Abstract 21196: Validity of the Seattle Heart Failure Model in Predicting Outcomes on the Transplant Candidate List
Background: Mortality in advanced heart failure (HF) patients (pts) on the transplant (TX) candidate list remains high and reliable assessment of prognosis to allow for timely life saving intervention such as implantation of a left ventricular assist device (LVAD) or urgent TX (UNOS Status 1A) remains a clinical challenge. The Seattle Heart Failure Model (SHFM) uses easily available clinical variables to derive one-, two-, and three-yr survival rates and average yrs of survival for ambulatory, stable, chronic HF pts. Though, SHFM has been shown to underestimate risk of death in advanced HF pts referred for TX, the use of SHFM in predicting absolute risk among listed TX candidates remains unknown.
Methods: Data on 88 consecutive pts (mean age 55± 11 yrs; 80% males; 88% caucasian; mean LVEF 19± 8%; 82% on angiotensin II modulation; 59% on beta blockers; 31% on aldosterone antagonists; 71% with ICD± CRT) listed for TX between January 1, 2006 and April 30th, 2009 were retrospectively abstracted to calculate the SHFM predicted 1 yr risk of death. This was compared to the observed 1 yr risk of death, and the observed cumulative risk of death or LVAD insertion or urgent TX. The chi-square test and the C statistic were used to evaluate the differences.
Conclusions: The SHFM tends to overestimate risk of death among TX candidates and significantly underestimates the cumulative risk of death/LVAD insertion/urgent TX. Other models are therefore needed to reliably assess prognosis in TX candidates.
- © 2010 by American Heart Association, Inc.