Concerns about the safety of the blood supply have focused attention on various strategies to reduce allogeneic transfusion. Transfusions expose patients to risk of various infectious diseases, including human immunodeficiency virus (HTV) and hepatitis, as well as to noninfectious risks such as allergic reaction, graft-versus-host disease, and alloimmunization.1"4 While the risks of transfusion-related diseases have decreased with improved screening and banking procedures, infectious and noninfectious transfusion risks will likely always persist. Despite numerous strategies to avoid allogeneic transfusion, children undergoing surgery for correetion of spinal deformity often require blood transfusion.3,5-10 In the late 1980s, confronted with reports of iatrogenic transmission of HIV and hepatitis to children via transfusion, we began to seek novel ways of avoiding transfusion in our pediatric patients undergoing major elective spine surgery.
Preoperative autologous donation (PAD), directed transfusion, hemodilution, hypotensive anesthesia, and intraoperative and postoperative Cell Saver have all been used in attempts to reduce transfusion rates in this setting.11,12 While some physicians advocate the routine use of PAD in children, we have found PAD to be onerous to arrange, poorly tolerated by the children, and often not effective given me smaller blood volumes and baseline anemia found in many of our patients. For this reason, we have sought other mechanisms by which to avoid transfusion for this population.
RECOMBINANT HUMAN ERYTHROPOIETIN
Encouraged by the efficacy of recombinant human erythropoietin (Epoetin alfa) at decreasing transfusion rates in various adult orthopedic procedures,13-18 we began using Epoetin alfa in children 8 years ago. There was a paucity of experience with the use of Epoetin alfa in children undergoing major surgery. Pediatric usage had been limited to the treatment of neonatal anemia, anemia secondary to leukemia and other cancers, or anemia in children on dialysis.12,13,19-21 Few studies had examined the potential benefits of preoperative Epoetin alfa in children, despite the fact that the consequences of transfusion-related diseases are even more serious in children given the longer life expectancy of this population.
Since 1990, we have routinely, but not universally, used preoperative Epoetin alfa for pediatric (< 18 years of age) patients undergoing scoliosis surgery at this institution.8,9,22 A retrospective review of our experience, examining a cohort of 178 patients who have undergone surgery to correct scoliosis, has recently been published.23
At the outset, we sought to use Epoetin alfa as widely as possible but often were unable to administer the drug because of patient inability to cover medication costs prior to approval, preexisting primary hematologic disease, or parental refusal. Thus, our experience to date is neither randomized nor well controlled; however, we feel the results summarized here strongly support the use of Epoetin alfa in pediatric patients undergoing major elective spine surgery.
Our protocol calls for the subcutaneous administration of 10,000 International Units (IU) Epoetin alfa by the child's pediatrician once per week for the 3 weeks before surgery. All patients were administered 325 mg ferrous sulfate three times each day for the 4 weeks before surgery regardless of treatment with Epoetin alfa. Hematocrit (HCT) level was monitored during the treatment phase. Therapy was stopped if the patient's preoperative HCT exceeded 50%, which, in our 8-year experience, has been the basis for discontinuation of treatment only once, No adverse effects of treatment were observed. This experience corroborates other well-controlled studies that did not observe complications resulting from Epoetin alfainduced polycythemia.18 Furthermore, theoretical risks of this therapy, such as hypertension, myocardial infarct, and thrombosis, are much less germane to a healthy pediatric population.
Fig: Effect of recombinant human erythropoietin (Epoetin alfa) on allogeneic transfusion rates in pediatric patients with various types of scoliosis. *P = .006.
Of 178 patients studied, 78 (44%) received Epoetin alfa treatment. The overall transfusion rate for the entire cohort was 26%, which is consistent with reported transfusion rates in this setting. However, for patients with idiopathic scoliosis, treatment with Epoetin alfa correlated with significantly fewer transfusions and significantly higher HCT levels throughout hospitalization. Only 4% of Epoetin alfa-treated patients received allogeneic transfusion compared with 24% of patients not treated with Epoetin alfa (P = .006) (Fig).
EFFECTIVENESS IN CHILDREN WITH NEUROGENIC SCOLIOSIS
Contrary to the effects observed in the idiopathic population, patients with congenital or neurogenic scoliosis did not appear to benefit from treatment with Epoetin alfa. In fact, there was a trend (not statistically significant) for a higher rate of allogeneic transfusion among the neurogenic patients who received Epoetin alfa (55%) than among those not so treated (45%) (Fig 1). This paradoxical effect may be the result of a selection bias inherent in our retrospective design. Patients with the worst degrees of curvature and/or the more rigid curves who were felt to be at the highest risk for excessive blood loss may have been more likely to receive Epoetin alfa in the preoperative setting. Alternatively, these patients may have been more likely to have anemia of chronic disease and thus elevated endogenous erythropoietin levels. It is possible that we have been underdosing this group of patients. For this reason, a prospective randomized study is currently in progress to compare the effect of 300 and 600 IU/kg Epoetin alfa doses in this population. Given that this group has the highest transfusion rate in the cohort and is the least likely to benefit from PAD, Epoetin alfa could play an important role in this population if proven efficacious.
