Omitting the Second Bladder Resection: A 3-Year Prospective Pilot Study

Abstract

Background/Objectives: The study aimed to find out if the patients without a reTURBT, due to changed protocol in our centre because of the COVID-19 pandemic, had presented higher rates of relapse or progression compared to patients treated by standard reTURBT protocol. Methods: A prospective study was conducted including 43 patients with high-risk T1 non-muscle invasive bladder cancer diagnosed between March 2020 and June 2021. Patients were divided into two groups: those who underwent reTURBT and those who did not, due to limitations during the COVID-19 pandemic. All patients received intravesical BCG induction therapy and were followed for 3 years. The institutional research ethics committee approved the study. Results: A total of 43 patients were included, 17 (39.5%) underwent reTURBT and 26 (60.5%) did not. No significant differences were observed in tumour characteristics between groups. Recurrence occurred in 43.8% of the reTURBT group and 15.4% of the non-reTURBT group. Tumour progression to T2G3 was observed only in the reTURBT group. Survival analysis showed no significant differences in recurrence-free survival between groups (p = 0.299). Conclusions: Omitting reTURBT in carefully selected patients did not result in significantly worse oncological outcomes; however, due to the small sample size, the study is underpowered and these findings should be interpreted with caution. Early BCG administration in the non-reTURBT group may have contributed to favourable recurrence-free survival. However, further prospective studies are needed to confirm these findings and define optimal criteria for safely omitting reTURBT.

1. Introduction

According to the European Association of Urology (EAU) Guidelines, a second transurethral resection of the bladder tumour (reTURBT) is recommended in selected cases, particularly for patients with T1 high-grade non–muscle invasive bladder cancer (NMIBC), for the following reasons: (1) to eliminate any residual tumour, (2) to confirm correct pathological staging and (3) to collect missing clinical information such as the involvement of the prostatic urethra. The guidelines emphasize that even in cases where detrusor muscle is present in the initial TURBT specimen, residual tumours can be found in up to 58% of patients and under-staging can occur in 11%, thus justifying the importance of reTURBT in high-risk diseases [1]. The American Urological Association Guidelines recommend a reTURBT in cases of an incomplete first resection, patient with high-risk, high-grade Ta tumours and T1 disease [2].

The efficacy of intravesical BCG therapy in NMIBC may be influenced by host-related factors, including systemic inflammation and smoking. A machine-learning CART algorithm identified smoking status and inflammatory biomarkers (LMR, NLR, PLR) as relevant variables to stratify BCG response [3]. Krajewski et al. in a multicentre study suggest that there might be no difference between patients with high-grade T1 disease who had reTURBT before BCG induction and patients with reTURBT after BCG induction [4].

The COVID-19 pandemic disrupted standard urological care pathways worldwide, leading to surgical delays and modified treatment protocols in many centres [5,6,7]. Although bladder cancer surgeries were among the least affected, local resource constraints at our institution required a temporary suspension of reTURBT in selected patients with T1 non–muscle invasive bladder cancer. Instead, these patients received immediate BCG therapy following a complete initial resection with non-affected muscle present in the specimen.

This prospective study was designed to assess whether omission of reTURBT under these exceptional circumstances led to higher rates of tumour recurrence or progression compared to patients treated according to standard protocols.

2. Materials and Methods

A prospective study of 612 patients who underwent TURBT in our institution between March 2020 and June 2021 were identified. The inclusion criteria were as follows: (1) patient age more than 18 years old and (2) indication of the second TURBT based on initial anatomy (no muscle-invasive bladder T1 tumour in the first TURBT). We exclude patients with the following: (1) recurrent tumours, (2) emergency procedures, (3) palliative surgery, (4) incomplete resection (visually incomplete resection in operative report or no muscle visualized at the resection base), or (5) absence of muscle in the first TURBT anatomy specimen (pathology report). The patients were followed for 3 years.

