfusions
Therefore, throughout the study, PAD patients had significantly lower mean Hb values compared with the Epoetin alfa group (P < .0001). Furthermore, at discharge, mean Hb levels were 1 g/dL lower in the PAD group despite the fact that PAD patients were administered 246 units of autologous blood. These differences in between Epoetin alfa patients and patients, at every point measured, port the need for ongoing clinical research that is investigating the tionship of higher Hb to recuperative power.
Incidence of adverse events observed in ≥ 2.5% of patients*
In this study, Epoetin patients exhibited a 6.3% reduction percentage of patients receiving geneic transfusion (12.9% 19.2%) and a 26-unit reduction in number of allogeneic units given (53 units versus 79 units). The difference allogeneic transfusion approached nificance (P = .078; a sample size 1100 patients was required to demonstrate an 8% reduction in allogeneic transfusion). These results occurred despite similar "transfusion triggers," similar reasons for transfusion, similar amounts of estimated blood loss, and despite autologous transfusions in 71% of patients in the PAD group. Considering autologous transfusion, which accounted for 246 additional units received, presents a more dramatic comparison in the receipt of overall transfusion requirements: 12.9% of Epoetin alfa-treated patients received a total of 54 units, compared with 74.4% of PAD patients receiving a total of 325 units.
The effectiveness of PAD in reducing patient exposure to allogeneic blood transfusion should be weighed against the limitations associated with its use. For example, the endogenous erythropoietic response to serial phlebotomy is often inadequate,12,22,23 resulting in progressive decreases in Hb and HCT levels up to the time of surgery.5,13,19 Low baseline Hb concentrations e.g., anemia) and lack of autodonated blood were found to be the most important predictors of allogeneic transfusion in the recently completed prospective study of 9482 total joint replacement patients.18 In addition, this study demonstrated that patients witli baseline Hb levels ≤ 13 g/dL received more allogeneic transfusions, even if they had participated in a PAD program. Results from Epoetin alfa studies have also demonstrated that anemic patients with Hb levels ≤ 13 g/dL have a narrower safety margin for perioperative blood loss, are more likely to require allogeneic blood, and benefit the most from Epoetin alfa, compared with patients with Hb levels > 13 g/dL.15"17
Transfusion of autologous blood is not without risks. In a recent review of all transfusion reactions reponed to The Cleveland Clinic Hospital's transfusion service during a 6-year period. Domen reported adverse reactions including febrile nonhemolytic and allergic reactions.24 Further, immune and nonimmune hemolysis,25·26 coagulopathies,27 fluid overload,28 and bacterial contamination29 have all been reported. Importantly, clerical errors also place autologous blood donors at risk of receiving allogeneic blood.24·30
Preoperative autologous blood donation is also complicated by the logistics of the PAD process. Preoperative autologous blood donation is time-consuming and inconvenient with respect to patient travel to and from the clinic and with respect to time patients spend during the actual autodonation process. Although the 3dose preoperative schedule of Epoetin alfa presents a new paradigm to the orthopedic surgeon, the time necessary to receive a SC injection is considerably less than the time required for the autodonation process. The PAD program also requires adniinistrative support (i.e., infrastructure), including program management, staffing, and a blood storage facility. Finally, the costeffectiveness of PAD has been questioned31,32 because of the substantial amount of blood that is discarded (i.e., collected but never used).11,18,33"35 Although in the present study the proportion of collected units discarded was not determined, previous investigators have reported that a high proportion of autologous blood is wasted.18,33,36,37
This study contributes substantially to the safety record of a regimen of 600 IU/kg of Epoetin alfa. The frequencies of adverse events were similar between treatment groups. The most common adverse events were nausea, constipation, and fever, which are expected in this patient population undergoing major elective surgery and receiving oral iron supplementation. Patients who participated in PAD were more likely to complain of dizziness than their counterparts receiving Epoetin alfa. These safety data are consistent with other investigations of perioperative Epoetin alfa.14-17
Despite aggressive anticoagulation therapy, the risks of DVT and pulmonary emboli are not eliminated in this patient population.38·39 The incidence of thrombotic/vascular events was similar between groups, occurring in 2.5% of Epoetin alfa-treated patients versus 2.0% of PAD patients. The incidence of DVT reported in this study is consistent with previous reports.40-42 Three patients in the PAD group exhibited pulmonary emboli, which resulted in the only two deaths that occurred in this study.
