Preoperative Autologous Blood Donation Program Research Articles

A survey on preoperative autologous blood donation policy in bone marrow stem cell donors in Italy.

The high safety of homologous blood components, together with the introduction of the Patient Blood Management strategy, has led to the progressive abandonment of preoperative autologous blood donation (PAD) in surgery. Furthermore, recent scientific publications provide evidence about the non-usefulness of PAD in the collection of hematopoietic stem cells (HSC) from bone marrow (BM), also in consideration of harvest procedure safety. Nevertheless, no conclusive studies have been published yet. Blood Establishments (BE) and Bone Marrow Collection Centers (BMCC) participated in a specific qualitative survey proposed by Italian National Blood and Transplant centers with the support of the relevant Italian Scientific Societies. The survey aimed at evaluating the policy adopted for PAD in related and unrelated adult HSC donors in Italy during the period 2018-2020. Forty-one BE corresponding to 37 BMCC filled in the questionnaire. Of 830 BM donors, 661 (80%) underwent 1063 PAD (mean 1.6 PAD/donor). The remaining 169 donors (20%) underwent BM harvest without PAD. No serious adverse events were reported for either donor group. In the case of ineligibility of donors for the PAD program, due to low hemoglobin values, 7/10 centers shifted donors to peripheral blood stem cell collection and three centers chose a different donor. Remarkably, only 51% of the PAD units requested were eventually transfused during the BM harvest process. Finally, the iron support policy among centers was heterogeneous. The results of this survey show that PAD is heterogeneously applied in Italian BMCC, as in other countries. However, all BMCC except two are willing to adopt a Patient Blood Management strategy as an alternative approach to adult related and unrelated BM donor harvests.

Safety and efficacy of preoperative autologous blood donation for high-risk pregnant women: experience of a large university hospital in Japan.

Preoperative autologous blood donation (PAD) has the advantages over allogeneic blood transfusion of theoretically no risk of viral infection and alloimmunization. However, there are some concerns regarding PAD in pregnant women, as they sometimes become anemic and adverse effects such as low blood pressure could be harmful to fetuses. In our hospital, the PAD program was implemented in 2006 and has been used in pregnant women at high risk of massive hemorrhage. In this study, the safety of PAD in pregnant women and its efficacy for avoiding allogeneic blood transfusion were investigated. The hospital records of pregnant women who delivered at our hospital from January 2009 to June 2012 were reviewed and those who were enrolled in the PAD program for predicted massive hemorrhage were analyzed. Among the total of 3095 deliveries, 69 cases enrolled in the PAD program were analyzed. Blood donation was performed 189 times for the 69 cases. The median donated blood volume was 1200 mL (range, 400-2000). The mean blood loss during delivery was 1976 ± 1654 mL. Autologous blood was transfused in 64 cases. Allogeneic blood transfusion was required in five cases of massive blood loss exceeding 5000 mL. In the other 64 cases, no additional allogeneic blood transfusion was required. No adverse events were observed in either the pregnant women or fetuses. For pregnant women at a high risk of massive hemorrhage, our PAD program was safe and effective for avoiding allogeneic blood transfusion.

Results of a preoperative autologous blood donation program for patients undergoing elective major spine surgery

Background and objectivesSpinal surgery has been shown to have a high blood transfusion requirement. Preoperative autologous blood donation (PABD) is a strategy to reduce the allogeneic transfusions in this subset of patients. Material and methodsWe retrospectively reviewed transfusion outcome of patients undergoing elective major spinal surgery from 2005 to 2011, and included in the PABD program. Transfusion outcome was compared with a group of patients that did not enter in the program during the same period. ResultsA total of 148 patients were included in the program during the analyzed period. Patients in the PABD program benefited from reduced exposure to allogeneic blood (Odds Ratio: 0.077, 95% confidence interval 0.043–0.140). However, 12.16% (n=18) of these patients received also allogeneic blood (total 40 red blood cell units). Univariate analysis showed the following parameters as significantly predictors of transfusion: inclusion in the program (p<0.000), number of levels fused (Odds Ratio: 1.143, p=0.010), and number of autologous red blood cells donated (Odds Ratio: 1.906, p<0.000). ConclusionsThe preoperative autologous blood donation program designed in our hospital was effective for reducing allogeneic transfusion in mostly young patients under major elective spinal surgery. However and as expected, their inclusion in the program increased the risk to be transfused.

