| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Medicine and Psychiatry and Behavioral Sciences, Stanford University, Stanford, California (C.B.T., A.C., E.N., A.D., J.G.-D., W.T.R., R.O., J.C., H.K., D.S.); the Department of Veterans Affairs Health Care System, Palo Alto, California (A.C., W.T.R.); the Institute for Psychology, Department of Clinical Psychology and Psychotherapy, University of Basel, Basel, Switzerland (F.H.W.); and the Department of Psychology, University of Würzburg, Würzburg, Germany (M.A.K.).
Address correspondence and reprint requests to C. Barr Taylor, MD, Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, 401 Quarry Rd., Room 1316, Stanford, CA 94305-5722. E-mail: btaylor{at}stanford.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: Forty-eight depressed participants and 20 controls with elevated cardiovascular risk factors underwent a psychological stress test during which cardiovascular variables were measured. Salivary cortisol was collected after each test segment. Traditional (e.g., lipids) and atypical (e.g., C-reactive protein) CVD risk factors were also obtained.
Results: At baseline, the groups did not differ on lipid levels, flow-mediated vasodilation, body mass index, or asymmetric dimethylarginine. However, the depressed patients had significantly higher C-reactive protein levels. Contrary to our hypothesis, there were no differences in baseline cortisol levels or diurnal cortisol slopes, but depressed patients showed significantly lower cortisol levels during the stress test (p = .03) and less cortisol response to stress. Compared with nondepressed subjects, depressed subjects also showed lower levels of respiratory sinus arrhythmia (RSATF) during the stress test (p = .02).
Conclusions: In this sample, older depressed subjects with elevated risk for CVD exhibited a hypocortisol response to acute stress. This impaired cortisol response might contribute to chronic inflammation (as reflected in the elevated C-reactive proteins in depressed patients) and in other ways increase CVD risk. The reduced RSATF activity may also increase CVD risk in depressed patients through impaired autonomic nervous system response to cardiophysiological demands.
Key Words: depression • cardiovascular risk • psychophysiology • cortisol
Abbreviations: ACTH = adrenocorticotropic hormone; ADH = antidiuretic hormone; ADMA = asymmetric dimethylarginine; ANS = autonomic nervous system; BMI = body mass index; BP = blood pressure; BRC = baroreflex control; CAD = coronary artery disease; CBT = cognitive behavioral therapy; CHD = coronary heart disease; CO = cardiac output; CON = nondepressed control; CPM = cycles per minute; CVD = cardiovascular disease; DBP = diastolic blood pressure; DISH = Depression Interview and Structured Hamilton; ECG = electrocardiogram; FMVD = flow-mediated vasodilation; HDL = high-density lipoprotein; HPA = hypothalamic–pituitary–adrenal axis; HR = heart rate; HRSD = Hamilton Rating Scale of Depression; HRV = heart rate variability; LDL = low-density lipoprotein; MDD = major depressive disorder; MI = myocardial infarction; NO = nitric oxide; PANAS = Positive and Negative Affect Schedule; pCO2 = partial pressure of carbon dioxide; PEP = preejection period; PSS = Perceived Stress Scale; RSA = respiratory sinus arrhythmia; RSATF = transfer function respiratory sinus arrhythmia; SBP = systolic blood pressure; SVR = systemic vascular resistance; TSST = Trier Social Stress Test; VLDL = very-low-density lipoprotein.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
A number of mechanisms have been proposed to explain how depression might increase CVD risk, including abnormalities in hypothalamic–pituitary–adrenal axis (HPA) function, increased inflammation, alterations of sympathetic and parasympathetic activity, increasing clotting/thrombus formation, increased traditional (e.g., blood pressure) and nontraditional (e.g., C-reactive protein) risk factors, singly or in combination, reflecting the multifactorial nature of atherosclerosis development (4,5).
In terms of HPA axis function, hypercortisolism has been reported in many studies of depressed patients (6). Abnormal HPA axis function may contribute to CVD risk through a variety of risk factors, including elevated blood pressure (BP), high lipid levels, insulin resistance, and abdominal obesity. In addition to long-term effects of such factors, inadequate HPA response to an acute MI may hamper recovery (7). Surprisingly, very few studies have compared HPA response with mental stress in depressed and nondepressed subjects. In these studies, depressed individuals tend to respond to stress with a lower cortisol response than nondepressed participants (8).
