Associations Between Age, Heart Rate Variability, and BOLD fMRI Signal Variability

Area of Science:

• Neuroimaging and Cardiovascular Physiology
• Functional Magnetic Resonance Imaging (fMRI)
• Heart Rate Variability (HRV)

Background:

• Age-related decreases in BOLD fMRI signal variance (SD_BOLD) are widely reported.
• Cardiovascular factors, particularly heart rate variability (HRV), may confound these age-SD_BOLD associations.
• Previous research has not sufficiently explored the direct relationship between HRV and SD_BOLD, especially within individuals.

Purpose of the Study:

• To investigate the association between heart rate variability (HRV) and BOLD fMRI signal variance (SD_BOLD).
• To determine if HRV partially explains the observed age-related decreases in SD_BOLD.
• To examine within-person coupling between HRV and SD_BOLD using sliding window analyses.

Main Methods:

• Analysis of resting-state fMRI data from two independent Midlife in the United States (MIDUS) samples (Core and Refresher).
• Partial Least Squares (PLS) analyses to assess HRV-SD_BOLD and age-SD_BOLD associations.
• Concordance and sliding-window analyses to examine within-person relationships between HRV metrics (SDNN, RMSSD, LF, HF) and SD_BOLD.

Main Results:

• Significant positive associations were found between HRV and SD_BOLD across both samples (permutation p<0.018).
• Whole-brain age-SD_BOLD associations were non-significant, contrasting with widespread age-related decreases in SD_BOLD (~70% of voxels).
• Sliding-window analyses revealed robust positive within-person associations between HRV (SDNN, RMSSD, LF) and SD_BOLD, independent of baseline levels.

Conclusions:

• Heart rate variability (HRV) is positively associated with BOLD fMRI signal variance (SD_BOLD).
• Cardiovascular factors, specifically HRV, partially mediate the relationship between age and SD_BOLD.
• Controlling for HRV, particularly low-frequency components or SDNN, is recommended for isolating neural effects in SD_BOLD analyses.