Diagnostic value of high-resolution CT in the evaluation of chronic infiltrative lung disease in children

OBJECTIVE. The purpose of this study was to evaluate the accuracy of CT in the diagnosis of chronic infiltrative lung disease in children. MATERIALS AND METHODS. Fifty-nine patients selected over a 14-year period ( 29 girls, 30 boys; mean age, 6 +/- 4.9 years; range, 2 months-18 years) had nine disorders. CT scans were evaluated independently by two experienced chest radiologists, who were unaware of pathologic or clinical data. The radiologists recorded specific CT findings of infiltrative lung disease and were asked to give the most likely diagnosis and up to two differential diagnoses. Descriptive statistic analysis was followed by logistic regression analysis for each elementary lesion on the grid of abnormalities. RESULTS. A correct first-choice diagnosis was made in 38% of CT observations. The correct diagnosis was among the three main choices in 59% of CT observations. Pulmonary alveolar proteinosis (n = 18) was most frequently correctly diagnosed; it was the first-choice diagnosis 47% of the time and among the three main choices 72% of the time. The correct first-choice diagnosis of idiopathic pulmonary fibrosis ( n = 16) was made 43% of the time; of hypersensitivity pneumonitis ( n = 4), 37% of the time; of sarcoidosis ( n = 7), 28% of the time; of idiopathic pulmonary hemosiderosis ( n = 6), 16% of the time; and of connective tissue diseases ( n = 5), 10% of the time. All single cases of pulmonary fibrosis with calcification, lymphangiectasia, and Langerhans' cell histiocytosis were correctly diagnosed. CONCLUSION. Our results showed there are limitations to diagnosing chronic infiltrative lung disease in children on the basis of CT data alone. We suppose that these differences are explained by the technical difficulties of high-resolution CT in children, the insufficient number of cases of and data on high-resolution CT of children, and the heterogeneity of lesions of a given cause.