Related Experiment Videos
Admission Systemic Immune-Inflammation Index and 90-Day AECOPD-Related Unplanned Readmission: A Single-Center
Qiong Chen1, Sufang Xu1, Jin Wang2
1Department of Medical Laboratory Center, Anhui No. 2 Provincial People's Hospital, Hefei, Anhui, 230041, People's Republic of China.
Background:
The systemic immune-inflammation index (SII) is derived from routine blood counts, but its incremental value for disease-specific early readmission after acute exacerbation of chronic obstructive pulmonary disease (AECOPD) remains uncertain. We evaluated the association between admission SII and 90-day AECOPD-related unplanned readmission and its value beyond available clinical variables.
Methods:
This single-center retrospective development cohort included 207 unique patients hospitalized for AECOPD from January 2024 to January 2025; only the first eligible hospitalization per patient was analyzed. Readmissions were ascertained from electronic records and structured telephone follow-up completed on June 30, 2025. SII was calculated from the first blood count obtained within 24 hours of arrival. Logistic regression, receiver operating characteristic analysis, 1000-resample bootstrap internal validation, calibration assessment and decision curve analysis were performed. Bootstrap validation assessed optimism in discrimination, calibration and net benefit.
Results:
Seventy-eight patients (37.7%) experienced AECOPD-related unplanned readmission within 90 days. Each 500-unit increase in SII was associated with higher adjusted odds of readmission (odds ratio 1.329, 95% confidence interval 1.153-1.530; p < 0.001). Apparent AUC was 0.809 (95% confidence interval 0.742-0.869) for the integrated model versus 0.752 (0.678-0.821) for the limited clinical model. The absolute increase was 0.057 (bootstrap 95% confidence interval 0.014-0.101), and adding SII improved model fit (likelihood-ratio p < 0.001). Optimism-corrected AUC, calibration intercept and slope were 0.785, -0.003 and 0.874. Otherwise identical NLR and PLR models had corrected AUCs of 0.775 and 0.765; pairwise differences were inconclusive.
Conclusion:
Higher admission SII was associated with 90-day AECOPD-related unplanned readmission and improved discrimination beyond a limited clinical model. These exploratory single-center findings require external validation, recalibration and prospective impact assessment before clinical implementation.