Elsevier

Clinical Neurophysiology

Volume 118, Issue 9, September 2007, Pages 2009-2015
Clinical Neurophysiology

Non-linear complexity measures of heart rate variability in acute schizophrenia

https://doi.org/10.1016/j.clinph.2007.06.012Get rights and content

Abstract

Objective

Cardiovascular mortality is significantly increased in patients suffering from schizophrenia. The mechanisms currently discussed contain unhealthy lifestyle with obesity and smoking, increased incidence of diabetes, adverse pro-arrhythmic effects of antipsychotic medication and altered autonomic function. It is therefore likely that the adaptation of the heart rate to different requirements is faulty in schizophrenia. One way to detect adaptive capabilities and thus stability of regulation is to measure complexity of heart rate fluctuations, with more complex heart rate fluctuations indicating better adaptability of the underlying system.

Methods

We calculated novel non-linear measures for beat-to-beat interval complexity from short-term ECG recordings in 20 unmedicated patients suffering from acute schizophrenia and compared them to those obtained from matched controls.

Results

Data from all mathematical models applied, i.e. joint symbolic dynamics, compression entropy, fractal dimension and approximate entropy, revealed significantly reduced complexity of heart rate time series in acute schizophrenia. When using heart rate as a covariate, only fractal dimension remained significantly altered, thus appearing to be a relatively more important heart rate independent parameter.

Conclusions

Complexity of heart rate modulation is significantly reduced in acute, untreated schizophrenia, thus indicating an increased risk for cardiovascular events in these patients.

Significance

These data might eventually add to the currently discussed monitoring of physical health in patients with schizophrenia, possibly providing a promising tool for cardio-arrhythmic risk stratification.

Introduction

Mortality in patients with schizophrenia is at least two times higher than in reference populations (Black and Fisher, 1992, Brown et al., 2000). Apart from suicide (Meltzer, 1998) cardiovascular causes account for the majority of these events (Brown et al., 2000, Enger et al., 2004). The latter have been associated with unhealthy lifestyle, adverse effects caused by antipsychotics and higher incidence of diabetes as well as metabolic syndrome (Ruschena et al., 1998, Glassman and Bigger, 2001, Grundy et al., 2004), which resulted in the recommendation to routinely monitor physical health in these patients (Marder et al., 2004). The impact of antipsychotic medication has been studied in the greatest extent due to these substances’ ability to significantly prolong QTc intervals (Titier et al., 2005). However, even unmedicated patients with schizophrenia face an increased incidence of sudden cardiac death (Ray et al., 2001, Jindal et al., 2005). As we and others already assumed previously, one possible underlying cause might be autonomic dysfunction during acute psychosis as reflected by altered heart rate variability and autonomic function parameters. These include reduced time and frequency measures of heart rate variability (HRV) (Valkonen-Korhonen et al., 2003, Bar et al., 2005, Mujica-Parodi et al., 2005), reduced baroreflex sensitivity (Bar et al., 2007a) and reduced QT variability (Bar et al., 2007c) in short-term recordings. Besides linear HRV parameters describing the variance of beat-to-beat intervals, non-linear complexity parameters have been developed to describe the regularity of heart rate time series. The application of these novel analyses has led to a higher sensitivity for detecting autonomic dysfunction (Baumert et al., 2004b, Hoyer et al., 2006) and patients at risk for sudden death (Voss et al., 1996) in different diseases. To date, there is limited information about complexity measures of HRV in schizophrenia. Reduced complexity was found in 24 h recordings in unmedicated patients (Boettger et al., 2006) and in clozapine-treated patients when applying the sample entropy parameter, which negatively correlated with the degree of positive psychotic symptoms (Kim et al., 2004).

Even less data exist on complexity measures obtained from short-term recordings, owing to the fact that the calculation of such parameters requires a significant amount of raw data. Here, we aimed to fill this gap employing novel non-linear complexity measures that have been shown to reveal reliable results even in short recording times in patients suffering from different cardiovascular diseases. These are compression entropy Hc (Baumert et al., 2004b), probability of high or low variability sequences (phvar and plvar, respectively) (Voss et al., 1996), fractal dimension (Katz, 1988) and approximate entropy (Pincus, 1991). These were applied to electrocardiographic data obtained from 20 unmedicated patients with acute schizophrenia and 20 matched controls.

