A Novel Imaging Biomarker in Cerebral Blood Flow in Patients with Sickle Cell Disease using fMRI
Poster No:
2608
Submission Type:
Abstract Submission
Authors:
Brianna Kish1, Jinxia Yao1, Andrew John Frels1, Jessica Budde1, Vidhya Vijayakrishnan Nair1, Andrew O’Brien2, Ying Wang2, Yunjie Tong1
Institutions:
1Purdue University, West Lafayette, IN, 2Indiana University School of Medicine, Indianapolis, IN
First Author:
Co-Author(s):
Introduction:
Sickle cell disease (SCD) is a genetic blood disorder characterized by production of abnormal hemoglobin (HgB) known as hemoglobin S (HgS), reducing blood oxygen-carrying capacity (Arkuszewski, 2013, DeBaun, 2006). To compensate the body increases cerebral blood flow but overcompensation can result in cerebral vascular shunting; blood velocity is increased until oxygen off-loading efficiency is reduced (Juttukonda, 2017). Studies have shown that shunting is more common in SCD patients than healthy individuals (Juttukonda, 2021). Systemic low-frequency oscillations (sLFOs) are a subset of low-frequency oscillations (0.01- 0.1 Hz) represented throughout the brain with high correlations and meaningful delay values in resting state BOLD-fMRI (Tong, 2017). In healthy subjects, sLFOs between the internal carotid arteries (ICA) and the superior sagittal sinus (SSS) have delays consistent with brain transit time (~5s) (Hoffmann, 2000). Using BOLD-fMRI, fluctuations in physiology can be measured via deoxy-hemoglobin concentration in blood; SCD is a blood disorder, so this allows changes to be captured (Heeger, 2002). In this study, we aim to investigate the hemodynamic mechanisms of sLFOs in SCD patients and their relationship with blood parameters.
Methods:
Data from 23 participants with SCD were collected. 10 age and gender matched non-SCD controls were also enrolled. Each subject underwent a T1-weighted MRI, a resting state fMRI, and a blood draw. The blood tests included complete blood cell count, reticulocytes, and HgB electrophoresis. The fMRI data were preprocessed using FSL to correct motion artifacts (Jenkinson, 2012). The SSS was identified and masked from the T1w image and then transformed into the fMRI space to extract the time series. The global mean signal (GMean) was generated by averaging the fMRI signal across the whole brain. Both signals were demeaned and filtered (0.01-0.1Hz) to acquire the sLFOs. Cross-correlation was performed between the SSS and GMean to calculate the maximum cross correlation coefficients (MCCC) and delay times. Pearson correlation was performed between the MCCC/delay times and 14 different blood measurements to determine any significant relationships.
Results:
90% of healthy subjects exhibit high positive MCCC values and negative delay times indicating that the GMean leads the SSS. However, only 11/23 SCD subjects displayed this standard behavior (MCCC 0.889, Delay -2.67 s). 12/23 SCD subjects show a negative MCCC value with the SSS leading the GMean (MCCC -0.748, Delay 3.708 s). In addition to a known relation with SCD, HgB, Hct, and HbS have significant correlations to MCCCs and delays. Negative MCCC subjects have significantly lower average HgB/Hct and higher HbS (Fig. 1A-C) compared with positive MCCC subjects. A significant positive correlation is observed between HgB/Hct and MCCC values in SCD patients with positive MCCCs while HbS exhibits a negative correlation to MCCC (Fig. 1D-F). Healthy subjects show a lower correlation. Similarly, the correlations between delays and blood measures in that cohort are weaker (Fig. 1G-I). Based on a particular model (Fig. 2A), fMRI signal flips sign when the oxygen saturation value from the corresponding voxel is greater than or equal to approximately 86% (Tong, 2019). This finding suggests that the SSS or GMean has a sign change. Since the SSS contains more venous blood, it is likely that oxygen saturation in the brain increases exceeding the threshold of 86%, thus flipping the GMean sign (Fig. 2B).
Conclusions:
The present study employed fMRI to investigate the hemodynamics of SCD patients and identified a novel anti-correlation between the SSS-sLFO and the GMean-sLFO. The proposed model linked different degrees of shunting with anti-correlation and explained the high correlation between the MCCC values and blood measures. These findings represent significant advances in understanding the pathophysiology of SCD and offer potential biomarkers for measuring disease severity.
Novel Imaging Acquisition Methods:
BOLD fMRI 2
Physiology, Metabolism and Neurotransmission :
Cerebral Metabolism and Hemodynamics 1
Keywords:
Blood
Cerebral Blood Flow
Data analysis
FUNCTIONAL MRI
Other - Sickle Cell Disease
1|2Indicates the priority used for review
Provide references using author date format
DeBaun, M.R. (2006), 'Etiology of strokes in children with sickle cell anemia', Ment Retard Dev Disabil Res Rev, vol. 12, no. 3, pp. 192–199
Heeger D. J. (2002), 'What does fMRI tell us about neuronal activity?', Nature Reviews Neuroscience, vol. 3, no. 2, pp. 142–151
Hoffmann O. (2000), 'Measurement of Cerebral Circulation Time by Contrast-Enhanced Doppler Sonography', Cerebrovascular Diseases, vol. 10, no. 2, pp. 142–146
Jenkinson M. (2011), 'FSL', Neuroimage, vol. 62, no. 2, pp. 782–790
Juttukonda M.R. (2017), 'Cerebral hemodynamics and pseudo-continuous arterial spin labeling considerations in adults with sickle cell anemia', NMR Biomed, vol. 30, no. 2, p. e3681
Juttukonda M.R. (2021), 'Reduced oxygen extraction efficiency in sickle cell anemia patients with evidence of cerebral capillary shunting', Journal of Cerebral Blood Flow & Metabolism, vol. 41, no. 3, p. 546
Tong Y. (2017), 'Perfusion information extracted from resting state functional magnetic resonance imaging', Journal of Cerebral Blood Flow and Metabolism, vol. 37, no. 2, pp. 564–576
Tong Y. (2019), 'The resting-state fMRI arterial signal predicts differential blood transit time through the brain', Journal of Cerebral Blood Flow and Metabolism, vol. 39, no. 6, pp. 1148–1160