Tilburg University
Chronic stress exposure in men and women, and implications for the course of fatigue
after percutaneous coronary intervention
Doedee, Fleur; Van Den Houdt, Sophie; Widdershoven, Jos; Kupper, Nina
Published in:
General Hospital Psychiatry: Psychiatry, Medicine and Primary Care
DOI:
10.1016/j.genhosppsych.2021.07.001
Publication date:
2021
Document Version
Publisher's PDF, also known as Version of record
Link to publication in Tilburg University Research Portal
Citation for published version (APA):
Doedee, F., Van Den Houdt, S., Widdershoven, J., & Kupper, N. (2021). Chronic stress exposure in men and
women, and implications for the course of fatigue after percutaneous coronary intervention: The THORESCI
study. General Hospital Psychiatry: Psychiatry, Medicine and Primary Care, 72, 45-52.
https://doi.org/10.1016/j.genhosppsych.2021.07.001
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal
Take down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
General Hospital Psychiatry 72 (2021) 45–52
Available online 13 July 2021
0163-8343/© 2021 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Chronic stress exposure in men and women, and implications for the course
of fatigue after percutaneous coronary intervention; the THORESCI study
Fleur Doedee
a, Sophie van den Houdt
a, Jos Widdershoven
a,b, Nina Kupper
a,*aDepartment of Medical & Clinical Psychology, Center of Research on Psychology in Somatic Diseases, Tilburg University, Tilburg, the Netherlands bDepartment of Cardiology, Elisabeth-TweeSteden Hospital, Tilburg, the Netherlands
A R T I C L E I N F O Keywords:
myocardial infarction chronic stress recovery
acute coronary syndrome stable angina
A B S T R A C T
Background: Fatigue is a prevalent symptom in patients with coronary heart disease (CHD). Individual differences in chronic stress may affect the experience and persistence of fatigue, and this may vary between the sexes. Therefore, we studied the effect of chronic stress on the course of fatigue over a 2-year period after percutaneous coronary intervention (PCI), and examined the moderating effects of sex.
Methods: 1682 patients (78% men, age = 67.1 ± 10.6) were recruited and filled out multiple self-report ques-tionnaires at baseline, one, 12, and 24 months post-PCI, including questions on demographics, fatigue (HCS). Multiple chronic stressors were assessed at baseline: work stress (ERI16), marital stress (MMQ-6), early life events (Life Events Questionnaire) and social stress. Latent class factor analysis (LatentGOLD) was used to construct a comprehensive chronic stress index. Linear mixed modeling examined the predictive quality of predictors and covariates.
Results: Fatigue was found to substantially decrease over the first month post-PCI, then stabilized at a moderate level. Chronic stress impacted both the level and course of fatigue by increasing its level and delaying recovery. Overall and across time, women reported more fatigue than men. The level and course effects of chronic stress and sex were independent of demographic, health behavioral, and medical covariates.
Conclusions: Individual differences in chronic stress impact both the level and course of fatigue post-PCI, with women being affected most. Future research could further explain the mechanisms underlying the observed relationships. Developing and testing interventions focusing on exercise and stress-reduction could be used to alleviate fatigue.
1. Introduction
In spite of major advancements in treatment options, coronary heart disease (CHD) is a leading public health burden and a major cause of death [1], with women’s risk being similar or worse than men’s despite differences in pathophysiology [2]. Besides classic cardiovascular risk factors, psychosocial characteristics convey increased risk for disease progression and mortality [3], among which, chronic stress [4,5]. A variety of chronic stressors, such as work stress [6,7], marital stress [6,8–10], early life events [11], and social stress [6,12,13] have shown to contribute to the pathogenesis and relative recurrent risk of CHD. Chronic stressors seem to promote CHD incidence and progression through health behavioral (e.g., smoking, physical inactivity; [14]), and biological pathways (e.g., increasing blood pressure and neurohormonal
arousal [6,15], promoting atherosclerotic process [8–10]). Stress, early trauma, and stress-related disorders have an almost double prevalence in women [16], and, as there is a clearly stress-related increased risk of CHD, women seem particularly vulnerable.
Illness-related fatigue is one of the most disturbing and long-lasting symptoms experienced by patients with CHD [17–20]. Elevated levels of fatigue have been associated with the progression of heart disease [21], and negatively affect patients’ quality of life [19,21,22]. Chronic stress has been recognized as a predisposing and perpetuating factor of fatigue through biological mechanisms, including continued activation of the HPA axis, and increased inflammation [23]. In the general pop-ulation, research shows early life events to be prospectively associated with increased fatigue [24]. An experience sampling study in healthy students further showed an equivalent, bidirectional association * Corresponding author at: Tilburg University, Department of Medical & Clinical Psychology, Center of Research on Psychology in Somatic diseases, PO box 90153, 5000LE Tilburg, the Netherlands.
E-mail address: h.m.kupper@tilburguniversity.edu (N. Kupper).
