3728
The Effect of End-tidal PETCO2 Clamping on Resting-state fMRI Connectivity1Department of Psychology, University of Toronto, Toronto, ON, Canada, 2Rotman Research Institute at Baycrest, Toronto, ON, Canada, 3Medical Biophysics, University of Toronto, Toronto, ON, Canada
Synopsis
The pressure of end-tidal CO2 (PETCO2) has a significant impact on the blood oxygenation level dependent (BOLD) signal and clamping it leads to detecting additional brain activation in task-based fMRI. However, the effect of PETCO2 clamping on the resting state connectivity has not been yet investigated. In this study we compared whole brain connectivity patterns of 13 healthy subjects during free-breathing and clamped PETCO2. Results showed clamping the PETCO2 removes the global and widespread effect of PETCO2 fluctuation in the gray matter, leading to weaker connectivity and more between-subject variability in the connectivity values.
Introduction
The pressure of end-tidal CO2 (PETCO2) has a significant impact on the blood oxygenation level dependent (BOLD) signal1. In our previous work we showed that removing the effect of PETCO2 improves reproducibility and separability of resting state brain connectivity2. However, retrospective correction of PETCO2 fluctuations may have a different effect on the connectivity patterns than actually clamping out PETCO2 fluctuations during acquisitions. The latter is shown to improve detection of task-related activation in additional brain regions3. The effect of PETCO2 clamping on whole brain connectivity however has not yet been thoroughly investigated. In this study we controlled the resting PETCO2 fluctuation by clamping it to a near constant value and compared whole-brain connectivity patterns generated with those generated from free-breathing data.Method
13 healthy subjects (age = 26.5 ± 6.5 years) were scanned using a Siemens TIM Trio 3T MRI scanner with a 32-channel head coil. For each subject two resting-state BOLD scans were collected using the simultaneous multi-slice GE-EPI BOLD technique. (TR/TE = 380/30 ms, flip angle = 40°, 20 5-mm slices, 64x64 matrix, 4x4x5 mm voxels, slice factor = 3, 950 volumes). During the clamped run, we clamped PETCO2 to the average PETCO2 level of each participant using the RespirActTM breathing circuit (Thornhill Research, Toronto, Canada). For the free-breathing run, the PETCO2 level is passively monitored but not controlled. rs-fMRI processing includes: motion correction, spatial smoothing, de-trending, and registering data into a 4mm3 MNI atlas. We limit the clamped and free-breathing comparison to voxels in the grey matter (GM). To compute whole-brain connectivity, each voxel is considered as a seed and correlated to every other voxel in the GM using Pearson correlation. The difference between clamped and free-breathing datasets is assessed using paired t-test. The number of voxels (N) showing significantly (p = 0.001) different connectivity is counted for both directions separately (clamped > free-breathing & free-breathing > clamped). Furthermore the number of voxels that has significantly different between-subject variance in connectivity values is counted with the same p value for both tails of the distribution. This procedure is repeated for all voxels in the gray matter. The output maps are thresholded with N = 12 for multiple comparisons correction.Result
Figure 1a shows the brain regions that have significantly stronger connectivity in free-breathing (top) and clamped (bottom) runs. Stronger connectivity during free breathing is found in most of the grey matter. Figure 1b represents the difference in the between-subject variability in connectivity values obtained using clamped and free-breathing data. PETCO2 clamping results in significantly higher between-subject variability compared to free-breathing data in almost all the gray matter.Discussion
Clamping PETCO2 leads to weaker connectivity and more between-subject variability in the connectivity values. This is consistent with previous findings suggesting stronger connectivity during free-breathing in the default mode network3. PETCO2 fluctuations have a global and widespread effect in the gray matter4, causing stronger correlations and hence more reproducible connectivity maps across subjects. Controlling PETCO2 fluctuations by clamping it would reveal the variability of brain connectivity across subjects. Our results question the practice of optimizing rs-fMRI connectivity pipelines through increasing reproducibility.Acknowledgements
No acknowledgement found.References
1. Chang C & Glover GH. Relationship between respiration, end-tidal CO2, and BOLD signals in resting-state fMRI. Neuroimage 47, 1381-1393 (2009).
2. Golestani AM, Kwinta JB, Khatamian YB, Chen JJ. The effect of low-frequency physiological correction on the reproducibility and specificity of resting-state fMRI metrics: functional connectivity, ALFF, and ReHo. Front Neurosci. 11: 546 (2017).
3. Madjar C, Gauthier CJ, Bellec P, Birn RM, Brooks JC, Hoge RD. Task-related BOLD responses and resting-state functional connectivity during physiological clamping of end-tidal CO2. Neuroimage 61(1): 41-49 (2012).
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 104: 266-277 (2015).
Figures
