Download Heart rate variability response to mental arithmetic stress in patients

Available online at www.sciencedirect.com
Schizophrenia Research 99 (2008) 294 – 303
www.elsevier.com/locate/schres
Heart rate variability response to mental arithmetic stress
in patients with schizophrenia
Autonomic response to stress in schizophrenia
Mariana N. Castro a,b,c , Daniel E. Vigo c,d , Hylke Weidema a,e , Rodolfo D. Fahrer b ,
Elvina M. Chu f , Delfina de Achával a , Martín Nogués a , Ramón C. Leiguarda a ,
Daniel P. Cardinali c,d , Salvador M. Guinjoan a,b,c,d,⁎
a
b
Department of Neurology, Fundación Lucha contra Enfermedades Neurológicas de la Infancia (FLENI), Buenos Aires, Argentina
Department of Psychiatry, Fundación Lucha contra Enfermedades Neurológicas de la Infancia (FLENI), Buenos Aires, Argentina
c
Department of Physiology, University of Buenos Aires School of Medicine, Buenos Aires, Argentina
d
CONICET, Argentina
e
Rijksuniversiteit Groningen, The Netherlands
f
Institute of Neurology, University College London, London, United Kingdom
Received 3 July 2007; received in revised form 23 August 2007; accepted 24 August 2007
Available online 30 October 2007
Abstract
Background: The vulnerability-stress hypothesis is an established model of schizophrenia symptom formation. We sought to
characterise the pattern of the cardiac autonomic response to mental arithmetic stress in patients with stable schizophrenia.
Methods: We performed heart rate variability (HRV) analysis on recordings obtained before, during, and after a standard test of
autonomic function involving mental stress in 25 patients with DSM-IV schizophrenia (S) and 25 healthy individuals (C).
Results: Patients with schizophrenia had a normal response to the mental arithmetic stress test. Relative contributions of lowfrequency (LF) HRV and high-frequency (HF) HRV influences on heart rate in patients were similar to controls both at rest (LF 64 ±
19% (S) vs. 56 ± 16% (C); HF 36 ± 19% (S) vs. 44 ± 16% (C), t = 1.52, p = 0.136) and during mental stress, with increased LF (S: 76 ±
12%, C: 74 ± 11%) and decreased HF (S: 24 ± 12%, C: 26 ± 11%) in the latter study condition. Whilst healthy persons recovered the
resting pattern of HRV immediately after stress termination (LF 60 ± 15%, HF 40 ± 15%, F = 18.5, p b 0.001), in patients HRV
remained unchanged throughout the observed recovery period, with larger LF (71 ± 17%) and lower HF (29 ± 17%) compared with
baseline (F = 7.3, p = 0.013).
Conclusions: Patients with schizophrenia exhibit a normal response to the mental arithmetic stress test as a standard test of
autonomic function but in contrast with healthy individuals, they maintain stress-related changes of cardiac autonomic function
beyond stimulus cessation.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Schizophrenia; Stress; Autonomic nervous system; Heart rate variability
⁎ Corresponding author. Cognitive Neurology & Neuropsychiatry Section, FLENI — Montañeses 2325 8th floor, C1428AQK Buenos Aires,
Argentina. Tel.: +54 11 57773200x2824; fax: +54 11 57773209.
E-mail address: [email protected] (S.M. Guinjoan).
0920-9964/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.schres.2007.08.025
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
1. Introduction
Schizophrenia is characterised by macroscopic and
histopathological abnormalities of cortical structures
concerned with autonomic control, including the hippocampi, and temporal, cingulate, and prefrontal areas.
(McDonald et al., 2004). Although it is well established
that schizophrenia is highly heritable (Kendler et al.,
1994; Cardno et al., 1999), the syndrome probably
requires the concerted effect of several susceptibility
genes, in addition to exposure to environmental noxious
stimuli (e.g. perinatal damage and psychosocial stressors) in order to be expressed (McDonald and Murray,
2000). Environmental stress influences the time of
emergence of symptoms of psychosis (Doering et al.,
1998), and according to this “vulnerability-stress model”
(Nuechterlein and Dawson, 1984), psychotic symptoms
appear whenever stressors exceed the afflicted person's
vulnerability level, which is considered a stable, or
“trait,” characteristic (Nuechterlein and Dawson, 1984).
The autonomic nervous system is a major bodily
mediator of stressful emotions, hence theories of
symptom development in schizophrenia have long
incorporated the notion of autonomic dysfunction as
central to the manifestation of psychosis. Kraepelin
(1899) identified extensive autonomic alterations in
schizophrenic patients, including pupillary, vasomotor,
sweating, heart rate, salivation, and temperature changes,
most of them suggesting increased sympathetic output,
decreased parasympathetic activity, or both. Later studies
on the subject have confirmed and extended those early
observations (Lindstrom, 1996). Most available data
about autonomic responses to environmental stress in
schizophrenia refer to alterations of electrodermal
responses (Venables, 1992; Grossberg, 2000). Although
skin conductance orienting “hyporesponsiveness” in
schizophrenia was suggested to be related to attentional
and arousal deficits characteristic of the disorder (Dawson
et al., 1994), skin lacks significant parasympathetic
innervation.