Contents lists available at ScienceDirect
General Hospital Psychiatry
journal homepage: www.elsevier.com/locate/genhospsych
https://doi.org/10.1016/j.genhosppsych.2021.07.001
General Hospital Psychiatry 72 (2021) 45–52
46
between stress and fatigue [25]. Importantly, sex differences exist, as a meta-analysis on studies in patients with acute myocardial infarction (MI; [26]) showed women to be more likely to report fatigue symptoms than men. A paucity of studies has examined the trajectories of fatigue after MI. Studies that did, showed that after MI, fatigue decreases slowly [18,20]. In patients with heart disease, the association between fatigue and chronic stress, and sex differences therein, has not been studied.
In sum, given the importance of fatigue as a long-term symptom in heart disease, and the potential role of chronic stress as a perpetuating factor, the current study aimed to examine the extent to which chronic stress explained individual differences in the course of fatigue over 2 years after percutaneous coronary intervention (PCI) for acute coronary syndrome (ACS). Importantly, as fatigue and psychosocial risk factors seem to claim a more prominent role in female patients, we examined the main effects and moderating role of sex.
We hypothesized that, based on prior cross-sectional [24] and experience sampling [25] findings from the general population, PCI patients with higher scores on chronic stress would have increased and more persistent levels of fatigue symptoms over the 2 years of follow-up, while the general tendency would be one of slowly reducing fatigue over the 2-year period [20]. Previous research found that women report more fatigue symptoms and have the tendency to appraise stressors as more severe in comparison to men [21,27]. We thus hypothesized that women would experience increased fatigue symptoms due to chronic stress. 2. Method
2.1. Participants and procedure – THORESCI study
The current study was part of a large prospective and ongoing cohort study, Tilburg Health Outcomes Registry of Emotional Stress after Cor-onary Intervention (THORESCI). The THORESCI study recruited par-ticipants at the Elisabeth-TweeSteden Hospital in Tilburg. All patients were enrolled in the clinical standard of care Percutaneous Coronary Intervention (PCI) registry of the hospital. Patients were included when they were scheduled for an elective or acute PCI for ≥1 coronary oc-clusions, were aged ≥18 and had sufficient understanding of the Dutch language to fill out the questionnaires. Patients were excluded when they had a life-threatening comorbidity (e.g., metastasized cancer), a severe cognitive disorder (i.e., dementia or Alzheimer’s disease) and/or a disability that prevented them from filling out questionnaires (e.g., blindness). During the stay in the hospital after the PCI, patients were approached by a member of the research team. Patients were verbally explained the content and the requirements of the study and were additionally given an information letter. After giving written informed consent, patients were asked to fill out a set of validated psychosocial questionnaires, spread over five measurements moments post-PCI. I.e., immediately following PCI (within 1 week post-PCI; Inclusion (T0)), at 1 month (T1), at 6 months (T2), 1 year (T3), and 2 years (T4) thereafter. Participants could choose if they wanted to receive the questionnaires digitally (~60%) or on paper (~40%). For the current study, data from the T0, T1, T3 and T4 measurement occasions were used. The THORESCI study was approved by the institutional medical ethics review board and the study protocol was in line with the Helsinki declaration [28].
2.2. Materials and measures
2.2.1. Demographic and clinical variables
Demographic variables age, sex, and working status were obtained from the self-report questionnaires at inclusion. Education level (highest level attained) and marital status (in long-term relationship or not) were obtained from a self-report question at baseline or 1 month-follow up. Health behaviors (exercise, diet) were reported as well. Furthermore, clinical variables, i.e. cardiac history (heart failure, and previous MI, coronary artery bypass grafting (CABG) and PCI), cardiovascular risk factors (genetic risk, hypertension, hypercholesterolemia, and smoking),
PCI profile (acute vs. elective procedure), beta-blocker use, and comorbidities (cancer in past 5 years, COPD, rheumatoid arthritis, anemia, diabetes mellitus type 2, liver disease, kidney disease) were extracted from medical records at inclusion.
2.2.2. Chronic stress
The co-occurrence of a multitude of stressors [29], and the variety in stress sources/measures makes chronic stress a challenging construct to assess [29]. In previous studies, chronic stressors have mostly been studied as separate entities. Taking a similar approach as the allostatic load theory [30], i.e., combining physiological variables that reflect stressor-induced activation, a comprehensive chronic stress measure may be derived by statistically merging different chronic stressor mea-sures into one index. We constructed a chronic stress index (see Statis-tical Analyses) based on baseline questionnaires on work-related stress, marital stress, early life events, and social stress. For patients who started at T1, these questionnaires were added with specific instructions that they referred to the period before their PCI.
