SEARCH

SEARCH BY CITATION

Keywords:

  • nephroureterectomy;
  • robot;
  • upper urinary tract urothelial carcinoma;
  • laparoendoscopic single-site (LESS) surgery

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Objective

  • To compare the peri-operative, pathological and oncological outcomes of laparoendoscopic single-site (LESS) robot-assisted nephroureterectomy (LESS-RALNU) with those of multiport robot-asssisted nephroureterectomy (M-RALNU).

Patients and Methods

  • A total of 38 patients with upper urinary tract urothelial carcinoma underwent LESS-RALNU (n = 17) or M-RALNU (n = 21) by a single surgeon at a tertiary institution.
  • Data were obtained from a prospectively maintained database.

Results

  • Patients' demographics and tumour characteristics were similar between the M-RALNU and LESS-RALNU groups.
  • The mean follow-up was 48.4 months for M-RALNU and 30.9 months for LESS-RALNU (P = 0). The mean operating time, estimated blood loss and length of hospitalization for M-RALNU and LESS-RALNU were 251 min, 192 mL, 6.5 days and 247 min, 376 mL and 5.4 days, respectively (P > 0.05).
  • Overall, there were no significant differences in complication rates, although three patients in the LESS-RALNU group required blood transfusion, whereas no patient in the M-RALNU group did (P = 0.081).
  • The proportion of patients with bladder recurrence, local recurrence and distant metastases was similar between the two groups.
  • There were no significant differences in the recurrence-free survival, cancer-specific survival and overall survival rates between the two groups.

Conclusions

  • Although the oncological and peri-operative outcomes of patients who underwent LESS-RALNU compared well with those who underwent M-RALNU and with series of other surgical approaches, LESS-RALNU might result in greater intra-operative blood loss.
  • We suggest careful selection of patient for this technique.

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

The surgical treatment of upper urinary tract urothelial carcinoma (UUT-UC) is evolving. The traditional treatment is open radical nephroureterectomy (ONU) with the en bloc excision of the ipsilateral bladder cuff, and this remains the ‘gold standard’, but since Clayman et al. [1] reported the first successful laparoscopic nephroureterectomy (LNU) in 1991, minimally invasive approaches have begun to slowly establish themselves as feasible alternatives to ONU, offering the additional benefits of smaller incision wounds, lower medication use, shorter length of hospital stay (LOH) and shorter convalescent time [2]. With increasing experience in laparoscopic surgeries, bold efforts were made to further reduce morbidities and improve cosmesis. Raman et al. [3] were the first to report their experiences in laparoendoscopic single-site (LESS) urological surgeries in 2007. They argued that, in addition to the potential cosmetic benefits of this ‘scarless’ surgery, port-related complications such as pain and incisional hernias could be reduced with a reduction in the number of ports needed. LESS surgery in general remained a technically challenging procedure because of the inline camera visualization, restricted triangulation of laparoscopic instruments and limited retraction of tissues; but these difficulties were compounded in LESS-nephroureterectomy as a result of the dual-site involvement (abdomen and pelvis), the restricted working space in the pelvis and the technically challenging management of the bladder cuff. Indeed, there have been very few series of LESS nephrouretectomy reported to date [4-7]. The introduction of the robotic platform could alleviate some of these difficulties, which should make the procedure less challenging and in turn achieve better outcomes.

As with all oncological surgeries, oncological clearance and patient safety must remain the surgeon's top priority and must never be compromised. To date, there have been few published series on robot-assisted laparoscopic nephroureterectomy (RALNU), and even fewer on LESS-RALNU [8, 9]. In the present study, we aimed to compare for the first time, the peri-operative, pathological and oncological outcomes between LESS-RALNU and multi-port RALNU (M-RALNU).

