This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (7) Citing Articles via Google Scholar Google Scholar Articles by Couvret, C. Articles by Fusciardi, J. Search for Related Content PubMed PubMed Citation Articles by Couvret, C. Articles by Fusciardi, J. Related Collections Blood Surgery

---

GENERAL ARTICLES

A Restrictive Use of Both Autologous Donation and Recombinant Human Erythropoietin Is an Efficient Policy for Primary Total Hip or Knee Arthroplasty

Claude Couvret, MD^*, Marc Laffon, MD^*, Annick Baud, MD^*, Valérie Payen, MD^*, Philippe Burdin, MD, and Jacques Fusciardi^*

Departments of *Anesthesiology and Critical Care and Orthopedic Surgery, Trousseau University Hospital, Tours, France

Address correspondence and reprint requests to Dr. Claude Couvret, MD, Department of Anesthesiology and Critical Care, Trousseau University Hospital, 37044 Tours Cedex 1, France. Address email to fusciardi{at}med.univ-tours.fr

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
A limitation of preoperative autologous blood donation (PABD) in nonanemics and the use of recombinant human erythropoietin (EPO) in anemics (baseline hematocrit [Hct] 39%) could be an efficient approach of the cost-benefit ratio of transfusion during primary total hip (THA) or knee (TKA) arthroplasties. We evaluated the consequences on transfusion rates and costs of two different applications of a transfusion policy based on personal requirements during primary THA or TKA. This quality assurance observational study compared two prospective successive time periods; each included successive patients treated by the same medical team and standardized care. In Study 1 (n = 182), PABD was indicated if there were insufficient estimated red blood cell reserve and a life expectancy 10 yr, no use of EPO, and identical criteria for any transfusion. Because this policy led to a 50% allogeneic transfusion rate when baseline Hct 37% and autologous blood wastage in the nonanemics (baseline Hct > 39%), 2 refinements were introduced in Study 2 (n = 708): EPO without PABD when baseline Hct 37%, and life expectancy 10 yr, and avoidance of PABD in nonanemics. This novel care induced a marked decrease in transfusion rates (respectively, from 41% to 7%, P < 0.0002, in nonanemics; from 58% to 27%, P < 0.003, in anemics; and from 43% to 12%, P < 0.0001, overall), with no change in allogeneic transfusion (10%) and discharge Hct, and a 39% financial savings. This saving effect is a result of the suppression of PABD in nonanemics, who represent 75% of this surgical population. Although erythropoietin is expensive, it can be used with cost savings in selected patients because the overall cost of transfusion is reduced.

IMPLICATIONS: During primary total hip or knee arthroplasty, the limitation of erythropoietin to patients with hematocrit 37%, the restrictive use of autologous donation to patients with 37% < hematocrit 39% and no autologous donation in the nonanemics may allow savings both in blood requirements and financial cost.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Large variations in perioperative blood transfusion have been observed among hospitals (1–3) during primary total hip (THA) or knee (TKA) arthroplasties. Efforts to avoid complications associated with allogeneic transfusion have promoted the use of preoperative autologous blood donation (PABD). Most patients will participate in a PABD program (4,5).

Nevertheless, PABD, based on standard predictive blood loss, may lead to preoperative anemia (6) and increased exposure to transfusion (7). It may also promote autologous red blood cell (RBC) waste when transfusion management is tailored to the individual (8–10). The prevention of PABD-induced preoperative anemia might be limited by reduction of PABD and one preferential use of recombinant human erythropoietin (EPO) for anemic patients, as Mercuriali et al. (11) showed a 50% reduction in allogeneic transfusion with EPO alone with a baseline hematocrit (Hct) between 33% and 39%. Reduction of autologous waste might be favored by avoidance of PABD for nonanemic patients (i.e., baseline Hct >39%) because waste is more frequent and allogeneic requirement unlikely in this subpopulation (4,10). Thus, one can assume that for most nonanemic patients undergoing primary THA or TKA, simply decreasing the transfusion Hct trigger might be enough to avoid the need for transfusion and, as a result, PABD would not be needed. Finally, the cost-benefit ratio of transfusion during primary THA or TKA could be a limitation of PABD and a preferential use of EPO for anemic patients (4). However, the financial cost-benefit ratio of such a strategy is presently unknown when applied to a population undergoing primary arthroplasty.

