Background: Targeted therapy of pulmonary hypertension (PH) should be considered in patients with pre-capillary PH and cardiovascular comorbidities, and it may be considered in patients with post-capillary PH caused by left heart disease in selected cases with markedly elevated pulmonary vascular resistance (PVR). Both patient groups may be characterised using the term “left heart phenotype”. However, the traditional threshold value for the pulmonary artery wedge pressure (PAWP) to distinguish pre- from post-capillary PH of 15 mmHg and the key role of PAWP for therapeutic decision making is currently under debate. The primary objective of the present study was to investigate possible associations of haemodynamic variables measured at rest and during exercise with clinical therapy response.
Methods: Consecutive patients with a left heart phenotype undergoing right heart catheterization (RHC) for evaluation of PH and receiving targeted PH therapy at one single centre were analysed. A positive clinical therapy response was defined as an improvement of Functional Class (FC) by ≥ 1 class in conjunction with at least stable N-terminal pro-brain natriuretic peptide (NT-proBNP) levels at follow-up, or decrease of NT-proBNP levels by ≥ 30% with at least stable FC.
Results: A total of 60 patients with complete resting haemodynamic data sets and baseline right atrial pressure (RAP) ≤ 15 mmHg were included in the analysis; 30 of them underwent additional exercise RHC. Patients were in mean 70 (±10) years of age, 53% were male, and 78% presented with symptoms of heart failure according to FC class III. Targeted PH treatment consisted of PDE5 inhibitors in 95% of patients. After a median follow-up time of 99 (IQR 65-137) days, 50% showed a positive clinical therapy response as defined above. Using the traditional PAWP threshold value, 82% had post-capillary PH, and 18% pre-capillary PH. The mean value for RAP was 9.7 (±3.4) mmHg, for mean PA pressure 39.0 (±8.2) mmHg, for PAWP 20.7 (±5.7) mmHg, for pulmonary vascular resistance (PVR) 5.3 (±2.2) Wood Units, and the median value for cardiac output was 3.5 (IQR 2.9-4.3) l/min. The median serum level of NT-proBNP was 2824 pg/ml (IQR 1283-4800).
Groupwise comparisons of responders vs non-responders showed only two significantly different parameters between groups: PVR at rest was higher in responders (5.9 vs 4.6; p=0.01), and RAP measured immediately post exercise was higher in non-responders (24.5 vs 17.6; p=0.03). All other investigated parameters, among them NT-proBNP levels at baseline, PAWP at rest and with exercise and others were not different. This was confirmed by logistic regression analysis: a positive therapy response was associated with PVR at rest (OR 1.38, 95%CI 1.07-1.84; p=0.01; AUC 0.68, p=0.02) and RAP immediately post exercise (OR 0.79, 95%CI 0.58-0.97; p=0.02; AUC 0.83, p=0.03).
Conclusions: In our cohort of patients with “left heart phenotype” PH, clinical response to targeted PH therapy seemed to be independent from PAWP at rest and during exercise. The key role of PVR for therapeutic decision making could be confirmed. A higher RAP post exercise was associated with non-response, possibly indicating advanced right ventricular dysfunction.
