Erythropoietin Alfa in Critically Ill Patients – Use of Patient-Oriented Evidence that Matters (POEMs) to Identify, Evaluate, and Apply Biomedical Literature to Patient Care
A POEM (patient oriented evidence that matters) is the process of evaluating primary literature in search of patient-specific outcomes that will directly impact the patient’s quality of life.1,2 Shaughnessy and Slawson, a pharmacist and a physician, developed the concept of using POEMs to evaluate literature for use in family medicine residency training programs. POEMs are intended to bridge the gap between learning about disease-oriented outcomes and the application of these disease-oriented outcomes to patient care. POEMs presented in the Journal consist of four key components: introduction of a patient-specific problem, literature search and identification of primary literature, literature evaluation, and clinical application.
An 82-year-old African American female presented to the emergency department with coffee ground emesis and abdominal pain. The patient had a past medical history significant for hypertension and past surgical history of valve replacement and laparotomy for partial intestinal resection two years earlier. The patient was found to have small bowel obstruction on CT scan and was admitted to the surgical ICU unit for exploratory laparotomy and small bowel resection for internal hernia and bowel ischemia. Postoperatively, she developed septic shock and received aggressive fluid resuscitation, vasopressors, corticosteroids, transfusions, and broad-spectrum antibiotics. Sepsis-induced organ dysfunction included cardiovascular, renal, and metabolic organ systems and the APACHE II score was 28. Drotrecogin alfa was administered for 96 hours without complications.
On hospital day four, the patient had a second surgery for re-exploration, washout, ileal resection and bowel anastamosis, followed by red blood cell and platelet transfusions due to critical illness and blood loss (Figure 1). Anemia was a persistent problem during the ICU stay and was treated with periodic transfusions, rHuEPO, and iron therapy. On hospital day 13, the patient was started on Epoetin Alfa 40,000 units subcutaneously weekly which continued for the remainder of her ICU stay. At that time, the patient’s serum iron level was 6 mcg/dL, TIBC 102 mcg/dL, ferritin 594 ng/mL, transferrin 68 mg/dL. Sodium ferric gluconate (Ferrlecit) 125 mg IVPB every other day for a total of 8 doses was started on hospital day 13.
Due to thrombocytopenia at baseline which worsened during her ICU stay (platelets level were as low as 35,000/cm2, down from 90,000/cm2 at baseline), pharmacologic venous thromobemolism prophylaxis was held and reinitiated on hospital day 17 with Heparin 5,000 units SC every 8 hours when her platelet levels increased to 83,000/cm2. However, patient developed lower extremity deep vein thrombosis on hospital day 23 and was started on enoxaparin (Lovenox) 60 mg subcutaneously every 12 hours at a reduced dose due to kidney dysfunction.
Anemia is common among critically ill patients and may be the result of blood loss, direct inhibition of erythropoiesis by inflammatory cytokines, nutritional deficiencies, and decreased production of erythropoietin due to kidney dysfunction.3,4 About 50% of ICU patients receive transfusions for anemia management. However, red blood cell transfusion has been shown to increase morbidity and mortality in critically ill patients and a restrictive transfusion strategy is recommended.5,6
More recently, recombinant human erythropoietin therapy (rHuEPO) has become a mainstay in the management of anemia secondary to chronic kidney disease and cancer and has been used for the non-FDA approved indication of anemia of critical illness. rHuEPO induces erythropoiesis by stimulating division and differentiation of erythroid progenitor cells and releases reticulocytes from bone marrow to blood stream, and has been suggested as a safer alternative to transfusion. However, the onset of action of rHuEPO is typically 2-4 weeks and may increase the risk of thrombosis. The purpose of this paper is to identify, evaluate, and apply primary literature to answer the question, “What is the role of rHuEPO in the management of anemia of critical illness for this patient?”
The search began with a PubMed search using the terms, “epoetin alfa, critically ill, blood transfusion, and mortality,” and yielded 20 results including 9 review articles, 3 free full text articles, and a meta-analysis of randomized controlled trials of erythropoietin receptor agonists in critically ill patients.7 The meta-analysis included nine studies, five of which were conducted in North America. The most frequent dose of epoetin alfa that was used was 40,000 units/week. The EPO-1 and EPO-2 trials were conducted in 1999 and 2002 and showed that epoetin alfa decreased number of red-cell transfusions.8,9 However, the more recent EPO-3 trial that was conducted in 2007 contradicted the outcome of the previous two studies.10 The conclusion of the meta-analysis was consistent with the findings of the 2007 study that epoetin alfa does not reduce mortality in critically ill patients. Additionally, the meta-analysis also reported that epoetin alfa does not significant decrease red blood cell transfusions in critically ill patients and therefore the use is not recommended. Based on the large sample size, study design, and findings, the EPO-3 study titled, “Efficacy and Safety of Epoetin Alfa in Critically Ill Patients,” was selected for evaluation.
