Limited, Extended, Superextended, Megaextended Pelvic Lymph Node Dissection at the Time of Radical Cystectomy: What Should We Perform?

By: and Alexandre R. Zlottaa b

Published online: 01 February 2012

Abstract Full Text Full Text PDF (82 KB)

Refers to article:

Clinical Nodal Staging Scores for Bladder Cancer: A Proposal for Preoperative Risk Assessment

Shahrokh F. Shariat, Behfar Ehdaie, Michael Rink, Eugene K. Cha, Robert S. Svatek, Thomas F. Chromecki, Harun Fajkovic, Giacomo Novara, Scott G. David, Siamak Daneshmand, Yves Fradet, Yair Lotan, Arthur I. Sagalowsky, Thomas Clozel, Patrick J. Bastian, Wassim Kassouf, Hans-Martin Fritsche, Maximilian Burger, Jonathan I. Izawa, Derya Tilki, Firas Abdollah, Felix K. Chun, Guru Sonpavde, Pierre I. Karakiewicz, Douglas S. Scherr and Mithat Gonen

Accepted 12 October 2011

February 2012 (Vol. 61, Issue 2, pages 237 - 242)

Lymph node (LN) staging unequivocally has prognostic implications in bladder cancer patients treated with radical cystectomy (RC). The histologic status of regional LNs is one of the most important predictors of survival, as demonstrated in countless studies. RC patients can be stratified into risk groups according to tumour stage, LN involvement, number of metastatic nodes, and LN density [1]. The number of LNs retrieved during pelvic LN dissection (PLND) is quite variable, and different cut-off numbers of LNs that should be dissected have been suggested. So far, there is no consensus regarding the standard number of LNs that should be obtained in patients with bladder cancer operated by RC.

Several studies have reported that extended PLND (ePLND; whatever that actually really means and how one defines it) decreases local recurrence and improves cure rates compared to limited and standard PLND [2]. Despite some evidence, there is no universally accepted metric of surgical quality for PLND that predicts survival in all bladder cancer patients undergoing RC [3]. Some authors have suggested that the LN technique is more important than the node count [4]. Whether the LN count does not represent merely another surrogate marker for the quality of the RC itself is also worth investigating.

Recent oncologic outcomes at two well-known and respected cystectomy centres where two different ePLND templates are practiced were compared to determine whether removing lymphatic tissue up to the inferior mesenteric artery confers an additional survival advantage [5]. Demographic data and pathologic subgroup distribution (pT2 and pT3) were similar in 554 University of Southern California (USC) patients and 405 University of Bern patients. USC patients had a higher median number of LNs removed than University of Bern patients (38 vs 22, p<0.0001) and a higher incidence of LN metastasis (35% vs 28%, p=0.02). However, the USC and University of Bern groups had similar 5-yr recurrence-free survival for pT2pN0–2 disease (57% vs 67%, p=0.55) and pT3pN0–2 disease (32% vs 34%, p=0.44). The overall recurrence rate was equal at the two institutions (38%). Meticulous ePLND up to the middle upper third of the common iliac vessels appeared to provide survival and recurrence outcomes similar to those of a superextended template up to the inferior mesenteric artery.

In the absence of randomised data, the jury is still out regarding the true survival benefit of PLND, ePLND, or even superextended PLND. A randomised clinical trial designed to test this question in invasive bladder cancer has been ongoing in Germany, with the target enrolment of 400 patients nearing completion. The primary end point of the trial is a 65% progression-free survival rate in patients who undergo ePLND compared with 50% in patients who undergo standard dissection with a follow-up of 5 yr. This study is closed and is currently being analysed. The first data are to be expected in about 1 yr (; pers. comm., A. Heidenreich, Aachen, Germany).

Investigators in the Southwest Oncology Group have also proposed a randomised trial comparing standard versus extended node dissection. The primary end point of the trial, which would include 315 patients in each arm, would be progression-free survival at 3 yr, with an increase of 10% considered clinically meaningful.

In this paper in European Urology, the authors propose a risk calculator allowing preoperative determination of the minimum number of LNs to avoid pN understaging [6]. Based on assessment of a very large series of 4335 patients, mostly from university centres, who underwent RC and PLND, the authors conclude that a 90% correct node staging can be obtained by removing 6, 9, and 25 LNs in stage Ta/TIS, T1, and T2 bladder cancer, respectively.

As one would anticipate, the risk of LN positivity increases with stage, and the more nodes one removes in higher stages, the better the chance of catching positive ones. More positive nodes are found when more nodes are examined: It is reassuring to observe that even with a very large retrospective database, logic is preserved.

