Margins of excision for cutaneous melanoma of the eyelid skin: the Collaborative Eyelid Skin Melanoma Group Report.

PURPOSE

To evaluate the practice patterns among surgeons who treat melanomas of the eyelid skin with respect to margins of excision and to look for possible correlation between margins of excision and the incidence of local and regional recurrence and distant metastasis.

METHODS

A retrospective survey of the members of the American Society of Ophthalmic Plastic and Reconstructive Surgery and the European Society of Ophthalmic Plastic and Reconstructive Surgery yielded 44 cases. The patients' age, sex, date of diagnosis, histologic classification of melanoma, Breslow thickness, Clark level, location of melanoma, size of margins of excision, and findings of local or regional recurrence or distant metastasis were recorded in each case. Patients were stratified on the basis of margins of excision: </=5 mm; >5 mm but <10 mm; and >/=10 mm. Patients were also stratified by Breslow thickness. A Cox regression model was used to evaluate the predictive value of each factor for recurrence. Main outcome measures were the incidences of local and regional recurrence and distant metastasis as a function of margins of excision and Breslow thickness.

RESULTS

The majority of patients for whom reliable information was available had excision margins of </=5 mm. The Breslow thickness of most of the tumors was </=1 mm. Eleven patients (25%) had local recurrence. Five patients (11%) had regional lymph node metastasis. All patients with regional nodal metastasis were men. Distant metastasis developed in 3 patients (7%)-2 men and 1 woman. The follow-up times ranged from 10 to 108 months (mean, 34 months; median, 21 months). The incidence of local recurrence was higher among patients with melanomas at least 2 mm thick and margins of excision </=5 mm than among patients with melanomas at least 2 mm thick but with margins >/=10 mm, but this difference was not statistically significant because very few patients had melanomas at least 2 mm thick. Breslow thickness was the only statistically significant predictor of local, regional, and distant metastasis. Margins of excision did not have a statistically significant effect on local, regional, or distant recurrence.

CONCLUSIONS

Breslow thickness is an important prognostic indicator for eyelid skin melanomas. A 5-mm margin of excision may be adequate for thin melanomas of the periocular skin, but because of the small number of patients in this series who had >5-mm margins, a definitive comparison of outcome with larger margins of excision cannot be made. For melanomas >/=2 mm, wider margins of excision may be prudent, and careful surveillance for local and regional recurrence is indicated.