Type 2 Diabetes and HbA1c Predict All-Cause Post-Metabolic and Bariatric Surgery Hospital Readmission

Machine learning analysis of lab tests to predict bariatric readmissions

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.

Obesity Metabolic Phenotypes and Unplanned Readmission Risk in Diabetic Kidney Disease: An Observational Study from the Nationwide Readmission Database

BACKGROUND AND AIM

Obesity is a potentially modifiable factor for reducing readmissions, with heterogeneity that varies according to the metabolic status. Our objective was to examine the independent or mutual relationship between obesity and metabolic abnormalities and diabetic kidney disease (DKD)-related hospitalizations.

METHODS

493,570 subjects with DKD were enrolled in the 2018 Nationwide Readmission Database (NRD, United States). The at-risk population was reclassified into refined obesity subtypes based on the body mass index (BMI) classification of metabolic abnormalities (hypertension and/or dyslipidemia) to investigate the 180 d readmission risk and hospitalization costs related to DKD.

RESULTS

The overall readmission rate was 34.1%. Patients with metabolic abnormalities, regardless of obesity, had a significantly higher risk of readmission compared to non-obese counterparts (adjusted HR, 1.11 [95% CI, 1.07-1.14]; 1.12 [95% CI, 1.08-1.15]). Hypertension appeared to be the only metabolic factor associated with readmission among individuals with DKD. Obesity without metabolic abnormalities was independently associated with readmission (adjusted HR,1.08 [1.01,1.14]), especially among males and those >65 years (adjusted HR,1.10 [1.01-1.21]; 1.20 [1.10-1.31]). Women or those ≤65 years with metabolic abnormalities (all p <0.050) had elevated readmission rates, regardless of obesity; however, no such trend was observed in obese subjects without metabolic abnormalities (adjusted HR, 1.06 [0.98,1.16]). Additionally, obesity and metabolic abnormalities were associated with elevated hospitalization costs (all p <0.0001).

CONCLUSIONS

Increased BMI and hypertension are positively associated with readmissions and related costs among patients with DKD, which should be considered in future studies.

Glycemic control and macular edema in patients undergoing cataract surgery

AIMS

Cataract, the most common cause of blindness, has higher prevalence among patients with diabetes mellitus. About 20% of cataract surgeries are performed on patients with diabetes. One of the complications of cataract surgery is pseudophakic cystoid macular edema (CME). This study examined whether patients' glycemic control (as indicated by HbA1c level before cataract surgery) is associated with CME incidence within one year post-surgery.

METHODS

We conducted a retrospective cohort study of 1285 diabetes patients over age 18 who underwent cataract surgery between January 2015 and January 2020. Data were obtained from medical records reporting glycated hemoglobin (HbA1c) level prior to surgery and post-operative CME with intraocular anti-vascular endothelial growth factor injections.

RESULTS

The patients with CME complications were younger, with longer duration diabetes, and higher percentages of type 1 diabetes and diabetic retinopathy. The main variables influencing risk of post-operative CME were found to be diabetic retinopathy and HbA1c level. Multivariate analysis revealed that HbA1c is an independent risk for post-operative CME with a relative risk of 2.01 when HBa1c is above 7 c (95% CI, 1.10-3.67).

CONCLUSION

The study demonstrates that pre-cataract surgery diabetes control, measured by HbA1c level, is an independent risk factor for developing post-surgery CME.