Construction and validation of a prediction model for cystitis glandularis using serological markers combined with histological indexes to predict recurrence risk.

Background: Cystitis glandularis (CG) is a chronic inflammatory condition of the bladder characterized by a high recurrence rate, imposing a substantial burden on patients. The mechanisms underlying recurrence remain unclear. Objectives: This study aims to identify markers associated with CG recurrence and develop a predictive model for recurrence risk. Design: Retrospective cohort study of patients with confirmed recurrence based on outpatient visits or readmissions were included in this study, which was subsequently divided into training and test set. Methods: Patients diagnosed with CG from four hospitals between 2013 and 2023 were retrospectively included and followed for one year. Recurrence was defined as the appearance of new nonneoplastic lesions on cystoscopy after complete resection of the primary disease. A total of 161 patients were divided into a training set ( Results: Among 161 patients followed for 12 months, the recurrence rate was 49.6% ( Conclusion: Serological indicators, specifically eosinophil and basophil counts, are closely linked to CG recurrence. A risk score model based on these markers was developed, providing effective prediction of recurrence in clinical practice.