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Patient Selection

From May 2007 to June 2012, 41 consecutive adult patients with UUT-UC underwent RALNU by a single surgeon (K.H.R) via either M-RALNU or LESS-RALNU approaches. The choice of approach was based solely on the surgeon's preference during the study period. Only M-RALNU was performed before June 2009. From July 2009 to June 2010, only LESS-RALNU was performed. After July 2010, distal ureteric tumours were preferentially managed with M-RALNU, while mid and upper ureteric tumours were managed with LESS-RALNU as we believed that M-RALNU allowed a more meticulous pelvic lymph node dissection (PLND) and bladder resection and reconstruction. Approval was obtained from our institutional review board. Data were obtained from a prospectively maintained database. Indications for surgery were patients aged ≥18 years, who had localized UUT-UC on preoperative staging. Patients with non-TCC, clinical ≥T3 disease, positive regional lymph nodes and metastatic disease on preoperative staging were excluded. A total of 38 patients met our study criteria. None of the patients had neoadjuvant chemotherapy before surgery. Informed consent was obtained from all patients before surgery. All patients underwent preoperative triphasic CT imaging for staging.

Operating Technique

Our M-RALNU and LESS-RALNU port placement and surgical techniques have been detailed in our previous publications [8, 10]. All procedures were carried out with the da Vinci S system (Intuitive Surgical, Sunnyvale, CA, USA). For LESS-RALNU, a homemade port was assembled using an Alexis wound retractor (Applied Medical, Rancho Santa Margarita, CA, USA) and a common size 7 surgical glove, applied over the external side of the Alexis wound retractor. Four trocars, including two 12-mm trocars (the third and fourth fingers) and two 8-mm robotic trocars (the first and fifth fingers), were inserted and secured through incisions on the fingertips of the surgical glove. For right-side cases (n = 6), an additional 5-mm port was inserted solely for the purpose of liver retraction. Our management of the distal ureter and bladder cuff mimicked those of the open approach in all cases. The bladder cuff was dissected circumferentially around the ureteric orifice and resected by the extravesical manoeuvre en bloc in all 38 cases. The bladder defect was then closed in two layers. For the first six cases in the M-RALNU group, the patient was repositioned from flank to lithotomy and the robot was redocked for excision of the distal ureter and bladder cuff. No patient repositioning, redocking of the robot or movement of the patient cart were necessary for the subsequent 15 patients in the M-RALNU group, after the use of the hybrid port technique as described in our previous publication [10]. This hybrid port technique allows easy reconfiguration of the robotic arms during the transition from the nephrectomy phase to the bladder cuff dissection phase of the procedure. No redocking of the robot was necessary for any of the patients in the LESS-RALNU group. All specimens were retrieved en bloc in an entrapment bag via the umbilical incision (extension of the umbilical incision was needed for M-RALNU). Ipsilateral PLND was performed for distal ureteric tumours. The template for PLND included the obturator and deep obturator nodes, and the external and internal iliac nodes, extending up to crossing of the ureter. Retroperitoneal PLND (RPLND) was not performed for any of the 38 cases. Intra-operative and postoperative complications were recorded and classified according to the Clavien–Dindo systems.

Follow-Up Regime

Patients were reviewed 1 month postoperatively with a complete blood count, serum electrolytes and creatinine tests and CT imaging. Patients with positive surgical margins, pathological T3 tumours or non-vesical recurrences were counselled for adjuvant therapies. Subsequent follow-ups were scheduled at 3rd, 6th, 9th and 12th months after surgery, with repeat CT imaging, urine analysis and urine cytology at each visit. Cystoscopy was performed if patients were symptomatic or if investigations were abnormal. Subsequent visits were 6-monthly for the next 2 years and annually thereafter.

Statistical Analysis

We compared patients' demographics, tumour characteristics, pathological features and oncological outcomes in patients who underwent M-RALNU or LESS-RALNU using Student's t-test or the Mann–Whitney U-test for continuous variables and the Pearson chi-squared test for categorical variables. The Kaplan–Meier method was used to estimate the recurrence-free survival (RFS), cancer-specific survival (CSS) and overall survival (OS) rates; and survival curves were compared using the log-rank test. Statistical analysis was conducted using IBM spss statistics software version 20.0; A P-value <0.05, from the use of two-sided statistical tests, was considered to indicate statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Table 1 shows the patient demographics and tumour characteristics of the whole study cohort and the comparison between the M-RALNU and LESS-RALNU groups. A total of 21 patients (55.3%) underwent M-RALNU, while 17 (44.7%) underwent LESS-RALNU. The majority of LESS-RALNU procedures were performed between 2009 and 2010 (76.5%). There were no significant differences in patients' demographics and tumour characteristics between the two groups except for age; patients who underwent LESS-RALNU were significantly older (P = 0.030).