The study is a prospective, randomized, placebo-controlled, multinational that aimed to assess the effect of epoetin alfa on the need for blood transfusion, safety, and clinical outcomes. The primary endpoint of the study was the percentage of patients who received any red-cell transfusion between days 1 and 29. Secondary endpoints were the number of red-cell units transfused between days 1 and 42, mortality at day 29 and 140, and the change in hemoglobin concentration from baseline to day 29. The study was conducted over a period of 3 years (2003-2006) and included a mixed ICU patient population.
Epoetin Alfa (Procrit) 40,000 U or an identical placebo were administered subcutaneously on day 1, 8, and 15. All patients received liquid iron by mouth or NG tube on day 1 or when they could tolerate NPO. However, IV iron was considered if patients did not respond to PO iron. The inclusion criteria were patient who remained in the ICU for 2 days, age ≥ 18 years, and Hgb concentration < 12g/dL. Patients were excluded if they were expected discharge from ICU within 48 hours after the 2nd day in ICU, acute ischemic heart disease, PE, DVT, hypercoagulable disorder, uncontrolled hypertension (>200/110 mmHg) on anti-hypertensive medications, new onset of seizure, third degree burns, pregnancy/lactation or treatment with epoetin alfa within the past 30 days.
A total of 1460 patients, 733 in the epoetin alfa group and 727 in the placebo group, were enrolled and were randomized to three admission groups: trauma, surgical, non-trauma and medical, non-trauma. All patients were followed for 140 days. Characteristics of patients in two groups were similar, however the SICU and MICU patient population were older than trauma patients. Mean APACHE II score for enrolled patients was about 20 and about 13% of patients had an initial diagnosis of sepsis. Ninety four percent of the participants completed the study.
In the treatment group, 28.2% of patients received one dose, 32.3% received two doses and 38.9% received three doses. The study found no significant difference between the groups for primary outcome (p=0.34). A total of 74.8% patients in the epoetin alfa group and 77.2% in the placebo group received red-cell transfusion with Hgb target between 7g/dL and 9 g/dL. The number of units transfused between the two groups was similar. There was an increase in hemoglobin concentration in the epoetin alfa group from baseline (P<0.001) at day 29. At day 42, the hemoglobin concentrations in the two study groups were not significantly different.
Mortality at day 29 was lower in the epoetin alfa group vs. placebo in the subgroup of trauma patients (3.5% vs. 6.6%, p=0.04) and at day 140 (6.0% vs. 9.2%, p=0.08). Mortality was similar in the SICU and MICU, non-trauma patients at day 29 and 140. The safety analysis identified an increased incidence of thrombotic vascular events in the treatment group compared to placebo (16.5% vs. 11.5% p=0.008). The event rates were similar in all three admission groups and a post hoc analysis showed that thrombotic events were most prevalent in patients who did not receive DVT prophylaxis at baseline (p=0.008) and most apparent among patients who received three doses of epoetin alfa (p=0.048).
The study failed to meet the primary endpoint of reduced transfusions in the treatment group although it did show transient, clinically insignificant increases in Hgb. The authors correctly cite changes in transfusion practices as a factor that may have contributed to efficacy shown in earlier studies. An increase in thrombotic events was not noted in previous trials but thrombotic events occurred more commonly in patients receiving rHuEPO despite the restrictive exclusion criteria. The finding of mortality benefit for trauma patients receiving rHuEPO in the pre-specified subgroup analysis is certainly hypothesis generating, at best. However, we disagree with the study investigators that this was the most important finding of the study. In November 2009, the manufacturer recently terminated a study due to slow enrollment, which was designed to answer this question.11 In response to the EPO-3 study, a published opinion cautions clinicians against the use of rHuEPO in critically ill patients without other clear indications.12 Institutions that remove rHuEPO from protocols for trauma and other critically ill patients may benefit from drug cost savings due to decreased usage.13
Application to Patient Case
The patient case presented here illustrates the complexity of critically ill surgical patients who are anemic due to feeding difficulty, blood loss, infection, and prolonged inflammatory states. Although well designed, the EPO-3 study leaves important unanswered questions. Given the delayed onset of action of the medication, for patients with surgery complications and prolonged ICU stays, could there be improved efficacy with greater exposure? Unfortunately, only one-third of study patients received rHuEPO therapy for more than three weeks and even these patients may not derive benefit until after one or two months. Although side effects were noted for rHuEPO, appropriate pharmacologic prophylaxis may mitigate any risks of thrombotic events. Still, the harms of transfusion are likely greater. Are we back where we began? What is the future of anemia management in the intensive care unit?
