British Journal of Plastic Surgery (2002), 55, 298-301 �9 2002 The British Association of Plastic Surgeons doi: 10.1054/hips.2002.3825

I B R I T I S H J O U R N A L P L A S T I C S U R G E R Y

The learning curve for sentinel node biopsy in malignant melanoma

G. L. Ross, T. Shoaib, J. Scott, D. S. Soutar, H. W. Gray* and R. MacKiet

Plastic Surgery Unit, Canniesburn Hospital; *Department of Nuclear Medicine, Glasgow Royal Infirmary; and q?Department of Dermatology, University of Glasgow, Glasgow, UK

SUMMARY. Sentinel node biopsy (SNB) has emerged as an accurate means of identifying nodal micrometastasis m cutaneous melanoma. In order to assess our learning curve, we compared our first 30 cases with our subsequent 30 cases. A total of 60 patients underwent SNB for cutaneous melanoma, using preoperative lymphoscintigraphy together with the intraoperative use of a Neoprobe and Patent Blue V dye. At least one sentinel node was identified in 93% of patients (90% in our first 30 cases; 97% in our subsequent 30 cases). Sentinel nodes contained tumour in 21% of cases. Of the sentinel nodes that contained tumour in the first 30 cases, 87% were identified by Neoprobe examination and 60% using blue dye. In the second 30 cases, the tumour-containing sentinel nodes were identified in all cases by both the Neoprobe and the blue dye. The sentinel node appeared to be the only involved node in 71% of patients. In the first 30 patients, one patient with a negative sentinel node developed nodal recurrence. These data confirm the feasibility of the sentinel-node technique in cutaneous melanoma. However, there is a learning curve, and the technique should be performed only by limited numbers of people with suitable training. �9 2002 The British Association of Plastic Surgeons

Keywords: sentinel node biopsy, melanoma.

Sentinel node biopsy (SNB) has emerged as an accurate means of identifying micrometastatic nodal disease in melanoma patients since its introduction by Morton et al.1

SNB has been performed in our unit to identify patients requiring therapeutic lymph-node dissection, as part of a prospective randomised trial, and to stage patients for adjuvant therapy. These patients have been evaluated to determine whether a learning curve exists in our practice in identifying positive sentinel nodes. We compared our first 30 cases with our subsequent 30 cases.

Patients and methods

A total of 60 patients underwent SNB for melanoma using preoperative lymphoscintigraphy together with the intraoperative use of both a Neoprobe and Patent Blue V dye. Similar methods of sentinel-node identification have been described previously. 2-s The patients all had a proven diagnosis of malignant melanoma using excision biopsy. Wider excision of the melanoma and SNB were performed at the same time.

The day before surgery, patients attended the Nuclear Medicine Department, where up to 40MBq of 99m Tc-99m labelled co l lo ida l human serum albumin (Nanocoll) with a mean particle size of 80 nm (Nycomed, Amersham, UK) in approximately 0.5-1 ml of saline was injected intradermally at as many points as necessary to

Presented at the British Association of Plastic Surgeons Summer Meeting, Stifling, UK, July 2001.

surround the excision-biopsy scar completely. Static lymphoscintigraphy was performed 15 min, 30min and 60min after injection, or until the first appearance of sentinel nodes.

During surgery, approximately 0.5-2 ml of Patent Blue V dye (Laboratoire Guerbet, Aulnay-Sous-Bois, France) was injected into the same site(s) as the radiocolloid, completely surrounding the previous excision-biopsy scar. Through a small incision over the identified regional basin, blue-stained lymphatic vessels were sought and followed up to the site at which they drained into a lymph node. The standardisation of the injection technique of radiocolloid and blue dye was arrived at by the initial involvement of a single practitioner in all injections.

The Neoprobe-1500 hand-held gamma probe (Ethicon Endosurgery, UK), fitted with a 14.0 mm diameter straight collimated probe, was used to identify radioactive sen- tinel nodes, including those marked preoperatively during lymphoscintigraphy. Sentinel nodes were identified in their lymph-node basin, and were labelled according to colour and the presence of radioactivity. Radioactivity was confirmed within the sentinel node ex vivo. If no blue-stained lymphatic vessels were found, the probe was used to guide the dissection. Once the sentinel lymph node had been excised, the Neoprobe was used to search the resection bed to ensure that there were no residual areas of high radioactivity. Once the operator was satis- fied that all first-echelon nodes had been completely removed, the wound was closed.

The sentinel nodes were sent to histology, where they were fixed in 10% neutral buffered formalin and, after fixation, bisected through their longest axis. If the halves

298

The learning curve for sentinel node biopsy 299

were more than 2mm thick, the slices were further trimmed to provide additional 2 mm thick blocks. One half was processed for histological examination. One haematoxylin-and-eosin (H&E) stained section was pre- pared from each block and examined. Multiple step section- ing and immunohistochemistry were performed if routine H&E staining proved negative.

Results

A total of 60 patients underwent SNB. The patients were divided into two groups, according to whether the node biopsy was performed within the first 30 cases or the subsequent 30 cases. The demographic data of the patients in each group are shown in Table 1. The mean Breslow thicknesses for the two groups were 3.06mm and 3.36 mm, respectively. Out of the 60 patients studied, lymphoscintigraphy identified more than one lymphatic draining basin in five patients. A total of 33 nodal basins were explored in each group.

