Association Between Type D Personality, Depression, and Oxidative Stress in Patients With Chronic Heart Failure
Kupper, Nina PhD; Gidron, Yori PhD; Winter, Jobst MD, PhD; Denollet, Johan PhD
From the CoRPS-Center of Research on Psychology in Somatic diseases (N.K., J.D.), Tilburg University, Tilburg, Netherlands; School of Health Sciences and Social Care (Y.G.), Brunel University, Uxbridge, Middlesex, UK; and the Department of Cardiology (J.W.), TweeSteden Hospital, Tilburg, Netherlands.
Address correspondence and reprint requests to Nina Kupper, Department of Medical Psychology, Tilburg University, Center of Research on Psychology in Somatic diseases, Warandelaan 2, PO box 90153, 5000 LE Tilburg, Netherlands. E-mail: firstname.lastname@example.org
Received for publication September 5, 2008; revision received May 20, 2009.
The present research was supported, in part by VICI Grant 453-04-004 from the Netherlands Organization for Scientific Research (NWO, The Hague, Netherlands) (J.D.) and Grant NHS 2003B038 from The Netherlands Heart Foundation (J.D.).
Objective: To examine whether markers of oxidative stress differ as a function of Type D personality, depression, and chronic heart failure (CHF) etiology. Type D (distressed) personality and depression are related to poor cardiac prognosis. Because patients with CHF are characterized by increased oxidative stress, this may be a candidate mechanism responsible for the adverse prognosis in emotionally distressed patients with CHF.
Methods: Serum levels of xanthine oxidase (XO), inducible heat shock protein (Hsp)70, and deoxyribonucleic acid damage marker 8-OHdG were measured in 122 patients, and effects of Type D, depression, and etiology were assessed.
Results: CHF patients with Type D personality had lower levels of Hsp70 than non-Type D patients (6.48 ng/mL versus 7.85 ng/mL, p = .04, d = 0.26), and in case of an ischemic etiology, higher levels of XO (13.57 ng/mL versus 9.84 ng/mL, p = .01, d = 0.98). There were no significant univariate differences for depression. When adding depression as an additional independent variable in the Type D analysis, the effect of Type D personality remained significant (F = 5.460, p = .02) and was independent of depression (F = 0.942, p = .33). The ratio of XO to Hsp70 was significantly higher in Type D patients with CHF as compared with non-Type D patients (6.14 versus 2.83, p = .03, d = 0.39), independent of etiology class.
Conclusion: CHF patients with Type D personality are characterized by an increased oxidative stress burden, apparent in the decreased antioxidant levels and an increased oxidative stress ratio.
8-OHdG = 8-hydroxy-2-deoxyguanosine; ACE = angiotensin-converting enzyme; CHF = chronic heart failure; ELISA = enzyme-linked immunosorbent assay; HF = heart failure; Hsp70 = heat shock protein 70; LVEF = left ventricular ejection fraction; MANOVA = multivariate analysis of variance; MDD = major depressive disorder; MI = myocardial infarction; NA = negative affectivity; ROS = reactive oxygen species; SI = social inhibition; TNF = tumor necrosis factor; XO = xanthine oxidase.
Oxidative stress refers to an imbalance between activity of reactive oxygen species (ROS) and antioxidant mechanisms, and has been implicated in the pathogenesis of chronic heart failure (CHF) (1–3). Xanthine oxidase (XO) is able to generate ROS that play a pathophysiological role in both experimental and clinical heart failure by influencing several characteristics of the failing heart, e.g., contractile function, interstitial fibrosis, endothelial dysfunction, and myocyte hypertrophy. XO expression is increased in heart failure due to ischemia (4), and inhibition of XO improves cardiac contractility in patients with CHF (3,5,6). Oxidative stress may also be quantified by examining the amount of oxidative stress-induced deoxyribonucleic acid (DNA) damage, as indicated by serum levels of the DNA-repair product 8-hydroxy-2-deoxyguanosine (8-OHdG) formed from 2-deoxyguanosine in DNA by hydroxyl free radicals. Animal research has shown the role of oxidative DNA damage in cardiac dysfunction (7). It was demonstrated that DNA damage was increased in nonischemic patients with CHF, and its levels reflected the severity of the cardiac dysfunction and cardiac remodeling (8).
Antioxidant defense mechanisms include the activation of heat shock proteins. When cardiomyocytes are exposed to acute hypoxia or ischemia, the production of inducible heat shock protein (Hsp) 70 is enhanced (9–11). Inducible Hsp70 has a protective role after ischemia (12) and is associated with a reduced risk of developing coronary artery disease (13). Animal research showed that, when chronic heart failure develops, inducible Hsp70 levels return to control values (14), which may contribute to a decrease in contractile function during the development of heart failure and the functional deterioration of the failing heart (15). Few studies have examined Hsp70 in humans with CHF, and yielded inconsistent results. Whereas one study found increased levels of Hsp70 in patients with CHF (16), another study reported equal levels of Hsp70 production in normal hearts versus hearts from end-stage heart failure patients (17), thus replicating the results found in the animal studies.
