Comparison of Pathological Outcome and Recurrence Rate between En Bloc Transurethral Resection of Bladder Tumor and Conventional Transurethral Resection: A Meta-Analysis

Simple Summary The conventional surgery for early-stage localized bladder tumor is transurethral resection of bladder tumor (TURBT), but this surgical method cannot preserve the integrity of tumor appearance. Therefore, it results in lower identification rate of detrusor muscle of bladder which plays an importance role in bladder cancer staging. The en bloc procedure, which has raised more concern recently, is performed by resecting the tumor completely rather than piece-by-piece. This method improved the pathological outcome and also improved the underestimation rate of bladder cancer stage. Considering its several advantages, we conducted a meta-analysis to analyze current published studies, and the results showed promising outcomes for bladder cancer staging. Abstract Current treatment for non-muscle invasive bladder cancer (NMIBC) is the conventional transurethral resection of bladder tumor (CTURBT), but the en bloc transurethral resection of bladder tumor (ERBT) has been gaining more attraction in recent years considering better specimen integrity. Thus, we conducted this meta-analysis to compare the safety and efficacy of ERBT versus CTURBT. Trials were collected from an online database. The primary outcomes included identification of detrusor muscle in specimen, residual tumor, 3, 12, and 24-month recurrence rates and same-site recurrence rate. A total of 31 trials were included. The ERBT group had a higher rate of identification of detrusor muscle in specimens (p = 0.003) and lower residual tumor (p < 0.001). Other than that, lower rates of 3-month (p = 0.005) and 24-month recurrence rate (p < 0.001), same-site recurrence rate (p < 0.001) and complications were also observed. For perioperative outcomes, shorter hospitalization time (HT) (p < 0.001), and catheterization time (CT) (p < 0.001) were also revealed in the ERBT group. No significant difference was found in operative time (OT) (p = 0.93). The use of ERBT showed better pathological outcomes and fewer complications, so it could be considered a more effective treatment option for NMIBC.


Introduction
Bladder cancer is the 10th most common cancer worldwide and remains the most common malignancy of the urinary tract. It accounts for approximately 573,000 new cases and 213,000 deaths annually [1]. The most common clinical presentation is gross hematuria, but other symptoms also contain microscopic hematuria or irritative voiding symptoms. There are many risk factors for bladder cancer including advanced age, cigarette smoking, male sex, race, exposure to chemicals, chronic inflammatory conditions owing to infection, pelvic radiation and so on [2].
The treatment strategies for localized bladder cancer could be classified into two different phenotypes, non-muscle invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC). For this reason, the presence or absence of detrusor muscle in bladder tumor specimens reflects the quality of resection in patients with localized bladder cancer.
En bloc transurethral resection of bladder tumor (ERBT) has raised concern in recent years and has been proven to achieve good prognosis by providing a complete resection and preserving the integrity of specimens which contain detrusor muscle. It is believed that the absence of detrusor muscle in specimens was associated with residual disease, early recurrence and tumor understaging [3,4].
Since the raising concern of the en bloc surgery, more and more studies were published recently. However, there are still relatively few studies included in the existing metaanalysis for comparing these two different surgical methods. Additionally, meta-analysis which included the identification of detrusor muscle as outcome parameter were rarer. Thus, it is necessary to perform an updated meta-analysis to compare the outcomes such as different period of recurrence rate, complications and most of all, the identification of detrusor muscle. Herein, we assessed the pathological outcomes and surgical safety of ERBT and CTURBT.