In the last two decades heart rate variability (HRV)
analysis has been developed as a tool to probe the peripheral autonomic output in the cardiovascular territory
(Boettger et al., 2006). Most available data obtained with
this tool in schizophrenia refer to acute states and show
vagal withdrawal or decreased baroreflex sensitivity
(Boettger et al., 2006), which support a status of baseline
autonomic “hyperarousal” (Williams et al., 2004) in acute
schizophrenia. We sought to probe the autonomic reaction
to stress in patients with schizophrenia using this tool.
In light of previous observations, we hypothesised that
in the patient sample the cardiovascular autonomic status
295
at rest would be characterised by an increased sympathetic/vagal balance and response to stress would be less
intense than in controls, paralleling the “hyporesponsiveness” observed in electrodermal studies. Patients with
schizophrenia have increased cardiovascular morbidity
and cardiac autonomic alterations have been implicated in
the increased cardiovascular mortality (Ewing, 1992;
Hennekens et al., 2005). Alterations of the cardiac
autonomic response to stress may therefore have a
cardiovascular prognostic value as well.
2. Methods
2.1. Participants
2.1.1. Patients
Psychiatry outpatients reaching DSM-IV diagnostic
criteria were invited to participate in the study if (a)
diagnosis was confirmed with a Composite International
Diagnostic Interview (Robins et al., 1988) administered
by a psychiatrist participating in the study, (b) aged 18 to
75 years, and (c) on stable medication for at least two
weeks. Exclusion criteria were (a) use of illegal substances in the previous 6 months or a history of substance abuse/dependence (b) active symptoms having
recently (b2 weeks) warranted antipsychotic dose
adjustment or admission to the hospital, (c) history of
mental retardation, (d) cardiac arrhythmias, (e) presence
of diseases associated with neuropathy or autonomic
nervous system involvement (e. g. diabetes), (f) use of
medication with anticholinergic activity in the preceding
week. Patients were identified by a staff-grade psychiatrist and referred only after both patient and accompanying first-degree relative signed an informed consent,
as approved by the local ethics committee. Severity of
schizophrenia symptoms was measured by the Positive
and Negative Syndrome Scale (PANSS) (Kay et al.,
1987) administered the day of the heart rate testing
session by a psychiatrist who was blind to the results of
heart rate variability analysis.
2.1.2. Controls
Healthy comparison individuals were recruited from
staff at FLENI, and from local community attendees to
free lectures related to health promotion as advertised in
posters and the media. Individuals were selected to
match patients' gender and age. All participants signed
an informed consent form as approved by the ethics
committee. Exclusion criteria included (a) the lifetime
presence of any DSM-IV Axis I anxiety, mood, or
psychotic disorder diagnosis as detected by a psychiatric
interview with a consultant psychiatrist and a medication
296
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
history of antidepressants, antipsychotics, anticonvulsants or lithium, (b) the current use of any psychopharmacologic agent, or any medication with anticholinergic
activity in the previous week, (c) cardiac arrhythmia, or
(d) medical conditions potentially associated with
autonomic nervous system dysfunction (including
diabetes mellitus and Parkinson's disease).
2.2. Testing session
Participants were tested between 0900 h and 1300 h, 2–
4 h post-prandial, and having abstained from smoking for
2 h, consuming caffeinated beverages for 6 h, and vigorous
exercise for 24 h. Testing was carried out whilst seated in a
quiet, dimly lit room after 10 min of rest. A baseline heart
rate variability recording with at least 550 beats was
obtained prior to the test. We used a standard test of autonomic function, the mental arithmetic task to induce mental
stress (Ewing, 1992), which consisted in subtracting serial
7's starting from 700. Patients were requested to keep their
eyes closed and pronounce aloud each response. The test
was chosen because it consistently induces sustained
psychological stress resulting in an autonomic profile of
sympathetic activation and parasympathetic inhibition,
manifested by sustained increases in heart rate and blood
pressure (Ewing, 1992) during the time necessary to collect
Table 1
Sample characteristics
Age (years)
Female
Body Mass Index
(kg/m2)
Education (years)
Smokers
Age at onset (years)
Disease duration
(years)
Total PANSS
Positive subscale
Negative subscale
Haloperidol
Novel antipsychotic
Risperidone
Olanzapine
Clozapine
Ziprasidone
Quetiapine
Selective Serotonin
Reuptake Inhibitor
Benzodiazepine
Patients
(n = 25)
Controls
(n = 25)
Statistic
p
34 ± 11
9 (36)
26 ± 6
34 ± 14
9 (36)
23 ± 1
t = − 058
0.954
χ2 = 0
1
t = − 2.485 0.016
11 ± 4
7 (29)
23 ± 6
10 ± 8
14 ± 3
2 (8)
–
–
t = 2.853
χ2 = 3.659
–
–
0.007
0.056
–
–
86 ± 24
16 ± 7
27 ± 11
5 (20)
–
–
–
–
–
–
–
–
–
–
–
–
10 (40)
8 (32)
3 (12)
2 (8)
1 (4)
8 (32)
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
12 (48)
–
–
–
PANSS: Positive and Negative Symptom Scale. Shown are mean ±
standard deviation or number (percentage).