Predonated autologous blood transfusion in elective orthopaedic surgery: A Hospital experience

Background: The use of homologous blood carries significant risk of viral infections and immune-mediated reactions. Preoperative autologous blood donation is an attractive alternative to homologous transfusion and has become common in elective orthopaedic surgery. Objective: To present our experience with the use of preoperative autologous blood donation program in elective orthopaedic surgery. Design/ Setting: A prospective study of patient charts and blood bank records carried out between March 1998 and August 1999 at the Department of Orthopaedic Surgery, Yopougon Teaching Hospital, Abidjan, Côte d\'Ivoire. Patients and methods: Twenty patients scheduled for elective orthopaedic procedures entered the program. There were 13 men and seven women with a mean age of 33.1 years. Fifteen of them were operated on and transfused. Main outcome measures: The number of units of blood collected, autologous or homologous transfusion requirements, side effects encountered during donation and transfusion of blood were documented. Results: Of the 62 units requested 57 (91.9%) were collected. We were able to meet 86.7% of transfusion requirements with autologous blood. The incidence of homologous transfusion was 13.3%. Thirty-eight (66.7%) units were transfused. The rate of wastage of units was 33.3%. There were no major complications as a result of the patients either giving their own blood or receiving any blood products. Conclusions: This study confirms that autologous blood transfusion is an effective way of reducing the need for homologous blood. The procedure of preoperative autologous blood donation is cost effective, safe, and should be widespread. Keywords: Autologous blood, Elective orthopaedic surgery, Homologous blood, TransfusionNigerian Journal of Orthopaedics and Trauma Vol. 6 (2) 2007: pp.52-54

Efectividad de la donación de sangre autóloga preoperatoria en prótesis total de rodilla

To assess the effectiveness in usual clinical practice of a preoperative autologous blood donation (PAD) program in total knee replacement (TKR). An observational study was carried out in which 127 patients undergoing TKR were followed from the pre-anaesthetic visit to hospital discharge. Number of patients transfused and number of autologous and homologous blood units transfused in the PAD group and in the control group were registered and compared. 73 patients (57.5% of the sample) entered in the PAD program, of which 18 (24.7%) received their own blood. Four patients in the PAD group (5.5%) also received one unit of homologous blood in comparison to 18.5% in the control group (p = 0.025). In the PAD group, 5.6% of patients who received antifibrinolytic treatment (n = 36) and 43.2% of those who did not (n = 37) were transfused (p < 0.001). Due to the low transfusion rate in TKR, especially when antifibrinolytic agents were administered, the use of PAD must be reconsidered in this indication.

Preoperative autologous blood donation

SUMMARYBlood‐conservation techniques are declining because of the increasingly safe allogeneic blood supply. However, they continue to be used because of the public’s perception of allogeneic blood risks and increasing blood shortage. Preoperative autologous blood donation (PABD) may decrease the risk of postoperative infection, either because immunomodulation is avoided, or because bacterial or viral infections have to be carefully screened before inclusion of a patient in a PABD program. However, the higher risk of mortality from allogeneic blood transfusion is now more likely due to administrative errors and to blood‐transmitted infection, and the risks associated with banked autologous blood units are similar to those associated with banked allogeneic blood units. Therefore, when PABD is performed inducing anemia, there is an increase of the number of transfused units, thus an increase of the risk related to any transfusion. PABD is most beneficial in procedures that cause significant blood loss, in association with recombinant human erythropoietin administration and blood salvage. Identification of surgical candidates who will need transfusion and will therefore truly benefit from blood conservation must be based on factors specific to the patient, such as the baseline hematocrit and the anticipated blood loss during surgery. Professor Francesco Mercuriali promoted and validated this concept. The appropriate use of autologous techniques should be targeted toward the best possible cost‐effectiveness ratio for an individual patient. Despite its reduced role and declining rationale, PABD must be periodically revisited to assess its role in a changing environment.

