Search

Abstract: Background: Differentiating melanoma metastasis from benign cutaneous lesions currently requires biopsy or costly imaging, such as positron emission tomography scans. Melanoma metastases have been observed to be subjectively warmer than similarly appearing benign lesions. We hypothesized that infrared (IR) thermography would be sensitive and specific in differentiating palpable melanoma metastases from benign lesions. Materials and methods: Seventy-four patients (36 females and 38 males) had 251 palpable lesions imaged for this pilot study. Diagnosis was determined using pathologic confirmation or clinical diagnosis. Lesions were divided into size strata for analysis: 0–5, >5–15, >15–30, and >30 mm. Images were scored on a scale from −1 (colder than the surrounding tissue) to +3 (significantly hotter than the surrounding tissue). Sensitivity and specificity were calculated for each stratum. Logistical challenges were scored. Results: IR imaging was able to determine the malignancy of small (0–5 mm) lesions with a sensitivity of 39% and specificity of 100%. For lesions >5–15 mm, sensitivity was 58% and specificity 98%. For lesions >15–30 mm, sensitivity was 95% and specificity 100%, and for lesions >30 mm, sensitivity was 78% and specificity 89%. The positive predictive value was 88%–100% across all strata, and the negative predictive value was 95% for >15–30 mm lesions and 80% for >30 mm lesions. Conclusions: Malignant lesions >15 mm were differentiated from benign lesions with excellent sensitivity and specificity. IR imaging was well tolerated and feasible in a clinic setting. This pilot study shows promise in the use of thermography for the diagnosis of malignant melanoma with further potential as a noninvasive tool to follow tumor responses to systemic therapies. [Copyright &y& Elsevier]