Understanding CBF Changes in Patients with Parkinson’s Disease using 3D pCASL-MRI

257 

Abstract Submission 

Saul Lopez¹, Binu Thomas², Corina Catiul¹, Zoltan Mari³, Virendra Mishra¹

¹The University of Alabama at Birmingham, Birmingham, AL, ²The University of Texas at Southwestern Medical Center at Dallas, Dallas, TX, ³Cleveland Clinic Foundation, Las Vegas, NV

Saul Lopez  
The University of Alabama at Birmingham
Birmingham, AL

Binu Thomas  
The University of Texas at Southwestern Medical Center at Dallas
Dallas, TX

Corina Catiul  
The University of Alabama at Birmingham
Birmingham, AL

Zoltan Mari  
Cleveland Clinic Foundation
Las Vegas, NV

Virendra Mishra  
The University of Alabama at Birmingham
Birmingham, AL

Parkinson's disease (PD) is the second most common progressive neurodegenerative disorder, characterized by the death of dopaminergic neurons in the substantia nigra leading to motor dysfunction(Kalia & Lang, 2015). Arterial Spin Labeling (ASL) MRI has been used to understand both disease severity(Pelizzari et al.) and disease progression(Joshi, Prasad, Saini, & Ingalhalikar, 2023) in PD. Due to the low spatial resolution of the acquisition of ASL-MRI, however, partial volume effects (PVE) occur. PVE bias the conclusion and the repeatability of the measures derived from ASL-MRI, such as cerebral blood flow (CBF) and arterial transit time (ATT)(Chappell et al., 2021). In this study, we investigated the effects of partial volume correction (PVC) to understand the correlations between CBF/ATT and disease severity of participants with PD.

Ten PD and 12 healthy control (HC) patients were recruited from the Cleveland Clinic Center for Brain Health, Las Vegas. 3D multi-delay pseudo-continuous ASL (pCASL) MRI were acquired with the following parameters: field of view (FOV)=224×224 mm^...2, matrix=64×64, 44 axial slices, thickness=3.5 mm, TR/TE=4130/39 msec, labeling duration=1.8 seconds, multiple post-labeling delays (PLD)=[0.5, 1.0, 1.5, 2.0, 2.5] seconds. CBF and ATT maps were generated from the pCASL-MRI images and PVC was performed using the BASIL (Bayesian Inference for Arterial Spin Labeling) toolbox in FSL. The analysis procedure followed the recommendations in the ASL white paper (Alsop et al., 2015) . We statistically compared the differences in CBF and ATT between HC and PD patients in addition to performing correlations between CBF/ATT measures and PD severity (measured by Movement Disorder Society Sponsored Unified Parkinson's Disease Rating Scale (MDS-UPDRS)-III and Hoehn and Yahr (H&Y) scores), disease duration, and levodopa equivalent daily dose (LEDD) in the PD group. All statistical analyses were done using the PALM toolbox from FSL. The results were considered significant at a familywise error (FWE) corrected p-value<0.05. Sex and affected side were used as regressors. Framewise displacement (FD) (Power, Barnes, Snyder, Schlaggar, & Petersen, 2012) was computed to check for head motion.

Handedness and sex were significantly different between the groups. PD patients were 70% right-handed, 20% left-handed, 10% ambidextrous and 90% male, while HC patients were 100% right-handed and 50% male. No significant difference was observed between the FD between the groups. Our analysis did not reveal any significant difference in either CBF or ATT between HC and PD. Non-PVC analysis revealed a statistically significant (p_...corr<0.05) negative correlation between CBF and H&Y scores (Fig.1b and Fig.2a). The regions showing correlations encompassed predominantly frontal and temporal cortices (Fig.1b). These correlations between non-PVC and H&Y were also observed in PVC analysis (Fig.1a and Fig.2a). However, the extent and strength of the correlations were greater as compared to non-PVC analysis (Fig.1a-b and Fig. 2a). Non-PVC corrected CBF showed an unexpected positive correlation with UPDRS-III (Fig.1c and Fig.2b) which was not observed with PVC analysis. ATT measures in PD were not correlated with any clinical measures in our sample.

No significant differences between HC and PD in either CBF or ATT measures suggest that the pCASL-derived measures are highly variable in PD, further suggesting a larger sample to discern these differences. The seemingly false-positive correlation between non-PVC-corrected CBF and MDS-UPDRS-III was not observed in PVC-corrected data albeit a stronger correlation between PVC-corrected CBF and H&Y was observed in our sample. Overall, the results of our study suggest that 3D multi-delay pCASL can be used to measure disease severity in patients with PD but efforts should be made to jointly analyze both PVC and non-PVC CBF and ATT measures.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ¹

Cerebral Metabolism and Hemodynamics ²

Cerebral Blood Flow

Data analysis

Degenerative Disease

Movement Disorder

Neurological