There is substantial evidence that inflammation plays an important role in atherogenesis (9,10). Clinical depression is associated with marked increases in systemic inflammation, as evidenced by elevations in circulating concentrations of C- reactive protein (11) and other inflammatory markers (12,13). However, not all studies have found a consistent relationship with C-reactive protein and CVD risk (14).
The argument for focusing on autonomic nervous system dysregulation as a possible explanation for the higher morbidity and mortality in depressed patients begins with observations that patients with greater CVD reactivity and dysregulation are at heightened risk for developing CVD and for accelerated disease progression (15,16). A number of studies have been undertaken to look for autonomic nervous system differences in depressed and nondepressed patients with and without manifest CVD. Depression has effects on the autonomic control of the cardiovascular system. Respiratory sinus arrhythmia (RSA), which is mainly controlled by parasympathetic activity (17), is one measure of this regulation. A high degree of RSA is observed in normal hearts with good cardiac function, whereas RSA can be significantly decreased in patients with severe coronary artery disease (CAD) or heart failure. The relative risk of sudden death after acute MI is significantly higher in patients with decreased RSA (18). Some studies have found an association between RSA and depression (19–21); others have not (22,23). The inconsistencies among studies might be partly related to the method of estimating RSA. Newer computation methods have been developed to adjust RSA for potential respiratory confounds to reduce error variance or systematic bias and increase its accuracy as a vagal index in within- and between-individual comparisons (24–26). Other measures of heart rate variability (HRV) may also be important (27).
Baroflex control (BRC) is another measure of sympathovagal cardiovascular regulation. The baroreflex results in heart rate (HR) slowing during transient BP increases. In patients after MI, reduced BRC increases the risk of sudden cardiac death (28). Lower control is associated with exaggerated BP and HR responses to psychological stressors. Depressive symptoms were associated with lower BRC in 60 stable patients with CAD (29).
In addition, depression, perhaps through increased sympathetic drive (30), might affect traditional risk factors, and in particular, systolic and diastolic blood pressure. However, the evidence for increase in such risk factors in patients with CVD is inconsistent. For instance, in a large longitudinal study, Gump et al. (31) found no difference in systolic blood pressure (SBP), diastolic blood pressure (DBP), or serum cholesterol across quintiles of Center for Epidemiologic Studies Depression Scale (CES-D) scores, a self-report measure of depressed mood.
Depression may affect some of the newer risk factors of interest such as C-reactive protein or endothelial function, which is a potential common pathway by which all risk factors may increase CVD risk (32). Endothelial function is strongly affected by nitric oxide (NO) activity. In addition to its vasodilator activity, NO inhibits key processes involved in vascular disease, including leukocyte adhesion, platelet aggregation, and vascular smooth muscle cell proliferation. Sherwood et al. (33) recently found higher levels of endothelial dysfunction associated with higher symptoms of depression. In animal models, alterations in vascular NO synthesis profoundly influence the progression of atherosclerosis and restenosis and impairment in the nitric oxide synthase (NOS) pathway and independently predicts cardiovascular events. Endothelial function is mediated by an endogenous inhibitor of NOS, a molecule known as asymmetric dimethylarginine (ADMA) (32). ADMA is an endogenous competitive inhibitor of NOS. Elevated ADMA plasma levels have been reported in connection with diseases associated with an impaired endothelial L-arginine–NO pathway. ADMA has been found to be elevated in depressed, as compared with nondepressed, healthy, middle-aged individuals (34).
Changes in many of the systems mentioned here interact or have multiple effects, and small changes across a variety of systems could increase risk even when the change within an individual system is minor (35). For instance, vagus nerve activity, which not only helps regulate heart rate, but also inhibits the production of proinflammatory cytokines from macrophages (36). Lower-level parasympathetic activity might affect both healthy response to changes in cardiovascular hemodynamics and may also fail to inhibit inflammatory response.
The purpose of this study was to examine atypical risk factors and compare cardiovascular, respiratory, and cortisol reactivity during psychological stress testing in depressed and nondepressed subjects at elevated risk for CVD but without manifest disease. Our primary hypotheses were that depressed subjects would show evidence of hypercortisolism, decreased parasympathetic activity, and no difference in sympathetic activity. Our secondary hypotheses were exploratory. We predicted that depressed subjects would have lower levels of BRC, higher levels of inflammation, as reflected in C-reactive protein, higher levels of ADMA, and greater endothelial dysfunction as reflected in lower levels of flow-mediated vasodilation (FMVD).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Measures
Sociodemographic and Medical Variables
Sociodemographics
Baseline demographics included education, ethnicity, and marital status.
