You are here

Article Text

Article menu
Heart failure and cardiomyopathies
Original article
Cortisol awakening and stress response, personality and psychiatric profiles in patients with takotsubo cardiomyopathy
  1. Sabrina Kastaun1,2,
  2. Niko P Schwarz1,
  3. Martin Juenemann1,3,
  4. Mesut Yeniguen1,3,
  5. Holger M Nef4,
  6. Helge Moellmann5,
  7. Christian W Hamm4,5,
  8. Gebhard Sammer6,
  9. Juergen Hennig7,
  10. Georg Bachmann1,8,
  11. Tibo Gerriets1,3
  1. 1Department of Neurology, Heart & Brain Research Group, University Hospital Giessen and Marburg, Giessen, Germany
  2. 2Department of Heart Surgery, Kerckhoff Heart and Thorax Center, Bad Nauheim, Germany
  3. 3Department of Neurology, Buergerhospital Friedberg, Friedberg, Germany
  4. 4Department of Cardiology, University Hospital Giessen and Marburg, Giessen, Germany
  5. 5Department of Cardiology, Kerckhoff Heart and Thorax Center, Bad Nauheim, Germany
  6. 6Department of Psychiatry, University Hospital Giessen and Marburg, Giessen, Germany
  7. 7Department of Psychology, Personality Psychology and Individual Differences, Justus Liebig University, Giessen, Germany
  8. 8Department of Radiology, Kerckhoff Heart and Thorax Center, Bad Nauheim, Germany
  1. Correspondence to Professor Tibo Gerriets, Department of Neurology, Heart & Brain Research Group, University Hospital Giessen and Marburg, Klinikstraße 33, Giessen 35392, Germany; tibo.gerriets{at}neuro.med.uni-giessen.de

Abstract

Objective Alterations in cortisol awakening and stress responses (CAR, CSR) are sensitive markers for the basal activity and responsiveness of the hypothalamus–pituitary–adrenal axis (HPAA) in psychopathological conditions. We investigated whether patients with takotsubo cardiomyopathy (TTC) differ in these markers when compared with non-ST-segment elevation myocardial infarction (NSTEMI) patients and healthy controls.

Methods 19 female TTC patients were compared with 20 female NSTEMI patients and with 20 healthy women, matched by age and index event date. Salivary sampling indicated cortisol release, questionnaires assessed personality, life events, chronic stress and psychiatric symptoms.

Results The groups did not differ relevantly in their basal HPAA activity, psychiatric or personality profiles. Despite increased heart rates in response to stress (median difference (MDdiff)=3.5, p=0.002) and higher nervousness scores (MDdiff=−3.0, p=0.024), TTC patients revealed a blunted CSR with a medium effect compared to the controls (MDdiff=−3.2 nmol/L, p=0.022, r=0.36); even when controlled for prestress cortisol differences (p=0.044, r=0.33). In comparison with NSTEMI patients, no significant differences in CSR (MDdiff=−1.9 nmol/L, p=0.127, r=0.25) or nervousness (MDdiff=2.0, p=0.107) can be observed. Stressful life events, for example, traumatic experiences, occurred more often in TTC (42%) than in NSTEMI patients and controls (both 10%, p=0.031).

Conclusions In this small exploratory trial, a trend for a blunted CSR and high incidences of stressful life events were observed in TTC patients. If these results can be confirmed in larger studies, chronic stress and the inhibitory influence of cortisol on catecholamine release might be significant for the pathogenesis of TTC.

Keywords
  • cortisol stress response
  • personality
  • psychiatric disorders

https://doi.org/10.1136/heartjnl-2014-305745

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Keywords

    Introduction

    The aetiology of takotsubo cardiomyopathy (TTC) is still poorly understood. The most established theory of catecholamine-mediated myocardial stunning being provoked by emotional or physiological stress is supported by supraphysiological levels of plasma catecholamines in TTC patients.1 ,2 However, no stressful trigger is observed in approximately one-third of the patients.3 TTC predominantly occurs in postmenopausal women, and recent studies suggest a predisposing influence of psychiatric disorders, psychosocial stress4 and type D personality5 in its pathogenesis.