Social stress may induce oxidative stress (18). The distressed personality (Type D), defined as the tendency to experience negative emotions, and to inhibit the expression of these emotions in a social context, has been shown to predict poor prognosis in coronary artery disease (19,20) and CHF (21), independent of other biomedical and psychosocial risk factors. Mechanisms responsible for the adverse outcome in Type D patients with CHF could depend on an underlying inflammatory process (22), but may also include oxidative stress. Hsp70 is involved in regulating the tumor necrosis factor (TNF)-α production cascade (23,24), and XO is implicated in cardiac remodeling and left ventricular dysfunction (5). Decreased levels of Hsp70 and increased levels of XO would result in increased oxidative damage, increased myocardial oxidative stress, and increased levels of TNF-α, hereby negatively affecting cardiac prognosis.
Further clues for a potential role of oxidative stress as a mediator between distress and prognosis come from studies reporting increased oxidant levels in emotionally distressed (25) and clinically depressed subjects (26–29). Mixed results are presented for antioxidant levels in depressed patients, as increases have been found (26,27) as well as unaltered levels (28,29). Increased oxidative DNA damage has been reported for people with depression and chronic work stress, and was found to be induced by stress in animals (30). The relationship between depression and oxidative stress has not yet been examined in patients with CHF.
To date, no prior study has examined the relationship between oxidative stress levels and emotional distress in patients with CHF. Therefore, the purpose of the present study was to determine whether serum levels of several markers of oxidative stress, i.e., XO, Hsp70, and 8-OHdG, and the ratio between XO and Hsp70 differ as a function of Type D personality and depression. Patients with systolic heart failure may have an ischemic or a nonischemic etiology of their disease. The process of cardiac remodeling and eventual progression of left ventricular dysfunction is complex, and there may be differences between patients with different etiologies (31). We therefore will test whether the above relationships depend on CHF etiology.
A total of 122 consecutive CHF outpatients visiting the cardiology outpatient clinic of the TweeSteden teaching hospital in Tilburg, Netherlands, were included in the blood collection substudy of the TweeSteden Heart Failure Study between October 17, 2003 and January 7, 2005. Patients were included when they were aged <80 years, pharmacologically stable 1 month preceding inclusion, and having a left ventricular ejection fraction (LVEF) of ≤40%. Because of their hearts' intact pump function during systole, and diminished filling capacity during diastole, patients with diastolic heart failure were excluded from the study. In addition, patients with inadequate proficiency of the Dutch language, with significant cognitive impairments, life-threatening comorbidities (e.g., chemotherapy-treated cancer), clinical signs of acute infections, with an active episode of gout or arthritis, or use of anti-inflammatory medication were excluded (about 10% of the initial sample, n = 136). All patients were treated following the most recent guidelines for the treatment of CHF (32), and were informed about the study by a cardiologist or a specialized heart failure nurse. All patients provided their written informed consent and participation was voluntary. The hospital's medical ethics committee approved the study protocol, and the study was carried out in accordance with the Helsinki Declaration.
Patients completed a psychological survey assessing Type D personality, depression, and sociodemographic characteristics. Patients were asked to fill out the survey at home and return it in a self-addressed envelope. All questionnaires were checked for completeness. Patients who had left open several questions were called to obtain the answers or they were mailed a copy of the items and asked to complete them. In case the questionnaires were not returned within 2 weeks, patients received a reminder telephone call. There is no information available on the Type D or depression status of those patients who needed to be reminded to fill out the questionnaire, nor is there information on the Type D or depression status of those patients who declined to participate.
Final analyses were based on data from a maximum of 110 patients, due to missing psychological data (n = 5), diastolic heart failure (n = 1), or because patients constituted outliers (>3 standard deviation [SD] from the mean; in four instances, this was the case for XO, in one instance for 8-OHdG, and in one instance for Hsp70).
Blood was allowed to clot at room temperature and centrifuged. Serum samples were stored at −80°C in anticipation of further processing. Hsp70 (sensitivity: 0.50 ng/mL) was assessed, using Stressgen's StressXpress Hsp70 ELISA (Sanbio, Uden, Netherlands), which is a quantitative sandwich immunoassay. We used competitive enzyme-linked immunosorbent assays (ELISAs) to measure serum levels of 8-OHdG (sensitivity: 0.13 ng/mL) (Orange Medical, Tilburg, Netherlands) and XO (sensitivity: 0.20 ng/mL) (DiaMed Eurogen, Turnhout, Belgium). All tests were performed in accordance with the manufacturers' recommendations. The sensitivity of all tests was calculated by the mean of 6 zero-values + 3 SD extrapolated on the standard curve. Values below sensitivity were raised to sensitivity value. The intra-assay variation was <9% for all ELISAs, and the interassay variation was <10%.
Type D Personality
Type D personality was assessed, using the Type D Scale (DS14) (33), which consists of 14 items that are divided into two subscales measuring two stable personality traits, negative affectivity (NA) (tendency to experience negative emotions across time and situations), and social inhibition (SI) (tendency to inhibit the expression of negative emotions in social interaction). The items are answered on a 5-point Likert scale ranging from “false” (0) to “true” (4). A standardized cutoff of ≥10 on both subscales indicates Type D caseness. The two subscales have good psychometric qualities, with Cronbach's α = 0.88/0.86 and 3-month test-retest reliability r = .72/0.82 for the NA and SI subscale, respectively (33). In the present data set, the internal consistency of the DS14 was equally high (Cronbach's α was 0.81 for NA and 0.87 for SI).
