Preoperative 15-MHz ultrasound assessment of tumor thickness in malignant melanoma

Abstract

Background: Tumor thickness is of great importance in the management of cutaneous malignant melanoma (MM): this variable not only affects prognosis but is also a key factor in planning surgical margins and selecting candidates for sentinel node biopsy. Breslow depth is the standard histologic measure of thickness, but technological advances have provided imaging techniques such as cutaneous ultrasound that can potentially assess tumor thickness and enable prompt initiation of definitive treatment.

Objectives: a) To evaluate the utility of ultrasound assessment of tumor thickness in MM, and b) to analyze histologic variables that affect ultrasound assessments of thickness.

Materials and methods: Retrospective study of a consecutive series of 79 primary cutaneous MMs in which tumor thickness had been assessed by 15-MHz ultrasound before surgery. We gathered data from histology reports, studying Breslow depth and the presence of ulceration, regression, inflammatory infiltrate, and associated nevi. Correlation coefficients were calculated to evaluate the strength of association between Breslow depth and thickness assessed by ultrasound. We also calculated the sensitivity, specificity, and positive and negative predictive values of ultrasound measurement in the diagnosis of MMs more than 1mm thick. Associations between histologic variables and the overestimation of thickness by ultrasound were also analyzed.

Results: The 79 primary MMs studied had a mean (SD) Breslow depth of 0.8 (1.4) mm. There was moderate correlation and agreement between Breslow depth and the ultrasound assessment of thickness (Pearson correlation coefficient, 0.678; intraclass correlation coefficient, 0.78). The tendency of ultrasound to overestimate thickness was nonsignificantly related to the presence of a moderate to intense infiltrate and associated nevi (P>.05). The sensitivity of ultrasound for the diagnosis of MM over 1mm thick was 82%; specificity was 80%, and positive and negative predictive values were 54% and 94%, respectively.

Conclusions: Ultrasound imaging quite correctly identifies thin MMs and can be useful for planning adequate surgical margins; however, there are limitations on its usefulness in the diagnosis of thick MMs. Additional studies are required to confirm whether certain histologic characteristics, such as the presence of a moderate to intense inflammatory infiltrate or associated nevi can lead to overestimation of thickness by ultrasound, limiting the clinical utility of this imaging technique in MM management.