Table 1. Comparison of demographics and tumour characteristics of patients in the M-RALNU and LESS-RALNU groups
 Cohort N = 38M-RALNU n = 21LESS-RALNU n = 17P
  1. ASA, American Society of Anesthesiologists.

Median (range) age, years66.5 (43–83)62 (43–83)71 (44–83)0.030
Gender, n (%)    
Male26 (68.4)15 (71.4)11 (64.7)0.734
Female12 (31.6)6 (28.6)6 (35.3)
Mean (se) body mass index, kg/m223.0 (0.5)22.5 (0.6)23.6 (0.7)0.272
ASA score, n (%)    
115 (39.5)7 (33.3)8 (47.1)0.655
221 (55.3)13 (61.9)8 (47.1)
32 (5.2)1 (4.8)1 (5.8)
40 (0.0)0 (0.0)0 (0.0)
Tumour side, n (%)    
Left25 (65.8)14 (66.7)11 (64.7)0.919
Right13 (34.2)7 (33.3)6 (35.3)
Tumour site, n (%)    
Pelvicalyceal17 (44.7)10 (47.6)7 (41.2)0.665
Pelvicalyceal-ureteral6 (15.8)4 (19.1)2 (11.7)
Ureteral15 (39.5)7 (33.3)8 (47.1)
Clinical tumour stage, n (%)    
T121 (55.3)13 (61.9)8 (47.1)0.513
T217 (44.7)8 (38.1)9 (52.9)
T30 (0.0)0 (0.0)0 (0.0)
T40 (0.0)0 (0.0)0 (0.0)
Year of surgery, n (%)    
20075 (13.2)5 (23.8)0 (0.0) 
20088 (21.1)8 (38.1)0 (0.0)
200912 (31.6)5 (23.8)7 (41.2)
20107 (18.4)1 (4.8)6 (35.3)
20115 (13.2)2 (9.5)3 (17.6)
20121 (2.5)0 (0.0)1 (5.9)

Table 2 shows the peri-operative, pathological and oncological outcomes of the whole study cohort and of the two groups. No patient required open conversion for the LESS approach or conversion to conventional M-RALNU. There were no significant differences in the mean operating times between the two groups. Overall, there were no significant differences in the total number of complications or number of complications in each grade category, even though there seemed to be a trend towards higher complication rates in the LESS-RALNU group. This was because three patients in the LESS-RALNU group required blood transfusion, while none of the patients in the M-RALNU group did (P = 0.081). The mean intra-operative blood loss was also higher in the LESS-RALNU group, although it did not reach significance (376 vs 192 mL, P = 0.177). Intra-operative complications occurred in one patient in the M-RALNU group. In this case, the small bowel was injured during dissection and was repaired primarily. No intra-operative complications were seen in the LESS-RALNU group. Postoperative complications developed in 10 patients (25.0%), including ileus in five patients (LESS-RALNU, n = 3; M-RALNU, n = 2), which resolved with conservative management, a non-fatal myocardial event in one patient (M-RALNU), which was treated with medication, postoperative blood transfusion in three patients (all LESS-RALNU), and the development of a small intra-abdominal abscess in one patient (LESS-RALNU), necessitating insertion of percutaneous drainage.

Table 2. Comparison of peri-operative, pathological and oncological outcomes of M-RALNU and LESS-RALNU
 Cohort, N = 38M-RALNU, n = 21LESS-RALNU, n = 17P
  1. *Clavien–Dindo classification.

