Predictive Capacity of Beat-to-Beat Blood Pressure Variability for Cardioautonomic and Vascular Dysfunction in Early Metabolic Challenge

Abstract

Diabetic patients present established cardiovascular disease at the onset of diagnostic metabolic symptoms. While premature autonomic and vascular deterioration considered risk factors for major cardiovascular complications of diabetes, present in initial stages of metabolic impairment, their early detection remains a significant challenge impeding timely intervention. In the present study, we examine the utility of beat-to-beat blood pressure variability (BPV) parameters in capturing subtle changes in cardiac autonomic and vascular control distinguishing between various risk categories, independent of the average BP. A rat model of mild hypercaloric (HC) intake was used to represent the insidious cardiovascular changes associated with early metabolic impairment. Invasive hemodynamics were used to collect beat-to-beat BP time series in rats of either sex with different durations of exposure to the HC diet. Linear (standard deviation and coefficient of variation) and nonlinear (approximate entropy, ApEn, and self-correlation of detrended fluctuation analysis, α) BPV parameters were calculated to assess the impact of early metabolic impairment across sexes and feeding durations. HC-fed male, but not female, rats developed increased fat:lean ratio as well as hyperinsulinemia. Unlike linear parameters, multivariate analysis showed that HC-fed rats possessed lower ApEn and higher α, consistent with early changes in heart rate variability and blunting of parasympathetic baroreceptor sensitivity, particularly in males. Moreover, logistic regression demonstrated the superiority of nonlinear parameters of diastolic BPV in predicting a prediabetic disease state. Our findings support the use of nonlinear beat-to-beat BPV for early detection of cardiovascular derangements in the initial stages of metabolic impairment.

Keywords: approximate entropy; autonomic dysfunction; blood pressure variability; metabolic challenge; self-correlation.

FIGURE 1

Metabolic parameters of different rat groups. Cumulative calorie intake (A), body weight (B), blood glucose level (C), fat:lean ratio (D), and serum insulin concentration (E) in control and hypercaloric diet (HC)-fed rats of either sex in different feeding duration groups. Statistical analysis was performed by three-way ANOVA followed by Sidak’s multiple comparison test. * denotes p < 0.05 vs. control group of the same sex and feeding duration group (control vs. HC), while # denotes p < 0.05 vs. the corresponding male group (males vs. females).

FIGURE 2

Gross myocardial, hemodynamic, and cardiac autonomic functions in response to HC-feeding as a function of feeding duration and sex. (A) Representative electrocardiograms of M-mode images (scale bar corresponds to 1 mm), (B) ejection fraction (EF) assessed by echocardiography, (C) average mean arterial pressure (MAP) assessed by invasive carotid artery catheterization, and slope of the best-fit linear regression of reflex bradycardic responses to a rise in mean arterial pressure (MAP) following increasing doses of phenylephrine (PE) or sodium nitroprusside (SNP) reflecting (D) parasympathetic nervous system (PSNS) or (E) sympathetic nervous system (SNS) activity, respectively, of control and hypercaloric diet (HC)-fed rats of either sex in different feeding duration groups. Statistical analysis was performed by three-way ANOVA followed by Sidak’s multiple comparison test. * denotes a p < 0.05 vs. control of the same sex and feeding duration group (control vs. HC), while # denotes a p < 0.05 vs. the corresponding male group (males vs. females).

FIGURE 3

Impact of HC feeding on average HR and linear and nonlinear parameters of beat-to-beat HRV in rats of either sex in different feeding duration groups. (A) Average, (B) standard deviation (SD), (C) coefficient of variation (CV), (D) approximate entropy (ApEn), and (E) α of detrended fluctuation analysis of HR in control and HC-fed rats of either sex in different feeding duration groups. Statistical analysis was performed by three-way ANOVA followed by Sidak’s multiple comparison test. * denotes a p < 0.05 vs. control of the same feeding duration and sex (control vs. HC), while # denotes a p < 0.05 vs. the corresponding male group (males vs. females).

FIGURE 4

Representative beat-to-beat blood pressure and heart rate tracings of selected rats. Raw arterial pressure, heart rate, and systolic and diastolic blood pressure recordings in 12-week and 24-week control and HC-fed (A) male and (B) female rats. (C) Downsampled beat-to-beat diastolic blood pressure (BP) tracing of control and HC-fed male rats of different feeding duration groups showing equal average, SD, and CV values for all groups with reduced ApEn values in 12-week HC-fed rats. (D) Beat-to-beat systolic blood pressure (BP) tracing of control and HC-fed female rats of different feeding duration groups showing equal average, SD, and CV values for all groups with reduced ApEn and α values compared to male rats.

FIGURE 5

Impact of HC-feeding on average and linear parameters of beat-to-beat systolic and diastolic blood pressure (SBP and DBP, respectively) in rats of either sex in different feeding duration groups. (A,B) Average, (C,D) standard deviation (SD), and (E,F) coefficient of variation (CV) of SBP and DBP, respectively, in control and HC-fed rats of either sex in different feeding duration groups. Statistical analysis was performed by three-way ANOVA followed by Sidak’s multiple comparisons test. * denotes a p < 0.05 vs. control of the same feeding duration and sex (control vs. HC), while # denotes a p < 0.05 vs. the corresponding male group (males vs. females).

FIGURE 6

Impact of HC-feeding on nonlinear parameters of beat-to-beat systolic and diastolic blood pressure (SBP and DBP, respectively) in rats of either sex in different feeding duration groups. (A,B) Approximate entropy (ApEn) and (C,D) α of detrended fluctuation analysis of SBP and DBP, respectively, in control and HC-fed rats of either sex in different feeding duration groups. Statistical analysis was performed by three-way ANOVA followed by Sidak’s multiple comparisons test. * denotes a p < 0.05 vs. control of the same feeding duration and sex (control vs. HC), while # denotes a p < 0.05 vs. the corresponding male group (males vs. females).