Modified approach to the characterization of adrenal nodules using a standard abdominal magnetic resonance imaging protocol.
Radiol Bras. 2017 Jan-Feb;50(1):19-25
Authors: Matos AP, Semelka RC, Herédia V, AlObaidiy M, Gomes FV, Ramalho M
OBJECTIVE: To describe a modified approach to the evaluation of adrenal nodules using a standard abdominal magnetic resonance imaging protocol.
MATERIALS AND METHODS: Our sample comprised 149 subjects (collectively presenting with 132 adenomas and 40 nonadenomas). The adrenal signal intensity index was calculated. Lesions were grouped by pattern of enhancement (PE), according to the phase during which the wash-in peaked: arterial phase (type 1 PE); portal venous phase (type 2 PE); and interstitial phase (type 3 PE). The relative and absolute wash-out values were calculated. To test for mean differences between adenomas and nonadenomas, Student's t-tests were used. Receiver operating characteristic curve analysis was also performed.
RESULTS: The mean adrenal signal intensity index was significantly higher for the adenomas than for the nonadenomas (p < 0.0001). Chemical shift imaging showed a sensitivity and specificity of 94.4% and 100%, respectively, for differentiating adenomas from nonadenomas. Of the adenomas, 47.6%, 48.5%, and 3.9%, respectively, exhibited type 1, 2, and 3 PEs. For the mean wash-in proportions, significant differences were found among the enhancement patterns. The wash-out calculations revealed a trend toward better lesion differentiation for lesions exhibiting a type 1 PE, showing a sensitivity and specificity of 71.4% and 80.0%, respectively, when the absolute values were referenced, as well as for lesions exhibiting a type 2 PE, showing a sensitivity and specificity of 68.0% and 100%, respectively, when the relative values were referenced. The calculated probability of a lipid-poor lesion that exhibited a type 3 PE being a nonadenoma was > 99%.
CONCLUSION: Subgrouping dynamic enhancement patterns yields high diagnostic accuracy in differentiating adenomas from nonadenomas.
PMID: 28298728 [PubMed - in process]