    Excessive or inadequate basal activity or responsiveness of the hypothalamus–pituitary–adrenal axis (HPAA) are sensitive indicators for health and psychological conditions. Free salivary cortisol is released in response to a stressor, with its peak after ∼20 min.6 It also follows a circadian rhythm, as it increases after awakening and continuously declines throughout the day.7 Whereas cortisol stress response (CSR) indicates HPAA reactivity, cortisol awakening response (CAR) provides insight into the basal physiological functioning of the HPAA. Altered CAR profiles are associated with depression8 ,9 and hyporesponsiveness of the HPAA has often been reported in stress-related disorders.7 ,10 ,11

    The present hypotheses-generating study investigates basal HPAA activity and HPAA responsiveness in TTC patients for the first time. The results were contrasted to those of patients with myocardial infarction and healthy controls. Moreover, we aimed to identify predisposing psychiatric and personality pattern, as well as stressful life events, that may be linked to altered cortisol profiles in patients with TTC.

    Methods

    Study population

    Data were collected from 19 female TTC patients diagnosed at Kerckhoff Heart and Thorax Center, Bad Nauheim, Germany. Inclusion criteria were in line with the Mayo Clinic criteria.3 Twenty women with a history of non-ST-segment elevation myocardial infarction (NSTEMI) and 20 cardiac-healthy female volunteers (eg, recruited via announcements from community facilities) served as control groups. All gave written, informed consent. The groups were matched by age (mean age 60.5±9.2), and the TTC and NSTEMI patients were additionally matched by their index event date in relation to study inclusion (mean months 18.4±8.5). We recorded the most prominent pharmaceuticals (if used by more than two patients) that are known to influence the physiological stress response.12 In advance, one TTC and two NSTEMI patients were excluded due to oestrogen therapy, and three TTC and seven NSTEMI patients refused to participate in the study. None of our participants took oral contraceptives or corticoids. The ethics committee of the Justus-Liebig-University Giessen approved the study.

    Measurements

    Stress hormones

    Saliva cortisol is a useful, non-invasive biomarker which reflects the HPAA activity and facilitates multi-sampling. Its concentrations are closely correlated to serum cortisol, when controlled for specific variables such as sex steroids.13

    Catecholamines are known to reflect the sympathetic activity and are elevated by psychological stressors. However, its concentrations in saliva seem to be a poor index of acute changes in sympathetic activity,14 and venipuncture itself can lead to activation of stress systems. Thus, we refrain from catecholamine measurements.

    Salivette sampling devices (Sarstedt, Rommelsdorf, Germany) were used for saliva collection. Samples were centrifuged at 3500 rpm for 5 min and frozen at −20°C until assayed. An ELISA determined free cortisol concentrations (ELISA, IBL International GmbH, Hamburg, Germany) according to standard procedures. The intraassay and interassay variability were <10%.

    Laboratory stress experiment

    The participants received introductions to study protocols, which spared details about the ‘stress experiment’, and were asked to provide their first salivary sample on the afternoon prior to the lab session. We advised them not to eat, drink (except non-carbonated water), or smoke 2 h prior to the samplings. Individual examinations were performed between 14:00 and 16:00. To reduce initial excitement, all participants listened to relaxation exercises for 15 min. The 15 min stress induction phase followed, always conducted by the same psychologist. Our self-designed stress experiment (see online supplemental data) included all aspects known to provoke robust stress responses.6

    Saliva was collected at 15 min intervals: at arrival, after relaxation, after stress and over the following 45 min while the subjects had to fill out the psychometric questionnaires. Heart rates and nervousness were measured simultaneously. Finally, the participants were told about the stress experiment and received instructions for a single collection of a morning saliva profile at home.

