Previous heart failure (HF) trials suggested that age influences patient characteristics and outcome; however, under-representation of elderly patients has limited characterization of this cohort. Whether standard prognostic variables have differential utility in various age groups is unclear.Methods and results
The PROTECT trial investigated 2033 patients (median age 72 years) with acute HF randomized to rolofylline or placebo. Patients were divided into five groups based on the quintiles of age: ≤59, 60–68, 69–74, 75–79, and ≥80 years. Baseline characteristics, medications, and outcomes (30-day death or cardiovascular/renal hospitalization, and death at 30 and 180 days) were explored. The prognostic utility of baseline characteristics for outcomes was investigated in the different groups and in those aged <80 years vs. ≥80 years. With increasing age, patients were more likely to be women with hypertension, AF, and higher EF. Increased age was associated with increased risk of 30- and 180-day outcomes, which persisted after multivariable adjustment (hazard ratio for 180-day death = 1.17; 95% confidence interval 1.11–1.24 for each 5-year increase). The prognostic utility of baseline characteristics such as previous HF hospitalization and serum sodium, systolic blood pressure, and NYHA class was attenuated in the elderly for the endpoint of 180-day mortality. An increase in albumin was associated with a greater reduction in risk in patients aged ≥80 years vs. <80 years.Conclusions
In a large trial of acute HF, there were differences in baseline characteristics and outcomes amongst patients of different ages. Standard prognostic variables exhibit different utility in elderly patients.