Work stress - The short version of the Effort-Reward Imbalance (ERI)
questionnaire was used to measure work-related stress [31]. The ques-tionnaire consists of 16 items (e.g. “the last few years my work has become
more demanding”), with three items measuring effort, seven measuring
reward and six measuring over-commitment [31,32]. All items were rated on a Likert scale from 1 (“completely disagree”) to 4 (“totally agree”), and 5 (“not applicable”). The items were summed and the total score could range from 0 to 64, with higher scores indicating more work- related stress. Questions in this instrument refer to the current or last job patients had. The questionnaire has previously been shown to be reliable and valid [31]. In the current study, the internal consistency was acceptable (Cronbach’s alpha Baseline = 0.78).
Relational stress was assessed using a 6-item version of the Maudsley
Marital Questionnaire (MMQ) (e.g. “Can you let your partner know your
true feelings?”) [33]. The items were rated on a seven-point Likert scale from 0 (e.g., “Frank and open with partner”) to 6 (e.g., “Conceal all the emotions all the time”) [33]. All items were summed and the total score could range from 0 to 36, with higher scores indicating more marital stress. Internal consistency in this study was good (Cronbach’s alpha Baseline = 0.89).
Life events were assessed using the Life events Questionnaire [34]. The Dutch questionnaire consists of 27 items that measure the experi-ence of certain life events (“have you been through the following life
events?” e.g. “decease parent/caretaker”, “suicide attempt of brother or sister”). Every item had 4 scores; “no, in none of those periods”, “yes,
before my 16th year of life”, “yes, between my 16th year of life and 1 year
ago”, “yes, in the past year”. The life events of each participant were summed and the total score could range from 0 to 27 life events, with higher scores indicating more experienced life events.
Social stress was assessed with questions from the psychosocial
Eu-ropean Society of Cardiology (ESC) screening interview [5]. From the ESC psychosocial screening interview the questions “do you live alone” and “is there a confidant in your life?” [5] were used. The questions are answered on a dichotomous scale by “yes” or “no”. When “do you live alone” was answered with “yes” that indicated social stress and “is there a confidant in your life” answered with “no” indicated social stress.
Past year’s stress experience – We asked participants to report on the
stress they had experienced over the past year, using four answer cate-gories: a lot of stress [4], a moderate amount of stress [3], relatively little stress [2], or almost no stress at all [1], in a Likert scale format [35].
2.2.3. Fatigue
Fatigue was measured with the fatigue subscale of the 24-item Health Complaints Scale (HCS) [36]. The fatigue subscale consisted of four items (“lately I have been feeling…” e.g. “tired”, “physically weak”), which were all rated on a five-point Likert scale from 0 (“not at all”) to 4 (“a lot”). The items were summed and the total score could range from 0 to 16. Higher scores indicate more fatigue complaints. Fatigue was
assessed at baseline, at 1 month follow-up (T1), at 1 year (T3) and 2 years follow-up (T4). Cronbach’s alpha for the fatigue subscale was excellent at all measurement points (T0 = 0.92, T1 = 0.92, T3 = 0.93 and T4 = 0.94).
2.3. Statistical analyses
Baseline characteristics were presented as Mean (SD) for continuous variables and as % (N) in case of categorical variables. Further, a sex stratified approach was adopted to examine whether there were signif-icant sex differences by independent t-tests and Chi-squared tests for the continuous and categorical variables, respectively. Education level was recoded into two categories, i.e., at least vocational education vs. lower. PCI indication was also recoded into two categories: elective (i.e., planned PCI) and acute (for ACS). Comorbid illnesses were counted and transformed into a count variable, counting none, one or two or more comorbidities. This index ranged from 0 to 4, with 61% of the sample having no comorbidities, 29% having one, and 10% having two or more.
2.3.1. Chronic stress index
As chronic stressors tend to co-occur [29], and different stress scales (with different levels of measurement) were used to assess a variety of chronic stressors, we constructed a chronic stress index. Because of the inclusion of continuous and ordinal scales, the assumptions of tradi-tional factor analysis (continuous variables, multivariate normality) are not met. However, in LatentGOLD®5.1 [37], mixture modeling can be used to estimate a factor model based on variables with different mea-surement levels. Typically, mixture models are used to estimate separate factor models in a number of latent classes, or clusters. Since we were not interested in finding clusters of patients with a typical factor loading, we forced LatentGOLD to estimate a 1-cluster mixture model, which estimates a factor model in only one cluster containing all participants. We entered all scale scores (work stress, marital stress, social stress, and early life events) as indicators. A 1-factor solution was then forced to obtain factor loadings for all participants, representing their combined chronic stress score, i.e., the chronic stress index, which was exported to IBM SPSS Statistics (SPSS).