    Cortisol awakening response

    Two TTC and four NSTEMI patients refused to return their samples. Saliva was collected immediately after awakening, as well as 15, 30, 45 and 60 min thereafter. We advised the subjects to collect their saliva on a quiet day and not to eat, drink or brush their teeth during the sampling period. Otherwise, they were free to follow their morning routines. All participants were asked to record their sleep duration, store their samples cool and return them promptly.

    Psychological assessment

    We investigated the previous histories and current statuses of psychiatric disorders by exploring the medical records and conducting direct interviews. Self-report questionnaires were used to assess psychiatric and somatic symptoms (Symptom Checklist-Revised, SCL-90-R),15 personality traits (Freiburger Personality Inventory-Revised, FPI-R),16 chronic distress within the 2 years before cardiac event (Trier Inventory for the Assessment of Chronic Stress, TICS)17 and current mood (‘calmness vs nervousness’ dimension, Multidimensional Mood State Questionnaire, MDMQ).18 Stressful life events were investigated using a shortened version of Tennant and Andrews's life-event inventory.19 Online supplementary table S1 provides detailed information about all psychometric dimensions.

    Statistical analysis

    Analysis of histograms and Quantil-Quantil-plots gave information about the data distribution. Shapiro-Wilk tests were used to verify if the normal distribution had to be rejected. Homogeneity of variance was tested using Levene tests. Group comparisons were planned a priori; the TTC group was contrasted against both control groups.

    Student t tests, or non-parametric Mann–Whitney tests, were used to analyse group differences for pairwise comparisons. For categorical variables, Pearson’s χ2 or Fisher’s exact test were used. Associations were examined using Pearson's correlation coefficient.

    To perform analysis of covariance (ANCOVA) or repeated measures ANOVA, heart rates, CAR and CSR data were natural log-transformed due to its positive skew. The repeated measures ANOVA was performed with Greenhouse–Geisser correction for the CAR data including five within-subject factors (awakening, 15, 30, 45 and 60 min).

    CSR was analysed initially by simple pairwise group contrasts for each poststress cortisol level (+15, +30, +45 min). To control the influence of group differences in prestress cortisol, the primary outcome parameter (cortisol level 15 min poststress) was additionally tested with ANCOVA by inserting the poststress cortisol level as the dependent variable, the prestress cortisol level (end of relaxation) as covariate and the group variable as the categorical factor. Pharmaceuticals were inserted as further covariates. To quantify the cortisol concentrations after awakening (awakening, 15, 30, 45 and 60 min) and during stress (end of relaxation, end of stress, +15, +30, +45 min) we calculated the area under the curve with respect to ground (AUCg) using the formula from Pruessner et al.20

    Group differences in heart rates were analysed in the same way as CSR. Additionally, paired-sample Wilcoxon test analysed the heart rates from prestress to poststress for each group.

    To compare the ratio between subjective and objective stress, MDMQ and cortisol data were z-transformed and adjusted; thus, greater scores represent higher stress levels. ‘Cortisol increase’ represents cortisol levels 15 min poststress minus cortisol levels prior to stress. MDMQ nervousness scores underwent the same procedure. Each group had its quotient calculated (▵ subjective stress/▵ objective stress).

    All p values reported are two-sided. The statistical significance was set at p<0.05. p Values of multiple testing were controlled using the false discovery rate from Benjamini and Hochberg. For the primary endpoints, effect sizes were calculated using Pearson’s correlation coefficient r.

    Results

    Clinical data, acute stressful trigger and life events

    Table 1 presents the clinical data. TTC patients showed fewer cardiovascular risk factors, such as hyperlipidaemia, arterial hypertension and tobacco use, than the NSTEMI patients. Heart attack reoccurred in two patients of both cardiac groups.

    View this table:
    Table 1

    Clinical characteristics

    At acute hospital admission, TTC patients had a significantly lower left ventricular EF than NSTEMI patients (median difference (MDdiff)=−20.0, 95% CIs −25.0 to −10.0), whereas the laboratory results from the NSTEMI patients showed higher concentrations of creatine kinase (MDdiff=−134.0, 95% CI −247.0 to −12.0) and creatine kinase-MB (MDdiff=−11, 95% CI −30.0 to −1.0).

