Integrating 4D-flow MRI and computational fluid hemodynamics to noninvasively estimate trans-stenotic pressure gradient in idiopathic intracranial hypertension

Abstract

Venous manometry, though invasive, is the gold standard for selecting idiopathic intracranial hypertension (IIH) patients for venous sinus stenting. Noninvasive flow-based imaging methods show promise but need validation. This study combined patient-specific 4D-flow MRI with computational fluid dynamics (CFD) to estimate trans-stenotic pressure gradients (PGs) in IIH noninvasively, assessing branch effects and sinus velocity distributions. In this retrospective study of 48 IIH patients, transient and steady-state CFD simulations using 4D-flow MRI inlet flows were compared to manometric PGs. Sensitivity analyses tested generic flows, branch omissions, and altered branch flowrates. Velocities were resampled to voxel grids for distribution analysis. Transient CFD agreed well with manometry (bias: −0.2 mmHg), similar to steady-state (−0.4 mmHg). Branches raised PGs up to 4.5 mmHg in high cases but without classification changes. Tributary flows (0.12–0.49 mL/s) had a linear but minimal impact on PG when <1 mL/s. Velocities at stenosis averaged 83.2 ± 41.9 cm/s, and at inlet was 13.0 cm/s (IQR: 9–17.8). Of the voxel-wise velocities, 7.8% exceeded 50 cm/s, 2.3% exceeded 80 cm/s, and 0.6% exceeded 120 cm/s. Thus, 4D-flow MRI–CFD provides accurate noninvasive PG estimation in IIH and omitting venous branches maintains diagnostic accuracy.