2.3.2. General linear mixed modeling analysis
General linear mixed modeling analysis in SPSS 27.0 (IBM) was used to estimate the effect of the chronic stress index on the course of fatigue, using maximum likelihood estimation and Satterthwaite approximation for df calculation (default in SPSS). The linear mixed modeling tech-nique is suitable for analysis of repeated measurements, as it takes the possibility of correlated data into account. In addition, in contrast to traditional repeated measures ANOVA, one missing measurement occasion does not automatically lead to exclusion of that patient from analysis, limiting bias and preserving statistical power. It also has the possibility to measure variables as fixed- or as time-varying factors. We treated the chronic stress index, sex, and the covariates (see below) as fixed variables, measured at the first time point. Fatigue was the dependent variable, and time-varying as we wanted to examine the course of fatigue. Five subsequent models were tested. First, we exam-ined the effect of time on the course of fatigue. Next, we tested whether there were sex differences in both the levels (between-subjects effect) and course (within-subject effect) of fatigue over time (time interaction). Then, we assessed the between- (main effect) and within-subjects (time interaction) effects of chronic stress and examined whether the effect of chronic stress differed for the sexes (interaction). Finally, we adjusted the analysis for the effects of sociodemographic (low education, age), health behavioral (low physical activity, BMI), and medical variables (acute PCI setting, cardiac history, comorbidity index, beta-blockers).
In linear mixed modeling, the fixed effects (F tests) reflect the single overall test of the usefulness of the explanatory variables, without focusing on individual levels, while the t-tests reflect the significance of the estimates of the fixed effects. We used IBM SPSS Statistics for
Windows, Version 27.0 [38] for the analyses. 3. Results
3.1. Sample characteristics
In total, 2852 patients receiving PCI in the Elisabeth-TweeSteden hospital have been approached to participate in the THORESCI study since December 2013. Out of these participants, 11% (n = 306) was excluded based on the a priori exclusion criteria, leaving 2546 eligible patients. While 7% of men were cognitively unable to participate (due to e.g. dementia), 13% of women were excluded for this reason. Presence of a life-threatening comorbidity was 7 and 6% respectively in men and women. Finally, there was a small difference in mastery of the Dutch language, with women being more often excluded because of this reason (15 vs 13%). Rare exclusions were due to failure of the PCI and being disabled (blind or deaf). Of the eligible sample, 34% refused to partic-ipate (a third of which were women), leading to a final sample size of 1682. Among this sample, there were 327 (17%) premature dropouts, of which a third was female. Further, 277 patients only started a month after their PCI and thus skipped the baseline assessment due to incon-venience or severity of the experienced MI (mostly). Four percent of the eligible sample passed away during the study period (no sex differ-ences). We had complete fatigue assessments for N = 1405 at T0, N = 1325 at T1, N = 1005 at T3, for N = 789 at T4. Note that this is an ongoing study, indicating that not all missing measures are dropouts. All participants (N = 1682) had a chronic stress index score at baseline.
Baseline characteristics, stratified for sex, are presented in Table 1. Of all participants included in the current study, 78% (n = 1311) was male. The average age of participants was 67 years (SD = 10.61), with men being significantly younger than women (Mdifference = − 2.4, t (463,59) = − 3.283, p = .001). The majority of the sample had a partner, though more men than women were in a relationship (χ2 =46.36, p < .001). Around 40% of the participants was employed, with men again Table 1 Baseline characteristics. Total Men (n = 1311) Women (n =371) Demographics Age 67.1 (10.6) 66.6 (10.3) 68.9 (11.3) Relationship (no) 19% (323) 16% (206) 32% (117) Working (fulltime or parttime) 40% (643) 43% (535) 30% (108) Education (low) 33% (496) 29% (351) 44% (145) Risk factors Smoking (y) 12% (169) 11% (126) 14% (43) Low physical activity 14% (209) 15% (168) 13% (41) BMI 27.7 (5.1) 27.6 (4.4) 27.9 (6.9) Hypertension (y) 41% (682) 39% (513) 46% (169) Hypercholesterolemia (y) 30% (508) 31% (403) 28% (105) Familial risk (y) 36% (598) 35% (453) 39% (145) Medical background (yes)
Acute PCI 73% (992) 73% (773) 71% (219) Cardiac historya 39% (648) 41% (430) 32% (118) Diabetes 18% (301) 18% (229) 19% (72) COPD 7% (118) 7% (85) 9% (33) Cancer past 5 yrs. 2% (37) 2% (28) 2% (9) Anemia 1% (11) 1% (9) 1% (2) Rheumatoid arthritis 19% (312) 18% (236) 21% (76) Liver disease 1% (12) 1% (8) 1% (4) Kidney disease 3% (52) 3% (38) 4% (14) Betablocker prescriptionb 49% (704) 47% (536) 55% (168)
Data are presented as n (%) or mean (standard deviation).
aPrevious myocardial infarction, coronary artery bypass grafting, percutaneous
General Hospital Psychiatry 72 (2021) 45–52
48
holding a job more often than women did (χ2 = 16.15, p < .001). Significantly more women than men had a low level of education (high school or less; χ2 =25.59, p < .001). With respect to the classical car-diovascular risk factors, only hypertension showed sex differences with a higher prevalence in women (χ2 =7.59, p = .023). Cardiac history was also different between the sexes, as men more often had experienced a PCI, CABG or MI in the past (χ2 =9.14, p = .003).