    On the date of study inclusion, TTC patients received significantly less frequent hypolipidemic statins, thyroid hormones and salicylates than the NSTEMI patients. Seven control subjects were treated with β blockers as long-term medication for age-related primary hypertension without a history of coronary artery disease.

    Sociodemographic characteristics, acute stressful trigger and stressful life events are listed in table 2. Compared to the NSTEMI patients, TTC patients underwent significantly more often an acute stressful trigger prior to their heart attack. That trigger was frequently a severe familial or professional argument.

    View this table:
    Table 2

    Sociodemographic characteristics, acute trigger and life events

    The life-event inventory shows that TTC patients experienced spousal loss and severe, recurrent familial or professional arguments more frequently than the controls. More traumatic life events (domestic abuse, forced abortion, car accident with severe injury and spousal loss) happened to TTC patients when compared with NSTEMI patients and controls. The groups did not differ relevantly in current or prior psychotherapeutic treatment. However, TTC patients more often reported domestic abuse as their predominant motivation for therapy than the NSTEMI patients or controls did.

    Personality traits, psychiatric status and chronic stress

    The psychometric results appear in the online supplementary table S1. By trend, TTC patients differ from the controls in four subscales (FPI-R: greater emotional lability; SCL-90-R: higher anxiety; TICS: more social isolation and chronic worrying) and tend to be less aggressive than NSTEMI patients. However, group differences in personality, in psychiatric profiles and in the experience of chronic work and social hassles (TICS) were not statistically significant.

    Cortisol awakening response

    The repeated measures ANOVA revealed comparable cortisol release profiles (figure 1) between the TTC and NSTEMI patients (time×group: p=0.918, r=0.05, 95% CIeffect −0.29 to 0.38) and between TTC patients and controls (time x group: p=0.747, r=0.09, 95% CIeffect −0.24 to 0.40). Group contrasts in the overall postawakening cortisol release (AUCg), and at each measurement, confirmed these results (table 3).

    Figure 1
    Figure 1

    Left: salivary cortisol profile (mean±SD) at awakening (0), 15, 30, 45 and 60 min after awakening. Right: salivary cortisol release in nanomoles per liter in response to stress (M±SD). 15, 30 and 45 min after stress exposition, takotsubo cardiomyopathy (TTC) patients showed a lower cortisol release than controls *(p<0.05).

    View this table:
    Table 3

    Cortisol data

    Stress response

    Cortisol data

    After α-correction (p<0.025), TTC patients tended to a significantly lower overall cortisol release (AUCg) than the controls (MDdiff=−10.3 nmol/L, 95% CI −21.3 to −0.8, r=0.35, 95% CIeffect 0.04 to 0.60; table 3, figure 1). No statistically relevant difference in the overall cortisol release was found in comparison with the NSTEMI patients (MDdiff=−5.8 nmol/L, 95% CI −17.4 to 3.7, r=0.23, 95% CIeffect 0.09 to 0.51).

    In response to the stressor (table 3, figure 1), the TTC patients showed significantly lower cortisol levels than the controls 15 min (MDdiff=−3.2 nmol/L, 95% CI −6.4 to −0.7, r=0.36, 95% CIeffect 0.05 to 0.61) and 45 min poststress (MDdiff=−2.9 nmol/L, 95% CI −4.9 to −0.8, r=0.38, 95% CIeffect 0.07 to 0.62), and a trend 30 min poststress (MDdiff=−2.6 nmol/L, 95% CI −5.4 to 0.3, r=0.29, 95% CIeffect −0.03 to 0.55). Even when controlled for non-significant group differences in the prestress cortisol level (covariate: prestress cortisol, dependent variable: cortisol 15 min poststress; estimate of the difference (Ediff)=−0.252, 95% CI −0.498 to −0.007, p=0.044, r=0.33, 95% CIeffect 0.02 to 0.58) and for the presence of β blockers (Ediff=−0.248, 95% CI −0.505 to 0.009, p=0.053, r=0.31, 95% CIeffect −0.06 to 0.57) a trend towards a significant reduced stress response with medium effect sizes could be observed. No statistically relevant differences were found in comparison with the NSTEMI patients in response to the stressor at 15 min (MDdiff=−1.9 nmol/L, 95% CI −5.4 to 0.6, r=0.25, 95% CIeffect −0.07 to 0.52), 30 min (MDdiff=−1.8 nmol/L, 95% CI −4.2 to 0.9, r=0.20, 95% CIeffect −0.12 to 0.48) and 45 min poststress (MDdiff=−1.6 nmol/L, 95% CI −3.9 to 0.6, r=0.25, 95% CIeffect −0.07 to 0.52, see also table 3), even when controlled for prestress group differences (Ediff=−0.151, 95% CI −0.415 to 0.114, p=0.255, r=0.20, 95% CIeffect −0.123 to 0.48) and the presence of β blockers, hypolipidemic statins, thyroid hormones or salicylates.