The patients were divided into two groups: those who underwent a second TURBT and those who did not. Both groups were followed up as high-risk NMIBC cases receiving induction with BCG instillation course post-TURBT. Patient’s epidemiological and clinical data were identified: sex, age and smoking history. As for operative items, we collected the date of the first TURBT and tumour characteristics during the first TURBT (architecture, number of lesions, size). Also, the pathology specimen grade of the tumour was collected (in accordance to World Health Organization (WHO) classification 2004/2016 and WHO 1973) [1]. In patients who underwent the second TURBT, identical items were collected. For adjuvant treatment, we identified the intravesical induction course therapy with BCG, the start and end date, the total number of installations and if the patient presented some complications. Time to the first BCG administration and relapse was calculated from the date of the first TURBT in both groups.

Follow-up tests were reviewed (cystoscopy, urine cytology, imaging if available). Patients with tumour relapse during follow-up were identified. Data on the new TURBT was collected. The minimum duration of follow-up was established at 3 years.

Mean and standard deviation (SD) were reported for continuous variables, proportions and percentages for the categorical. The univariate analysis of the groups was performed with Student’s t-test (numerical variables with normal distribution) and a Chi-square test (for categorical variables). The univariate effect of the absence of reTURBT in the tumour recurrence was explored by a log-rank test. The statistical significance threshold was set at p < 0.05. The analysis was performed with SPSS 26.0 (Chicago, IL, USA).

All procedures were conducted in accordance with the Declaration of Helsinki and Spanish legislation. The institutional research ethics committee (Clinical Drug Trials Committee of the Getafe University Hospital) approved the study (approval number 21/47). Each patient signed an individual informed consent to undergo TURBT and to allow data from their clinical chart to be collected for research purposes. Recruitment of participants began under a consultation of research ethics committee in the context of urgent pandemic-related circumstances. Formal ethical approval was subsequently obtained on 31 March 2022.

3. Results

A total of 43 patients were included. In total, 17 patients (39.5%) underwent a reTURBT, and 26 (60.5%) did not due to the COVID-19 pandemic. The mean age was significantly higher in the reTURBT group, 75.29 years (SD ± 11.02), compared to 68.65 years (SD ± 7.44) in the group without reTURBT (p = 0.023). There were no significant differences regarding sex distribution between groups; there were 14 male patients (82.4%) in the reTURBT group compared to 18 male patients (69.2%) in the group without reTURBT (p = 0.48).

Regarding smoking habits at diagnosis, the most common status among patients without reTURBT was never-smoker (42.3%), whereas in the reTURBT group, ex-smokers (35.3%) were slightly more frequent, closely followed by smokers (23.5%) and never-smokers (29.4%). No significant differences were observed in smoking habits between the two groups (p = 0.874).

Tumour characteristics at initial TURBT showed sessile configuration as the most frequent (94.1% in reTURBT and 80.8% without reTURBT, p = 0.221), with a predominantly papillary architecture (76.5% reTURBT and 84.6% without reTURBT, p = 0.439). Most patients presented a single tumour, particularly in the absence of the reTURBT group (61.5% vs. 41.2% in reTURBT, p = 0.170). The tumour size commonly ranged between 11 and 30 mm, occurring in 35.3% of reTURBT cases and 46.2% of those without reTURBT, without statistically significant differences (p = 0.217).

Detailed data on the baseline characteristics and first TURBT findings are presented in

Table 1. Baseline characteristics of patients and first transurethral resection data.

	ReTURBT N = 17 (39.5%)	Absence of ReTURBT N = 26 (60.5%)	p-Value
Age (years), mean ± SD	75.29 ± 11.02	68.65 ± 7.44	0.023
Sex, n (%)			0.48
Male	14 (82.4%)	18 (69.2%)
Female	3 (17.6%)	8 (30.8%)
Smoking habit, n (%)			0.874
Never-smoker	5 (29.4%)	11 (42.3%)
Smoker	6 (23.5%)	8 (30.8%)
Ex-smoker	6 (35.3%)	7 (26.9%)
Tumour configuration, n (%)			0.221
Sessile	16 (94.1%)	21 (80.8%)
Flat	1 (5.9%)	5 (19.2%)
Tumour Architecture, n (%)			0.439
Papillary	13 (76.5%)	22 (84.6%)
Solid	0	1 (3.8%)
Mixed	4 (23.5%)	3 (11.5%)
Number of tumours, n (%)			0.170
1	7 (41.2%)	16 (61.5%)
2	5 (29.4%)	3 (11.5%)
3	3 (17.6%)	6 (23.1%)
More than 5	2 (3.8%)	1 (3.8%)
Tumour size, n (%)			0.217
<5 mm	2 (11.8%)	2 (7.7%)
6–10 mm	2 (11.8%)	6 (23.1%)
11–30 mm	6 (35.3%)	12 (46.2%)
31–50 mm	4 (23.5%)	6 (23.1%)
>50 mm	3 (17.6%)	0

