A nomogram to predict postoperative pulmonary complications after cardiothoracic surgery

Area of Science:

• Cardiovascular Surgery
• Critical Care Medicine
• Pulmonary Complications

Background:

• Critically ill patients undergoing cardiac and major vascular surgery face a significant risk of postoperative pulmonary complications.
• Developing a predictive tool is crucial for proactive management and improved patient outcomes.

Purpose of the Study:

• To develop and validate a novel scoring system predictive of postoperative pulmonary complications.
• To identify key perioperative predictors for risk stratification in this patient population.

Main Methods:

• A retrospective analysis of 17,433 patients undergoing major cardiovascular surgery.
• Elastic net logistic regression and five-fold cross-validation were used to build and refine the prediction model.
• Model performance was assessed using C-statistics and Brier scores on training and independent validation datasets.

Main Results:

• Postoperative pulmonary complications occurred in 9.6% of patients.
• The developed prediction model demonstrated strong predictive efficacy with a C-statistic of 0.87.
• The model maintained satisfactory discrimination and calibration when validated on an independent cohort.

Conclusions:

• A novel prediction nomogram accurately predicts postoperative pulmonary complications following major cardiac and vascular surgery.
• This tool can aid intensivists in implementing proactive and preventative interventions for at-risk patients.