Lymph node retrieval rates in melanoma: a quality assessment parameter

Short Communication

Lymph node retrieval rates in melanoma: a quality assessment parameter

D. Berger-Richardson, MD CM MSc*, E. Cordeiro, MD MSc, M. Ernjakovic, MD*, A.M. Easson, MD MSc*




Regional lymph node dissection (rlnd) for melanoma with nodal metastasis is a specialized procedure that is associated with improved disease-specific survival in selected patients. Furthermore, there is evidence that a higher lymph node retrieval rate (lnrr) is associated with improved local control. Currently, no consensus has been reached on the definition of an adequate lnrr. A minimum lnrr has been proposed as a quality assessment parameter that has to be validated.


We conducted a retrospective cohort analysis at the Princess Margaret Cancer Centre (University Health Network, Toronto, ON). The lnrrs for all patients who underwent rlnd for malignant cutaneous melanoma during 2000–2010 were recorded. Indications for rlnd were a positive sentinel lymph node biopsy or clinical lymphadenopathy (palpable or radiologically detected).


Of the 207 identified rlnds, 146 (70.5%) were subsequent to a positive sentinel lymph node biopsy, and 61 (29.5%) were performed for clinical lymphadenopathy. The median lnrr was 24 nodes (range: 9–47 nodes; 10th percentile: 14 nodes) for axillary rlnd, 12 nodes (range: 5–30 nodes; 10th percentile: 8 nodes) for inguinal rlnd, and 16 nodes (range: 10–21 nodes; 10th percentile: 11 nodes) for ilioinguinal rlnd. The results were similar when comparing patients with positive sentinel lymph nodes and those with clinical lymphadenopathy, and the same surgical techniques were used in both groups.


The lnrrs at our institution are similar to rates reported at other tertiary-care melanoma centres. A minimum acceptable lnrr can be considered a quality assessment parameter in the surgical management of melanoma with nodal metastasis.

KEYWORDS: Melanoma, lymph node retrieval rates, quality assessment


The incidence of melanoma is increasing1. Prognosis is poor in patients with advanced or recurrent disease2. Regional lymph node dissection (rlnd) is currently recommended for patients with isolated gross nodal disease (palpable or detected by medical imaging) and for patients with a positive sentinel lymph node biopsy (slnb)3,4. A higher lymph node retrieval rate (lnrr) has been shown to be an independent predictor of local disease control, and assessment of the burden of nodal disease is an important element of staging5. The lnrr has therefore been proposed as a quality indicator, whereby surgeons and pathologists should aim to retrieve and identify more than a predetermined minimum number of lymph nodes if a resection is to be considered adequate6.

A similar metric has been used for other solid tumours. For example, in colon cancer, the minimum lnrr required for staging is 12, and that number has been used as a marker of a quality resection and pathology assessment7. However, in the setting of rlnd for melanoma, no minimum required number of lymph nodes has yet been set—a deficiency that was previously noted by Spillane and colleagues6,8 from the Melanoma Institute of Australia. They reported the lnrr from cases in their prospective database and proposed that the minimum number of excised lymph nodes should be greater than the 10th percentile of the number of lymph nodes excised in their patient cohort.

An editorial in Annals of Surgical Oncology encouraged other institutions to publish their lnrrs to validate the Australian data by assessing lnrrs in various populations and countries9. Subsequently, Rossi et al.10 published the lnrrs from 9 Italian centres, obtaining results similar to those in the Australian study. Our aim was to compare the lnrrs at our specialized Canadian cancer centre with previously published rates so as to provide external validity in establishing quality standards for rlnd in metastatic melanoma. Additionally, we sought to determine whether the minimum lnrrs can be applied to all patients by examining whether the lnrrs were different for patients with clinical lymphadenopathy (cl—that is, palpable or radiographic nodal disease) and for patients undergoing completion lymph node dissection after a positive slnb.


Our retrospective cohort analysis, conducted at the Princess Margaret Cancer Centre (Toronto, ON) included all patients who underwent rlnd for cutaneous malignant melanoma during 2000–2010. Patients were identified from prospective surgeon-maintained databases of all patients treated with surgery for melanoma at our institution. All procedures were undertaken by a group of fellowship-trained surgical oncologists who performed similar comprehensive axillary (levels 1–3) and complete inguinal and ilioinguinal node dissections4. Data were collected from clinical notes, operative notes, and pathology reports. The study protocol was approved by the institution’s research ethics board.

