A Co-Operative Evaluation of Mammography in Seven Teaching Hospitals

Abstract

The diagnostic criteria and radiographic technic of mammography were described in the American literature by Leborgne (1) in 1951 and Gershon-Cohen (2) in 1953. Prior to Egan's report in 1960 (3), however, the examination was not performed as a routine in the majority of radiology departments in hospitals of the Philadelphia area. Many radiologists had attempted mammography only to abandon it after failing to achieve the accuracy of interpretation of published reports (4). Because of the unchanging death rate from breast cancer and the possibility of earlier accurate diagnosis suggested by Egan, a group of radiologists representing seven teaching hospitals2 in Philadelphia undertook in 1962, at the suggestion of Dr. Eugene P. Pendergrass, to study the value of mammography as a diagnostic tool. The primary purpose of the project was to determine the accuracy of mammography in diagnosis and its usefulness in the management of the breast patients of the physicians and surgeons practicing in the participating hospitals. Materials and Methods Description of Patients and Conduct of the Study: A total of 3,882 breasts in 2,022 women were examined radiographically. For selection of these patients, surgeons, gynecologists, internists, hospital ward, and clinic services were recruited as referral sources. Each source agreed to refer for mammography all adult female patients appearing with breast complaints of any type, regardless of the suspected diagnosis or the plan for clinical management. The age distribution and menopausal status of the subjects are given in Table I. The mean age of the group was forty-five years: three-fourths of the women were between thirty and sixty years, and 60 per cent were premenopausal. The referring sources assigned each breast of each patient to one or more of the following categories according to physical findings and clinical impression: 1. Clinically obvious carcinoma 2. Dominant mass, not obviously carcinoma 3. More than one breast mass 4. Breast complaint, no dominant mass 5. Established fibrocystic disease (by biopsy or follow-up) 6. Previous complete mastectomy 7. Very large breasts 8. Control; no breast complaint or finding in either breast 9. No positive findings in this breast; positive findings in opposite breast. These classifications were recorded on standard forms and forwarded to the project center. No attempt was made to define the classifications more specifically or to impose absolute uniformity in their use, since there were more than 100 referring sources. Table I shows the total number of breasts examined in each of the diagnostic categories listed above. The radiologist listed his findings on standard forms which were forwarded with the mammograms to the project center. Radiologic diagnoses were recorded as follows: 1. No disease 2. Benign 3. Malignant 4. Uncertain 5. Follow-up examination advised in __months