A nomogram to predict pulmonary complications after gastrointestinal surgery: a retrospective study

Area of Science:

• Surgical Oncology
• Pulmonary Medicine
• Medical Informatics

Background:

• Postoperative pulmonary complications (PPC) are a significant concern following gastrointestinal surgery.
• Accurate prediction of PPC is crucial for optimizing patient management and improving surgical outcomes.
• Existing predictive models may lack sufficient accuracy or generalizability for diverse patient populations.

Purpose of the Study:

• To develop and validate a nomogram for predicting the risk of postoperative pulmonary complications (PPC) in patients undergoing gastrointestinal surgery.
• To identify independent risk factors associated with PPC development after gastrointestinal surgery.

Main Methods:

• Retrospective analysis of 563 patients who underwent gastrointestinal surgery.
• Development of a nomogram using Least Absolute Shrinkage and Selection Operator (LASSO) and logistic regression on a training cohort.
• Validation of the nomogram using receiver operating characteristic (ROC) curves and decision curve analysis (DCA) in both training and validation cohorts.

Main Results:

• The incidence of PPC was 17.6%.
• Key predictors identified include age, hypertension, respiratory disease history, preoperative albumin, intraoperative blood loss, ICU time, PaO2, and tracheal intubation time.
• The nomogram demonstrated high predictive accuracy with C-indices of 0.857 (training) and 0.936 (validation), and strong performance on ROC curves.

Conclusions:

• The developed nomogram is a valuable tool for predicting PPC after gastrointestinal surgery.
• It provides a scientific reference to aid clinicians in risk stratification and perioperative decision-making.
• Further prospective studies can further validate its clinical utility.