Background: The hypothesis that physiological and biochemical changes associated with proliferating malignant tumors may cause an increase in total tissue sodium concentration (TSC) was tested with non-invasive, quantitative sodium ((23)Na) magnetic resonance imaging (MRI) in patients with benign and malignant breast tumors.
Methods: (23)Na and (1)H MRI of the breast was performed on 22 women with suspicious breast lesions (> or =1 cm) at 1.5 Tesla. A commercial proton ((1)H) phased array breast coil and custom solenoidal (23)Na coil were used to acquire (1)H and (23)Na images during the same MRI examination. Quantitative 3-dimensional (23)Na projection imaging was implemented with negligible signal loss from MRI relaxation, or from radio-frequency field inhomogeneity, in less than 15 min. Co-registered (1)H and (23)Na images permitted quantification of TSC in normal and suspicious tissues on the basis of (1)H MRI contrast enhancement and anatomy, with histology confirmed by biopsy.
Results: Sodium concentrations were consistently elevated in (N = 19) histologically proven malignant breast lesions by an average of 63% compared to glandular tissue. The increase in sodium concentration in malignant tissue was highly significant compared to unaffected glandular tissue (P < 0.0001, paired t-test), adipose tissue, and TSC in three patients with benign lesions.
Conclusion: Elevated TSC in breast lesions measured by non-invasive (23)Na MRI appears to be a cellular-level indicator associated with malignancy. This method may have potential to improve the specificity of breast MRI with only a modest increase in scan time per patient.