    Heart rate

    For each group, a significant increase in heart rates (bpm) from prestress to poststress could be observed (figure 2), which confirmed the efficiency of the stress experiment: TTC (median,IQR) (66.0, 59.0–78.0 vs 70.0, 65.0–84.0; MDdiff=3.5, 95% CI 1.0 to 6.5, p=0.002), NSTEMI (66.0, 57.0–79.0 vs 69.0, 59.0–80.0; MDdiff=2.0, 95% CI 0.5 to 3.5, p=0.014) and controls (68.0, 56.5–73.0 vs 72.0, 59.0–79.0; MDdiff=4.0, 95% CI 2.0 to 5.0, p=0.004). No relevant group differences could be found in pairwise comparisons (all p>0.20), even when controlled for the prestress heart rates and the use of above-named pharmaceuticals.

    Figure 2
    Figure 2

    Left: Mood scores (M±SD) during the examination. Lower scores represent increased nervousness. Takotsubo cardiomyopathy (TTC) patients felt significantly more nervous than controls at arrival, poststress and at the end of examination (*p<0.05, **p<0.01). Right: Heart rates in beats per minute in response to stress (M±SD). A significant increase (*p<0.05, **p<0.01) of from relaxation to stress could be observed for each group.

    Mood

    In figure 2, lower scores in the MDMQ represent increased nervousness. At arrival (median, IQR) (12.0, 10.0–15.0), poststress (10.0, 7.0–13.0) and at the end of the examination (16.0, 12.0–18.0), TTC patients felt greater nervousness than the controls (arrival: 17.0, 14.0–19.0, MDdiff=−4.0, 95% CI −7.0 to −2.0, p=0.001; poststress: 12.0, 10.3–16.5; MDdiff=−3.0, 95% CI −5.0 to 0.0, p=0.024; end: 18.0, 16.3–20.0; MDdiff=−2.0, 95% CI −5.0 to −1.0, p=0.006). The comparison of TTC and NSTEMI patients revealed no significant differences (arrival: 15.0, 12.0–16.8, MDdiff=−2.0, 95% CI −4.0 to 0.0, p=0.101; poststress: 12.0, 9.0–15.0; MDdiff=−2.0, 95% CI −5.0 to 0.0, p=0.107; end: 16.5, 15.3–18.0; MDdiff=−1.0, 95% CI −3.0 to 1.0, p=0.283).

    Subjective versus objective stress

    The mean ratio between subjective and objective stress response (▵ ‘nervousness’/▵ cortisol) in TTC patients was 1.9±6.1 SD, with no significant difference when contrasted with the NSTEMI patients (−1.8±13.4 SD) but with a trend towards a significant larger ratio when contrasted with the controls (−2.3±8.4 SD; mean difference=4.19, 95% CI −0.64 to 9.02, p=0.087).

