Occurrence Proportion and Associated Factors of Respiratory Failure During Hospitalization in Chinese Patients With AECOPD: A Systematic Review and Meta‐Analysis
Dandan Liu, Minzuo Chen, Kaosi Wen, Jingmei Pan, Xiangyang Duan, Tao Yang, Shengru TaoABSTRACT
Objective
To systematically evaluate the occurrence proportion and associated factors of respiratory failure during hospitalization in Chinese patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD).
Methods
Eight databases (CNKI, Wanfang, VIP, CBM, PubMed, Web of Science, Embase, and Cochrane Library) were searched from inception to March 1, 2026. Studies reporting multivariable‐adjusted odds ratios (ORs) for factors associated with respiratory failure in AECOPD patients were included. The Newcastle–Ottawa Scale was used for quality assessment. A single‐group rate meta‐analysis with Freeman–Tukey double arcsine transformation was performed to pool the occurrence proportion. Subgroup analyses were conducted by study design, geographical region (Qinling–Huaihe Line), and respiratory failure type. Factors associated with respiratory failure were stratified by definitions and cutoff values; only subgroups with ≥ 3 studies were pooled.
Results
Twenty‐three studies (5 cohort, 18 case–control) with 4212 patients were included. The pooled occurrence proportion of respiratory failure was 41% (95% CI: 0.37–0.46). Meta‐analysis identified nine factors associated with respiratory failure: comorbid cardiac disease (OR = 2.12, 95% CI: 1.68–2.67), advanced age (OR = 1.22, 95% CI: 1.13–1.33), elevated NLR (OR = 1.87, 95% CI: 1.50–2.33), elevated PCT (OR = 2.27, 95% CI: 1.29–4.00), elevated D‐dimer (OR = 1.72, 95% CI: 1.01–2.90), ≥ 2 annual exacerbations (OR = 2.48, 95% CI: 1.71–3.60), higher APACHE II score (OR = 1.29, 95% CI: 1.17–1.43), elevated CRP (OR = 1.13, 95% CI: 1.02–1.24), and decreased FEV1% predicted (OR = 0.81, 95% CI: 0.67–0.99). Model dependence analysis showed that ALB was statistically significant only under the fixed‐effects model but not under the random‐effects model. NE was not statistically significant under either model. Therefore, ALB and NE were not included among the nine factors that were statistically significant in the primary analysis.
Conclusion
The occurrence proportion of respiratory failure in hospitalized Chinese AECOPD patients is high and associated with comorbidities, clinical characteristics, severity scores, and inflammatory/coagulation markers. These findings support risk identification and monitoring, but prospective studies are needed to validate their predictive and interventional value.