Unveiling the prognostic potential: Naples score for predicting in-hospital mortality in intensive care unit patients with acute exacerbation of COPD

To estimate in-hospital mortality and identify its independent predictors among patients with acute exacerbation of chronic obstructive pulmonary disease (AE-COPD) admitted to the general intensive care unit (ICU), focusing on the Naples Prognostic Score (NPS). This retrospective analysis was conducted in the general ICU of a tertiary university hospital between 2023 and 2024, including patients with AE-COPD. Demographic data, comorbidities, pulmonary function test results, arterial blood gas levels, and laboratory markers (including albumin, total cholesterol, neutrophil-to-lymphocyte ratio [NLR], lymphocyte-to-monocyte ratio [LMR], and lactate levels) were extracted. NLR and LMR were evaluated individually and as components of the Naples Prognostic Score (NPS) to comprehensively assess the combined impact of systemic inflammation and nutritional status on patient outcomes. The NPS was calculated as the sum of points (0–4), with 1 point assigned for each of the following: albumin level <4.0 g/dL, total cholesterol level ≤180 mg/dL, NLR≥2.96, and LMR≤4.44. We confirmed that albumin values were converted from g/L to g/dL for the calculation. Associations with in-hospital mortality were examined using univariate analyses and multivariate regression, and NPS discrimination was assessed using receiver operating characteristic (ROC) analysis (area under the curve [AUC], sensitivity, and specificity). We analyzed data of 153 patients with AE-COPD who were admitted to the ICU and had an in-hospital mortality rate of 8.05%. Compared with survivors, non-survivors were older, had higher lactate levels, more frequent invasive mechanical ventilation, and higher NPS scores. In multivariable models, NPS (odds ratio [OR], 4.851; 95% confidence interval [CI], 1.934–12.164; p=0.001), age (OR, 1.062; 95% CI, 1.017–1.109; p=0.007), and lactate levels (OR, 1.553; 95% CI, 1.117–2.103; p=0.008) independently predicted in-hospital mortality. NPS showed good discrimination (area under the curve [AUC] = 0.822; sensitivity = 84.1%; specificity = 77.1%). In patients with AE-COPD, higher NPS, older age, and elevated lactate levels independently predicted in-hospital mortality. Because NPS components are routinely available, incorporating this composite nutrition-inflammatory index alongside basic clinical variables may facilitate early risk stratification and guide monitoring and support in the ICU setting.