Prevalence of suspicious ultrasound findings in patients with mammographic focal asymmetry at Aga Khan University Hospital, Nairobi

Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Prof. Sudhir Vinayak

Second Supervisor/Advisor

Prof. Zul Premji

Third Supervisor/Advisor

Dr. Rose Ndumia


Imaging and Diagnostic Radiology (East Africa)


Background: Breast cancer has become the leading cancer in women in both economically developed and developing countries, accounting for 25% of all cancers diagnosed worldwide in 2012. The cornerstone of breast cancer control remains early detection in order to improve outcomes and survival. Thus far the only breast cancer screening method that has proved to be effective is mammography. Although mammography is the mainstay of early detection, a fundamental limitation is its low inherent contrast difference between the soft tissue structures in the breast. Mammographic specificity relies on the ability to distinguish benign from malignant breast lesions based on their margins and morphological features. When breast malignancy presents with subtle mammographic features such as focal asymmetry, its specificity is reduced. Overall, the larger portion of false -negative mammograms comprises of cancers which are visible in retrospect as ‘minimal sign’ cancers. Additional imaging with ultrasound is useful to further characterize areas of mammographic focal asymmetry, and sonographic findings are used to determine the need for subsequent biopsy for histological confirmation. However, mammographic focal asymmetry has historically not been subjected to adequate follow up, and there is limited data in radiology literature regarding ultrasound findings in its evaluation.

Objective: To determine the prevalence of suspicious ultrasound findings in patients with mammographic focal asymmetry.

Methods: This was a cross-sectional study, whereby women presenting for mammograms at the Radiology Department with a mammographic descriptor of focal asymmetry (as per the American College of Radiology guidelines) and recommendation for additional imaging evaluation with breast ultrasound were consecutively recruited. The whole breast ultrasound images were evaluated for normal and abnormal findings, and allotted a final imaging assessment categorisation using the ACR Breast Imaging – Reporting and Data System (BI-RADS®) ultrasound lexicon. Ultrasound BI-RADS® 1-3 categories were assigned as benign findings, while ultrasound BI-RADS® 4 and 5 were assigned as suspicious findings. A total of 141 patients were enrolled.

Analysis: Data collected were entered into a spreadsheet application (Excel for Microsoft Windows, Microsoft Corporation) and analysed using Stata® version 11.2. The proportion of suspicious ultrasound findings in mammographic focal asymmetries was calculated.

Results: A total of 141 patients met the eligibility criteria during the study period and were enrolled into the study. The median age was 50 years, with a range of 31 to 79 years. The prevalence of suspicious ultrasound findings in patients with mammographic focal asymmetry was 7.1%. There was equal involvement of right and left breasts in patients with suspicious ultrasound lesions, with the most commonly involved location being the right upper outer quadrant in 40%. Overall, mammographic focal asymmetries were found to involve the left breast only slightly more than the right (in 51.8% of patients), and were most frequently located in the left upper outer quadrant (in 35.5% of all patients). The most common benign sonographic finding was fibrocystic breast disease, seen in 56% of all patients.

Discussion: Mammographic focal asymmetry is not uncommon as an early sign of breast cancer, but has historically not been subjected to adequate follow up. As a result there is limited data in radiology literature regarding findings on further ultrasound evaluation. The available data is also widely disparate, with reported prevalence of suspicious ultrasound findings of mammographic asymmetry ranging between 0 and 76%. To the authors’ knowledge, this is the first study in East Africa aimed at estimating the prevalence of suspicious ultrasound findings in mammographic focal asymmetry. Previous studies have been retrospective in nature, and have only utilised secondary data. Additionally, only 25% of the total asymmetries were found to represent normal glandular tissue, contrary to literature that has described mostly normal findings on additional breast ultrasound evaluation. Other than the inherent limitations of retrospective studies and use of secondary data, imaging criteria for the definition of mammographic focal asymmetry and the clinical significance of individual ultrasound characteristics have been updated to be more comprehensive and intelligible, and data utilising current ACR guidelines is scant.

Conclusion/recommendations: Additional ultrasound evaluation of mammographic focal asymmetry is desirable and important as it encompasses adequate imaging follow-up assessment, with no additional radiation risk. It also provides a means for final imaging categorisation, facilitating decision-making regarding which patients to reassure of normal or benign findings, and which patients require biopsy for histological analysis. The authors therefore recommend that focal mammographic asymmetry be subjected to routine ultrasound follow-up. This study also forms a basis for the need of a population-based study in which different population demographics are appropriately represented; with the goal of formalising policy on appropriate imaging follow-up of mammographic focal asymmetry.

This document is available in the relevant AKU library