Background: Because the early detection of cutaneous melanoma can dramatically improve survival, identification and surveillance of persons at risk have received much attention.
Objective: Our purpose was to examine the influences of personal or family history, patterns of detection, and prior skin biopsies (considered to be a measurement of surveillance by medical personnel) on the Breslow depth of cutaneous melanomas.
Methods: A retrospective cohort analysis of 218 patients with a history of at least one invasive cutaneous melanoma who visited the Yale Pigmented Lesion Clinic between January 1995 and January 1996 was performed. Data on patterns of detection, melanocytic nevi, and skin biopsies before and after the initial diagnosis of melanoma were collected, and patients with a family history of melanoma were compared with sporadic patients.
Results: Initial melanomas discovered by dermatologists were more likely to be 0.75 mm or less in depth than those found by other physicians (P = .03). Although patients detected 45% of the initial primary melanomas (98/218), dermatologists discovered 80% of the second primary tumors (33/41; P = .001). A personal history of melanoma was predictive of a thinner Breslow depth (P = .01), but a family history of melanoma was not. Having a biopsy of any type or combination of types of skin lesion(s) performed in the 5 years, 2 years, or 1 year before the first diagnosis of melanoma did not predict a melanoma of thinner Breslow depth among either familial or sporadic patients. The mean number of skin biopsies performed per patient was 8 times higher in the 5-year period after (5.6) versus the 5-year period before (0.7) the initial diagnosis of melanoma, with a peak in the first year after the diagnosis (2.3 vs 0.25 in the prior year). In 27 patients, one or more skin biopsies were performed in the year before the initial diagnosis of melanoma; 41% of these biopsies (23/56) were of lesions in normally exposed sites (eg, the face, neck, and forearms) compared with 22% of the melanomas (6/27).
Limitations: Since an invasive melanoma (with the possible exception of a nodular melanoma) would likely have been present for at least a year, plausible explanations for why evidence of previous dermatologic care did not appear to result in earlier detection include performance of a limited rather than a total body skin examination as well as subtle clinical features of early melanomas. However, this study cannot give weight to these explanations because at the time new Pigmented Lesion Clinic patients were not routinely asked about previous total body skin examinations.
Conclusions: The disappointing trends seen in this study, with neither the well-established risk factor of a family history of melanoma nor previously having a skin biopsy predicting thinner melanomas, highlight the need to establish criteria defining the subset of patients for whom appropriate management requires periodic total body skin examination.