ARRS 2022 Abstracts

RETURN TO ABSTRACT LISTING


1431. Iodine Density Threshold to Discriminate Between Enhancing and Non-Enhancing Renal Lesions With Dual-Layer Dual-Energy CT
Authors * Denotes Presenting Author
  1. Ezra Margono *; Boston Medical Center
  2. Muhammad Qureshi; Boston Medical Center
  3. Avneesh Gupta; Boston Medical Center
Objective:
This study aims to determine the optimum threshold iodine density to distinguish between enhancing and non-enhancing renal masses using dual-layer dual-energy CT (dlDECT).

Materials and Methods:
In this retrospective, HIPAA-compliant, institutional review board-approved study, 295 consecutive renal mass studies obtained with dlDECT (Philips iQon) from 9/5/2018 through 6/8/2021 were reviewed, yielding 33 enhancing solid renal masses (increase of at least 20 Hounsfield units [HU] on nephrogenic phase compared to non-contrast). Thirty-six simple renal cysts from 34 consecutive studies in the same interval served as the control group. Region-of-interest (ROI) density measurements of all lesions in HU were recorded on unenhanced and nephrogenic phases using conventional CT reconstructions. Iodine density ROI measurements of all renal lesions and the abdominal aorta for normalization were obtained on iodine density maps using dedicated reconstruction software (Philips IntelliSpace Portal). All measurements were recorded by a single investigator (E.M.). Mean lesion attenuation values and absolute normalized iodine densities were compared between enhancing and non-enhancing renal lesions using a two-sample t-test. Diagnostic accuracy of iodine thresholds was assessed by calculating sensitivity and specificity, with receiver operating characteristic curve and area under the curve (AUC) analysis and 95% confidence interval.

Results:
There were 29 women and 40 men (mean age 65.8±12.4 years) with 33 enhancing lesions and 36 non-enhancing lesions. The mean HU difference was 72.8±39.7 and 3.9±5.1 HU for enhancing and non-enhancing lesions, respectively. The mean absolute iodine density was 3.2±1.8 mg/ml and 0.19±0.22 mg/ml for enhancing and non-enhancing lesions, respectively, and the difference was statistically significant (p<0.0001). Normalized against the aorta, the mean iodine density was 0.63±0.34 mg/ml and 0.03±0.04 mg/ml for enhancing and non-enhancing lesions, respectively, and the difference was statistically significant (p<0.0001). An absolute iodine threshold of 0.7 mg/mL demonstrated an AUC of 0.999 with a sensitivity 100% and specificity 97.2% for discriminating between enhancing and non-enhancing renal lesions. When normalized to the aorta, an iodine density threshold of 0.115 mg/mL demonstrated an AUC of 0.998, with a sensitivity 100% and specificity 97.2% for discriminating between enhancing and non-enhancing renal lesions.

Conclusion:
Enhancing and non-enhancing renal lesions can be distinguished with very high sensitivity and specificity using absolute and normalized region-of-interest iodine density measurements using dlDECT.