Optimization of the Contrast Agent Injection Protocol for Carotid Artery Dynamic Contrast-Enhanced Magnetic Resonance Imaging

Background The injection protocol used in previous carotid artery dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) studies varied. Purpose To investigate the effect of contrast injection protocol and optimize this protocol for carotid artery DCE-MRI. Study Type Prospective. Subjects Digital phantom and seven patients with carotid atherosclerosis. Field Strength/Sequence 3 T, spoiled gradient recalled echo sequence. Assessment Different injection doses (0.01-0.3 mmol/kg) and effective injection rates (0.01-1 mmol/sec) were tested using a digital carotid plaque phantom considering the contrast pharmacokinetics, DCE-MRI imaging, contrast variation and flow-related imaging artifacts, random time delay between the contrast injection and image acquisition, and pharmacokinetic analysis process. For each injection protocol, combining the root mean square relative error (RMSRE) of the measured Ktrans and vP maps within the adventitial vasa vasorum from 10 tested time delays by the root mean square produced RMSREoverall-vv which was used to measure the overall accuracy of the pharmacokinetic parameters. In vivo validation was performed on seven patients with carotid atherosclerosis by imaging them twice using the traditional commonly used protocol and the recommended protocol found by simulation. Statistical Test Student's t-test, chi-square test, and paired t-test, P < 0.05 was considered statistically significant. Results A low region of RMSREoverall-vv with the combination of medium injection dose and low effective injection rate was found. The protocol with injection dose of 0.07 mmol/kg and effective injection rate of 0.06 mmol/sec achieved the minimal RMSREoverall-vv (4.29%), thus was recommended, which showed more accurate arterial input function. Coinciding with the simulation results, this recommended protocol in in vivo experiments produced significantly fewer image artifacts, lower Ktrans and vP (P all <0.05) than traditional protocol which overestimated these parameters in simulation. Data Conclusion The contrast injection protocol influenced the accuracy of the pharmacokinetics parameter estimation in carotid artery DCE-MRI. The injection protocol with injection dose of 0.07 mmol/kg and effective injection rate of 0.06 mmol/sec was recommended. Level of Evidence 2 Technical Efficacy Stage 1