Whether or not a minimum of 25 nodes is truly necessary in cT2 is a matter of debate; unfortunately, issues as simple as processing (even the interest and zeal of the technician in charge matter) may dramatically affect this number. The major limitation can be found in the retrospective nature of the evaluation over a long period of time (starting in 1980). The data about clinical staging also seem to be rather unreliable. It might have been interesting to stratify this cohort by decade and to analyse data accordingly, as one would anticipate that some of the different outcomes might have been driven by earlier procedures. It is also a pity that survival data are not provided. Unfortunately, the most important issue, namely, the relation of PLND to survival and outcome, is not addressed. At the end of the day, this is probably what really matters.

Working myself in two institutions located, literally, next to each other and connected by a 10-m-long bridge, with the same uropathologist reviewing the cases, with the same surgeon, I have always been puzzled by the difference in node counts when performing an identical procedure. Every single thing, from the surgeon to the pathologist, is the same, but the technicians are different. This difference has as a consequence in that the node count may often be 50% higher when reported in one institution compared to the other. Amazing, really amazing. Suffice it to say that this is a type of bias (flying under the radar for many of us) that one has to keep in mind when eventually reporting results or attempting to translate data observed in one institution to another.

It is also somehow difficult to imagine the clinical applicability of the model proposed.

The authors [6] combined a lot of data spread out from many centres, from two continents, from many countries, and with an impressive number of different surgeons and pathologists. It goes without saying that this combination implies a very heterogeneous dataset. Considering the enormous variation in patients and preoperative imaging (if any), surgical techniques (“the extent of the LND was at the surgeon's discretion”), surgeons, pathologic preparation, and pathologic assessment, the results of such a study can be significant only because of the large amount of patients. Another significant limitation of the paper is the fact that the locations of the LN samples could not be defined because the information required was not available.

The authors and all other centres scientifically analysing the value and benefit of PLND should be encouraged to look at their own data and to compare them with centres where a different technique is practised. This would make sense, as long as positive margin rates and disease-specific survival appear similar in cancer stages less likely to be influenced by LN metastasis risk (less than pT2 disease).

Even while acknowledging some limitations inherent to randomised trials (many patients simply do not fit into the inclusion criteria), we clearly need prospective or comparative studies rather than retrospective data to address this crucial question. To make things worse, even when level 1 evidence is available, our specialty has been notoriously slow to implement it into clinical practice for various reasons [7].

Conflicts of interest

The author has nothing to disclose.


  • [1] M. May, E. Herrmann, C. Bolenz, et al. Lymph node density affects cancer-specific survival in patients with lymph node-positive urothelial bladder cancer following radical cystectomy. Eur Urol. 2011;59:712-718 Abstract, Full-text, PDF, Crossref.
  • [2] H. Abol-Enein, D. Tilki, A. Mosbah, et al. Does the extent of lymphadenectomy in radical cystectomy for bladder cancer influence disease-free survival? A prospective single-center study. Eur Urol. 2011;60:572-577 Abstract, Full-text, PDF, Crossref.
  • [3] S.P. Kim, R.J. Karnes. Standard or extended lymphadenectomy in radical cystectomy: what metric of surgical quality for lymph node dissections matters in improving patient outcomes?. Eur Urol. 2011;60:953-954 Abstract, Full-text, PDF, Crossref.
  • [4] R.P. Dorin, S. Daneshmand, M.S. Eisenberg, et al. Lymph node dissection technique is more important than lymph node count in identifying nodal metastases in radical cystectomy patients: a comparative mapping study. Eur Urol. 2011;60:946-952 Abstract, Full-text, PDF, Crossref.
  • [5] P. Zehnder, U.E. Studer, E.C. Skinner, et al. Super extended versus extended pelvic lymph node dissection in patients undergoing radical cystectomy for bladder cancer: a comparative study. J Urol. 2011;186:1261-1268 Crossref.
  • [6] S.F. Shariat, B. Ehdaie, M. Rink, et al. Clinical nodal staging scores for bladder cancer: a proposal for preoperative risk assessment. Eur Urol. 2012;61:237-242 Abstract, Full-text, PDF, Crossref.
  • [7] A. Feifer, J.M. Taylor, M. Shouery, et al. Multi-institutional quality-of-care initiative for nonmetastatic, muscle-invasive, transitional cell carcinoma of the bladder: phase I. J Clin Oncol. 2011;29(Suppl 7) Abstract 240


a Mount Sinai Hospital, Murray Koffler Urologic Wellness Centre, Joseph and Wolf Lebovic Building, 60 Murray Street, 6th Floor, Box 19, Toronto, Ontario M5T 3L9, Canada

b University Health Network, Princess Margaret Hospital, Department of Surgical Oncology, Urology, 610 University Ave. Toronto, Ontario M5G 2M9, Canada