When analyzing the data found in the biomedical literature, it is important to consider what the outcomes of the studies mean to the patient as an individual. A patient is less concerned about correcting lab values and more concerned how the lab values or treatment options will affect the quality and quantity of his or her life. In regards to the presented case, our patient would be most concerned with the consequences of untreated anemia, the adverse effects of the treatment options including red blood cell transfusions, epoetin alfa, and iron replacement therapy, and decreasing her ICU and hospital lengths of stay.
Does epoetin alfa reduce the incidence of red blood cell transfusion in critically ill patients? This question was raised while treating this patient due to lack of efficacy and the occurrence of a potential adverse event. After conducting a comprehensive literature search, evaluating patient oriented outcomes, and applying this to the patient case, our POEM suggests a limited role for rHuEPO in the treatment of anemia of critical illness.
Shaughnessy AF, Slawson DC, Bennett JH. Becoming an information master: A guidebook to the medical information jungle. J Fam Pract. 1994;39:489-99. Slawson DC, Shaugheness AF. Becoming an information master: Using POEMs to change practice with confidene. J Fam Pract. 2000;49(1):63-7. Asare K. Anemia of critical illness. Pharmacotherapy. 2009;28(10):1267–82. Bateman AP, McArdle F, Walsh TS. Time course of anemia during six months follow up following intensive care unit discharge and factors associated with impaired recovery of erythropoiesis. Crit Care Med 2009;37(6):1906-12. Marik PE, Corwin HL. Efficacy of red blood cell transfusion in the critically ill: A systematic review of the literature. Crit Care Med. 2008;36(9):2667-74. Hebert 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(6):409-17. Zarychanski R, Turgeon A, et al. Erythropoietin-receptor agonists in critically ill patients: a meta-analysis of randomized controlled trials. CMAJ. 2007;177(7): 725–34. Corwin HL, Gettinger A, Rodriguez RM, et al. Efficacy of recombinant human erythropoietin in the critically ill patient: a randomized double blind placebo controlled trial. Crit Care Med. 1999;27:2346-50. Corwin HL, Gettinger A, Rodriguez RM, et al. Efficacy of recombinant human erythropoietin in the critically ill patient: a randomized double blind placebo controlled trial. JAMA. 2002;288:2827–35.
10. Crowin H, Gettinger A, Fabian TC, et al. Efficacy and safety of epoetin alfa in critically ill patients. N Engl J Med. 2007;357(10):965-76.
11. ClinicalTrials.gov. NCT00210626: Assess functional outcomes in anemic, critically ill, trauma patients when taking epoetin alfa. National Institutes of Health. Accessed at http://clinicaltrials.gov/ct2/show/NCT00210626.
12. Cook D, Crowther M. Targeting anemia with erythropoietin during critical illness. N Engl J Med. 2007;357(10):1037-9.
13. Christmas AB, Camp SM, Barrett MC. Removal of erythropoietin from anaemia trauma practice guideline does not increase red blood cell transfusions and decreases hospital utilization costs. Injury. 2009;40:1330–35.
Figure 1: Hematocrit Values During ICU Admission
The following items correspond to the numbers on the above figure: 1- Blood loss following exploratory, segmental bowel resection, and release of ventral hernia received 2 units of pRBCs. 2 – 500 mL Blood loss following exploratory laparotomy, segmental small bowel resection and primary anastomosis, received 2 units pRBCs and 1 unit platelets, 6 liters of IV fluid. 3- Minor blood loss in oral cavity. 4- Anastamotic leak, exploratory laparotomy, abdominal washout, repair graham patch for proximal anastomotic leak, received RBC transfusion. 5- RBC transfusion. 6- Blood loss following surgery. 7- RBC transfusion. 8- Bleeding from left JP site, RBC transfusion.