Comparison of MRI features in lipid-rich and lipid-poor adrenal adenomas using subjective and quantitative analysis

Wendy Tu; Rosalind Gerson; Jorge Abreu-Gomez; Amar Udare; Rachel Mcphedran; Nicola Schieda

doi:10.1007/s00261-021-03161-1

Comparison of MRI features in lipid-rich and lipid-poor adrenal adenomas using subjective and quantitative analysis

Abdom Radiol (NY). 2021 Oct;46(10):4864-4872. doi: 10.1007/s00261-021-03161-1. Epub 2021 Jun 12.

Authors

Wendy Tu¹, Rosalind Gerson², Jorge Abreu-Gomez³, Amar Udare⁴, Rachel Mcphedran², Nicola Schieda^{5

6}

Affiliations

¹ Department of Radiology, Brigham and Women's Hospital, Boston, MA, USA.
² Department of Medical Imaging, The Ottawa Hospital, University of Ottawa, Ottawa, ON, Canada.
³ Joint Department of Medical Imaging, The University Health Network, Toronto, ON, Canada.
⁴ Juravinski Hospital, Hamilton Health Sciences, Hamilton, ON, Canada.
⁵ Department of Medical Imaging, The Ottawa Hospital, University of Ottawa, Ottawa, ON, Canada. nschieda@toh.on.ca.
⁶ C1 Radiology, The Ottawa Hospital, University of Ottawa, 1053 Carling Avenue, Ottawa, ON, K1Y 4E9, Canada. nschieda@toh.on.ca.

PMID: 34120206
DOI: 10.1007/s00261-021-03161-1

Abstract

Objective: To compare MR-imaging features in benign lipid-rich and lipid-poor adrenal adenomas.

Materials and methods: With institutional review board approval, we compared 23 consecutive lipid-poor adenomas (chemical shift [CS] signal intensity [SI] index < 16.5%) imaged with MRI to 29 consecutive lipid-rich adenomas (CS-SI index ≥ 16.5%) imaged during the same time period. A blinded radiologist measured T2-weighted (T2W) SI ratio (adrenal adenoma/psoas muscle), dynamic enhancement wash-in (WI) and wash-out (WO) indices, and T2W texture features. Two blinded Radiologists (R1/R2) assessed T2W-SI (relative to renal cortex) and T2W heterogeneity (using 5-Point Likert scales). Comparisons were performed between groups using independent t tests and Chi-square with Holm-Bonferroni correction.

Results: There was no difference in age or gender between groups (p = 0.594, 0.051 respectively). Subjectively, all lipid-rich and lipid-poor adenomas were rated hypointense or isointense compared to renal cortex and T2W-SI did not differ between groups (p = 0.129, 0.124 for R1, R2). Agreement was substantial (Kappa = 0.67). There was no difference in T2W SI ratio (1.8 ± 0.9 [0.5-4.3] lipid rich versus 2.2 ± 1.0 [0.6-4.3] lipid poor, p = 0.139). Enhancement WI and WO did not differ comparing lipid-rich and lipid-poor adenomas (p = 0.759, 0.422 respectively). There was no difference comparing lipid-rich and lipid-poor adenomas T2W heterogeneity judged subjectively (p = 0.695, 0.139 for R1, R2; Kappa = 0.19) or by texture analysis (entropy, kurtosis, skewness; p = 0.134-0.191) with all adenomas except for one rated as mostly or completely homogeneous.

Conclusions: There is no difference in T2W signal intensity, enhancement pattern or T2W heterogeneity judged subjectively or by quantitative texture analysis comparing lipid-poor and lipid-rich adrenal adenomas.

Keywords: Adrenal gland neoplasms; Magnetic resonance imaging; Metastasis; Neoplasm.

MeSH terms

Adenoma* / diagnostic imaging
Adrenal Gland Neoplasms* / diagnostic imaging
Diagnosis, Differential
Humans
Lipids
Magnetic Resonance Imaging
Retrospective Studies
Sensitivity and Specificity

Substances

Lipids