Depressive symptoms were assessed with the Beck Depression Inventory (BDI) (34). The BDI is a 21-item self-report questionnaire developed to assess presence and severity of depressive symptoms, from which a cognitive (questions 1–13) and a somatic subcomponent (questions 14–21) can be extracted. Each item is rated on a 0 to 3 scale. A total score is obtained by summing all the items. The BDI is a reliable and well-validated measure of depressive symptomatology (35), and is a widely used self-report measure of depression in patients with cardiovascular disease. In the current study, we used a standardized cutoff score of ≥10, indicative of at least mild-to-moderate symptoms of depression (36).
Diagnosis of lifetime major depressive disorder (MDD) was determined with the World Health Organization-authorized Dutch version of the Composite International Diagnostic Interview (CIDI) (37) based on the diagnostic criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (38). The CIDI has acceptable interrater and test-retest reliability for most nonpsychotic diagnoses including MDD (39).
Demographics and Clinical Covariates
Sociodemographic information included sex, age, educational level, occupational status, and social situation (with partner versus alone). Smoking status and anthropometric characteristics were assessed by means of self-report. Information on clinical variables including LVEF, risk factors (obesity, hypertension, diabetes mellitus, and hyperlipidemia), comorbid diseases (atrial fibrillation, liver disease, renal disease, chronic obstructive pulmonary disease, gout, rheumatoid arthritis, gastrointestinal disease, and peripheral arterial disease), and medication (β blockers, aspirin, diuretics, ACE inhibitors, calcium antagonists, statins, digoxin, AT-II antagonists, allopurinol, and psychotropic medication [tricyclic antidepressants, benzodiazepines, selective serotonin reuptake inhibitors]) were obtained from the patients' medical records or the treating cardiologist/CHF nurse. An underlying etiology of heart failure (HF) (ischemic versus nonischemic) was recorded at the time of inclusion based on clinical judgment of the treating cardiologist and the results of multiple invasive and noninvasive tests (cardiac ultrasound, coronary angiogram, nuclear studies). Patients were classified as having an ischemic etiology if there was a confirmed history of coronary artery disease (myocardial infarction [MI], angina pectoris, previous coronary artery bypass graft/percutaneous coronary intervention). All patients received a MIBI scan to exclude current (acute or chronic) ischemia. Classification took place regardless of other nonischemic causes of HF, such as chronic hypertension.
Demographic and CHF characteristics, medication use, prevalence of comorbid diseases, and established biomedical risk factors were considered as potential covariates in our analysis. To this end, discrete variables were compared with χ2 tests and are presented in Table 1 as percentages (n). When the prevalence of one of the variables within the Type D or non-Type D group was <5, the Fisher exact test was used. Continuous variables were compared with Student's t tests and are presented in Table 1 as mean (SD) values. XO, Hsp70, and 8-OHdG were modeled as continuous variables. Because of skewed data distributions, Hsp70 and XO were normalized by a natural logarithmic transformation (referred to as Hsp70 and XO for simplification). Furthermore, the interrelationships between the three oxidative stress markers and between the markers and the potential covariates were calculated, using Pearson/Spearman correlations.
Multivariate analyses of variance (MANOVAs) were used to compare between-group differences (Type D versus non-Type D and depressed versus nondepressed) on serum levels of Hsp70, XO, and 8-OHdG. We adjusted for any significant biomedical (risk factors, medication, or comorbidities) or demographic covariates reflecting initial group differences by adding them to the MANOVA either as a covariate or as an additional independent variable when the covariate was nominal. Depression was added as an additional independent variable in the Type D analyses to determine whether the effect of Type D would be independent of depression. The influence of etiology on the relationship between Type D personality and the oxidative stress markers was examined by adding etiology and the interaction between Type D personality and etiology as additional independent variables. In analogue with recent studies that demonstrated the importance of increased pro- to anti-inflammatory cytokine ratio for poor prognosis in acute coronary syndrome (40,41) as well as for progression of HF subsequent to MI (42), we finally tested the relationship between Type D personality, depression, and the imbalance between oxidant (i.e., XO) and antioxidant (i.e., Hsp70) marker levels by taking their ratio (XO/Hsp70), with the appropriate covariate(s) in an analysis of covariance. All tests were two-tailed, and a p < .05 was considered statistically significant. Cohen's effect sizes were calculated for all significant results regarding Type D, depression, and etiology group differences. All analyses were performed, using SPSS 17.0 for Windows.