Mean (se) preoperative serum creatinine,mg/dL1.33 (0.15)1.41 (0.22)1.24 (0.19)0.590
Mean (se) postoperative serum creatinine, mg/dL1.52 (0.17)1.67 (0.26)1.35 (0.19)0.355
Mean (se) EBL, mL264 (67)192 (46)376 (137)0.177
Mean (se) operating time, min249 (8)251 (11)247 (14)0.831
Number converted to open000 
Complications*, n (%)11 (28.9)4 (19.0)7 (41.2) 
Grade 15 (13.2)2 (9.5)3 (17.6)0.173
Grade 24 (10.5)1 (4.8)3 (17.6)
Grade 32 (5.3)1 (4.8)1 (5.9)
Grade ≥40 (0.0)0 (0.0)0 (0.0)
Blood transfusion, n (%)3 (7.9)0 (0.0)3 (17.6)0.081
Mean (se) LOH, days6.0 (0.6)6.5 (0.7)5.4 (0.9)0.321
Mean (range) follow-up duration months40.6 (6–65)48.4 (15–65)30.9 (6–41)0.000
Mean (se) pathological tumour size, cm2.89 (0.24)3.10 (0.40)2.72 (0.28)0.432
Pathological tumour stage, n (%)    
Ta4 (10.5)4 (19.0)0 (0.0)0.184
T111 (28.9)4 (19.0)7 (41.2)
T27 (18.5)4 (19.0)3 (17.6)
T316 (42.1)9 (42.9)7 (41.2)
T40 (0.0)0 (0.0)0 (0.0)
Pathological tumour grade, n (%)    
Low grade7 (18.4)4 (19.0)3 (17.6)0.973
High grade31 (81.6)17 (81.0)14 (82.4)
Carcinoma in situ, n (%)2 (5.3)1 (4.8)1 (5.9)1.000
Multifocal diseases, n (%)    
Yes8 (21.1)5 (23.8)3 (17.6)0.709
No30 (78.9)16 (76.2)14 (82.4)
Positive surgical margins, n (%)1 (2.5)0 (0.0)1 (5.9)0.447
Ipsilateral PLND performed, n (%)11 (29.0)9 (42.9)2 (11.8)0.070
Lymph nodes yield from ipsilateral PLND7.27.84.5 
Local recurrence, n (%)4 (10.5)4 (19.0)0 (0.0)0.113
Bladder recurrence, n (%)10 (26.3)4 (19.0)6 (35.3)0.293
Distant metastases, n (%)11 (28.9)7 (33.3)4 (23.5)0.486

All 38 patients had pathologically confirmed UUT-UC. High grade lesions were present in 31 patients (81.6%), multifocal diseases were noted in eight (21.1%) and positive surgical margins in one (2.6%) from the LESS-RALNU group. The majority (42.1%) of the tumours were T3 disease on pathological staging (Table 2). There were no significant differences in the pathological characteristics between the two surgical groups. The mean (range) follow-up was 40.6 (2–65) months in the study cohort; however, follow-up durations were significantly longer for patients who underwent M-RALNU (P = 0) because we only started performing LESS-RALNU after July 2009.

Of the 38 patients, 11 (28.9%) died, including seven (18.4%) cancer-related deaths (LESS-RALNU, n = 2; M-RALNU, n = 5). Recurrences developed in 17 (44.7%) patients (LESS-RALNU, n = 7; M-RALNU, n = 10) at a median (range) of 8 (1–23) months. Of the recurrences, 10 (26.3%) were urothelial (LESS-RALNU, n = 6; M-RALNU, n = 4) at a median (range) of 8.0 (5–12) months and 11 (28.9%) were non-urothelial (LESS-RALNU, n = 4; M-RALNU, n = 7) at a median of 11 (1–28) months. Of the four local recurrences, two patients had renal bed recurrences, one had a retroperitoneal recurrence and one had pelvic recurrence. Four patients underwent adjuvant systemic chemotherapy and 10 underwent salvage radiotherapy and/or chemotherapy at recurrence.

