ABSTRACT
INTRODUCTION: Mammography was pioneered by surgeons but is now the domain of radiologists. With ever-increasing cost pressures it must be examined whether interpretation of mammography by clinicians and radiation technologists is comparable to that of breast radiologists. We present the largest series of surgeon-read screening mammography to date. METHODS: All mammography performed between 2003 and 2009 at a comprehensive breast centre was recorded prospectively. First assessment was by a radiation technologist and consensus established after second reading by a breast surgeon, who took responsibility for the reading. Data recorded were: age, hormonal replacement therapy, prior breast surgery, indications for mammography and outcomes. Outcomes were classified based using the Breast Imaging Reporting and Data System (BIRADS). Indeterminate lesions were imaged further or underwent tissue acquisition. All BIRADS 5 lesions underwent tissue acquisition. RESULTS: Of 11,948 mammograms, 538 were reported as indeterminate/compatible with malignancy; 240 biopsies were performed, and 87 cancers diagnosed. In 40-49-year-old women (4,956 mammograms), the recall rate was 4.2%, the biopsy rate was 1.6%, the malignancy rate of biopsy was 23.7% and the cancer diagnosis rate was 3.6/1,000 examinations; for 50-69-year-old women these figures were 6,546, 4.7, 2.2, 44.1% and 10.0/1,000, respectively, and in women older than 70 years, they were 446, 5.6, 3.4, 33.3% and 11.2/1,000, respectively. Of all cancers, 32.2% were non-invasive; of invasive cancers, 49.1% were 10 mm or less in diameter and 75% were node negative. CONCLUSIONS: These results are similar to those in high-quality organized screening programs. The role of breast surgeons in mammography interpretation should be expanded.
