A nomogram to predict postoperative nausea and vomiting in the ward following laparoscopic bariatric surgery.

Abstract

Postoperative nausea and vomiting (PONV) is a common and distressing complication of laparoscopic bariatric surgery (LBS). However, there is a lack of effective integrated prediction models for preventing and treating PONV in patients after LBS. Based on a randomized controlled trial conducted between November 1, 2021, and May 13, 2022, we included 334 participants who underwent LBS according to the inclusion criteria. The database was divided randomly into training and validation cohorts in a ratio of 7:3. Least absolute shrinkage and selection operator plus multivariable logistic regression were used to identify independent predictors and construct a nomogram. The performance of the nomogram was assessed and validated by the area under the receiver operating characteristic curve (AUC), the concordance index (C-index), calibration plots, and a decision curve analysis (DCA). We also explored specific risk factors for PONV in patients with diabetes. The subjects were divided randomly into training (n = 234) and validation (n = 100) cohorts. Age, history of diabetes, type of surgery, and sugammadex use were incorporated to construct a nomogram prediction model. In the training cohort, the AUC and the optimism-corrected C-index were 0.850 [95% confidence interval (CI) 0.801-0.899] and 0.848, while in the validation cohort they were 0.847 (95%CI 0.768-0.925) and 0.844, respectively. The calibration plots showed good agreement between the predicted and actual observations. The DCA results demonstrated that the nomogram was clinically useful. The type of surgery, sugammadex use, and insulin level at 120min were predictors of PONV in patients with diabetes with an AUC of 0.802 (95%CI 0.705-0.898). We developed and validated a prediction model for PONV in patients after LBS. A risk factor analysis of PONV in patients with diabetes provides clinicians with a more precise prophylactic protocol.