Non-cardiovascular comorbidities were not significantly different for men and women, although occurrences always were higher for women. This also becomes evident in the comorbidity index, which shows that women tend to have more comorbidities than men, albeit marginally significant (χ2 =4.71, p = .095).
3.2. Chronic stress index
All variables significantly contributed to the factor, with factor loadings being smallest for social stress and largest for experienced stress in past 12 months (Table 2). There was a significant difference in the mean score of the chronic stress index between men and women (t = − 4.926, df = 1680, p < .001). Examining sex differences in the input of the chronic stress index showed that women reported more stress on all components, except work stress and life events (data not shown).
3.3. Sex and chronic stress as predictors of fatigue over time
In the first model, the effect of time on the course of fatigue was tested. Fatigue reduced substantially over the first month (F(3, 4524) = 30.365, p < .001) from M = 5.3 (95% CI [5.1, 5.6]) to M = 4.2 (95% CI [4.0, 4.4]), and then stabilized over the 1-year follow-up (M = 4.2, 95% CI [4.0, 4.5]), and 2-year follow-up period (M = 3.8, 95% CI [3.5, 4.1]). See also Fig. 1.
3.3.1. Sex
Next, we tested whether there were sex differences both the levels and course of fatigue. Overall, women (F(1,4519) =100.041, p < .001; M =5.6, 95% CI [5.2, 6.0]) were significantly more fatigued than men (M =4.1, 95% CI [3.8,4.4]), with the trajectory over time being different between men and women as well (F(3, 3104.484) = 3.801, p = .026). Custom hypothesis testing for the effects of this interaction revealed that women differed most from men at baseline (Mdifference =1.8, 95% CI [1.3,2.3]) and at the 2-year follow-up (Mdifference =1.3, 95% CI [0.7,1.9]; all p < .001).
3.3.2. Chronic stress
Then, we assessed the between- and within-subjects effects of chronic stress and examined whether the effect of chronic stress differed for the sexes. The chronic stress index significantly impacted the level of fatigue across the follow-up period (F(1, 4518) =349.376, p < .001), with higher levels of chronic stress adding substantially to the level of experienced fatigue (B = 1.64, 95% CI [1.40, 1.89], t = 13.232, p < .001). Chronic stress also interacted with time, i.e., the within-person change in fatigue (F(3, 3040.817) = 5.556, p < .001), especially from baseline to 1 month follow-up (B = 0.49, 95% CI [0.15, 0.82], t = 7.96, p
<.001). The between-subjects effect of chronic stress was present in
both sexes (p < .001), while the within-subjects change was only sig-nificant in men (F(3, 2.399.629) =3.879; p < .009) as compared to women
(F(3, 647.769) =0.986; p = .399) Though, the main effect estimate for chronic stress was larger for women (B = 1.89; 95% CI [1.1, 2.7],t = 4.554, p < .001) than men (B = 1.28; 95% CI [0.90, 1.66], t = 6.559, p < .001). To visualize the effect of the different chronic stress levels on fatigue over time, we calculated three levels of chronic stress to display the course of fatigue for different levels of chronic stress separately for men and women (Fig. 1).
3.3.3. Covariates model
The estimates of the fully adjusted model are presented in Table 3
and the effects of the covariates are visualized in Fig. 2. Because the effect of beta-blockers was nonexistent (est = − 0.008 (− 0.38–0.40), t = 0.042, p = .967), and because it was important to preserve sample size, we ran the full model without the beta-blocking medication variable (predictor and covariate effects did not change due to addition of the beta-blocking medication variable. The full model indicated that fatigue reduced significantly over time, each time point being significantly different from baseline (i.e., first week after PCI). Chronic stress increased the level of fatigue significantly, with a 2.2-point difference in fatigue with one unit increase in chronic stress. Chronic stress also affected the recovery of fatigue, which was less pronounced with increased chronic stress. Women also reported higher levels of fatigue, and there was a significant interaction with time for the first, but not the second year of follow-up. The effect of chronic stress was similar for men and women alike. Low education, comorbid illnesses, elective PCI pro-cedure, younger age, and low physical activity all increased the level of fatigue significantly. While the effect of female sex was reduced a little bit by adding the covariates, the effect of chronic stress became stronger. This means that the covariates explained part of why women report increased fatigue. In general, all covariates but cardiac history signifi-cantly explained individual differences in fatigue (Fig. 2).