Figures 1 and 2 show the comparison of survival curves between patients who underwent M-RALNU and those who underwent LESS-RALNU. There were no significant differences in the RFS rates, including either urothelial or non-urothelial RFS rates, between the two groups (Fig. 1A–C). Similarly, the CSS and OS rates were similar between the two surgical groups (Fig. 2A and B).

figure

Figure 1. Comparison of RFS rates between the M-RALNU and LESS-RALNU groups. (A) RFS. (B) Urothelial RFS. (C) Non-urothelial RFS.

Download figure to PowerPoint

figure

Figure 2. Comparison of (A) CSS and (B) OS between patients who underwent M-RALNU and LESS-RALNU.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

Although UUT-UC is uncommon and accounts for only 5% of all urothelial malignancy, it is often aggressive, necessitating radical treatment. Despite this treatment, the Upper Tract Urothelial Carcinoma Collaboration group reported the 5-year CSS after RNU to be only 73% [11]. As with all oncological surgeries, oncological outcomes after radical nephroureterectomy are of paramount importance. A recent meta-analysis comparing ONU and LNU reported that the bladder recurrence, local recurrence and distant metastases rates of LNU were 19–43, 4–15 and 10–28%, respectively, all of which were not significantly different from open series [12]. RALNU is still in its infancy, and current available studies are limited to safety and feasibility studies. A recent review of all available RALNU studies reported bladder recurrence and local recurrence rates of 0–44 and 2.4% [13], which could be deemed similar to open or LNU series. In the present study, we observed bladder recurrence, local recurrence and distant metastases rates of 26.3, 10.5 and 28.9%, respectively, and positive surgical margins in one patient, results that seemed to be equivalent to the existing literature [12]. In particular, there were no significant differences in the bladder, local and metastases rates between the M-RALNU and the LESS-RALNU groups.

Pure laparoscopic LESS-nephroureterectomy is technically very challenging and is carried out in very few centres worldwide. Management of the distal ureter and bladder cuff is especially difficult, both laparoscopically and endoscopically. There is still a paucity of evidence regarding LESS-nephroureterectomy to date, with most published studies consisting of very small numbers. Desai et al. [4] described two cases of LESS-nephroureterectomy in their series of 100 patients who underwent LESS procedures, one for UUT-UC and the other for severe reflux and recurrent pyelonephritis [4]. Operating time ranged from 90 to 200 min, estimated blood loss (EBL) ranged from 75 to 300 mL and LOH ranged from 1 to 5 days. One case needed insertion of an additional 5-mm port. White et al. [5] also performed seven LESS-nephroureterectomies out of 100 LESS procedures, and two of these cases were robot-assisted [5]. They reported a mean tumour size of 2.73 cm, EBL of 396 mL, operating time of 198 min and LOH of 3.9 days, and one case required conversion to standard laparoscopy. Lee et al. [7] reported in their series of 10 patients a mean operating time of 226 min, EBL of 188 mL and LOH of 4.8 days. Two cases required conversion to open surgeries. Khanna et al. [9] performed three LESS-RALNU procedures and reported an operating time of 300 min, EBL of 183 mL and one conversion to the standard laparoscopic approach. In the largest series of LESS-RALNU procedures published to date, consisting of 101 patients from various international centres, the mean operating time was 221 min, EBL was 232 mL and LOH was 6.3 days [14]. In that series, there was no bladder cuff excision in 20.8% of the cases and the disease recurrence rate was 22.8%, with a mean time to recurrence of 11.5 months. In the present series of 18 LESS-RALNU procedures, the mean operating time was 244 min, EBL was 358 mL and LOH was 5.3 days. These results were largely similar to the above-mentioned LESS-nephroureterectomy studies, despite the initial learning curve. Our LOH was seemingly longer than reported in the literature. The main reason for this is more social and cultural than medical; because of the extensive health insurance coverage with little out-of-own pocket expenses during their hospital stay, most Korean patients prefer to remain in hospital until they are almost independent in their daily living activities, even though the majority of our patients were medically fit for discharge on postoperative day 1 or 2. Compared with the disease recurrence rate of 22.8% in the multi-intuitional study by Park et al. [14], our LESS-RALNU disease recurrence of rate of 41% might seem higher; however, the majority of the patients in the former study had pathological stage ≤T1disease (55.4%) as opposed to the present study where 41.2% of patients had pT3 disease. The nonsignificantly higher EBL in the LESS-RALNU group as compared with the M-RALNU group might be reflective of the increased technical difficulty of the procedure. In our experience, the dissection of the distal ureter and bladder cuff and PLND were technically more challenging via the LESS approach and might have resulted in the increased EBL observed.

