Characterization of Resting State Default Mode Network in Individuals with HbSS and HbSC, and Healthy Controls Using 7-Tesla Human MRI

Abstract

Sickle cell disease (SCD) is complicated by accelerated brain aging and cerebrovascular complications leading to structural brain damage and altered resting state functional connectivity (RS-FC). Investigation of the Default Mode Network (DMN), the most studied of the RS-FC networks, by functional MRI may provide insight into the neurological and neuropsychiatric complications of SCD. Prior studies have largely focused on the HbSS genotype of SCD as opposed to the HbSC genotype, which accounts for 30% of the total SCD cases in the US. Both genotypes are expected to impact brain disease, but in different manner.The aim of this study was to characterize the DMN in patients with HbSC, HbSS and healthy race and age-matched controls using state of the art 7-Tesla (7T) magnetic resonance imaging (MRI) with high contrast- and signal- to-noise ratios. On the day of the MRI scanning, participants were administered the brief pain inventory questionnaire (BPI), the digit symbol substitution test of cognitive function, and underwent basic blood work. Controls (n= 43, 53% female, mean age = 36+/-12 years), patients with HbSS (n=19, 63% female, mean age= 36+/-12) and HbSC (n=16 (50% female, mean age=36+/-13) were scanned using a 7T Siemens Magnetom scanner with the Tic Tac Toe radiofrequency coil system (Ibrahim et al., 2013). We obtained whole brain structural T1-weighted images and blood oxygenations level dependent resting state functional MRI (RS-fMRI). During the RS-fMRI, participants were instructed to keep their eyes open, look at a fixation cross, and lie without motion in the scanner. To identify significant differences in RS-FC between the groups, we conducted voxel-wise nonparametric analysis of variance. In the cortical areas with significant clusters, we ran ad-hoc t-tests (controls vs. HbSS, HbSC vs. HbSS, 1000 permutations, uncorrected cluster forming threshold = p < 0.001, family-wise error (FWE) rate=0.05) and we added age and sex as confounders. Then, we correlated RS-FC and the following variables: age, hemoglobin, BPI total, and DSST score.We found a significant difference between the three groups in the DMN (Fig 1) (p=0.0002). The ad-hoc t-tests showed that in the DMN, the RS-FC for the medial prefrontal cortex (mPFC) was significantly higher for the controls when compared to patients with HbSS (p=0.002). RS-FC for the mPFC was also significantly higher in patients with HbSC as compared to HbSS (p=0.001). There was a non-significant trend towards lower RS-FC in patients with HbSC as compared to controls. We also found weak non-significant negative correlation, (r= -0.24, p=0.1315), a strong significant negative correlation (r=-0.5, p=0.04) and very strong significant negative correlation (r=-0.8, p=0.001) between age and RS-FC in the mPFC for controls, HbSC, and HbSS patients, respectively. (Fig 2). We did not find any significant correlation between the RS-FC in the three groups and Hb levels, BPI total, and DSST score.In this work and for the first time, using 7T human MRI we identified altered RS-FC between SCD genotypes and between controls and patients with HbSS. It is notable that the RS-FC for HbSC patients was intermediate between that of the other two groups. Our results imply the RS-FC levels might be susceptible to disease genotype and not necessarily Hb levels, cognition and pain. The finding of a stronger correlation between age and FC in patients with SCD as compared to controls corroborates the notion that accelerated brain aging is operant in SCD. [Display omitted] DisclosuresNovelli: Novartis Pharmaceuticals: Consultancy.