Materials and Methods
The inclusion criteria were as follows: retrospective, prospective and randomized controlled trials; studies published in English language; studies that compared ERBT and conventional TURBT in the patients with NMIBC; trials that included efficacy, feasibility or pathology outcomes. The exclusion criteria in this meta-analysis included original articles published in other languages; single-arm trials, case reports, animal experiments, expert opinions, systematic reviews, conference abstracts and other meta-analyses.
We comprehensively searched the online databases such as PubMed, Cochrane Library and Embase for relevant articles published through April 2021 by using MESH terms including (bladder tumor OR bladder cancer) AND (en bloc resection) AND (transurethral resection). Overall, two authors (C.W. Wang, C.W. Wu) were involved in literature review and data extraction independently. Other than these databases, we also reviewed the previous four meta-analysis articles that compared the efficacy and feasibility of ERBT and conventional TURBT, and the articles that met our inclusion criteria collected in these meta-analysis articles but not in our literature search were included as additional records identified through other sources. After the studies were identified by the steps mentioned above, one of the authors reviewed the data from every study and assessed whether they met the criteria. The following data were extracted: surgical techniques, tumor characteristics, operation time, hospitalization time, catheterization time, 3/12/24-month recurrence rate, same-site recurrence rate, perioperative complications which included bladder irritation, bladder perforation, and obturator nerve reflex and pathological report which contained detrusor muscle in specimen and residual tumor.
The quality of the included randomized controlled trials (RCT) in our study was assessed with the Revised Cochrane risk-of-bias tool for randomized trials (RoB 2) tool [5]. On the other hand, non-randomized studies were assessed by using The Risk Of Bias In Non-randomized Studies -of Interventions (ROBINS-I) assessment tool [6]. The plots for the results above were generated with the robvis web app [7]. The parameters were divided into continuous and dichotomous variables. For the continuous one, standard mean difference with random-effect model was used to assess the difference. For the dichotomous one, the Mantel-Haenszel Test and odds ratio with 95% confidence interval and random-effect model were applied. Data analysis was performed by using Review Manager, version 5.4, and forest plot was used to evaluate the outcome between the two types of surgery.

Studies Selection
Overall, 504 potentially relevant studies were identified. Among them, 495 studies were acquired in the primary literature search, and nine studies were included from other sources. After assessment for the eligibility, inclusion and exclusion criteria, 31 studies were finally included for data analysis. The study selection flow diagram is shown in Figure 1.

Characteristics of the Studies
Of all these articles, there were sixteen retrospective studies, nine prospective cohort studies and six RCTs. A total of 4195 patients were recruited; 2024 cases underwent ERBT and 2171 conventional TURBT. The characteristics of these studies including surgical methods, number of patients included in each study, sex, mean age and postoperative intravesical chemotherapy are shown in Table 1. The tumor characteristics such as size, numbers, location, stage and grade are shown in Table 2. As for surgical techniques of ERBT in these articles, electrocautery, holmium, thulium, KTP laser and hybrid knife were used; as for instilled chemotherapy, mitomycin, epirubicin, pirarubicin and BCG were used [8].

Tumor Complete Resection Outcome
This section contained two outcome parameters; one is identification of detrusor muscle in specimen and the other one is the residual tumor. The former one was available in 13 trials (Figure 2a). The result showed that the ERBT group had a higher identification rate of detrusor (OR 0.26; 95% CI = 0.10 to 0.63; I2 = 88%; p value = 0.003) and lower residual tumor (OR 0.30; 95% CI = 0.16 to 0.57, I2 = 0%; p value < 0.001) (Figure 2b).