Section snippets

Participants

Twenty patients suffering from paranoid schizophrenia and 20 healthy controls matched with respect to age, sex, weight, smoking habits and education (see Table 1) were included in this study. Patients were examined unmedicated in the acute stage on the day of admission to the hospital and were only included in the study, when they had not taken antipsychotic medication for at least 8 weeks prior to the study. Control subjects were recruited from hospital staff and medical students. Neither

Multivariate statistic

Using multivariate ANOVA to compare schizophrenic patients with matched controls, we found a significant main effect for the between-subjects factor GROUP (patients with schizophrenia versus healthy controls) [F(4, 35) = 5.33, p = 0.002].

Univariate statistics

Follow-up univariate ANOVAs showed significant main effects for lnHc [F(1, 41) = 13.33, p < 0.0001] (Fig. 2a), lnphvar10 [F(1, 41) = 13.46, p < 0.001] (Fig. 2b), lnApEn [F(1, 41) = 10.13, p < 0.003] (Fig. 2c) and lnFD [F(1, 41) = 22.49, p < 0.001] (Fig. 2d), indicating that all

Discussion

In this study, we present further evidence for reduced complexity in beat-to-beat interval (BBI) time series obtained from short-term electrocardiographic recordings in patients with acute schizophrenia. Recently, we were able to show a similar reduction of complexity in patients with schizophrenia (Boettger et al., 2006). However, these measures needed to be calculated from 24 h Holter ECG recordings. Here, we assessed complexity from short BBI time series.

High complexity of biosignals reflects

Acknowledgement

We are grateful to Dr. R. Vollandt (IMSID, Institute for Medical Statistics, University of Jena, Germany) for his advice on statistics.

References (45)

  • K.J. Bar et al.

    Decreased baroreflex sensitivity in acute schizophrenia

    J Appl Physiol

    (2007)
  • Bar KJ, Koschke M, Boettger MK, Kabisch A, Sauer H, Voss A, et al. Acute psychosis leads to increased QT variability in...
  • M. Baumert et al.

    Forecasting of life threatening arrhythmias using the compression entropy of heart rate

    Methods Inf Med

    (2004)
  • M. Baumert et al.

    Estimating the complexity of heart rate fluctuations – an approach based on compression entropy

    Fluctuation Noise Lett

    (2004)
  • F. Beckers et al.

    Fractal dimension in health and heart failure

    Biomed Tech (Berl)

    (2006)
  • S. Brown et al.

    Causes of the excess mortality of schizophrenia

    Br J Psychiatry

    (2000)
  • N.H. Covell et al.

    Health monitoring for patients who have schizophrenia. Summary of the Mount Sinai Conference recommendations

    Postgrad Med

    (2006)
  • C. Enger et al.

    Serious cardiovascular events and mortality among patients with schizophrenia

    J Nerv Ment Dis

    (2004)
  • M.B. First

    Clinical utility: a prerequisite for the adoption of a dimensional approach in DSM

    J Abnorm Psychol

    (2005)
  • A.H. Glassman et al.

    Antipsychotic drugs: prolonged QTc interval, torsade de pointes, and sudden death

    Am J Psychiatry

    (2001)
  • S.M. Grundy et al.

    Definition of metabolic syndrome: report of the national heart, lung and blood institute/American Heart Association conference on scientific issues related to definition

    Arterioscler Thromb Vasc Biol

    (2004)
  • R. Jindal et al.

    Cardiac risk and schizophrenia

    J Psychiatry Neurosci

    (2005)
  • Cited by (123)

    • On the role of artificial intelligence in psychiatry

      2023, British Journal of Psychiatry
    • Longitudinal analysis of heart rate and physical activity collected from smartwatches

      2024, CCF Transactions on Pervasive Computing and Interaction
    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text