Resting heart rate (variability) predicts cognitive function and suggests different heart‐brain connections in cognitively healthy individuals with abnormal CSF amyloid/tau

Abstract Background Both resting heart rate variability (HRV) and resting heart rate (HR) are regulated by the central autonomic network linking prefrontal cortex with cardiac regulation. To better understand the interactions between the heart and cognition we determined the relationships between resting HR/HRV and cognitive function. Method We compared and correlated resting HR with Mini‐Mental State Exam (MMSE) in cognitively healthy (CH, n=137) individuals when their CSF amyloid/tau ratios were normal (CH‐NATs, n=51) or pathological (CH‐PATs, n=68), in participants with mild cognitive impairment (MCI, n=54), and Alzheimer’s disease (AD, n=42). We also compared and correlated HRV with behavioral performance (accuracy and response time) or alpha event‐related desynchronization (ERD) during Stroop (interference) or task switching testing between subsets of CH‐NATs and CH‐PATs. Cognitive tasks include: 1) Stroop during low load (congruent trials) and high load (incongruent trials); 2) task switching during low load, repeat trials (color‐color or word‐word) and high load, switch trials (color‐word or word‐color). HRV was assessed from 5‐minute resting ECG recording, including: HR and root mean squared of successive differences (RMSSD) that reflect vagal activities. Result No differences of resting HR or HRV were observed between CH‐NATs, CH‐PATs, MCI, and AD. Interestingly, resting HR negatively correlated with MMSE only in CH‐PATs (r=‐0.57, p=0.002), but not in CH‐NATs (r=0.07, p=0.71), CH (r=‐0.09, p=0.57), MCI (r=‐0.14, p=0.41), or AD (r=‐0.11, p=0.59) ( Fig 1 ). For Stroop interference testing, resting HR (preceding task) positively correlated with frontal alpha ERD during high load trails (r=0.78, p=0.017) in CH‐PATs, but not in CH‐NATs ( Fig 2 ). For task switch testing, resting RMSSD positively correlated with accuracy during switch trials (r=0.64, p=0.004) in CH‐NATs, but not in CH‐PATs ( Fig 3 ). Conclusion These results suggest: 1) resting HR is linked with cognition in CH‐PATs; 2) resting HR is related to inhibitory control in CH‐PATs: higher HR linked with less negative alpha ERD (less activation); 3) resting HRV predicts task switching performance in CH‐NATs: higher RMSSD linked with higher accuracy. Overall, resting HR and HRV potentially reflect cognition and executive functions that differentiate CH‐NATs and CH‐PATs, suggesting different cerebral‐cardiovascular integration in these two populations.