Background
Decannulation for people in a persistent vegetative state (PVS) is challenging and relevant predictors of successful decannulation have yet to be identified.
Objective
This study aimed to explore the predictors of tracheostomy decannulation outcomes in individuals in PVS and to develop a nomogram.
Method
In 2022, 872 people with tracheostomy in PVS were retrospectively enrolled and their data was randomly divided into a training set and a validation set in a 7:3 ratio. Univariate and multivariate regression analyses were performed on the training set to explore the influencing factors for decannulation and nomogram development. Internal validation was performed using 5-fold cross-validation. External validation was performed using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) on both the training and validation sets.
Result
Data from 610 to 262 individuals were used for the training and validation sets, respectively. The multivariate regression analysis found that duration of tracheostomy tube placement≥30 days (Odds Ratio [OR] 0.216, 95 % CI 0.151–0.310), pulmonary infection (OR 0.528, 95 %CI 0.366–0.761), hypoproteinemia (OR 0.669, 95 % CI 0.463–0.967), no passive standing training (OR 0.372, 95 % CI 0.253–0.547), abnormal swallowing reflex (OR 0.276, 95 % CI 0.116–0.656), mechanical ventilation (OR 0.658, 95 % CI 0.461–0.940), intensive care unit (ICU) duration>4 weeks (OR 0.517, 95 % CI 0.332–0.805), duration of endotracheal tube (OR 0.855, 95 % CI 0.803–0.907), older age (OR 0.981, 95 % CI 0.966–0.996) were risk factors for decannulation failure. Conversely, peroral feeding (OR 1.684, 95 % CI 1.178–2.406), passive standing training≥60 min (OR 1.687, 95 % CI 1.072–2.656), private caregiver (OR 1.944, 95 % CI 1.350–2.799) and ICU duration<2 weeks (OR 1.758, 95 % CI 1.173–2.634) were protective factors conducive to successful decannulation. The 5-fold cross-validation revealed a mean area under the curve of 0.744. The ROC curve C-indexes for the training and validation sets were 0.784 and 0.768, respectively, and the model exhibited good stability and accuracy. The DCA revealed a net benefit when the risk threshold was between 0 and 0.4.
Conclusion
The nomogram can help adjust the treatment and reduce decannulation failure.
Registration
Clinical registration is not mandatory for retrospective studies.