Search

Background Pulmonary artery wedge pressure (PAWP) during exercise, as a surrogate for left ventricular (LV) end-diastolic pressure (EDP), is used to diagnose heart failure with preserved ejection fraction (HFpEF). However, LVEDP is the gold standard to assess LV filling, end-diastolic PAWP (PAWPED) is supposed to coincide with LVEDP and mean PAWP throughout the cardiac cycle (PAWPM) better reflects the haemodynamic load imposed on the pulmonary circulation. The objective of the present study was to determine precision and accuracy of PAWP estimates for LVEDP during exercise, as well as the rate of agreement between these measures. Methods 46 individuals underwent simultaneous right and left heart catheterisation, at rest and during exercise, to confirm/exclude HFpEF. We evaluated: linear regression between LVEDP and PAWP, Bland–Altman graphs, and the rate of concordance of dichotomised LVEDP and PAWP ≥ or < diagnostic thresholds for HFpEF. Results At peak exercise, PAWPM and LVEDP, as well as PAWPED and LVEDP, were fairly correlated (R2>0.69, p2 mmHg·L−1·min−1. Conclusions During exercise, PAWP is accurate but not precise for the estimation of LVEDP. Despite a good rate of concordance, these two measures might occasionally disagree.

Background and aims Pulmonary hypertension due to left heart disease (PH-LHD) is the most frequent form of PH. As differential diagnosis with pulmonary arterial hypertension (PAH) has therapeutic implications, it is important to accurately and noninvasively differentiate PH-LHD from PAH before referral to PH centres. The aim was to develop and validate a machine learning (ML) model to improve prediction of PH-LHD in a population of PAH and PH-LHD patients. Methods Noninvasive PH-LHD predictors from 172 PAH and 172 PH-LHD patients from the PH centre database at the University Hospitals of Leuven (Leuven, Belgium) were used to develop an ML model. The Jacobs score was used as performance benchmark. The dataset was split into a training and test set (70:30) and the best model was selected after 10-fold cross-validation on the training dataset (n=240). The final model was externally validated using 165 patients (91 PAH, 74 PH-LHD) from Erasme Hospital (Brussels, Belgium). Results In the internal test dataset (n=104), a random forest-based model correctly diagnosed 70% of PH-LHD patients (sensitivity: n=35/50), with 100% positive predicted value, 78% negative predicted value and 100% specificity. The model outperformed the Jacobs score, which identified 18% (n=9/50) of the patients with PH-LHD without false positives. In external validation, the model had 64% sensitivity at 100% specificity, while the Jacobs score had a sensitivity of 3% for no false positives. Conclusions ML significantly improves the sensitivity of PH-LHD prediction at 100% specificity. Such a model may substantially reduce the number of patients referred for invasive diagnostics without missing PAH diagnoses.