The final study cohort consisted of 196 patients who had undergone a total of 207 rlnds. In 9 patients, a synchronous bilateral rlnd was performed; 2 patients underwent an initial rlnd and then subsequently developed a regional recurrence at a different basin, undergoing a second rlnd at that time. The primary outcome was the lnrrs for axillary, inguinal, and ilioinguinal rlnds. We also compared lnrrs between patients with cl and patients with a positive slnb. The patient groups were compared with respect to patient and tumour characteristics and also with respect to the burden of nodal disease measured by the number of positive nodes and the ratio of positive nodes to the total number of nodes examined (lymph node ratio).The total number of nodes retrieved included all nodes retrieved at the time of rlnd and slnb if applicable.

A univariable analysis was performed to compare patient characteristics in both the positive slnb and cl cohorts. The chi-square test or Fisher exact test (in cases in which the expected count was less than 5) was used to compare categorical variables. A 2-sample t-test or Wilcoxon rank-sum test was used to examine continuous variables (normally and non-normally distributed respectively). Results were considered significant at the level of p ≤ 0.05. All statistical analyses were performed using the SAS software application (version 9.3: SAS Institute, Cary, NC, U.S.A.).


Of the 207 rlnds, 146 (70.5%) were subsequent to a positive slnb, and 61 (29.5%) were performed in the setting of cl. Mean age was 53.2 years in the slnb group compared with 58.6 years in the cl group (p = 0.017). The most common location of the primary was the trunk in patients with a positive slnb; the most common location in the cl group was the leg. Ulceration of the primary was present in 42.5% of patients in the slnb group and in 66.7% of patients in the cl group (p = 0.014). No significant difference between groups was observed in sex distribution, tumour thickness, or histologic subtype (Table i).

TABLE I Patient and tumour characteristics


The median number of positive lymph nodes was higher in the cl group than in the slnb group (2 vs. 1,p = 0.015). The median lymph node ratio was also higher in the cl group than in the slnb group: 14% (range: 2%–100%) and 9% (range: 2%–92%) respectively (p = 0.026).

The median lnrr was 24 nodes (range: 9–47 nodes; 10th percentile: 14 nodes) for axillary rlnd, 12 nodes (range: 5–30 nodes; 10th percentile: 8 nodes) for inguinal rlnd, and 16 nodes (range: 10–21 nodes; 10th percentile: 11 nodes) for ilioinguinal rlnd (Table ii). The lnrr for each procedure type was not significantly different for patients undergoing rlnd for either positive slnb or cl: The mean lnrr in axillary rlnd was 25.3 ± 1.0 nodes for positive slnb (n = 85) compared with 26.3 ± 1.6 nodes for cl (n = 30), p = 0.59; in inguinal lnrr, it was 12.4 ± 0.6 nodes for positive slnb (n = 55) compared with 13.4 ± 0.9 nodes for cl (n = 26), p = 0.32; and in ilioinguinal lnrr, it was 17.3 ± 1.5 nodes for positive slnb (n = 6) compared with 14.0 ± 1.7 nodes for cl (n = 5), p = 0.17.

TABLE II Comparison of lymph node (LN) retrieval between centres



Delivering quality health care is a priority. There is increasing awareness and adoption of the National Surgical Quality Improvement Program from the American College of Surgeons11. Participating hospitals report risk-adjusted outcomes data for patients undergoing surgical procedures. Hospitals adopting the program have benefited with enhanced performance12. There is interest in incorporating new outcomes variables that are oncology-specific, and lnrr might be a suitable candidate as a measure of surgical quality for melanoma surgery. Although the anatomic boundaries of each basin’s rlnd have been described, awareness of the anatomic land-marks might not be sufficient to ensure an adequate resection4. Furthermore, although the Multicenter Selective Lymphadenectomy Trial ii is incomplete, some surgeons are offering ultrasonography surveillance of the nodal basin after positive sentinel lymph node biopsy13. As a result, surgeons are performing fewer rlnds, which have become specialized procedures. However, rlnd is still recommended for some patients (for example, in the presence of cl), and therefore an objective quality measure would be valuable in assessing the performance of this uncommon procedure.