Erythropoietin as adjuvant to pre-operative autologous blood donation in total hip arthroplasty: New algorithm for use

Erythropoietin (rhEPO) has been used in different surgical procedures as a method for saving allogeneic blood, with variable efficacy. Forty consecutive patients entered the pre-operative autologous blood donation (PABD) program, and during donations hemoglobin fell below 115 g/l; they received rhEPO 40,000 U every week for three or four weeks (group 1). As control group, 35 consecutive patients who entered the PABD program were studied; during donations, hemoglobin levels in these patients fell below 115 g/l, but rhEPO was not administered (group 2). Pre-surgery hemoglobin levels were higher in patients who received rhEPO (134 g/l vs. 121 g/l; p < 0.0002), and an average of 3.47 doses were administered. The number of transfused autologous units was 1.6 in group 1 and 2.1 in group 2 ( p < 0.05), while the number of allogeneic units was 0.9 and 0.1, respectively ( p < 0.0005), so that only 5% of patients treated with rhEPO required some allogeneic unit, as compared to 40% of those who did not receive rhEPO ( p < 0.0005). There were no relevant adverse effects, but in two patients from group 1, rhEPO treatment had to be discontinued because the level of hemoglobin exceeded 150 g/l. The mean duration of hospital admission was shorter in the patients who received rhEPO than in those who only underwent PABD (8 days vs. 11.8 days; NS). When adequately used, rhEPO is an effective and safe alternative to the use of allogeneic blood.

Frozen preoperative autologous blood donation for heart transplantation at the Mayo Clinic from 1988 to 1999.

Preoperative autologous blood donation (PABD) has been used to reduce the need for allogeneic RBC transfusion, decreasing risk and conserving supply. A frozen PABD program for heart transplant patients was instituted at the Mayo Clinic in 1988, but participation has steadily declined. The aims of this study were to determine how the availability of PABD influenced the transfusion RBC components, whether the availability of PABD reduced exposure to allogeneic RBC components, and the costs of providing PABD units. A retrospective review of all heart transplant cases from 1988 to 1999 was performed (n = 141). Data on collection and transfusion practices were compared for patients with (n = 88, 62%) and without PABD (n = 53, 38%). Total RBC transfusion requirements did not differ between the groups. Patients with frozen PABD received fewer allogeneic units, but they also had less blood salvaged and reinfused. Twenty patients (23%) completely avoided exposure to allogeneic RBCs in the PABD group versus three patients (6%) in the group without PABD. Although patients in the PABD group successfully donated a total of 423 units, 41 percent were discarded. Over 11 years, the need for 251 units of allogeneic RBCs was avoided ($27,610), but $283,500 was spent to have the frozen PABD units available. PABD can be performed for heart transplantation, but it is expensive.

An autologous blood donation program for paediatric scoliosis patients in Hong Kong.

A retrospective audit was conducted to determine the safety, efficacy and patient satisfaction related to a preoperative autologous blood donation program for children and teenagers undergoing corrective surgery for scoliosis. Forty-five of the 77 patients donated the requested amount of blood. These 45 compliant patients had been requested to donate fewer units of blood than noncompliant patients (mean 4.0 vs 4.6 respectively, P = 0.02). Twelve patients required allogeneic blood transfusion. Two patients had surgery delayed making the collected autologous blood unavailable. The extent of the operation was associated with the need for allogeneic blood transfusion. Six and a half percent of all donated units of blood were discarded. No major complications were reported. Overall, 93% of patients were satisfied with the program. With careful patient selection, good inter-departmental coordination and teamwork, preoperative autologous blood donation in paediatric patients undergoing extensive corrective surgery for scoliosis is safe and effective.

