Background: Complex sclerosing lesion (CSL) and its smaller counterpart, the radial scar (RS), are frequently seen pathological entities. They are clinically asymptomatic and, prior to the implementation of mammographic screening, were most commonly found incidentally during pathological examination of other biopsied lesions. Complex sclerosing lesions are being detected regularly on mammograms due to widespread screening; many of their features resemble those of malignancy. Management varies and has been controversial.
Methods: Twenty-three cases of CSL detected during the first prevalent round of screening at BreastScreen Western Sydney (from February 1993 until June 1995) are presented and reviewed. Assessment was by a combination of radiological, clinical and cytological work-up prior to surgical biopsy. In addition, 126 spiculated carcinomas detected in the same period were reviewed and compared.
Results: Fourteen RS/CSL (62%) had lucent centres and nine (38%) had a central mass; three had been diagnosed provisionally as RS/CSL. Spicule lengths ranged from 25 to 90 mm; central masses ranged from 5 to 50 mm; and mass:spicule length ratio ranged from 1.2:1 to 1:10. Calcification (benign or indeterminate) was present in six cases (29%). No RS/CSL contained 'suspicious' calcifications, whereas 120 of 126 carcinomas (95%) had a central mass and six (5%) had a lucent centre (spicule lengths: 10-90 mm; central mass: 5-40 mm; and mass:spicule length ratio: 1.1:1-1:10). Twenty-one spiculated carcinomas (17%) contained microcalcifications (14 benign or indeterminate; seven suspicious). Provisional radiological diagnosis (PRD) after mammogram, with or without ultrasound, for histologically confirmed RS/CSL, was RS/CSL in 18 cases (78%), carcinoma in four cases (17%) and equivocal in one case (5%). For eight (6.5%) spiculate carcinomas the PRD was RS/CSL prior to histological diagnosis. The RS/CSL were detected with equal frequency in right and left breasts, and 22 (96%) lesions occurred in the upper breast. Seven RS/CSL (31%) and 83 spiculated carcinomas (65%) had been described as 'palpable' but most were subtle. Twelve fine-needle aspiration biopsies were performed (six 'palpable' lesions (no radiological guidance); four with ultrasound guidance and two with stereotactic guidance), and five (62.5%) of eight adequate lesions were reported as benign, two (25%) were reported as atypical, and one (12.5%) was reported as suspicious.
Conclusions: Definitive mammographic and sonographic differentiation of RS/CSL and stellate-type carcinoma is impossible. For screen-detected lesions that may be RS/CSL, the appropriate surgical procedure is a small but adequate biopsy using guidewire or other localization methods with optimal cosmetic incision.