Is Robot-assisted Radical Cystectomy the Preferred Minimal Invasive Procedure for Non-advanced Bladder Urothelial Carcinoma(T1-T2N0M0)? A Single Center Retrospective Study.


 Background: To investigate the perioperative efficacy and cost of robot-assisted radical cystectomy(RARC) and laparoscopic RC(LRC) in patients with non-advanced bladder urothelial carcinomaMethods: 156 patients with non-advanced bladder urothelial carcinoma undergoing minimally invasive radical cystectomy in our center between January 2015 and April 2020 were included. Perioperative data and hospitalization expenses were extracted from our database. All analyses were performed using SPSS 23.0 software, and p < 0.05 was considered statistically significant.Results: The proportion of male patients was 86.5%(135/156) and the median age was 65(IQR 59-71) years old. RARC had a lower PSM rate (0 vs 5.3%,P=0.051), longer median operation time(370 vs 305 min,P<0.001) and higher median hospitalization cost(20565.2 vs 15532.4$,P<0.001). There were no significant differences in intraoperative transfusion rate, anesthesia resuscitation in ICU, postoperative hospital stay, 30-d complications and postoperative treatment expenses between the two groups(P=0.815,0.715, 0.817,0.92 and 0.543,respectively.)Conclusion: Short operation time and low hospitalization costs are favorable factors for LRC, but RARC may be the preferred surgical procedure for non-advanced bladder urothelial carcinoma considering the potentially low PSM rate.Trial registration: A complete informed consent was obtained from the patient and their families before the surgery.Informed consent was signed for all patients.This study was approved by the Ethics Review Committee of the Second Xiangya Hospital of Central South University

All data were obtained from our electronic medical record systems. The histologic type, tumor stage, number of lymph node, positive surgical margin(PSM) and presence of lymphovascular invasion were derived from postoperative pathological reports. Physical condition and anesthesia risk were analyzed using age, American Society of Anesthesiologists(ASA) score and underling diseases. Baseline hemoglobin and albumin level were used to assess anemia and nutritional status followed by transfusion volume and perioperative changes in hemoglobin for estimating intraoperative blood loss. The maximum value of hemoglobin in patients before operation and the value on the day of discharge were obtained, and calculated the decreased value. The postoperative 30-day complications were collected and evaluated by comprehensive complication index(CCI) and Clavien-Dindo classi cation(CDC) system [15]. The expenses of diagnosis, anesthesia, surgery, blood transfusion, postoperative treatment and consumables were calculated respectively. Other perioperative indicators include gender, operation time, incidence of anesthesia resuscitation in ICU, restart autonomous deambulation and postoperative hospital stay.
Continuous variables conforming to the normal distribution and homoscedasticity were analyzed by T test and showed as mean ± standard deviation(SD), otherwise, Wilcoxon rank sum test and median(Interquartile range,IQR) were performed. Categorical variables were presented as the frequency(ratio) and compared by chi-squared or sher exact method. SPSS 23.0 software was used for statistical analysis and p < 0.05 meant a statistical difference in the study. While the RARC group had a higher median hemoglobin value (133 vs 130 g/L,P = 0.042), this mini difference was not clinically signi cant. Gender, age, ASA score,albumin level and other comorbidities between the two groups were no signi cantly different(P all > 0.05). 1.Fisher exact method ASA = American Society of Anesthesiologists, LRC = laparoscopic radical cystectomy, RARC = robotic-assisted radical cystectomy, IQR = interquartile range.

Results
Pathological outcomes are described in Table 2. The two groups achieved a good balance in histological subtype and tumor stage(P = 0.876,0.955). Patients receiving RARC seemed to have a lower PSM rate(0 vs 5.3%,P = 0.051) and higher median lymph node output (11 vs 10,P = 0.062).

