An analysis of diagnostic practices in a mammography unit in a tertiary hospital in South Africa

• Published in: SA journal of radiology 2017, Vol.21 (1), p.1-8
• Main Authors: Surridge, Clare A., Goodier, Matthew D.M.
• Format: Article
• Language: English
• Subjects: Age; Analysis; Benign; BI-RADS; Biopsy; Breast biopsy; Breast cancer; Breasts; Core needle biopsy; Data processing; Diagnosis; Fibroadenoma; Health Care Sciences & Services; Hospitals; Internet; Invasiveness; Mammography; Medicine, General & Internal; Mortality; Patient outcomes; Patients; Radiology; Radiology, Nuclear Medicine & Medical Imaging; Ultrasonic imaging; Ultrasound; Ultrasound guided core needle biopsy; Validation studies; Women's health
• ISSN: 1027-202X; 2078-6778; 2078-6778
• DOI: 10.4102/sajr.v21i1.1059

Background: Breast cancer is the most common cancer in females in South Africa. The reporting of breast imaging has been standardised internationally using the Breast Imaging and Reporting Data System (BI-RADS), which includes guidelines for reporting of breast lesions and further management. Ultrasound-guided core-needle breast (UGCNB) biopsy is a widely used method of obtaining histological diagnoses of breast lesions to assist with planning definitive management.Objectives: To perform an audit of the UGCNB biopsies performed at the Grey’s Hospital Mammography Department and assess the accuracy of the radiologists’ use of the BI-RADS scoring system.Methods: Records of all patients who underwent UGCNB biopsy between 01 January 2014 and 31 October 2015 were reviewed. A retrospective study was performed.Results: A total of 304 UGCNB biopsies were performed on 291 patients. The mean age was 49.2 (s.d. = 15.9) years. Tissue samples from 303 lesions were adequate for histological assessment, and of these, 51% of the lesions were malignant whilst 49% were benign. The most common malignant and benign diagnoses were invasive ductal carcinoma and fibroadenoma respectively. The BI-RADS scoring of the radiologists demonstrated a positive predictive value of 61.6% for the identification of possible malignant lesions.Conclusion: This study describes the patient and lesion profile and unit practices in a tertiary hospital setting in South Africa. The radiologists’ application of the BI-RADS scoring largely conforms to the BI-RADS guidelines. The study highlights several challenges encountered by a breast imaging programme in an under-resourced setting as well as making recommendations in overcoming these challenges.