| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Psychiatry (R.D.L., C.L.F.) and Psychology (L.S.), University of Arizona, Tucson, Arizona; the International Heart Institute of Montana, Missoula, Montana (C.L.); the National Center for Complementary and Alternative Medicine, National Institutes of Health, Bethesda, Maryland (M.A.C.); the Sarver Heart Center, University of Arizona, Tucson, Arizona (F.I.M., K.G., R.D.L.); the Department of Cardiology, University of Pavia and Policlinico S. Matteo IRCCS, Pavia, Italy (S.G.P., P.J.S.); and the Department of Epidemiology and Public Health, University College London, London, U.K. (A.S.).
Address correspondence and reprint requests to Richard D. Lane, MD, PhD, Department of Psychiatry, 1501 N. Campbell Ave., Tucson, AZ 85724-5002. E-mail: lane{at}email.arizona.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: We interviewed 25 IVF survivors (12 men, 13 women) and 25 matched comparison patients regarding life events during the 6 months and 24 hours preceding the cardiac event. The comparison group consisted of patients with an acute myocardial infarction or angina pectoris requiring angioplasty but without cardiac arrest. Judges independently rated written summaries of these interviews for psychosocial stress at each time point on a three-point scale (low, moderate, severe).
Results: During the 6 months before the cardiac event, 20 patients sustaining IVF had severe/moderate stress and five had low stress, whereas 10 comparison patients had severe/moderate stress and 15 had low stress (Fisher exact p = .008). During the preceding 24 hours, nine patients with IVF had severe/moderate stress and 16 had low stress, whereas two comparison patients had severe/moderate stress and 22 had low stress (Fisher exact p = .04) (one silent myocardial infarction could not be precisely dated).
Conclusion: These data suggest that psychosocial stress is playing a role in otherwise unexplained cardiac arrest.
Key Words: idiopathic ventricular fibrillation • sudden cardiac death • psychologic stress • coronary heart disease
Abbreviations: CHD = coronary heart disease; DSM-IV = Diagnostic and Statistical Manual, Fourth Edition; ICD = implantable cardioverter defibrillator; IVF = idiopathic ventricular fibrillation; IVF-US = Idiopathic Ventricular Fibrillation Registry of the United States; UCARE = Unexplained Cardiac Arrest Registry of Europe.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
A small proportion of cardiac arrests and sudden cardiac deaths occur in patients with apparently normal hearts and no evidence of coronary heart disease (8–10). Registries of survivors of idiopathic ventricular fibrillation (IVF) have been established in Europe and in the United States (10). The etiology of IVF remains poorly understood (11), and the factors precipitating an episode are unclear. More than 20 years ago, life-threatening ventricular arrhythmias in the absence of structural heart disease were associated with increased risk of acute emotional disturbance during the 24 hours preceding onset (12). However, a systematic examination of the role of stress in IVF has not previously been undertaken.
Sudden death is thought to result from an interaction between a substrate and a trigger, as exemplified by the Long QT Syndrome (13). Substantial evidence also suggests that as the severity of the substrate increases, the intensity of the trigger required to induce an event decreases (12,15,16). We were therefore keenly interested in evaluating the role of psychologic stress preceding cardiac arrest in patients with IVF because in the absence of an identifiable substrate, the principle of an inverse relationship between substrate and trigger suggests that psychological stress plays a prominent precipitating role in this context.
We therefore predicted that psychological stress would be observed frequently before cardiac arrest in patients with IVF. Because the precise methodology that we used to evaluate psychosocial stress in patients with IVF was adapted to the unique circumstances of this patient group and because this exact approach has not been used in previous studies of the circumstances preceding cardiac events, we needed to study a comparison group to ensure that our findings were valid. Given that a variety of studies have shown that psychosocial stress occurs before acute coronary events in patients with CHD, we used the same methodology to compare our findings in patients with IVF with those in a sample of patients with CHD who had sustained an acute myocardial infarction or angina requiring angioplasty. A combination of interview-based measures and standardized questionnaires was used to assess the emotional climate in the hours, days, and months preceding the cardiac events. Measures were also obtained of depression, work stress and hostility, because these have been associated with cardiovascular disease in previous research (3,4).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Our original goal was to use each patient as his or her own control in a case–crossover methodology (5), so a modified case–crossover interview was administered. Unfortunately, the time elapsed between interview and event in some cases precluded the kind of detailed information needed to enable each subject to serve as his or her own control. We therefore needed a comparison group to ensure that the results observed in the patients with IVF were not an artifact of our interview methodology. For this purpose, we selected a comparison group of patients with CHD who had experienced an acute coronary event, given the reports that stress has been significantly associated with myocardial infarction (4).