Patient characteristics stratified for Type D personality are displayed in Table 1. The prevalence of Type D personality was 18.2%. On average, the patients were diagnosed with HF 4.3 years ago (SD = 4.4). Ischemic etiology was equally prevalent in the Type D and non-Type D group, with 12 patients being Type D and having an ischemic etiology of HF. The group of Type D CHF patients did not differ from the group of non-Type D patients on demographics, disease characteristics, or biomedical risk factors. In addition, there were no group differences in the prevalence of comorbid disease, such as atrial fibrillation (overall prevalence 27%; p = .80), liver disease (overall prevalence 2.7%; p = .99), renal disease (overall prevalence 11%; p = .99), peripheral arterial disease (overall prevalence 15.5%; p = .99), gout (overall prevalence 5.4%; p = .71), rheumatoid arthritis (overall prevalence 7.2%; p = .17), gastrointestinal disease (overall prevalence 7.3%; p = .64), or chronic obstructive pulmonary disease (overall prevalence 10.9%; p = .69). Type D and non-Type D patients did not show differences in medication use, except for psychotropic medication (p = .04), that was used more often by Type D patients with CHF (Table 1). As selective serotonin reuptake inhibitors and tricyclic antidepressants may induce oxidative stress in vitro (43) and in vivo (44), antidepressant use was included as a covariate in further analyses assessing between-group differences. Furthermore, patients with a Type D personality had higher depression scores (p = .002). The difference between non-Type D and Type D patients in cognitive symptom subscore was on average 3.5, whereas the difference for the somatic symptom subscore was on average 0.95. There were no significant group differences in lifetime depression diagnosis, although depression diagnosis was more prevalent in Type D patients.
Clinical Characteristics and Oxidative Stress
None of the relationships between the oxidative stress markers were significant. Nevertheless, correlations between Hsp70 and 8-OHdG and between XO and 8-OHdG were in the expected direction. Hsp70 levels were negatively related to 8-OHdG (r = −.14, p = .14) and XO (r = −.07, p = .45). XO and 8-OHdG levels showed a slight positive correlation (r = .14, p = .16).
There were no significant sex differences in the levels of the three oxidative stress markers (XO: F = 0.031, p = .86; Hsp70: F = 0.491, p = .49; 8-OHdG: F = 3.274, p = .07) nor were there significant effects of age (XO: r = .03, p = .75; Hsp70: r = .10, p = .30; 8-OHdG: r = .11, p = .24). As Table 2 shows, XO correlated positively with LVEF, agreeing with correlations in previous studies (16). The use of digoxin and allopurinol were related to Hsp70 levels, as was incidence of liver disease. Both 8-OHdG and XO were negatively associated with taking ACE inhibitors.
Emotional Distress, Etiology, and Markers of Oxidative Stress
Table 3 displays the mean (SD) values for all three markers and for the ratio between the oxidant (XO) and antioxidant marker (Hsp70), stratified for Type D personality, whereas Table 4 displays these mean values stratified for depression.
MANOVA showed that Type D patients with CHF had significantly lower serum levels of Hsp70 compared with non-Type D patients and that there were no significant differences for 8-OHdG and XO (overall multivariate model: F = 1.546, p = .21) (Table 3). When adding the significant covariate (psychotropic medication use) to the model, the difference between Type D and non-Type D CHF patients for Hsp70 remained significant (F = 4.580, p = .03), and was a small-to-medium effect according to Cohen's d. Results for XO and 8-OHdG were unmodified.
Despite previous results that both diabetes and hyperlipidemia are related to increased oxidative stress (45,46), together with the fact that these comorbidities occurred more often in the Type D patients (although not significantly), diabetes and hyperlipidemia did not affect levels of Hsp70, XO, or 8-OHdG, nor did they affect the relationship between Type D personality and Hsp70 as covariates.
A similar MANOVA was performed for depression to examine whether there were differences in oxidative stress levels between depressed and nondepressed patients. MANOVA results (overall multivariate model: F = 0.347, p = .79) (Table 4) showed that there were no significant differences for Hsp70, XO, and 8-OHdG. Results were not affected by adding psychotropic medication use as a covariate. Nonparametric correlations between depression and Hsp70 (r = .08), XO (r = −.04), and 8-OHdG (r = .03) were all nonsignificant. Analyses were repeated for lifetime depression diagnosis as assessed with the CIDI, rendering similar, nonsignificant results (results not shown).
When adding depression as an additional independent variable in the Type D analysis (overall multivariate model: F = 2.072, p = .11 for Type D and F = 0.524, p = .67 for depression), the effect of Type D personality remained significant (F = 5.460, p = .02), and was independent of depression (F = 0.942, p = .33).
When etiology was added to examine its main effects on the oxidative stress markers and the interaction effects of Type D personality with etiology (depression was dropped from this analysis) (overall multivariate model: F = 1.804, p = .15 for Type D, F = 2.444, p = .07 for etiology and F = 2.766, p = .046 for the interaction effect of etiology*Type D), the main effect of Type D personality on Hsp70 remained significant (F = 5.268, p = .02). Etiology did not significantly affect Hsp70, and there was no significant interaction with Type D personality. Regarding XO, there was no main effect for Type D personality, but there was a significant main effect for etiology (F = 3.984, p = .049), with ischemic patients having slightly higher levels of XO compared with nonischemic CHF patients. Furthermore, a significant interaction effect of etiology with Type D personality was observed for XO (F = 6.640, p = .01): Whereas for non-Type D CHF patients, there was no difference in XO levels between the two etiology classes, Type D CHF patients with an ischemic etiology had significantly higher XO levels compared with nonischemic CHF patients with a Type D personality, and this was a large effect according to Cohen's d (Fig. 1). There were no significant main or interaction effects of personality and etiology for 8-OHdG (data not shown). Adding LVEF or New York Heart Association class to the model as an additional covariate did not significantly affect the results.