For this reason, we conducted a prospective observational study that examines all patients exposed to similar perioperative conditions and compared a standardized policy based on individual requirements to novel care, including the introduction of EPO and further limitations of PABD. Our aim was primarily to assess the effect of such a blood-saving strategy on transfusion requirements. We expected a reduction in overall transfusion with no worsening in discharge anemia and no increase in allogeneic requirements. Second, we aimed to determine the proper role of anemic and nonanemic patients in the impact of this new strategy.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
We performed a prospective observational study in consecutive patients undergoing primary THA or TKA in the Department of Orthopedic Surgery of our institution. Every patient included gave written, informed consent, and the study was conducted after IRB approval. Two successive studies were considered. They were conducted with the same surgical and anesthetic teams, identical principles for clinical care, and a transfusion management tailored to the individual, but two different strategies were used to apply the blood-saving principle.

Study 1
This study included every ASA physical status I–III patient scheduled for a cemented primary unilateral THA or TKA during a 10-mo period (from April, 1999, to February, 2000). Refusal to participate was the single criterion for noninclusion. These patients were already included in a previously published work (10).

The preoperative anesthesia evaluation was performed 2 mo before surgery. At that time, the baseline Hct was measured and a PABD decision was made. Indication for PABD was based on comparison of each patient’s RBC reserve with mean estimated perioperative RBC loss. The patient’s estimated RBC reserve was calculated. The median RBC loss was previously estimated to be 538 mL (range, 100–1212 mL) for THA and 693 mL (range, 272–1535 mL) for TKA. PABD was indicated if RBC reserve was <800 mL (THA) or 1000 mL (TKA). PABD was indicated by the anesthesiologist in case of insufficient RBC reserve, baseline Hct >33%, an estimated life expectancy of 10 yr, no medical contraindication, and consent of the patient. Two units of blood were collected preoperatively with each collection performed a week apart and with 2 wk between the last collection and the day of surgery. All patients had 320 mg of oral ferrous sulfate given twice a day beginning 3 wk before starting PABD. Antiplatelet drugs and oral anticoagulant treatments were planned to be stopped 8 days before surgery, and a switch to low-molecular-weight heparin was started on the day before surgery and continued throughout hospitalization.

Neither acute normovolemic hemodilution nor intraoperative reinfusion of autologous salvaged blood was used. Transfusion of postoperative autologous salvaged blood from wound drainage was performed during the first 6 postoperative hours in TKA.

A standardized general anesthesia was used in every case. All anesthetic inductions were performed with propofol, sufentanil, atracurium, tracheal intubation, and controlled ventilation. For maintenance, isoflurane was administered in 50% oxygen/50% N₂O, and sufentanil reinjections were given as needed. A 3-in-one nerve block with 30 mL of bupivacaine 0.25% and clonidine 1 µg/kg was used for postoperative analgesia. Forced air warming set at 43°C via a blanket applied on the upper part of the body was used throughout the procedure. Neither systemic controlled hypotension nor antifibrinolytic drugs were used.

Transfusion, either autologous or allogeneic, was indicated in case of Hct <24%; for patients with risk factors for myocardial ischemia, it was indicated in case of an Hct value between 24% and 30% (12) and one of the following symptoms: dyspnea, excessive weakness impeding deambulation or rehabilitation, evidence of myocardial ischemia or overt congestive heart failure, or postoperative neuropsychological impairment.

The collected data were as follows: patients’ age, sex, weight, ASA physical status, type of surgery, duration of the procedure, and length of hospitalization. Potential clerical errors in the transfusion were noted. Hct values were obtained at preoperative anesthesia evaluation and before inclusion for PABD (baseline Hct), the day before surgery (admission Hct), and at Day 5 or 8 after surgery (discharge Hct). Preoperative anemia was defined by baseline Hct 39%, and no anemia by baseline Hct >39%. The number of collected autologous units was noted, as was the number of autologous and allogeneic units transfused during the perioperative periods. The volume of the postoperative salvaged blood transfused was also noted, and RBC loss was calculated as previously indicated (see Calculation).

We calculated the costs of collected autologous RBC units and transfused allogeneic RBC units based on the cost of a single unit: $188.13 for one autologous RBC unit and $155.27 for one allogeneic RBC unit.

Study 2
After evaluation of blood requirements, effectiveness and costs resulting in a personalized transfusion policy in Study 1 (control period), a second prospective study was initiated to evaluate the effects of some changes in this policy (Study 2: intervention period).