All of the patients with CHD were recruited from the outpatient clinic in the Department of Cardiology at the University of Arizona in Tucson, Arizona, and were studied subsequent to the collection of data in patients with IVF. Patients with CHD were eligible if they had a documented myocardial infarction or had undergone angioplasty in the context of acute exacerbation of angina pectoris. Exclusion criteria for each group consisted of neurologic or cognitive impairment that precluded participation in the interview, completion of questionnaires, or recall of distant events. Identical procedures were followed for both groups. Information about the patient’s perception of stress that would enable us to compare those patients with CHD who participated versus those who did not is not available. The characteristics of patients who were studied are described in Table 1.
|
The vast majority of patients in each group were very pleased to participate to advance knowledge through this research. All participants understood clearly the need for truthfulness and for as much accuracy of recall as they were capable of providing. At the time of the interview, nearly all participants had a significant other or spouse with them to assist with recall and provide support. At the conclusion of the interview, many expressed their gratitude for the opportunity to participate.
The diagnosis of IVF is based on the clinical presentation of cardiac arrest accompanied by negative findings from a workup of possible causes of sudden death after successful resuscitation. The minimum evaluation required consists of a detailed clinical history, including information on drug intake, blood chemistry (cardiac enzymes, thyroid function tests, serum alcohol level, sedimentation rate, inflammatory indices, glucose, serum electrolytes, white blood cell count), 12-lead electrocardiogram (ECG), 24-hour ambulatory ECG, exercise stress test, coronary angiography, electrophysiological study including assessment of AV conduction, and ventricular biopsy in the presence of suggestive findings obtained in noninvasive tests. Minimal abnormalities that are compatible with the diagnosis of IVF include mitral valve prolapse (with lack of regurgitation, redundant valves, QT or ST-T abnormalities), modest regional dyskinesia, thickening (<10% above normal) of septum or left ventricular wall, paroxysmal or chronic atrial fibrillation, atrioventricular (AV) block (first or second degree), bundle branch block, age >60 years, hypertension (no hypertrophy), and nonspecific abnormalities at myocardial biopsy (11). Among 20 patients from the IVF-US Registry, there were eight patients with mitral valve prolapse (one also had AV nodal dysfunction and another also had right bundle branch block), three had minor wall motion abnormalities, and one had very mild left ventricular hypertrophy consistent with athlete’s heart. Among the five UCARE patients, two had mild left ventricular hypertrophy (one also had left bundle branch block) and one had a minor wall motion abnormality. Potentially silent heart diseases that may be present in patients with IVF include focal cardiomyopathy, focal myocarditis, focal fibrosis, local tumors/cysts, local biochemical/metabolic/neural alterations, transient electrolyte abnormalities, or silent ischemia (10). In addition, although there was no evidence on the 12-lead ECG of genetic abnormalities affecting depolarization or repolarization, it is possible that such abnormalities may have been present that were either unknown or undetected at the time of evaluation.
Patients were interviewed by telephone or in person about the circumstances surrounding their cardiac event. One person interviewed the UCARE patients and a second person interviewed the American patients with IVF and CHD comparison patients. Subjects were asked to recall the time of day and day of the week on which the cardiac event took place, their location and activity. They were asked about events in their lives both at work and at home in the 6 months before onset (chronic stress) and over the 24 hours before the cardiac event (acute stress). They were also asked questions that systematically surveyed the frequency with which they exerted themselves at different levels of intensity at specific time intervals (6 weeks, 1 week, 1 day, and 1 hour before the event) and the frequency with which they experienced different levels of anger, tension, and depression at these same specific time intervals. The interviewer wrote a summary for later qualitative scoring using notes taken during the interview. The case–crossover data were helpful in generating the qualitative interviewer summaries.
