Abstract:

Objective

To develop a nomogram prediction model for weaning failure in mechanically ventilated critical ill patients in plateau areas.

Methods

A total of 255 patients requiring invasive mechanical ventilation admitted to the ICU of the People's Hospital of Xizang Autonomous Region between January 1 and December 31，2023，were enrolled.Based on weaning outcomes，patients were divided into a successful weaning group （184 cases） and a failed weaning group （71 cases）.Demographic characteristics，comorbidities，vital signs，and organ function parameters were collected and compared between the two groups.Statistically significant variables in univariate analysis were incorporated into multivariate logistic regression to identify independent risk factors for weaning failure.And a nomogram prediction model for weaning failure was established.

Results

Compared with the successful weaning group，the failed weaning group had significantly longer ICU stays，higher hospitalization costs，and increased 28-day mortality （all P＜0.001）.The multivariate regression analysis identified three independent risk factors for weaning failure：tidal volume （Vt） before weaning （P＜0.001，95%CI：0.979-0.992），respiratory rate （RR） before weaning（P=0.006，95%CI：1.034-1.218），and fluid balance in 24 hours before weaning （P=0.033，95%CI：1.000-1.001）.The area under the ROC curve of the prediction model was 0.883 （95%CI：0.841-0.925）.

Conclusions

Vt，RR，and 24-hour fluid balance levels before weaning significantly influence the incidence of weaning failure in mechanically ventilated critically ill patients in plateau areas.The nomogram prediction model developed for weaning failure in this study based on these three parameters，demonstrates a good predictive value.

Key words: Plateau area, Weaning failure, Risk factors, Prediction model