Routine biopsy of sonographically benign breast lesions greater than 3cm is necessary for the diagnosis of malignancy in women less than 40 years of age

Abstract

Palpable solid breast masses that are circumscribed and not calcified on mammogram or ultrasound are probably benign. There is controversy therefore, whether these deserve tissue diagnosis. More data is required to determine whether short term follow up can replace the need for biopsy. Benign appearing lesions greater than 3cm in diameter on ultrasound continue to undergo biopsy due to fear that a malignancy or phyllodes tumour might be missed. Published research reflects patients from Europe and North America, and no relevant data from Africa exists. AIM: This study aims to determine the histological spectrum of sonographically benign lesions greater than 3cm, which were biopsied, in our local population (majority of black patients) and to determine whether biopsy is indicated based on the local cancer risk. The study also aims to characterise the results by age and population group as well as correlate the histological result with the size of the lesion on ultrasound, the HIV status, family history and the seniority of the examining radiologists. MATERIALS AND METHODS: A retrospective descriptive study of biopsy results of sonographically benign breast masses was undertaken using biopsy procedural recording sheets. . The size of the lesions (continuous variables) mean with standard deviations was determined. The prevalence of lesions was expressed as a percentage. Other categorical variables were summarized as frequency and percentage. The vi histological spectrum of the lesions was determined. The HIV status and family history of the patients as well as the seniority of the reviewing radiologist was assessed. A Krusskal Wallis test and separate logistic regression analysis was used. RESULTS: A total of 68 patients (below 40 years of age) were included from a total of 13112 patients (of all ages) seen between 2007 and the end of 2010. 73 lesions were identified (65 benign and 8 malignant). The prevalence of benign lesions was 89.7%. .The prevalence of malignant lesions was 10.29%.There was little evidence to support lesion size for predicting histology (p value = 0.22) or benignity. There was little evidence that the family history and HIV status were significant. CONCLUSION: There was a high prevalence (10.29%) of malignancies in lesions classified by ultrasound as benign. The size of the lesion did not correlate with histological subtype or whether the lesion was benign or malignant. Training of sonographers, standardization of technique for established users and double reading, may produce a different result, as both junior and senior radiologists mistook malignant lesions for benign ones on ultrasound. Repeating this research using double reading after training may demonstrate whether there is a true higher prevalence of malignancy in ultrasonically benign breast lesions in our community. Until then, routine biopsy of these lesions is recommended.