IMAGING

doi: 10.1590/S1677-553820100002000018

Utility of diffusion-weighted MRI in characterization of adrenal lesions
Miller FH, Wang Y, McCarthy RJ, Yaghmai V, Merrick L, Larson A, Berggruen S, Casalino DD, Nikolaidis P
Department of Radiology, Northwestern Memorial Hospital, Northwestern University Feinberg School of Medicine, Chicago, IL, USA
AJR Am J Roentgenol. 2010; 194: W179-85.

• Objective: The purpose of our study was to evaluate the utility of apparent diffusion coefficient (ADC) values for characterizing adrenal lesions and determine if diffusion-weighted imaging (DWI) can distinguish lipid-rich from lipid-poor adenomas.
  Materials and Methods: We retrospectively evaluated 160 adrenal lesions in 156 patients (96 women and 60 men; mean age, 63 years). ADCs and signal intensity (SI) decrease on chemical shift imaging were measured in adrenal lesions with a wide variety of pathologies. Lipid-rich and lipid-poor adenomas were identified by unenhanced CT. The overall predictive power of ADC, SI decrease, and lesion size were determined by receiver operating characteristic (ROC) analysis. Areas under the ROC curve (AUC) were compared for equivalence using nonparametric methods. Sensitivity, specificity, and positive and negative predictive values were calculated. Correlation coefficients were used to assess ADCs versus percentage SI decrease and ADCs versus CT attenuation.
  Results: ADCs of adrenal malignancies (median, 1.67 x 10(-3) mm(2)/s; interquartile range, 1.41-1.84 x 10(-3) mm(2)/s) were not different compared with those of benign lesions (1.61 x 10(-3) mm(2)/s; 1.27-1.96 x 10(-3) mm(2)/s; p > 0.05). Cysts (2.93 x 10(-3) mm(2)/s; 2.70-3.09 x 10(-3) mm(2)/s) showed higher ADCs than the remaining adrenal lesions (p < 0.05). The median ADCs of lipid-rich adenomas did not differ from those of lipid-poor ones (p > 0.05). The CT attenuation had no negative or positive correlation with the ADCs of adrenal adenomas (r = -0.05, p = 0.97).
  Conclusion: Unlike lesion size and percentage decrease in SI, the ADCs were not useful in distinguishing benign from malignant adrenal lesions. Lipid-poor adenomas could not be distinguished from lipid-rich adenomas and all other nonfatty lesions of the adrenal gland with DWI.

• Editorial Comment
  Diffusion-Weighted MRI (DWI), is a technique used to detect the state of molecular translational motion of water in the tissue. In some tumors, densely packed malignant cells, causes restricted diffusion of water relative to that of normal tissue. DWI is quantified by the apparent diffusion coefficient map-ADC. Since apparent diffusion coefficient (ADC) reflects primarily diffusion coefficient of extra-cellular water, ADC values tend to be lower for tumors compared to normal tissue. Contrary to cancer, in benign lesions, extra-cellular space volume is higher, thus ADC values are higher as well. For this reason, DWI is an important complementary tool in the evaluation of pathologic conditions in the abdomen and is increasingly used in routine imaging. The authors of this study showed that lipid-poor adenomas could not be distinguished from lipid-rich adenomas and all other non-fatty lesions of the adrenal gland with DWI. They showed that ADCs were not useful in distinguishing benign from malignant adrenal lesions. Similarly recent report has been shown that this technique has also limitation in other abdominal organs since a lesion with restricted diffusion was found to be benign in about 22% of the lesions (1). Fortunately, radiological characterization of an adrenal incidentaloma can be done with high sensitivity and specificity using well established techniques such, CT attenuation without contrast enhancement, wash-out CT technique and chemical-shift MR imaging. Thus, further evaluation with diffusion-weighted MRI is not essential.
  Reference
  1. Sebastian Feuerlein S, Pauls S, Markus S. Juchems MS, et .al, Pitfalls in Abdominal Diffusion-Weighted Imaging: How Predictive is Restricted Water Diffusion for Malignancy. AJR 2009; 193:1070-1076

Dr. Adilson Prando
Head, Department of Radiology and
Diagnostic Imaging, Vera Cruz Hospital
Campinas, São Paulo, Brazil
E-mail: adilson.prando@gmail.com