Mental Stress–Induced Ischemia and All-Cause Mortality in Patients With Coronary Artery Disease
Results From the Psychophysiological Investigations of Myocardial Ischemia Study
Background— Ischemia during laboratory mental stress tests has been linked to significantly higher rates of adverse cardiac events. Previous studies have not been designed to detect differences in mortality rates.
Methods and Results— To determine whether mental stress–induced ischemia predicts death, we evaluated 196 patients from the Psychophysiological Investigations of Myocardial Ischemia (PIMI) study who had documented coronary artery disease and exercise-induced ischemia. Participants underwent bicycle exercise and psychological stress testing with radionuclide imaging. Cardiac function data and psychological test results were collected. Vital status was ascertained by telephone and by querying Social Security records 3.5±0.4 years and 5.2±0.4 years later. Of the 17 participants who had died, new or worsened wall motion abnormalities during the speech test were present in 40% compared with 19% of survivors (P=0.04) and significantly predicted death (rate ratio=3.0; 95% CI, 1.04 to 8.36; P=0.04). Ejection fraction changes during the speech test were similar in patients who died and in survivors (P=0.9) and did not predict death even after adjusting for resting ejection fraction (P=0.63), which was similar in both groups (mean, 56.4 versus 59.7; P=0.24). Other indicators of ischemia during the speech test (ST-segment depression, chest pain) did not predict death, nor did psychological traits, hemodynamic responses to the speech test, or markers of the presence and severity of ischemia during daily life and exercise.
Conclusions— In patients with coronary artery disease and exercise-induced ischemia, the presence of mental stress–induced ischemia predicts subsequent death.
Received November 9, 2001; revision received February 4, 2002; accepted February 6, 2002.
Epidemiological research has shown that mental stress is associated with adverse events in patients with coronary artery disease.1–3⇓⇓ States of chronic mental stress characterized by depression, lack of social support, or hostility also are tied to an increased event rate in this population.4–7⇓⇓⇓ Studies have shown that mental stress induced in the laboratory triggers ischemia in 40% to 70% of stable coronary patients who have positive exercise tests.8–10⇓⇓ Other investigations have shown that the development of ischemia in response to mental stress in the laboratory is independently associated with significantly higher rates of fatal and nonfatal cardiac events and predicts events over and above exercise-induced ischemia.11–13⇓⇓ However, these studies have used composite outcomes and have not been powered to detect differences in mortality rates. Composite outcomes have included recurrent myocardial infarction, need for angioplasty or bypass surgery, and hospitalization for unstable angina, in addition to death. The Psychophysiological Investigations of Myocardial Ischemia (PIMI) study was designed to investigate psychological characteristics influencing the manifestations and expression of myocardial ischemia in patients with coronary artery disease. We examined the data collected to determine whether mental stress–induced ischemia predicts an increased risk of death.
The PIMI protocol has been described in detail elsewhere by Kaufmann et al14; essential features are summarized below. The institutional review board at each participating institution approved the PIMI protocol and informed-consent procedures.
At 4 participating PIMI sites, we studied 196 patients with coronary artery disease (170 men and 26 women) who met eligibility criteria, including the following: >50% narrowing in at least 1 major coronary artery or verified myocardial infarction, evidence of myocardial ischemia (ST-segment depression ≥0.1 mV on an exercise treadmill test conducted off anti-ischemic medications), and informed consent to participate. Patients were excluded if they had a serious noncardiac illness, unstable angina, or neurological disease; the inability to discontinue medications that might influence cardiac function; or if they had previously undergone PTCA or thoracic surgery. Eligibility was established during a qualifying visit, at which time a standard exercise treadmill test was administered. After no more than 3 months, mental and exercise stress tests were conducted in the morning, after fasting, on 2 separate days occurring within 2 weeks. Cardiac function data, psychological tests, and serum samples were collected at specified times before, during, or after stress tests.
Two different physical stress tests were administered during the study: a standard exercise treadmill test at the qualifying visit and, at a later visit, a maximal bicycle exercise tolerance test with radionuclide imaging. The sequence was designed to minimize distractions and discomfort from extraneous procedures that might influence the results of neuroendocrine assays or other measures. The patient received a light breakfast and completed the Beck Depression Inventory15 and the Spielberger State Anxiety Scale.16 After a 30-minute rest period, blood was drawn for determination of resting levels of epinephrine and norepinephrine and for other assays. Baseline radionuclide images were then obtained, after which the bicycle exercise test was performed with workload increasing in 3-minute stages and images recorded during the last 2 minutes of each stage. Radionuclide images were read in a central laboratory with semiautomated assessment of ejection fraction and qualitative assessment of wall motion changes by experienced raters blinded to experimental condition. ECGs and ambulatory ECGs recorded during stress tests also were read at central laboratories.