We further analyzed the apparent imbalance between the oxidant and antioxidant marker by examining between-group differences for the XO/Hsp70 ratio. The high ratios reported in Table 3 are due to high concentrations of XO in combination with low Hsp70 levels. Analysis of covariance results demonstrated that Type D patients with CHF had a significantly higher ratio (p = .03), indicating a larger oxidative stress burden in these patients (Table 3). Cohen's d showed that this difference between the groups represented a small-to-medium effect size, and indicated a 20% to 25% nonoverlap in data distributions between Type D and non-Type D groups. When adding etiology as an additional independent variable, Type D still was a significant factor explaining between-group differences in this ratio (F = 4.108, p = .045). There was no main effect of etiology (F = 0.178, p = .67), nor was there a significant interaction effect of etiology by Type D classification (F = 1.838, p = .18) for the XO/Hsp70 ratio.
This preliminary study is the first to report that the level of oxidative stress marker Hsp70 and the XO/ Hsp70 ratio in patients with CHF are related to Type D personality, but not to depression. As our results demonstrate, CHF patients with a Type D personality had lower levels of Hsp70. Type D patients with an ischemic etiology also had higher levels of XO. The influence of Type D personality was found to be independent of depression. Furthermore, CHF patients with a Type D personality had a significantly higher XO/ Hsp70 ratio. Because imbalance between oxidant and antioxidant mechanism markers has been demonstrated in patients with CHF (47), these initial findings suggest that, in CHF patients with Type D personality, this imbalance may be emphasized.
Type D patients with CHF typically have a worse prognosis compared with non-Type D CHF patients (20–22), which might have to do with this imbalance between oxidants and antioxidants, and specifically with the observed decreased Hsp70 levels. Hsp70 is cardioprotective, especially after acute ischemic or hypoxic stress (12,13), and may exert its protective effect by regulating TNF-α levels (23,24). TNF-α is known to depress cardiac contractility by inducing nitric oxide production (48). Hsp70 controls the activation of nuclear factor-κB, the primary transcription factor controlling TNF-α gene expression, thereby negatively affecting the TNF-α production cascade (24). This regulatory cardioprotective effect of Hsp70 on TNF-α may be diminished in the high-risk Type D patients with CHF.
The cytokine hypothesis of HF suggests inflammation to play an important role in the progression and prognosis of HF. Both depression (49,50) and Type D personality (51,52) have been associated with increased levels of cytokines and their receptors in patients with CHF, suggesting a role of cytokines as a mechanism by which depression and Type D personality influence prognosis. Hsp70 is involved in the TNF-α production cascade. The regulatory role of Hsp70 on TNF-α, together with the lower levels of Hsp70 observed in the present study, may explain previous findings that Type D is related to increased levels of TNF-α in patients with CHF (22) and to poor prognosis in CHF (21,22).
Another mechanism by which a decrease in Hsp70 would be able to affect adversely HF is through its protective role in apoptotic cell death. Although the exact mechanisms are still elusive, there are some suggestions that Hsp70 is involved in protecting cells from apoptosis by stabilizing lysosomes and by regulating iron homeostasis (53). Future studies may wish to test these parameters in Type D patients.
A substantial body of literature indicates that emotional distress is associated with increased levels of oxidative stress. However, no previous studies have been performed on Hsp70 levels in relation to depression, depressive symptoms, or personality. Our results in Type D patients with CHF are in correspondence with the differential levels of antioxidants in other emotionally distressed patient groups (25–27), although in the current study no differences were found for depression. Previous research suggested that both XO (44) and 8-OHdG (54) may be increased in patients with MDD. Chronic work stress and depressive symptoms have been related to elevated 8-OHdG repeatedly, although it also has been suggested that, in men with high tension and anxiety levels, oxidative DNA damage levels are lower (28). The results of the current study do not support these previous findings, as we found no effects of Type D personality on XO and 8-OHdG levels, and no effect of depression on all measures of oxidative stress. Further research is essential as this is the first study to examine emotional distress in relationship to oxidative stress in the context of a cardiac disease.
An obvious question regarding the results of this paper is why Type D did and depression did not significantly affect oxidative stress levels. It has been established that Type D personality and depression are two separate constructs that differentially affect outcome, as multiple studies that have directly compared the effects of Type D personality and depression have reported distinct differences between Type D and depression in their effects on outcome variables (55–57). It is conceivable that the mechanisms through which depression and Type D personality exert their detrimental effect on prognosis may also (partly) differ.
In the present study, imbalance between antioxidant mechanism marker (Hsp70) and ROS producer (XO) in Type D CHF patients of ischemic etiology did not result in increased levels of 8-OHdG (also reflected in the low correlations between XO, Hsp70, and 8-OHdG), which, with respect to magnitude, might have been expected based on previous research (44). A possible explanation for this lack of association between DNA damage levels and XO and Hsp70 production may be that patients all had stable CHF, in which cardiac remodeling was already advanced. Although oxidative stress still characterizes these patients, it may be that, at this stage, this is not accompanied by major cell loss. In addition, many factors may influence DNA damage, XO and Hsp70 being only a few of them. The amount of explained variance may therefore also be limited.
