Abstract Body (Do not enter title and authors here): Introduction: Pulmonary hypertension (PH) is a heterogeneous disease characterized by impaired gas exchange (CO₂, O₂) in the pulmonary circulation, leading to dyspnea and exercise intolerance. We hypothesize that specific hemodynamic metrics correlate with poor gas exchange in PH phenotypes, which can be leveraged to define therapeutically relevant hemodynamic targets.

Aim: This study aims to define hemodynamic correlates of poor gas exchange (peak VO₂, V_E/VCO₂ slope) in heart failure with preserved ejection fraction (HFpEF) and pre-capillary PH.

Methods: We retrospectively analyzed 170 participants with invasive cardiopulmonary exercise testing: HFpEF (n=91), pre-capillary PH (n=54), and non-cardiac dyspnea (NCD, n=25). Linear regression models, adjusted for patient groups with interaction, were used to assess the association of peak VO₂ and V_E/VCO₂ slope with predictors (pulmonary vascular resistance [PVR], distensibility, and pulmonary arterial compliance [PAC]).

Results: In the order of NCD vs. HFpEF vs. pre-capillary PH, rest PAC (5.5±1.6 vs 3.8±1.7 vs 2.6±1.3 mL/mmHg), rest PVR (1.4±0.6 vs 3.2±2.0 vs 5.7±3.0 Woods unit), and distensibility (1.5±0.5 vs 0.8±0.4 vs 0.5±0.2 % per mmHg). In comparison to a non-significant correlation with rest PVR and distensibility, peak VO₂ (%predicted) showed significant correlation with rest PAC (p<0.01) with significantly positive slope of +6 in HFpEF and +10 in pre-capillary PH. Similarly, significant slopes were noted with rest PAC and V_E/VCO₂ relationship (p<0.01) in HFpEF (-3) and pre-capillary PH (-4).

Conclusion: Pulmonary arterial compliance (PAC) is the best correlate of poor gas exchange. Improving rest PAC by 1.0 mL/mmHg can lead to improvement in peak VO₂ predicted (+10% in pre-capillary PH, +6% in HFpEF), and V_E/VCO₂ slope (-4 in pre-capillary PH, and -3 in HFpEF). In comparison to pre-capillary PH, more limited improvement in HFpEF is likely due to multi-organ dysfunction leading to exercise intolerance.