ABSTRACT
BACKGROUND: A multicenter study was conducted to test the ability of electrical impedance scanning to differentiate between benign and malignant skin lesions. The performance of a dual electrical impedance scanning/image analysis device was also assessed. METHODS: Electrical impedance scanning measurements of 449 preoperative lesions found on 382 patients and including 53 melanomas from the trunk and extremities were performed. Results were correlated with histopathologic findings. In addition, ABCD parameters for the lesions were automatically calculated by the system. RESULTS: Electrical impedance scanning detected melanomas of the trunk and extremities with 91 percent sensitivity and 64 percent specificity. Moreover, sensitivity of electrical impedance scanning was increased to 100 percent for in situ and thin melanomas of smaller size (n = 27). Visual examination identified as malignant only 67 percent of these early tumors (p = 0.002). Clinical examination detected 96 percent of the larger or thicker melanomas (n = 26), whereas electrical impedance scanning detected only 81 percent of them. Combined electrical impedance scanning and image analysis detected 100 percent of the melanomas, independent of their thickness, and with no significant decrease of specificity. Because of electrical differences between the head/neck and the rest of the body, the assessed electrical impedance scanning parameters were not adequate for the diagnosis of melanomas from the head and neck. CONCLUSIONS: A validation study proved the value of electrical impedance scanning as a noninvasive technique for detection of melanoma lesions of the trunk and extremities, specifically, of in situ and thin type. In addition, image analysis was shown to be a valuable, complementary procedure. New parameters should be designed to optimize the performance of electrical impedance scanning for melanomas of the head and neck.
