Meeting Brochure and registration form      SMDM Homepage

Tuesday, 17 October 2006


Marya D. Zilberberg, MD, Ortho Biotech Clinical Affairs, Bridgewater, NJ and Andrew F. Shorr, Washington Hospital Center, Washington, DC.

Purpose: Recent epidemiologic studies indicate that nearly half of all ICU patients require allogeneic blood.1 While it is well known that blood transfusion (TF) is associated with an increased risk of complications, only one randomized controlled trial (RCT) comparing two different transfusion thresholds is published that helps quantify the acute complications directly attributable to incremental blood administered.2 In a Phase 3 study, erythropoietin (EPO) reduced TF incidence as well as frequency and number of blood transfusions in anemic ICU patients.3 We wanted to estimate the potential annual reduction in transfusion-attributable serious acute complications (TSACs) in US ICUs with use of EPO 40,000 U SQ QW in selected patients.

Methods: Utilizing inputs from published literature, a model was constructed. Step 1 calculated total number of potential EPO-eligible patients in US ICUs annually (defined as all adult ICU patients who are alive and at risk of TF on day 7 or later4). In step 2, the number of packed red blood cell (pRBC) units avoided in the EPO arm of the RCT3 was extrapolated to the population in all US ICUs annually. Step 3 quantified excess complications (adult respiratory distress syndrome (ARDS) and cardiovascular (CV) complication rates were higher in patients who experienced more transfusions in the ICU2; this excess of complications was defined as the TSACs) and the number of pRBC units transfused per one TSAC based on the randomized transfusion threshold trial data.2 Lastly, step 4 translated EPO-related avoidance of pRBC units to a reduction in TSACs.

Results: 1,624 additional units transfused resulted in 49 incremental incident cases of TSACs or 33 units per TSAC. The base-case analysis showed that 10,315 TSACs could potentially be avoided annually in US ICUs by using EPO in selected ICU patients. Sensitivity analyses based on ranges of inputs yielded the lower and upper bound estimates of 1,783 and 37,905 TSACs avoided, respectively.

Conclusions: Based on this model, judicious use of EPO in selected ICU patients in the US may result in a reduction in TSACs.

1Corwin CCM 2004; 2Hebert NEJM 1999; 3Corwin JAMA 2002; 4Millbrandt Crit Care 2006

Disclosure: Funded by Ortho Biotech Clinical Affairs, LLC

See more of Poster Session III
See more of The 28th Annual Meeting of the Society for Medical Decision Making (October 15-18, 2006)