RC combined with bilateral pelvic lymph node dissection (PLND), as the standard treatment for radical bladder cancer, has been validated clinically with multiple studies demonstrating similar oncologic results versus both open and minimally invasive procedures. However, the most important issue after RC is performing a proper UD or reconstruction as only when this link is completed, can the radical operation for bladder cancer be considered as over. Radical resection of bladder cancer combined with UD is considered as one of the most complex and difficult surgeries in the field of surgery due to the large amount of surgical trauma caused, the extensive resection needed, and the complexity of the surgical procedure, including resection and reconstruction, as well as the involvement of the intestinal tract. It is not difficult to infer therefore, that the rate of surgical complications is very high. The high complication rates in RC have been confirmed by many studies,with an overall complication rate of approximately 30–70%[29, 33]. A high incidence of complications is associated with urinary tract reconstruction or diversion, with a reported duct-related complication rate of up to 32.7%[34]. IC UD is currently the most widely used technique, accounting for the highest proportion of all UD surgeries[13, 14, 16].
The main problem with classic IC is in the placement of ileal segment and uretero-ileal anastomosis. With classic IC, the left ureter is widely dissociated and pulled towards the right side for anastomosis with ileum, increasing the potential risk for complications related to uretero-ileal anastomosis. This complication has also been reported by many other studies[18, 24]. To avoid such complications, many scholars have modified the Bricker IC. One method is to make the stoma extraperitoneal to avoid stomata-related complications. Another one is to pull IC from the posterior sigmoid colon towards the left side to reduce complications of left ureteral-ileal anastomosis. Combining the advantages of the two above procedures, we externalised the peritoneum of the stoma and pulled the input side of IC towards the left, significantly reducing the incidence of ileal conduit-related complications and ureteral-ileal anastomose-related complications. In this study, 103 cases (48.8%) had complications, including 16 cases (7.6%) of ileal conduit-related complications and 12 cases (5.7%) of ureteral anastomotic complications, which was lower than that reported in previous studies[18, 20, 24]. It is noteworthy that no cases of parastomal hernia were observed. Hussein et al.[35] reported that the incidence of parastomal hernia after RC and IC was as high as 20%, of which 15% patients required surgery. The absence of parastomal hernia in this study was mainly due to our modifications.
To reduce the complications associated with ostomy and ureteral-ileal anastomosis, to improve surgical safety, and to improve postoperative quality of life, we optimised and modified surgical procedures for UD after RC, with some parts completely changed. Based on an experience with 211 cases of modified IC over 8 years, we suggest the following: 1) If both ends cannot be taken into consideration, the outlet end of IC should be guaranteed first. Since IC enters the retroperitoneum from anterior sacrum, it needs a relatively long ileum and mesentery. Some obese patients, or people with short mesenteries, have difficulty taking care of both sides. After the peritoneal extraperitoneum of the IC, the position of uretero-ileal anastomosis should be preliminarily estimated under the condition that the exit is long enough. At this time, it may not be suitable to pull the left side of the IC towards the left ureter, but it can be as close to it as possible. That is, the position of uretero-ileal anastomosis can be slightly adjusted according to the specific length of the IC; 2) At the skin stoma, the aponeurosis of rectus abdominis should be sutured intermittently with 1 − 0 silk thread. The suture should be retained and used to fix the ileostomy. This can not only prevent the stoma from retraction or protrusion but can also prevent parastomal hernia. Extraperitoneal stoma, suture choice, and fixation of the outlet are the key techniques to avoid parastomal hernia; 3) Uretero-ileal anastomosis does not require anti-reflux. Continuous suture can be performed directly with 4 − 0 absorbable suture. The single J tubes were routinely placed inside as the stent for internal drainage, thus reducing the chance of urine leakage and anastomotic stenosis; 4) Under tension-free conditions, the maximum feasible amount of residual ureter should be removed to ensure a good blood supply to the remaining ureter, preventing lower ureter and anastomotic ischaemia, thereby reducing the occurrence of anastomotic stenosis and hydronephrosis.
To ensure negative surgical margins as far as possible, we had a wide range of resected area. Since Bricker IC does not need to consider the issue of urinary control, extended resection is feasible. The bladder was dissociated on both sides of the pelvic floor fascia and placed in the extraperitoneal space. The neurovascular bundle was generally not retained unless the patient requested it. After a cystectomy, the pelvic floor was almost skeletonized. In this study, none of the patients had a positive surgical margin. In addition, since IC was pulled towards the left retroperitoneum, the left ureteral stump was excised 8–10 cm, further ensuring a negative margin.