3.3.4. Power
We conducted three post-hoc statistical power analyses in R with the ‘lme4’ and ‘simr’ packages [39], with 250 replications, to establish the level of power for the stress-by-time interaction, the sex-by-time inter-action, and the stress-by-sex interinter-action, which revealed the following respective power levels: post hoc power for stress-by-time = 96% (95% CI: 92.77–98.07) given an effect size of − 0.3; posthoc power levels for sex-by-time was 69% (95%CI: 63.07–74.86) given an effect size of 0.37; and the stress-by-sex interaction had a power hoc power level of 36% (95%CI: 30.05–42.29) given an effect size of 0.45. It is of note, that the ‘simr’ package only works with complete datasets, so we only ran the power analysis based on the smaller complete data set (I.e., with all missing values excluded). This implies that the actual power is likely to be higher.
4. Discussion
The present study examined predictors of the level and course of fatigue in the 2-year period after undergoing PCI for either chronic stable angina or ACS, focusing on chronic stress and sex differences. The results revealed that fatigue reduced substantially over the first month after the PCI procedure, and then stabilized at a moderate level. Women reported significantly more fatigue than men, and the course of fatigue showed sex differences as well, particularly across the first month. Table 2
Descriptive statistics and factor loadings for variables included in chronic stress index (N = 1682).
Variables Measurement level Mean (SD)/% (N) Wald p value Factor loadings R2 Past year stress experience (range: 1–4) Ordinal 2.2 (0.98) 22.076 <0.001 0.707 0.51
Work stress of pre-PCI job (ERI; range 16–64) Continuous 39.8 (7.14) 80.887 <0.001 0.514 0.26
Negative life events (0–17) Count 0.85 (1.38) 36.651 <0.001 0.333 0.20
Relational stress (0–34) Continuous 3.9 (5.39) 72.853 <0.001 0.360 0.13
Lack of close confidant (0/1) Dichotomous 14% (371) 5.253 0.022 0.086 0.01
Chronic stress impacted both the level and course of fatigue by increasing its level and slowing its recovery post-PCI, with fatigue in women being affected more. The effects of sex and chronic stress on the level and course of fatigue were independent of demographic, health behavioral, and medical covariates, that all (except cardiac history) individually added to the explained variance.
Fatigue is common in coronary heart disease, often coinciding in a triad of symptoms with angina and dyspnea [40,41], and recovering slowly after ACS [18,20]. The current study specifies that in the first month, fatigue rapidly recovers, while after that month, fatigue remains fairly stable. Our study is in concordance with previous work from the Netherlands showing that fatigue decreases slowly post-PCI from the moment of cardiac rehabilitation to a year later, with only a minority reporting severe fatigue [20]. Extant literature further report the pres-ence of sex differpres-ences, as meta-analytic and review evidpres-ence [21,26] show that women with coronary heart disease were more inclined to report fatigue symptoms than their male counterparts. It is of note that women also report more fatigue than men prior to the diagnosis of coronary heart disease [42]. Our results are in line with these findings. As for one possible explanation of these sex differences, there may be consistent differences between men and women in symptom reporting. Women tend to experience and communicate psychological distress in the form of somatic symptoms like fatigue more often than men [43], which may explain an elevated report of fatigue. On the other hand, there is some evidence in other populations (shiftwork, heart failure), that sex differences are also present in objective measures of fatigue [44,45]. Future research may want to examine the prevalence and course of objective measures of fatigue (i.e. exercise capacity, muscle fatigue, mental fatigue) during the short- and long-term recovery from ACS.
Chronic stress has been recognized as a predisposing and perpetu-ating factor of fatigue in the general population [24]. Our results confirm this strong association between chronic stress and fatigue, showing both level and course effects, independent of covariates. Bio-logical processes associated with chronic stress such as continued HPA
axis activation and, as a consequence, increased low grade inflammation may explain this link [23]. Chronic stress is known to cause inappro-priate, prolonged activation of the innate immune system [4], resulting in persistently increased levels of proinflammatory cytokines [46], accompanied by symptoms of sickness behavior, among which, fatigue [47]. The relationship between stress and fatigue may be reciprocal [25], what may contribute to a self-sustaining system, with persistent levels of chronic stress and fatigue in patients coronary heart disease [17–20]. Future studies may want to add measures of HPA axis and immune activation as potential mediators in the relation between chronic stress and fatigue. As there is growing evidence for the capacity of stress reduction interventions such as mindfulness-based stress reduction to diminish low grade inflammation [48], stress biomarkers [49], as well as fatigue [50], future studies may want to examine whether in patients with coronary heart disease such an intervention is able to reduce inflammation and cortisol, and thereby fatigue.
General Hospital Psychiatry 72 (2021) 45–52
50
vocational education reported significantly more fatigue than patients who had vocational education [52]. It is known that people with a lower education are more inclined to experience work stress, due to job strain and insufficient reward [29,53]. Considering work stress (i.e. effort reward imbalance) has been included in the chronic stress index in the current study, the level of education may moderate the effect of chronic stress on fatigue. Furthermore, we found that younger patients reported significantly more fatigue symptoms over the 2-year follow-up period in comparison to older patients, potentially due to the greater variety of responsibilities (work, family, care) that come with a midlife age, in comparison to pensioners. Evidence in the literature so far shows no consistent variation of general fatigue with age in the general population [54–57]. Exertion fatigue seems to be higher with older age in studies in patients with heart failure [56,57].