Psychiatric Diagnosis
A structured clinical interview focusing on depression and modified to include the Hamilton Depression Interview was used (DISH) (37). Prospective interviewers were given extensive training and each DISH was reviewed by a senior psychologist or psychiatrist.
Medical Status
A standard questionnaire assessing coronary risk factors, medications, and medical diagnoses was completed at baseline by all subjects. Subjects also underwent a physical examination by a cardiologist who reviewed and confirmed diagnoses. A resting 12-lead electrocardiogram (ECG) was recorded. Subjects with CVD ischemic events or arrhythmias evident in their ECG were excluded. Patients were diagnosed as having hypertension and/or hypercholesterolemia based on history and current medications.
Medications
Patients were asked to bring all of their medications, which were then reviewed by the project physician and research assistant. Medications were grouped into one of 28 drug classes (e.g., lipid-lowering, antihypertensive) based on pharmacologic action and potential effect on the outcome measures.
Blood Pressure
Baseline resting BP was measured in triplicate by auscultatory techniques using a mercury sphygmomanometer. The first reading was discarded and the latter two readings were averaged. The dominant arm was used.
Body weight (in kilograms) and height (in meters) were obtained in a standing position with shoes removed and subjects wearing street clothes. These were used to calculate body mass index (BMI; in kg/m2).
C-reactive protein was measured using a high-sensitivity assay (Dade, Behring, Marburg, Germany). (Some subjects were initially run with an older immunoassay so these data were not included.) The final sample for the hsCRP assay was run on 81% (35 of 43) of depressed subjects and 70% (14 of 20) control subjects.
Lipids
Fasting plasma levels of total cholesterol and triglycerides were measured using standard enzymatic procedures. High-density lipoprotein (HDL) cholesterol was measured by dextran sulfate–magnesium precipitation followed by enzymatic measurement of the nonprecipitated cholesterol. Low-density lipoprotein (LDL) cholesterol was calculated as total cholesterol minus the sum of HDL cholesterol plus very-low-density lipoprotein (VLDL) cholesterol.
Asymmetric Dimethylarginine
ADMA was measured by immunoassay (38).
Cortisol
At baseline, each participant was scheduled to obtain saliva samples using cotton swabs in "salivette devices" at the time of waking, 30 minutes later, and then at 12:00 PM, 5:00 PM, and 9:00 PM on each of 2 baseline saliva collection days. Participants were asked to refrigerate each sample immediately after collection; not to eat, drink, smoke, brush their teeth, or use mouthwash in the 30 minutes before collection; and not to drink alcohol during the 8 to 10 hours before collecting samples or during the days of collection. Saliva samples were obtained 10 times during the stress task. Salivary cortisols were assayed using luminescence immunoassay (LIA) reagents provided by Immuno-Biological Laboratories, Inc. (Hamburg, Germany). Assay sensitivity was 0.015 µg/dL.
Flow-Mediated Vasodilation
Participants were instructed to report to the hospital between 9:00 AM and 12:00 PM after a 12-hour fast with medication withheld. Vascular function was assessed by measurement of flow as well as nitroglycerin-mediated vasodilation. FMVD of the brachial artery was measured using a Siemens Acuson Sequoia C256 high-resolution ultrasound machine with a 14-MHz probe and established techniques (39). Measurements of brachial artery diameter were performed 30, 45, and 60 seconds after cuff deflation. Arterial diameter was measured using electronic calipers. FMVD is expressed as percent change in vessel diameter from rest to postreactive hyperemia. After the brachial artery diameter had returned to baseline measurements, 0.6 mg nitroglycerin was administered and the brachial artery diameter was recorded for 5 minutes.
Psychosocial Variables
Depression
Depression was measured by self-report using the Beck Depression Inventory and by interview with the Hamilton Depression Inventory obtained during the DISH (37).
State Emotions
State emotions were measured with the Positive and Negative Affect Schedule (PANAS), an adjective list of emotions with a Likert-type scale (1 = slightly to 5 = extremely) and the instruction to rate feelings in the moment. It has 20 items that yield a negative and positive affect score. The PANAS was administered when the subjects first arrived for stress testing and after each stress task.