EFFECT ON LENGTH OF HOSPITALIZATION
On average, the length of hospitalization was 9.3 days for Epoetin alfa-treated patients, compared with 10.0 days for patients not receiving Epoetin alfa (not statistically significant: P = .6). However, the effect of Epoetin alfa treatment on the length of hospitalization was dependent on the type of scoliosis. In the idiopathic group, Epoetin alfa treatment was associated with a significantly shorter length of stay compared with no Epoetin alfa treatment (6.7 vs. 9.4 days; P = .02). In contrast, patients in the neurogenic and congenital groups treated with Epoetin alfa exhibited a trend toward a somewhat increased length of hospitalization that was not statistically significant. There was no statistically significant difference in length of stay in the intensive care unit between groups treated and not treated with Epoetin alfa. Given that these data are based on a retrospective, uncontrolled study, however, they must be interpreted with discretion. Future, controlled studies should rigorously examine the effect of Epoetin alfa treatment on length of hospitalization, as well as on transfusion rates.
COMPARISON OF COSTS OF EPOETIN ALFA TREATMENT WITH ALTERNATIVES
In an attempt to examine the longrange implications of Epoetin alfa treatment, including averted morbidity and costs of treatment, we are currently developing a decision analytic model of Epoetin alfa use in the previously described patient population. Using Decision Maker™ 7.0 software,24 we have designed a model to assess the short- and long-term costs and benefits of treatment with Epoetin alfa against a control group and against a group of patients undergoing PAD. Ranges of plausible event probabilities and associated costs were obtained from our retrospective experience and from the literature. Though the work is ongoing, our analysis suggests that compared with the control group, preoperative administration of Epoetin alfa theoretically saves 1.1 years of life per 1000 treatments. This outcome was garnered from averting 28 cases of chronic hepatitis, 5.5 cases of cirrhosis, 2. 1 cases of hepatocellular carcinoma, and 1.7 cases of HIV infection per 1,000,000 treatments. As expected, these projections were sensitive to the values for Epoetin alfa effectiveness, control transfusion rate, and the average number of transfusions per patient. When PAD was included in the model, the Epoetin alfa treatment was approximately equal to PAD in cost-effectiveness. Our analysis suggests that the use of Epoetin alfa compares favorably with PAD, in terms of both efficacy and cost-effectiveness. Additionally, as discussed previously, our retrospective experience suggests that use of Epoetin alfa in patients with idiopathic scoliosis may result in shorter hospitalization times. A reduction in hospitalization of only 0.8 hospital days would make Epoetin alfa treatment absolutely cost-effective.
IDENTIFYING PATIENTS FOR EPOETIN ALFA USE
Clearly, rates of transfusion vary from one type of patient to another and depend on clinical characteristics such as age, type of scoliosis, degree of curve, number of levels involved, associated comorbidities, and operative approach. As such, cohorts of patients exist who are at particularly high risk of transfusion. As this risk increases, the probability that Epoetin alfa can reduce or eliminate transfusion also increases. This implies not only decreased risk, and decreased associated costs, of transfusion-related diseases, but also decreased expense of allogeneic transfusion. In fact, our model indicates that Epoetin alfa use may indeed be costeffective for patients with a high risk of transfusion. We are currently reviewing a much larger cohort of pediatric spine surgery patients to develop a clinical predictive model of the risks of transfusion and to identify groups of patients for whom Epoetin alfa treatment would be particularly appropriate.
Our experience to date indicates that preoperative use of Epoetin alfa in children undergoing major spine surgery is a safe and convenient method of decreasing transfusion rates in children with idiopathic scoliosis. It is presently unclear if Epoetin alfa treatment is equally effective for patients with neurogenic scoliosis, though work is ongoing to examine the effectiveness of higher dosing regimens. For Epoetin alfa to be cost-effective, the cost of treatment either must offset the cost of transfusion or impact the length of hospitalization. For a subset of patients at very high risk of transfusion, it is possible that the cost of Epoetin alfa will be offset by the cost of multiple transfusions, and we are currently attempting to define this population of "high risk" patients via a clinical predictive model. Additionally, one might hypothesize that decreased rates of transfusion would be associated with shorter length of hospitalization; however, a controlled, prospective study is necessary to assess this hypothesis. Our studies suggest that Epoetin alfa is no less cost-effective than other commonly used means of averting transfusion, including PAD. Furthermore, we feel that Epoetin alfa treatment is more easily applied to children than PAD, is better accepted and tolerated, and is likely more effective for a broader range of children. In summary, we feel there is a definite role for the preoperative use of Epoetin alfa in children undergoing spine surgery, and that treatment with Epoetin alfa compares favorably with other means of avoiding allogeneic transfusion in this population.