Fig. 1. Non-spectral measures of heart rate variability obtained in
resting conditions in patients with schizophrenia and healthy
individuals. ⁎: p b 0.05 as compared to healthy individuals, independent-samples t test. SDNN: Standard deviation of all normal beats.
rMSNN: Root-mean-square of normal beat-to-beat differences. See the
text for details.
at least 550 beats, which was estimated to be 7 min in a pilot
study. Heart rate signal data was collected for an additional
7 min after participants were instructed to stop the task to
assess the autonomic recovery from mental stress.
Respiratory rate was visually monitored simultaneously.
2.3. Heart rate data acquisition and analysis
We analysed HRV on recordings of successive RR
intervals of sinus node origin as described previously
(Task Force, 1996; Guinjoan et al., 2004). Patients were
connected to an interphase which detected R-waves in a
surface electrocardiographic signal (as obtained from a
V4 or a V5 lead) and fed into a computer which stored
RR intervals sampled at 1250 Hz (VariaDat® HRV
package, University of Entre Ríos School of Bioengineering). Premature and lost beats were individually
identified and manually tagged in the file of RR
intervals. These abnormal beats were replaced for RR
intervals resulting from linear interpolation. Time-(non-
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
297
Fig. 2. Spectral components of heart rate variability in resting conditions in patients with schizophrenia and normal individuals. Panel A depicts
recordings of successive normal beats (top) and their corresponding power spectra (bottom) obtained in resting conditions in an individual with
schizophrenia (left) and a healthy person of similar age (right). Panel B plots relative contributions of low- and high-frequency heart rate variability
components in the whole group of participants. See the text for details.
spectral) and frequency-(spectral) domain measures of
HRV were obtained (Task Force, 1996).
2.3.1. Non-spectral HRV analysis
Time domain measures of HRV included mean RR
interval (RR), standard deviation of all normal RR
intervals (SDNN), and rMSSD (square root of the mean
squared differences of successive normal RR intervals).
SDNN estimates global HRV, mediated by sympathetic
and parasympathetic nervous systems. rMSNN reflects
short-term, beat-to-beat variations in heart rate.
2.3.2. Spectral HRV analysis
Frequency domain measures of HRV were quantified
through fast Fourier transform (Task Force, 1996) and
included low-frequency power (LF, 0.03–0.15 Hz),
298
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
Fig. 3. Non-spectral measures of heart rate variability obtained during
mental stress and immediately after it in patients with schizophrenia
and healthy individuals. ⁎: p = 0.022 and ⁎⁎: p b 0.01 vs. controls,
independent-samples t test; a: F = 10.7, p b 0.001 vs. baseline
and post-task, b: F = 29.3, p b 0.001 vs. baseline and post-task, and
c: F = 11.7, p = 0.005 vs. baseline, repeated-measures ANOVA
followed by post hoc Bonferroni analysis. SDNN: Standard deviation
of all normal beats. rMSNN: Root-mean-square of normal beat-to-beat
differences. See the text for details.
high-frequency power (HF, 0.15–0.40 Hz), and the LF/
HF ratio (L/H). The purpose of the Fourier analysis is to
decompose a complex time series f(t) with cyclical
components into the underlying sinusoidal functions of
particular wavelengths. Mathematically, the process of
Fourier analysis is represented by the Fast Fourier
Transform, which is the sum over all time of the signal f
(t) multiplied by a complex exponential. The results of
the transform are the Fourier coefficients; the squared
coefficients for each frequency are the power spectral
density of the signal f(t). The goal of spectral estimation
is to describe the distribution (over frequency) of the
power contained in a signal. The average power of a
signal over a particular frequency band can be found by
integrating the power spectral density over that band.
