Effects of High-Resolution CT Changes on Prognosis Predictability in Acute Respiratory Distress Syndrome with Diffuse Alveolar Damage

Abstract

Diffuse alveolar damage (DAD) is the pathological hallmark of acute respiratory distress syndrome (ARDS). DAD is independently correlated with higher mortality compared with the absence of DAD. Traction bronchiectasis in areas of ground-glass opacity or consolidation is associated with the late fibroproliferative or fibrotic phase of DAD. This study examined whether the 60-day mortality related to DAD could be predicted using high-resolution computed tomography (HRCT) findings and HRCT scores. A total of 34 patients with DAD who received HRCT within 7 days of ARDS diagnosis were enrolled; they were divided into a 60-day survival group and a nonsurvival group, with 17 patients in each group. Univariate and multivariate binary regression analyses and the receiver operating characteristic curve revealed that only the total percentage of the area with traction bronchiectasis or bronchiolectasis was an independent predictor of 60-day mortality (odds ratio, 1.067; 95% confidence interval (CI), 1.011–1.126) and had favorable predictive performance (area under the curve (AUC): 0.784; 95% CI, 0.621–0.946; cutoff, 21.7). Physiological variables, including age, days from ARDS to HRCT, the sequential organ failure assessment (SOFA) score, the PaO2/fraction of inspired oxygen (FiO2) ratio, dynamic driving pressure, and dynamic mechanical power, were not discriminative between 60-day survival and nonsurvival. In conclusion, the extent of fibroproliferation on HRCT in early ARDS, presented as the total percentage of area with bronchiectasis or bronchiolectasis, is an independent positive predictor with a favorable predictive ability for the 60-day mortality of DAD.