Abstract TP25: Model-Free Dynamic Contrast-Enhanced MRI in Cerebral Small Vessel Disease: Test-Retest Reliability and Association With WMH Lesion Burden

Abstract

Introduction: Dynamic contrast enhanced (DCE)-MRI has been used to study blood-brain barrier (BBB) disruption in cerebral small vessel disease (CSVD), but complex kinetic modeling of DCE MRI is sensitive to noise and is not robust in detecting subtle BBB impairments. In contrast, model-free DCE is simple and does not require arterial input function and complex computation. In this study, we aimed to evaluate the test-retest repeatability of model-free DCE and its association with white matter hyperintensity (WMH) burden. Methods: Patients with CSVD were recruited. DCE-MRI, FLAIR, and T1 MPRage scans were acquired. A subgroup of patients had two repeated DCE MRI scans with an average interval of 41 days. WMH segmentation was performed to obtain the relative volume of WMH (rVwmh). Ten subjects had DCE-MRI data in two sessions for test-retest repeatability analysis. DCE-MRI were acquired using a Siemens volumetric interpolated breath-hold examination (VIBE) gradient echo sequence with a temporal resolution of 49 seconds and an acquisition time of ~20 minutes. The wash-out slope (WOS), which indicates the rate of contrast agent leaving the tissue, was computed. Results: 64 patients (mean (SD) age = 68.6 (8.51), 37 females) with CSVD were studied. WOS demonstrated good test-retest repeatability for WM and GM with an intra-class correlation coefficient (ICC) of 0.626 and 0.671 in cerebrum WM and GM, respectively. WMH WOS has a significant negative correlation with rVwmh (r=-0.31; p<0.02). Conclusions: A slow contrast wash-out (small WOS) is expected in the presence of BBB impairment. WMH burdens are associated with BBB impairments manifested by decreased WOS. WOS is computationally less intense and demonstrates good repeatability. It may offer an alternative to model-based DCE in evaluating BBB dysfunction in patients with CSVD.