.

All patients included had a T1 tumour in the first TURBT specimen. According to the WHO 2004/2016 grading system [1], high-grade tumours were present in all 17 patients (100%) of the reTURBT group and in 22 patients (84.6%) in the group without reTURBT, showing no significant difference (p = 0.140). Using the WHO 1973 grading system [1], grade three tumours predominated in both groups, accounting for 12 patients (70.6%) in the reTURBT group and 17 patients (65.4%) in the absence of reTURBT group, with no significant differences observed (p = 0.704). Additionally, concomitant carcinoma in situ (CIS) was similarly frequent in both groups, appearing in three patients (15%) in the reTURBT group and four patients (15.4%) in the group without reTURBT.

Detailed pathology outcomes from the initial TURBT are available in

Table 2. Pathology outcome of the first transurethral resection.

	ReTURBT N = 17 (39.5%)	Absence of ReTURBT N = 26 (60.5%)	p-Value
WHO 2004/2016, n (%)			0.140
Low grade	0	4 (15.4%)
High grade	17 (100%)	22 (84.6%)
WHO 1973, n (%)			0.704
G1	0	1 (3.8%)
G2	5 (29.4%)	8 (30.8%)
G3	12 (70.6%)	17 (65.4%)
Concomitant CIS, n (%)	3 (15%)	4 (15.4%)

.

The anatomy specimen of the second resection of the reTURBT group was also analyzed. In total, fourteen patients (82.4%) were a T0, two presented carcinoma in situ (11.8%) and one had a T1 tumour. No patient had a T2 tumour. All patients underwent a cystoscopy check after the induction course or 3 months after the reTURBT.

All patients received a BCG induction after TURBT. Also, all patients who had not presented a relapse had received a maintenance course of BCG. The median time to the first instillation was shorter in the no-reTURBT group, at 60 days vs. 139 days in the reTURBT group (p = 0.004). Data regarding complications is listed in

Table 3. Complications during the administration and time to the first instillation.

	ReTURBT N = 17 (39.5%)	Absence of ReTURBT N = 26 (60.5%)	p-Value
Complications, n (%)
Infection	3 (17.6%)	2 (7.7%)
Haematuria	0	1 (3.8%)
Not tolerate	1 (5.9%)	2 (7.7%)
Median time (days) to first instillation from first TURBT, median ± SD	139.22 ± 25.09	60 ± 11.92	0.004

.

The total number of recurrences in the no reTURBT group was four (15.4%), lower than in the reTURBT group where seven patients presented a relapse of the disease (43.8%). In

Table 4. Induction installations, complications during the administration and time to the first instillation.

	ReTURBT N = 17 (39.5%)	Months After ReTURBT	Absence of ReTURBT N = 26 (60.5%)	Months After TURBT
Relapse, n (%)	7 (43.8%)		4 (15.4%)
CIS	2	9 18	1	11
TaG1/LG	0	-	1	15
TaG2/LG	2	19 23	1	4
T1G2/HG	1	20	0	-
T1G3/HG	0	-	1	6
T2G3	2	9 12	0	-

a detailed summary of the anatomy and number of months after relapse of both groups can be consulted. The survival analysis using Cox regression showed differences between patients undergoing reTURBT and those who did not, although these differences were not statistically significant (Log-rank test, Mantel–Cox; Chi-square = 1.080, df = 1, p = 0.299). The mean survival time was higher in patients without reTURBT (620.57 ± 182.49 days; 95% CI: 262.89–978.25) compared to patients undergoing reTURBT (280.75 ± 138.96 days; 95% CI: 8.39–553.11). Similarly, median survival times were longer in the group without reTURBT (590 days; 95% CI: 20.29–1159.71) compared to the reTURBT group (150 days; 95% CI: 41.22–258.78). Despite this observed trend toward better survival in patients not undergoing reTURBT, the results indicate that performing or omitting the procedure did not lead to statistically significant differences in survival time within the observed period.