Declining value of preoperative autologous donation.

Preoperative autologous blood donation (PABD) has been shown to decrease allogeneic blood transfusion requirements in major elective surgery. Changes in transfusion practice motivated an examination of blood use from 1993 to 2000 of patients participating in the Héma-Québec PABD program. Blood donation and transfusion, type of surgery, and demographic characteristics were prospectively entered into a computer database for patients participating in the Héma-Québec PABD program. Autologous donations represented from 0.8 to 2 percent of total blood collections and have declined by 26 percent after peaking in 1995. The mean number of units collected per patient declined, as did the number of units transfused per patient and the utilization rate. For radical prostatectomy, knee replacement surgery, hip replacement surgery, and scoliosis, utilization rates were 72, 60, 83, and 78 percent in 1993 compared with 50, 50, 58, and 58 percent in 2000, respectively. In 2000, 18 percent of patients were receiving a 1-unit autologous transfusion. Depending on the surgical procedure, 85 to 95 percent of patients avoided allogeneic transfusion; this did not change significantly from 1993 to 2000. Patients participating in the PABD program successfully avoided allogeneic transfusion in over 85 percent of cases. However, declining utilization rates and frequent 1-unit transfusions demonstrate the decreasing utility of PABD over time.

Utility of red blood cell apheresis in autologous blood donation

Although autologous blood donation is an alternative to allogeneic transfusion, some authors had questioned its cost-effectiveness. New techniques, like red blood cell apheresis could improve the cost-effectiveness of autologous blood transfusion, therefore we have valued the efficiency of this procedure in autologous blood donation. Materials and Methods. We studied 131 patients undergoing different types of surgery who entered the preoperative autologous blood donation program over a one year period. Apheresis was performed with the MCS 3p from Haemeonetics. Results. We were able to collect 304 red blood cell units from 131 patients. The average yield per procedure was two units (88 cases, 67.2%). In 41 patients (31.3%), we collected 3 units and, in two cases, 4 units were collected. The mean volume of the units was 255 (191–280). 18 (13.7% patients had an adverse reaction. Most of these were mild. Only in one case was it necessary to stop the procedure. 202 units (66.4%) were transfused to 97 patients (74%). 12 (9.2%) patients also used allogeneic transfusions (mean units: 0.18±0.05 with a range 1–5). Conclusion. Red blood cell apheresis is a useful procedure in autologous blood donation.

Erythropoietin in Cancer Surgery

SUMMARY Patients with gastrointestinal cancer usually have a sideropenic anemia unresponsive to iron supplementation which is characterized by low levels of endogenous erythropoietin. Anemic patients are not suitable for preoperative autologous blood donation programs and are at high risk for perioperative homologous blood transfusions. Preoperative treatment with recombinant human erythropoietin (r-HuEPO) has allowed anemic patients to donate autologous blood and reduced homologous blood transfusion after surgery. A dose-response study showed that 400 U/kg of r-HuEPO, divided in four doses administered every four days, is the minimum effective dose which increases hemoglobin levels about 10 g/L/week. The red blood cell (RBC) gain was 268 ± 79 mL, which is grossly equivalent to the RBC content of two units of predonated autologous blood. The total charge for this treatment ($280) was substantially lower than in other experiences in which higher doses of r-HuEPO were given. To maximize erythropoietic response, it is necessary to provide supplementary IV iron (125 mg daily). This results in a normalization of circulating and stored iron, suggesting that the amount was adequate both to sustain the production of new RBCs and to replace tissue stores during r-HuEPO treatment.

