Abstract
BACKGROUND: Tuberculosis infection (TBI) is a significant cause of bronchiectasis (BE). Identifying risk factors for radiological BE (RBE) could enhance the early detection of high-risk individuals following TB infection. This study aimed to develop and validate a novel Inflammation-Nutrition Risk Score (INRS) and a corresponding nomogram model to predict the risk of RBE after TBI. PATIENTS AND METHODS: We enrolled 2,210 post-TBI patients from two medical centres. Data from 1,825 patients at Wuhan Jinyintan Hospital were used to develop the INRS and the RBE nomogram. An independent cohort of 385 patients from Wuhan Union Hospital served as an external validation set. RESULTS: The INRS was derived from four parameters: PNI, HALP score, Lg(SII) and CAR. Multivariate analysis identified the following independent risk factors for RBE: age ≥60 years (OR = 1.19, p = 0.030), current smoking (OR = 1.71, p = 0.009), COPD (OR = 3.13, p < 0.001), RDW-CV ≥12.8% (OR = 1.09, p = 0.005), ALB <35.5 g/L (OR = 1.04, p = 0.003) and INRS ≥1.86 (OR = 5.04, p < 0.001). The RBE nomogram model demonstrated strong discriminatory power, accuracy and clinical utility across the development, internal validation and external validation cohorts. CONCLUSION: In post-TBI patients, the INRS represents a novel predictive biomarker for RBE. The INRS-based nomogram is a clinically applicable and efficient tool for risk stratification and guiding follow-up management to prevent RBE progression.