Using breath-hold hypercapnia to scale the fMRI BOLD response to visual stimulation in older adults with cardiovascular disease
Schwarz, Nicolette Françoise
MetadataShow full item record
Cardiovascular Disease (CVD) is a heart and blood vessel disease, and has been associated with deterioration in brain function and structure. These alterations, in turn, are implicated in the cognitive decline that can be observed in individuals with CVD. Functional magnetic resonance imaging (fMRI) is a powerful tool used to investigate cognition in older adult populations. However, the blood-oxygenation-level-dependent signal (BOLD) measured with fMRI is sensitive to variations in cerebral blood flow and volume. Impaired cardiovascular health may affect the validity of fMRI and complicate the interpretation of fMRI results. Hypercapnia challenges such as breath-holding can be used to evaluate cerebrovascular reactivity and the resulting estimates can be used as a scaling factor. The current investigation examined the effects of breath-hold hypercapnia scaling of brain activation in response to visual stimulation in a group of fourteen older adults with CVD (mean age=70±9, 3 women) and 14 age-matched healthy older adults (67±7, 8 women). In functional regions of interest, hypotheses were tested that scaling reduces activation magnitude, spatial extent, and variability. Additional exploratory analyses were carried out on the whole brain to investigate global effects. Hypothesis-testing yielded mixed results with brain activation estimates to visual stimulation being significantly reduced after scaling, while potential changes in activation extent could not be investigated due to spuriously high percent signal changes after scaling. These spurious values also resulted in increased variability. At the whole-brain level, evidence for scaling-induced reductions in visual stimulation amplitude, extent, and variability were observed, and a cluster of significantly different brain activation between groups revealed only after scaling. The global application of this correction method may be useful in detecting regions sensitive to vascular artifacts, including those in task-positive and task-negative networks as well as sinuses. The present findings highlight that conclusions derived from BOLD fMRI studies in older adult populations need to be cautiously interpreted and provide evidence for the utility and feasibility of incorporating a simple hypercapnia-based correction method in routine fMRI pipelines.