BACKGROUND Collagen vascular disease-associated interstitial lung disease (CVD-ILD) must be differentiated from idiopathic pulmonary fibrosis (IPF) since prognosis and treatment strategies differ between these two conditions. However, differentiating between CVD-ILD and IPF is often difficult. OBJECTIVE To examine the utility of expiratory high-resolution computed tomography (HRCT) for differentiating between CVD-ILD and IPF. METHODS Seventy patients were examined with expiratory and inspiratory HRCT with CVD-ILD (n = 36) or IPF (n = 34). Associated diagnoses in patients with CVD-ILD were rheumatoid arthritis (n = 22), Sjögren syndrome (n = 3), scleroderma (n = 2), polymyositis/dermatomyositis (n = 1), and unspecified connective tissue disease (n = 8). Parenchymal abnormalities on inspiratory HRCT and visual extent of air trapping on expiratory HRCT were evaluated, statistical differences in HRCT findings between the two conditions were determined, and air trapping CT scores were correlated with the results of pulmonary function testing. RESULTS Air trapping was found in 27 (75%) of 36 cases of CVD-ILD and four (12%) of 34 cases of IPF. Seventeen of the 27 cases for which air trapping was exhibited with CVD-ILD were diagnosed with rheumatoid arthritis. A significant difference in frequency of air trapping was seen between CVD-ILD and IPF (P < 0.0001). Frequency of centrilobular nodules was significantly higher in CVD-ILD than in IPF (P = 0.021). In contrast, frequencies of interlobular interstitial thickening and traction bronchiectasis were significantly higher in IPF than in CVD-ILD (P = 0.005, P = 0.007, respectively). Correlations were seen between visual extent of air trapping and pulmonary function test results such as air trapping index (P = 0.004, r = 0.34), closing volume/vital capacity (P = 0.0002, r = -0.47), and closing capacity/total lung capacity (P < 0.0001, r = -0.51). CONCLUSIONS The presence of air trapping on expiratory HRCT suggests CVD-ILD rather than IPF.