This study is the first randomized clinical trial directly comparing Epoetin alfa with PAD. Epoetin alfa administered at 600 IU/kg SC was safe and, compared with PAD, significantly increased both preoperative and postoperative Hb levels. In addition, a reduction in allogeneic transfusion requirements was observed in anemic patients receiving Epoetin alfa compared with patients enrolled in a PAD program.
1 . Goodnough LT, Despotie GJ. Establishing practice guidelines for surgical blood management. Am J Surg. 1995; 170(suppl 6?): 16S^20S.
2. Spence RK. Surgical red blood cell transfusion practice policies. Blood Management Practice Guidelines Conference. Am J Surg. 1995; 170(stippl6A):3S-15S.
3. Wyman A, Rogers K. Randomized trial of laser scalpel for modified radical mastectomy. Br J Surg. 1993;80:871-873.
4. Han CD, Shin DE. Postoperative blood salvage and reinfusion after total joint arthroplasty. J Arthroplasty. 1997; 12:511-516.
5. Goodnough LT, Shafron D, Marcus RE. The impact of preoperative autologous blood donation on orthopaedic surgical practice. Vox Sang. 1990; 59:65-69.
6. Surgenor D. The patient's blood is the safest blood [editorial]. N Engl J Med. 1987; 316:542-544.
7. Woolson ST, Watt JM. Use of autologous blood in total hip replacement. A comprehensive program. / Bone Joint Surg Am. 1991; 73:76-80.
8. Murray DJ, Forbes RB, Titone MB, Weinstein SL. Transfusion management in pediatric and adolescent scoliosis surgery. Efficacy of autologous blood. Spine. 1997; 22:2735-2740.
9. Healy JC, Frankforter SA. Graves BK, Reddy RL, Beck JR. Preoperative autologous blood donation in total-hip arthroplasty. A costeffectiveness analysis. Arch Pathol Lab Med. 1994;118:465-470.
10. Woolson ST, Pottorff G. Use of preoperatively deposited autologous blood for total knee replacement. Orthopedics. 1993; 16:137-141.
11. Renner SW, Howanitz PJ, Bachner P. Preoperative autologous blood donation in 612 hospitals. A College of American Pathologists' Q-Probes study of quality issues in transfusion practice. Arch Pathol Lab Med. 1992; 1 16:613619.
12. Kickler TS, Spivak JL. Effect of repeated whole blood donations on serum immunoreactive erythropoietin levels in autologous donors. JAMA. 1988;260:65-67.
13. Kanter MH, van Maanen D, Anders KH. et al. Preoperative autologous blood donations before elective hysterectomy. JAMA. 1996; 276:798-801.
14. Goldberg MA, McCutchen JW, Jove M, et al. A safety and efficacy comparison study of two dosing regimens of Epoetin alfa in patients undergoing major orthopedic surgery. Am J Orthop. 1996; 25:544-552.
15. de Andrade J. Jove M, Landon G, et al. Baseline hemoglobin as a predictor of risk of transfusion and response to Epoetin alfa in orthopedic surgery patients. Am J Orthop. 1996; 25:533-542.
16. Fans P, Ritter M, Abels R, The American Erythropoietin Study Group. The effects of recombinant human erythropoietin on perioperative transfusion requirements in patients having a major orthopaedic operation. J Bone Joint Surg Am. 1996; 78-A.62-72.
17. Canadian Orthopedic Perioperative Erythropoietin Study Group. Effectiveness of perioperative recombinant human erythropoietin in elective hip replacement. Lancet. 1993; 341:1227-1232.
18. Bierbaum BE, Galante JO, Rubash HE, Tooms RE1 Welch RB. Prediction of red cell transfusion in orthopaedic surgery. Paper presented at 65th Annual Meeting of the American Academy of Orthopaedic Surgeons: March 19, 1998-March 23, 1998; New Orleans, LA. Paper 380.
