[Dual sequence method for analysis of spontaneous baroreceptor reflex sensitivity in patients with dilated cardiomyopathy]

Using machine learning to estimate the calendar age based on autonomic cardiovascular function

Introduction

Aging is accompanied by physiological changes in cardiovascular regulation that can be evaluated using a variety of metrics. In this study, we employ machine learning on autonomic cardiovascular indices in order to estimate participants' age.

Methods

We analyzed a database including resting state electrocardiogram and continuous blood pressure recordings of healthy volunteers. A total of 884 data sets met the inclusion criteria. Data of 72 other participants with an BMI indicating obesity (>30 kg/m²) were withheld as an evaluation sample. For all participants, 29 different cardiovascular indices were calculated including heart rate variability, blood pressure variability, baroreflex function, pulse wave dynamics, and QT interval characteristics. Based on cardiovascular indices, sex and device, four different approaches were applied in order to estimate the calendar age of healthy subjects, i.e., relevance vector regression (RVR), Gaussian process regression (GPR), support vector regression (SVR), and linear regression (LR). To estimate age in the obese group, we drew normal-weight controls from the large sample to build a training set and a validation set that had an age distribution similar to the obesity test sample.

Results

In a five-fold cross validation scheme, we found the GPR model to be suited best to estimate calendar age, with a correlation of r=0.81 and a mean absolute error of MAE=5.6 years. In men, the error (MAE=5.4 years) seemed to be lower than that in women (MAE=6.0 years). In comparison to normal-weight subjects, GPR and SVR significantly overestimated the age of obese participants compared with controls. The highest age gap indicated advanced cardiovascular aging by 5.7 years in obese participants.

Discussion

In conclusion, machine learning can be used to estimate age on cardiovascular function in a healthy population when considering previous models of biological aging. The estimated age might serve as a comprehensive and readily interpretable marker of cardiovascular function. Whether it is a useful risk predictor should be investigated in future studies.

Risk Stratification in Idiopathic Dilated Cardiomyopathy Patients Using Cardiovascular Coupling Analysis

Cardiovascular diseases are one of the most common causes of death; however, the early detection of patients at high risk of sudden cardiac death (SCD) remains an issue. The aim of this study was to analyze the cardio-vascular couplings based on heart rate variability (HRV) and blood pressure variability (BPV) analyses in order to introduce new indices for noninvasive risk stratification in idiopathic dilated cardiomyopathy patients (IDC). High-resolution electrocardiogram (ECG) and continuous noninvasive blood pressure (BP) signals were recorded in 91 IDC patients and 49 healthy subjects (CON). The patients were stratified by their SCD risk as high risk (IDC_HR) when after two years the subject either died or suffered life-threatening complications, and as low risk (IDC_LR) when the subject remained stable during this period. Values were extracted from ECG and BP signals, the beat-to-beat interval, and systolic and diastolic blood pressure, and analyzed using the segmented Poincaré plot analysis (SPPA), the high-resolution joint symbolic dynamics (HRJSD) and the normalized short time partial directed coherence methods. Support vector machine (SVM) models were built to classify these patients according to SCD risk. IDC_HR patients presented lowered HRV and increased BPV compared to both IDC_LR patients and the control subjects, suggesting a decrease in their vagal activity and a compensation of sympathetic activity. Both, the cardio -systolic and -diastolic coupling strength was stronger in high-risk patients when comparing with low-risk patients. The cardio-systolic coupling analysis revealed that the systolic influence on heart rate gets weaker as the risk increases. The SVM IDC_LR vs. IDC_HR model achieved 98.9% accuracy with an area under the curve (AUC) of 0.96. The IDC and the CON groups obtained 93.6% and 0.94 accuracy and AUC, respectively. To simulate a circumstance in which the original status of the subject is unknown, a cascade model was built fusing the aforementioned models, and achieved 94.4% accuracy. In conclusion, this study introduced a novel method for SCD risk stratification for IDC patients based on new indices from coupling analysis and non-linear HRV and BPV. We have uncovered some of the complex interactions within the autonomic regulation in this type of patient.

