A CT-based score for predicting histopathological usual interstitial pneumonia features and predicting disease progression in smokers with fibrotic idiopathic interstitial pneumonia
摘要
Histopathological usual interstitial pneumonia (UIP) features are associated with progressive fibrosis but may be overlooked in smokers, where emphysema and smoking-related fibrosis complicate CT-based UIP classification. This study aimed to develop and validate a CT-based score to detect histopathological UIP features and to evaluate its association with disease progression in smokers with fibrotic idiopathic interstitial pneumonia (IIP).
Materials and methodsThis retrospective multicentre study included two biopsy-proven cohorts of smokers with fibrotic IIP: derivation (n = 242; nationwide registry) and validation (n = 126; single-centre database, 2008–2013). Three predefined HRCT findings—upper-lobe irregular lines (U), irregularity of pleural surface (I), and pleural-based basal reticulation (P)—were scored (0/1) and summed as the U-I-P score (0–3). Associations with histopathological UIP features and outcomes were assessed using logistic regression, receiver operating characteristic analysis, Cox proportional hazards models, and bootstrap internal validation.
ResultsCohorts were similar (median age 65 vs. 64 years; 88% vs. 87% male). Histopathological UIP features were present in 208/242 (86%) and 109/126 (87%) patients. The U-I-P score independently predicted histopathological UIP features in both cohorts (area under the curve, 0.83 and 0.89; both p < 0.001). Bootstrap validation demonstrated minimal optimism, supporting model stability. The U-I-P score independently predicted disease progression at 1 and 2 years in the validation cohort (p = 0.002 and p = 0.001, respectively).
ConclusionThe U-I-P score detects histopathological UIP features and predicts disease progression in smokers with fibrotic IIP, providing a practical imaging biomarker, even in cases with atypical CT patterns.
Key Points