Effectiveness Study of rHuEPO in the ICU

Purpose: To evaluate the clinical outcomes and resource use in ICU patients receiving rHuEPO in a naturalistic setting.
Methods: A retrospective, case-matched (1:2 ratio) study compared patients receiving rHuEPO to a control group. Patients admitted between January 2000 and July 2002 with an ICU length of stay (LOS) ¡Ý3 days were identifi ed by an electronic data repository. Patients, who received rHuEPO prior to ICU admission, had chronic renal failure or were <18 years of age were excluded. Patients were matched by age (¡À5 years), sex, admission year and ICU type. Collected data included patient demographics, admission date, ICU and hospital mortality and LOS, mechanical ventilation days, serum creatinine concentration, hemoglobin concentration, number of blood transfusions, and ICU resource use. Results: rHuEPO-treated patients (n=391) were matched with 782 controls. Patients receiving rHuEPO had higher Simplifi ed Acute Physiology Scores II (46.2 vs 38.8; p <0.001) and received signifi cantly more blood transfusions than control patients (19 vs 6; p <0.001). After adjusting for severity of illness in a linear regression model, rHuEPO was signifi cantly associated with increased blood transfusions and higher mortality risk. Patients receiving rHuEPO had signifi cantly longer hospital and ICU LOS, mechanical ventilation duration, and higher hospital and ICU mortality rate and hospital resource use (p <0.001). Conclusions: In this real-world retrospective analysis, critically ill patients treated with rHuEPO did not experience clinical benefi ts; however, patients were sicker and received rHuEPO late in their ICU stay. Monitoring prescribing patterns and patient selection of rHuEPO treatment in critically ill patients in clinical practice is recommended to optimize rHuEPO use and outcomes.