We had no conversions to the standard laparoscopic or open approaches, suggesting that the use of the robotic system could have made the surgery more manageable. Superior ergonomics, three-dimensional magnification of the operative field, tremor filtering and the enhanced degrees of freedom afforded by the EndoWrist (Intuitive Surgical, Sunnyvale, CA, USA) instruments are several advantages of the da Vinci Surgical System over conventional laparoscopy; however, limitations such as intracorporeal space constraints, external collision of the robotic arms and limited triangulation of instruments may still pose significant restrictions. Joseph et al. [15] presented a novel ‘chopstick’ technique that allowed the use of the current da Vinci system through a single incision without external collision. To maximize the triangulation of instruments and minimize robotic arm collisions externally, several modifications were made to our LESS technique. These included changing the system configuration to fine-tuning mode should there be excessive collisions between the robotic arms, using a 30-degrees upwards scope to reduce instrument clashing intracorporeally, making our incision ∼4–5 cm long with a larger fascia incision of 5–6 cm long, and finally configuring the ports as far apart as possible from one another in our homemade single-port device. Finally, several technological breakthroughs in the robotic system that may overcome these limitations are in the pipeline. Robotic instruments dedicated to LESS, flexible robots and miniature in vivo robots are some advancements that are on the horizon [16, 17] and are likely to change the current landscape of LESS surgery.

When compared with conventional laparoscopy, currently no clear benefits on postoperative course and patient convalescence have been definitively proven in LESS surgery. The only potential benefit of LESS surgery might be improvement in cosmetic outcomes [18, 19]. Previously, this claimed improvement in cosmesis largely entailed subjective assessment by the surgeon [4-6, 18, 20], but studies of patients' objective assessment of cosmetic outcomes are accumulating. Park et al. [21] showed that scar satisfaction after LESS surgery is higher than that for alternative surgical approaches and 86% of patients who underwent LESS surgery would undergo future LESS surgery at equivalent risk, compared with 20% who would choose open surgery in future after having undergone the same. Nevertheless, when the theoretical risk of LESS surgery was increased, a decreased preference for it resulted, whereas preference for laparoscopy and open surgery increased. Indeed, Golkar et al. [22] reported that patients' utmost concern was safety and, although LNU could be performed with several smaller incisions, intact specimen retrieval would require extension of one of the smaller incisions, negating any potential benefits of multiple smaller incisions over a single larger LESS incision. In fact, a study on morbidly obese patients undergoing LESS gastric banding reported minimal wound-related complications such as infections and incisional hernias after a minimum 2-year follow-up [23], even in this high-risk group of patients, reinforcing the notion that wound-related morbidity of LESS surgeries is marginal. In the present study, there were no wound-related complications in our LESS-RALNU cohort. We have shown in the present study that LESS-RALNU could potentially improve cosmetic outcomes with similar complication rates and peri-operative and oncological outcomes to those of M-RALNU, although the small sample size may limit definitive conclusion.

