Transradial versus transfemoral access for diagnostic cerebral angiography: Lesion-specific radiation metrics and outcomes.

OBJECTIVE

To compare system-reported radiation metrics and diagnostic outcomes of right transradial access (TRA) versus transfemoral access (TFA) for first-time diagnostic whole-brain digital subtraction angiography (DSA) performed under a modified standardized protocol, and to determine whether access-related differences vary according to anticipated catheterization workflow.

METHODS

This retrospective single-center cohort included consecutive adults undergoing first-time whole-brain DSA for moyamoya disease (MMD), intracranial aneurysm (IA), brain arteriovenous malformation (AVM), or dural arteriovenous fistula (DAVF) between September 2021 and August 2025. Access selection was pragmatic rather than randomized. Radiation metrics, including total fluoroscopy time (TFT), total radiation exposure dose (TRED), and fluoroscopy-only dose (TFD), were extracted from the same angiography platform. Disease-stratified comparisons were analyzed as workflow-stratified comparisons according to predefined target-vessel requirements. The DAVF subgroup requiring external carotid artery (ECA) branch catheterization was analyzed exploratorily.

RESULTS

Among 1055 included patients (TRA, n = 635; TFA, n = 420), overall diagnostic adequacy was 98.9% (1043/1055). In IA angiography, TRA was associated with lower TFT (235 ± 128 vs 278 ± 122 s), TRED (877.8 ± 283.8 vs 1074.0 ± 402.9 mGy), and TFD (60.4 ± 18.1 vs 73.7 ± 23.4 mGy) than TFA (Holm-adjusted P < 0.001 for all). Radiation metrics were similar for MMD, AVM, and overall DAVF. In the exploratory DAVF subgroup requiring ECA branch catheterization (TRA, n = 15; TFA, n = 8), the overall multivariate comparison was borderline (Hotelling's T² P = 0.0498), and right TRA showed higher unadjusted TFT and TFD; however, the univariate differences did not remain statistically significant after Holm correction. No clinically evident symptomatic radial artery occlusion, femoral pseudoaneurysm, femoral arteriovenous fistula, or refractory puncture-site pain was recorded.

CONCLUSION

In this nonrandomized single-operator cohort, access-related radiation differences appeared workflow-dependent rather than purely disease-dependent. The findings support tailoring access selection to anticipated catheterization demands while avoiding causal claims about the superiority of either access route.