Perceived Stress Scale
The PSS measures global perception of stress during the previous month. It has a short version comprised of 10 items, e.g., "In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?" Response options were assessed using a 5-point Likert-type scale (0 = never to 4 = very often).
Stress Task
A modified version of the Trier Social Stress Test (TSST) (40) was used in this study. The TSST is a standardized social and cognitive stressor composed of 5 minutes of anticipatory stress and then 5 minutes of public speaking (simulated job interview) and 5 minutes of mental arithmetic, both done before a panel of two evaluators. Subjects were sitting in a comfortable chair throughout the entire procedure.
Cardiovascular and Respiratory Physiology Data Recording and Analysis
Placement of electrodes/sensors, data recording, and data reduction followed conventions established for psychophysiological research and published guidelines (41). Physiological channels were A/D-converted, sampled at 400 Hz, and simultaneously streamed to disk and displayed on a PC monitor. A standard lead-II ECG was obtained. A second-generation Minnesota-type impedance cardiograph (HIC-2000; Instrumentation for Medicine, Inc., Old Greenwich, CT) measured electrical impedance changes in the thoracic cavity using four spot electrodes attached to the neck and thorax. A Finapres 2300 blood pressure monitor (Ohmeda, Inc., Madison, WI) obtained the continuous arterial pulse pressure waveform using the volume clamp method. Respiratory pattern data were measured using thoracic and abdominal bellows (Lafayette Instrument, Inc., Lafayette, IN) connected to pneumographic transducers (James Long Company, Inc., Caroga Lake, NY). Expiratory pCO2 was measured continuously by a calibrated infrared capnograph (N-1000; Nellcor, Hayward, CA) into which air was drawn with a flow rate of 150 mL/min through a 1.2-mm diameter plastic tube ending in a dual nostril prong. In addition to these continuous measurements, an automatic blood pressure monitor (Dinamap 1846SX; Critikon & GE Healthcare, Chalfont St. Giles, U.K.) measured blood pressure with a cuff around the upper right arm. Inflation was triggered 2 minutes after onset of each test segment.
Physiological signals were analyzed and averaged for each 5-minute period using an integrated suite of biosignal analysis programs written in MATLAB (Mathworks, Inc., Natick, MA). Rate–pressure product was calculated as HR x SBP. The impedance cardiogram (ICG) dZ/dt-signal was ensemble averaged in alignment with the R-wave time over 5-minute periods. Characteristic points (B, Z, X) of the inverted dZ/dt signal of ensemble-averaged beats were identified automatically after exclusion of abnormal beats and edited when necessary. Preejection period (PEP) (in milliliters) was calculated as the interval from the ECG Q-point to the ICG B-point. PEP is inversely related to left-ventricular contractility and beta-adrenergic sympathetic influences on the myocardium. Thoracic and abdominal respiration channels were converted to calibrated lung volume change using comparison data from a fixed-volume bag calibration procedure performed by each subject. Tidal volumes (in milliliters) were calculated between peaks and valleys of valid breaths, and total time of each breath was calculated and converted to respiratory rate (in cpm). End-tidal pCO2 (mm Hg) was determined from the capnometry signal as the levels at which pCO2 stopped rising at the end of expirations. Only breaths in which pCO2 waveforms reached a distinct plateau were considered to reflect arterial values.
Transfer function respiratory sinus arrhythmia (RSATF) was quantified by fast Fourier transform and the averaged periodogram method as the magnitude of the transfer function relating RR interval to lung volume oscillations at the prominent respiratory frequency (in ms/mL) (24). Spectral coherence between lung volume and RR interval at this frequency was required to be at least 0.5 for the estimate to be valid (less coherence indicates sources for RR interval variation other than respiration). Nonrespiratory adjusted RSA, the high-frequency (HF) power of heart period variability (in ms2), was computed as the summed power spectral density of RR interval between 0.15 to 0.5 Hz for normative comparison. Similarly, spectral density of RR interval was summed over the low (LF, 0.07–0.15 Hz, in ms2) and very low (VLF, 0.0033–0.07 Hz, in ms2) frequency bands. HF, LF, and VLF power measurements were normalized by natural logarithmic transformation. BRC (in mm Hg/ms) was estimated as the magnitude of the transfer function relating RR interval to systolic pressure oscillations at the prominent systolic pressure frequency in the 0.07- to 0.15-Hz band. Spectral coherence at this frequency was required to be at least 0.5. BRC was estimated for baseline periods only because initial analyses indicated that as a result of instability in SBP measurement during stress periods, coherence often was below 0.5. (In contrast, the transfer function RSA method during mental stress does not rely on SBP and showed relatively good coherence between IBI and respiratory oscillations.)
