Detection of recurrent breast carcinoma using unenhanced breast MRI

Abstract

Abstract Background Early detection of locally recurrent breast carcinoma has been shown to significantly improve long-term survival. Surgery and radiation therapy made treated breast prone to several modifications. This can complicate the interpretation of ultrasound and mammographic images, especially when local recurrence is suspected. The aim of this work is to assess the role of unenhanced MRI (T1WI, T2WI, STIR, and DWI) in differentiating recurrent breast cancer from benign post-operative lesions. Results The presence of fat SI within the lesions had 100% sensitivity, 90.9% specificity, 94.1% PPV, 100% NPV, and 96.2% accuracy in differentiating fat necrosis from recurrent breast carcinoma. A cutoff ADC value of 1 × 10–3 mm2/s for observer one had 80% sensitivity, 90.9% specificity, and 88.9% accuracy in diagnosis of recurrent breast carcinoma. For observer two, a cutoff ADC value of 1.25 × 10–3 mm2/s had sensitivity of 80%, specificity of 88.6%, and diagnostic accuracy of 87.03% in differentiating recurrent breast carcinoma from benign post-operative changes. Unenhanced MRI had 81.8% sensitivity, 97.7% specificity, 90% PPV, 95.5% NPV, and 94.5% accuracy in the diagnosis of recurrent breast carcinoma. Conclusion Unenhanced MRI including T1WI, T2WI, STIR, DWI, and ADC map had high sensitivity, specificity, and diagnostic accuracy in diagnosis of recurrent breast carcinoma and differentiating it from benign post-operative changes.

Keywords

• Unenhanced MRI
• Recurrent breast carcinoma
• Benign post-operative changes
• DWI
• ADC
• T1WI