Aim: To evaluate the efficacy of flexible ureteroscopic lithotripsy (FURL) and extracorporeal shock wave lithotripsy (ESWL) in the treatment of ureteral calculi based on decision tree model.
Methods: A total of 600 patients with ureteral calculi, including 289 treated with FURL and 311 cases with ESWL in Anqing Municipal Hospital from June 2021 to August 2023, were selected as study subjects. Perioperative indicators and stone clearance rate of the two groups were compared, and the preoperative and postoperative (24 and 72 hours) changes of serum creatinine, cystatin C (Cys-C) and microalbumin were observed. The complications during and 7 days after treatment, the influence of perioperative indexes, total stone removal rate and renal function indexes were analyzed using decision tree method, and a complication risk prediction model was constructed.
Results: The operation time, length of hospital stays and postoperative hematuria time in FURL group were shorter than those in ESWL group (p < 0.001), and the usage of painkillers was less frequent in FURL group than in ESWL group (p = 0.002). The total stone removal rate in the FURL group was higher than that in the ESWL group (p < 0.001). Serum creatinine, urinary microalbumin and Cys-C in both groups were lower before surgery than at 24 h and 72 h after surgery (p < 0.05). Serum creatinine, urinary microalbumin and Cys-C in FURL group were lower than those in the ESWL group at 24 and 72 h after operation (p < 0.001). The overall complication rate in the FURL group was lower than that in the ESWL group (p = 0.028). Decision tree model analysis showed that four explanatory variables, including preoperative creatinine, urinary microalbumin, Cys-C and surgical method were identified by screening. The risk statistic of the model was 0.027, and the accuracy, sensitivity and specificity of the model in predicting postoperative complications in patients with ureteral calculi were 97.33%, 97.73% and 97.30%, respectively.
Conclusions: FURL has significant advantages over ESWL in the treatment of ureteral calculi, and has less impact on renal function and fewer complications. Preoperative creatinine, urinary microalbumin, Cys-C and surgical methods may adversely influence the occurrence of postoperative complications. These identified factors can be employed to build a decision tree model for predicting the occurrence of postoperative complications.