The power spectral density of the original heart rate
signal thus provides information of how variance distributes as a function of frequency (Task Force, 1996),
and was calculated with Matlab V6 software. The highfrequency (HF) component of HRV is related to
respiratory sinus arrhythmia (mediated by parasympathetic activity), and the low-frequency (LF) component
is related to Mayer waves of blood pressure, which
influence heart rate via the baroreceptor reflex. Thus LF
depends on both sympathetic and parasympathetic
influences, although in most instances (with the notable
exceptions of strenuous physical exercise and heart
failure) it is thought to reflect mainly sympathetic activation (Malliani et al., 1991; Montano et al., 1994; Task
Force, 1996). We report absolute (natural logarithm of
HRV amplitude, ms2) values and normalised units of
each component. Normalised units are calculated in
relative terms as follows: LF: LF / (Total power − very lowfrequency power) × 100, and HF: HF / (Total power − very
low-frequency power)× 100. This form of expression
permits the comparison of groups of individuals with
different absolute values of HRV, because when the
spectral components are expressed in absolute units, the
changes in total power influence LF and HF in the
same direction and prevent the accurate appreciation of
the fractional distribution of the energy (Task Force,
1996).
2.4. Statistical analysis
Discrete variables in patients and controls were
compared using the chi-square test (χ2), and continuous
variables were compared with an independent-samples t
test. This method was also used to compare HRV
measurements in patients on risperidone with those who
were not taking it, as this antipsychotic was the most
commonly used in the present sample and may affect heart
rate. Within-subject comparisons of HRV data obtained
during baseline, mental stress, and recovery phases were
performed with a repeated-measures ANOVA followed
by a post hoc Bonferroni test. PANSS scores were correlated with HRV measures with a Pearson's r test. Significance was assumed at α = 0.05, after a Bonferroni
correction was applied to multiple measurements. All
results are two-tailed.
3. Results
Table 1 shows the demographic and clinical characteristics of patients and controls. Two women and one man
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
declined to participate in the study. Patients showed a
trend towards greater body mass index (BMI), and were
less educated than controls. All patients had been on
stable doses of antipsychotics for at least two weeks; three
were receiving two different antipsychotics simultaneously. Respiratory rate ranged between 13 and 21 cpm (0.22–
0.35 Hz) in all participants throughout testing.
3.1. Baseline heart rate variability (HRV)
Baseline non-spectral data for the entire sample of
patients and controls are shown in Fig. 1. Patients had a
higher heart rate (Fig. 1 top panel) and a globally
decreased HRV (reflected by smaller SDNN and
rMSNN, Fig. 1 bottom panel). Fig. 2A shows the
tachograms (i.e., the series of sinus RR intervals, top
panels) and their corresponding power spectra (bottom
panels) obtained at baseline in a 34-year-old male with
schizophrenia (left), and a 27-year-old male in the
healthy comparison group (right). The patient displays a
higher heart rate (i.e., shorter RR intervals) and
decreased HRV at baseline compared to the healthy
person. Spectral measures of HRV suggest that these
changes in heart rate and non-spectral HRV are mostly
due to high-frequency (HF) oscillations (4.03 ± 1.28 ln
ms2 vs. 5.02 ± 1.37 ln ms2 in healthy persons, t = − 2.64,
p = 0.011), as HRV in the low-frequency (LF) spectrum
exhibited a tendency to be lower in patients (4.68 ± 1 ln
ms2 vs. 5.34 ± 1.3 ln ms2 in healthy persons, t = − 2,
p = 0.051). Relative HF and LF contributions to total
HRV, as expressed by normalised units, were on the
other hand similar in patients and controls (Fig. 2 B).
Accordingly, L/H was similar in both groups as well
(2.7 ± 2.3 in patients vs. 2 ± 2.3 in controls).
3.2. Mental stress-related changes in HRV
Patients and controls had a normal response to the
mental arithmetic test, as indicated by the significant
tachycardia associated with it (i.e., shortening of average
RR interval), which was immediately followed by a
recovery of the baseline heart rate (Fig. 3, top panel). Fig. 3
(bottom panel) depicts mental stress-related changes in
non-spectral HRV, which reveals smaller variability in
patients. Fig. 4A displays oscillations of heart rate during,
and immediately after, a 7-minute period of mental stress
in the same individuals as in Fig. 2A. Shown are the
tachograms (top), the power spectra corresponding to the
task itself (middle) and post-task recovery phase (bottom)
in the same patient (left) and healthy person (right) as in
Fig. 2A. Mental stress brought about a shortening of RR
intervals, followed by a noticeable interval lengthening
299
after the task (Fig. 4A, top graphics). On the other hand,
the recovery of frequency components of HRV after the
period of stress is most noticeable in the normal individual
(Fig. 4A, right middle and bottom graphics), whereas the
patient with schizophrenia maintains a similar pattern of
HRV during mental stress (Fig. 4A, left middle graphic)
and during recovery (Fig. 4A, left bottom graphic). In the
sample as a whole, the response of each autonomic
component was different in healthy individuals and
persons with schizophrenia. In normal individuals, mental
stress evoked an increase in absolute LF power, which
abated immediately after the task (5.34 ± 1.3 ln ms2
baseline vs. 6.09± 1.03 ln ms2 mental stress vs. 5.44± 1.25
ln ms2 post-task, F = 8.3, p b 0.012), and these changes
were paralleled by equal changes in LF HRV relative to
total HRV as expressed by normalised units (Fig. 4, panel
B). Whereas absolute HF power did not change
significantly during the task, the relative contribution of
HF in terms of normalised units showed a significant
withdrawal during the mental stress-inducing task followed by restoration to baseline levels immediately after
mental stress cessation (Fig. 4, panel B). HRV response to
the mental stress-inducing task in patients was similar to
that attained by healthy persons, namely, an increase in
absolute power in the LF range (4.68 ± 1 ln ms2 baseline
vs. 5.36± 0.95 ln ms2 mental stress), and both an increase
in the relative LF contribution to total variability, and
a significant decrease in relative HF HRV as measured
by normalised units, so that both relative HRV components were indistinguishable from controls during stress
(Fig. 4, panel B). This is also reflected in a similar
sympathovagal balance in both groups as measured by the
L/H quotient (4.5± 3.2 in patients vs. 3.7 ± 2.5 in controls).
