Abstract:

Objective To investigate the association of early changes of quadriceps muscle thickness （QMT）with prolonged weaning in elderly patients with sepsis undergoing mechanical ventilation（MV）. Methods A prospective observational study was conducted and 329 elderly patients with sepsis undergoing MV admitted to the intensive care unit（ICU）in the Second Affiliated Hospital of Hainan Medical University from January 2021 to August 2022 were enrolled. The patients were assigned to prolonged weaning group（n = 93）and non ⁃ prolonged weaning group（n = 236）. The age，sex，body mass index（BMI），Charlson comorbidity index（CCI）score，sequen⁃ tial organ failure assessment（SOFA）score，acute physiology and chronic health evaluation Ⅱ（APACHEⅡ）scoreat ICU admission，and rapid shallow breathing index，duration of MV，laboratory indicators and organ support at the first spontaneous breathing trial（SBT）were recorded and analyzed. QMT was measured by ultrasound at MV startup and the first SBT，and the change rate of QMT was calculated. Univariate analysis，multivariate logistic regression analysis and receiver operating characteristic（ROC）curve were used to explore the relationship between QMT parameters and prolonged weaning in elderly patients with sepsis undergoing MV. And decision curve analysis was used to determine the clinical practicability of QMT parameters in predicting prolonged weaning in elderly patients with sepsis undergoing MV. Results A total 329 patients were enrolled and the incidence of prolonged weaning was 28.27% . The proportion of age ≥ 75 years and CCI score ≥ 3 points，APACHE Ⅱ score，SOFA score，length of stay in ICU，28⁃day mortality，and serum lactate level in the prolonged weaning group were higher or longer than those in the non⁃prolonged weaning group，while BMI was lower than that in the non⁃prolonged wean⁃ ing group，and the difference between the two groups was statistically significant（P < 0.05）. Multivariate logistic regression analysis showed that age ≥ 75 years（OR = 2.521，95%CI：1.270～5.001，P = 0.008），CCI score ≥ 3 points （OR = 2.139，95%Ⅱ：1.062～4.311，P = 0.033），the decreased value of QMT at the first SBT（OR = 0.383， 95%Ⅱ：0.280 ~ 0.524，P < 0.001），and the elevated change rate of QMT（OR = 1.355，95%Ⅱ：1.254～1.465， P < 0.001）were independent risk factors for prolonged weaning in elderly patients with sepsis undergoing MV. According to ROC curve analysis，the area under the ROC curve of change rate of QMT for predicting prolonged weaning in elderly patients with sepsis undergoing MV was 0.887（95% CI：0.848～0.919），which was signifi⁃ cantly higher than that of QMT at the first SBT（0.758，95%CI：0.758～0.803），with statistically significant differ⁃ ences（Z = 3.507，P < 0.001）. According to decision curve analysis，when the threshold value was 0.10-0.70， the net benefit rate of change rate of QMT for predicting prolonged weaning in elderly patients with sepsis under⁃ going MV was better than that of QMT at the first SBT. Conclusion QMT at the first SBT and change rate of QMT are closely related to the risk of prolonged weaning in elderly patients with sepsis undergoing MV，which could be used to predict the risk of prolonged weaning in elderly patients with sepsis undergoing MV，and the pre⁃ dictive efficacy and net benefit rate of change rate of QMT are higher than those of QMT at the first SBT.

Key words: elderly,  , sepsis,  , mechanical ventilation,  , quadriceps muscle thickness,  , prolonged weaning