In the first 30 patients the sentinel node was identified in 90% (27/30). There were three cases in which a sen- tinel node could not be identified, two in the axillary- node basin and one in the cervical basin. In the second 30 patients the sentinel node was identified in 97% (29/30). One patient had a sentinel node that could not be detected within the cervical basin (Fig. 1).

More than one sentinel node was identified in 31 patients. A total of 100 sentinel nodes were biopsied and analysed. Overall, 21% (21/100) of sentinel nodes analysed contained tumour, and 79% (79/100) were negative for tumour. We identified 87% (13/15) of positive sentinel nodes in the first 30 patients by Neoprobe examination,

Table 1 Comparison of prognostic factors between the first and second 30 patients

Prognostic factor First 30 patients Second 30 patients

sex 20 M/10 F 11 M/19 F mean age (years) 55 52 primary site

head and neck 8 4 trunk 8 9 extremities 14 17

Breslow thickness <1 ram 3 1 1.01-2.0 mm 8 6 2.01-4.0 mm 13 14 > 4 . 0 m m 6 9

and 60% (9/15) using blue dye (Fig. 2). All seven posi- tive nodes in the second 30 patients were identified both by Neoprobe examination and using blue dye. Subsequent therapeutic lymph-node dissection confirmed that, over- all, in 71% (15/21) of patients the sentinel node was the only lymph node containing micrometastatic disease (57% (8/14) in the first 30 patients and 100% (7/7) in the second 30 patients (Fig. 3)).

Follow-up of these patients is relatively short. The mean follow-up for the first 30 patients was 24 months, and for the second 30 patients it was 6 months. One patient in the first group has developed regional failure in the lymph node basin staged negative by SNB.

Discussion

There is a distinct learning curve for SNB in the treat- ment of cutaneous melanoma. Standardisation of the technique was maintained by limiting the number of practitioners who perform SNB.

Histological diagnosis was obtained by excision biopsy in all cases before proceeding with wide excision and SNB. SNB was performed before the wide excision of the tumour in an attempt to minimise the disruption to the lymphatics and to preserve the blue-dye reservoir. 9 Wide excision prior to SNB is thought to increase the failure rate of sentinel-node detection, 1 although identifi- cation of the sentinel node by SNB in 98.5% of cases has been reported after wide excision. 10

Our experience has shown that the injection method forms a crucial part of establishing a good SNB programme. We used an intradermal peri-scar injection technique for both the radiocolloid and the blue dye.

100

90

80 70

60

50

40

30

20

!0 0

DYE NEOPROBE

~FZRST30 [ PATZENTS [

|SECOND 30 PATZENTS

Figure 2--Percentage of positive sentinel nodes identified using Neoprobe and blue dye.

[ ] FZRST30 PATZENTS

I S E C O N D 3 0 PATIENTS

~ONLY SENTINEL NODE POSITIVE

I I LYMPH NODE BASZN POSZTIVE

UNIDENT][FlrED SENTlrNEL NODES

Figure 1--Numbers of patients where no sentinel lymph node was found, by group.

FIRST 30 SECOND :30

Figure 3~Comparison of positive sentinel-node basin and subsequent lymph node basin specimens.

300 British Journal of Plastic Surge~T

Figure 4--Scar following excision biopsy of a malignant melanoma, prior to injection.

Figure 6--Scar surrounded by blue dye on completion of the injection technique.

Figure 5--Blue dye is injected into the dermis.

The scar was injected around its periphery as previously described (Figs 4-6). 11,12 Different lymphatic channels may drain different parts of the tumour, so it is important that the tumour is completely surrounded. This is particu- larly important for tumours of the trunk, where metasta- sis is frequently to more than one nodal basin. Injection directly into the scar is discouraged because of the high pressure required. We found that approximately four to six injections were usually required. This is similar to other reported injection techniquesJ 3 A 25 G syringe was used, and the dermis was infiltrated by holding the syringe at approximately 20 ~ to the horizontal. It is important to withdraw on the syringe while removing it,

in order to avoid spillage. Excessive amounts of blue dye may result in permanent tattooing of the patient's skin. The ideal injection uses the smallest possible amount of blue dye to completely surround the tumour; this is particularly important when treating lesions on the face.

Lymphatics are richer in the dermis than in the subcu- taneous system, and clearance is faster. Variation in the depth of injection will lead to variations in the distribu- tion of the colloid and dye to the sentinel node and in the speed of drainage through the lymphatic systemJ 4 A constant injection technique is important in reliably infiltrating the dermis and producing reproducible results. This is important for the nuclear medicine depart- ment who performed the lymphoscintigraphy and for the surgical team who can be guided by both the radiocolioid and the intraoperative blue dye.

The learning curve has been suggested to be 30 cases, ~,3,8,15 and a similar figure has been suggested in the treatment of breast cancer, where a 90% sentinel-node identification rate was obtained after 23 patients and a 95% rate after 53 patients. 16-18 Limiting the number of practitioners who perform SNB allows the development of suitable expertise in the nuclear medicine, pathology and surgical departments.

