Automated injection improves consistency of brain perfusion parameters in dynamic susceptibility contrast magnetic resonance imaging in dogs in a randomized trial.

Abstract

Objective: To compare perfusion characteristics in morphologically normal canine brains with dynamic susceptibility contrast (DSC) MRI following manual versus automated contrast medium injection.

Methods: Client-owned dogs undergoing routine 3-tesla brain MRI protocol including DSC were prospectively enrolled, randomly allocated to manual or automated contrast medium injection between November 2023 and February 2025. Gadoteric acid (0.1 mmol/kg) was injected at weight-dependent rates: 1 mL/s (1 to 5 kg), 2 mL/s (6 to 15 kg), and 3 mL/s (> 15 kg). Perfusion parameters evaluated included time of contrast medium arrival (T0), time to peak contrast medium enhancement (TTP), mean transit time of contrast medium (MTT), relative cerebral blood flow (relCBF), and relative cerebral blood volume (relCBV) at 3 regions of interest: thalamus, piriform lobe, and semioval center. Mann-Whitney U tests with Holm-Bonferroni correction were used. Rank-biserial correlations were calculated as effect sizes.

Results: Dynamic susceptibility contrast studies of 29 dogs with morphologically normal brains were included (16 manual and 13 automated injections). Manual injection showed longer T0, TTP, and MTT in all regions of interest and higher relCBV in the semioval center compared to automated injection. The relCBF was higher in the thalamus and piriform lobe with automated injection.

Conclusions: With manual injection, the perfusion curves appeared flatter and wider compared to automated injection, resulting in prolonged T0, TTP, MTT, and reduced relCBF.

Clinical relevance: Automated injection achieves superior reproducibility of DSC curves. However, manual injection is also feasible, but wider DSC curves should be expected, and appropriate reference parameters should be considered when evaluating brain pathology.