19. Covens A, Pinkerton P, Osborne R, DePetrillo A. Review of autologous and allogeneic blood transfusion practices in patients undergoing radical hysterectomy. Eur J Gynaecol Oncol. 1997; 18:449-452.
20. Cohen JA, Brecher ME. Preoperative autologous blood donation: benefit or detriment? A mathematical analysis. Transfusion. 1995; 35:640-644.
21. Beris P. Epoetin alfa as an adjuvant to autologous blood donation. Semin Hematol. 1996;33:27-29.
22. Goodnough LT, Brittenham GM. Limitations of the erythropoietic response to serial phlebotomy, implications for autologous blood donor programs. J Lab Clin Med. 1990; 115:28-35.
23. Goodnough L, Bravo J, Hsueh Y, Keating L, Brittenham G, Red blood cell mass in autologous and homologous blood units. Implications for risk/benefit assessment of autologous blood crossover and directed blood transfusion. Transfusion. 1989;29:821-822.
24. Domen RE. Adverse reactions associated with autologous blood transfusion: evaluation and incidence at a large academic hospital. Transfusion. 1998; 38:296-300.
25. Baussaud V, Mentec H, Fourcade C. Hemolysis after autologous transfusion. Ann Intern Med. 1996; 124:931-932.
26. Cregan P, Donegan E, Gotelli G. Hemolytic transfusion reaction following transfusion of frozen and washed autologous red cells. Transfusion, 1991; 31:172-175.
27. Murray DJ, Gress K, Weinstein SL. Coagulopathy after reinfusion of autologous scavenged red blood cells. Anesth Analg. 1992; 75:125-129.
28. Popovsky MA, Audet AM, Andrzejewski C Jr. Transfusion-associated circulatory overload in orthopedic surgery patients: a multi-institutional study, lmmunohematology. 1996; 12:8789.
29. Richards C, Kolins J, Trindale CD. Autologous transfusion-transmitted Yersinia enterocoiitica. JAMA. 1992; 268:252-154.
30. Sculco TP. Blood management in orthopedic surgery. Am J Surg. 1995; I70(suppl 6A):60S-63S.
31. Etchason J, Petz L, Keeler E, et al. The cost effectiveness of preoperative autologous blood donations [published comment appears in N Engl J Med. 1995; 332:740-742]. N Engl J Med. 1995;332:719-724.
32. AuBuchon JP, Birkmeyer JD. Controversies in transfusion medicine. Is autologous blood transfusion worth the cost? Con Transfusion. 1994; 34:79-83.
33. Bernstein L. Coles M, Ganata A. The Bridgeport Hospital experience with autologous transfusion in orthopedic surgery. Orthopedics. 1997; 20:677-680.
34. Goh M, Kleer CG, Kielczewski P, et al. Autologous blood donation prior to anatomical radical retropubic prostatectomy: is it necessary? Urology. 1997;49:569-573.
35. Lemos MJ, Healy WL. Blood transfusion in orthopaedic operations. J Bom Joint Surg Am. 1996; 78:1260-1270.
36. Knight JL, Sherer D, Guo J. Blood transfusion strategies for total knee arthroplasty: minimizing autologous blood wastage, risk of homologous blood transfusion, and transfusión cost. J Arthroplasty. 1998; 13:70-76.
37. Pinkerton PH. Use of autologous blood in support of orthopaedic surgery using a hospitalbased autologous donor programme. Transfus Med. 1995;5:139-144.
38. Wilson MG, Pei Linda F, Malone KM, et al. Fixed low-dose versus adjusted higher-dose warfarin following orthopedic surgery. / Arthroplasty. 1994; 9:127-130.
39. Wells PS, Lensing AW. Davidson BL, Prins MH, Hush J. Accuracy of ultrasound for the diagnosis of deep venous thrombosis in asymptomatic patients after orthopedic surgery. A meta-analysis. An« Intern Med. 1995; 122:47-53.
40. Davidson BL, Elliott CG, Leasing AW. Low accuracy of color Doppler ultrasound in the detection of proximal leg vein thrombosis in asymptomatic high-risk patients. Ann intern Med. 1992; 117:735-738.