Finally, XO expression is known to be increased in HF due to an ischemic event in rodents (4), the current study confirming this association in patients with CHF. The finding of the elevated XO in combination with lower levels of Hsp70 in ischemic Type D patients may indicate that such patients might particularly be at risk for poor prognosis, due to poor regulatory processes of oxidative stress and its potential adverse cardiac effects.
The present findings should be interpreted with the appropriate caution. The small sample size of 110 CHF patients (of which 20 were Type D) makes the analysis more vulnerable for Type II errors. In addition, patients with CHF suffer from multimorbidity and polypharmacy, and in our study, it seemed that in Type D subjects this was more pronounced (although not significantly) compared with non-Type D patients. It is conceivable that all these factors may affect disease severity and oxidative stress levels additively. For example, comorbidities, such as hyperlipidemia and diabetes, are themselves related to oxidative stress (57,58), although not significantly in the current study. Because of the limited sample size, we were not able to include them in the analysis. The complex interactions of disease characteristics and comorbidities of patients with CHF warrant the attention of future research. A further limitation is the cross-sectional study design, implying that no conclusions can be drawn on causality, nor in relation to direction of effects, although personality is a stable trait not expected to change due to illness or disease severity. Recently, the latter was formally tested in a study in patients with MI reporting that Type D personality was not related to LVEF, whereas post-MI depression was (58), which was confirmed in the current study.
In conclusion, decreased Hsp70 levels in CHF patients with Type D personality and an increased XO in those with ischemic etiology suggest an increased oxidative stress burden in Type D patients. Whether oxidative stress levels mediate the link between Type D personality and prognosis in CHF merits the attention of future investigation.
We thank Dr. Hooijkaas from the Erasmus Medical Centre in Rotterdam for providing his laboratory facilities to our analyst.
1. Nojiri H, Shimizu T, Funakoshi M, Yamaguchi O, Zhou H, Kawakami S, Ohta Y, Sami M, Tachibana T, Ishikawa H, Kurosawa H, Kahn RC, Otsu K, Shirasawa T. Oxidative stress causes heart failure with impaired mitochondrial respiration. J Biol Chem 2006;281:33789–801.
2. Takenaka H, Kihara Y, Iwanaga Y, Onozawa Y, Toyokuni S, Kita T. Angiotensin II, oxidative stress, and extracellular matrix degradation during transition to LV failure in rats with hypertension. J Mol Cell Cardiol 2006;41:989–97.
3. Cingolani HE, Plastino JA, Escudero EM, Mangal B, Brown J, Perez NG. The effect of xanthine oxidase inhibition upon ejection fraction in heart failure patients: La Plata study. J Card Fail 2006;12:491–8.
4. de Jong JW, Schoemaker RG, de Jonge R, Bernocchi P, Keijzer E, Harrison R, Sharma HS, Ceconi C. Enhanced expression and activity of xanthine oxidoreductase in the failing heart. J Mol Cell Cardiol 2000;32:2083–9.
5. Baldus S, Mullerleile K, Chumley P, Steven D, Rudolph V, Lund GK, Staude H, Stork A, Koster R, Kahler J, Weiss C, Munzel T, Meinertz T, Freeman BA, Heitzer T. Inhibition of xanthine oxidase improves myocardial contractility in patients with ischemic cardiomyopathy. Free Radic Biol Med 2006;41:1282–8.
6. Ekelund UE, Harrison RW, Shokek O, Thakkar RN, Tunin RS, Senzaki H, Kass DA, Marban E, Hare JM. Intravenous allopurinol decreases myocardial oxygen consumption and increases mechanical efficiency in dogs with pacing-induced heart failure. Circ Res 1999;85:437–45.
7. Tsutsui H, Ide T, Shiomi T, Kang D, Hayashidani S, Suematsu N, Wen J, Utsumi H, Hamasaki N, Takeshita A. 8-oxo-dGTPase, which prevents oxidative stress-induced DNA damage, increases in the mitochondria from failing hearts. Circulation 2001;104:2883–5.
8. Watanabe E, Matsuda N, Shiga T, Kajimoto K, Ajiro Y, Kawarai H, Kasanuki H, Kawana M. Significance of 8-Hydroxy-2′-Deoxyguanosine levels in patients with idiopathic dilated cardiomyopathy. J Card Fail 2006;12:527–32.
9. Dillmann WH, Mestril R. Heat shock proteins in myocardial stress. Z Kardiol 1995;84 Suppl 4:87–90.
10. Nakano M, Mann DL, Knowlton AA. Blocking the endogenous increase in HSP 72 increases susceptibility to hypoxia and reoxygenation in isolated adult feline cardiocytes. Circulation 1997;95:1523–31.
11. Latchman DS. Heat shock proteins and cardiac protection. Cardiovasc Res 2001;51:637–46.
12. Williams RS, Thomas JA, Fina M, German Z, Benjamin IJ. Human heat shock protein 70 (hsp70) protects murine cells from injury during metabolic stress. J Clin Invest 1993;92:503–8.
13. Zhu J, Quyyumi AA, Wu H, Csako G, Rott D, Zalles-Ganley A, Ogunmakinwa J, Halcox J, Epstein SE. Increased serum levels of heat shock protein 70 are associated with low risk of coronary artery disease. Arterioscler Thromb Vasc Biol 2003;23:1055–9.