Fatigue is often underappreciated by physicians, possibly because fatigue knows a multifactorial etiology, often outside the scope of
cardiology, and is difficult to treat. From a clinical cardiology perspec-tive, the present study contributes by removing some of the elusiveness of fatigue in patients with heart disease, showing the diversity and range of significant predictors of fatigue, all explaining some of the variance. Our research indicates the importance of assessing chronic stress in patients, as patients with chronic stress are more likely to suffer from long-lasting elevated fatigue symptoms, likely as a consequence of low- grade inflammation, which promotes atherosclerosis and decreases quality of life [19,21,22]. To improve the clinical practice, interventions focusing on stress-reduction or increasing physical exercise may be offered to improve fatigue, especially in the female and younger patients.
Several limitations and strengths characterize our study. Both fatigue and chronic stress measures were obtained by self-report, thus no objective measures have been included. Objective measures of fatigue (e.g., dual tasking or continued performance experiments, or concen-tration of oxidized hemoglobin in muscle during physical activity), though more time-consuming, might provide more detail in individual differences in (predictors of) fatigue. Nevertheless, previous research indicated a strong association between objective and self-reported fa-tigue [58]. Another limitation concerns the decline in participants across the study period. THORESCI is an ongoing study, and therefore about half of the patients did not reach the 2-year mark yet. The mixed linear modeling we chose to execute our analysis uses maximum like-lihood estimation which is robust for such a decline in numbers. In addition, there were relatively more women among refusals and drop-outs (1 in 3 for both) than among participants (1 in 5). It was therefore important to examine the statistical power of our analysis post-hoc. This power analysis showed that the sex-by-time interaction was not suffi-ciently powered. However, because the current power analysis could only be performed on the data set that includes only complete mea-surements, the actual level of power is likely to be higher. With respect to generalizability, the difference in prevalence of women in the included vs the excluded/refusal group could have induced bias in the results, with more fatigued women refusing participation. The presented results would them be an underestimation. The larger number of women who were excluded because of dementia/Alzheimer’s is illustrative for the later age of onset of heart disease in women.
We successfully constructed a chronic stress index, based on multiple questionnaires assessing the level of experienced chronic stress in different contexts (i.e., work, relation, life events), which is a novel approach. This chronic stress index could then be used as a more comprehensive, yet parsimonious indicator of overall chronic stress. We did not use one of the comprehensive measures of chronic stress, like the Trier Inventory for Chronic Stress (TICS) [59], Chronic Burden scale [60] or the very comprehensive STRAIN measure [61], either because they were not available yet at the time of study design, or due to page limitations of our survey. In such a case, our method is a good option to combine separate questionnaires for the different contexts of chronic stress, instead of one item for assessing chronic stress in a particular context [59,60]. An important sidenote is that the TICS only includes the work and the social contexts, disregarding life events, while the Chronic burden scale’s scope is very broad from one’s own health stress to drug and alcohol problems in the family. STRAIN produces a very compre-hensive stress exposure measure [61], the downside being that it takes quite some time to complete.
In conclusion, the present study showed that chronic stress affected the level and 2-year course of fatigue in patients who underwent PCI, with women being impacted most. The effects of sex and chronic stress on fatigue were independent of demographic, health behavioral, and medical covariates, most of them individually adding to the explained variance. Future research should focus on further explaining the mechanism of these relations, and on developing and testing in-terventions focusing on physical activity and stress-reduction to treat fatigue.
Table 3
Fully adjusted model of chronic stress and sex effects on the level and course of fatigue over 2 years of follow-up, adjusted for covariates.
