RESUMO
OBJECTIVE: To determine the predictive value of esophageal dilatation as observed in high-resolution computed tomography (HRCT) for the diagnosis of systemic sclerosis (SSc) in patients with interstitial lung disease (ILD). METHODS: Our sample consisted of patients diagnosed with SSc and proven interstitial lung involvement with available HRCT exams (n = 20). Individuals with other forms of rheumatic ILD were included as a control group (n = 20). Two blinded radiologists independently reviewed the images for the presence of esophageal dilatation, measured at 3 different levels. Interobserver agreement was tested with Lin's concordance correlation coefficient (CCC). Independent t-test was used to compare maximum esophageal diameters between groups. Friedman's test was used to evaluate differences between the 3-level measurements. Receiver operating characteristic analysis was performed. RESULTS: There was a substantial correlation between both readers (CCC = 0.9802-0.9919). Esophageal dilatation was significantly associated with SSc (P = .0012). The optimal calculated cut-off value to differentiate SSc from other ILDs was 18.5 mm (sensitivity and specificity of 70 and 90%, respectively; area under the curve 0.819), measured 1 cm above the diaphragmatic hiatus. CONCLUSION: HRCT may have a discriminative role in the presence of both ILD and esophageal dilatation for the diagnosis of SSc. Our results suggest that a cut-off value for the esophageal diameter of 18.5 mm might propose the diagnosis of SSc with reasonable confidence.
