3720
Cerebrovascular Reactivity (CVR) mapping using intermittent breath modulation1Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States
Synopsis
Cerebrovascular reactivity (CVR) mapping is typically performed using CO2-inhalation or breath-holding as a vasoactive challenge while collecting BOLD images. Recently, resting-state BOLD has been used to map CVR by utilizing spontaneous fluctuations in breathing pattern, but the results could be noisy in subjects with little fluctuation in their spontaneous breathing. Here we developed a new technique for CVR mapping that does not require gas-inhalation yet provides substantially higher sensitivity than resting-state CVR mapping. This new technique is largely based on resting-state scan, but introduces intermittent modulation of breathing pattern in the subjects to enhance fluctuations in their end-tidal CO2 level.
Introduction
Cerebrovascular reactivity (CVR), an index of vessel’s dilatory function, has provided valuable information in various cerebrovascular conditions1-3. CVR mapping is typically performed using hypercapnic gas inhalation1,2 or breath-holding3 as a vasoactive challenge while collecting BOLD images, but both methods require considerable subject cooperation and could be discomfortable for some subjects. More recently, there have been some attempts to use resting-state BOLD data to map CVR by utilizing spontaneous fluctuations in breathing pattern4-6. However, in subjects who have little fluctuation in their spontaneous breathing, the CVR results could be noisy or even undetectable. Therefore, we aim to develop a new technique for CVR mapping that does not require gas-inhalation yet provides substantially higher sensitivity than resting-state CVR mapping. This new technique is largely based on resting-state scan, but introduces intermittent modulation of breathing pattern in the subjects to enhance fluctuations in their end-tidal CO2 (EtCO2) level. Here we examined the sensitivity and comfort level of this technique.Methods
Study design: Eight healthy young subjects (4 females, age 25.5±5.4y) were scanned on a Philips 3T. In each subject, five CVR mapping techniques were performed, which were 1) free breathing; 2) intermittent breath modulation of 6s/breath (see more explanations below); 3) intermittent breath modulation of 12s/breath; 4) breath-holding; 5) CO2-gas inhalation. The order of the first three tasks was pseudo-randomized across subjects. EtCO2 was recorded during the entire session using a capnograph device.
Breathing tasks: During the free breathing task, the subjects were instructed to breathe at their own pace. During the intermittent breath modulation task, the subjects were asked to breathe at their own pace except for the periods when pacing instructions, e.g. “breathe in”, “breathe out”, appear on the screen. The pacing instructions appear for 12 seconds after every 30-60s free breathing period (Figure 1). Two different pacing frequencies, specifically 6s/breath (3s in/3s out) and 12s/breath (6s in/6s out), were employed in separate scans. In the breath-holding task, the subjects were instructed to hold breath for 20s in each minute. In the CO2-gas inhalation task, the subjects were fitted with a nose clip, and breathed room air and hypercapnia gas (5%CO2) in an interleaved fashion (50s CO2, 70s room air, repeated four times) through a mouthpiece.
CVR scans: BOLD data was collected during each breathing task using identical imaging parameters (TR/TE=1500/21ms, 9.3min for non-CO2 scans and 7min for CO2 scan). To measure the comfort level of each CVR technique, subjects were asked to rate their comfort level after each scan.
Data analysis: CO2-CVR was obtained from the CO2-inhalation task following standard analysis described previously2,7. For non-gas inhalation scans, whole-brain BOLD time course was obtained and low-pass filtered (cutoff at 0.08Hz). General linear model (GLM) analysis was then performed on each voxel with the filtered whole-brain time course as independent variable, voxel-wise BOLD time course as dependent variable, and motion vectors as covariates, which yielded relative CVR map. Absolute whole-brain CVR values were obtained from the GLM analysis using the filtered whole-brain time course and recorded EtCO2 time course. The product between relative CVR map and whole-brain absolute CVR value then yields voxel-wise absolute CVR map.
Results
Figure 2 summarizes comfort levels of breathing during the five CVR scans, ranked in decreasing order. Intermittent breath modulation of 6s/breath was comparable to the most comfortable free breathing scan (p=0.52). Intermittent breath modulation with 12/s breath was slightly less comfortable although not significant. The comfort levels of both breath-holding and CO2 inhalation were significantly lower than free breathing and intermittent breath modulations.
Figure 3 displays the relative CVR maps from all CVR scans of all eight subjects. Visual inspection suggests that intermittent modulation of 6s/breath scan consistently yielded superior CVR maps compared to free-breathing or 12s/breath scans. This is confirmed by the quantitative comparison of the average Z scores (Z statistics from the GLM analysis) shown in Figure 4a, which is listed in descending order. The 6s/breath modulation showed the highest Z scores, while free-breathing showed the lowest.
Using CVR maps obtained with CO2-inhalation as a “gold standard”, we found that intermittent modulation of 6s/breath showed the highest spatial correlation (0.89±0.05) with CO2-CVR map (Figure 4b) and, once again, free-breathing slightly the lowest (0.74±0.21).
Furthermore, absolute CVR values obtained from intermittent modulation of 6s/breath showed a significant correlation with CO2-CVR (Figure 5).
Discussion and conclusion
In this work, we developed a new CVR technique that does not require gas inhalation (thus simpler and more comfortable) while preserving signal sensitivity. This new CVR mapping method may be a practical approach to detect vascular deficits in clinical applications.Acknowledgements
No acknowledgement found.References
1. Mandell DM, Han JS, Poublanc J, Crawley AP, Stainsby JA, Fisher JA, Mikulis DJ. Mapping cerebrovascular reactivity using blood oxygen level-dependent MRI in Patients with arterial steno-occlusive disease: comparison with arterial spin labeling MRI. Stroke. 2008;39(7):2021-8.
2. Yezhuvath US, Uh J, Cheng Y, Martin-Cook K, Weiner M, Diaz-Arrastia R, van Osch M, Lu H. Forebrain-dominant deficit in cerebrovascular reactivity in Alzheimer's disease. Neurobiol Aging. 2012;33(1):75-82.
3. Geranmayeh F, Wise RJ, Leech R, Murphy K. Measuring vascular reactivity with breath-holds after stroke: a method to aid interpretation of group-level BOLD signal changes in longitudinal fMRI studies. Hum Brain Mapp. 2015;36:1755-1771.
4. Golestani AM, Chang C, Kwinta JB, Khatamian YB, Chen JJ. Mapping the end-tidal CO2 response function in the resting-state BOLD fMRI signal: spatial specificity, test-retest reliability and effect of fMRI sampling rate. Neuroimage. 2015;104: 266-277.
5. Jahanian H, Christen T, Moseley ME, Pajewski NM, Wright CB, Tamura MK, Zaharchuk G, Group, S.S.R. Measuring vascular reactivity with resting-state blood oxygenation level-dependent (BOLD) signal fluctuations: A potential alternative to the breath-holding challenge? J Cereb Blood Flow Metab. 2017;37:2526-2538.
6. Liu P, Li Y, Pinho M, Park DC, Welch BG, Lu H. Cerebrovascular reactivity mapping without gas challenges. Neuroimage. 2017;146:320-326.
7. Lu H, Liu P, Yezhuvath US, Cheng Y, Marshall O, Ge Y. MRI mapping of cerebrovascular reactivity via gas inhalation challenges. J Vis Exp. 2014;94.
Figures