Preoperative Autologous Blood Donation in Patients with Rheumatoid Arthritis: Analysis on the Effectiveness in Increasing Red Blood Cell Mass

Background: Analysis on the effectiveness of preoperative autologous blood donation (PABD) on increase in red blood cell mass (+RBC) in patients with rheumatoid arthritis (RA). Patients and Methods: 199 RA patients (52 men and 147 women) participating in a PABD program prior to primary or revision arthroplasty of the hip or knee joint were included into the study. RA of the patients was known for many years and diagnosed according to the criteria of the American Rheumatism Association. Data were collected either retrospectively from the patients’ records or from actual records of patients being scheduled for PABD and surgery. Patients’ ability to predeposit was verified by an anesthetist in charge of the autologous blood bank. Patients considered suitable for this elective surgical intervention were suitable for PABD as well according to the standard criteria for PABD (compensated organ function, no signs of systemic infections, relevant laboratory parameters within their physiological limits except for a ‘normal’ hemoglobin/hematocrit (Hb/ hct) level), and the ASA physiological status score. Data on +RBC were analyzed for differences between men and women as well as between manual and mechanical PABD. Statistical significance was considered if p < 0.05 (with Bonferroni correction where appropriate). Results: When starting with PABD, the hct of the patients amounted to hct_init = 33.6 ± 3.9%; immediately before operation, hct was hct_preop = 30.7 ± 4.0% (p < 0.05). +RBC for all patients was 51 ± 82 ml, with no difference between men (45 ± 97 ml) and women (53 ± 76 ml). The interval between donation and surgery (IDOP) was 18.3 ± 12 days for all patients, with no difference between men (17.6 ± 11 days) and women (18.6 ± 12 days). 48.7% of the patients generated a +RBC of ≤50 ml, while 13.0% only produced a +RBC of >150 ml, including those 4.5% of patients generating +RBC at least equal to 1 U of packed red blood cells (190 ml). Patients with mechanical PABD generated a higher +RBC than those with manual PABD (70 ± 68 ml versus 38 ± 88 ml; p < 0.05); IDOP did not differ between both groups (mechanical PABD 19 ± 10 days, manual PABD 17 ± 14 days). Among a variety of parameters, only patient’s RBC massinit, IDOP, and RBC mass_donated were shown to have an impact on +RBC; neither age nor gender nor ASA physiological status score were of relevance for +RBC. Conclusions: Our analysis demonstrates that in RA patients the 1-unit PABD concept is poorly effective in increasing total RBC mass. To rise effectiveness, IDOP should be prolonged to at least 4 weeks prior to surgery, and blood volume_donated/RBC mass_donated should be increased in order to stimulate erythropoiesis more intensively. Additionally, our data suggests to perform only 1 PABD in RA patients, but with a blood volume_donated/RBC mass_donated being greater than in the routine PABD procedure.

Preoperative autologous blood donation by 1073 elderly patients undergoing elective surgery: a safe and effective practice.

Preoperative autologous blood donation (PABD) aims at avoiding the risks associated with exposure to allogeneic blood. While its use is extremely common among adult patients in connection with elective surgery, it is still uncommon in elderly patients, because of a series of coexisting pathologies. A retrospective study was made of 1073 consecutive elderly patients at a city hospital from 1990 to 1996. Their responses to the PABD program were evaluated by analysis of the incidence of complications and the demand for allogeneic blood. The PABD program was interrupted in 79 (7.4%) of 1073 patients because of the onset of anemia, vasovagal reactions, lack of accessible superficial veins, or cardiovascular complications. Seven hundred eighty-four (73.1%) of 1073 patients were given autologous blood; 151 (14.1%) patients also required allogeneic blood. The onset of anemia (6.5%) was the main contraindication for continuing the PABD program: incidence increased with age. PABD in connection with elective surgery is both feasible and effective in a high percentage of elderly patients.

