Divergent Determinants of Mortality in Connective Tissue Disease‑associated Versus Connective Tissue Disease-non-associated Pulmonary Arterial Hypertension: An Interaction Analysis on Hematologic and Inflammatory Markers

Divergent Determinants of Mortality in Connective Tissue Disease‑associated Versus Connective Tissue Disease-non-associated Pulmonary Arterial Hypertension: An Interaction Analysis on Hematologic and Inflammatory Markers

Objectives: Prognostic models for pulmonary arterial hypertension (PAH) are largely derived from idiopathic or mixed populations. We investigated whether the determinants of mortality in PAH are etiology‑specific, focusing on the differences between connective tissue disease‑associated PAH (CTD‑PAH) and non‑CTD‑PAH. Methods: This retrospective cohort study included 138 PAH patients stratified by etiology (CTD‑PAH, n = 41; non‑CTD‑PAH, n = 97). Multivariable Cox proportional hazards models and interaction analysis, adjusted for demographics and PAH‑targeted therapies, were performed to evaluate the prognostic value of clinical and laboratory parameters for 3‑year mortality. Results: Despite a more severe baseline phenotype (higher functional class and inflammatory burden), CTD‑PAH patients exhibited 3‑year survival comparable to non‑CTD patients. Interaction analysis revealed distinct prognostic profiles. In non‑CTD‑PAH, higher hemoglobin was protective (adjusted HR [aHR]: 0.690, P < 0.001), and elevated neutrophil‑to‑lymphocyte ratio (NLR) predicted mortality (aHR: 1.161, P < 0.001). However, these associations were abrogated in CTD‑PAH (Pinteraction = 0.046 and < 0.001, respectively), where neither marker predicted outcome. Conversely, red blood cell distribution width–standard deviation (RDW‑SD) emerged as a consistent predictor across both subgroups (CTD: aHR = 1.094; non‑CTD: aHR = 1.074) without significant interaction (Pinteraction = 0.741). Conclusions: Traditional prognostic markers are not universally applicable. While the prognostic utility of anemia and NLR is notably attenuated in CTD‑PAH–potentially due to confounding from systemic autoimmunity or therapy–RDW‑SD emerges as a robust, etiology‑independent predictor. These findings highlight the need to move beyond a“one‑size‑fits‑all”strategy toward precision risk stratification that accounts for the specific inflammatory background of connective tissue diseases.