By direct comparison to results from the Melanoma Institute of Australia6,8 and the Italian Melanoma Intergroup10 studies, the mean and 10th percentile axillary and inguinal lnrrs at our institution were at least as good as those achieved by the other two groups. Our lnrrs for ilioinguinal dissection were lower, but those procedures numbered too few in our cohort (n = 11) to permit any definite conclusions to be drawn. Globally, it seems that our lnrrs are comparable to those at other large centres. Although our sample size might be too small to provide significant external validity, our results successfully demonstrate that a reproducible lnrr might be achievable and could be used as a quality indicator.

An important consideration in applying the lnrr as a quality indicator is that lnrr has not been correlated with outcomes in our patients. It is possible that, beyond a certain lnrr, removal of additional nodes might be of no oncologic benefit. Furthermore, removal of additional nodes puts patients at increased risk for lymphedema.

The usefulness of reviewing an institution’s lnrr is that surgeons and pathologists who might benefit from additional education to improve their technique could be identified. The lnrr can also serve as an entry criterion for patients enrolling in randomized controlled trials for adjuvant therapy. Furthermore, as novel surgical approaches such as minimally invasive inguinal lymph node dissection are incorporated into surgical care, reviewing the lnrr of those procedures for comparison with our published lnrrs in a conventional open approach can be used to quickly validate the completeness of the new techniques14.

Research into quality improvement in surgery has focused on the effect of surgeon and hospital volume on outcomes15,16. Spillane et al.6,8 demonstrated a higher lnrr for surgeons at the high-volume Sydney Melanoma Unit than for other Australian surgeons. Similarly, at our own institution, where our lnrr for ilioinguinal dissections compares poorly with those at reference centres, the difference might be a result of our lower volume for that specific procedure. Our results and those of the other published groups can be used as a benchmark that other surgeons and hospitals can use in a comparison of their lnrrs for rlnds in melanoma. In cases in which fewer than the minimum expected number of nodes is retrieved, surgeons should critically assess their surgical technique and perhaps refrain from performing the procedure if volumes are low and consider referring patients to specialty centres.

A limitation of our conclusions is that the lnrr does not exclusively depend on surgical technique. Pathologists at different centres might process surgical specimens differently and with variable scrutiny. A study investigating the lnrr in colorectal cancer specimens revealed that repeat examination of the specimen by pathologists can reveal previously unseen nodes17. As a quality indicator of surgical technique, lnrr might therefore have to be a combined quality indicator of the surgeon and the pathologist. All specimens included in our study were examined by experienced melanoma-specific pathologists.

At our institution, the surgical procedure is the same whether the indication is a positive slnb or cl, and therefore, not surprisingly, the lnrr was not significantly different between those patient groups. The use of adjuvant systemic therapies has led some authors to advocate observation alone or limited axillary dissection for patients with positive sentinel lymph nodes13,18,19. It might therefore be appropriate that, in future, different standards for lnrr be set based on the indication for the procedure.

We encourage the U.S. National Comprehensive Cancer Network to critically review our data and the previously published data with respect to lnrr in melanoma to establish guidelines that can aid surgeons and hospitals in assessing the quality of their surgical treatment of melanoma with lymph node metastases. A rlnd is a specialized procedure, and surgeons need quantitative feedback on their performance.


There is a reproducible minimal number of lymph nodes retrieved that defines an adequate lymph node dissection for metastatic melanoma that can be applied to patients with a positive slnb or cl.


We have read and understood Current Oncology’s policy on disclosing conflicts of interest, and we declare that we have none


*Division of General Surgery, Department of Surgery, University of Toronto, Toronto;,
Division of General Surgery, Department of Surgery, University of Ottawa, Ottawa; and,
Department of Surgical Oncology, Princess Margaret Cancer Centre, University Health Network, Toronto, ON..


1. Erdmann F, Lortet-Tieulent J, Schuz J, et al. International trends in the incidence of malignant melanoma 1953–2008— are recent generations at higher or lower risk? Int J Cancer 2013;132:385–400.

