A novel scoring model for predicting prolonged mechanical ventilation in cardiac surgery patients: development and...

Area of Science:

• Cardiovascular Surgery
• Critical Care Medicine
• Health Services Research

Background:

• Prolonged mechanical ventilation (PMV) is a major complication after cardiac surgery, increasing mortality and healthcare costs.
• Effective prediction of PMV risk is crucial for optimizing patient management and resource allocation.

Purpose of the Study:

• To develop and validate a novel scoring model for predicting the risk of prolonged mechanical ventilation in adult cardiac surgery patients.
• To identify key clinical and surgical factors associated with PMV in this population.

Main Methods:

• Retrospective analysis of 5,206 adult patients undergoing coronary artery bypass grafting, valve, or aortic surgery across 14 hospitals.
• Development of a nomogram using LASSO and multiple logistic regression, with model performance assessed via C-index, calibration plots, and decision curve analysis (DCA).

Main Results:

• The developed nomogram identified 9 significant predictors of PMV, including demographic, preoperative, and surgical variables.
• The model demonstrated strong predictive performance with C-indices of 0.79 (training/internal validation) and 0.75 (external validation).
• Calibration plots and DCA confirmed the model's accuracy and clinical utility for decision-making.

Conclusions:

• A validated scoring model for PMV risk in cardiac surgery patients has been developed, integrating multiple clinical predictors.
• This model enables early identification of high-risk patients, facilitating tailored perioperative strategies.
• The tool has the potential to improve patient outcomes and enhance resource utilization in cardiac surgery settings.