MR Imaging of Adrenal Lesions

Aoife Kilcoyne, MB BCh BAO, FFR, RCSI; Shaunagh McDermott, MB BCh BAO, FFR, RCSI; Michael A. Blake, MB BCh BAO, MRCPI, FRCR, FFR, RCSI


Appl Radiol. 2017;46(4) 

In This Article

MR Imaging Techniques

At our institution all abdominal MR imaging studies are performed with a dedicated torso phased array coil (to optimize the signal to noise ratio), high-field imaging, and a power injector when contrast is required. Breath-holding or respiratory navigation sequences are used to minimize artifacts secondary to respiratory motion.

A breath-hold or respiratory-navigated, heavily T2-weighted sequence is performed in the coronal and axial planes to help provide an anatomic roadmap. A breath-held, dual-echo T1-weighted, two-dimensional (2D) gradient-echo (GRE) sequence is acquired in-phase (IP) and out-of-phase (OOP) to detect intracellular lipid. Both echoes are acquired within the same breath-hold, allowing reliable coregistration of the corresponding IP and OOP images. This sequence is used to differentiate lipid rich adrenal adenomas from non-lipid containing adrenal lesions. A baseline, unenhanced fat-suppressed, T1-weighted sequence is used to differentiate between hemorrhage and fat, to provide a baseline comparison to postcontrast images to allow determination of presence or absence of contrast enhancement.

Chemical Shift Imaging

Chemical shift imaging (CSI) is the mainstay of MR evaluation of solid adrenal lesions. Chemical shift MRI relies on detecting intracellular lipid in an adenoma for accurate adenoma identification.[11] Chemical shift MRI exploits the different resonant frequencies of fat and water protons in a given voxel, rather than usingattenuation differences as with CT for lesion characterization. Fat protons precess at a lower frequency than water protons. When fat and water molecules occupy the same voxel, differences in precessional frequency quench MR signal in that voxel during opposed-phase gradient-echo MR imaging.[12–16] This decrease in signal intensity is appreciated when OOP images are compared with the corresponding IP images. Korobkin et al[17] demonstrated an inverse linear relationship between the percentage of lipid-rich cells and the relative change in MR signal intensity on chemical shift images. This relationship holds as long as the number of lipid protons in a voxel does not exceed the number of water protons; however, if the lipid protons were to outnumber the water protons, the result would be a smaller drop in signal as it is the cancellation of signals from pairs of lipid and water protons that produces the drop in signal.

The CSI signal loss can be measured quantitatively, as the adrenal-to-spleen chemical-shift ratio, by dividing the lesion-to-spleen signal intensity ratios on the IP images by the OOP images.[18] A CSI ratio of less than 0.71 at 1.5T field strength indicates a lipid-rich adenoma.[15] The alternative adrenal signal intensity index is calculated as [(IP signal intensity – OOP signal intensity)/(IP signal intensity)] x 100%; using this formula, a measurement of greater than 16.5% at 1.5T is consistent with a lipid-rich adenoma.[19] In clinical practice, however, most radiologists evaluate chemical shift change visually or qualitatively using muscle or spleen as the internal reference organ. This has been demonstrated to be as effective as quantitative methods.[20,21] The liver should not be used as the internal reference organ, as it can also demonstrate drop-off of OOP signal with fatty infiltration of the liver.[20,22] The sensitivity and specificity of CSI for the differentiation of adrenal incidentalomas are reported at 81% to 100% and 94% to 100%, respectively.[14,21,23] It should be remembered that adrenocortical carcinoma, pheochromocytoma, as well as clear cell renal cancer and fat-containing HCC metastases may occasionally demonstrate signal loss on OOP images.[11]

Dynamic Contrast Enhancement

Little has been published regarding the utility of gadolinium-enhanced MRI in characterizing adrenal masses. One study evaluating the MRI enhancement features of adrenal masses found that neither absolute enhancement nor gadolinium washout kinetics were useful in differentiating adenomas from non-adenomas.[16] Another, more recent study[24] concluded that although the diagnostic value of dynamic MRI is lower than chemical shift MRI, contrast enhancement and time-to-peak and wash-in rates derived from SI-time curve of dynamic MRI may be contributory to the results of chemical shift MRI in atypical cases.