14. Tanonaka K, Yoshida H, Toga W, Furuhama K, Takeo S. Myocardial heat shock proteins during the development of heart failure. Biochem Biophys Res Commun 2001;283:520–5.
15. Tanonaka K, Furuhama KI, Yoshida H, Kakuta K, Miyamoto Y, Toga W, Takeo S. Protective effect of heat shock protein 72 on contractile function of perfused failing heart. Am J Physiol Heart Circ Physiol 2001;281:H215–22.
16. Genth-Zotz S, Bolger AP, Kalra PR, von Haehling S, Doehner W, Coats AJS, Volk H, Anker SD. Heat shock protein 70 in patients with chronic heart failure: relation to disease severity and survival. Int J Cardiol 2004;96:397–401.
17. Knowlton AA, Kapadia S, Torre-Amione G, Durand J-B, Bies R, Young J, Mann DL. Differential expression of heat shock proteins in normal and failing human hearts. J Mol Cell Cardiol 1998;30:811–8.
18. Yamaguchi T, Shioji I, Sugimoto A, Yamaoka M. Psychological stress increases bilirubin metabolites in human urine. Biochem Biophys Res Comm 2002;293:517–20.
19. Denollet J, Pedersen SS, Vrints CJ, Conraads VM. Usefulness of Type D personality in predicting five-year cardiac events above and beyond concurrent symptoms of stress in patients with coronary heart disease. Am J Cardiol 2006;97:970–3.
20. Denollet J, Brutsaert DL. Personality, disease severity, and the risk of long-term cardiac events in patients with a decreased ejection fraction after myocardial infarction. Circulation 1998;97:167–73.
21. Schiffer AA, Smith ORF, Pedersen SS, Widdershoven JW, Denollet J. Type D personality and cardiac mortality in patients with chronic heart failure [published online ahead of print January 20, 2009]. Int J Cardiol.
22. Conraads VM, Denollet J, De Clerck LS, Stevens WJ, Bridts C, Vrints CJ. Type D personality is associated with increased levels of tumour necrosis factor (TNF)-alpha and TNF-alpha receptors in chronic heart failure. Int J Cardiol 2006;113:34–8.
23. Heimbach JK, Reznikov LL, Calkins CM, Robinson TN, Dinarello CA, Harken AH, Meng X. TNF receptor I is required for induction of macrophage heat shock protein 70. Am J Physiol Cell Physiol 2001;281:C241–7.
24. Meng X, Harken AH. The interaction between Hsp70 and TNF-alpha expression: a novel mechanism for protection of the myocardium against post-injury depression. Shock 2002;17:345–53.
25. Cernak I, Savic V, Kotur J, Prokic V, Kuljic B, Grbovic D, Veljovic M. Alterations in magnesium and oxidative status during chronic emotional stress. Magnes Res 2000;13:29–36.
26. Bilici M, Efe H, Koroglu MA, Uydu HA, Bekaroglu M, Deger O. Antioxidative enzyme activities and lipid peroxidation in major depression: alterations by antidepressant treatments. J Affect Disord 2001;64:43–51.
27. Khanzode SD, Dakhale GN, Khanzode SS, Saoji A, Palasodkar R. Oxidative damage and major depression: the potential antioxidant action of selective serotonin re-uptake inhibitors. Redox Rep 2003;8:365–70.
28. Srivastava N, Barthwal MK, Dalal PK, Agarwal AK, Nag D, Seth PK, Srimal RC, Dikshit M. A study on nitric oxide, β-adrenergic receptors and antioxidant status in the polymorphonuclear leukocytes from the patients of depression. J Affect Disord 2002;72:45–52.
29. Ozcan ME, Gulec M, Ozerol E, Polat R, Akyol O. Antioxidant enzyme activities and oxidative stress in affective disorders. Int Clin Psychopharmacol 2004;19:89–95.
30. Gidron Y, Russ K, Tissarchondou H, Warner J. The relation between psychological factors and DNA-damage: a critical review. Biol Psychol 2006;72:291–304.
31. Swynghedauw B. Molecular mechanisms of myocardial remodeling. Physiol Rev 1999;79:215–62.
32. Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, Komajda M, Tayazzi L, Smiseth OA, Gavazzi A, Haverich A, Hoes A, Jaarsma T, Korewicki J, Lévy S, Linde C, Lopez-Sendon JL, Nieminen MS, Piérard L, Remme WJ, Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): The Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J 2005;26:1115–40.
33. Denollet J. DS14: standard assessment of negative affectivity, social inhibition, and Type D personality. Psychosom Med 2005;67:89–97.
34. Beck AT, Steer RA. Manual for the Revised Beck Depression Inventory. San Antonio: Psychological Corporation; 1993.
35. Beck AT, Steer RA, Garbin MC. Psychometric properties of the Beck Depression Inventory: twenty-five years of evaluation. Clin Psychol Rev 1988;8:77–100.
36. Frasure-Smith N, Lesperance F, Juneau M, Talajic M, Bourassa MG. Gender, depression, and one-year prognosis after myocardial infarction. Psychosom Med 1999;61:26–37.
37. World Health Organization. Composite International Diagnostic Interview (CIDI). Geneva: WHO; 1990.
38. American Psychiatric Society. Diagnostic and Statistical manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Society; 1994.
