Purpose: This study aims to compare the accuracy of proton-density fat fraction (PDFF) measurements with chemical shift magnetic resonance imaging (CSI) for quantifying the fat content of adrenal nodules and for differentiating adenomas from nonadenomas. Materials and methods: Oil-saline phantom measurements was performed to compare the correlation between PDFF and CSI in detecting and quantifying fat content. 43 consecutive patients who had known adrenal nodules were imaged on a 3.0-T MR scanner. PDFF was measured, and the signal intensity (SI) index (SII), SI adrenal-to-liver ratio (ALR) and SI adrenal-to-spleen ratio (ASR) of the adrenal nodules were calculated. Results: In the phantom study, PDFF ranged from 12.6% to 99.1% and the SII was between 0.72 and 1.23. There was good correlation between these two methods (R square = 0.972, p < 0.0001). The PDFF of adrenal adenoma was significantly increased compared with that of nonadenoma (p < 0.001). PDFF was an effective tool for distinguishing adenoma from nonadenoma, with an area under the curve (AUC) of 0.98. In comparing SII, ALR and ASR the AUC was 0.94, 0.95 and 0.93, respectively. No significant difference was noted between these two methods (p > 0.05). Conclusion: PDFF measurements provide an accurate estimation of fat content in discriminating adenomas from nonadenomas compared with CSI, avoiding complicated data calculations and offering a simpler technique using 3T. (C) 2016 Elsevier Ireland Ltd. All rights reserved.