Global brain physiology does not impact regional BOLD activations Yufen Jennie Chen Xue Wang Todd B. Parrish 1 53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
Jan 04, 2016
1
Global brain physiology does not impact regional BOLD activations
Yufen Jennie Chen
Xue Wang
Todd B. Parrish
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
2
Disclosure
The authors have no relevant financial interest or relationship to disclose with regard to the subject matter of this presentation.
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
3
Purpose Blood-oxygenation level dependent (BOLD) fMRI is a
complex interplay between multiple physiological processes including cerebral blood flow, cerebral blood volume, oxygen consumption and neural activity. Intersubject variability in baseline local CBF accounts for
variability in BOLD response to tasks1
Recently developed MRI techniques can quantitatively measure global cerebral metabolic rate of oxygen (CMRO2)2
The purpose of this study is to investigate whether global CMRO2 will affect the amplitude of task-related BOLD response
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
4
Materials & MethodsSubjects
Group 1: 9 subjects (4F, 36 ± 9 years old)Group 2: 8 subjects (4F, 29 ± 4 years old)
TasksSynonyms task (group 1): alternating 30s epochs
of word pairs and letter strings, repeated 8 times (scan time: 8’06”)
Story comprehension (group 2): alternating 24s epochs of stories and scrambled words, repeated 8 times (scan time: 6’30”)
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
5
Materials & MethodsImaging
3.0T Siemens TIM Trio & 32 channel head coilBOLD fMRI: gradient-echo echo planar imaging
(EPI) Group 1: 1.7x1.7x3mm3, TE=20ms, TR=2s, 35 slices Group 2: 1.7x1.7x3mm3, TE=20ms, TR=2.4s, 42 slices
Venous oxygenation: postsat-TRUST3
1.4x1.4x5mm3, TE=18ms, TR=4s, TEeff=0/40/80/160 ms, 1 slice through sagittal sinus
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
6
Materials & MethodsImaging (Cont.)
Global blood flow: retro-gated phased contrast MRI (PC-MRI) 0.5x0.5x5mm3, Venc=80cm/s, 20 phases, TE=3.81ms,
TR=66.3ms, 1 slice positioned perpendicular to carotids and vertebrals based on non-contrast time-of-flight scan
Multiple scans were run to ensure perpendicular cross-section for all major arteries
Sample position planning of scan for carotids
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
7Materials & MethodsBOLD analysis
Preprocessing: coregistration, motion correction, smoothing (6mm), non-linear spatial normalization to MNI152_T1 template
First-level GLM analysis for each subject Preprocessing and first-level analysis performed
via online pipeline4 Second-level GLM analysis on the contrast
images of the group % signal change extracted from two ROIs using
MarsBar5
Location of ROIs
1
2
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
8Materials & MethodsData analysis for TRUST
Coregistration between TEeff data and motion-correction within TEeff
datapairwise subtraction between control
& tag imagesExtract mean signal of 6 highest
signal voxels in sagittal sinusFit for blood T2 and Yv3
Example of data & fit
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
9
Materials & MethodsData analysis for PC-MRI
Calculate total CBF by summation of flow in bilateral carotids and vertebral arteries, average over cardiac cycle
Estimate brain volume from tissue segmentation of high resolution anatomical T1
Calculate normalized CBF per tissue mass
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
10
Materials & MethodsCMRO2 calculation
CMRO2 = CBF * (Ya-Yv) * Ca
Ya = arterial oxygenation ~ 100% Yv = venous oxygenation (calculated from TRUST)Ca = amount of oxygen/volume of blood = 8.337 mol O2/ml of blood
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
11Results: Correlations with physiology
Total flow is positively correlated with both venous oxygenation (Yv) and cerebral metabolic rate of oxygen (CMRO2)
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
12Results: Correlations with %BOLD
ROI1 ROI2
r p r p
Group 1
CBF (ml/100g/min)
0.33 0.38 0.14 0.71
Yv (%) 0.35 0.36 0.22 0.57
CMRO2 (mol/100g/min)
0.26 0.50 0.10 0.79
Group 2
CBF (ml/100g/min)
0.03 0.94 0.15 0.72
Yv (%) 0.02 0.97 0.28 0.51
CMRO2 (mol/100g/min)
0.14 0.74 -0.37 0.36
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
13
ConclusionGlobal CBF is positively correlated with venous oxygenation and oxygen consumption
None of the physiological metrics correlated with %BOLDGlobal physiology is preserved in spite of local
disturbances such as functional activationCurrent CMRO2 technique is a static measure, which
may not be sensitive to local dynamic changes queried by BOLD
Balance in global physiology may be disrupted with aging and disease of the neurovascular system
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
14
Future WorkImplement technique for local physiological metrics to better understand their effects on functional activity
Extend findings in older healthy controls and chronic stroke patients
53rd ASNR meeting 2015, Chicago IL Poster #:EP-20
15
References1. Liau J., Liu TT. Inter-subject variability in hypercapnic normalization
of the BOLD fMRI response. Neuroimage 45 (2009) 420–430
2. Xu F., Ge Y., Lu H. Noninvasive Quantification of Whole-Brain Cerebral Metabolic Rate of Oxygen (CMRO2) by MRI. Magn Reson Med 62 (2009) 141–148
3. Xu F., Uh J., Liu P., Lu H. On Improving the Speed and Reliability of T2-Relaxation-Under-Spin-Tagging (TRUST) MRI. Magn Reson Med 68 (2012) 198–204
4. Song X., et al. Robust web-based parallel-optimized minimal preprocessing and analysis pipeline for fMRI big data. OHBM 2015, Honolulu, Hawaii
5. Brett M., et al. Region of interest analysis using an SPM toolbox. OHBM 2002, Sendai, Japan