Abstract
Background: Differentiating nontuberculous mycobacterial pulmonary disease (NTM-PD) from pulmonary tuberculosis (PTB) remains challenging due to overlapping clinical features, particularly in resource-limited settings where diagnostic errors are frequent. This retrospective case-control study (January 2023-June 2024) aimed to identify key clinical predictors and develop a diagnostic model to distinguish NTM-PD from PTB. Methods: Patients initially presumed to have PTB (meeting clinical-radiological criteria but lacking bacteriological confirmation at admission) at a tertiary tuberculosis hospital were enrolled. Final diagnoses of NTM-PD (n = 105) and PTB (n = 105) were confirmed by mycobacterial culture identification. Clinical, laboratory, and radiological data were compared using univariate analysis. Variables showing significant differences (p < 0.05) were entered into multivariable logistic regression. Diagnostic performance was evaluated via receiver operating characteristic (ROC) curve analysis. Results: Female sex (odds ratio [OR] = 2.51, 95% confidence interval [CI] 1.12-5.60), hemoptysis (OR = 2.20, 1.05-4.62), bronchiectasis (OR = 5.92, 2.56-13.71), and emphysema/pulmonary bullae (OR = 2.69, 1.16-6.24) emerged as independent predictors of NTM-PD, while systemic symptoms favored PTB (OR = 0.45, 0.20-0.99). The model demonstrated 91.4% specificity and 68.6% sensitivity with an area under the curve [AUC] of 0.871. Conclusions: This high-specificity model helps prioritize NTM-PD confirmation in females with hemoptysis and structural lung changes (computed tomography evidence of bronchiectasis and/or emphysema) while maintaining PTB suspicion when systemic symptoms (fever, night sweats, weight loss) dominate. The approach may reduce misguided antitubercular therapy in resource-limited settings awaiting culture results.