For each subject, a spreadsheet was assembled containing physiological parameters for each test segment for review by a senior psychophysiologist. Improbable or inconsistent values prompted reanalysis for validation. In a last step, outliers more than 2 standard deviations from the group mean that appeared inconsistent or improbable for that individual or the measure were eliminated. Less than 1% of data were excluded on this basis.
Data Analysis
After data reduction, a random-effects model assuming autoregressive covariance was used to examine potential differences among groups as recommended by Bagiella et al. (42) for analyzing psychophysiological data in repeated-measures experiments. Effects were examined for group (depressed or nondepressed), time, gender, and their interactions. Baseline differences were examined with parametric or nonparametric statistics as appropriate. A 5% two-tailed significance level was used for all tests. No correction for multiple tests was included. Note that the time course of measurement varied for the three data sets obtained during the stress test.
Psychophysiological signals were averaged for baseline, anticipation, speech, math, and the first two 5-minute segments of recovery. Affectivity was recorded after hookup, before baseline, before the anticipation period, after the anticipation period, after the math test, and after 10, 20, and 30 minutes of recovery. Cortisols were measured at these times and also after 45 and 60 minutes of recovery.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
|
|
Baseline Measures
Psychological Measures
Table 1 shows the baseline psychological measures for depressed and nondepressed subjects. The depressed group scored significantly higher on the HRSD (p < .0001), BDI (p < .000), the negative affect subscale of the PANAS (p < .000), and also reported significantly more distress on the PSS (p < .0001).
Cardiovascular Risk
There were no significant differences in baseline SBP, DBP, lipids, percent FMVD, or percent nitroVD change or ADMA between depressed and control patients (Table 2). However, high sensitivity C-reactive protein was significantly higher in depressed compared with nondepressed subjects (Z = –2.8, p = .005).
Daytime Cortisol
There were no group differences in waking, 30-minute postwaking, or cortisol slopes, nor any group x gender interactions.
Stress Test Baseline and Reactivity
Table 3 shows the results of the random-effects model for the major psychophysiological variables, cortisol, and the negative affect scores. There were significant overall differences between groups in negative affect, cortisol, and RSATF. Depressed subjects reported more negative affect to the acute stressor than nondepressed subjects. RSATF (Fig. 1) was lower in depressed subjects in reaction to the stressor. There was a striking time effect for all variables (except log VLF power and tidal volume), reflecting significant reactivity induced by the stress test paradigm. There was also a significant group x time effect for cortisol, with depressed subjects showing less cortisol reactivity than nondepressed subjects (Fig. 2). In addition, there was a significant gender x time effect for cortisol reactivity. An area under the curve analysis of log cortisol with a two-way analysis of variance (group, gender) also revealed a significant difference between cortisol patterns in depressed compared with nondepressed subjects (F = 6.32, p = .015). There were no significant differences in BRC between depressed and nondepressed subjects at baseline.
|
|
|
There also were significant gender effects for SBP, DBP, and cardiac output (CO). Women had lower SBP, DBP, and CO than did men. (CO is normally adjusted by surface area to create CI, which is generally not different between genders but this adjustment was not made in this study.) Over time, women had less increase in PEP than did men, and there was a significant gender x group x time effect for HR with depressed women having higher sustained HR than other groups.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Our hypothesis, that we would find evidence of hypercortisolemia in depressed patients, was not confirmed; in fact, we found a hypocortisol response to acute stress in depressed individuals. Hypercortisolemia has been most consistently found in patients with "severe forms of depression," for instance, those with psychotic features, which none of our patients had, or with melancholia (43–45). Patients with more moderate levels of depression may not show hypercortisolemia (12). The mean Hamilton scores in our study (M = 18.2) were similar to those in the Miller article (M = 17.8) (12). Gold and Chourous (45) and others (46) have argued that patients with some types of depression, e.g., those with atypical features (lethargy, fatigue, lifetime course with few remissions), may have downregulated HPA activity. Appels (47), in discussing why "vital exhaustion," which has been found to be one of the precursors of myocardial infarction and other cardiac events, noted that "these feelings probably reflect decreased activity of the hypothalamic–pituitary–adrenal axis." In exploring this relationship, Nicolson and van Diest (48) found lower basal cortisol levels in subjects with vital exhaustion, although the cortisol responses to the speech task were similar in the vital exhaustion group and in control groups in their study. We did not assess vital exhaustion in our sample, but most of the patients reported high levels of fatigue.
