The religious beliefs of Jehovah's Witnesses prohibiting the acceptance of blood or blood products by transfusion are well known. In many instances, the treatment options for these patients are limited by these beliefs. This is particularly true for orthopaedic operative procedures associated with a great deal of blood loss.
Erythropoietin is a naturally occurring glycoprotein hormone that regulates the production of erythrocytes in the human body. It is produced primarily by the kidneys in response to hypoxia and, therefore, its level in the systemic circulation is inversely proportional to the degree of tissue oxygenation. Recent advances in the technology and cloning of recombinant DNA have provided the basis for the development and production of recombinant human erythropoietin, which is now available worldwide. Erythropoietin has been used successfully to treat anemia in patients who have end-stage renal disease6. Its use in the treatment of both medically and operatively induced anemia has been described2,12, and its use as an adjunct to facilitate the preoperative collection of autologous blood from patients who are scheduled to have a total hip or knee arthroplasty has been well established9,13. Supplemental erythropoietin therapy for the treatment of anemia has been widely described in the medical literature and is considered to have a favorable risk-benefit ratio1; however, erythropoietin has been shown to stimulate the proliferation of megakaryocyte cell lines in vitro10. A slight but significant (p < 0.0005) increase in the platelet count was demonstrated after the use of erythropoietin to treat anemia in patients who required hemodialysis7. Consequently, there is concern whether the use of exogenous erythropoietin increases the already well established risk of deep venous thrombosis associated with a total joint arthroplasty3. To our knowledge, there have been no reports of a significant increase in the prevalence of deep venous thrombosis for patients who have been managed with exogenous erythropoietin.
A previous report by the senior one of us (C. L. N.) demonstrated that primary and revision total hip replacements can be accomplished safely in Jehovah's Witnesses without the transfusion of allogenic blood11. However, revision total hip replacements have proved to be difficult in these patients because of the considerable intraoperative blood loss that frequently occurs. Occasionally, despite hypotensive anesthesia, careful operative technique, meticulous hemostasis, and techniques to reinfuse blood intraoperatively with a cell-saver device, two-stage revisions have been necessary in these patients because of the potential for considerable intraoperative blood loss. Clearly, a single-stage operation is preferable and could be performed more readily and safely in patients who had a relatively high hematocrit (0.45 to 0.50) preoperatively, as this increases the margin of safety with regard to blood loss. The hypothesis that was addressed in this study was that optimization of the hematocrit by the administration of exogenous erythropoietin preoperatively is a beneficial and safe adjunct in the management of Jehovah's Witnesses who have revision total hip arthroplasty.
*Although none of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received, but are directed solely to a research fund, foundation, educational institution, or other non-profit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was Ortho Biotic, Raritan, New Jersey.
†Department of Orthopaedics, University of Arkansas for Medical Sciences, 4301 West Markham, Slot 531, Little Rock, Arkansas 72205-7199.
The study included five Jehovah's Witnesses (four women and one man) who were scheduled to have a revision total hip arthroplasty. For all five patients, the indication for the revision consisted of pain or radiographic signs of loosening of at least one component, or both. The average age at the time of the revision was 66.4 years (range, fifty-eight to seventy-eight years).
After a patient had been identified as a potential participant in the study, several laboratory values were determined: a complete blood-cell count with differential, a reticulocyte count, total iron-binding capacity, and levels of serum folate, ferritin, and iron. A hematocrit of 0.45 or more, a transferrin saturation of more than 20 per cent, or a ferritin level of more than sixty micrograms per liter were criteria for exclusion from the study. A history of myocardial infarction (within the previous three months), decompensated congestive heart failure, unstable angina pectoris, seizure disorder, transient ischemic attacks, stroke, or uncontrolled hypertension (a systolic blood pressure of greater than 180 millimeters of mercury [24.00 kilopascals] or a diastolic blood pressure of greater than ninety-five millimeters of mercury [12.67 kilopascals]); current pregnancy or lactation; or clinically important ongoing blood loss (such as gastrointestinal bleeding) were also criteria for exclusion. However, no patient was excluded from the study.