2. United States, Department of Health and Human Services, National Institutes of Health, National Cancer Institute (nci), Surveillance, Epidemiology, and End Results (seer) Program. Cancer Stat Facts: Melanoma of the Skin [Web page]. Bethesda, MD: nci; 2016. [Current version available at:; cited 13 July 2017]

3. Morton DL, Thompson JF, Cochran AJ, et al. on behalf of the mslt group. Final trial report of sentinel-node biopsy versus nodal observation in melanoma. N Engl J Med 2014;370:599–609.
cross-ref  pubmed  pmc  

4. Easson AM, Rotstein LE, McCready DR. Lymph node assessment in melanoma. J Surg Oncol 2009;99:176–85.

5. Grotz TE, Huebner M, Pockaj BA, Perkins S, Jakub JW. Limitations of lymph node ratio, evidence-based benchmarks, and the importance of a thorough lymph node dissection in melanoma. Ann Surg Oncol 2013;20:4370–7.
cross-ref  pubmed  

6. Spillane AJ, Cheung BL, Stretch JR, et al. Proposed quality standards for regional lymph node dissections in patients with melanoma. Ann Surg 2009;249:473–80.
cross-ref  pubmed  

7. McFadyen C, Lankshear S, Divaris D, et al. Physician level reporting of surgical and pathology performance indicators: a regional study to assess feasibility and impact on quality. Can J Surg 2015;58:31–40.
cross-ref  pmc  

8. Spillane AJ, Haydu L, McMillan W, Stretch JR, Thompson JF. Quality assurance parameters and predictors of outcome for ilioinguinal and inguinal dissection in a contemporary melanoma patient population. Ann Surg Oncol 2011;18:2521–8.
cross-ref  pubmed  

9. Sondak VK. How do we know when a lymph node dissection is adequate? Ann Surg Oncol 2011;18:2419–21.
cross-ref  pubmed  

10. Rossi CR, Mozzillo N, Maurichi A, et al. Number of excised lymph nodes as a quality assurance measure for lymphadenectomy in melanoma. JAMA Surg 2014;149:700–6.
cross-ref  pubmed  

11. Jones RS, Brown C, Opelka F. Surgeon compensation: “Pay for performance,” the American College of Surgeons National Surgical Quality Improvement Program, the Surgical Care Improvement Program, and other considerations. Surgery 2005;138:829–36.
cross-ref  pubmed  

12. Fuchshuber PR, Greif W, Tidwell CR, et al. The power of the National Surgical Quality Improvement Program—achieving a zero pneumonia rate in general surgery patients. Perm J 2012;16:39–45.
cross-ref  pubmed  pmc  

13. Morton DL. Overview and update of the phase iii Multicenter Selective Lymphadenectomy Trials (mslt-i and mslt-ii) in melanoma. Clin Exp Metastasis 2012;29:699–706.
cross-ref  pubmed  pmc  

14. Jakub JW, Terando AM, Sarnaik A, et al. Safety and Feasibility of Minimally Invasive Inguinal Lymph Node Dissection in patients with melanoma (safe-milnd): report of a prospective multi-institutional trial. Ann Surg 2017;265:192–6.

15. Mesman R, Westert GP, Berden BJ, Faber MJ. Why do high-volume hospitals achieve better outcomes? A systematic review about intermediate factors in volume-outcome relationships. Health Policy 2015;119:1055–67.
cross-ref  pubmed  

16. Liu CJ, Chou YJ, Teng CJ, et al. Association of surgeon volume and hospital volume with the outcome of patients receiving definitive surgery for colorectal cancer: a nationwide population-based study. Cancer 2015;121:2782–90.
cross-ref  pubmed  

17. Fan L, Levy M, Aguilar CE, et al. Lymph node retrieval from colorectal resection specimens for adenocarcinoma: is it worth the extra effort to find at least 12 nodes? Colorectal Dis 2011;13:1377–83.

18. Tsutsumida A, Takahashi A, Namikawa K, et al. Frequency of level ii and iii axillary nodes metastases in patients with positive sentinel lymph nodes in melanoma: a multi-institutional study in Japan. Int J Clin Oncol 2016;21:796–800.
cross-ref  pubmed  

19. Nessim C, Law C, McConnell Y, Shachar S, McKinnon G, Wright F. How often do level iii nodes bear melanoma metastases and does it affect patient outcomes? Ann Surg Oncol 2013;20:2056–64.
cross-ref  pubmed  

Correspondence to: Alexandra M. Easson, Princess Margaret Hospital, Room 3-130, 610 University Avenue, Toronto, Ontario M5T 2M9. E-mail:

(Return to Top)

Current Oncology, VOLUME 24, NUMBER 4, August 2017

Copyright © 2017 Multimed Inc.
ISSN: 1198-0052 (Print) ISSN: 1718-7729 (Online)