Association between autonomic dysfunction, 12-year mortality and 10-year incident cardiovascular disease

Abstract Background Standing from a supine position requires a co-ordinated response in peripheral and central haemodynamic signals to protect the brain and body from sudden drops in blood pressure (BP) and has previously been shown to be an important measure of neuro-cardiovascular health (1-3). Previous studies have found an association between BP and heart rate recovery (HRR) to standing with all-cause mortality, cardiovascular disease and other health outcomes (4-8). However, almost all of these studies have exclusively investigated the association of consensus-defined HR/BP phenotypes measured at specific timepoints during standing. Purpose We hypothesised that assessing information from the entire haemodynamic response to orthostasis - instead of discrete timepoints or consensus definitions - may uncover novel associations between BP/HR responses to active stand and neuro-cardiovascular health in older adults. This may afford better physiological understanding of the longitudinal relationships between the response to standing and both Cardiovascular Disease (CVD) and mortality. Methods We used functional Principal Components Analysis (FPCA) to examine the determinants of variability in the complete haemodynamic response to active stand and examine which components were related to mortality and incident CVD in >4500 community-dwelling adults aged 50+. Results Figure 1 shows the mean response curve according to mortality status and Figure 2 the mean curve for participants free of CVD at baseline who later have CVD vs those who remain CVD-free during 10 year follow-up. All-cause mortality was associated with functional principal components which discriminate higher baseline HR, blunted HR peak and impaired HRR up to ~25 seconds after standing. Components that discriminated high baseline and impaired recovery of SBP from the post-stand nadir to 60 seconds post stand were also associated with all-cause mortality. This association remained even after covariate adjustment. Interestingly, neither baseline SBP nor orthostatic hypotension (OH) were associated with mortality. Components which discriminate high baseline and blunted HR peak; elevated stroke volume index and impaired recovery of DBP from ~25 seconds post stand were associated with 10-year incident CVD. Conclusions To our knowledge this is the first study to incorporate data driven information over the entire trace of the haemodynamic response to standing to investigate associations with incident CVD and mortality. The fact that components which discriminated high BP and impaired recovery of BP were associated with 12-year mortality and 10-year incident CVD when traditional summary measures of OH and baseline BP were not; suggests that incorporating information across the entire response to standing allows for richer and more flexible associations with health outcomes to be uncovered and allows for a better understanding of the underlying physiology leading to such associations.Figure 1:Mean trace of Mortality StatusFigure 2:Mean Trace of CVD Status