The ideal extent for lymph node dissection has not yet been determined. Some scholars reported that enlarged lymph node dissection yielded more lymph nodes and had a higher positive rate than standard lymph node dissection. In patients with lymph node positivity, the five-year recurrence-free survival rate after expanded lymph node dissection was significantly higher than that after localised lymph node dissection[36, 37]. However, a randomised phase 3 clinical trial involving 198 patients with extended lymph node dissection and 203 patients with limited lymph node dissection yielded the opposite result. The former had no significant advantage in terms of 5-year RFS,CSS and OS[38]. The current guidelines recommend a minimum of standard lymph node dissection, defined as the removal of lymphoid fat tissue below the bifurcation level of the iliac vessels. Guidelines also recommend that at least standard lymph node dissection should be performed, which is the removal of adipose tissue below the level of the iliac vascular bifurcation[12]. In this study, we used standard lymph node dissection. The median number of harvested lymph nodes was 14 (IQR, 6–18), and 16 (7.6%) patients were lymph node positive. Although there is no consensus regarding the scope of lymph node resection, the impact of lymph node involvement on tumour prognosis is clear[39]. Patients with lymph node metastasis had worse RFS and OS than those without lymph node metastasis and that difference was statistically significant. Univariate and multivariate regression analyses further confirmed that lymph node positivity was an independent predictor of the prognosis in RC[8, 40, 41].
The prognosis of bladder cancer after radical resection has been verified via long-term clinical practice[7, 8, 42]. The introduction of minimally invasive surgery in the latest two decades, such as laparoscopic RC and robot-assisted RC for bladder cancer, has not only improved perioperative parameters, but has also achieved similar oncologic results as open surgery[43]. This has been further demonstrated in recent randomised clinical trials[9, 10, 28]. In our study, a modification of IC significantly reduced associated complications, especially those related to ureteral ileum anastomosis and the stoma. The tumour prognosis was also encouraging. Our study further confirmed that the most accurate predictors of tumour recurrence and death were the pathological stage of the primary tumour and regional lymph node status. Through univariate and multivariate regression analyses, the study also found that preoperative comorbidities had significant effects on RFS and OS, with poorer RFS and OS outcomes for patients with CCI ≥ 1. This was consistent with the poor physical status mentioned in the RAZOR trial as an important predictor of 36-month progression-free survival[29].
Minimally invasive treatment of tumours has become widely used in practice and it is most widely used in the field of urology. Minimally invasive surgery provides many important advantages, such as reducing intraoperative bleeding, accelerating recovery, shortening hospital stay and reducing wound complications[14, 29]. However, in addition to these perioperative advantages, there is scant evidence that these methods have better oncological outcomes, with outcomes still depending largely on the local tumour stage, the biological characteristics of the disease, and perhaps also on the surgeon's experience[9]. In fact, it is not difficult to speculate that, in a sense, minimally invasive surgery only achieves the same range of results as open surgery through less invasive methods, and in some aspects, with less resection under certain refined procedures, such as uterus preservation. In this study, there were 136 (64.5%) cases of minimally invasive surgery, compared with 75 (35.5%) cases of open surgery, RFS and OS were comparable, which was the same as previously reported results[9, 10, 28].
Currently, although the application of robotic surgery in large medical centres is accelerating around the world, its high cost hinders further popularisation. Although our hospital is a cancer treatment centre in central China, Da Vinci robotic surgery has started only recently. Therefore, it is reasonable to speculate that laparoscopic surgery will remain the main mode of minimally invasive surgery for the years to come. It is worth mentioning that open surgery is known for its relatively short operating time. For patients with a history of abdominal surgery, extensive intestinal adhesions, or cardiopulmonary diseases that do not allow long-term surgery with high CO2 pressure in the abdominal cavity, open surgery will continue to play an irreplaceable role for quite a long time.
Since the current guidelines for bladder cancer in China are based on the 2010 TNM staging, which has not been updated, we refer to the seventh edition of the TNM staging system for clinical and pathological staging. The prognosis of patients with higher stages was significantly worse than that of patients with lower stage.
So far, although this is the largest single-centre study of modified IC after RC, it still has some inherent limitations. This study is retrospective in nature, making it prone to selection biases and differences in care. Inconsistent follow-up and lack of multicentre results also limit the wider applicability of this study. The OS and RFS of this study were higher than those reported before, which may be related to the relatively substantial number of patients who were lost to follow-up. In the past 20 years, China's economic and social development has been in a transitional period. Some patients, especially those in rural areas, have frequently changed their address and mobile phone number. Further, some patients living in remote areas may have not received regular follow-up, resulting in some mild complications not being recorded. In addition, under our country’s current medical system, a three-level referral system is not yet complete. Further, patients can seek medical treatment anywhere, which poses challenges to follow-up. The number of complications may thus be less than the actual number of occurrences, and some may not have been recorded. We did not calculate the cancer-specific survival rate, mainly because a considerable number of patients died at home and the cause of death was unknown. However, except for a few deaths, the follow-up time was more than half a year, which showed that the complication reports within 180 days were accurate. Therefore, the results of the incidence of postoperative complications are reliable. However, most of the patients lost to follow-up were mainly concentrated before 2014,and these patients were lost to follow-up due to changes of their phone numbers and addresses. There were relatively few patients lost to follow-up after 2015, so the 3-year and 5-year RFS and OS were basically unaffected. In addition, the proportion of total patients lost to follow-up is still within the statistical range; therefore, the impact of loss to follow-up on survival rate is limited.