However, in contrast to healthy participants, patients with
schizophrenia displayed a prolonged autonomic activation
as indicated by sustained increase in absolute LF power
(5.25 ± 1.01 ln ms2, F = 11.6, p b 0.001 vs. baseline LF),
and decreased normalised HF variability throughout the
observed recovery period (Fig. 4, panel B). Normalised
LF in patients was also significantly higher than that
observed in controls as far as the post-task period (Fig. 4,
panel B). Consistent with these observations, sympathovagal balance reflected by L/H quotient was higher in
patients with schizophrenia after stress cessation (4.1± 3.9
vs. 2.2 ± 2.2 in controls, t = 2.13, p = 0.039).
3.3. Relationship between psychotic symptoms or
antipsychotic use and HRV
No positive or negative correlations were found
between the negative and positive subscales or total
PANSS scores, and measures of HRV during baseline,
300
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
mental stress, or post-task recordings. Average duration
of RR intervals and different non-spectral and spectral
measures of HRV were similar in patients whether
taking risperidone or other antipsychotics.
4. Discussion
This study shows that patients with schizophrenia
exhibit (a) a decreased baseline HRV and an increased
baseline heart rate probably related to impaired parasympathetic input to the heart (as measured by HF-HRV), (b) a
normal heart rate response to the standard mental
arithmetic stress test of autonomic function, (c) a normal
pattern and intensity of the cardiac autonomic response to
mental stress in terms of increased sympathetic input (as
inferred from increases in LF-HRV) and decreased vagal
input to the heart (as indicated by decreases in HF-HRV),
and (d) an abnormal prolongation of this pattern once
stress is terminated, as compared to healthy individuals
who exhibit an immediate recovery of the cardiac
autonomic balance characteristic of baseline resting
conditions.
In humans, stimulation of the insula, medial prefrontal
cortex and anterior cingulate, and medial temporal lobe
elicit changes in blood pressure and heart rate that are
occasionally accompanied by subjective mood changes
(Critchley et al., 2000). Critchley et al. (2000) additionally
characterised activation of brain areas in association to
stressful mental arithmetic, confirming significant involvement of cingulate cortex (particularly on the right
anterior region), cerebellar vermis, and pons. Most of
these areas have been implicated by structural and
functional brain studies of schizophrenia (Lawrie and
Abukmeil, 1998; Harrison, 1999; Wright et al., 2000) and
the autonomic alterations described may ultimately reflect
the alteration of those central structures, expressed as an
inability to “switch off” the cardiovascular autonomic
profile characteristic of mental stress. Williams et al.
(2004) proposed that schizophrenia might be characterised by a disjunction of autonomic arousal and
amygdala–prefrontal circuits that impede appropriate
processing of stressful signals. Excessive arousal together
with decreased amygdala activation in schizophrenia
301
might reflect a dysregulation of the normal positive feedback between amygdala function and autonomic activity
due to failure of regulation by prefrontal structures, thus
leading to a perseveration and exacerbation of arousal
responses (Williams et al., 2004). This phenomenon
would then lead to an internally generated cycle of
hypervigilance and misattribution that feeds into paranoid
cognition (Williams et al., 2004; Phillips et al., 2000).
Regarding negative symptom generation, some of the
central structures potentially involved in autonomic
control additionally mediate social and emotional behaviour; hence both these areas of functioning are compromised when disrupted (Damasio et al., 1991). It has been
recently documented that right ventromedial prefrontal
lesions result in excessive sympathetic responses to
stressful stimuli (Hilz et al., 2006). Orbitofrontal damage
reduces anticipatory arousal to emotive stimuli, whilst
lesions of the amygdala block autonomic responses that
accompany conditioning (Damasio et al., 1991). Lesions
of these areas are associated with changes in social
behaviour and emotional responses, suggesting that
feedback of altered autonomic arousal (represented in
cortical regions such as the orbitofrontal/ventromedial
prefrontal cortex) may directly influence social behaviour
and decision making (Damasio et al., 1991).
