The impact of preoperative CT-based 3D visualization model on ureteric access sheath placement failure without pre-stenting

BACKGROUND: Placement failure of the flexible and navigable suction ureteral access sheath (FANS) with negative pressure suction may occur during retrograde intrarenal surgery (RIRS) due to ureteral stricture, hemorrhage, injury and other factors. This study aimed to investigate the predictive value of parameters from a 3D visualization model based on preoperative computed tomography (CT) for the failure of FANS placement without pre-stenting. METHODS: We retrospectively analyzed the clinical data of 113 patients who underwent RIRS at Chengdu Hospital of Integrated Traditional Chinese and Western Medicine. A 3D visualization model was constructed based on preoperative CT data to measure the lateral ureteral angle, intramural ureteral diameter, intramural ureteral length and pelvic ureteral diameter. A predictive model for failure of FANS placement without pre-stenting was established after univariate and multivariate analyses, and the model performance was evaluated by bootstrap internal validation, calibration curves, decision curve analysis (DCA) etc. RESULTS: Among the 113 patients, significant intergroup differences were observed between the successful placement group (n = 96) and the failed placement group (n = 17) in age (P = 0.010), lateral ureteral angle (P = 0.018), intramural ureteral diameter (P < 0.001) and pelvic ureteral diameter (P = 0.008). No significant intergroup differences were found in body mass index (BMI), stone CT value, stone size, stone number, gender, hydronephrosis, history of ipsilateral lithotripsy, preoperative fever history or intramural ureteral length (P > 0.05). Multivariate analysis identified intramural ureteral diameter (OR = 0.06, 95% CI: 0.01~ 0.29, P < 0.001) and pelvic ureteral diameter (OR = 0.26, 95% CI: 0.08 ~ 0.78, P = 0.017) as protective factors against placement failure, while the lateral ureteral angle (OR = 1.17, 95% CI: 1.04 ~ 1.31, P = 0.006) was a risk factor for placement failure. Due to sample size limitations, the lateral ureteral angle and intramural ureteral diameter were included in the predictive model based on the Bayesian information criterion (BIC). The area under the curve (AUC) of the model was 0.852 (95% CI: 0.763 ~ 0.941), and the AUC of internal validation via Bootstrap resampling (n = 1000) was 0.846 (95% CI: 0.808 ~ 0.860). CONCLUSIONS: The intramural ureteral diameter and lateral ureteral angle measured by the 3D visualization model based on preoperative CT are powerful predictors for the failure of FANS placement without pre-stenting. The constructed predictive model exhibits good performance. However, further validation with multi-center and large-sample studies is still required.