Criteria for inclusion and exclusion were the same as those described in Study 1. The same anesthesia and surgical teams were involved in the two studies. The medical, surgical and anesthetic procedures previously described in Study 1 were also applied to Study 2.

The analysis of Study 1 results led us to assume that two changes might improve our transfusion policy. First, using EPO instead of PABD when baseline Hct 37% with a life expectancy of almost 10 yrs and, second, the abolition of PABD in case of baseline Hct >39%. EPO (Epoetine alfa; Ortho Biotech SA, Raritan, NJ) was only offered to a subpopulation of anemic patients with baseline Hct 37%, a life expectancy of almost 10 yr, and no contraindications. Potential contraindications were uncontrolled hypertension, toxic anemia, and refusal to participate. Treatment included 3 subcutaneous injections of 600 UI/kg. The first one was initiated 3 wk before surgery, and the next 2 on a weekly basis. The last injection was done 1 wk before the day of hospitalization. Oral ferrous sulfate 320 mg daily was given in association with EPO. PABD was planned at the preoperative anesthesia evaluation and its modalities were those described in Study 1. However, the indications were limited to patients with 37% < baseline Hct 39%. No PABD was performed in Study 2 patients with baseline Hct >39%. Triggers for any transfusion (autologous or allogeneic) were identical to those in Study 1.

The collected data, the calculations of RBC reserve and RBC loss, and costs of RBC units were identical to those in Study 1. The percentage of EPO use and the inclusion of EPO cost (EPO 40000 UI = $431), were also considered in Study 2, in addition to the cost of autologous and allogeneic units.

Calculations
The estimated RBC reserve and RBC loss were calculated using Mercuriali and Inghilleri’s formulas (13). The estimated RBC reserve (mL) is the volume of tolerated RBC loss corresponding to the minimal Hct value at which no allogeneic transfusion is required (target Hct). Estimated RBC reserve (mL) = EBV (mL) x (baseline Hct – target Hct), where EBV = patient’s estimated blood volume (body weight in kilograms x 70 mL/kg); the target Hct value at discharge was chosen to be 30%.

Estimated RBC loss (mL of RBC) = EBV (mL) x (admission Hct – discharge Hct) + transfused RBC (mL), where admission Hct was Hct during hospitalization before operation, transfused RBC = 150 mL x number of autologous and/or allogeneic RBC units, and postoperative autologous salvaged blood volume (V) transfused in the 6 postoperative hours after TKA. The mean Hct of this unwashed blood was 30%. So, the postoperative salvaged autologous RBC was calculated as V x 0.3. Thus, this calculated RBC loss (mL) included total perioperative blood loss.

Demographic and biological data were expressed as mean ± SD except for variables not normally distributed (duration, RBC loss, and collected, transfused, or wasted RBC units); for these, median and range values were used.

The following tests were performed with StatView (SAS Institute, Cary, NC). Comparisons of quantitative variables, such as Hct of patients with and without autologous or allogeneic transfusion, or Hct of patients between Study 1 and 2, used the Student’s t-test. Comparisons of the qualitative variables with two or more classes, such as autologous and allogeneic transfusions, sex, surgery, and anemia, were assessed with the ² test or Fischer’s exact test in case of insufficient calculated values. Using the Mann-Whitney U-test and Wilcoxon’s test compared quantitative variables not normally distributed. Changes in Hct value within a group or between groups were analyzed with one-way or two-way analysis of variance and then paired two-sided tests. A P value < 0.05 was considered statistically significant.

Calculation of sample size in Study 2 was aimed at gaining a sufficient power to show a decrease in allogeneic requirement in patients treated by EPO. To determine the incidence of allogeneic transfusion with EPO, we planned to include 182 successive patients in Study 2, and then to conduct an interim assessment of the data. The sample size was then determined, based on these preliminary data. Interim assessment found 6 allogeneic transfused patients of 23 with baseline Hct 37% (26%), versus 8 of 16 (50%) in Study 1. The sample size of Study 2 was then determined using this preliminary result and the following principles: reject of the null hypothesis test; estimated difference of clinical interest: 24%; type I error: 0.05; and a power of 90%. Eighty-one patients with baseline Hct 37% were needed, indicating a sample size of 640 patients in Study 2. Taking into account a potential risk of 10% of patients that would not be able to be evaluated, the final estimated sample size considered was 712 for Study 2.

