Abstract 5664: Circulating Soluble Endoglin is a Sensitive Predictor of Pulmonary Arterial Hypertension and Exercise Function
Rationale: Circulating biomarkers of inflammation and endothelial dysfunction are sensitive for the presence of systemic vascular disease and may aid in the early assessment of pulmonary arterial hypertension (PAH). Current biomarkers for PAH such as NT-proBNP provide useful prognostic data but detect ventricular strain associated with advanced or severe disease.
Objectives: We tested the hypothesis that circulating protein markers of endothelial dysfunction or inflammation – soluble endoglin (sEng), osteoprotegerin (OPG), high-sensitivity C-reactive protein (hsCRP) and interleukin 15 (IL-15) – may predict the presence or severity of PAH.
Methods: Circulating protein levels were measured by immunoassay in a total of 84 individuals: 42 with WHO diagnostic group 1 PAH of various etiologies (see table⇓), 11 of their first-degree relatives, and 31 healthy age- and gender-matched controls.
Measurements and Main Results: Circulating levels of sEng, hsCRP, OPG, and NT-proBNP were significantly elevated in PAH patients versus controls, whereas IL-15 levels were elevated only in PAH associated with connective tissue disease (p<0.001). sEng was superior to NT-proBNP for predicting PAH by receiver operating characteristic analysis (p=0.015). A sEng threshold of 3.7 ng/ml had a sensitivity of 86% and specificity of 74% for the presence of PAH, compared to a sensitivity of 70% and specificity of 74% for an NT-proBNP threshold of 180 pg/ml. Moreover, sEng levels correlated inversely with six minute walk distances (r=−0.48, p<0.05).
Conclusions: Serum markers of inflammation or vascular dysfunction are sensitive predictors for the presence of PAH of all etiologies and in the case of sEng correlate with reduced exercise capacity. Levels of IL-15 are elevated in PAH associated with connective tissue disease. Markers of inflammation may be more sensitive for PAH than those of ventricular strain and may provide unique noninvasive data.