The objective of this study was to determine clinical predictors of interstitial lung disease in patients with systemic sclerosis (SSc) and pulmonary involvement as defined by presence of a decreased diffusing capacity for carbon monoxide (DLCO). Forty subjects with SSc were retrospectively evaluated. Patients were categorized according to their level of DLCO (< o > or = 80% of predicted). Sensitivity of dyspnea to detect a decreased DLCO was 46.6% and specificity 90%, whereas oxygen desaturation showed a sensitivity of 71.4% and a specificity of 80%. Patients with decreased DLCO (n = 18) were not different in age (51.1 +/- 13.5 vs. 53.5 +/- 9.3 y, p = 0.5182), sex (male 13.6%, p = 0.6088), prevalence of Raynaud (86.6% vs. 85%, p = 0.6272), sicca syndrome (6.2% vs. 10.5% p = 1.0000) diffuse cutaneous involvement (94.1% vs. 83.3%, p = 0.6026) or esophageal dilatation. The duration of symptoms since diagnosis was no different. Prevalence of pulmonary hypertension assessed by Doppler echocardiography or abnormal nailfold capillaroscopic findings were identical in both populations. Patients with low DLCO had a significatly higher prevalence of anti topoisomerase antibodies. (5/9 vs. 0/11, p = 0.0081) and restrictive lung disease. Patients with low DLCO showed a significantly higher prevalence of abnormal HRCT findings suggestive of ILD (82.3% vs. 5.8%, p < or = 0.0001). We conclude that a low DLCO is a frequent finding in SSc patients, strongly associated with HRCT signs of ILD. We have not found clinical factors predictive for a low DLCO.