Research reportFive-minute recordings of heart rate variability in obsessive–compulsive disorder, panic disorder and healthy volunteers
Introduction
In all anxiety disorders, symptoms are present which suggest the involvement of the autonomous nervous system (ANS). In panic attacks, which can occur in several anxiety disorders, ANS symptoms are prevalent, such as palpitations, chest pain and shortness of breath. In panic disorder (PD), which is characterized by spontaneous panic attacks (American Psychiatric Association, 1994), several studies have investigated the involvement of the ANS (Bystritsky et al., 2000; Katerndahl, 1990; Papp et al., 1993; Roth et al., 1986; Seier et al., 1997; Weissman et al., 1987; Yeragani et al., 1989).
Recent studies have used the analysis of heart rate variability (HRV) as a tool to study the functioning of the ANS. This analysis of a simple, noninvasive electrocardiogram (ECG) elucidates the influence of the ANS on short-term regulation of the cardiovascular system (Appel et al., 1989; van Ravenswaaij-Arts et al., 1993). HRV is primarily controlled by the continuous interplay of the sympathetic and parasympathetic (vagal) branches of the ANS (Friedman and Thayer, 1998a). Spectral analysis of HRV quantifies the relative influence of each branch on heart rate (HR), as these rhythms occur at different frequencies. Two frequency bands are usually distinguished: the high frequency band (HF: around 0.25 Hz) is thought to be mediated by cardiac vagal tone, which depends on respiration, and a low frequency band (LF: around 0.10 Hz) which is mediated by both branches of the ANS (Akselrod et al., 1981; Pomeranz et al., 1985). The ratio of LF power to HF power (LF/HF) is commonly seen as an index of cardiac sympathovagal balance (Pagani et al., 1986).
In several studies it has been reported that PD patients are characterized by a reduced HRV, reduced HF power, increased LF or an elevated LF/HF ratio; for an extensive review see Friedman and Thayer (1998a). These HRV abnormalities seem to point to a disturbed balance between the vagal and the sympathetic branches of the ANS, with possible decreased vagal activity and increased sympathetic activity.
Not all studies have consistently shown this pattern: one research group reported no HRV abnormalities in PD patients (Asmundson and Stein, 1994; Stein and Asmundson, 1994). Possible explanations mentioned for these discrepant findings were factors such as apprehensiveness of subjects at the time of testing (Ito et al., 1999), age, fitness, illness duration and illness severity (Asmundson and Stein, 1994; Stein and Asmundson, 1994), or even methodological differences in the analysis of HRV (Friedman and Thayer, 1998b).
HRV has not been investigated very thoroughly in other anxiety disorders. To date, no studies in obsessive–compulsive disorder (OCD) have been carried out, using spectral analysis of HRV. Patients with OCD are characterized by intrusive thoughts and irresistible urges to perform ritualized actions. When patients attempt to control these thoughts and urges, anxiety is often reported and elevated ANS activity is evident (Boulougouris et al., 1977; Rabavilas et al., 1977; Zahn et al., 1996). Earlier studies on ANS functioning in OCD patients have shown conflicting results: some studies found higher levels of skin conductance or heart rate, indicative of elevated autonomic arousal (Benkelfat et al., 1991; Insel et al., 1985), whereas others found no differences with healthy controls (Hollander et al., 1991).
In OCD three studies have investigated baseline differences in heart inter-beat intervals (IBIs) as a marker of ANS function. Hoehn Saric et al. (1995) could not find baseline differences in IBIs between adult OCD patients and controls. They did find that OCD patients showed a nonspecific decreased physiological flexibility on provocation with two stressful psychological tasks. McCarthy et al. (1995) used an intake-rejection attentional paradigm to compare heart rate activity between OCD patients and high and low trait anxious controls. IBIs, measured at baseline and during tasks, did not differ between the groups. Zahn et al. (1996) measured autonomic changes during rest, mild stress periods and a signal detection task in a substantial group of children and adolescents with OCD. They were unable to find evidence for a specific dysfunction in HRV underlying childhood OCD.
