Abstract:

Objective: To investigate the discriminative value of quantitative computed tomography (QCT) parameters combined with the peripheral blood neutrophil-to-lymphocyte ratio (NLR) for pulmonary function grading in patients with chronic obstructive pulmonary disease (COPD). Methods: A total of 535 patients with COPD admitted to Guoyang County People’s Hospital between January 2022 and April 2024 were retrospectively enrolled. General clinical data, laboratory parameters obtained within 24 hours of admission (including white blood cell count, neutrophil count, lymphocyte count, and the calculated NLR), and QCT parameters were collected, including total lung volume, emphysema volume, and the percentage of low attenuation area (LAA%). Pulmonary function grading (GOLD 1-2 vs. GOLD 3-4) was defined as the dependent variable. Univariable analysis and multivariable logistic regression modeling were performed. Model performance was evaluated using receiver operating characteristic curves (ROC), calibration curves, and decision curve analysis (DCA) in terms of discrimination, calibration, and clinical net benefit. Results: Univariable ROC analysis showed that emphysema volume had the highest discriminative performance for pulmonary function grading (AUC=0.740), followed by LAA% (AUC=0.717), total lung volume (AUC=0.679), NLR (AUC=0.637), and albumin (AUC=0.565). The final combined model (total lung volume+LAA%+NLR+albumin) achieved an AUC of 0.765, which was higher than that of any single indicator. Multivariable logistic regression analysis demonstrated that total lung volume, LAA%, and NLR were independently associated with an increased risk of severe grading, whereas albumin was associated with a decreased risk (all P<0.05). Specifically, the adjusted odds ratios (aORs) were 1.053 per 100 ml increase in total lung volume (95%CI: 1.033-1.072, P<0.001), 1.092 per 1% increase in LAA% (95%CI: 1.062-1.123, P<0.001), 1.060 per 1-unit increase in NLR (95%CI: 1.014-1.109, P=0.011), and 0.935 per 1 g/L increase in albumin (95%CI: 0.882-0.990, P=0.022). Bootstrap internal validation indicated good model calibration (mean absolute error≈0.019), and decision curve analysis showed that the combined model provided higher clinical net benefits within the threshold probability range of 0.05 to 0.30. Conclusion: QCT parameters combined with NLR demonstrate moderate discriminative ability in distinguishing GOLD 3-4 from GOLD 1-2 COPD patients, with good calibration and potential clinical applicability after internal validation, providing a reference for risk stratification and clinical decision-making in COPD patients.

Key words: Pulmonary disease, chronic obstructive, Tomography, X-ray computed, Neutrophil, Lymphocyte, Pulmonary function