Discussion
In this meta-analysis, we comprehensively reviewed the past studies and provided a meta-analysis containing the most studies to compare ERBT with CTURBT for the participants with NMIBC to determine which had better pathological outcome and safety. For the perioperative outcomes, all of the three variables showed a significant test for heterogeneity, so the results would be conservative. The shorter HT and CT were found in the ERBT group and this finding was compatible with the previous meta-analysis in 2016 and 2020 [37,38]. Second, there was no significant difference in terms of 12-month RR, but ERBT showed lower 3-month, 24-month and same-site RR. Though 12-month RR showed no statistical significance, it revealed a trend that ERBT had lower 12-month RR. On the other hand, with respect to the identification of detrusor muscle in specimen, it was rarely analyzed in the previous meta-analysis owing to the small number of trials in the past. Recently, many trials were published containing the pathological outcome [12,19,23,24,[28][29][30][31][32][34][35][36], so in this meta-analysis, 13 trials were enrolled, and the results were significantly superior in the ERBT compared to CTURBT group, but the test for heterogeneity was significant (p < 0.00001, I 2 = 88%). Although we had this amazing result, the explanation of the result should be cautious and conservative. Furthermore, among the eight studies which showed a better rate of identification of detrusor muscle in the ERBT group, five of them performed ERBT by monopolar electrode and the others by laser fiber. It appears that a higher proportion of monopolar usage was observed, but the specific reasons are still unclear. Regarding the last primary outcome, residual tumor rate, the results show that the ERBT group had a significantly lower rate compared to the CTURBT group. Finally, ERBT showed fewer complications in bladder perforation and residual tumor and no statistical significance was noted in urethral stricture [39]. However, as for obturator nerve reflex and bladder irritation, though significant results were identified, the high heterogeneity could not be ignored. There may be some bias existing such as patients' characteristics, demographic difference and different surgeons. Our results were compatible with the past meta-analyses; the ERBT group showed significantly better outcomes than the conventional group.
Lower incidence of obturator nerve reflex and bladder perforation was found in our study and the results were similar to the past studies. The results could be attributive to the prevention of thermal injury and the shorter contact time between the laser fiber or electrode and tumor tissue as compared to the conventional approach [26]. Herr et al. proposed three ways to assess the quality of TURBT: complete resection, deep muscle presence in the specimen and same-site recurrence rate after previous TURBT [40]. The conventional TURBT is performed by resecting the tumor piece-by-piece and it may cause incomplete tumor resection, fragmentation of the tumor tissue and also floating cancer cells. These conditions may lead to the absence of the detrusor muscle and higher tumor recurrence rate. The en bloc resection could maintain the tumor tissue integrity containing the lamina propria and detrusor muscle and enhancing the accuracy of pathological staging. By performing ERBT, we can prevent floating cancer debris and residual tumor in the tumor base and thus lower the recurrence rate, leading to a better prognosis.
The current treatment strategies for different phenotypes of bladder cancer vary greatly. NMIBC, which comprises Ta, T1 and CIS, can be treated by local resection. However, despite the new techniques introduced in this field, the postoperative recurrence rate within 12 months is up to 50% by using conventional resection [41]. Therefore, en bloc resection has caught urologists' attention in recent years. The main difference between the two is the preservation of the tumor integrity. In the CTURBT group, the tumor is resected piece-by-piece and the tumor debris is scattered in the bladder and floated. This kind of surgical procedure violates the principle of completion resection of tumor. Furthermore, electrocautery is usually used in CTURBT, and it can cause eschar to the tumor fragments leading to difficulty in interpretation of specimen. Owing to this, the detrusor muscle is difficult to distinguish and thus it is possible to underestimate the depth of tumor invasion [29]. On the other side, the ERBT technique is performed by making an incision near the tumor down to the muscular layer and completely resecting the lesion. Recent studies have shown that it achieved good prognosis and better efficacy and safety compared with CTURBT [3].
Because of the shortcomings of the conventional TURBT, en bloc resection of bladder tumor technique has gained increasing popularity in the last decade. Theoretically, it provides a better quality of specimen by preserving the integrity of the tumor. ERBT can be performed by using eletroresection [26], laser resection or hydrodissection [31]. The procedure starts with circular coagulation on the mucosa around the tumor with a distance within 0.5-1.0 cm away from the tumor edge. Then, the incision is carried out through the submucosa layer until the detrusor muscle is exposed. The muscular fibers are cautiously dissected from the periphery to the center of the tumor with the tip of a laser fiber. The lesion is then lifted up and detached from bladder surface. After detaching completely, the tumor is retrieved by using various exit strategies. Regarding the learning curve, as most en bloc surgeries are conducted by using laser fiber, there are fewer complications such as bleeding and it provides a better surgical vision. Pankaj N. Maheshwari et al. presented a single-center study showing that the learning curve of Holmium laser ERBT is not steep and is around twenty cases [42]. Therefore, wider adoption of laser ERBT for NMIBCs is feasible and it may help to provide a better prognosis to patients with NMIBCs.
There are some limitations in this meta-analysis. First of all, this meta-analysis included retrospective studies, prospective studies and randomized controlled trials. The difference in the nature of the study types may contribute to the bias across the studies. Secondly, different techniques used in operation such as different types of lasers, electrocauterization and hybrid knife were all included in this study and as we know, laser transurethral resection has better outcomes and fewer complications than the electroresection group [10]. Thirdly, the articles collected in this meta-analysis were from the published database, PubMed, Cochrane Library and Embase, so there was no regional limitation. Thus, the patient demographics, epidemiology in different countries, and tumor characteristics can be factors that influenced the outcome. Finally, though this meta-analysis included the most articles, only six RCTs were analyzed and this has a big impact on certain outcomes. For example, as our primary endpoints, the number of RCTs which contained the result of the identification of detrusor muscle were only three articles. With such a small number, it is easy to cause bias in the results.

Conclusions
As a promising surgical technique of NMIBC, current evidence indicated ERBT is superior to CTURBT in terms of both pathological and clinical outcomes. ERBT provides a higher rate of detrusor muscle in specimen and lower rates of short-term (3 months) and intermediate-term (24 months) cancer recurrence. Furthermore, the ERBT group is associated with fewer postoperative complications, including bladder perforation and bladder irritation. However, we could not ignore the significant heterogeneity in some of these outcomes; these promising results should be explained more cautiously.

Data Availability Statement:
The data presented in this study are available in this article.

Conflicts of Interest:
The authors declare no conflict of interest.