Abstract 1714: The Influence of Systolic Blood Pressure on the Long-term Prognostic Interpretation of Cardiopulmonary Exercise Test Variables in Patients With Heart Failure
Introduction: Cardiopulmonary exercise testing is used to assess prognosis in patients with heart failure (HF). However, exercise studies commonly do not integrate or even report systolic blood pressure (SBP) measurements and do not include long-term follow-up with sufficient sample sizes.
Hypothesis: We assessed the hypothesis that SBP would affect the predictive ability of exercise variables.
Methods: HF patients (n = 399, 51 ± 10 years) underwent exercise testing and had mortality and cardiac transplant follow-up for 10 years.
Results: SBP (Mean ± SD = 114 ± 19 mm Hg) had a threshold effect at 100 mm Hg, which improved the Chi Square from 8 as a continuous variable to 25 as a dichotomous variable. Cox regression identified a model including Log VE/VCO2 slope (Hazard Ratio (HR) = 61.5, p = 0.00), and SBP≤100mm Hg (HR = 2.0, p = 0.00), but not peak VO2 (HR = 1.0, p = 0.12), that best predicted death or emergent transplant. The Figure⇓ shows the combination of dichotomous SBP at 100 mm Hg and tertile VE/VCO2 slope (<31.9, ≥31.9and <38.9, ≥38.9) predicted mortality better than tertile peak VO2 (<13.9, ≥13.9and <17.8, ≥17.8) and tertile VE/VCO2 slope (Chi Square = 72.0 vs. 58.1) and the combination of slope and SBP more clearly identified risk over the 10 years of follow-up.
Conclusions: SBP provided more predictive information than peak VO2, whether considered by itself or within a multivariable model. SBP should not be overlooked when considering prognosis because it has a significant effect on the predictive ability of cardiopulmonary exercise data.