To our knowledge, this study is the first to develop and validate a predictive nomogram to identify patients at increased risk of ureteral restenosis after balloon dilation. The model incorporated five items that should be evaluated with priority in clinical practice, including narrow nature, balloon size, NIT, CKD, and stent retention time.
In our study, 97 patients relapsed during follow-up, with the overall ureteral re-stricture rate was 30.2%. The results are superior to a patency rate of 41.1% reported by Campschroer et al.[1 2] Similarly to Stehman, we observed that benign stenosis is more suitable for balloon dilation than malignant stenosis (recurrence rate:25.8%vs51.0%). [3 4] As was shown in Fig. 1, The lowest recurrence rate had seen in stenosis secondary to both ureteral polyps (14.2%) and vascular compression (14.2%), followed by secondary to lithiasis (25.9%) and surgery for benign gynecologic disease (27.7%). All of them belonged to benign stricture. The long-term patency rate of dilation for benign stenosis was comparable to that of open surgery, which indicated that balloon dilation could replace open surgery to treat benign ureteral strictures. However, 14.6% of patients failed immediately after the stent was withdrawing, and a new stent was required to be inserting. These patients were considered not suitable for balloon dilation unless not tolerate open surgery. 41.67% of the above patients belonged to malignant. Therefore, the best indication for balloon expansion is benign stenosis, such as surgery secondary to stones, ureteral polyps, vascular compression, and benign gynaecological diseases.
Some strictures (19 5.81%) were observed secondary to the laparoscopic radical cystectomy (LRC 12 3.67% recurrence rate:66.67%) or the robot-assisted radical cystectomy (RARC 7 2.14% recurrence rate:28.5%). These strictures were possibly secondary to ischemia in the context of a widespread incision. It is worth noting that stenosis secondary to RARC has a lower recurrence rate after balloon dilation than other stenoses secondary to LRC.
In univariate analysis, no significant statistical difference had revealed among different stenosis sites or the number of stents retained. It is similar to previous studies that reported that there was no apparent association between stricture location (upper, mid, or distal ureter) with long-term success rate. [3 6–10] M.J.van Son's study reported that the stricture side was an independent risk factor for stricture recurrence (P = 0.009, HR 0.35, 95%CI 0.16–0.77). However, there was no significant difference in SFS between patients with left ureter stricture and those with the right ureter stricture in our study (P = 0.504). Interestingly, our results found no significant difference in SFS between one and two stents retention (P = 0.293). Two stents retention might cause ureteral ischemia that leads to a bad outcome.
Stavros I. reported that when indwelling stents for prolonged periods, stents may prevent tissue repair for inflammation. In our study, balloon sizes were associated independently with SFS, which is consistent with several previous studies showing that appropriate size could reduce the damage to the blood supply and rupture the fibrous tissue in the stenosis. [11 12]
CKD is also considered to be an independent risk factor for patients with stenosis recurrence after balloon dilation, which is consistent with previous studies that kidney function was considered an important predictor. [2 13] Moreover, our study showed a significant positive correlation between the five CKD stages.
Our study possesses several strengths. First, our nomogram had established through the analysis of a relatively large patients cohort. Furthermore, the calculation method of follow-up time used more accurately reflected the recurrence of stenosis. Also, the nomogram is easily applicable in clinical practice. Therefore, it can use as an essential screening tool that allows us to improve clinical care decisions for patients with a ureteral stricture. Surgeons and patients could make better choices through this usable rating tool before surgery. However, our research also has some limitations. First, the developed nomogram was based on data obtained retrospectively from a single centre. Second, due to missing data, some important factors such as the length and diameter of the stenosis were not included in this study. Although this model showed good predictive ability through internal validation, it still needs external validation.