	Results

Top Abstract Introduction Methods Results Discussion References

 
In Study 1, 182 consecutive patients (101 THA and 81 TKA) were eligible, included and evaluated. In Study 2, among the 712 eligible patients, 4 were not included because they wanted PABD despite a baseline Hct >39%. Thus, 708 patients were included and evaluated (441 THA and 267 TKA).

The Study 2 period lasted from June 2000 to September 2002. The main characteristics of the patients are summarized in Table 1. There were no between-studies differences in general characteristics either at study entry or for operative time, RBC loss, or the type of arthroplasty. When compared with THA, TKA characteristics were older patients, more often female gender, longer-lasting operation and more bleeding, whatever the study.

View larger version (25K): [in this window] [in a new window]   Figure 1. Changes of policy between Studies 1 and 2 resulted in a reduction of overall transfusions from 43% to 12%. It was mainly attributable to a 15-fold reduction of exclusive autologous transfusion. No significant variation in the rate of allogeneic transfusion was observed (P = 0.22). *P < 0.0001, Study 1 vs Study 2.

Transfusion Outcomes According to Autologous Donors and Nondonors
Transfusion outcomes for donors and nondonors are shown in Table 2. There was a 10-fold reduction in donors in Study 2, in accordance with the very restrictive policy for PABD in Study 2. This policy did not increase the exposure of donors to allogeneic transfusion. In the two studies, donors were more exposed to transfusion than were nondonors. The collection was complete (2 RBC units) for 90% of donors in Study 1 and 81% in Study 2.

As expected by the change of PABD policy, the between-studies decrease in donors was associated with a marked reduction of collected autologous units (172 in Study 1 versus 56 in Study 2). The median of wasted autologous units per donor was not different between the two studies: 1 (0–2), P = 0.14. Forty-five percent of autologous units were wasted in Study 1 and 61% in Study 2 (P = 0.14); although the absolute number of wasted units was larger in Study 1 (n = 79) than in Study 2 (n = 34). Eighty percent of Study 1 patients with at least one wasted autologous RBC unit had a baseline Hct >39%, whereas allogeneic transfusion was infrequent (8%) in this subpopulation (Fig. 2).

View larger version (21K): [in this window] [in a new window]   Figure 2. Overall transfusion rates decreased from Study 1 to Study 2: from 43% to 12% (P < 0.0001) for the whole populations, from 41% to 7% in patients with baseline hematocrit (Hct) >39% (nonanemics) (P < 0.0002), and from 58% to 27% in patients with baseline Hct 39% (anemics) (P < 0.003). Figure 2 shows that the decrease in transfusion was attributable to no autologous transfusion in nonanemics and to a decrease in autologous transfusion in anemics, in accordance with the Study 2 policy. In addition, the allogeneic transfusion rate was decreased in Study 2 when baseline Hct 37%, because recombinant human erythropoietin was introduced in 48 of 89 patients of this subpopulation. *P < 0.05, P < 0.005, **P < 0.0001.

Admission and Discharge Hematocrit
Table 5 shows that the mean admission Hct was higher in Study 2 than in Study 1. The admission Hct was also higher in nondonors than in donors (P < 0.0001) in the two studies. The mean discharge Hct did not differ between studies, both for the whole population, the nonanemics, the anemics, and for nondonors; whereas it was slightly higher in Study 1 donors. The discharge Hct did not differ according to allogeneic or autologous transfusion, both in Study 1 (30% ± 2% and 31% ± 2%, respectively, P = 0.09) and in Study 2 (29% ± 3% and 29% ± 3%, respectively, P = 0.90).