In addition, a series of standardized measures was verbally administered during the interview. Psychological stress during the 1 month before onset was assessed with the four-item Perceived Stress Scale, a standard measure that has been widely used in clinical and population research (17). Because recall of cardiac events might be affected by memory, we administered a word recall task. This consisted of 10 words that were read to the subject and then repeated from memory by the subject immediately, 1 minute later, and 5 minutes later. The score was the number correct at each time point. Other measures included a 14-item self-rated measure of work stress (18) and the 15-item ISEL measure of social support (19), which were both rated for the period preceding the cardiac event. Standard instructions were given for the 27-item Cook Medley measure of hostility (20). Depression and anxiety in the 3 months before the cardiac event were assessed using the Structured Clinical Interview for Diagnostic and Statistical Manual, Fourth Edition Axis I Disorders (21).
Interview summaries of the patients’ circumstances in the 6 months and 24 hours before the cardiac event were rated by two independent judges who were each experienced in stress research (R.D.L. and A.S.). Because of the temporal order of data collection in the two groups, the raters were not blind to patient category. A three-point scale was devised for both acute and chronic stress, in which 1 = none or mild stress: patients were carrying out their usual work and experienced only mild social hassles; 2 = moderate stress: heightened demands over the period that were being coped with effectively; and 3 = severe stress: unexpected threatening events or chronic stressors disruptive of everyday life and normal social functioning. Stress ratings were based only on the narrative interview summary; questionnaire and other objective measures were not taken into account in the stress ratings.
Discrepancies between the ratings of the two judges were resolved by discussion. Stress data were analyzed by combining the severe and moderate stress categories because we were interested in comparing the effect of significant stress with that of everyday circumstances (absent or mild stress). Data were analyzed using analysis of variance and nonparametric tests as appropriate.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Overall, 20 patients were rated as having none or mild stress on the chronic stress measure, with 12 rated moderate stress and 18 rated severe stress. The corresponding proportions for ratings of acute stress were 38 patients: none or mild stress; seven patients: moderate stress; and four patients: severe stress. Examples of stress at each level of severity in each of the two time contexts are presented in Table 2. The distribution of acute and chronic stress ratings in the two groups is shown in Table 3, along with the co-occurrence of acute and chronic stress. Patients with IVF were more likely to have experienced moderate or severe stress in the 6 months before their cardiac arrest than were the CHD group (Fisher exact p = .008). Fourteen (56%) of the patients with IVF experienced severe stress and six (24%) moderate stress, compared with four (16%) severe and six (24%) moderate stress in the CHD group. A difference was also obtained for the 24 hours before cardiac arrest, with nine (36%) patients with IVF compared with two (8%) patients with CHD experiencing moderate or severe stress (Fisher exact p = .04). The combined data showed a significant effect for trend (p = .003); only three (12%) patients with IVF experienced no stress over both time periods, and seven (28%) had moderate or severe stress in both time periods. In comparison, 14 (58%) patients with CHD were free of stress during the acute and chronic periods, and two (8%) experienced moderate or severe stress in both periods. Ratings on the Perceived Stress Scale for the month preceding the cardiac event were greater in the IVF than CHD groups (p < .01).
|
|
Psychiatric assessments with the SCID indicated that eight (five IVF, three CHD) patients (16%) had experienced mild or marked depression in the 3 months before their cardiac event, but this proportion did not differ between groups. In the years after the cardiac event, 10 (six IVF, four CHD) patients (20%) experienced mild and 11 (six IVF, five CHD) patients (22%) had marked depression, anxiety, or panic syndromes. There was no difference between IVF and CHD groups in the number of patients with psychiatric problems at any time point.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
We have referred to the patients with CHD as a "comparison group" rather than a "control group" because the latter would have implied that the CHD group differed from the IVF group only in the nature of the cardiac event, which was not the case. Rather, we studied a comparison group of patients with CHD to determine whether our method for assessing stress yielded results that were comparable to those of other patients with CHD in the literature. In fact, the proportion of patients with CHD who experienced moderate or severe stress in the 24 hours before their cardiac event was 8%, the same as the 8% prevalence of anger described in the Onset study (5) and similar to the 19% reported by Rissanen and colleagues (22). Although our patients with CHD were older than the patients with IVF, their mean age (57.1 years) was strikingly similar to the mean age (57.0 years) of the patients with CHD studied by Rissanen and colleagues (22). Similarly, although fewer of our patients with CHD were unmarried (37.1%) compared with the IVF group, our sample of patients with CHD was quite comparable in this respect to the sample of 896 patients with CHD (28.3% unmarried, mean age 59.4 years) studied by Lesperance and colleagues (7). Thus, our patients with CHD were quite comparable to those of other patients with CHD in the literature. The comparability of our findings in our comparison group to other similar studies in patients with CHD support the validity of our methods for evaluating the prevalence of psychosocial stress before cardiac events and therefore support our conclusion that psychosocial stress is often a precipitating factor in cardiac arrest in patients with IVF.