The rationale for PLND in UUT-UC is mainly extrapolated from evidence of the therapeutic benefits of PLND for bladder cancer [24], but the true role of RPLND in UUT-UC is controversial. While some authors have recommended RPLND in higher stage tumours [25-27], others have failed to show any survival benefits associated with RPLND [28-31]. In our previous series of 152 patients with UUT-UC treated with RNU, of whom 63 underwent RPLND, we found no difference in RFS and CSS between those who underwent RPLND and those who did not [31]. In the largest of these studies, Lughezzani et al. [28] compared the survival outcomes of 2824 patients with pN0 vs pNx nodal status after radical nephreoureterectomy for UUT-UC. There were no differences in CSS between pN0 and pNx patients at 5 years. There were also no differences in CSS between the two groups after a subgroup analysis of patients with muscle-invasive UUT-UC. RPLND is associated with significant morbidities. In the present study, we included only patients with clinical ≤T2 tumours. Kondo et al. [32] reported that the incidence of lymphatic involvement for Tis-1, T2, T3 and T4 tumours were 0, 5, 24 and 84%, respectively. We believed that lymph node metastases were indicative of systemic metastases and performance of RPLND would be unlikely to improve survival. Rather than subjecting these patients with clinical ≤T2 tumours to the possible complications of RPLND, we think that these patients would be better managed with systemic chemotherapy. Some authors argue that RPLND may aid in prognostication of disease for adjuvant chemotherapy [26], but there is no level 1 evidence, to date, to show that adjuvant chemotherapy improves RFS or CSS rates. In fact, studies have shown that adjuvant chemotherapy did not confer additional OS or CSS benefits for T3 or pN+ tumours [33, 34]. Indeed, a recent review concluded that there seemed to be no consistent benefit in outcomes when comparing patients who either did or did not undergo regional lymphadenectomy although regional lymphadenectomy allowed the accurate staging of UUT-UC. The benefits of adjuvant systemic therapies for UUT-UC remain unproven [24]. Nevertheless, the lack of nodal status in the present study remained a major limitation.

Despite limiting our patient cohort to only those with ≤cT2 tumours, almost 50% of our patients were eventually found to have pT3 tumours. This result is reflective of the inadequacy of the current preoperative staging. Studies have shown that accuracy in clinical staging for preoperative CT imaging was only ∼50% [35, 36]. Because of the tiny sample sizes from cold-cup biopsies, accurate T-staging of tumours > T1 is almost impossible. In a study where 39 tumours were biopsied, none of the biopsy specimens had muscle identified, making diagnosis of ≥T2 tumours impossible preoperatively [37].

The present study has several other limitations, in addition to the inherent biases of retrospective studies and small sample sizes. Firstly, the patients were not randomized, but the baseline characteristics between the two surgical approaches were not significantly different. Secondly, although our findings were comparable with published open, laparoscopic and LESS nephroureterectomy series, the fact that this was a single-surgeon study may limit its extrapolation and the results need further external validation. Thirdly, we needed an additional 5-mm port for right-sided cases. Although purists will claim that this cannot be considered LESS in the strictest sense, the addition of an extra port has been documented in many LESS series [4-6]. In fact, in the consensus statement of the consortium for LESS surgery, the use of an additional 2-mm port and needle instruments were allowed [19]; however in our study, we found that needle instruments could not provide adequate liver retraction and hence a 5-mm port was inserted instead for patient safety. Despite these limitations, the present study, representing the largest and the only comparative M-RALNU and LESS-RALNU cohort, will provide valuable reference point for future comparative or descriptive studies of a similar nature.

In conclusion, the intermediate-term oncological outcomes and peri-operative outcomes of patients who underwent LESS-RALNU compared well, not only with M-RALNU but also with those in other RALNU, LNU and ONU series; however, LESS-RALNU is technically more challenging to perform and may result in more blood loss and higher blood transfusion rates. We suggest careful selection of patients for LESS-RALNU and, in more complex cases or in patients with lower ureteric tumours, M-RALNU should be performed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References

This study was supported by a faculty research grant from Yonsei University College of Medicine for 2012 (6-2012-0181).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of Interest
  9. References
Abbreviations
LESS

laparoendoscopic single-site

UUT-UC

upper urinary tract urothelial carcinoma

ONU

open radical nephroureterectomy

LNU

laparoscopic nephroureterectomy

RALNU

robot-asssisted nephroureterectomy

M-RALNU

multiport robot-asssisted nephroureterectomy

PLND

pelvic lymph node dissection

RPLND

retroperitoneal lymph node dissection

RFS

recurrence-free survival

CSS

cancer specific survival

OS

overall survival

EBL

estimated blood loss

LOH

length of hospital stay