Discussion
The perioperative values of RARC and LRC have been fully proved-less blood loss, rapid intestinal recovery, low analgesic needs and short hospital stay [16][17][18][19]. In theory, robotic devices could reduce muscle tremors caused by fatigue in the operators' arm, while weakening haptic feedback. However, two simultaneous mete-analysis comparing the perioperative and tumor outcomes of the two minimally invasive surgical approaches reached divergent conclusions [9,10]. Feng et al [9] considered that RARC was superior to LRC in complications, length of stay, lymph node yield and mortality. Peng and colleagues found no signi cant difference on the efectiveness and safety between RARC and LRC. We hypothesized that heterogeneity of the included literature and neoplasm staging led to the above discrepancy. In our study, there was good comparability between the two groups in age, gender, ASA score, coexisting disease, tumor stage and histological type. Hydronephrosis is one of the secondary prognostic factors in BC patients and affects the choice of treatment options [20]. Only one of the 4 PSM patients had hydronephrosis before operation, so hydronephrosis was not a signi cant factor for PSM in this study(P = 0.475). Perioperative complications and postoperative survival outcomes were associated with tumor progress even with the use of minimally invasive surgical tools [11][12][13]. The PSM rate and 30d mortality of pT4 patients increased to 31.5% and 4.2% [12]. Previous studies suggested that there was no signi cant difference in PSM between the RARC and LRC groups [16,21,22]. Here, RARC had a potentially lower PSM rate(0 vs 5.3%,P = 0.051) in patients with non-advanced diseases, which might owe to the exible mechanical arms and stereoscopic vision. Although the median lymph node yield in the RARC group was higher (11 vs 10,P = 0.062), but this failed to achieve statistical difference and signi cant clinical signi cance. And the lymph node yield depended on the extent of pelvic lymph node dissection.
The surgery time of RARC and LRC uctuated in 328-511 and 301-533 minutes respectively [10]. There are many factors affecting intraoperative time [23]. Although the time-consuming of RARC tends to be shortened [23], it doesn't omit the necessary surgical steps but requires more time in assembling, debugging and disassembling instruments compared with LRC. RARC required obviously longer median operative time than LRC in non-advanced group (370 vs 305 min,P < 0.001), while it did not signi cantly increase the risk of anaesthetic resuscitation(11.1 vs 9.3%,P = 0.715). Some scholars pointed out that intraoperative blood transfusion was associated with survival parameters and robots were expected to reduce the demand for blood products [24]. In non-advanced patients, the intraoperative blood transfusion rate was similar in the RARC and LRC groups(20 vs 18.5%,P = 0.815), which is line with Su's retrospective study of 315 patients [21]. And there was no statistically signi cant difference in hemoglobin decline and transfusion cost (P = 0.489,0.898), suggesting similar intraoperative blood loss.
Khan et al [22] showed no signi cant difference in 30-d major complication(Clavien 3-5) among open surgery and minimally invasive surgical approaches(P = 0.20). Peng and colleagues [10] further demonstrated that the postoperative complication rates were similar between RARC and LRC(P = 0.61). However, other voices insisted that LRC was associated with a higher complication rates(p = 0.02) [9]. There existed a higher morbidity in LRC group(50% vs 36.5%,P = 0.017) [21]. Vetterlein and colleagues [15] reported that 99% of patients with RC would experience some complications, with a fatality rate of 2.4% within 30 days after surgery. However, the concept of postoperative meaningful complications has not been clearly de ned, resulting in a wide variation in the reported rate of complications in different institutions. We found no statistically signi cant difference in the rate and severity of 30-d complications between RARC and LRC in patients strati ed for tumor stage. Likewise, postoperative hospital stay was similar. Inconsistent with the above studies, lymphorrhagia was a common issue in minimal Invasive RC in our retrospective study(supplement Table 1). Compared with open surgery, RARC has advantages in the less blood loss, mild complication and short hospital stay, which cannot completely counteract the high cost of equipment [25]. Patients with RARC spend much more than those with LRC [21]. There is no study about cost analysis between RARC and LRC. The costs of surgery and consumables were the reason for the expensive hospitalization fees of RARC.
Therefore, we has compared the effectiveness and cost of RARC and LRC. Compared with previous studies, LRC and RARC had better comparability in terms of intraoperative transfusion rates, complications, postoperative hospital stay and treatment costs. Meantime, RARC showed better pathological outcomes, which need to be further validated with long-term survival data.
Inevitably, there were several limitations in this study. First, Our study was a single-center retrospective study with its inherent drawbacks, but we provided much detailed perioperative data. Second, different surgical habits and treatment ideas among the ve surgeons may lead to some bias. Third, instead of collecting estimated blood lose, we assessed the intraoperative blood transfusion, perioperative hemoglobin level and blood fees, which we believe can re ect intraoperative condition objectively. Nevertheless, this study enriches the perioperative data of LRC and RARC and opens a new perspective for patients to choose surgical tools.

Conclusion
Short operation time and low hospitalization costs are favorable factors for LRC, but RARC may be the preferred surgical procedure for non-advanced bladder urothelial carcinoma considering the potentially low PSM rate. Consent fore publication A complete informed consent was obtained from the patient and their families before the surgery.Informed consent was signed for all patients.

Availability of date and materials
The datasets used and analysed during the current study are available from the corresponding author on reasonable request

Competing interests
Xiao Ming, Zhong Zhaohui, Ren Jiannan, Xiong Wei have no con ict of interest to declare. Funding.
No funding was received.

Authors' contribution
Xiao Ming was responsible for study concepts,study design and the writing of the manuscript. Zhong Z was responsible for reviewing the article and proposing changes. Ren J collected and analyzed data. Xiong W was responsible for the starring revision of the article and study design. All authors read and approved the nal manuscript.

Figure 1
Patients selection process

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupplementaryTable.pdf