Preoperatively, recombinant human erythropoietin, 100 international units per kilogram of body weight, was administered subcutaneously three times per week. The protocol developed before the initiation of the study dictated that, for each patient, the duration of preoperative therapy be determined by the hematocrit at presentation: the therapy was supposed to be given for four weeks if the initial hematocrit was 0.30 to 0.36, for three weeks if it was 0.361 to 0.399, and for two weeks if it was 0.40 to 0.449. The hematocrit, hemoglobin level, and blood pressure were monitored weekly, beginning at the end of the second week of treatment. If the hematocrit exceeded 0.42, the dose of erythropoietin was supposed to be reduced to fifty international units per kilogram of body weight. The erythropoietin therapy was to be stopped if the hematocrit exceeded 0.45. These were thought to be the ideal parameters for the implementation of the study. All patients received the preoperative erythropoietin therapy from their individual local physicians, usually in other states, under the direction and supervision of our orthopaedic department. All administering physicians were provided with a copy of the desired dosing regimen. The therapy was withheld or extended if the hematocrit increased rapidly, if it did not increase, or if the operation was delayed after the initiation of the erythropoietin therapy. The average duration of erythropoietin therapy was twenty-six days preoperatively and 3.6 days postoperatively. All of the patients took iron sulfate (325 milligrams) and vitamin C (250 milligrams) orally three times per day and folic acid (one milligram) orally once per day for the duration of the erythropoietin therapy (average, approximately thirty days).
Operative Planning and Technique
Preoperative anteroposterior radiographs of the pelvis and the femur were used to discuss and develop detailed plans and options for all patients. Hypotensive anesthesia was induced in all patients, with methods that have been previously described in reports from our institution11. A Cell Saver (Haemonetics, Braintree, Massachusetts) was used to reinfuse blood intraoperatively in three of the five patients; the remaining two patients refused this adjunct on the basis of religious beliefs.
At the time of the operation, each patient was placed in the lateral position and an anterolateral approach was employed, with use of the scar from the previous operation whenever feasible. The anterior half of the insertion of the gluteus medius into the greater trochanter and the surrounding scar tissue were released to facilitate exposure. Careful operative technique was employed at all times, including extensive use of electrocautery, meticulous hemostasis whenever possible, and diligent tamponade of all cut or reamed bone surfaces with use of sponges soaked in a solution of epinephrine (one part in one million). The operative time was minimized as much as possible without compromising the quality of the operation. One or two large drains were placed in the wound before closure. Only gravity drainage was used, to minimize the postoperative blood loss.
Preoperatively, all but one of the patients had received an antibiotic (cefazolin, one gram given intravenously) prophylactically. In the fifth patient, infection was suspected and the administration of antibiotics was delayed until culture specimens had been obtained intraoperatively by swabbing. The intravenous administration of antibiotics was continued until all of the drains had been removed. For the two patients who were managed during the early period of the study, prophylaxis against deep venous thrombosis was achieved with warfarin, given orally each day, beginning the night before the operation. For the three patients who participated in the study after low-molecular-weight heparin had become clinically available, enoxaparin (thirty milligrams) was given subcutaneously, twice a day, beginning on the day of the operation. All prophylactic treatment was discontinued after the absence of deep venous thrombosis had been confirmed by venous Doppler examination of the lower extremities on the fifth, sixth, or seventh postoperative day.
The hematocrit and hemoglobin level were monitored closely throughout the study period, and the values before the erythropoietin therapy was started, at the time of admission to the hospital (one or two days preoperatively), immediately postoperatively, and at the time of discharge were recorded for this investigation. The lowest hematocrit in each patient during the perioperative period was also noted. All blood samples were collected in pediatric specimen tubes to minimize the loss of red blood cells.
All five revision total hip arthroplasties were performed successfully without the transfusion of blood, although three patients permitted the reinfusion of autologous blood intraoperatively with use of the Cell Saver. No patient had any complications associated with an excessive loss of blood or a low hematocrit.
The average hematocrit was 0.395 (range, 0.317 to 0.447) before the erythropoietin therapy was started, 0.476 (range, 0.431 to 0.509) at the time of admission to the hospital, 0.368 (range, 0.272 to 0.424) immediately postoperatively, and 0.308 (range, 0.294 to 0.327) at the time of discharge. For the three patients who received autologous blood through the use of the Cell Saver, 300, 300, and 750 milliliters of blood was reinfused. The average total blood loss, consisting of the intraoperative blood loss and the postoperative drainage, was 919 milliliters (range, 692 to 1705 milliliters). In three of the five patients, the hematocrit had reached its lowest point and had begun to increase by the time of discharge (average, 9.2 days postoperatively) (Table I).