The finding of a lack of cortisol response to stress is consistent with the Burke et al. (49) meta-analysis of seven studies that examined the association between depression and cortisol responses to psychological stressors. They found significant time-of-day effects for studies that performed stress tests in the afternoon. Patients with MDD had blunted reactivity and impaired recovery to stress reactivity. The blunted reactivity was most pronounced in older and more severely depressed patients. In a study with a comparable age range, and also using the TSST, Gotthardt et al. (8) reported only a 10% rise from prestress in the depressed group compared with over 100% in the nondepressed group, although the depressed group had higher baseline cortisols. However, control subjects in the Gotthardt et al. (8) sample showed a more rapid return to baseline and not the prolonged elevation evidenced here. Of note, we used a longer and more intensive stressor (social stress followed by math stress) than did Gotthardt and colleagues. Our results are also consistent with findings by Miller et al. (12) who recently reported a lower cortisol response in a stress test (conducted in the morning) in young, healthy, depressed and nondepressed volunteers. Unfortunately, we did not collect lifetime course data that would have allowed us to distinguish atypical from typical depression. The lack of hypercortisolemia may also reflect long-term adaptation of the HPA axis to chronic stress or depression (50).
Why might this lack of cortisol response be important? First, as suggested by Miller et al. (12), cortisol helps suppress an inflammatory response. Although it is important for the immune system to react to a pathogen, sustained proinflammatory activity might promote atherogenesis as discussed in the introduction. Miller et al. (12) suggested that acute stress boosts C-reactive protein levels in the blood that remain elevated longer in depressed subjects compared with nondepressed subjects. Consistent with this hypothesis, the high-sensitivity CRP protein levels were significantly higher in depressed compared with control subjects. Alsesci et al. (51) have recently shown that plasma IL-6 is a good predictor of future risk for CVD and was significantly elevated in patients with MDD, although there were no differences in morning cortisols in our sample. Second, as mentioned in the introduction, a lack of cortisol response in patients who experience an MI may be related to increased risk of death during resuscitation (7). If so, this would suggest that cardiovascular mortality should be higher in depressed patients having an MI.
Our second major finding was a lower RSATF level in depressed compared with nondepressed subjects. As noted in the introduction, a high degree of RSA is observed in normal hearts with good cardiac function, whereas RSA can be significantly decreased in patients with severe CAD or heart failure. The relative risk of sudden death after acute MI is significantly higher in patients with decreased RSA (18), although the exact mechanisms are not known. Some studies have found an association between RSA and depression (19–21); others have not (22,23). Some of the inconsistencies in prior RSA findings relating to depression might be the result of inadequate adjustment for respiratory confounding factors in the relationship between RSA and vagal control. In contrast to the HF power estimate of RSA, the RSATF measure, which adjusts for these factors, was a good discriminator of depressed versus nondepressed groups in the present study. The HF power RSA measure simply sums all sources of heart period variation in a frequency band assumed to be associated with respiration (9–30 cpm) without measuring respiration to confirm cardiorespiratory coherence, the defining feature of RSA, or that a subject actually was breathing in the assumed frequency range. In patients with arrhythmias, beat-by-beat editing is necessary, as we did, to eliminate this source of error. It is possible that other studies that assessed RSA for relating depression to CVD risk were affected by confounds from clinical or subclinical cardiac pathology and respiratory deviances. Our failure to find HRV differences using spectral power estimates is consistent with a recent, large 24-hour study comparing depressed and nondepressed individuals with stable CHD (52).
In contrast to several other studies of depression, we did not find any differences in BRC between the depressed and nondepressed groups at baseline. This may have been the result of the higher percentage of hypertensive subjects in the nondepressed group, which rendered their baroreflex less sensitive and masked more subtle effects of psychiatric diagnosis on BRC.
