Breast Lesions Detected via Molecular Breast Imaging: Physiological Parameters Affecting Interpretation

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Molecular breast imaging (MBI) is utilized for breast cancer detection.
• Understanding the positive predictive value (PPV) of MBI findings is crucial for accurate diagnosis.
• Descriptor characteristics of MBI require evaluation for their correlation with cancer detection rates.

Purpose of the Study:

• To investigate the relationship between MBI findings and the positive predictive value (PPV) for breast cancer.
• To assess how lesion patterns, background parenchymal uptake (BPU), and breast density influence cancer detection accuracy.

Main Methods:

• A retrospective analysis of 193 suspicious MBI findings from 153 women was conducted.
• Associations were examined between lesion characteristics (mass vs. nonmass), BPU (homogeneous vs. heterogeneous), breast density (dense vs. non-dense), and their respective PPVs.
• Statistical analysis was performed to determine the significance of these correlations.

Main Results:

• Overall PPV1 was 71.9% and PPV3 was 67.4%.
• Mass findings showed higher suspicion for cancer than nonmass findings (P<.001), but PPV3 was comparable (71.7% vs. 61.3%).
• Nondense breasts had significantly higher PPV3 (85.4%) compared to dense breasts (60.6%) (P=.0025).
• No significant difference in PPV3 was observed between homogeneous and heterogeneous BPU (72.5% vs. 60.7%).
• Dense breasts were more associated with heterogeneous BPU.

Conclusions:

• MBI findings, including mass/nonmass characteristics and BPU, are not significant predictors of malignancy.
• Dense breast tissue is associated with lower MBI predictability and a higher likelihood of heterogeneous BPU.
• Further research may refine the interpretation of MBI in dense breasts.