39. Wittchen HU. Reliability and validity studies of the WHO–Composite International Diagnostic Interview (CIDI): a critical review. J Psychiatr Res 1994;28:57–84.
40. Kilic T, Ural D, Ural E, Yumuk Z, Agacdiken A, Sahin T, Kahraman G, Kozdag G, Vural A, Komsuoglu B. Relation between proinflammatory to anti-inflammatory cytokine ratios and long-term prognosis in patients with non-ST elevation acute coronary syndrome. Heart 2006;92:1041–6.
41. Heeschen C, Dimmeler S, Hamm CW, Fichtlscherer S, Boersma E, Simoons ML, Zeiher AM, CAPTURE Study Investigators. Serum level of the antiinflammatory cytokine interleukin-10 is an important prognostic determinant in patients with acute coronary syndromes. Circulation 2003;107:2109–14.
42. Kaur K, Sharma AK, Singal PK. Significance of changes in TNF-alpha and IL-10 levels in the progression of heart failure subsequent to myocardial infarction. Am J Physiol Heart Circ Physiol 2006;291:H106–13.
43. Slamon ND, Pentreath VW. Antioxidant defense against antidepressants in C6 and 1321N1 cells. Chem Biol Interact 2000;127:181–99.
44. Herken H, Gurel A, Selek S, Armutcu F, Ozen ME, Bulut M, Kap O, Yumru M, Savas HA, Akyol O. Adenosine deaminase, nitric oxide, superoxide dismutase, and xanthine oxidase in patients with major depression: impact of antidepressant treatment. Arch Med Res 2007;38:247–52.
45. Niedowicz D, Daleke D. The role of oxidative stress in diabetic complications. Cell Biochem Biophys 2005;43:289–330.
46. Warnholtz A, Mollnau H, Oelze M, Wendt M, Münzel T. Antioxidants and endothelial dysfunction in hyperlipidemia. Curr Hypertens Rep 2001;3:53–60.
47. Demirbag R, Yilmaz R, Erel O, Gultekin U, Asci D, Elbasan Z. The relationship between potency of oxidative stress and severity of dilated cardiomyopathy. Can J Cardiol 2005;21:851–5.
48. Kumar A, Paladugu B, Mensing J, Kumar A, Parrillo JE. Nitric oxide-dependent and -independent mechanisms are involved in TNF-alpha-induced depression of cardiac myocyte contractility. Am J Physiol Regul Integr Comp Physiol 2007;292:R1900–6.
49. Moorman AJ, Mozaffarian D, Wilkinson CW, Lawler RL, McDonald GB, Crane BA, Spertus JA, Russo JE, Stempien-Otero AS, Sullivan MD, Levy WC. In patients with heart failure elevated soluble TNF-receptor 1 is associated with higher risk of depression. J Card Fail 2007;13:738–43.
50. Redwine LS, Mills PJ, Hong S, Rutledge T, Reis V, Maisel A, Irwin MR. Cardiac-related hospitalization and/or death associated with immune dysregulation and symptoms of depression in heart failure patients. Psychosom Med 2007;69:23–9.
51. Denollet J, Schiffer AA, Kwaijtaal M, Hooijkaas H, Hendriks EH, Widdershoven JW, Kupper N. Usefulness of Type D Personality and kidney dysfunction as predictors of interpatient variability in inflammatory activation in chronic heart failure. Am J Cardiol 2009;103:399–404.
52. Denollet J, Vrints CJ, Conraads VM. Comparing Type D personality and older age as correlates of tumor necrosis factor-alpha dysregulation in chronic heart failure. Brain Behav Immun 2008;22:736–43.
53. Doulias P-T, Kotoglou P, Tenopoulou M, Keramisanou D, Tzavaras T, Brunk U, Galaris D, Angelidis C. Involvement of heat shock protein-70 in the mechanism of hydrogen peroxide-induced DNA damage: the role of lysosomes and iron. Free Radic Biol Med 2007;42:567–77.
54. Baek SH, Kim JY, Choi JH, Park EM, Han MY, Kim CH, Ahn YS, Park YM. Reduced glutathione oxidation ratio and 8-OHdG accumulation by mild ischemic pretreatment. Brain Res 2000;856:28–36.
55. Molloy GJ, Perkins-Porras L, Strike PC, Steptoe A. Type D personality and cortisol in survivors of acute coronary syndrome. Psychosom Med 2008;70:863–8.
56. Schiffer AA, Pedersen SS, Widdershoven JW, Denollet J. Type D personality and depressive symptoms are independent predictors of impaired health status in chronic heart failure. Eur J Heart Fail 2008;10:802–10.
57. Denollet J, de Jonge P, Kuyper A, Schene AH, van Melle JP, Ormel J, Honig A. Depression and Type D personality represent different forms of distress in the Myocardial INfarction and Depression Intervention Trial (MIND-IT). Psychol Med 2009;39:749–56.
58. de Jonge P, Denollet J, van Melle JP, Kuyper A, Honig A, Schene AH, Ormel J. Associations of type D personality and depression with somatic health in myocardial infarction patients. J Psychosom Res 2007;63:477–82.
heat shock proteins; oxidative stress; DNA damage; Type D personality; chronic heart failure; depression
Copyright © 2009 by American Psychosomatic Society
Highlight selected keywords in the article text.