The lack of differences in measures that partly reflect sympathetic activity (SBP, PEP reactivity) between depressed and nondepressed subjects is consistent with the meta-analysis of Kibler and Ma (53) of the effects of depression on cardiovascular reactivity. In their analysis, the only study that included patients with depression, found no relationship between mental arithmetic and systolic blood pressure, diastolic blood pressure, or heart rate reactivity, although the stress tests did elicit substantial cardiovascular reactivity. In a study with a comparable design, Strike et al. (54) found that systolic blood pressure increased by an average of 37% in patients with CAD compared with 24% in healthy controls. HR increased by 28% versus 8% in controls. In our study, SBP increased by 32% for depressed patients and 40% for controls; and HR by 22% and 18%, respectively. Our levels of FMVD are much lower than were reported by Sherwood et al. (33), perhaps reflecting a higher CVD risk in our population. We also observed gender effects for SBP, DBP, and CO.
We selected a control group that was not different from the depressed patients in baseline traditional CVD risk factors. Both groups were at elevated risk for CVD based on a history of hypertension and/or hypercholesterolemia and had similar levels of baseline total, LDL, HDL, triglyceride, SBP, DBP, HR, and BMI. Without such matching, one cannot attribute physiological differences between groups to depression.
The study has a number of limitations. It included a small number of males in the CVD risk condition. Given the importance of gender effects, future studies need to have adequate numbers of males and females. We also did not adjust for experiment-wise error.
Patients with CVD risk factors take a variety of medications that might affect the variables measured. We followed standard practice in psychophysiological practice to stop ß-blockers and any other adrenergic or cholinergic medications 24 hours before stress testing to be able to test cardiovascular reactivity. Although these medicines taken the day before and discontinued may still have some effect, it was unlikely to affect our results because the two groups used these drugs equally. Continuing drug effects might possibly influence HR reactivity but not BP reactivity (55) or RSA (56).
Our primary findings are a lower cortisol response to stress, a lower RSATF, and higher C-reactive protein levels in older depressed patients with elevated cardiovascular risk. We do not know if these differences are isolated to these systems or represent a more general dysregulation. Nonetheless, under conditions of stress, reduced cortisol response coupled with altered vagal regulation may increase risk in vulnerable patients. Chronic cortisol hyporesponsivity to stress might counteract the normal reduction in inflammatory response after a stress, thus increasing CVD risk. Reduced parasympathetic activity might also contribute to prolonged inflammation. Future research should examine interactions of endocrine, autonomic, and immune systems in depressed patients at risk for CVD. These results need to be replicated in a larger sample with more males and females. We have included effect sizes in the tables to help determine sample size for variables of interest.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This study was supported by NIA/NCI Program Project AG18784 and in part by a grant 5 M01 RR000070 from the National Center for Research Resources, National Institutes of Health.
DOI:10.1097/01.psy.0000222372.16274.92
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. H. Lichtman, J. T. Bigger Jr, J. A. Blumenthal, N. Frasure-Smith, P. G. Kaufmann, F. Lesperance, D. B. Mark, D. S. Sheps, C. B. Taylor, and E. S. Froelicher Depression and Coronary Heart Disease: Recommendations for Screening, Referral, and Treatment: A Science Advisory From the American Heart Association Prevention Committee of the Council on Cardiovascular Nursing, Council on Clinical Cardiology, Council on Epidemiology and Prevention, and Interdisciplinary Council on Quality of Care and Outcomes Research: Endorsed by the American Psychiatric Association Focus, July 1, 2009; 7(3): 406 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Lichtman, J. T. Bigger Jr, J. A. Blumenthal, N. Frasure-Smith, P. G. Kaufmann, F. Lesperance, D. B. Mark, D. S. Sheps, C. B. Taylor, and E. S. Froelicher Depression and Coronary Heart Disease: Recommendations for Screening, Referral, and Treatment: A Science Advisory From the American Heart Association Prevention Committee of the Council on Cardiovascular Nursing, Council on Clinical Cardiology, Council on Epidemiology and Prevention, and Interdisciplinary Council on Quality of Care and Outcomes Research: Endorsed by the American Psychiatric Association Circulation, October 21, 2008; 118(17): 1768 - 1775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Giese-Davis, F. H. Wilhelm, A. Conrad, H. C. Abercrombie, S. Sephton, M. Yutsis, E. Neri, C. B. Taylor, H. C. Kraemer, and D. Spiegel Depression and stress reactivity in metastatic breast cancer. Psychosom Med, September 1, 2006; 68(5): 675 - 683. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