1. Dodd RY. The risk of transfusion-transmitted infection [editorial]. N Engl J Med. 1992; 327:419421.
2. Holland PV. Transfusion transmitted infectious diseases. J Fla Med Assoc. 1993; 80:33-36.
3. Klein HG. Allogeneic transfusion risks in the surgical patient. Am J Surg. 1995: 170(suppl 6A):21S-26S.
4. Wylie BR. Transfusion transmitted infection: viral and exotic diseases. Anaesth Intensive Care. 1993:21:24-30.
5. Cowell HR. Swickard JW. Autotransfusion in children's orthopaedics. J Bone Joint Surg Am. 1974; 56:908-912.
6. Faris 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.
7. MacEwen GD, Bennett E, Guille JT. Autologous blood transfusions in children and young adults with low body weight undergoing spinal surgery. J Pediatr Orthop. 1990; 10:750-753.
8. Rothstein P, Roye DP Jr, Verdisco L. Experience with preoperative administration of erythropoietin in adolescents undergoing scoliosis surgery. In press.
9. Roye DP Jr. Rothstein P, Rickert JB, Verdisco L, Farcy JP. The use of preoperative erythropoietin in scoliosis surgery. Spine. 1992; 17:S204-S205.
10. Simpson MB, Georgopoulos G, Orsini E, Eilert RE. Autologous transfusions for orthopaedic procedures at a children's hospital. J Bone Joint Surg Am. 1992;74:652-658.
11. Mann DC, Wilham MR. Brower EM. Nash CL Jr. Decreasing homologous blood transfusion in spinal surgery by use of the Cell Saver and predeposited blood. Spine. 1989; 14:1296-1300.
12. Pierelli L. Menichella G, Scambia G. et al. In vitro and in vivo effects of recombinant human erythropoietin plus recombinant human G-CSF on human haernopoietic progenitor cells. Bone Marrow Transplant. 1994; 14:23-30.
13. Evans RW, Rader B. Manninen DL. The quality of life of hemodialysis recipients treated with recombinant human erythropoietin. Cooperative Multicenter EPO Clinical Trial Group. JAMA. 1990; 263:825-830.
14. Goodnough LT, Price TH, Friedman KD, et al. A phase HJ trial of recombinant human erythropoietin therapy in nonanemic orthopedic patients subjected to aggressive removal of blood for autologous use: dose, response, toxicity, and efficacy. Transfusion. 1994;34:66-71.
15. Goodnough LT, Vizmeg K, Verbrugge D. The impact of autologous blood ordering and blood procurement practices on allogeneic blood exposure in elective orthopedic surgery patients [published comment appears in Am J Clin Pathol. 1994; 101:249]. Am J Clin Pathol 1994; 101 :354-357.
16. Kyo S. Omoto R, Hirashima J, Eguchi S. Fujita T. Effect of human recombinant erythropoietin on reduction of homologous blood transfusion in open-heart surgery. Circulation. 1992; 86(suppl II):II-413-II-418.
17. Levine EA, Rosen AL, Gould SA, et al. Recombinant human erythropoietin and autologous blood donation. Surgery. 1988; 104:365-369.
18. Mercuriali F, Zanella A, Barasi G, et al. Use of erythropoietin to increase the volume of autologous blood donated by orthopedic patients. Transfusion 1993;33:55-60.
19. Alexander SR. Pediatric uses of recombinant human erythropoietin: the outlook in 1991. Am J Kidney Dis. 1991; 18(suppl 1, pt 4):42-53.
20. McNamee P, van Doorslaer E, Segaar R. Benefits and costs of recombinant human erythropoietin for end-stage renal failure: a review. Benefits and costs of erythropoietin. Int J Technol Assess Health Care. 1993; 9:490-504.
21. Strauss RG. Erythropoietin in the pathogenesis and treatment of. neonatal anemia. Transfusion. 1995; 35:68-73.
22. Rothstein P, Roye D, Verdisco L, Stem L. Preoperative use of erythropoietin in an adolescent Jehovah's Witness. Anesthesiology. 1990; 73:568570.
23. Vitale MG, Stazzone EJ, Gelijns AC, Moskowitz AJ, Roye DP. The effectiveness of preoperative erythropoietin in avoiding allogeneic blood transfusion among children undergoing scoliosis surgery. J Pediatr Orthop (Part B). 1998: 7:203-209.
24. Sonnenberg FA, Pauker SG. [Decision MakerJ. ver. 7.0. New England Medical Center. Boston. MA: Division of Clinical Decision Making.