This study has a number of strengths. First, data were collected from a well-characterized group of patients with IVF who had survived cardiac arrest and had no evidence of structural heart disease such as CHD or left ventricular hypertrophy despite extensive investigation. Second, we compared the patients with IVF with a group of patients with CHD who had also experienced a life-threatening cardiac event. Because the prevalence of psychological stress preceding coronary events in patients with CHD has been examined in a number of previous studies, this represented a logical and informative comparison group. Third, we used a systematic interview-based method of evaluating stress before cardiac events. The technique of having independent judges rather than interviewers rate the stress experienced by patients is analogous to the contextual assessments used in life events research (23).
The main weakness of the study is that stress measures were reliant on retrospective recall by the patients themselves. Although the time elapsed between the cardiac event and the interview averaged just over 5 years, the two groups did not differ in the elapsed time. Furthermore, ratings of stress were based on judges’ ratings of life events rather than the patients’ ratings of those events. Nevertheless, it is possible that selective recall bias or memory loss may have been operating.
Survivors of acute cardiac events develop personal explanations for the causes of their problems, and stress is frequently cited as a factor in CHD (4). Still others may deny that stress was playing a role when independent observers such as spouses may believe otherwise. In the case of IVF, a causal explanation for the event cannot be provided by physicians, creating an explanatory vacuum that could potentially be filled by attributions of psychologic stress. The question therefore arises whether the patients with IVF had a greater bias toward stress as a causal explanation than had patients with CHD. If patients with IVF were more likely to make psychologic causal attributions, then differences in depression levels before or after the cardiac event might have been expected, but these were not present. Unfortunately, a more detailed assessment of the psychological response to the cardiac event was beyond the scope of this study. In the absence of such data, given the lack of group differences in depression and the time elapsed between the event and the interview, there is no solid basis for hypothesizing that the psychological sequelae of the cardiac event influenced the reporting characteristics of one group more than the other.
Another possibility is that patients with IVF, who were younger, may have been better able to recall emotionally salient events because their ability to recall past personal experiences may have been superior to those of patients with CHD. Our formal test of memory does not speak to this point directly because it involved short-term memory. This is different from the autobiographic memory tapped in the interview because of the time elapsed between encoding and retrieval and the absence of emotional content. In fact, emotional content enhances recall of personal experiences, and the literature is ambiguous as to whether emotion-laden memories are recalled less accurately in older than younger subjects (24). Research on "flashbulb memories" (shocking events such as September 11, 2001, recorded as if captured by a camera flashbulb), which may be most comparable to the life-threatening circumstances under investigation in this study, have produced mixed results, with some studies showing greater clarity in older than younger adults (25) and greater-than-expected accuracy in older adults (26). Based on these findings, one cannot conclude that our results are biased as a result of selective recall bias or superior recall in patients with IVF.
