Correspondence between BOLD fMRI task response and cerebrovascular reactivity across the cerebral cortex

Abstract

BOLD sensitivity to baseline perfusion and blood volume is a well-acknowledged fMRI confound. Vascular correction techniques based on cerebrovascular reactivity (CVR) might reduce variance due to baseline cerebral blood volume, however this is predicated on an invariant linear relationship between CVR and BOLD signal magnitude. Cognitive paradigms have relatively low signal, high variance and involve spatially heterogenous cortical regions; it is therefore unclear whether the BOLD response magnitude to complex paradigms can be predicted by CVR. The feasibility of predicting BOLD signal magnitude from CVR was explored in the present work across two experiments using different CVR approaches. The first utilized a large database containing breath-hold BOLD responses and 3 different cognitive tasks. The second experiment, in an independent sample, calculated CVR using the delivery of a fixed concentration of carbon dioxide and a different cognitive task. An atlas-based regression approach was implemented for both experiments to evaluate the shared variance between task-invoked BOLD responses and CVR across the cerebral cortex. Both experiments found significant relationships between CVR and task-based BOLD magnitude, with activation in the right cuneus (R ² = 0.64) and paracentral gyrus (R ² = 0.71), and the left pars opercularis (R ² = 0.67), superior frontal gyrus (R ² = 0.62) and inferior parietal cortex (R ² = 0.63) strongly predicted by CVR. The parietal regions bilaterally were highly consistent, with linear regressions significant in these regions for all four tasks. Group analyses showed that CVR correction increased BOLD sensitivity. Overall, this work suggests that BOLD signal response magnitudes to cognitive tasks are predicted by CVR across different regions of the cerebral cortex, providing support for the use of correction based on baseline vascular physiology.

Keywords: BOLD; attention; cerebral blood blow; cerebrovascular reactivity; cognition; functional magnet resonance imaging (fMRI); hypercapnia; vascular physiology.

FIGURE 1

Schematic diagram summarizing the pre-processing and first-level analysis pipeline for experiment 1 (left) and experiment 2 (right).

FIGURE 2

ROIs included in the regression analyses for the SCAP (A), stop-signal (B) and task-switch (C) tasks are shown in cyan-magenta. ROIs where the task did not result in significant activation in at least 10 participants are not coloured. Images in coronal, sagittal and axial planes are shown in neurological orientation with MNI coordinates. The color bar indicates p-values from the regressions; regions in magenta did not show a significant linear relationship between breath-hold (BH) BOLD PSC and task PSC. The scatter plots in (A) demonstrate the PSC (breath-hold and SCAP) from the left and right insula to the SCAP task. The regression for the left insula (shown in magenta in the coronal plane) was non-significant, while the right insula (shown in cyan in the coronal plane) was significant. The scatter plots in (B) show an example of a significant (left lateral occipital cortex) and non-significant (right lateral occipital cortex) ROI for the stop-signal task. The scatter plots in (C) also show the left and right lateral occipital cortex, but to the task-switch paradigm. For this task, the regression for the left lateral occipital cortex was non-significant, while the right was significant. For all scatter plots, each data point represents one subject’s averaged PSC from that ROI.

FIGURE 3

Group activation maps for the SCAP (A, B), stop-signal (C, D) and task-switch (E, F) paradigms for experiment 1. The upper rows of each task (A, C, E) show the activation maps which were corrected for CVR by including the breath-hold contrast images as covariates. The lower row for each task (B, D, F) are the standard, uncorrected for CVR, group activation maps. All images cluster-corrected for multiple comparisons (p < 0.05). The colour bar indicates t-values.

FIGURE 4

An example CVR map of a single subject is shown in (A). Whole-brain average group maps for the alerting contrast of the ANT-R shown in (B). The ROI analyses of the alerting contrast shown in (C) and (D) included explicit masks comprised of the five significant ROIs from the linear regression analyses for experiment 2. For the vascular corrected ROI analysis in (C), the mean CVR values for each ROI were entered into the model as covariates. The vascular uncorrected ROI analysis shown in (D) included the explicit masks only with no covariates. The group maps in (B, C, D) were thresholded at p < 0.001 uncorrected for multiple comparisons. The upper colour bar indicates %BOLD/mmHg for the CVR map in (A). The lower colour bar indicates t-values for (B, C, D).

FIGURE 5

Right (upper) and superior (lower) views of the significant ROIs (in cyan) from the alerting contrast of the ANT-R task. Reduced opacity images on the right highlight medial aspects. Scatter plots show the linear relationship between alerting PSC (y-axis) and CVR (x-axis) for each ROI. The (*) next to right cuneus and right paracentral gyrus indicate that these two ROIs were significant following FDR-correction for multiple comparisons. A = anterior, P = posterior.

FIGURE 6

Polar plot displaying the R ² values for all ROIs from all linear regression analyses, across both experiments. Each ROI shows its corresponding R ² value for the left and right hemisphere separately. The R ² values for the SCAP are shown in blue, stop-signal in red, task-switch in black and ANT-R in green. The open-coloured markers (‘o’) indicate that the ROI was not significant after FDR correction for multiple corrections, while the closed markers were significant. The concentric circles indicate the R ² value, with values increasing with radius to a maximum of R ² = 0.80.