The psychological stress day was designed to follow a pattern similar to that of the physical stress day, except that two 5-minute mental stress tests (Stroop and speech) were substituted for the bicycle exercise test. The sequence of the 2 stress tests was randomly determined and instructions were provided from a standard script read by staff. The results provided in this article relating to psychological stress testing were based on the speech stressor because this proved to be the more potent stimulus for myocardial ischemia.17 For the speech task, patients were asked to speak for 5 minutes on an assigned topic while being observed and critically evaluated by laboratory staff. The topic required role-playing relating to a difficult scenario in which a close relative was being mistreated in a nursing home.
The initial design of the PIMI study did not include follow-up to ascertain cardiac events. Several years after the baseline PIMI studies concluded (range, 2.6 to 4.3 years), in a test of the feasibility of ascertaining events, we contacted 178 of 196 patients or relatives by telephone to determine study participants’ vital status. During the interview, verbal consent was obtained for participation in the follow-up study. Consenting patients were sent a written informed consent form to sign and return. If the patient had died, the date of death was obtained from their family members. Several years later (range, 4.3 to 6.0 years after the baseline study), Social Security death records were queried for all 196 patients originally enrolled to identify additional fatalities as of December 31, 1998 (average follow-up for death, 5.2±0.4 years).
Characteristics of patients with and without wall motion abnormalities during the speech test were compared by means of χ2 tests for categorical data and Wilcoxon rank tests for continuous data. The Cox proportional hazards model18 was used to test the association of death with patient characteristics and stress test results. The Kaplan-Meier method19 was used to estimate the distribution of time to death in patients with and without wall motion abnormalities during the speech test. Because of the limited number of reported deaths (n=17), extensive multivariate modeling of survival could not be performed. Instead, the association between death and presence of speech wall motion abnormalities was examined, controlling for one potential confounding variable at a time, with the use of the Cox model. Because the numbers of patients with data available varied from test to test as the result of technical problems or patient nonresponse, numbers (n) in individual tables vary slightly, depending on the measures reported. For each analysis relating a baseline variable or set of variables to death, all patients with data available for the particular measures considered were used.
Wall motion abnormalities were observed during the speech test in 37 (20%) of 182 patients for whom this outcome was ascertained. Compared with patients who had normal wall motion responses during the speech test, patients with wall motion abnormalities were more often female (24% versus 12%, P=0.05) and more often had a history of diabetes (27% versus 12%, P=0.03) but did not differ in other clinical or psychological characteristics or in ischemic responses to exercise or daily life (Table 1).
Patients with wall motion abnormalities during the speech test had higher mortality rates (rate ratio=2.79; 95% CI, 1.00 to 7.85; P=0.04) during follow-up than did patients without such abnormalities (Table 2 and Figure). Other indicators of ischemia during the speech test (ejection fraction response, ST-segment depression) did not predict death. No significant association was observed between death and hemodynamic or catecholamine responses to the speech test. No association was observed between mortality rates among PIMI patients and psychological characteristics (depression, anxiety, or anger) that have been reported to predict cardiac events (Table 3). Markers of the presence and severity of ischemia during daily life and exercise stress testing were not significantly associated with death (Table 4). As noted above, wall motion abnormalities were associated with a history of diabetes, but even in nondiabetics, wall motion abnormalities remained associated with death (rate ratio=3.7; 95% CI=1.2 to 11.8; P=0.02).
Of the 17 deaths noted, all were in men; none of the 26 women enrolled died (P=0.08). Deaths were slightly more common among patients with a history of myocardial infarction, hypertension, or diabetes, but the differences in mortality rates compared with patients without these conditions were not statistically significant (data not shown).
In Table 5, the association between death and presence of new or worsened wall motion abnormality during the speech test is examined with and without adjusting for history of myocardial infarction, hypertension, or diabetes; duration of treadmill exercise; and resting ejection fraction (average of preexercise and premental stress baseline measures). Because of the limited number of deaths, only one of these risk factors at a time was used to adjust the estimated mortality rate ratio for presence of speech wall motion abnormalities. Adjustment for these characteristics had little effect on the estimated association between death and speech-induced wall motion abnormalities. Adjustment for history of myocardial infarction had the largest effect on the rate ratio, reducing it from 2.79 (95% CI, 1.00 to 7.85; P=0.04) unadjusted to 2.59 (95% CI, 0.92 to 7.33; P=0.07) adjusted.
