Abstract Rationale Pulmonary hypertension (PH) associated with COPD (PH-COPD) leads to reduced exercise capacity, impaired quality of life, and increased mortality. Severe PH-COPD is currently defined by invasive hemodynamic thresholds (PVR 5 WU), and the ability of noninvasive measures to identify high-risk patients remains uncertain. This study evaluated whether noninvasive indices, including non-invasive cardiopulmonary exercise testing (niCPET) variables, predict adverse clinical outcomes in PH-COPD. Methods A retrospective analysis was performed of patients undergoing right heart catheterization between 2019-2025. Patients were included if they had COPD by GOLD criteria and met diagnostic criteria for PH per the 2022 ESC/ERS Guidelines. Associations between clinical variables and a composite outcome (heart failure hospitalization or all-cause mortality) were assessed using Cox proportional hazards models. To evaluate the relative contribution of significant variables (p 0.05), and identify combinations best discriminating risk, a survival decision-tree analysis using CART was performed. Models were constructed with and without PVR to determine whether noninvasive parameters alone could stratify prognosis. Results 68 patients were included (mean age 69 ± 9 years; 53% female). Mean FEV1 was 59.8 ± 21.1% predicted, and corrected DLCO was 42.4 ± 20.9% predicted. Common comorbidities included obesity (mean BMI 28.8 ± 7.7 kg/m²), hypertension (55.9%), obstructive sleep apnea (32.4%), and atrial fibrillation (30.9%). Most patients had pre-capillary (62.2%) or combined post- and pre-capillary pulmonary hypertension (30.9%). 50% had a pulmonary vascular resistance (PVR) 5 Wood units. Several noninvasive indices were significantly associated with adverse outcomes (p 0.05) including pulmonary function test parameters (lower DLCOc); niCPET variables (lower gas exchange derived pulmonary vascular compliance (GXCAP), lower oxygen uptake efficiency, and lower submaximal VO2; and echocardiographic measures (higher eRVSP and lower TAPSE/eRVSP). In the decision-tree model, patients with GXCAP 243 and eRVSP ≥ 45 mmHg had the highest risk (relative HR 4.4). A second high-risk group included those with preserved GXCAP but reduced DLCOc 48 % and TAPSE/RVSP 0.22 (relative HR 1.4). While higher PVR correlated with adverse outcomes in Cox modeling, it did not improve risk stratification in the decision-tree. Conclusions In PH-COPD, combining GXCAP with eRVSP identified high-risk individuals with poor outcomes better than invasive hemodynamics. Reduced GXCAP reflects impaired pulmonary vascular compliance, while elevated eRVSP reflects increased right-ventricular afterload. Together, these measures capture the physiologic consequences of pulmonary vascular remodeling and offer a noninvasive means to identify advanced disease. This abstract is funded by: None
Tarras et al. (Fri,) studied this question.