Baroreflex Coupling Assessed by Cross-Compression Entropy

Estimating interactions between physiological systems is an important challenge in modern biomedical research. Here, we explore a new concept for quantifying information common in two time series by cross-compressibility. Cross-compression entropy (CCE) exploits the ZIP data compression algorithm extended to bivariate data analysis. First, time series are transformed into symbol vectors. Symbols of the target time series are coded by the symbols of the source series. Uncoupled and linearly coupled surrogates were derived from cardiovascular recordings of 36 healthy controls obtained during rest to demonstrate suitability of this method for assessing physiological coupling. CCE at rest was compared to that of isometric handgrip exercise. Finally, spontaneous baroreflex interaction assessed by CCE_BRS was compared between 21 patients suffering from acute schizophrenia and 21 matched controls. The CCE_BRS of original time series was significantly higher than in uncoupled surrogates in 89% of the subjects and higher than in linearly coupled surrogates in 47% of the subjects. Handgrip exercise led to sympathetic activation and vagal inhibition accompanied by reduced baroreflex sensitivity. CCE_BRS decreased from 0.553 ± 0.030 at rest to 0.514 ± 0.035 during exercise (p < 0.001). In acute schizophrenia, heart rate, and blood pressure were elevated. Heart rate variability indicated a change of sympathovagal balance. The CCE_BRS of patients with schizophrenia was reduced compared to healthy controls (0.546 ± 0.042 vs. 0.507 ± 0.046, p < 0.01) and revealed a decrease of blood pressure influence on heart rate in patients with schizophrenia. Our results indicate that CCE is suitable for the investigation of linear and non-linear coupling in cardiovascular time series. CCE can quantify causal interactions in short, noisy and non-stationary physiological time series.

Central- and autonomic nervous system coupling in schizophrenia

The autonomic nervous system (ANS) dysfunction has been well described in schizophrenia (SZ), a severe mental disorder. Nevertheless, the coupling between the ANS and central brain activity has been not addressed until now in SZ. The interactions between the central nervous system (CNS) and ANS need to be considered as a feedback-feed-forward system that supports flexible and adaptive responses to specific demands. For the first time, to the best of our knowledge, this study investigates central-autonomic couplings (CAC) studying heart rate, blood pressure and electroencephalogram in paranoid schizophrenic patients, comparing them with age-gender-matched healthy subjects (CO). The emphasis is to determine how these couplings are composed by the different regulatory aspects of the CNS-ANS. We found that CAC were bidirectional, and that the causal influence of central activity towards systolic blood pressure was more strongly pronounced than such causal influence towards heart rate in paranoid schizophrenic patients when compared with CO. In paranoid schizophrenic patients, the central activity was a much stronger variable, being more random and having fewer rhythmic oscillatory components. This study provides a more in-depth understanding of the interplay of neuronal and autonomic regulatory processes in SZ and most likely greater insights into the complex relationship between psychotic stages and autonomic activity.

Altered cardiovascular coupling in patients with sudden sensorineural hearing loss in comparison to healthy subjects

The causes of idiopathic sudden sensorineural hearing loss (SSNHL) are still unclear while SSNHL seems to be a multicausal disease. To date limited information about autonomic regulation and, especially, cardiovascular coupling (CVC) are available for those patients. The objective of this study was to characterize short-term (30 min) CVC in 23 SSNHL patients in comparison to 23 healthy age and gender matched normal hearing control subjects (CON). Further on, the results from CVC should be compared with those from standard heart rate variability (HRV) and blood pressure variability (BPV) analyses. The results showed that HRV is not affected by the disease whereas BPV analysis revealed significant differences between both groups (p<0.01) whereby SSNHL exhibit a decreased short-term BPV. Results from CVC analysis demonstrated that especially the applied nonlinear methods exhibit an increased short-term CVC in SSNHL patients (p<0.01) indicating more complex interactions of short-term HR and BP regulatory processes. In conclusion, this study was the first to show a changed and decreased short-term BPV and increased nonlinear CVC in SSNHL patients. Our findings might help to improve diagnostic strategies for hearing loss caused by vascular factors.

Cardiovascular coupling analysis with high-resolution joint symbolic dynamics in patients suffering from acute schizophrenia