An additional potential clinical implication refers to
mounting evidence on the role of cardiac autonomic
dysfunction as a culprit for increased cardiovascular mortality observed in individuals with schizophrenia. Increased sympathetic and/or decreased parasympathetic
input to the heart have been repeatedly shown to increase
the proclivity to malignant ventricular arrhythmias and
myocardial ischaemia (Schwartz and Priori, 1990; Smith
et al., 2005), ultimately resulting in increased incidence of
cardiovascular deaths (Gerritsen et al., 2001). Patients with
schizophrenia indeed have an excessive cardiac mortality
when compared with healthy cohorts (Brown et al., 2000;
Hennekens et al., 2005). The emerging evidence in the
field indicates that schizophrenia patients consistently
display a decreased vagal input to the heart compared to
normal individuals, both in acute phases of the disease in
which patients have not yet received antipsychotics and in
stable phases whilst taking the latter (Toichi et al., 1999;
Fig. 4 Spectral components of heart rate variability obtained during mental stress and immediately after it in patients with schizophrenia and normal
individuals. Panel A shows recordings of successive normal beats obtained during performance of the mental arithmetic stress test and immediately
after its termination (top) and power spectra corresponding to the period of mental stress (middle) and recovery from it (bottom) in the same
participants as in Fig. 2A. Panel B plots relative contributions of low- and high-frequency heart rate variability components in the whole group of
participants. *: p b 0.03 vs. controls, independent-samples t test; a: F = 18.5, p = 0.001 vs. baseline and post-task; b: F = 7.3, p = 0.013 vs. baseline;
c: F = 7.3, p = 0.054 vs. baseline; d: F = 7.42, p = 0.013 vs. baseline, and e: F = 7.42, p = 0.036 vs. baseline, repeated-measures ANOVA followed by
post hoc Bonferroni analysis. See the text for details.
302
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
Bar et al., 2005; Boettger et al., 2006). Along with
accumulating evidence of reduced vagal tone, our results
indicative of protracted stress-related cardiac sympathetic
responses suggest that this factor should also be taken into
account in future studies as a potential mechanism for the
association. In general, typical high-potency antipsychotics like haloperidol and most novel agents have limited
effects on HRV recordings (Rechlin et al., 1994; Bar et al.,
2005; Boettger et al., 2006), whereas clozapine is
consistently associated with a significant reduction in
total HRV (Rechlin et al., 1994; Cohen et al., 2001;
Mueck-Weymann et al., 2002). In our study, with the main
finding being the presence of a protracted stress-related
HRV profile, it therefore seems unlikely that the latter was
caused by clozapine, used in three patients.
Anxiety (including obsessive–compulsive) disorders
are sometimes comorbid with schizophrenia and in most
cases have been shown to exhibit diagnosis-specific
abnormalities of heart rate variability or heart rate responses to stress (Buhlmann et al., 2007; Friedman,
2007). Whilst our patients did not exhibit comorbid
anxiety disorders, further studies that investigate the
contribution of anxiety to the described findings are
probably warranted.
In summary, in a standard test of autonomic function
involving mental stress, patients with schizophrenia
exhibit an autonomic response that is abnormally maintained beyond the cessation of the stress source, as compared to healthy individuals. It remains to be further
elucidated to what extent this pattern is related to increased
cardiovascular mortality and precipitation of psychotic
episodes in schizophrenia.
Role of the funding source
MNC is a medical student research fellow from the University of
Buenos Aires (Project M084). DEV is a fellow from the Argentine
National Council on Scientific and Technological Research, CONICET. This study was supported in part by FLENI Foundation,
UBACyT (Project M084, SMG), and CONICET (SMG, DPC). These
founding sources had no involvement in the study design, in the
collection, analysis and interpretation of data, in the writing of the
report, and in the decision to submit the paper for publication.
Contributors
SMG and MNC designed the study and wrote the protocol. SMG,
MNC, HW, and EC managed the literature searches and analyses. SMG,
MNC, HW, RDF, MN, DA, and RCL collected experimental data. DEV,
SMG, MNC, DPC, and HW undertook the statistical analysis, and MNC
wrote the first draft of the manuscript. All authors contributed to and have
approved the final manuscript.
Conflict of interest
All authors declare that they have no conflicts of interest.
Acknowledgement
No acknowledgements.
References
Bar, K.J., Letzsch, A., Jochum, T., Wagner, G., Greiner, W., Sauer, H.,
2005. Loss of efferent vagal activity in acute schizophrenia. J.
Psychiatr. Res. 39, 519–527.
