DOI: 10.4103/ijmy.ijmy_90_26 ISSN: 2212-5531

Discriminatory Performance of Novel Inflammatory–Nutritional Indices for Identifying Chronic Obstructive Pulmonary Disease Comorbidity in Patients with Nontuberculous Mycobacterial Pulmonary Disease: A Retrospective Cross-sectional Study

Zexiao Ye, Fei He
Show PDF Cite

Background:

The coexistence of nontuberculous mycobacterial pulmonary disease (NTM-PD) and chronic obstructive pulmonary disease (COPD) represents a severe clinical phenotype. This study aimed to investigate the factors associated with COPD comorbidity in patients with NTM-PD and evaluate the discriminatory performance of novel composite inflammatory–nutritional indices, specifically the C-reactive protein-to-albumin ratio (CAR) and the neutrophil-to-lymphocyte ratio-to-albumin ratio (NAR).

Methods:

In this single-center, retrospective, cross-sectional study, we reviewed the clinical records of 153 patients with NTM-PD diagnosed between January 2019 and December 2024. Eligible patients were categorized into the NTM-PD alone group and the NTM-PD with comorbid COPD group based on postbronchodilator pulmonary function testing. A clinically driven multivariable logistic regression model was used to identify factors associated with COPD comorbidity after adjustment for potential confounders, including smoking index, comorbidity score, and NTM species. Discriminatory performance was assessed using receiver operating characteristic analysis, with internal validation by 1000-resample bootstrapping. Model calibration was evaluated using the Brier score and calibration plot, and clinical utility was explored using decision curve analysis.

Results:

Of the 153 included patients, 75 (49.0%) had NTM-PD alone, and 78 (51.0%) had comorbid COPD. In the multivariable model, male sex (odds ratio [OR] = 4.93, 95% confidence interval [CI]: 2.01–12.09, P < 0.001), the presence of computed tomography (CT) cavities (OR = 3.82, 95% CI: 1.57–9.30, P = 0.003), and a higher NAR level (per 0.1-unit increase, OR = 1.85, 95% CI: 1.16–2.93, P = 0.009) were independently associated with COPD comorbidity. CAR and NAR showed similar discriminatory performance, with area under the curves (AUCs) of 0.768 and 0.767, respectively. The multivariable model showed good discrimination after internal validation by 1000-resample bootstrapping (bootstrapped AUC = 0.858, 95% CI: 0.796–0.910) and acceptable calibration (Brier score = 0.164).

Conclusions:

Male sex, the presence of CT cavities, and elevated NAR were independently associated with COPD comorbidity in patients with NTM-PD. In addition, CAR and NAR showed acceptable discriminatory performance as potential adjunctive screening markers for identifying patients who may warrant further evaluation for concurrent COPD. However, external validation is still required before broader clinical application.

More from our Archive