Bladder cancer (BCa) is the ninth most common cancer worldwide and ranks 13th in terms of annual mortality from cancer (1, 2). Transurethral resection of bladder tumor (TURBT) followed by intravesical adjuvant chemotherapy or immunotherapy is the standard diagnostic and treatment method for non-muscle invasive bladder cancer (NMIBC) (3, 4). However, TURBT represents a challenge for urologists due to the high incidence of residual tumors (5, 6). A systemic review that contained 31 studies on 8409 patients with NMIBC revealed that the incidence of residual tumor was 17-67% in patients with Ta and up to 20-71% in patients with T1 after first TURBT (7).

Residual tumor following TURBT has been considered to be partly responsible for recurrence (8), and the European Association of Urology (EAU) guidelines recommend re-TURBT for patients with high-risk BCa (9). During re-TURBT, the base of the primary tumor should be second resected by the operating surgeon to eradicate residual disease and ensure accurate pathological staging. However, the incidence of residual tumor in re-TURBT specimens is low, especially when detrusor muscle (DM) is present in the first TURBT specimen. In addition, re-TURBT may impose an additional economic and emotional burden on patients, and a second resection at the base of the primary tumor will increase the risk of bladder perforation, especially for women (10).

Furthermore, a systematic meta-analysis of six studies detailing 3257 participants recently showed that re-TURBT did not improve survival outcomes in patients with T1 BCa (11). Similarly, a retrospective study conducted by Gontero et al. noted that in patients with high-grade T1 BCa treated with intravesical BCG, re-TURBT did not improve oncological outcomes (12). Furthermore, Calò et al. also documented that re-TURBT did not bring a survival benefit in patients with completely resected high-risk BCa (13).

These findings raise questions regarding the necessity of re-TURBT and second resection of the base of the primary tumor in patients with high-risk BCa. Thus, identifying patients who might benefit from re-TURBT and second resection of the base of the primary tumor would be very valuable. In the present study, we aimed to assess the rate and location of residual tumors in re-TURBT specimens and to explore the risk factors. Moreover, a risk stratification tool was developed to identify patients who would likely benefit from re-TURBT and second resection at the base of the primary tumor.

Re-TURBT is an important part of the optimal management of high-risk BCa and is recommended by several international guidelines. However, several recent publications suggest that re-TURBT could be avoided in well-selected patients with high-risk BCa and identifying them would be very valuable (12, 15, 16).

The presence of DM in resection samples is a suitable indicator of complete resection and indirectly reflects the quality of the first TURBT. Dutta et al. showed that staging inaccuracy was critically dependent on the absence of DM in the specimen, with upstaging at radical cystectomy in 62% and 30% of patients without or with DM during TURBT, respectively (17). Furthermore, a retrospective study conducted by Huang et al. revealed that the rate of residual tumor was 51.8% in patients without DM in the first TURBT specimen, which was significantly higher than that in patients with DM in the first specimen (51.8% vs. 20.9%, OR: 15.537, 95% CI: 2.814-85.789, P=0.002) ( 18). Similarly, Ayati et al. documented that the risk of residual tumor was increased by 21-fold when DM was absent in the first TURBT specimen (19). Moreover, they found that the absence of DM in the first resection specimen was associated with upstaging (OR: 8.123, 95% CI: 1.478-44.632), indicative of the presence of residual tumor at the base of the primary tumor. Subsequently, our study also confirmed that the absence of DM in the first resection specimen was associated with the risk of residual tumor, and this residual tumor was likely to be located at the base of the primary tumor (OR: 7.247, 95% CI: 0.923-56.926, P=0.06) ( Supplementary Table 3 ). In addition, a recent study revealed that re-TURBT may not be necessary in patients with T1-HG/G3 if DM is present in the first TURBT specimen (12). These results strongly indicate that the absence of DM in the first resection is an important surrogate marker of residual tumor. Patients without DM in the first TURBT specimen should receive re-TURBT, and surgeons should ensure that DM is obtained from the base of the primary tumor to confirm the depth of invasion during re-TURBT.

The adequacy and completeness of TURBT depend on the experience of the surgeon (20). In a retrospective study, Zurkirchen et al. analyzed the data of 214 patients treated with re-TURBT and found urologists in training had an equally low rate of residual tumor compared to senior urologists (27% vs. 37%, P=0.08) ( 21). However, various studies have documented that senior surgeons are likely to achieve a higher rate of DM presence and decrease the residual tumor rate (22–24). To our knowledge, this was the first study to explore the relationship between residual tumor and the expertise of surgeons. Furthermore, we found that UC subspecialists are more likely to achieve clean resection than non-UC subspecialists. The major reason for this phenomenon is that the UC subspecialists were experienced and confident in performing resections that were sufficiently wide and deep while ensuring technical safety. Another reason is that UC subspecialists were more likely to obtain DM in the first resection (OR: 8.721, 95% CI: 2.525-30.118, P=0.001) ( Supplementary Table 2 ). Moreover, the risk of residual tumor located at the base of the primary tumor increased by 4.5-fold when the first resection was performed by non-UC subspecialists ( Supplementary Table 3 ). These results indicate that the surgeon’s experience and expertise are important indicators of the rate and location of residual tumors.