Personalized Approach to Define Transfusion Support to Surgical Patients

Preoperative autologous blood donation (PABD) is accepted as a standard of care to reduce the risk of allogeneic blood transfusion. However, autologous blood is considered more costly than allogeneic blood. PABD can be made more cost-effective by reducing the cost of collection and by avoiding overcollection of units. When MSBOS and SOPCAB are used to identify and exclude from PABD procedures associated with low transfusion requirements, and to define the number of units to be collected, the overall AB wastage is still around 15% (ranging from 6% to 15% for different surgical procedures). To optimize the PABD program we have developed a more personalized approach to define each patient's transfusion requirement based on the predicted blood loss in mL of RBCs, calculated per surgical operation in the previous 6-12 months, along with the blood loss that the patient can tolerate. The latter depends on the baseline circulating RBC mass and the RBC mass compatible with the patient's clinical and cardiocirculatory condition. To fulfill transfusion requirements for each patient the most effective strategy can be selected according to the type and time to surgery, PABD applicability; age and clinical status of the patients taking into account what each of the different techniques can provide in terms of volume of RBCs produced or conserved.

Preoperative autologous blood donation.

Preoperative autologous blood donation (PABD) is the donation of the patient's own blood before surgery for use in planned surgical procedures. Appropriate patients for this form of blood conservation are those having elective surgery for which blood is usually transfused, whose hemoglobin is no less than 11 g/dL, and who display no signs or symptoms of bacteremia. Patients should begin donation 4–6 weeks before surgery, donate every 3–7 days, take oral iron, and donate the last unit more than 72 hours before surgery. Vesovagal reactions occur in 2–5% of donors. In this article, the authors review indications and associated risks of PABD. Also discussed are segregated processing, testing, disposition of unused units, platelet-rich plasma, recombinant erythropoietin, indications for autologous versus allogeneic transfusion, and cost-effectiveness of preoperative autologous blood donation. Nurses play important roles in PABD. Provision of information to patients regarding when, where, and how autologous donations are made along with prevention of errors in blood availability and transfusion are discussed. Nursing and blood bank staff members must establish regular communication to ensure the provision of a quality PABD program.

Physician recommendation as the key factor in patients' decisions to participate in preoperative autologous blood donation programs

To determine whether the physician or patient had initiated the discussion regarding preoperative autologous blood donation (PABD) and to assess the relative importance of the physician's recommendation in patients' decision to donate, responses were obtained from 254 of 409 patients (62%) who had donated preoperatively during the 3 study months. Nearly all (96%) strongly agreed they would donate again for themselves and nearly all (94%) strongly agreed they would recommend PABD to others. Patients initiated the discussion about PABD 23% of the time, while 71% indicated strong surgeon input. The remaining respondents said their surgeon had "mentioned it, but said it was up to me." The importance of avoiding transfusion reactions was rated significantly greater among those whose surgeons had initiated the discussion, as was the relative weight of the surgeon's recommendation. The desire to alleviate the blood shortage was rated much less important among patients who had initiated the discussion themselves.

Preoperative autologous donation: Surgery clinic staff knowledge/attitudes

Preoperative autologous blood donation (PABD) is both under- and overused. Although the decision to order PABD lies with the surgeon, it is quite likely that other surgery clinic personnel influence patient acceptance and enrollment into PABD programs. Accordingly, we measured knowledge, attitudes, and the referral practice of clinic personnel pertaining to PABD. We administered a questionnaire to 102 nurses and 33 clerks working in surgery clinics at three university medical centers--one center in an area with a high incidence of AIDS and two centers in areas of low incidence of AIDS. Knowledge of PABD was poor when assessed by six questions. Only 6% each of nurses and clerks answered all questions correctly; 55% of nurses and 54% clerks missed three or more of the six questions. Surprisingly, no differences (P > .05) in knowledge deficits were noted when personnel from high and low AIDS areas were compared--indicating an overall need for education about PABD. In general, attitudes about PABD were positive, as most respondents (63%) gave favorable answers. Clinic personnel from the high AIDS area had even more favorable attitudes (P = .02). Because of these favorable attitudes, it seems likely that educational programs dealing with PABD would be readily accepted by clinic personnel. Greater knowledge should enhance the effectiveness of clinic staff in identifying, counseling, and referring eligible patients for this service.

