Quantitative Analysis of Iodine Image of Dual-energy Computed Tomography at Rest

Purpose: Dual-energy computed tomography (DECT) can be used for visual determination of iodine distribution in the myocardium (iodine image); however, the accuracy and reproducibility of the process remains debatable. Because of the low contrast-to-noise ratio of CT, we hypothesized that quantitative measurement may be more accurate for detecting myocardial ischemia. In this study, we evaluated our quantitative method by comparing it with visual analysis using 99mTc-tetrofosmin (TF) stress-rest single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) as the reference standard. Materials and Methods: Forty-three patients who had a significant stenosis on cardiac rest DECT and had received 99mTc-TF stress-rest SPECT MPI within 1 month were retrospectively analyzed. The regions of interest were set on iodine images in accordance with the American Heart Association (AHA) 17-segment model (a total of 731 segments). The regions of interest values were divided by the amount of iodine (mg) per unit weight (kg) and defined as perfusion value (perfusion value analysis). All segments were also visually analyzed and receiver operating characteristic curve analysis performed to identify the superior analysis. Results: The receiver operating characteristic curve analysis showed that perfusion value analysis is significantly superior to visual analysis [the area under the curve: 0.921 (95% confidence interval, 0.860-0.981) versus 0.685 (95% confidence interval, 0.580-0.791), respectively, P<0.05], with 93.8% sensitivity, 99.1% specificity, 98.9% accuracy, 83.3% positive predictive value, and 99.7% negative predictive value (P<0.01). Conclusions: Quantitative analysis of the iodine image of rest DECT, called perfusion value analysis, is more accurate than visual analysis when compared with 99mTc-TF SPECT MPI as the reference standard.