Non-inferior low-dose coronary computed tomography angiography image quality with knowledge-based iterative model reconstruction for overweight patients.

In Kyung Park,Tae Hoon Kim,Joohee Lee,Chul Hwan Park,Jeffrey Park,Kyunghwa Han,Chisuk Oh,Li Zeng

doi:10.1371/journal.pone.0209243

Abstract

We investigated the feasibility of low-dose coronary computed tomography angiography (CCTA), using a prospective electrocardiogram (ECG)-triggered axial scan protocol, knowledge-based iterative model reconstruction (IMR), and fixed tube current, in overweight subjects. Forty non-overweight (group A; body-mass index [BMI] < 25 kg/m2) and 40 overweight individuals (group B; BMI = 25–30 kg/m2), who underwent CCTA for coronary artery disease screening, were retrospectively and consecutively enrolled. A 64-slice CT scanner was used at 100-kVp tube voltage and 150-mA tube current, and images were reconstructed using IMR techniques. Image noise, attenuation at the aorta, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) at the proximal right and left main coronary arteries (pRCA and LMCA) were calculated. CCTA images were qualitatively evaluated using a four-point scale (1, poor; 4, excellent) and analyzed using a non-inferiority test with a pre-defined non-inferiority margin of -0.2. The mean CCTA radiation dose (Group A: 1.33 ± 0.02 mSv; Group B: 1.35 ± 0.10 mSv; p = 0.151) and mean aortic root CT attenuation values (Group A: 447.9 ± 81.6 HU; Group B: 439.5 ± 63.6 HU; p = 0.571) did not differ significantly between the two groups. The mean noise in groups A and B was 26.0 ± 4.8 HU and 29.2 ± 4.4 HU, respectively (p = 0.005). The noise reduction ratio in the groups, compared to filtered back projection, was 65.0% and 68.1%, respectively. The mean grade of image quality did not differ significantly (3.75 ± 0.04 vs. 3.71 ± 0.04, p = 0.478). Group B CCTA image quality was non-inferior (mean difference = -0.043, 95% CI = -0.162–0.077) to that of Group A. We concluded that low-dose CCTA with prospective ECG-triggering and IMR might be applied to overweight subjects, as well as to normal-weight subjects, by using a fixed tube current without an increase in tube current based on the patient’s body size.

Highlights

Exposure to ionizing radiation is the main drawback of coronary computed tomography angiography (CCTA), which detracts from the advantages of CCTA, such as non-invasiveness and high diagnostic accuracy [1]
Excellent interobserver reliability was proven in the quantitative analysis of CCTA image quality
Our study showed that CCTA with prospective ECG-triggering and iterative model reconstruction (IMR) can be applied to overweight subjects as well as to normal weight subjects, using a fixed tube current, without the need for tube current modulation based on the patient’s body size

Summary

Introduction

Exposure to ionizing radiation is the main drawback of coronary computed tomography angiography (CCTA), which detracts from the advantages of CCTA, such as non-invasiveness and high diagnostic accuracy [1]. Advanced iterative reconstruction algorithms, such as model-based iterative reconstruction (MBIR) or knowledgebased iterative model reconstruction (IMR), were introduced to overcome such limitations. These approaches attempt to identify the image that is the best fit to the original data, while reducing image noise. IMR, which uses a fully iterative algorithm and knowledge-based approach, can reduce reconstruction time and has become available for routine clinical practice. This algorithm can reduce image noise by up to 80% as compared with FBP in clinical CCTA studies [11]. Since radiation dose is proportional to the tube current [13], it is essential to optimize the tube current

Objectives

Methods

Results

Conclusion