Radiographic features on baseline and follow-up computed tomography associated with treatment failure in patients with Mycobacterium avium complex lung disease.

OBJECTIVES

Nontuberculous mycobacteria, particularly the Mycobacterium avium complex (MAC), have emerged as significant pathogens in chronic pulmonary infections. This study aimed to investigate the association between CT features at baseline and their interval changes during treatment with clinical outcomes in patients with MAC lung disease.

MATERIALS AND

METHODS

We retrospectively analyzed 149 MAC-LD patients treated between 2006 and 2023 at a single tertiary center with two affiliated sites. CT scores for bronchiectasis, bronchiolitis, consolidation, cavities, and nodules were assessed at baseline and after treatment. Baseline values and interval changes were calculated. Logistic regression identified predictors of sputum culture conversion, and ROC analysis evaluated model performance.

RESULTS

Treatment failure (TF) occurred in 63/149 patients (42.2%). The TF group had higher baseline cavitation score and greater interval increases in cavitation, bronchiectasis extension, bronchiolitis, and consolidation, compared to the treatment success group. Bronchiectasis severity worsened in both groups. Baseline cavitation diameter (p = 0.002) and bronchiectasis extension (p = 0.008) were independent predictors of failure; bronchiectasis extension reached significance only in multivariable analysis. Interval changes in cavitation diameter (p = 0.028), bronchiectasis extension (p = 0.013), and consolidation extension (p = 0.007) were also associated with poor outcomes. Receiver operating characteristic analysis showed AUC values of 0.763 for interval change in consolidation extension, 0.748 for interval change in bronchiectasis extension, 0.643 for interval change in cavitation diameter, and 0.663 for baseline cavitation diameter.

CONCLUSIONS

Baseline and interval CT features might serve as early indicators of poor treatment response in MAC-LD, assisting clinical decision-making.