RESUMO
BACKGROUND: Although quantitative computed tomography (CT) has been used to analyze the lungs of patients with confirmed diagnoses of acute respiratory distress syndrome (ARDS), there are few reports to show the diagnosis during the early stage of ARDS. Using a canine model and quantitative CT, we aimed to develop an oleic acid (OA) induced ARDS regarding the early stage of ARDS that could improve in the early diagnosis of ARDS. METHODS: Fourteen healthy beagle dogs underwent CT. Their lung tissue was manually partitioned into four compartments, i.e., non-aerated, poorly aerated, normally aerated, and hyper-aerated lung compartments. The mean CT attenuation value Hounsfield unit (HU), tissue mass (g), residual volume (ml), and percentage of lung area were automatically determined for each lung compartment and compared between groups by receiver operating characteristic curve (ROC) analyses using area under curve (AUC). The optimized cut-off point for each parameter was determined by Youden's index. RESULTS: Regarding lung compartments during the expiratory phase, the percentage of non-aerated lung area in the ARDS group was higher vs. controls at all time points (T1 to T6). CT attenuation values for the ARDS group increased with time during both respiratory phases compared with controls. During both respiratory phases, tissue mass within the ARDS group significantly increased compared with controls at T3-T6. CONCLUSIONS: Quantitative CT analysis can detect ARDS at an early stage with high sensitivity and specificity, providing a minimum of assistance in the early diagnosis of ARDS.