Poster Z20 - Exploring Brain Connectivity and Pain Variability in Sickle Cell Disease Using 7-Tesla Resting State Imaging

Friday, October 25, 2024

3:30 PM - 4:30 PM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

Joel Dzidzorvi Kwame Disu, MSc (he/him/his)

PhD Candidate (Doctoral Student)
Carnegie Mellon University
Pittsburgh, Pennsylvania, United States

Co-Author(s)

CJ

Charles Jonassaint, PhD MHS

Associate Professor of Medicine with Tenure
Department of Medicine, University of Pittsburgh, PA
NF

Nadim Farhat, PhD

Postdoctoral Associate
University of Pittsburgh, Department of Bioengineering, Pittsburgh, PA
TS

Tales Santini, PhD

Postdoctoral Associate
University of Pittsburgh, Department of Bioengineering, Pittsburgh, PA
TI

Tamer Ibrahim, PhD

Associate Professor
University of Pittsburgh, Department of Bioengineering, Pittsburgh, PA
OO

Olubusola Oluwole, MD

Internal Medicine/Pediatric Specialist
University of Pittsburgh Medical Center, Department of Medicine, Vascular Medicine Institute, Pittsburgh, PA

Primary Investigator(s)

Sossena Wood, PhD (she/her/hers)

Assistant Professor
Carnegie Mellon University
Pittsburgh, Pennsylvania, United States

Introduction: Pain is the most common experience in Sickle Cell Disease (SCD) and the primary reason for healthcare utilization. However, there is significant individual variability in the location, intensity, and perception of SCD pain [1]. As more patients with SCD develop chronic pain, understanding the mechanisms accounting for differences in pain outcomes becomes increasingly important. Chronic pain in SCD varies widely, underscoring the need for personalized treatment strategies tailored to the subjective nature of SCD pain.

Individual differences in pain perception may be driven by differences in brain connectivity. Recent studies have linked chronic pain in SCD to variations in brain connectivity, particularly in the default mode and salience networks at rest. The default mode network is active during rest and self-referential activities, while the salience network integrates sensory, emotional, and cognitive information[2]. Farhat et al. confirmed altered functional connectivity in the medial prefrontal cortex (mPFC) (default mode network) between patients with different SCD genotypes, suggesting that disease severity may influence connectivity[3]. Thus, there is a bidirectional relationship between disease progression and brain function/structure.

Our study aims to use 7-Tesla resting-state data to identify differences in brain activation and functional connectivity in SCD patients at rest and to ascertain whether these differences correlate with their pain diagnosis. We utilized the Painimation app, a novel tool for body mapping that assists with visual representations and dynamic simulations[4][5]. Interpreting the relationship between brain connectivity, pain location, and intensity provides valuable insights into the neural mechanisms underlying chronic pain perception in SCD.

Materials and

Methods: Participants were 40-55 years old, including 41 controls and 27 patients with SCD. Pain location, intensity, and quality were assessed using the Painimation instrument. We acquired a T1-weighted MRPAGE and resting-state EPI functional MRI with a 7T Siemens Magnetom scanner using the Tic Tac Toe RF head coil system[6]. Participants kept their eyes open during the resting-state acquisitions.

Resting-state data were analyzed using CONN software and MATLAB. Functional data were preprocessed with realignment using SPM’s realign and unwarp procedure, coregistration to a reference image with a 6-parameter transformation, and resampling with B-spline interpolation for motion and magnetic susceptibility correction. Temporal misalignment was corrected with SPM's slice-timing correction using sinc temporal interpolation. Outlier scans were identified with ART based on framewise displacement and global BOLD signal changes. Data were normalized into MNI space, segmented into tissue classes, and resampled to 2 mm isotropic voxels using SPM's unified segmentation and normalization algorithm. Functional data were smoothed with a 3 mm FWHM Gaussian kernel and denoised using a bandpass frequency filter of 0.008-0.12 Hz [6].

To identify differences in resting-state functional connectivity between controls and patients, we performed first- and second-level analyses, measuring connectivity strength using Fisher-transformed bivariate correlation coefficients (z-scores) from a weighted general linear model. Seed-based analysis focused on the default mode and salience networks. Imaging results were thresholded at p< 0.001 voxel level and p< 0.05 cluster size (family-wise error corrected). We compared connectivity results with Painimation data (pain intensity scores) using correlation coefficients (r > 0.5 for strong correlation).

Results, Conclusions, and Discussions: Our second-level analysis represents significant differences in the anterior cingulate cortex (ACC) between controls and patients with SCD. We observed increased functional connectivity between the ACC and frontal operculum cortex in the salience network for SCD (p=0.000081). For the most reported location by patients (lower back chest), we observed increased functional connectivity between the mPFC and the right parietal operculum cortex (p=0.000100) and the ACC and the left middle temporal gyrus (p=0.000016) in the default and salience networks respectively for SCD.

The parietal operculum cortex plays an important role in the cortical processing and the localization of pain therefore the observation of increased connectivity between the mPFC and parietal operculum may reflect enhanced awareness of the painful sensations experienced by patients with SCD and lower back chest pain [7]. The middle temporal gyrus (MTG) plays a role in the regulation, cognition, and experience of pain. The increase in ACC-left MTG connectivity could indicate enhanced integration and processing of the pain-related information experienced by the SCD population [8].

When we examined the association between connectivity and painimation data (pain intensity scores), our comparative analysis revealed a negative correlation between functional connectivity and pain intensity scores for patients with different pain locations (r=-0.042, p=0.835). While this correlation is suggestive, it's crucial to acknowledge that our sample size may not have provided sufficient statistical power to detect significant correlations, but also the heterogeneity in pain may be genotype-specific. Therefore, further investigation with a larger dataset is required to validate and strengthen these preliminary findings. Despite the low statistical significance between neuroimaging and Painimation data, making correlations in the brain enhances our understanding of the complexities associated with chronic pain perception in SCD. Our findings emphasize the importance of continued research better to address the complex nature of pain in SCD and improve patient outcomes by providing a foundation for targeted interventions based on neural activity patterns. Identifying brain biomarkers for pain location and perception could enable objective pain assessment, leading to improved clinical care and management for patients with SCD.

Acknowledgements (Optional):