Background/Purpose: Pulmonary hypertension (PH) is an important cause of morbidity and mortality in patients with systemic sclerosis (SSc). Different screening algorithms have been proposed for identifying patients who have a high probability of PH and require right heart catheterization (RHC), which is the gold standard for diagnosing PH. The aim of this study was to compare the performance of PH screening algorithms in patients with SSc.

Methods: Forty-eight consecutive pts fulfilling ACR/EULAR 2013 SSc criteria have been screened for PH until now, using the 2015 ESC/ERS, DETECT and ASIG algorithms. Pulmonary function tests (PFT), diffusing capacity of the lung for carbon monoxide (DLCO), trans-thoracic echocardiography, serum NT-proBNP, serum uric acid assay and high-resolution computed tomography (HRCT) were performed as needed. Pts with known PH, severe interstitial lung disease and severe left ventricular dysfunction (LVD) were not included. RHC was performed in all patients with positive screening according to any one of the screening algorithms. Pts with PH were classified according to the updated PH classification criteria. Sensitivity and specificity of the 3 screening algorithms were evaluated according to the established cut-off value of 25 mmHg for mean systolic pulmonary artery pressure and for the recently proposed cut-off value of 20 mmHg.

Results: Among the 48 SSc pts, 15 were excluded due to already diagnosed PH (n=4), LVD (n=4), no measurable tricuspid regurgitation velocity (TRV) (n=5) and coexisting lung cancer (n=2). Among the remaining 33 patients, 14 required RHC according to at least one of the screening algorithms. Demographic and clinical features were summarized in Table 1. Number of patients who had suspected PH and required RHC according to ESC/ERS 2015, DETECT and ASIG algorithms were 8 (%25), 9 (%27), and 13 (%41) respectively (Figure 1). Among the 14 who had RHC, PH was present in 3 pts according to the 25-mmHg cut-off (Group1 in 1, Group 2 in 1, Group 3 in 1) and in 8 pts according to the 20-mmHg cut-off (Group 1 in 5, Group 2 in 2, Group 3 in 1). The sensitivity,specificity, positive and negative predictive values of each algorithm are presented in Table 2. Sensitivity was similar at 100% for the 3 algorithms, but the ESC/ERS algorithm had better specificity, when PH was diagnosed with the 25-mmHg cut-off.  For the 20-mmHg cut-off, sensitivity was better with ASIG and the specificity was better with the ESC/ERS algorithm. For both cut-offs ESC/ERS had the best positive predictive value, and the best negative predictive value for the 20-mmHg cut-off. The negative predictive values were similar for the 3 algorithms.

Conclusion: The ESC/ERS algorithm seems to perform better in detecting PH in patients with SSc. However our numbers are still small for a firm conclusion. Another limitation of this study was that RHC was not performed in patients who did not fulfill criteria according to any of the screening algorithms. The sensitivities may be lower than what we propose, if there are patients with PH who are asymptomatic and not captured with any of the algorithms.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/comparison-of-different-pulmonary-hypertension-screening-algorithms-in-patients-with-systemic-sclerosis/