Résumé : PURPOSE: To prospectively compare computed tomographic (CT) quantification of pulmonary emphysema in elastase-treated rats with morphometry and to evaluate the information yielded by CT quantification and pulmonary function tests. MATERIALS AND METHODS: This study was approved by the local committee on care and use of animals in research. Thirty-six animals were used. Emphysema was produced by means of one or two tracheal injections of 300 IU of elastase, 8 weeks apart, in seven and 12 rats, respectively. As a control group, 10 rats received an injection of normal saline. The dynamic resistance, dynamic compliance, and static compliance were measured. CT was performed with 1-mm section thickness and 3-mm intervals. Relative areas of lung with attenuation coefficients lower than nine chosen thresholds (from -900 to -980 HU) and eight percentiles (from 1st to 18th percentiles) of the distribution of attenuation coefficients were compared with measurements of alveoli size--that is, mean interwall distance (MIWD) and mean perimeter per field (MP). Correlations between data obtained with thresholds and percentiles and MIWD and MP were investigated by means of Spearman coefficients (r(s)). Values of pulmonary function tests, most appropriate relative area threshold, and percentile were investigated by means of stepwise multiple regressions. RESULTS: For thresholds, relative surface area with attenuation coefficients less than -940 HU (RA(940)) showed the strongest correlations with findings at microscopy (r(s) = 0.676, P < .001 for MIWD; r(s) = -0.720, P < .001 for MP). For percentiles, the 3rd percentile showed the strongest correlations (r(s) = -0.647, P < .001 for MIWD; r(s) = 0.701, P < .001 for MP). Dynamic compliance and RA(940) or 3rd percentile were complementary for predicting microscopic measurements. CONCLUSION: In rats, RA(940) and the 3rd percentile reflect the extent of elastase-induced pulmonary emphysema and are complementary to dynamic compliance to predict microscopic extent.