Main Article Content

A review of computed tomography patterns of metastatic breast cancer patients undergoing treatment at a private oncology centre in Ghana


Yaw Boateng Mensah
Clement Edusa
Josephine Nsaful
Naa Adjeley Mensah

Abstract

Introduction: breast cancer is the commonest malignant disease in Ghanaian women and accounts for 17% of cancer-related deaths in the country. It has been classified into molecular subtypes depending on the presence or absence of hormone receptors and the human epidermal growth factor receptor 2. Computed tomography is often the preferred modality for monitoring metastatic disease due to its ability to determine the extent of local and metastatic disease.


Methods: this was a retrospective study conducted at Sweden Ghana Medical Centre (SGMC). Hospital records and chest and abdominal computed tomography (CT) scan images of breast cancer patients who had been managed at SGMC between June 2016 and August 2019 were used to document age, gender, histological group, type of surgical intervention done, molecular subtypes of the disease and imaging findings. Microsoft Excel 2016 and SPSS version 20.0 were used to summarise the data obtained into tables, charts and to test for significant associations.


 


Results: the most common site of breast cancer metastasis was lymph nodes. The three commonest sites of distant metastases were the lung seen in 44 patients (55.3%), bone in 37 patients (44.6%) and liver in 33 patients (39.8%). Chi square test for association between the molecular subtypes of the breast cancer and proportion of patients that showed a particular type of metastases revealed that, the differences noted for lung, pleural and cardiac metastases were statistically significant, that for bone and liver were not.


Conclusion: breast cancer commonly metastasised to lymph nodes, lung, bone, liver, pleura and heart in descending order. The commonest CT patterns for metastases were multiple nodules for lung, effusion for pleura and heart and osteolytic lesions for bone.


Journal Identifiers


eISSN: 1937-8688