Abstract 4143487: Refining Risk Prediction in Systemic Sclerosis Pulmonary Arterial Hypertension

Background: Given the high morbidity and mortality of pulmonary arterial hypertension (PAH), several risk stratification tools have been developed to enhance patient prognostication and management. PAH commonly complicates patients with systemic sclerosis (SSc). However, the disease-specific performance of these scores is unknown. Research Question: To determine the prognostic value of clinical PAH risk scores in SSc and further determine if echocardiographic coupling metrics improve risk score performance. Methods: We included a well-characterized cohort of SSc-PAH patients from Johns Hopkins Medicine with echocardiograms within six months of invasive confirmation of PAH status. Conventional and speckle-tracking echocardiographic analysis was performed. Four scores were calculated: COMPERA 2 4-Risk Strata, REVEAL 2.0, REVEAL Lite 2, and French Criteria. Cox regression with the corresponding C-statistic was used to assess each hazard ratio and predictive power. A new model incorporating coupling parameters with the risk score was generated and compared to the original risk model with Likelihood Ratio (LR) tests. Results: Our cohort consisted of 174 SSc-PAH patients, 63.6 ± 12.7 years, 85.6% female, and 76.4% White, 86% limited SSc, with moderate-severe SSc-PAH on average. Table 1 . Across risk scores, C-statistics ranged from 0.63-0.69 in SSc-PAH, Table 2 . The incorporation of RVGLS/PASP, RVFWS/PASP, and mid-RVFWS/PASP to each risk score model significantly improved mortality prediction (LR test p-value<0.05) with RVGLS/PASP providing the greatest difference in C-statistic across all models. Conclusion: Established risk scores have a modest predictive capacity for mortality in SSc-PAH. Incorporating echocardiographic parameters of right ventricular to pulmonary arterial coupling enhances the performance of these risk scores, aiding in the prognostication of mortality.