BACKGROUND: Pre-operative determination of primary melanoma thickness could be a tool to identify those patients who could be treated with radical primary tumour excision and sentinel lymph node biopsy in a single procedure. An excellent correlation between sonographic and histological measurement of maximal tumour thickness has been achieved using 20-MHz transducers. OBJECTIVE: To show that widely available high resolution ultrasound with 12-15 MHz linear probe could also reliably assess the thickness of primary melanoma. METHODS: Sixty-nine patients underwent ultrasound evaluation of 70 clinically and dermoscopically suspicious pigmented skin lesions before surgical excision. RESULTS: The sensitivity, specificity, positive and negative predictive values of ultrasound to detect melanoma > 1 mm were 92%, 92%, 95% and 81% respectively. The correlation between ultrasound and histological tumour thickness was very good [Pearson's correlating index 0.823 (P < 0.001)]. Mean difference between sonographic and histological measurements was 0.045 mm with limits of agreement estimated at -1.4 and +1.49, and a bias between two methods 45 microm. CONCLUSION: Ultrasound examination with a 12-15 MHz linear transducer can reliably differentiate primary melanoma > 1 mm from those <or= 1 mm.
BACKGROUND: Pre-operative determination of primary melanoma thickness could be a tool to identify those patients who could be treated with radical primary tumour excision and sentinel lymph node biopsy in a single procedure. An excellent correlation between sonographic and histological measurement of maximal tumour thickness has been achieved using 20-MHz transducers. OBJECTIVE: To show that widely available high resolution ultrasound with 12-15 MHz linear probe could also reliably assess the thickness of primary melanoma. METHODS: Sixty-nine patients underwent ultrasound evaluation of 70 clinically and dermoscopically suspicious pigmented skin lesions before surgical excision. RESULTS: The sensitivity, specificity, positive and negative predictive values of ultrasound to detect melanoma > 1 mm were 92%, 92%, 95% and 81% respectively. The correlation between ultrasound and histological tumour thickness was very good [Pearson's correlating index 0.823 (P < 0.001)]. Mean difference between sonographic and histological measurements was 0.045 mm with limits of agreement estimated at -1.4 and +1.49, and a bias between two methods 45 microm. CONCLUSION: Ultrasound examination with a 12-15 MHz linear transducer can reliably differentiate primary melanoma > 1 mm from those <or= 1 mm.