41. Hull R, Raskob G. Pineo G, et al. A comparison of subcutaneous low-molecular-weight heparin with warfarin sodium for prophylaxis against deep- vein thrombosis after hip or knee implantation. N Engl J Med. 1993; 329:13701376.
42. Imperiale TF, Speroff T A meta-analysis of methods to prevent venous thromboembolism following total hip replacement [published erratum appears in JAMA 1995; 273:288]. JAMA. 1994;271:1780-1785.
The authors thank the nurses, study coordinators, and patients for their participation. In addition, we are grateful to the primary study nwestigator at each of the participating institutions: H. Steinman, MD, Morton Plant Hospital; C. Colwell, MD, La Jolla, CA; W. Lamer, MD, Orthopedic International, LTD; W. Bose. MD. Alabama Orthopedic Clinic; R. Brand, MD, Augusta Orthopedic Specialists; M. Jove, MD, Atlanta Knee and Sports Medicine; P. Lotke, MD, Hospital of the University of Pennsylvania; E. Miller, MD, Wellington Orthopaedic and Sports Medicine; D. Tsukayama, MD, Hennepin County Medical Center; T. Monk, MD, Washington University School of Medicine; M. Brodersen, MD, Mayo Clinic Jacksonville; P. Kirk, MD, University Orthopaedic Consultants of Cincinnati; C. Savory, MD, Hughston Sports Medicine Clinic; R. Thomberry. MD. Tallahasse Orthopedic Surgical Center; J. McCutchen, MD, Jewett Orthopaedic Clinic; H. Cates. MD, St. Mary's Hospital; A. Rosenberg, MD, Rush Presbyterian/St. Lukes Medical Center; K. Joffe, MD, Birmingham, AL; D. Fisher, MD, Orthopedics Indianapolis; R. Friedman, MD, Medical University of South Carolina; G. Wang, MD. University of Virginia HSC; P. Ireland, MD, Indianapolis, IN; M. Ritter, MD, The Center for Hip and Knee Surgery; C. Stowell, MD, Massachusetts General Hospital; S. Belknap, MD, The University of Illinois; D. Armstrong, MD, Mesa, AZ; R. Emerson, MD, Piano, TX; C. Kavolus, MD, Piedmont Orthopaedics; D. Griffin, MD, Vero Beach Orthopedic Group; J. Bargren, MD, The Orthopedic Center; B. Magsamen, MD, Fort Collins, CO; R. Buechel, MD, Bone and Joint Clinic; J. Davies, MD, Milwaukee Orthopedic Group, LTD; J. Benjamin, MD, University of Arizona; R. Strain, MD, Memorial Regional; H. Reynolds. MD. Webster Orthopaedic Medical Group; R. Truluck, MD, The Orthopeadic Clinic of Columbia; K. Chillag, MD, The Moore Orthopaedic Clinic, PA; H. Simon. MD, Toms River, NJ; W. Goldstein, MD, Center for Orthopaedic Surgery; M. Stachniw, MD, Galesburg, IL; A. Spitzer, MD, Centinela Hospital Medical Center; R, B. Sorrels, MD, Little Rock, AR; R. Conn, MD, Southern Bone & Joint Specialists; D. Rhoads, MD, Albany Orthopedics; T. Sculco. MD, Hospital for Special Surgery: D. Ayers, MD, Dept. of Orthopedic Surgery, SUNY; F. Burke, MD, Bluegrass Orthopaedics; J. Murphy, MD, North Alabama Bone & Joint Clinic; C. Creasman, MD, Arizona Orthopedic and Fracture Surgeons: J. Biundo, MD, New Orleans, LA; L. Pollak, MD, Sharpe Memorial Hospital; R. Barrack, MD, Tulane Medical School; C. Hikes, MD, Portland Joint Reconstruction Clinic; R. Johnston, MD, Park Nicollet Clinic; R. Spence. MD, Staten Island University Hospital; H. Kim, MD, Whittier. CA.
Patient demographic and baseline characteristics*
Surgical procedures, estimated blood loss, and blood salvage*
Summary of units transfused per patient*
Pretransfusion hemoglobin ("transfusion trigger")
Reasons for transfusions
Incidence of adverse events observed in ≥ 2.5% of patients*