The major finding of this study is that mental stress–induced ischemia, as detected by new or worsened wall motion abnormality during a laboratory-induced psychological stressor, is associated with close to a 3-fold relative increase in mortality rate in patients with coronary artery disease and exercise-induced ischemia when controlling for resting ventricular function. This finding is in agreement with previous work published by Jain et al,11 Jiang et al,12 and Krantz et al13 showing a >2-fold risk of subsequent cardiac events in patients with mental stress–induced ischemia in the laboratory. The previous studies, however, were not powered to detect differences in mortality rates.
It is also noteworthy that other demographic and clinical variables did not distinguish patients who died from those who survived. These included psychological variables, exercise test variables, and hemodynamic changes during exercise and psychological stress. Thus, the predictive value of mental stress testing for subsequent cardiac death cannot be attributed to differences in standard risk factors or hemodynamic changes during the stressor. Similarly, there was no difference in the cumulative duration of ischemia during ambulatory ECG monitoring in those who died versus those who lived. PIMI participants were selected to be clinically stable and to have ECG evidence of ischemia on a treadmill test, which may account for the lack of power of other exercise testing outcomes to predict death. This suggests that mental stress testing may provide added prognostic information beyond that given by a standard exercise test and is consistent with previous data.20
Because patients who died were not identified by differences in exercise response or ambulatory ischemic responses, mental stress–induced ischemia most likely conveyed high risk because of one of several other possible mechanisms. Mental stress elevates systolic blood pressure to a degree similar to that produced during exercise. However, the rate-pressure product achieved during mental stress testing is usually less than that achieved during maximal exercise testing because heart rate increases are generally less with mental stress than with exercise.20 Mental stress also usually produces higher epinephrine responses than exercise does,17,21⇓ and the change in blood pressure occurs more rapidly, which may predispose to plaque rupture in certain situations. Previous studies have shown that large blood pressure responses to mental stress predict subsequent adverse cardiovascular events among coronary patients.22,23⇓ Our data did not support a difference in blood pressure response as being predictive of subsequent cardiac events or death. The lack of a relation between hemodynamic responses during mental stress or exercise hemodynamic or ischemic responses and subsequent cardiac outcome also was found by Jiang et al.12 However, previous work published by our group has shown a good correlation between increased systemic vascular resistance and mental stress–induced myocardial ischemia.17
It has been shown previously that mental stress also may reduce myocardial oxygen supply.24 At present, it is unknown whether increases in coronary vascular resistance, either involving epicardial vessels or smaller vessels in the coronary microcirculation, correlate with changes in systemic vascular resistance.25,26⇓ If so, this association could account for the effects of mental stress on myocardial oxygen supply.
Our conclusions are limited by several considerations. There were relatively few deaths, and none occurred in the 26 women in this study despite the higher prevalence of mental stress–induced ischemia in women versus men. Thus, the results reported here understate the increased risk of death in men with mental stress–induced ischemia versus those without, and we could not test whether these findings apply to women as well as men. In addition, because the study was not prospectively designed as a follow-up study, we did not collect complete data on nonfatal events and therefore cannot know whether mental stress–induced ischemia also predicts nonfatal cardiovascular events. Moreover, we did not have resources to classify deaths as to cause, so we analyzed associations with total mortality rather than cardiac death. If mental stress ischemia is associated with cardiac but not noncardiac death, as seems reasonable to speculate, this association will be stronger than is shown in our data. The inability of exercise parameters to predict death or events probably is related to the low-risk characteristics of the population14 and restriction of range due to the requirement of a positive exercise test for entry to the study.
In summary, in our study, patients who had ischemia during mental stress had an increased mortality rate compared with those who did not, although the mechanism by which mental stress response increases risk of death is still unclear. Future research should focus on reproducing this finding in other populations, elucidating the mechanisms of mental stress–induced ischemia, and searching for an inexpensive surrogate for myocardial imaging, which would make mental stress testing more attractive for routine clinical use. In addition, a better understanding of the mechanisms responsible for mental stress–induced ischemia could lead to therapeutic interventions to modify these responses.
This study was supported by the National Heart, Lung, and Blood Institute, Bethesda, Md, through contract numbers HV18114, HV18119, HV18120, HV18121, and HV28127. This material also is the result of work supported with resources and the use of facilities at the Malcom Randall Veterans Affairs Medical Center, Gainesville, Fla. The authors thank Melanie Fridl Ross, MSJ, ELS, for editorial assistance.
- ↵Meisel SR, Kutz I, Dayan KI, et al. Effects of Iraqi missile war on incidence of acute myocardial infarction and sudden death in Israeli civilians. Lancet. 1991; 338: 660–661.
