The purpose of this study was to develop a machine learning model for predicting 30-day readmission after bariatric surgery based on laboratory tests. Data were collected from patients who underwent bariatric surgery between 2018 and 2023. Laboratory test indicators from the preoperative stage, one day postoperatively, and three days postoperatively were analyzed. Least absolute shrinkage and selection operator regression was used to select the most relevant features. Models constructed included support vector machine (SVM), generalized linear model, multi-layer perceptron, random forest, and extreme gradient boosting. Model performance was evaluated and compared using the area under the receiver operating characteristic curve (AUROC). A total of 1262 patients were included, of which 7.69% of cases were readmitted. The SVM model achieved the highest AUROC (0.784; 95% CI 0.696–0.872), outperforming other models. This suggests that machine learning models based on laboratory test data can effectively identify patients at high risk of readmission after bariatric surgery.

Background: Metabolic bariatric surgery (MBS) is associated with high risk of postoperative nausea and vomiting (PONV). We aimed to investigate the impact of aprepitant-based triple prophylaxis on PONV after MBS.Methods: We reviewed a retrospective cohort of patients who underwent primary MBS between December 28, 2023 and May 31, 2024. The eligible patients were divided into two groups based on whether receiving additional oral single 125 mg dose of aprepitant preoperatively to the dual prophylaxis (ondansetron 8 mg and dexamethasone 10 mg). Multivariable and propensity score-adjusted analyses were performed to compare the composite PONV endpoints between the groups.Results: A total of 207 patients were included in the study. Of these, 129 patients received dual prophylaxis, while the remaining 78 patients received additional single 125 mg dose of aprepitant. Similar to multivariable logistic regression analysis, propensity-adjusted logistic regression analysis revealed that the aprepitant-based triple prophylaxis group had a significantly higher complete response rate (82.1% vs. 24.0%, adjusted OR 10.312, 95% CI 4.186–25.399, p < 0.001), a lower incidence of PONV (59.0% vs. 85.3%, adjusted OR 0.287, 95% CI 0.125–0.663, p = 0.004), and required fewer rescue antiemetics (7.7% vs. 37.2%, adjusted OR 0.155, 95% CI 0.052–0.457, p < 0.001) compared to the dual prophylaxis group. Propensity score-adjusted analysis demonstrated that the addition of aprepitant to dual prophylaxis significantly reduced the incidence of PONV, vomiting frequency, and both the Nausea VAS and Nausea Subscale scores (all p < 0.05).Conclusion: Our findings indicate that the addition of a single preoperative dose of aprepitant to a dual antiemetic prophylaxis of dexamethasone and ondansetron might be associated with a further improve outcomes related to composite PONV endpoints in patients undergoing metabolic bariatric surgery.