A second limitation of this study is that the groups differed in marital status as well as age. However, when variance resulting from age and marital status are removed, the significant between-group difference on the Perceived Stress Scale persists, consistent with the conclusions that the degree of stress differed between the groups and that memory in both groups was within normal limits. Furthermore, as noted previously, our sample of patients with CHD was comparable to other CHD samples in the literature with regard to age and marital status. A third limitation is that the study was confined to survivors of IVF and myocardial infarction, so we do not know whether similar factors might have been implicated in fatal episodes. A fourth limitation is the relatively small size of the two patient groups. We believe that the size of our sample of patients with IVF is sufficiently large that we can be confident in our main conclusion, which is that psychosocial stress plays a precipitating role in otherwise unexplained cardiac arrest. Although the between-group differences in stress were reliable, they should be considered preliminary given the sample sizes. Finally, we did not attempt to dissect out more subtle emotional responses, so we could not assess the specific associations between cardiac events and anger described in the literature. Future studies would also benefit from systematic information about the characteristics of eligible patients who did and did not participate, and stress ratings by judges who were blind to patient diagnosis.
The mechanisms underlying the associations between psychological stress and cardiac arrest in IVF are likely to involve autonomic and neurohumoral stimulation of ventricular arrhythmias (12,27). There is substantial evidence that psychologic stress can stimulate ventricular tachycardia in vulnerable individuals (27), including those with no structural heart disease (28). A recent study of emotional triggers of ventricular arrhythmia involved repeated measures of moods in patients fitted with ICDs (6). A proportion experienced confirmed ventricular arrhythmias requiring shock, and it was found that high levels of anger were more frequent in the 15 minutes before shock than in control periods.
The definition of IVF is continually evolving, and concealed arrhythmogenic disorders of genetic origin may underlie some cases (11). Acute psychological stress is also associated with high levels of circulating catecholamines (27), so this may have played a role in some cases of IVF. Chronic stress may be associated with decreases in vagal tone (27), which in itself is an established risk factor for cardiac mortality (29). Cardiac vulnerability to arrhythmogenic stimuli in animal models is mediated through frontocortical (30) and/or amygdala (31) projections to the brain stem, and stress-induced autonomic arousal in humans is associated with increased activity in several regions of the brain, including the cerebellar vermis, brain stem, and right anterior cingulate gyrus (32).
In Walter Cannon’s famous study of "voodoo death," sudden death occurred within 24 to 48 hours of a curse being inflicted on a transgressing tribesman. Cannon hypothesized that the physiological mechanism of death consisted of massive sympathetic activation associated with extreme fear (33). Autopsies were not performed to determine whether those victims had preexisting heart disease. In the current study, we cannot exclude the possibility that a substrate is present, but one can conclude that because structural heart disease that could cause sudden death is absent, the severity of the substrate is likely less than that of CHD. The findings from this study are therefore consistent with the principle that the severity of the preexisting substrate and the intensity of the trigger are inversely related. However, because we did not quantify the extent of the substrate in either group, and because the outcomes in the two groups differed, the findings from this study do not directly address the relationship between substrate and trigger.
To address this question, a more extensive study would be needed. A more definitive test of the inverse substrate-trigger hypothesis would be to compare psychosocial stress preceding IVF with that of patients with CHD who survived cardiac arrest and to quantify the extent of the substrate in each group. Inclusion of patients with CHD with a coronary event but without cardiac arrest as well as a matched group of healthy volunteers, based on a 2 (versus no CHD) x 2 (cardiac arrest versus no cardiac arrest) design, would also permit specific conclusions about the psychologic characteristics of each group, an estimate of the probability of different levels of stress in the general population, and a disentangling of the contribution of stress to ischemia-related events as opposed to arrhythmic events and sudden cardiac death. In our view, the results of the present investigation suggest that an expanded study such as this would be worthwhile.
We thank Dr. Gordon Ewy for referral of his patients with CHD for this study.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This project was funded by grants from the Fetzer Institute, Kalamazoo, Michigan; the Flinn Foundation, Phoenix, Arizona; Medtronic, Inc., Minneapolis, Minnesota; Ventrotex, Inc., Heartstream, Inc.; and CEE BMH1-CT92–Unexplained Cardiac Arrest Registry of Europe.
DOI:10.1097/01.psy.0000160476.67536.41
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. C. Ziegelstein Acute Emotional Stress and Cardiac Arrhythmias JAMA, July 18, 2007; 298(3): 324 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Gray, P. Taggart, P. M. Sutton, D. Groves, D. R. Holdright, D. Bradbury, D. Brull, and H. D. Critchley A cortical potential reflecting cardiac function PNAS, April 17, 2007; 104(16): 6818 - 6823. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
