Development and validation of predictive models for meige syndrome patients based on oxidative stress markers

Meige syndrome (MS) is a complex neurological disorder with unclear etiology. Accurate prediction of MS risk is essential for facilitating early diagnosis. This study aimed to develop and validate a nomogram for predicting the risk of MS based on oxidative stress markers.

This retrospective, cross-sectional study included 424 patients with MS and 848 age- and sex-matched healthy controls, with data collected from January 2022 to December 2023. Clinical and laboratory data were extracted from electronic medical records. The MS patients and healthy controls were randomly allocated to the training and validation sets at a 7:3 ratio using random stratified sampling. A nomogram was developed using a multivariate logistic regression model based on data from the training set. Model performance was validated through fivefold cross-validation, receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA).

Univariate and multivariate logistic regression analyses identified albumin, gamma-glutamyl transferase (GGT), total bilirubin (TBIL), and the urea nitrogen-to-creatinine ratio as independent predictors of MS. A nomogram was constructed based on these four variables. The cross-validation confirmed the model's reliability. The model demonstrated high predictive accuracy, with an area under the curve (AUC) of 0.930 for the training set and 0.914 for the validation set. The calibration curve and DCA results indicate that the model has strong consistency and significant potential for clinical application.

This study developed a nomogram based on four risk predictors, GGT, TBIL, albumin, and the urea nitrogen-to-creatinine ratio, to forecast the risk of MS and enhance the accuracy of MS risk prediction.