Calculation of Costs
Based on the costs of autologous and allogeneic RBC unit, and EPO 40,000 UI, the calculated individual cost was $213 with the first transfusion policy (Study 1) and $130 per patient with the second one (Study 2). The saving was thus $83 on average per patient with the novel care. The costs of PABD, allogeneic transfusion, and EPO use in both studies are detailed in Table 4. The saving was only the result of the least expensive novel care applied to nonanemic patients; whereas the novel care was more expensive in anemic patients. In these later patients the mean individual overcharge was $226 because the cost of EPO was more than savings resulting from the PABD limitation.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Although PABD has been widely used, concerns have arisen regarding increase in preoperative anemia, wasted units, financial constraints, and risks. Recent evolution in perioperative transfusion have reduced the use of PABD, with collection based on predicted individual requirements (14,15,10), the use of an homogeneous trigger for any transfusion (16,10), and the inclusion of EPO in the care of anemic patients (11). Our first study applied the principles, except the use of EPO. Although its effectiveness in reducing the need of allogeneic transfusion was already clearly established (11,17), its financial cost-clinical benefit impact for daily use in an overall surgical population was not clearly defined. The rate of overall allogeneic transfusion we observed (13%) was in the 12%–27% range of frequency reported by others when using predefined transfusion criteria (1,3,18). After completion of Study 1, we assumed that our transfusion policy could be improved: first, by using EPO without PABD when baseline Hct 37% because as much as 50% of this subpopulation received allogeneic blood (Fig. 2), and second, by avoidance of PABD when baseline Hct >39% because the allogeneic transfusion rate was infrequent (Fig. 2) and autologous waste was frequent in this subpopulation. Study 2 was thus designed, on the basis of a quality assurance principle, to evaluate the saving effects of these two refinements. This new strategy induced a 10-fold reduction in PABD (Table 2) and a 15-fold reduction in autologous transfusion (Fig. 2), with no changes in both allogeneic requirement and mean discharge Hct value. The transfusion rate was reduced in nonanemics (from 43% to 7%) and in anemic patients (from 53% to 27%), as shown in Table 3. The discharge Hct values were similar whatever the transfusion (allogeneic or autologous) and the baseline Hct value (Table 5). Thus, Study 2 indicates that PABD is no longer warranted in our institution when baseline Hct >39%. In Study 2, PABD was limited to patients with 37% < baseline 39% and no contraindication. Because there was no allogeneic transfusion when PABD was performed in Study 1 anemic patients with 37% < baseline 39%, we decided to continue performing PABD in Study 2, and the rate of allogeneic transfusion was 3%. Patients with baseline Hct 37% are a subpopulation at increased risk of allogeneic transfusion. Study 1 shows that PABD is not efficient in these patients because 50% received allogeneic blood. In addition, the ability of anemic patients to give blood is limited or even impossible. Thus, using EPO to increase the RBC reserve and the admission Hct are often the only recourse. Study 2 shows that a restrictive use of EPO, without PABD, may be sufficient to avoid allogeneic transfusion in primary THA or TKA. This policy results in an overall allogeneic transfusion rate of 10%, a rather small frequency when compared with the 12%–27% range otherwise reported (1,3,18). It also induced a 91% reduction in the relative risk of allogeneic transfusion in this subpopulation with baseline Hct 37% (Fig. 2). Finally, a main adjunct of our study was to support previous suggestions (4) regarding the limited indications for PABD during primary THA and TKA.

Although PABD is less expensive than EPO, its cost-effectiveness has been questioned (5,6). We found that, even with a transfusion management tailored to the individual, a strategy based on limitation of PABD and EPO indications (Study 2) is less expensive than a strategy based on larger indications of PABD without EPO (Study 1). The financial difference is 39%, and the mean individual saving is $83 (Table 4). The suppression of PABD in nonanemics, who represent 75% of this surgical population, allows both this financial saving and the EPO financing for the anemic population. However, we chose to use PABD, rather than EPO, for the novel care of patients with 37% < baseline Hct 39%. What would be the cost of using EPO rather than PABD in these patients? EPO could have been indicated in 54% of this subpopulation (46 new patients) after elimination of contraindications. Giving EPO to these 46 new patients would suppress PABD cost (minus $10,535) but add an EPO cost of $48,943. With the hypothesis of an allogeneic transfusion rate similar to that observed with EPO in Study 2, this EPO policy would induce an additional cost of $69 per patient. Thus, the mean individual cost of such a Study 2 policy would be $199 instead of $130. Finally, although different, these two strategies are less expensive than the one used in Study 1 (Table 4).

