Surgical margins in patients with cutaneous melanoma--assessing the adequacy of excision

Appropriate margins of excision for melanoma have been well defined on the basis of prospective clinical trials. Various factors may result in a discrepancy between the intended clinical margin and the pathologic margin ex vivo, however, making it difficult to determine whether adequate excision margins have been obtained. We reviewed the clinical and pathologic margins in 220 patients evaluated at our institution for localized primary cutaneous melanoma between 1995 and 1997. We calculated the percentage difference between the intended clinical margins (C) and the measured pathologic margin (P). We then divided this number by the intended clinical margin to determine the percentage margin discrepancy. We asked whether the margin discrepancy varied as a function of the Breslow depth or anatomic site of the primary lesion, age or sex of the patient, or surgeon performing the excision. Student's t-test was used to determine whether the differences observed between the groups were significant. Pathologic margins were narrower than clinical margins in 59% (129/220) of specimens, equal in 12% (27/220) and wider in 29% (64/220). The median difference between the two measurements was 10% for the group. There was no significant impact of tumor or patient features on margin discrepancy. Margin discrepancy varied by surgeon, but this was not statistically significant. It can be concluded that, in most cases, the discrepancy between the pathologic and clinical margin of a wide excision specimen approximates to 10%. Larger differences should alert the clinician to the possibility of inadequate excision margins.