Abstract 20581: Population Based Characterization of the Definition, Prevalence and Associated Mortality of Pulmonary Hypertension in the Setting of Left Heart Disease
Purpose: Patients with left heart disease can develop pre-capillary pulmonary vascular remodeling, resulting in elevations of pulmonary vascular resistance (PVR). Pre-capillary pulmonary vascular disease has been variably defined by elevations in the transpulmonary or diastolic pulmonary pressure gradients (TPG or DPG) or PVR, but the optimal hemodynamic definition and predictor of mortality remains uncertain.
Methods and Results: We retrospectively analyzed a large cohort of 10,023 subjects who had a right heart catheterization at our institution between 2005 and 2012, with linkage of catheterization data to electronic health records with natural language-based search and social security death index. Pulmonary hypertension in the setting of heart failure with preserved ejection fraction (PH-HFpEF) was defined as mPAP ≥25 mmHg, PCWP ≥15 mmHg and LVEF ≥ 45%, with variable TPG, PVR and DPG cut-off values to quantify severity of PH. Cox proportional hazard regression analysis was used. In cases of non-linear associations with mortality (PVR and DPG), hazard ratios were calculated based on spline models fit to a parameter.
Of the 10,023 subjects, 21.4% have PH-HFpEF and 10.4%, 7.4% and 2.9% have PH-HFpEF with an elevated TPG (>12), PVR (≥3) and DPG (≥7), with 5-year survival rates of 50%, 46% and 45%, respectively (Figure 1). Predictive value was confirmed over a wide range of values by spline-modeled Cox analysis: TPG HR 1.02 (95%CI 1.01-1.02; P < 0.0001) across all values; PVR HR 1.06 (95%CI 1.04-1.08; P < 0.0001) in subjects with a PVR > 1 (91% of observations); DPG HR 1.03 (95%CI 1.02-1.04; P < 0.0001) in subjects with a DPG ≥ 0 (47% of observations). PVR ≤ 1 and DPG < 0 were not significantly associated with mortality.
Conclusions: PH-HFpEF is highly prevalent in patients referred for right heart catheterization. TPG, PVR and DPG are highly predictive of mortality by standard cut-offs and as continuous variables in a large PH-HFpEF cohort.
Author Disclosures: R.R. Vanderpool: None. M.I. Saul: None. S. Nouraie: None. M.T. Gladwin: None. M.A. Simon: None.
- © 2016 by American Heart Association, Inc.