- ↵Kaprio J, Koskenvuo M, Rita H. Mortality after bereavement: a prospective study of 95,647 widowed persons. Am J Public Health. 1987; 77: 283–287.
- ↵Leor J, Poole WK, Kloner RA. Sudden cardiac death triggered by an earthquake. N Engl J Med. 1996; 334: 413–419.
- ↵Anda R, Williamson D, Jones D, et al. Depressed affect, hopelessness, and the risk of ischemic heart disease in a cohort of US adults. Epidemiology. 1993; 4: 285–294.
- ↵Frasure-Smith N, Lespérance F, Talajic M. Depression and 18-month prognosis after myocardial infarction. Circulation. 1995; 91: 999–1005.
- ↵Everson SA, Goldberg DE, Kaplan GA, et al. Hopelessness and risk of mortality and incidence of myocardial infarction and cancer. Psychosom Med. 1996; 58: 113–121.
- ↵Barefoot JC, Schroll M. Symptoms of depression, acute myocardial infarction, and total mortality in a community sample. Circulation. 1996; 93: 1976–1980.
- ↵Burg MM, Jain D, Soufer R, et al. Role of behavioral and psychological factors in mental stress-induced silent left ventricular dysfunction in coronary artery disease. J Am Coll Cardiol. 1993; 22: 440–448.
- ↵Rozanski A, Bairey CN, Krantz DS, et al. Mental stress and the induction of silent myocardial ischemia in patients with coronary artery disease. N Engl J Med. 1988; 318: 1005–1012.
- ↵Blumenthal JA, Jiang W, Waugh RA, et al. Mental stress–induced ischemia in the laboratory and ambulatory ischemia during daily life: association and hemodynamic features. Circulation. 1995; 92: 2102–2108.
- ↵Jain D, Burg M, Soufer R, et al. Prognostic implications of mental stress-induced silent left ventricular dysfunction in patients with stable angina pectoris. Am J Cardiol. 1995; 76: 31–35.
- ↵Jiang W, Babyak M, Krantz DS, et al. Mental stress-induced myocardial ischemia and cardiac events. JAMA. 1996; 275: 1651–1656.
- ↵Krantz DS, Santiago HT, Kop WJ, et al. Prognostic value of mental stress testing in coronary artery disease. Am J Cardiol. 1999; 84: 1292–1297.
- ↵Kaufmann PG, McMahon RP, Becker LC, et al. The Psychophysiological Investigations of Myocardial Ischemia (PIMI) study: objective, methods, and variability of measures. Psychosom Med. 1998; 60: 56–63.
- ↵Beck AT, Steer RA, Garbin MG. Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clin Psychol Rev. 1988; 8: 77–100.
- ↵Spielberger CD. State-Trait Anxiety Inventory. Palo Alto, Calif: Consulting Psychologists Press; 1991.
- ↵Goldberg AD, Becker LC, Bonsall R, et al, for the PIMI Investigators. Ischemic, hemodynamic, and neurohormonal responses to mental and exercise stress: experience from the Psychophysiological Investigations of Myocardial Ischemia Study (PIMI). Circulation. 1996; 94: 2402–2409.
- ↵Cox DR. Regression methods and lifetables (with discussion). J R Stat Soc Series B. 1972; 34: 187–220.
- ↵Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958; 53: 457–481.
- ↵Rozanski A, Blumenthal JA, Kaplan T. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation. 1999; 16: 2192–2217.
- ↵Dimsdale JE, Moss J. Plasma catecholamines in stress and exercise. JAMA. 1980; 243: 340–342.
- ↵Keys A, Taylor HL, Blackburn H, et al. Mortality and coronary heart disease among men studied for 23 years. Arch Intern Med. 1971; 128: 201–214.
- ↵Manuck SB, Olsson G, Hjemdahl P, et al. Does cardiovascular reactivity to mental stress have prognostic value in postinfarction patients? A pilot study. Psychosom Med. 1992; 54: 102–108.
- ↵Yeung AC, Vekshtein VI, Krantz DS, et al. The effect of atherosclerosis on the vasomotor response of coronary arteries to mental stress. N Engl J Med. 1991; 325: 1551–1556.
- ↵Boltwood MD, Taylor CB, Burke MB, et al. Anger report predicts coronary artery vasomotor response to mental stress in atherosclerotic segments. Am J Cardiol. 1993; 72: 1361–1365.
- ↵Dakak N, Quyyumi AA, Eisenhofer G, et al. Sympathetically mediated effects of mental stress on the cardiac microcirculation of patients with coronary artery disease. Am J Cardiol. 1995; 76: 125–130.