    Discussion

    Psychological approaches to reveal the aetiology of TTC are sparse and reveal conflicting results.4 ,21 In the present study, TTC patients were screened in detail for predisposing attributes, but our findings neither lend support for psychiatric or personality characteristics nor for an increased occurrence of general social or work hassles. Moreover, TTC patients seem to have a basal HPAA activity comparable to NSTEMI patients and cardiac-healthy controls. Our results hereby confirm the findings of Madhavan and colleagues,22 who did not find basal urinary cortisol levels to be elevated in TTC patients on hospital days 1–3. However, because of the short half-life of cortisol, no conclusions about cortisol in response to the stressful trigger event can be drawn from these data. When compared with controls, our data indicate a blunted CSR and a high incidence of uncontrollable stressful life events (severe, recurrent conflicts and trauma experience: predominantly childhood or domestic abuse) in TTC patients, which was aligned with an increased motivation for psychotherapeutic consulting. Could the influence of long-term stress and depressed cortisol stress levels contribute to the pathogenesis of TTC?

    Chronic stress, blunted cortisol and its potential role in TTC

    Stress response plays a key role in the adaptation of an organism to environmental stressors. Hyporesponsiveness (hypocortisolism) of the HPAA has been reported frequently in patients with stress-related diseases, such as post-traumatic stress disorder (PTSD),10 chronic fatigue syndrome11 and fibromyalgia.7 Hellhammer and Wade23 suggested that a hyporeactive HPAA may develop as a result of prolonged stress periods with a HPAA activation and excessive glucocorticoid release. In rats, prolonged periods of stress or morphine treatment initially lead to a hyper-reactive HPAA, whereas animals subsequently develop hypocortisolism in response to stress,24 but noticeably, not under basal conditions.25

    Hypocortisolism has a known impact on the sympathetic nervous system, especially considering an overactivity. Findings in adrenalectomied rats have demonstrated that endogenous glucocorticoids suppress catecholamine turnover and release under stress.26 Studies in humans have shown inhibitory effects of glucocorticoids on norepinephrine release.27

    In the present study, TTC patients seem to differ from controls in their appraisal of stressful situations. They felt more nervous during stress but apparently had lower cortisol levels. Cortisol is known to have a protective influence on situational appraisals during stress (mood-buffering cortisol effect).28 Might blunted cortisol increase the negative appraisal of the situation that triggers TTC and leads to hyperarousal?

    This hypothesis could be of particular interest regarding the excessive catecholamine release,1 aligned with typical histological signs of catecholamine toxicity and specific alterations in Ca2+-regulatory proteins which might be crucial for the contractile dysfunction in TTC.2 The influence of mental stress on left ventricular wall motion abnormalities and perfusion defects has already been demonstrated in TTC patients.29

    Limitations

    Our results may be limited by the small sample size and, therefore, should be interpreted with caution. However, moderate effect sizes (r=0.20–0.38) were found for the primary endpoint, suggesting that studies with adequate power levels might negotiate the lack of highly significant results, and expose whether the deviating means in contrast with NSTEMI patients, can reach statistical significance.

    It could be discussed, if NSTEMI, instead of STEMI patients are the right comparator. We aimed to identify psychological and hormonal parameters in TTC patients contrasted with patients who have made an equivalent experience of a heart attack. To our knowledge, no differences had been described so far between patients with or without ST-segment elevation in this regard. Recent psychological studies in TTC patients included acute coronary syndrome patients or a mixed myocardial infarction group as cardiac controls.4 ,5 Due to our small sample size, we believe, that one of the strengths of our study is the inclusion of homogeneous cardiac and non-cardiac control groups.

    This is a retrospective analysis, so we do not know whether the psychometric ratings would have been similar if the assessments had taken place during the in-hospital stay. Though, at least the stability of personality traits in adulthood is a widely accepted assumption, and intraindividual changes are generally unrelated to adverse life events.30

    Conclusion

    Despite the small sample size, a trend for a blunted cortisol stress response and high incidences of stressful life events can be observed in TTC patients. If our results can be confirmed in larger studies, chronic stress, the inhibitory influence of cortisol on catecholamine release and its mood-buffering effects during stress might be significant for the pathogenesis of TTC.

