Hemodynamic Response Biomarkers for Aging Brain Health: Bogalusa Heart Study

Abstract

AbstractBackgroundIn functional magnetic resonance imaging (fMRI), a hemodynamic response function describes how neuron firing events are translated into a blood oxygenation response that is detected through imaging. The HRF varies between individuals and between brain regions, and its characteristics are believed to be valuable biomarkers of neurovascular aspects of brain health. However, the impact of cardiovascular risks on HRF characteristics among diverse cognitively healthy middle aged and older adults is not well studied.Method137 Bogalusa Heart Study participants performed a Stroop task during fMRI (Table.1). The HRFs of each participant within eight regions of interest (ROIs) previously identified as activated by this task [1] were calculated using a deep learning method. Each HRF was characterized by four main parameters (Figure 1): time to peak (TTP), peak magnitude (PM), trough magnitude (TM), and area under the HRF curve (AUHRF). Linear regression models related HRF parameters to demographic, cardiometabolic, and cognitive measures.ResultTable 2 shows the associations between demographic, cardiometabolic, and cognitive measures and HRF parameters. Younger age was associated with lower AUHRF in the inferior frontal gyrus (p‐value = 0.052). Whites (compared to African Americans) had lower PM in the angular gyrus and lower AUHRF in the anterior insula (maximum p‐value = 0.022). Better systolic blood pressure and diastolic blood pressure were associated with shorter TTP in the angular gyrus and cerebellum (maximum p‐value = 0.029). APOE 3‐3 genotype and better BMI were associated with shorter TTP in the middle frontal gyrus and fusiform gyrus (maximum p‐value = 0.025). Better white matter hyperintensity volume was associated with lower PM and TM in the anterior insula and cerebellum (maximum p‐value = 0.032). Better digit span backwards score, digit coding score, vocabulary score, and word reading score were associated with shorter TTP and lower AUHRF in the angular gyrus, middle frontal gyrus, and/or inferior frontal gyrus (maximum p‐value = 0.038).ConclusionIn a diverse, cognitively healthy, middle‐aged community sample, HRF parameters tracked closely with recognized indicators of brain health. Overall, lower TTP, PM, TM, and AUHRF represent better brain health, as there is no delay and alteration in neural‐vascular coupling unit.