No patient had evidence of deep venous thrombosis during the period of hospitalization or the follow-up period. Due to geographical constraints, the follow-up of one patient was by telephone conversation and by correspondence with the patient's local physician only. Follow-up of two patients consisted of visits to our clinic, telephone conversations, and correspondence; follow-up of the remaining two patients consisted of visits to our clinic only. The average duration of follow-up was 19.8 months (range, six to thirty-four months). All patients had a stable hematocrit and were making satisfactory progress at the time of the most recent follow-up examination.
Increased rates of morbidity and mortality have been associated with operations in patients who have anemia4. Similarly, intraoperative blood loss in excess of 500 milliliters has been shown to increase the rate of mortality14. Not surprisingly, blood transfusions have become commonplace in joint-replacement operations and particularly so with the increasing number of revision arthroplasties, in which considerable intraoperative blood loss is accepted as essentially unavoidable5.
Since the appearance of the human immunodeficiency virus in the general blood supply in the early 1980's, much expense and effort has been directed toward reducing the number of transfusions of allogenic blood in the United States. The indications for transfusion have become more stringent, techniques to reduce operative blood loss have been developed, and extensive networks have been established to accommodate the donation of autologous blood. Although much progress has been made in reducing the frequency of transfusions of allogenic blood, the expense and limited effectiveness of current methods leave room for additional investigation8.
The refusal of Jehovah's Witnesses to accept blood transfusions presents a particular challenge to the orthopaedic surgeon when these patients have failure of a total hip arthroplasty. The dilemma is compounded if the patient has anemia. Hypotensive anesthesia and the use of a cell-saver device intraoperatively and of plasma expanders may be helpful, but none of these increases the hematocrit preoperatively. Additionally, not all Jehovah's Witnesses allow the transfusion of autologous blood with a cell-saver device; two of the five patients in the current series refused such intervention. Exogenous erythropoietin therapy enables the direct optimization of the hematocrit before the operation.
Erythropoietin therapy in patients who have renal failure has resulted in a limited number of adverse reactions, including hypertension, seizures, and, rarely, allergic reactions1. Both seizures and exacerbation of hypertension appear to be associated with rapid increases in the hematocrit. Consequently, the manufacturer's guidelines for the use of erythropoietin recommend that the hematocrit increase by no more than 3 to 4 per cent over any two-week period of treatment and that the blood pressure be closely monitored throughout the duration of therapy. None of these complications was encountered in our patients. From our limited experience, it appears that recombinant human erythropoietin, when administered under the current guidelines, may be a safe and effective method of increasing the hematocrit preoperatively, thereby facilitating the performance of revision hip arthroplasties in patients who are Jehovah's Witnesses.
The results of this study are preliminary, and the sample was small. However, the potential applications for recombinant human erythropoietin are promising. The elective nature and the moderately flexible timing of joint replacements and revisions make preoperative treatment with erythropoietin feasible. In our study, the average delay of the operation of less than four weeks to increase the hematocrit by an average of 8.1 per cent proved to be prudent and fruitful. From our data, it appears that the hematocrit can be increased to a safer level over a relatively short period of time.
The cost of erythropoietin therapy will certainly influence its level of acceptance in today's health-care environment. No direct costs to the patient were incurred in this study because the manufacturer donated the erythropoietin. However, on the basis of an approximate cost of ninety dollars per vial of 10,000 international units of erythropoietin, and our experience that 50,000 to 150,000 international units of erythropoietin is usually needed to achieve the desired hematocrit, most patients would incur a cost of 450 to 1350 dollars if they were managed according to the existing guidelines. There are, of course, additional costs associated with the administration of the drug, the determinations of the hematocrit preoperatively, and the monitoring of the blood pressure; these costs are difficult to quantify but must be considered in any discussion of expense. Although the total costs are substantial, established methods for the collection, storage, and reinfusion of autologous blood can be just as expensive when multiple units are required.
The unique needs of Jehovah's Witnesses who have a total joint replacement have accelerated the application of erythropoietin therapy in orthopaedics. Clearly, additional refinement of protocols, larger series of patients, and extended follow-up are all needed to draw reliable conclusions. In our study, the optimization of the preoperative hematocrit through therapy with recombinant human erythropoietin appeared to be both feasible and safe. In the future, exogenous erythropoietin therapy may provide a practical and powerful tool in the efforts to provide revision arthroplasty to all appropriate patients, regardless of religious preference, without the use of blood transfusion and its associated risks.