    Key messages

    What is known on this subject?

    • Sympatheticallly mediated myocardial damage has been proposed as causative mechanism of takotsubo cardiomyopathy (TTC) that is often triggered by a stressful event. In neuroendocrine research, blunted cortisol stress response due to chronic stress and the inhibitory influence of cortisol on catecholamine release are familiar.

    What might this study add?

    • TTC patients showed a trend for a blunted cortisol stress response and had impressive incidences of stressful or traumatic life events.

    How might this impact on clinical practice?

    • If our findings can be confirmed in larger studies, they may provide new avenues for an interdisciplinary research in TTC.

    References

    1. ↵
      1. Wittstein IS,
      2. Thiemann DR,
      3. Lima JA,
      4. et al
      . Neurohumoral features of myocardial stunning due to sudden emotional stress. N Engl J Med 2005;352:539–48.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Nef HM,
      2. Mollmann H,
      3. Troidl C,
      4. et al
      . Abnormalities in intracellular Ca2+ regulation contribute to the pathomechanism of Tako-Tsubo cardiomyopathy. Eur Heart J 2009;30:2155–64.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Madhavan M,
      2. Prasad A
      . Proposed Mayo Clinic criteria for the diagnosis of Tako-Tsubo cardiomyopathy and long-term prognosis. Herz 2010;35:240–3.
      OpenUrlCrossRefPubMed
    4. ↵
      1. Delmas C,
      2. Lairez O,
      3. Mulin E,
      4. et al
      . Anxiodepressive disorders and chronic psychological stress are associated with Tako-Tsubo cardiomyopathy- New Physiopathological Hypothesis. Circ J 2013;77:175–80.
      OpenUrlCrossRefPubMed
    5. ↵
      1. Compare A,
      2. Bigi R,
      3. Orrego PS,
      4. et al
      . Type D personality is associated with the development of stress cardiomyopathy following emotional triggers. Ann Behav Med 2013;45:299–307.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Dickerson SS,
      2. Kemeny ME
      . Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychol Bull 2004;130:355–91.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Riva R,
      2. Mork PJ,
      3. Westgaard RH,
      4. et al
      . Comparison of the cortisol awakening response in women with shoulder and neck pain and women with fibromyalgia. Psychoneuroendocrinology 2012;37:299–306.
      OpenUrlCrossRefPubMed
    8. ↵
      1. Huber TJ,
      2. Issa K,
      3. Schik G,
      4. et al
      . The cortisol awakening response is blunted in psychotherapy inpatients suffering from depression. Psychoneuroendocrinology 2006;31:900–4.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Stetler C,
      2. Miller GE
      . Blunted cortisol response to awakening in mild to moderate depression: regulatory influences of sleep patterns and social contacts. J Abnorm Psychol 2005;114:697–705.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Rohleder N,
      2. Joksimovic L,
      3. Wolf JM,
      4. et al
      . Hypocortisolism and increased glucocorticoid sensitivity of pro-Inflammatory cytokine production in Bosnian war refugees with posttraumatic stress disorder. Biol Psychiatry 2004;55:745–51.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Roberts AD,
      2. Wessely S,
      3. Chalder T,
      4. et al
      . Salivary cortisol response to awakening in chronic fatigue syndrome. Br J Psychiatry 2004;184:136–41.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Granger DA,
      2. Hibel LC,
      3. Fortunato CK,
      4. et al
      . Medication effects on salivary cortisol: tactics and strategy to minimize impact in behavioral and developmental science. Psychoneuroendocrinology 2009;34:1437–48.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Hellhammer DH,
      2. Wust S,
      3. Kudielka BM
      . Salivary cortisol as a biomarker in stress research. Psychoneuroendocrinology 2009;34:163–71.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Schwab KO,
      2. Heubel G,
      3. Bartels H
      . Free epinephrine, norepinephrine and dopamine in saliva and plasma of healthy adults. Eur J Clin Chem Clin Biochem 1992;30:541–4.
