Am J Transl Res. 2026 Oct 15;18(10):9542-9552. doi: 10.62347/EWET5726. eCollection 2026.
ABSTRACT
OBJECTIVES: To develop and temporally validate an admission-based neutrophil-to-lymphocyte ratio (NLR)-albumin index for estimating invasive mechanical ventilation risk in patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and type II respiratory failure.
METHODS: We analyzed a development cohort of 155 patients (January 2023-June 2025) and a temporal validation cohort of 197 patients (July 2025-March 2026). A predicted probability (PRE) was derived from admission NLR and albumin and evaluated using logistic regression, variance inflation factors, 1,000 bootstrap resamples, receiver operating characteristic analysis, calibration, decision curve analysis, and Bonferroni correction. PRE was also combined with the Acute Physiology and Chronic Health Evaluation II (APACHE II) score.
RESULTS: Invasive ventilation was required in 43/155 (27.7%) development-cohort patients and 53/197 (26.9%) validation-cohort patients. NLR, albumin, and APACHE II were independently associated with invasive ventilation; each 0.10 increase in PRE was associated with an odds ratio of 1.620 (95% confidence interval, 1.311-2.002). Development-cohort areas under the curve were 0.794 for PRE and 0.819 for PRE plus APACHE II, with bootstrap-corrected values of 0.788 and 0.808; corresponding validation values were 0.845 and 0.881. Validation-cohort invasive ventilation rates increased from 3.3% to 24.7% and 60.4% across PRE risk groups (P<0.001). Exploratory 24-hour areas under the curve for PRE and ΔPRE were 0.877 and 0.890 in development and 0.919 and 0.932 in validation.
CONCLUSIONS: Admission PRE, particularly when combined with APACHE II, may facilitate early risk stratification in hospitalized patients with AECOPD and type II respiratory failure; multicenter prospective validation is required.
PMID:43068060 | PMC:PMC13788502 | DOI:10.62347/EWET5726