Clinical Analysis of an Ultrasound System in the Evaluation of Skin Cancers: Correlation With Histology

Abstract

Background: The assessment of skin cancers in the clinical setting is often difficult, with important features such as depth and width remaining unknown until the biopsy with pathology reports are received. When we remove skin cancers, with those especially involving the face, aesthetics and invasion to surrounding structures such as bone and cartilage are important features for deciding the optimal surgical procedure and future reconstructive options. The aim of the study was to compare the accuracy of the ultrasound system in vivo and to correlate the results with the histopathological tumor thickness measured in skin cancer patients. Patients and Methods: From March 2010 to February 2012, we reviewed 40 patients who comprised a total of 49 skin lesions involving the face, neck, and scalp. Each skin lesions were classified by 9 facial aesthetic units. The patient's various skin lesions were scanned using an ultrasound system device (Philips iU22 xMatrix US), with a 5-17-MHz compact linear transducer. Using the ultrasound system, we analyzed the shape, depth, echogenicity, size, invasion skin level, and vascularity of the skin cancer lesions. The results were correlated with the histology, with special note to the depth of involvement. Results: Of the 40 patients recruited, 15 were male and 25 were female, ranging in age from 53 to 92 years (mean +/- SD 78.7 +/- 13.7 years). Clinically, 49 lesions suspicious of skin cancer were identified and ultrasounds were performed preoperatively. Depth was measured by ultrasound and histology. Mean ultrasound depth of skin lesion was 3.97 +/- 3.15 mm (range 0.80-14.00), and it was found to be 4.04 +/- 2.92 mm (range 1.00-14.00) based off of histology. There was excellent correlation (interclass correlation coefficient, 0.953) between the depth of the skin lesions measured histologically and by using the ultrasound. Conclusion: The ultrasound is not meant to replace histologic evaluations, but it can be used as another diagnostic tool to provide improved preoperative planning. It can be used as a noninvasive, easy, and low-cost screening method for various skin cancers, and provides valuable information such as lesion margins, shape, layers of involvement, and vascularity patterns.