Our experience has shown that the identification rate of positive nodes by both neoprobe examination and using blue dye was lower during the first 30 cases. At present, the follow-up is too short to evaluate the role of SNB in terms of a potential survival benefit to patients with melanoma. Our experience confirms that SNB can be performed in over 90% of cases, but this figure can be improved with experience.

Acknowledgements

We thank all the consultants at Canniesburn for allowing their patients to be included in this study.

The learning curve for sentinel node biopsy 301

References

1. Morton DL, Wen D-R, Wong JH, et al. Technical details of intra- operative lymphatic mapping for early stage melanoma. Arch Surg 1992; 127: 392-9.

2. Albertini JJ, Cruse CW, Rapaport D, et al. Intraoperative radio- lymphoscintigraphy improves sentinel lymph node identification for patients with melanoma. Ann Surg 1996; 223: 217-24.

3. Morton DL, Thompson JF, Essner R, et al. Validation of the accuracy of intraoperative lymphatic mapping and sentinel lymphadenectomy for early-stage melanoma: a multicenter trial. Ann Surg 1999; 230: 453-65.

4. Thompson JF, McCarthy WH, Bosch CMJ, et al. Sentinel lymph node status as an indicator of the presence of metastatic melanoma in regional lymph nodes. Melanoma Res 1995; 5: 255-60.

5. Jansen L, Nieweg OE, Kapteijn AE, et al. Reliability of lym- phoscintigraphy in indicating the number of sentinel nodes in melanoma patients. Ann Surg Oncol 2000; 7: 624-30.

6. Gershenwald JE, Tseng C, Thompson W, et al. Improved sentinel lymph node localization in patients with primary melanoma with the use of radiolabeled colloid. Surgery 1998; 124: 203-10.

7. Sabel MS, Gibbs JF, Cheney R, McKinley BE Lee JS, Kraybill WG. Evolution of sentinel lymph node biopsy for melanoma at a National Cancer Institute-designated cancer center. Surgery 2000; 128: 556-63.

8. White WL, Loggie BW. Sentinel lymphadenectomy in the manage- ment of primary cutaneous malignant melanoma: an update. Dermatol Clin 1999; 17: 645-55.

9. Rees WV, Robinson DS, Holmes EC, Morton DL. Altered lymphatic drainage following tymphadenectomy. Cancer 1980; 45: 3045-9.

10. Karakousis CP, Grigoropoulos P. Sentinel node biopsy before and after wide excision of the primary melanoma. Ann Surg Oncol 1999; 6: 785-9.

11. Berman CG, Norman J, Cruse CW, Reintgen DS, Clark RA. Lymphoscintigraphy in malignant melanoma. Ann Plast Surg 1992; 28: 29-32.

12. Norman J, Cruse CW, Espinosa C, et al. Redefinition of cutaneous lymphatic drainage with the use of lymphoscintigraphy for malignant melanoma. Am J Surg 1991; 162: 432-7.

13. Keshtgar MRS, Waddington WA, Lakhani SR, Ell PJ. The Sentinel Node in Surgical Oncology. Berlin: Springer-Verlag, 1999: 49-59.

14. Abdel-Dayem HM, Siraj QH, Collier BD. Lymphoscintigraphy. In Maisey MN, Britton KE, Collier BD, eds. Clinical Nuclear Medicine, 3rd ed. London: Chapman & Hall, 1998: 541-52.

15. Thomas JM, Patocskai El. The argument against sentinel node biopsy for malignant melanoma: its use should be confined to patients in clinical trials. BMJ 2000; 321: 3-4.

16. Cox CE, Bass SS, Boulware D, Ku NK, Berman C, Reintgen DS. Implementation of new surgical technology: outcome measures for lymphatic mapping of breast carcinoma. Ann Surg Oncol 1999; 6: 553-61.

17. Bass SS, Cox CE, Ku NN, Berman C, Reintgen DS. The role of sentinel lymph node biopsy in breast cancer. J Am Coil Surg 1999; 189: 183-94.

18. Bass SS, Cox CE, Reintgen DS. Learning curves and certification for breast cancer lymphatic mapping. Surg Oncol Clin N Am 1999; 8: 497-509.

The Authors

Gary Ross MRCSEd, Head and Neck Research Fellow Taimur Shoaib FRCSEd, Specialist, Registrar in Plastic Surgery John Scott FRCS(Plast), Consultant Plastic Surgeon David S. Soutar ChM, Consultant Plastic Surgeon

Plastic Surgery Unit, Canniesburu Hospital, Switchback Road, Bearsden, Glasgow G61 1QL, UK.

Henry Gray FRCP, Consultant in Nuclear Medicine

Department of Nuclear Medicine, Glasgow Royal Infirmary, Glasgow, UK.

Rona Mackie MD, FRCP, FRCPath, Consultant Dermatologist

Department of Dermatology, Robertson Building, University of Glasgow, Glasgow, UK.

Correspondence to Mr Gary Ross.

Paper received 4 May 2001. Accepted 22 February 2002.