Nomogram prediction model for the risk of ICU-acquired weakness in patients with sepsis
10.3969/j.issn.1006-2483.2026.05.032
- VernacularTitle:脓毒症患者获得性肌无力发生风险的列线图预测模型
- Author:
Zhen TAN
1
;
Ling YANG
1
;
Lin LI
2
;
Tingting XIANG
2
;
Guirong FANG
3
;
Ying ZHOU
3
;
Qunrong SONG
4
Author Information
1. Department of Infectious Diseases, 363 Hospital, Chengdu, Sichuan 610000, China
2. Department of Critical Care Medicine, 363 Hospital, Chengdu, Sichuan 610000, China
3. Department of Nursing, 363 Hospital, Chengdu, Sichuan 610000, China
4. Department of Gastroenterology, 363 Hospital, Chengdu, Sichuan 610000, China
- Publication Type:Journal Article
- Keywords:
Sepsis;
ICU-acquired weakness;
Nomogram;
Bootstrap resampling;
Internal validation
- From:
Journal of Public Health and Preventive Medicine
2026;37(5):150-154
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate the risk factors for ICU-acquired weakness (ICU-AW) in patients with sepsis, and to construct a nomogram prediction model. Methods A total of 360 patients with sepsis admitted from January 2020 to December 2025 were enrolled as study subjects. The risk factors for ICU-AW were identified by logistic regression analysis. A nomogram prediction model was constructed, and the model was validated. Results Multivariate logistic regression analysis indicated that age, APACHE II score, duration of mechanical ventilation, duration of immobilization, duration of sedation, and NLR were risk factors for the occurrence of ICU-AW (P<0.05), while ALB was a protective factor (P<0.05). A nomogram model was constructed based on these seven variables. After validation via Bootstrap resampling, ROC curve showed an AUC of 0.949 (0.929–0.969) and a corrected AUC of 0.942 (0.940–0.948), indicating good discrimination degree of the model. Hosmer-Lemeshow test for calibration curve revealed a chi-square value of 10.188 (P=0.252), suggesting that the predicted value of fitting was generally consistent with the actual value. Decision curve analysis demonstrated that when the threshold was in the range of 0.05–0.9, the model had a good net benefit. Conclusion The nomogram prediction model for sepsis ICU-AW based on seven indicators including age, APACHE II score, NLR, ALB, and the durations of mechanical ventilation, immobilization, and sedation exhibits good predictive efficiency and stability. This nomogram prediction model is helpful for the clinical rapid screening of ICU-AW high-risk groups in patients with sepsis, and meets clinical needs for early warning and precise prevention and control.