One might conclude that ANS dysfunction is not a prominent feature of OCD, but this conclusion seems premature. A methodological flaw in the studies discussed above is that the authors reported on mean IBIs, measured at baseline, as a marker of ANS function. Mean IBIs, or mean heart rate, do not yield much information on the functioning of the ANS. They do offer information on autonomic arousal when used as a measure before and after a provocation of the ANS (Porges and Bohrer, 1990). Therefore, spectral analysis of HRV is needed in order to quantify the effect on heart rate of both branches of the ANS, and thus ANS functioning.
In the present study we compared ANS system functioning in PD patients, OCD patients and normal controls. This was done by means of 5-min recordings of HRV, analyzed with spectral analysis, with subjects sitting on a hospital bed. Because of the lack of consistency in reports on ANS dysfunction in OCD, we hypothesized that OCD patients do not differ from normal controls, matched on age. Furthermore, we expected to replicate earlier findings of reduced HRV in PD patients, as compared to normal controls.
As this comparison yielded unexpected results, we performed a second comparison, with substantially larger groups of PD patients and normal controls, to verify our first measurement.
Section snippets
Subjects: comparison I
For the comparison of HRV data of PD patients, OCD patients and normal controls, the HRV data was pooled of two studies, which are published separately (Boshuisen et al., 2001; Nielen et al., 2002), together with a set of measurements in normal controls, matched on age. Pooling data was straightforward, as all HRV measurement and subsequent analyses were carried out using the same protocol, as described below. Only subjects with valid HRV data were used in the pooled data set: data of four PD
Comparison I
As can be seen in Table 1, normal controls were measured at a later time of day, as compared to PD and OCD patients. The time of day did not have any influence on the HRV parameters (Pearson’s r<0.1).
The analysis of variance, with time of measurement and age as covariates, of heart rate and the HRV parameters yielded no statistically significant differences between the groups.
Comparison II
The larger sample comparison of PD patients (n=53) and normal controls (n=54) yielded similar results to Comparison I,
Discussion
To our knowledge, this is the first study which investigated HRV, by means of spectral analysis, in OCD patients. The main finding in this study was that mean heart rate and HRV parameters of OCD patients do not differ from normal controls. To our surprise this was also the case in PD patients, who did not differ from normal controls, even in a large sample.
As hypothesized, resting OCD patients were not characterized by HRV abnormalities in our static measurement of HRV functioning. Mean heart
Conclusion
On the basis of this study, one can only conclude that OCD and PD patients are not characterized by ANS abnormalities, as no evidence was found of diminished HRV in a large sample of resting OCD and PD patients, measured sitting on an hospital bed. However, this study does not rule out the possibility that alterations in ANS function in anxiety disorder patients are evident only when patients are acutely anxious or when they experience panic attacks. A further study should investigate this
References (42)
- et al.
Beat to beat variability in cardiovascular variables: noise or music?
J. Am. Coll. Cardiol.
(1989) - et al.
Plasma catecholamines and their metabolites in obsessive–compulsive disorder
Psychiatry Res.
(1991) - et al.
Autonomic dysregulation in panic disorder and in post-traumatic stress disorder: application of power spectrum analysis of heart rate variability at rest and in response to recollection of trauma or panic attacks
Psychiatry Res.
(2000) - et al.
Anxiety and autonomic flexibility: a cardiovascular approach
Biol. Psychol.
(1998) - et al.
Autonomic balance revisited: panic anxiety and heart rate variability
J. Psychosom. Res.
(1998) - et al.
Noradrenergic function in obsessive–compulsive disorder: behavioral and neuroendocrine responses to clonidine and comparison to healthy controls
Psychiatry Res.
(1991) Comparison of panic symptom sequences and pathophysiologic models
J. Behav. Ther. Exp. Psychiatry
(1990)- et al.
Altered heart rate variability in panic disorder patients
Biol. Psychiatry
(1995) - et al.
Relaxation-induced EEG alterations in panic disorder patients
J. Anxiety Disord.
(1997) - et al.
Cognitive influences on electrocortical and heart rate activity on obsessive–compulsive disorder
Int. J. Psychophysiol.
(1995)