IntroductionThis study aimed to evaluate the predictive value of abdominal fat characteristics, as measured by computed tomography (CT), in identifying early recurrence within one year following initial transurethral resection of bladder tumor (TURBT) in patients with non-muscle-invasive bladder cancer (NMIBC). Additionally, the study sought to develop a predictive model that integrates abdominal fat features with clinical factors, to guide individualized diagnostic and treatment plans. Materials and methodsA retrospective analysis of 203 NMIBC patients from two medical centers was conducted using 3D Slicer software to analyze abdominal CT plain images. Spearman correlation, logistic regression, and the Lasso algorithm were employed for data analysis. The predictive efficacy was assessed using the area under the curve (AUC) of receiver operating characteristic (ROC) and decision curve analysis (DCA). Calibration of the predictive model was evaluated using the Hosmer-Lemeshow test. ResultsStatistically significant differences in abdominal fat characteristics were observed between the recurrence and non-recurrence groups. All abdominal fat features showed positive correlations with body mass index (BMI). Among these, bilateral perirenal fat thickness (PrFT) demonstrated superior predictive performance for early recurrence in NMIBC compared to other fat area features. Multivariate logistic regression analysis identified several independent correlates of early recurrence, including tumor number, perfusion chemotherapy within 24 hours post-surgery, left PrFT, right PrFT, visceral fat area (VFA) at the umbilicus level, and VFA at the renal hilum level. The Lasso-based fusion model exhibited an AUC of 0.904, indicating excellent net gain and calibration. This model outperformed existing predictive models, including the EORTC and CUETO risk prediction models. ConclusionAbdominal fat characteristics, particularly bilateral PrFT, show a strong correlation with early recurrence in NMIBC following initial TURBT. The Lasso-based fusion model, which integrates fat thickness characteristics with clinical factors, demonstrates superior predictive efficacy, net clinical gain, and calibration compared to existing risk prediction models. These findings suggest the model's potential for enhancing individualized treatment strategies for NMIBC patients.Micro-AbstractWe accessed the predictive value of abdominal fat characteristics in identifying early recurrence within 1 year post-initial transurethral resection of bladder tumor (TURBT) in non-muscular-invasive bladder cancer (NMIBC).203 patients from two medical centers was conducted to analyze abdominal CT plain images. Abdominal fat characteristics, especially bilateral PrFT, were strongly associated with early recurrence of NIMBC after the initial TURBT surgery.