A potential weakness of this observational study is bias that may result from nonrandom assignment, unblinding of the physicians, and potential for time-linked improvement in the management of patients. Nevertheless, we chose to observe two successive groups of patients for several reasons: first, this study took place in the context of a continuing quality assurance assessment. Second, because of the difficulties in conducting a randomized clinical trial as a result of bias introduced by clinical judgment of the investigators and unwillingness of some patients to be randomized without PABD. Third, because observational studies have been regarded as valuable alternatives in these cases (19,20). We tried to minimize the potential confounding factors by studying a large population of consecutive patients operated on by the same members of the surgical team using homogeneous and time-honored surgical principles and standardized perioperative care. An important point is that both the overall and autologous transfusion rates observed in the control period (Study 1) (43% and 30% respectively) are rather less than rates otherwise reported (14,21,1,22). These rates decreased by 72% and 93% respectively after application of novel care without worsening postoperative anemia. We also found that lengths of hospitalization were identical in both studies and in Study 2 patients whatever their baseline Hct (> 39% or 39%). Thus, we suggest that in-hospital postoperative outcome was not different between studies and between nonanemic and anemic patients subjected to novel care. However, as our study did not examine out-of-hospital outcome, we cannot determine which method is better. Namely, considering Study 2 subgroups according to baseline Hct (Table 5), we observe that patients with 37% < Hct 39% had a mean discharge Hct 2 points less than others (29% versus 31%). Does it make a difference? Or is there evidence that maintaining a discharge Hct 30% has a positive effect on morbidity or mortality? Neither the current study nor the literature in general can give a definitive answer to this question. According to Green et al. (23) a postoperative hemoglobin value in the range 9.0–11.0 g/dL does not affect postoperative recovery after an orthopedic operation. Lotke et al. (5) showed that nonsurgical complications are more frequent after TKA when the hemoglobin value is <9.0/dL and that there is a tendency for a more rapid progression in physical therapy when postoperative hemoglobin values are in the range 10.7–11.4 g/dL. Conversely, an Hct value < 30% might be beneficial to some patients, including those with coronary disease (22,24). Therefore, we believe it is unlikely that a discharge Hct of 29% ± 3% in a Study 2 subgroup of patients may have favored morbidity.

The current study may impact the clinical management of primary THA and TKA. On the basis of a predicted mean perioperative RBC loss 600–800 mL (roughly 2000 mL blood loss at Hct 30%) and a transfusion threshold based on individual tolerance, we suggest applying the following principles. When baseline Hct >39%, no sparing method is required except reinfusion of drains in TKA; the predicted allogeneic transfusion rate is 7%. PABD must be only considered in the rare situations that associate an insufficient RBC reserve (more often females with small weight) and a sufficient life expectancy to allow the development of potential and infrequent risk of allogeneic blood-induced acquired viral disease. When baseline Hct 37%, EPO alone with 600 UI/kg given on a weekly basis from Day D – 3 wks and oral ferrous sulfate may be appropriate. When 37% < baseline Hct 39%, either PABD or EPO, depending on willingness or ability to predonate blood, is appropriate. These basic principles must be challenged according to the local mean blood loss and the calculated individual RBC reserve.

In summary, we evaluated the consequences of 2 refinements in a personal requirement policy during primary THA and TKA: no blood-sparing method when baseline Hct >39%, and limited use of PABD or EPO when baseline Hct 39%. Such changes induce a 72% relative reduction in transfusion rate, a 10% allogeneic transfusion rate, a 39% absolute reduction in transfusion costs, and no uncontrolled anemia on discharge. These improvements are mainly attributable to nonanemic patients, who rarely need PABD. The limited inclusion of either EPO or PABD in a global management program is possible and favorable to the financial cost-clinical benefit ratio, inasmuch as no blood-sparing method is used in patients with baseline Hct >39%.

	Acknowledgments

 
The authors acknowledge Dr. Philippe Bertrand, MD, for his helpful statistical comments and advice. We are indebted to Mrs. Andrée Verrier for typing the manuscript.