En bloc resection possesses a better hemostatic effect, clearer surgical fields of vision, and high-quality histological specimens than TURBT, as the integrity and architecture of the tumor can be maintained. Kramer et al. showed that DM was found in specimens of 97.3% of patients who underwent en bloc resection for a bladder tumor (25). Moreover, various studies have demonstrated that en bloc resection can result in a high rate of DM presence (theoretically up to 100%) and complete tumor removal (24, 26, 27). The same conclusion was reached in the present study: en bloc resection was associated with the presence of DM (OR: 2.467, 95% CI: 1.046-5.814; P=0.039) ( Supplementary Table 2 ). All these results suggest that en bloc resection can achieve high-quality resection and may decrease the number of re-TURBT procedures.

Recently, the management of NMIBC has been improved by new endoscopic technologies, such as photodynamic diagnosis (PDD) and narrow-band imaging (NBI). Owing to the advantage of tumor visualization, PDD and NBI could improve the quality of TURBT and may be promising technologies to avoid unnecessary re-TURBT. A literature review that included 44 studies showed that the rate of residual disease after PDD resection was only 4.5–32.7% compared to 25.2–53.1% after white-light resection (28). Furthermore, they also pointed out that the odds ratio of residual tumor for PDD was 0.28 compared to white-light but the relative risk of residual disease was 2.77-fold compared to white-light. Moreover, Ma et al. analyzed the data of 124 patients with NMIBC, and found that NBI-assisted TURBT could significantly reduce the rate of residual disease (29). However, a recent study showed that the rate of residual tumor was as high as 58.7% at the second TURBT with PDD (30). Thus, whether PDD or NBI could avoid re-TURBT still remains unconcluded and further studies are required.

Multiparametric magnetic resonance imaging (mp-MRI) for BCa could provide a high tissue contrast resolution and effectively differentiate bladder wall layers (31, 32). Thus, mp-MRI can assist urologists to identify the infiltration depth of tumor. Giudice et al. prospectively collected the data of 231 patients who underwent mp-MRI before initial TURBT and found that the sensitivity, specificity of mp-MRI to identify patients with MIBC at re-TURBT was 85% (95% CI: 62-96.8%) and 93.6% (95% CI: 86.6-97.6), respectively (33). Moreover, a study also documented that mp-MRI with a vesical imaging reporting and data system (VI-RADS) was an effective and reliable method to determine the patients who could benefit from re-TURBT (34). Furthermore, in the era of big data and precision medicine, not only mp-MRI, but also artificial intelligence (35) and molecular biomarkers (36) have become promising tools to identify patients with BCa who will benefit more from re-TURBT.

In our study, the rate of residual tumor during re-TURBT was 22.2% (32/144), and only 15 patients (10.4%) had a residual tumor at the base of the primary tumor. Several previous studies have reported that residual tumors are found in up to 50% of patients (17, 37), and between 30% and 60% of T1G3 BCa will become muscle invasive at radical cystectomy (38), which was higher than was identified in our study. The complete preoperative evaluation in our institution, including the acquisition of tumor characteristics and lesion biopsy during the preoperative cystoscopy examination, as well as the selection of an appropriate operative method, is the major reason. Moreover, these results indicate that a proportion of high-risk BCa patients may not benefit from re-TURBT and second resection at the base of the primary tumor. The patients in our study were categorized into risk groups based on the expertise of the surgeon, the operative method, and the presence or absence of DM in the first specimen. Half of the patients in the high-risk group had a residual tumor at re-TURBT, which was significantly higher than that of patients in the low- and intermediate-risk groups. Moreover, the rate of residual tumor at the base of the primary tumor was still higher in the high-risk group than in the low- and intermediate-risk groups (23.5% vs. 2.0% vs. 10.2%). Hence, these patients could benefit from re-TURBT and second resection at the base of the primary tumor.

Similar to other retrospective studies, this study was limited by a retrospective study design and a small sample size from a single center, which might lead to a selection bias. Furthermore, the role of this risk stratification as a diagnostic tool for candidates for re-TURBT was not externally validated. Therefore, further prospective multicenter studies are warranted to support our findings.

Non-UC subspecialists and the absence of DM in the first TURBT specimen are risk factors for residual tumor at re-TURBT. Furthermore, en bloc resection may improve the rate of DM presence in the first TURBT specimen. Risk stratification based on the above three factors may help identify patients who might benefit from re-TURBT and second resection the base of the primary tumor.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

JJF: data collection, data analysis, and manuscript writing. XZ: data collection and manuscript writing. JHF: performed operations and manuscript editing. LL: performed operations and manuscript editing. DH: project development and manuscript editing. KW: project development, performed operations, and manuscript editing. All authors contributed to the article and approved the submitted version.

This study was supported by the Clinical Research Award of the First Affiliated Hospital of Xi’an Jiaotong University, China (No. XJTU1AF-CRF-2015-002 to DH).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.