      OpenUrlPubMed
    15. ↵
      1. Derogatis LR
      . SCL-90-R, Symptom Checklist-90-R: administration, scoring and procedures manual. 3rd edn. Minneapolis, MN: NCS Pearson, 1994.
    16. ↵
      1. Fahrenberg J,
      2. Hampel R,
      3. Selg H
      . Das Freiburger Persönlichkeitsinventar (FPI-R). 7., überarb. u. neu norm. Aufl. ed. Göttingen: Hogrefe, 2001.
    17. ↵
      1. Schulz P,
      2. Schlotz W
      . The Trier Inventory for the Assessment of Chronic Stress (TICS): scale construction, statistical testing, and validation of the scale work overload. Diagnostica 1999;45:8–19.
      OpenUrlCrossRef
    18. ↵
      1. Steyer R
      . Der mehrdimensionale Befindlichkeitsfragebogen: (MDBF); Handanweisung. Göttingen [u.a.]: Hogrefe, 1997.
    19. ↵
      1. Tennant C,
      2. Andrews G
      . A scale to measure the stress of life events. Aust N Z J Psychiatry 1976;10:27–32.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Pruessner JC,
      2. Kirschbaum C,
      3. Meinlschmid G,
      4. et al
      . Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology 2003;28:916–31.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Del Pace S,
      2. Parodi G,
      3. Bellandi B,
      4. et al
      . Anxiety trait in patients with stress-induced cardiomyopathy: a case-control study. Clin Res Cardiol 2011;100:523–9.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Madhavan M,
      2. Borlaug BA,
      3. Lerman A,
      4. et al
      . Stress hormone and circulating biomarker profile of apical ballooning syndrome (Takotsubo cardiomyopathy): insights into the clinical significance of B-type natriuretic peptide and troponin levels. Heart 2009;95:1436–41.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Hellhammer DH,
      2. Wade S
      . Endocrine correlates of stress vulnerability. Psychother Psychosom 1993;60:8–17.
      OpenUrlCrossRefPubMed
    24. ↵
      1. Fries E,
      2. Hesse J,
      3. Hellhammer J,
      4. et al
      . A new view on hypocortisolism. Psychoneuroendocrinology 2005;30:1010–16.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Houshyar H,
      2. Cooper ZD,
      3. Woods JH
      . Paradoxical effects of chronic morphine treatment on the temperature and pituitary-adrenal responses to acute restraint stress: a chronic stress paradigm. J Neuroendocrinol 2001;13:862–74.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Kvetnansky R,
      2. Fukuhara K,
      3. Pacak K,
      4. et al
      . Endogenous glucocorticoids restrain catecholamine synthesis and release at rest and during immobilization stress in rats. Endocrinology 1993;133:1411–19.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Golczynska A,
      2. Lenders JW,
      3. Goldstein DS
      . Glucocorticoid-induced sympathoinhibition in humans. Clin Pharmacol Ther 1995;58:90–8.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Het S,
      2. Schoofs D,
      3. Rohleder N,
      4. et al
      . Stress-induced cortisol level elevations are associated with reduced negative affect after stress: indications for a mood-buffering cortisol effect. Psychosom Med 2012;74:23–32.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Sciagra R,
      2. Parodi G,
      3. Del Pace S,
      4. et al
      . Abnormal response to mental stress in patients with Takotsubo cardiomyopathy detected by gated single photon emission computed tomography. Eur J Nucl Med Mol Imaging 2010;37:765–72.
      OpenUrlCrossRefPubMed
    30. ↵
      1. Cobb-Clark DA,
      2. Schurer S
      . The stability of big-five personality traits. Economics Letters 2012;115:11–15.
      OpenUrlCrossRef

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Contributors All authors contributed substantially, read, and approved this paper.

    • Funding This work was supported by the German Heart Foundation/German Foundation of Heart Research, Frankfurt am Main, Germany.

    • Competing interests None.

    • Ethics approval Ethics committee of the Justus-Liebig-University Giessen, Germany.

    • Provenance and peer review Not commissioned; externally peer reviewed.