A 3 mL low-dose gadolinium-based contrast agent combined with 3D-TWIST sequence yields diagnostic accuracy equivalent to standard-dose CE-MRV in patients weighing <70 kg: a non-randomized controlled study

To address the safety concerns regarding gadolinium deposition associated with standard-dose gadolinium-based contrast agents in cerebral contrast-enhanced magnetic resonance venography (CE-MRV), this study aims to investigate the feasibility of a low-dose gadolinium protocol combined with a 3D time-resolved angiography with stochastic trajectories (TWIST) sequence for intracranial venous sinus imaging, and to evaluate its impact on image quality and diagnostic accuracy.

A prospective, single-center, non-randomized controlled study was conducted on 196 patients who underwent examinations on a Siemens Skyra 3.0 T MR scanner at Department of Radiology of Suzhou Hospital Affiliated to Nanjing Medical University from November 2022 to May 2024. All participants were divided into 3 groups: Group A (n=20), Group B (n=78), and Group C (n=98). Groups B and C were further stratified by body weight at 10 kg intervals (40 to <100 kg) into subgroups B4-B9 and C4-C9, respectively. Group A served as the reference standard: standard-dose (0.1 mmol/kg) gadopentetate dimeglumine administered undiluted, with a 3D-TWIST sequence using 10% central/20% peripheral k-space sampling. Group B received 3 mL of gadolinium-based contrast agent diluted to a total volume of 15 mL, with the same acquisition parameters as Group A. Group C received 3 mL of gadolinium-based contrast agent diluted to 15 mL, with a 3D-TWIST sequence using 40% central/20% peripheral k-space sampling. Objective comparison of the signal-to-noise ratio (SNR) of the maximum signal intensity in intracranial venous sinuses was conducted between different body weight subgroups of Groups B and C versus Group A. Subjective image quality was independently assessed by 2 senior radiologists using a 3-point scoring scale: 1=intracranial venous sinus segments not visible or indistinguishable; 2=moderate image quality with discernible venous sinus segments; 3=high-quality delineation of intracranial venous sinus segments. Furthermore, pairwise comparisons of diagnostic accuracy for intracranial venous sinus evaluation were performed between different subgroups of Group B and Group C versus Group A, respectively.

Comparison between Group A and Group B: Except for the inferior sagittal sinus in subgroups B4 and B7, the SNR of all venous sinuses in Group A was higher than that in each subgroup of Group B. Comparison between Group A and Group C showed that among the C4, C5, and C6 subgroups, the SNR of the inferior sagittal sinus in the C4 subgroup was higher than that in group A (P<0.0028), while the differences in SNR for the remaining venous sinuses were not statistically significant (all P>0.0028). In the C7, C8, and C9 subgroups, the SNR of each venous sinus was lower than that in group A, except for the inferior sagittal sinus in the C7 subgroup. The SNR of all subgroups within group C was higher than that of the corresponding subgroups in group B with the same body weight. All subjective image quality scores were ≥2, with no significant differences were found between group A and subgroups B4-B6 and C4-C8 (all P>0.00417). The diagnostic accuracy rates for groups A, B, and C were 100.00%, 98.71%, and 100.00%, respectively, with no statistically significant differences between any 2 groups (all P>0.0167).

A low-dose (3 mL) gadolinium-based contrast agent combined with an optimized 3D-TWIST sequence (40% central k-space sampling) is feasible for cerebral CE-MRV and may substantially reduce the risk of gadolinium deposition. This protocol is recommended for patients weighing <70 kg as an alternative to the conventional standard-dose regimen. For patients weighing 70-89 kg, subjective image quality meets diagnostic requirements, however, further improvements in SNR are warranted.