Boettger, S., Hoyer, D., Falkenhahn, K., Kaatz, M., Yeragani, V.K.,
Bar, K.J., 2006. Altered diurnal autonomic variation and reduced
vagal information flow in acute schizophrenia. Clin. Neurophysiol.
117, 2715–2722.
Brown, S., Inskip, H., Barraclough, B., 2000. Causes of the excess
mortality of schizophrenia. Br. J. Psychiatry 177, 212–217.
Buhlmann, U., Wilhem, S., Deckersbach, T., Rauch, S.L., Pitman, R.K.,
Orr, S.P., 2007. Physiologic responses to loud tones in individuals with
obsessive–compulsive disorder. Psychosom. Med. 69, 166–172.
Cardno, A.G., Marshall, E.J., Coid, B., Macdonald, A.M., Ribchester, T.R.,
Davies, N.J., Venturi, P., Jones, L.A., Lewis, S.W., Sham, P.C.,
Gottesman, I.I., Farmer, A.E., McGuffin, P., Reveley, A.M., Murray,
R.M., 1999. Heritability estimates for psychotic disorders: the
Maudsley twin psychosis series. Arch. Gen. Psychiatry 56, 162–168.
Cohen, H., Loewenthal, U., Matar, M., Kotler, M., 2001. Association
of autonomic dysfunction and clozapine. Heart rate variability and
risk for sudden death in patients with schizophrenia on long-term
psychotropic medication. Br. J. Psychiatry 179, 167–171.
Critchley, H.D., Corfield, D.R., Chandler, M.P., Mathias, C.J., Dolan, R.J.,
2000. Cerebral correlates of autonomic cardiovascular arousal: a
functional neuroimaging investigation in humans. J. Physiol. 523 (Pt 1),
259–270.
Damasio, A.R., Tranel, D., Damasio, H.C., 1991. Frontal Lobe Function
and Dysfunction. Oxford University Press, Oxford, pp. 217–229.
Dawson, M.E., Nuechterlein, K.H., Schell, A.M., Gitlin, M., Ventura,
J., 1994. Autonomic abnormalities in schizophrenia. State or trait
indicators? Arch. Gen. Psychiatry 51, 813–824.
Doering, S., Muller, E., Kopcke, W., Pietzcker, A., Gaebel, W.,
Linden, M., Muller, P., Muller-Spahn, F., Tegeler, J., Schussler, G.,
1998. Predictors of relapse and rehospitalization in schizophrenia
and schizoaffective disorder. Schizophr. Bull. 24, 87–98.
Ewing, D.J., 1992. Autonomic Failure: A Textbook of Clinical
Disorders of the Autonomic Nervous System, third edn. Oxford
Medical Publications, Oxford, pp. 326–327.
Friedman, B.H., 2007. An autonomic flexibility-neurovisceral integration model of anxiety and cardiac vagal tone. Biol. Psychiatry
74, 185–199.
Gerritsen, J., Dekker, J.M., TenVoorde, B.J., Kostense, P.J., Heine, R.J.,
Bouter, L.M., Heethaar, R.M., Stehouwer, C.D., 2001. Impaired
autonomic function is associated with increased mortality, especially
in subjects with diabetes, hypertension, or a history of cardiovascular
disease: the Hoorn Study. Diabetes Care 24, 1793–1798.
Grossberg, S., 2000. The imbalanced brain: from normal behavior to
schizophrenia. Biol. Psychiatry 48, 81–98.
Guinjoan, S.M., de Guevara, M.S., Correa, C., Schauffele, S.I.,
Nicola-Siri, L., Fahrer, R.D., Ortiz-Fragola, E., Martinez-Martinez, J.A., Cardinali, D.P., 2004. Cardiac parasympathetic dysfunction related to depression in older adults with acute coronary
syndromes. J. Psychosom. Res. 56, 83–88.
Harrison, P.J., 1999. The neuropathology of schizophrenia. A critical
review of the data and their interpretation. Brain 122 (Pt 4), 593–624.
Hennekens, C.H., Hennekens, A.R., Hollar, D., Casey, D.E., 2005.
Schizophrenia and increased risks of cardiovascular disease. Am.
Heart J. 150, 1115–1121.
Hilz, M.J., Devinsky, O., Szczepanska, H., Borod, J.C., Marthol, H.,
Tutaj, M., 2006. Right ventromedial prefrontal lesions result in
M.N. Castro et al. / Schizophrenia Research 99 (2008) 294–303
paradoxical cardiovascular activation with emotional stimuli.
Brain 129, 3343–3355.
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative
syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13,
261–276.
Kendler, K.S., Gruenberg, A.M., Kinney, D.K., 1994. Independent
diagnoses of adoptees and relatives as defined by DSM-III in the
provincial and national samples of the Danish Adoption Study of
Schizophrenia. Arch. Gen. Psychiatry 51, 456–468.
