Reduction of radiation exposure and improvement of image quality with BMI-adapted prospective cardiac computed tomography and iterative reconstruction

Abstract

Purpose: To assess the impact of body mass index (BMI)-adapted protocols and iterative reconstruction algorithms (iDose) on patient radiation exposure and image quality in patients undergoing prospective ECG-triggered 256-slice coronary computed tomography angiography (CCTA).

Methods: Image quality and radiation exposure were systematically analyzed in 100 patients. 60 Patients underwent prospective ECG-triggered CCTA using a non-tailored protocol and served as a 'control' group (Group 1: 120 kV, 200 mAs). 40 Consecutive patients with suspected coronary artery disease (CAD) underwent prospective CCTA, using BMI-adapted tube voltage and standard (Group 2: 100/120 kV, 100-200 mAs) versus reduced tube current (Group 3: 100/120 kV, 75-150 mAs). Iterative reconstructions were provided with different iDose levels and were compared to filtered back projection (FBP) reconstructions. Image quality was assessed in consensus of 2 experienced observers and using a 5-grade scale (1=best to 5=worse), and signal- and contrast-to-noise ratios (SNR and CNR) were quantified.

Results: CCTA was performed without adverse events in all patients (n=100, heart rate of 47-87 bpm and BMI of 19-38 kg/m2). Patients examined using the non-tailored protocol in Group 1 had the highest radiation exposure (3.2±0.4 mSv), followed by Group 2 (1.7±0.7 mSv) and Group 3 (1.2±0.6 mSv) (radiation savings of 47% and 63%, respectively, p<0.001). Iterative reconstructions provided increased SNR and CNR, particularly when higher iDose level 5 was applied with Multi-Frequency reconstruction (iDose5 MFR) (14.1±4.6 versus 21.2±7.3 for SNR and 12.0±4.2 versus 18.1±6.6 for CNR, for FBP versus iDose5 MFR, respectively, p<0.001). The combination of BMI adaptation with iterative reconstruction reduced radiation exposure and simultaneously improved image quality (subjective image quality of 1.4±0.4 versus 1.9±0.5 for Group 2 reconstructed using iDose5 MFR versus Group 1 reconstructed using FBP, p<0.05).

Conclusions: Prospective ECG-triggered 256-slice CCTA allows for visualization of the coronary artery tree with high image quality within a wide range of heart rates and BMIs. The combination of BMI-adapted protocols with iterative reconstruction algorithms can reduce radiation exposure for the patients and simultaneously improve image quality.