	Footnotes

 
Presented, in part, at the annual meeting of the European Society of Anesthesiologists, Nice, France, April 2002.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Churchill WH, McGurk S, Chapman RH, et al. The collaborative Hospital Transfusion Study: variations in use of autologous blood account for hospital differences in red cell use during primary hip and knee surgery. Transfusion 1998; 38: 530–9.[Web of Science][Medline]
2. Baele P, Beguin C, Waterloos H, et al. The Belgium BIOMED Study about transfusion for surgery. Acta Anaesth Belg 1998; 49: 243–303.[Medline]
3. Bierbaum BE, Callaghan JJ, Galante JO, et al. An analysis of blood management in patients having a total hip or knee arthroplasty. J Bone Joint Surg Am 1999; 81: 2–10.[Abstract/Free Full Text]
4. Earnshaw PH. Transfusion alternatives in orthopedic surgery. TATM 1999; 1: 13–22.
5. Lotke PA, Barth P, Garino JP, Cook EF. Predonated autologous blood transfusions after total knee arthroplasty. J Arthroplasty 1999; 14: 647–50.[Web of Science][Medline]
6. 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.[Medline]
7. Cohen J, Brecher ME. Preoperative autologous blood donation: benefit or detriment? A mathematical analysis. Transfusion 1995; 35: 640–4.[Web of Science][Medline]
8. Etchason J, Petz L, Keeler E, et al. The cost effectiveness of preoperative autologous blood donations. N Engl J Med 1995; 332: 719–24.[Abstract/Free Full Text]
9. Birkmeyer JD, Goodnough LT, Aubuchon JP, et al. The cost-effectiveness of preoperative autologous blood donation for total hip and knee replacement. Transfusion 1993; 33: 544–51.[Web of Science][Medline]
10. Couvret C, Tricoche S, Baud A, et al. The reduction of preoperative autologous blood donation for primary total hip or knee arthroplasty: the effect on subsequent transfusion rates. Anesth Analg 2002; 94: 815–23.[Abstract/Free Full Text]
11. Mercuriali F, Zanella A, Barosi G, et al. Use of erythropoietin to increase the volume of autologous blood donated by orthopedic patients. Transfusion 1993; 33: 55–60.[Web of Science][Medline]
12. Goodnough LT, Brecher ME, Kanter MH, Aubuchon JP. Transfusion medicine: first of two parts–blood transfusion. N Engl J Med 1999; 340: 438–47.[Free Full Text]
13. Mercuriali F, Inghilleri G. Proposal of an algorithm to help the choice of the best transfusion strategy. Curr Med Res Opin 1996; 13: 465–78.[Web of Science][Medline]
14. Larocque BJ, Gilbert K, Brien WF. A point score system for predicting the likelihood of blood transfusion after hip or knee arthroplasty. Transfusion 1997; 37: 463–7.[Web of Science][Medline]
15. Nuttal GA, Santrach PJ, Oliver WC, et al. The predictors of red cell transfusions in total hip arthroplasties. Transfusion 1996; 36: 144–49.[Web of Science][Medline]
16. Spahn D, Casutt M. Eliminating blood transfusions: new aspects and perspectives. Anesthesiology 2000; 93: 242–55.[Web of Science][Medline]
17. Price T, Goodnough LT, Vogler W, et al. Improving the efficacy or preoperative autologous blood donation in patients with low hematocrit: a randomized, double blind, controlled trial of recombinant human erythropoietin. Am J Med 1996; 101: 22S–7.
18. Goodnough LT, Despotis GJ, Merkel K, et al. A randomized trial comparing acute normovolemic hemodilution and preoperative autologous blood donation in total hip arthroplasty. Transfusion 2000; 40: 1054–7.[Web of Science][Medline]
19. Pocock SJ, Elbourne DR. Randomized trials or observational tribulations? N Engl J Med 2000; 342: 1907–9.[Free Full Text]
20. Radford MJ, Foody JM. How do observational studies expand the evidence base for therapy? JAMA 2001; 286: 1228–30.[Free Full Text]
21. Sculco TP, Gallina J. Blood management experience: relationship between autologous blood donation and transfusion in orthopedic surgery. Orthopedics 1999; 22: s129–34.[Web of Science][Medline]
22. Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med 1999; 340: 409–17.[Abstract/Free Full Text]
23. Green D, Lawler M, Rosen M, et al. Recombinant human erythropoietin: effect on the functional performance of anemic orthopedic patients. Arch Phys Med Rehabil 1996; 77: 242–6.[Medline]
24. Spiess BD, Ley C, Body SC, et al. Hematocrit value on intensive care unit entry influences the frequency of Q-wave myocardial infarction after coronary artery bypass grafting. J Thorac Cardio Vasc Surg 1998; 116: 460–7.[Abstract/Free Full Text]

This article has been cited by other articles:

				V. Martinez, A. Monsaingeon-Lion, K. Cherif, T. Judet, M. Chauvin, and D. Fletcher Transfusion strategy for primary knee and hip arthroplasty: impact of an algorithm to lower transfusion rates and hospital costs Br. J. Anaesth., December 1, 2007; 99(6): 794 - 800. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (7) Citing Articles via Google Scholar Google Scholar Articles by Couvret, C. Articles by Fusciardi, J. Search for Related Content PubMed PubMed Citation Articles by Couvret, C. Articles by Fusciardi, J. Related Collections Blood Surgery