One-year use of the Bloodloc system in an orthopedic institute

Human error in patient or specimen identification due to fatigue, stress and lack of attention by technologists, nurses, interns, and physicians, can cause routinely safety procedures to be circumvented. Clerical errors may occur during the specimen collection, the issue of blood unit and the transfusion of blood. The introduction in an increasing number of hospital of preoperative autologous blood donation programs further increases the chance of error, because a single patient can predeposit multiple units of blood. In this cases there is a greater commitment not only to transfuse any blood unit that is ABO compatible but to transfuse the specific units the patient previously donated for his own use. Human error has been recognized as a significant cause of transfusion-associated fatalities. The persistence of the frequency and type of errors observed in spite of extensive efforts to eradicate them, suggests that errors are inevitable as long as large number of repetitive procedures are performed unless major system changes are adopted. A system (Bloodloc System) that physically prevents the possibility of error was adopted since January 1993 and cuncurrently a quality improvement program (QI) was implemented specifically designed to monitor: 1. the absence of the code on the blood samples, 2. the blood bank error in setting the Bloodloc, 3. the misidentification of blood samples, 4. any attempt to transfuse the wrong blood unit, 5. any attempt to transfuse, the wrong patients. 4895 blood units (2469 autologous and 2426 allogeneic units) were transfused to 1478 patients (849 predeposited an average of 3.3 +/- 2.0 units). The methodological errors (absence of three-letter code on the patient's specimen tube, wrong transcription of the code on the blood sample, wrong setting of the Bloodloc in the blood bank)--41 cases--were limited at the first four months of implementation of the system. In the same period however have been reported 3 potentially fatal errors which have been avoided by the Bloodloc. Two cases of misidentification of blood samples at the moment of the specimen collection, and one attempt to transfuse the wrong units to the wrong patients. The Bloodloc system is effective in preventing potential transfusion-associated fatalities caused by units or recipients misidentification.

Preoperative autologous blood donation: Linkage of the public and private hospital sectors

To evaluate the safety and efficacy of a preoperative autologous blood donation (PABD) program with liberal patient exclusion criteria. Prospective patient accrual from September 1990 to May 1992. A PABD program in a public tertiary care hospital linked with a specialist orthopaedic practice in a private hospital. One hundred and twenty patients consecutively scheduled for orthopaedic surgery were referred for PABD. There were 77 women (median age, 63 years) and 43 men (median age, 62 years). One-third of patients were 70 or more years old. Coexistent medical disease, mainly cardiovascular, was present in 46% of patients and 3% were excluded because of coexisting morbidity. Total hip or total knee replacement was performed in a private hospital with intraoperative or postoperative blood salvage in 37% of the patients. Adverse effects of PABD, autologous blood collected and used and homologous blood transfused. One hundred and sixteen patients donated 267 units of autologous blood. In 70 patients undergoing total hip replacement, 78% donated three and 20% donated two units, with 95% of autologous blood being used. In 38 patients undergoing total knee replacement, 42% donated two units and 55% donated one unit, with 87% of autologous blood being transfused. Seventy-eight per cent of all patients only received autologous blood. In the remaining patients, homologous blood use was confined to two units or less in 80%. Hypotensive episodes associated with phlebotomy occurred in 3% of patients, and 13% of patients had preoperative haemoglobin levels ranging from 85-100 g/L without adverse clinical effects. A hospital-based PABD program with less strict patient exclusion criteria does not prejudice the clinical status of the donors. Its linkage to private hospitals can significantly reduce the use of homologous blood in selected elective surgery.