4. Discussion

This indication of reTURBT in the EAU guidelines is based on two meta-analyses. The first one performed by Cumberbatch et al. demonstrates a 51% risk of persistence and an 8% risk of under-staging in T1 tumours [8]. Naselli et al. identified that a persistent tumour was found in 58% and under-staging occurred in 11% of cases [9]. Recently a new metanalysis performed by Lin et al. focused on T1 tumours that showed a better RFS, cancer-specific survival (CSS) and progression-free survival (PFS) in the short term but no differences in the long term. The residual and upstaging rates of T1 in the second TURBT were around 50% and 10%, respectively [10].

The SARS-CoV-2 pandemic greatly affected specialized care and the number of surgeries performed in Spain [11]. Our hospital was not able to comply with the EAU’s recommendations, which is why the protocol described above was proposed [12].

ReTURBT in high-grade tumours offers the possibility of resecting unidentified tumours during the first resection and helps to establish the depth of the tumour [13]. The multicentre study by Soria et al. and one performed by Calo et al. opened the possibility that well selected patients, such as those with complete resection, muscle present in the anatomical specimen, absence of CIS and where en bloc resection is performed could omit the second resection [14,15]. Similar findings are presented by Raja et al. who claim that in procedures carried out in expert hands, the reTURBT could be omitted [16]. If the muscle layer is not present, reTURBT should be performed based on the study by Gontero et al. [17]. However, if the resection is complete and the muscle layer is present, the meta-analysis by Krajewski et al. supports that in such cases there is no need for reTURBT [18]. The conditions proposed in our study are along these lines. ReTURBT was omitted in very well selected patients to minimize the risk of understating. As a quality parameter of the resection and considering the data shown in Table 4, there were no early high-grade recurrences in the non-reTURBT group.

There are no head-to-head comparative studies of patients with and without reTURBT who have received an induction course of BCG. The studies report heterogeneous cycles and treatments when comparing the two groups. Early administration of BCG has been shown to reduce recurrence in patients with negative reTURBT [19]. Recent metanalyses report no improvement in long-term survival parameters with the reTURBT (RFS, PFS, CSS and overall survival). Lin et al.’s analysis does show improvement in RFS, PFS and CSS at a short follow-up time [10,18]. Our study follows along these lines. There were no differences in RFS and only two patients had progression in the reTURBT group. This may be due to the benefit of earlier BCG administration, which is significantly shorter than in the case of the reTURBT group. It is important to note that the time to BCG instillation was measured from the date of the initial TURBT in both cohorts. Therefore, the longer interval observed in the reTURBT group is expected, as it includes the time needed to perform and recover from the second resection. While early BCG initiation is desirable, the timing in the reTURBT group followed standard clinical practice and reflects a structured approach rather than a delay. Nevertheless, this temporal difference might still have influenced outcomes and should be considered when interpreting the results.

New molecular tests may change the indications for reTURBT in the future, adjusting the criteria for the procedure [20]. New techniques such as narrow-band imaging and photodynamic diagnosis can also help to perform more accurate resections [21]. Although narrow-band imaging and photodynamic diagnosis were not employed in this study, their relevance in the contemporary management of NMIBC is undeniable. Both technologies have demonstrated superior tumour detection rates and more complete resections compared to standard white light cystoscopy, particularly for flat lesions and carcinoma in situ [20,21]. Their use may help identify residual disease more precisely during the initial TURBT, potentially reducing the need for routine reTURBT in selected patients. Future studies should aim to incorporate these advanced imaging modalities, possibly in combination with molecular or inflammatory biomarkers, to refine risk stratification and optimize treatment pathways in NMIBC.

