Development of a Predictive Model for Postoperative Nausea and Vomiting in the PACU After Painless Endoscopy:...

To identify independent risk factors for PONV following painless esophagogastroduodenoscopy (EGD) and to develop and internally validate a nurse-facing predictive model to support PACU clinical decision-making.

Prospective, single-center cohort study.

Demographic, clinical, and procedural data were collected from electronic medical records. PONV during PACU stay was assessed serially using a nausea numeric rating scale (NRS; ≥1 indicates nausea) and by documenting any emetic episode. Multivariable logistic regression identified independent risk factors; model performance was evaluated by the area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis (DCA). Internal validation used 1000 bootstrap resamples.

A total of 1070 adult patients undergoing elective, painless esophagogastroduodenoscopy (EGD) in a tertiary hospital in China, under total intravenous anesthesia with propofol and sufentanil participated in the study. The overall incidence of PONV was 16.8%. Female sex (OR = 3.10), older age (OR = 1.07 per year), higher BMI (OR = 1.14 per unit), and longer procedure duration (OR = 1.12 per minute) were independent predictors of increased risk, whereas smoking history was protective (OR = 0.35). The model demonstrated good discrimination (apparent AUC = 0.81; optimism-corrected AUC = 0.80; 95% CI, 0.78-0.84), good calibration (slope = 0.98), and net clinical benefit across threshold probabilities of 10-40% on decision curve analysis.

This nurse-facing prediction model demonstrated good discrimination, calibration, and clinical usefulness and can support risk stratification and focused monitoring in preoperative nursing assessment and PACU workflows. Further external validation and implementation studies are warranted before routine clinical adoption.