ARRS 2022 Abstracts

RETURN TO ABSTRACT LISTING


1206. Holistic Multi-Parametric MRI Scoring System-based Approach in Differentiating Adrenal Adenoma From Other Adrenal Lesions
Authors * Denotes Presenting Author
  1. Ishaq Al Salmi; Oman Medical Specialty Board
  2. Fatma Al Hajri *; Oman Medical Specialty Board
Objective:
This study aims to propose and validate a holistic multiparametric MRI (mp-MRI) scoring system to differentiate adrenal adenomas from other adrenal lesions.

Materials and Methods:
A retrospective observational study was performed after obtaining ethical approval in a single center. All patients who underwent adrenal MRI and had a confirmed diagnosis of adrenal adenoma either by pathology or typical CT or MRI findings or other diagnosis confirmed by pathology were included. Initially, a scoring system was designed based on previously published typical MR imaging features of adrenal adenoma. Subsequently, the MR images were blindly reviewed by a single fellowship-trained radiologist, and each adrenal lesion was scored according to their imaging features. A total score of 3 or more was considered diagnostic for adrenal adenoma. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated.

Results:
A total of 77 patients (22 men and 55 women) were included (age range 19-84 years, mean 39) with a total of 86 lesions. Of those, 70 lesions were adenomas, and 16 lesions were non-adenomas. On the basis of using a score of 3 and above as diagnostic value for adrenal adenoma, the calculated sensitivity, specificity, PPV, and NPV were 98.6%, 93.8%, 98.6%, and 93.8%, respectively. Interestingly, we found that homogenous signal on T2 weighted images is a highly specific imaging feature for adrenal adenomas that was not previously described, to our knowledge.

Conclusion:
Combining different MRI features of adrenal lesions yields a higher diagnostic accuracy in differentiating adrenal adenomas from other adrenal lesions. Using this proposed scoring system can be helpful; however, further validation is required.