In recent years, there has been growing interest in incorporating molecular and inflammatory biomarkers into risk stratification models for NMIBC. A notable example is the study by Russo et al., which investigated the prognostic value of the Systemic Inflammation Response Index (SIRI) in patients undergoing radical cystectomy for urothelial carcinoma [22]. Their results demonstrated that elevated preoperative SIRI was significantly associated with adverse pathological features and poorer relapse-free survival, even when adjusted for standard clinical variables. Importantly, the inclusion of SIRI in prognostic models improved predictive accuracy and clinical net benefit, as shown through a decision curve analysis [22]. These findings highlight the potential utility of systemic biomarkers in refining patient selection for adjuvant treatments or alternative strategies such as reTURBT omission in selected low-inflammatory-risk patients. Another study performed by Russo et al. demonstrated that a high preoperative Systemic Immune-Inflammation Index (SII) was independently associated with advanced pathological stage, nodal invasion and worse recurrence-free and overall survival after radical cystectomy [23]. These findings reinforce the role of inflammatory biomarkers in risk stratification.

Although the non-reTURBT group showed lower rates of recurrence (15.4%) compared to the reTURBT group (43.8%), notable differences emerged regarding tumour progression. Two patients in the reTURBT group progressed to T2G3 tumours, while no such progression was observed in the non-reTURBT group. However, the non-reTURBT group did present a recurrence of T1G3/HG in one patient, highlighting potential risks associated with omitting reTURBT.

Interestingly, however, the recurrence rate was higher in the reTURBT group—a finding that contrasts with previous studies and meta-analyses supporting a protective role of reTURBT in recurrence and progression [10,13]. Several hypotheses may explain this discrepancy. First, selection bias could have influenced the results, as patients referred for reTURBT may have had more aggressive disease features at baseline, prompting clinicians to opt for closer monitoring or more conservative follow-up. Second, the group not undergoing reTURBT might have had more favourable tumour characteristics (e.g., complete initial resection, presence of detrusor muscle, less multifocality), even if not explicitly captured in baseline variables. Additionally, variations in the timing of BCG instillation and patient adherence during the COVID-19 period might have inadvertently favoured the non-reTURBT group.

Our study has several limitations. An important limitation of this study is the non-randomized assignment of patients to the reTURBT and non-reTURBT groups, which was dictated by hospital resource constraints during the COVID-19 pandemic. This introduces a potential selection bias. It is plausible that, in the context of reduced operating room availability and heightened clinical scrutiny, surgeons exercised increased diligence in ensuring complete resections during the initial TURBT. This heightened technical care, combined with stricter patient selection, could have conferred a prognostic advantage to the non-reTURBT group that is not fully captured by baseline clinical variables. As such, this procedural vigilance may partly explain the comparable outcomes observed despite the omission of reTURBT. While our inclusion criteria aimed to mitigate this bias by requiring muscle presence and complete resection, we acknowledge that the lack of randomization limits the generalizability and internal validity of our findings. The small sample size is a major limitation of our study. With only 43 patients included, the statistical power to detect significant differences in recurrence or progression is limited. Therefore, the absence of statistically significant differences between groups should not be interpreted as proof of equivalence, but rather as preliminary findings that require confirmation in larger studies. As mentioned, in new studies with homogeneous reTURBT, surveillance criteria should be conducted in a clinical trial setting.

5. Conclusions

Our exploratory study suggests that omitting reTURBT in well selected patients may not lead to significantly worse oncological outcomes. Importantly, the omission of reTURBT allowed earlier initiation of BCG treatment, which might have contributed to the favourable outcomes observed in the non-reTURBT group. Although our findings suggest potential safety in carefully selected patients, the small sample size must be acknowledged. These findings are hypothesis generating and should be validated with prospective trials with larger cohorts, and standardized inclusion criteria are necessary to confirm these preliminary findings, given that our study was underpowered to detect subtle but clinically relevant differences.