Kraepelin, E., 1899. Psychiatrie. Ein Lehrbuch fur Studirende und
Aertze, Psychiatry. A Textbook for Students and Physicians, 1991
edn. Watson Publishing International, Canton, MA, pp. 110–111.
Lawrie, S.M., Abukmeil, S.S., 1998. Brain abnormality in schizophrenia. A systematic and quantitative review of volumetric magnetic
resonance imaging studies. Br. J. Psychiatry 172, 110–120.
Lindstrom, L.H., 1996. Clinical and biological markers for outcome in
schizophrenia: a review of a longitudinal follow-up study in
Uppsala schizophrenia research project. Neuropsychopharmacology 14, 23S–26S.
Malliani, A., Pagani, M., Lombardi, F., Cerutti, S., 1991. Cardiovascular neural regulation explored in the frequency domain.
Circulation 84, 1482–1492.
McDonald, C., Murray, R.M., 2000. Early and late environmental risk
factors for schizophrenia. Brain Res. Brain Res. Rev. 31, 130–137.
McDonald, C., Bullmore, E.T., Sham, P.C., Chitnis, X., Wickham, H.,
Bramon, E., Murray, R.M., 2004. Association of genetic risks for
schizophrenia and bipolar disorder with specific and generic brain
structural endophenotypes. Arch. Gen. Psychiatry 61, 974–984.
Montano, N., Gnecchi Ruscone, T., Porta, A., Lombardi, F., Pagani,
M., Malliani, A., 1994. Power spectrum analysis of heart rate
variability to assess the changes in sympathovagal balance during
graded orthostatic tilt. Circulation 90, 1826–1831.
Mueck-Weymann, M., Rechlin, T., Ehrengut, F., Rauh, R., Acker, J.,
Dittmann, R.W., Czekalla, J., Joraschky, P., Musselman, D., 2002.
Effects of olanzapine and clozapine upon pulse rate variability.
Depress. Anxiety 16, 93–99.
Nuechterlein, K.H., Dawson, M.E., 1984. A heuristic vulnerability/
stress model of schizophrenic episodes. Schizophr. Bull. 10,
300–312.
303
Phillips, M.L., Senior, C., David, A.S., 2000. Perception of threat in
schizophrenics with persecutory delusions: an investigation using
visual scan paths. Psychol. Med. 30, 157–167.
Rechlin, T., Claus, D., Weis, M., 1994. Heart rate variability in
schizophrenic patients and changes of autonomic heart rate
parameters during treatment with clozapine. Biol. Psychiatry 35,
888–892.
Robins, L.N., Wing, J., Wittchen, H.U., Helzer, J.E., Babor, T.F.,
Burke, J., Farmer, A., Jablenski, A., Pickens, R., Regier, D.A.,
1988. The Composite International Diagnostic Interview. An
epidemiologic instrument suitable for use in conjunction with
different diagnostic systems and in different cultures. Arch. Gen.
Psychiatry 45, 1069–1077.
Schwartz, P.J., Priori, S.G., 1990. Sympathetic nervous system and
cardiac arrhythmias. In: Zipes, D.P., Jalife, J. (Eds.), Cardiac
Electrophysiology: From Cell to Bedside. Saunders, Philadelphia,
pp. 330–343.
Smith, L.L., Kukielka, M., Billman, G.E., 2005. Heart rate recovery
after exercise: a predictor of ventricular fibrillation susceptibility
after myocardial infarction. Am. J. Physiol, Heart Circ. Physiol.
288, H1763–H1769.
Task Force, 1996. Heart rate variability. Standards of measurement,
physiological interpretation, and clinical use. Task Force of the
European Society of Cardiology and the North American Society
of Pacing and Electrophysiology. Eur. Heart J. 17, 354–381.
Toichi, M., Kubota, Y., Murai, T., Kamio, Y., Sakihama, M., Toriuchi,
T., Inakuma, T., Sengoku, A., Miyoshi, K., 1999. The influence of
psychotic states on the autonomic nervous system in schizophrenia. Int. J. Psychophysiol. 31, 147–154.
Venables, P.H., 1992. Hippocampal function and schizophrenia. Experimental psychological evidence. Ann. N.Y. Acad. Sci. 658, 111–127.
Williams, L.M., Das, P., Harris, A.W., Liddell, B.B., Brammer, M.J.,
Olivieri, G., Skerrett, D., Phillips, M.L., David, A.S., Peduto, A.,
Gordon, E., 2004. Dysregulation of arousal and amygdala–
prefrontal systems in paranoid schizophrenia. Am. J. Psychiatry
161, 480–489.
Wright, I.C., Rabe-Hesketh, S., Woodruff, P.W., David, A.S., Murray,
R.M., Bullmore, E.T., 2000. Meta-analysis of regional brain
volumes in schizophrenia. Am. J. Psychiatry 157, 16–25.