Besides the well-known cardiac risk factors for schizophrenia, increasing concerns have been raised regarding the cardiac side-effects of antipsychotic medications. A bivariate analysis of autonomic regulation, based on cardiovascular coupling, can provide additional information about heart rate (HR) and blood pressure regulatory patterns within the complex interactions of the cardiovascular system. We introduce a new high-resolution coupling analysis method (HRJSD) based on joint symbolic dynamics (JSD), which is characterized by three symbols, a threshold (individual dynamic variability, physiological) for time series transformation and eight coupling pattern families. This is based on a redundancy reduction strategy used to quantify and characterize cardiovascular couplings. In this study, short-term (30 min) HR and systolic blood pressure (SP) time series of 42 unmedicated (UNMED) and 42 medicated patients (MED) suffering from acute schizophrenia were analysed to establish the suitability of the new method for quantifying the effects of antipsychotics on cardiovascular couplings. We were able to demonstrate that HRJSD, applying the threshold based on spontaneous baroreflex sensitivity (BRS) estimation, revealed eight significant pattern families that were able to quantify the anti-cholinergic effects of antipsychotics and the related changes of cardiovascular regulation (coupling) in MED in comparison to UNMED. This was in contrast to the simple JSD, BRS (sequence method) and only partly to standard linear HR variability indices. HRJSD provides strong evidence that autonomic regulation in MED seems to be, to some extent, predominated by invariable HR responses in combination with alternating SP values in contrast to UNMED, indicating an impairment of the baroreflex control feedback loop in MED. Surrogate data analysis was applied to test for the significance and nonlinearity of cardiovascular couplings in the original data due to medical treatment with antipsychotic drugs in MED. In conclusion, the application of HRJSD revealed detailed information about short-term nonlinear cardiovascular couplings and cardiovascular physiological regulatory mechanisms (patterns) of autonomic function due to the anti-cholinergic effects of antipsychotics in patients with acute schizophrenia.

The altered complexity of cardiovascular regulation in depressed patients

Major depressive disorders (MDD) are associated with an increased risk for cardiovascular morbidity and mortality. Even if it is known that MDD are accompanied by an autonomic dysbalance with increased sympathetic and/or reduced parasympathetic activity, to date only limited information is available about the degree and complexity of cardiovascular regulation. The aim of this study was to investigate the influence of MDD on the autonomous nervous system and cardiovascular complexity by means of linear and nonlinear indices from heart rate and blood pressure variability (HRV, BPV). From 57 non-medicated patients and 57 matched healthy controls with respect to age and gender HRV and BPV in time and frequency domain, symbolic dynamics, compression entropy, multiscale entropy, detrended fluctuation analysis, Poincaré plot analysis and baroreflex sensitivity were analysed from 30 min short-term recordings. Complexity indices from nonlinear dynamics demonstrated considerable changes in autonomous regulation due to MDD. For the first time we could show that non-medicated depressed patients who were matched with respect to age and gender reveal a significantly changed short-term as well as long-term complexity of cardiovascular regulation. These results suggest substantial changes in autonomic control probably due to a change of interactions between different physiological control loops in MDD.

Linear and nonlinear analysis of autonomic regulation in depressed patients

To investigate the effect of depression and medication on the autonomic nervous system (ANS) we compared 25 patients suffering from depression with 72 matched control subjects (CON). From every patient three times ECG and non-invasive blood pressure were recorded in: acute non-mediated state (T1), medicated state (T2) and clinical recovered state (T3). Cardiac autonomic regulation was evaluated by linear and nonlinear analysis methods. The results show that non-medicated depressed patients differ significantly from CON (p0.05). In the medicated state considerably high significant differences between depressed patient and CON were found (p0.0001). After an 18 months follow-up a normalization of autonomic parameters was achieved in depressed patients. These results show that depression influences autonomic parameters significantly and that medical treated clinical recovered patients approximate to healthy controls in autonomic regulation.

Bimodal dose-dependent effect on autonomic, cardiac control after oral administration of Atropa belladonna

This single-blind placebo-controlled study was designed to investigate the dose-dependent vagolytic and vagotonic effects after a single oral administration of Atropa belladonna tincture (ABT, 0.1 mg/ml alkaloid concentration, atropine/scopolamine = 20:1). In eight healthy young subjects, heart rate and noninvasive arterial finger blood pressure were recorded simultaneously over 4 h after oral application of four different doses of ABT (day 1: 2 ml, day 2: placebo, day 3: 5 ml, day 4: 1 ml). On each day, 14 20-min sequences under controlled experimental conditions were performed. Among others, mean RR interval (RR), high-frequency spectral power of heart rate variability (HF), and noninvasive baroreflex sensitivity (BRS) were calculated during metronome breathing in supine position. These parameters were robust markers of vagal activity. One hour after 5ml ABT, RR, HF and BRS decreased clearly in six of eight subjects. This effect was interpreted as vagolytic response. After 1 and 2 ml ABT, and after placebo, RR and HF increased markedly. The increase after ABT was much higher than the increase solely due to adaptation after placebo administration, and it could be clearly identified as an augmentation of vagal cardiac activity caused by low-dose ABT. In conclusion, low doses of orally administered ABT can be effectively used to stimulate parasympathetic activity in man. The mode of vagal activation changes between 2 and 5 ml ABT from vagotonic to vagolytic. ABT has no or very little effect on blood pressure control.