Send to

Choose Destination
See comment in PubMed Commons below
J Card Fail. 2004 Dec;10(6):460-6.

Risk stratification after hospitalization for decompensated heart failure.

Author information

  • 1Duke Clinical Research Institute, 2400 Pratt Street, Room 0311, Terrace Level, Durham, NC 27705, USA.



Decompensated heart failure (HF) is among the most common indications for hospitalization in the United States, but little is known about features on admission that predict adverse events. We used data from the Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure (OPTIME-CHF) study to develop a model that would predict outcomes in patients with decompensated HF.


OPTIME-CHF randomized 949 patients hospitalized with decompensated HF for 48 to 72 hours of infusion of either milrinone or placebo. We used multivariable modeling to evaluate variables on admission that would be predictive of 60-day mortality or the composite of death or rehospitalization at 60 days. Variables at presentation that predicted death at 60 days were increased age, lower systolic blood pressure, New York Heart Association class IV symptoms, elevated blood urea nitrogen (BUN), and decreased sodium. Predictors of the composite of death or rehospitalization within 60 days were the number of HF hospitalizations in the preceding 12 months, elevated BUN, lower systolic blood pressure, decreased hemoglobin, and a history of percutaneous coronary intervention (PCI). The discriminatory power of the model was substantial for the mortality model (c-index .77) but less for the composite endpoint (c-index .69).


Risk stratification of patients with decompensated HF may be accomplished using easily assessed clinical variables. Further research into the validity of this model in independent samples will potentially aid in the development of risk stratification strategies.

[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk