The aim was to evaluate computed tomography (CT)-measured pulmonary artery diameter (PAD) and lung density as predictors of pulmonary hypertension (PH) in subjects with systemic sclerosis (SSc). We compared these PAD values with normal values and between SSc subgroups with PH and/or interstitial lung disease (ILD). We investigated whether PAD predicts PH and whether lung densitometry, by using the 85th percentile density value (Perc85) as a measure for ILD, can predict PH. PAD and Perc85 were measured in axial CT scans and compared between 54 SSc and 76 control subjects. Four SSc subgroups were defined on the basis of PH (systolic PA pressure ≥35 mm Hg) and/or ILD (fibrosis score ≥7): PH-/ILD-, PH-/ILD+, PH+/ILD-, and PH+/ILD+. The association of PAD with age, body mass index, Perc85, lung function, and hemodynamic measures was investigated using univariate correlation along with the predictive value of these measures with respect to PH. PAD in SSc was larger than that in controls (30.1±4.9 vs. 26.9±2.7 mm, P<0.001). PH+ patients showed increased PAD compared with PH- patients (34.2±4.2 vs. 28.6±4.3 mm, P<0.001), where PH+/ILD+ subjects showed the widest diameter (34.6±4.1 mm). In SSc patients, hemodynamic measures, age, body mass index, Perc85, and lung function correlated with PAD. PAD was best explained by Perc85, together with age (R=0.358). PAD best predicted PH (AUC, 0.877; P<0.001), and PAD≥30.7 mm showed 80% sensitivity and 87% specificity. Perc85 also predicted PH (AUC, 0.733; P=0.024). In subjects with SSc, lung density and PAD are CT markers, each with predictive value for PH.