Parameter Estimate (B) 95%CI t(df) p-value Time Baseline Reference +1 month −1.75 − 2.26 to − 1.24 − 6.77 (985.76) <0.001 +12 months −1.56 − 2.18 to − 0.93 − 4.88 (904.23) <0.001 +24 months −1.43 − 2.05 to − 0.81 − 4.56 (804.42) <0.001
Chronic stress (between
subjects effect) 2.20 1.58 to 2.83 6.77 (1260.09) <0.001 Chronic stress (within
subjects effect) Chronic stress * baseline Reference * +1 mo 0.36 − 0.03 to 0.69 −(952.89) 2.17 0.030 * +12 mo 0.59 − 0.19 to 0.98 − 2.93 (855.53) 0.004 * +24 mo 0.40 − 0.01 to 0.79 − 2.01 (776.54) 0.045 Female sex (between
subjects effect) 1.13 0.55 to 1.71 3.84 (1144.72)
<0.001
Female sex (within subjects
effect)
Female sex * baseline Reference
* +1 mo −0.75 − 1.33 to − 0.18 − 2.58 (987.29) 0.010 * +12 mo −0.72 − 1.43 to − 0.03 − 2.03 (898.22) 0.042 * +24 mo −0.45 − 1.15 to − 0.25 1.26 (802.94) 0.208 Female sex * chronic
stress 0.54 0.26 to 1.05 1.59 (1054.52) 0.111 Covariates Low education 1.00 0.58 to 1.41 4.73 (1112.89) <0.001 Cardiac historya 0.22 0.19 to 0.61 1.05 (1070.81) 0.295 Comorbidity indexb 0.79 0.49 to 1.09 5.23 (1087.13) <0.001 Elective PCI 0.62 0.20 to 1.04 2.88 (1102.27) 0.004 Age −0.03 − 0.05 to 0.01 −(1124.86) 2.52 0.012 Low physical activity 0.87 0.48 to
1.27 4.37 (1098.29)
<0.001 Note: aCardiac history includes previous PCI, MI, CABG, atrial fibrillation, heart
failure, and implanted pacemaker. bComorbidity index included the presence of
diabetes, COPD, cancer during the past 5 years, rheumatoid arthritis, anemia, liver disease, and kidney disease. PCI = percutaneous coronary intervention. CABG = bypass surgery, MI = myocardial infarction.
Funding
This study is funded by the NWO Aspasia grant (Dutch Research Council) granted to dr. N. Kupper (grant number: 015008055) and by the Gender and Prevention grant awarded by The Netherlands Organi-sation for Health Research and Development (grant number: 555003012) to dr. N. Kupper.
References
[1] Townsend N, Wilson L, Bhatnagar P, Wickramasinghe K, Rayner M, Nichols M. Cardiovascular disease in Europe: epidemiological update 2016. Eur Heart J 2016; 37(42):3232–45.
[2] Vaccarino V, Badimon L, Corti R, de Wit C, Dorobantu M, Manfrini O, et al. Presentation, management, and outcomes of ischaemic heart disease in women. Nat Rev Cardiol 2013;10(9):508–18.
[3] Piepoli MF, Hoes AW, Agewall S, Albus C, Brotons C, Catapano AL, et al. 2016 European Guidelines on cardiovascular disease prevention in clinical practice: The
Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts)Developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR). Eur Heart J 2016;37(29):2315–81.
[4] Wirtz PH, von K¨anel R. Psychological stress, inflammation, and coronary heart disease. Curr Cardiol Rep 2017;19(11):111.
[5] van Montfort E, Denollet J, Vermunt JK, Widdershoven J, Kupper N. The tense, the hostile and the distressed: multidimensional psychosocial risk profiles based on the ESC interview in coronary artery disease patients - the THORESCI study. Gen Hosp Psychiatry 2017;47:103–11.
[6] Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 1999;99(16):2192–217.
[7] Li J, Zhang M, Loerbroks A, Angerer P, Siegrist J. Work stress and the risk of recurrent coronary heart disease events: a systematic review and meta-analysis. Int J Occup Med Environ Health 2015;28(1):8–19. https://doi.org/10.2478/s13382- 014-0303-7.
[8] Matthews KA, Gump BB. Chronic work stress and marital dissolution increase risk of posttrial mortality in men from the multiple risk factor intervention trial. Arch Intern Med 2002;162(3):309–15.
General Hospital Psychiatry 72 (2021) 45–52
52 [9] Wang H-X, Leineweber C, Kirkeeide R, Svane B, Schenck-Gustafsson K, Theorell T,
et al. Psychosocial stress and atherosclerosis: family and work stress accelerate progression of coronary disease in women. The Stockholm Female Coronary Angiography Study. J Intern Med 2007;261(3):245–54.
[10] Orth-Gom´er K, Wamala SP, Horsten M, Schenck-Gustafsson K, Schneiderman N, Mittleman MA. Marital stress worsens prognosis in women with coronary heart disease: the Stockholm female coronary risk study. JAMA 2000;284(23):3008–14. [11] Waterland RA. Is epigenetics an important link between early life events and adult
disease? Horm Res Paediatr 2009;71(Suppl. 1):13–6.
[12] Case RB, Moss AJ, Case N, McDermott M, Eberly S. Living alone after myocardial infarction: impact on prognosis. JAMA 1992;267(4):515–9.
[13] Williams RB, Barefoot JC, Califf RM, et al. Prognostic importance of social and economic resources among medically treated patients with angiographically documented coronary artery disease. JAMA 1992;267(4):520–4.
[14] Dimsdale JE. Psychological stress and cardiovascular disease. J Am Coll Cardiol 2008;51(13):1237–46.
[15] Steptoe A, Kivim¨aki M. Stress and cardiovascular disease: an update on current knowledge. Annu Rev Public Health 2013;34(1):337–54.
[16] Vaccarino V, Bremner JD. Behavioral, emotional and neurobiological determinants of coronary heart disease risk in women. Neurosci Biobehav Rev 2017;74(Pt B): 297–309.
[17] Bartels MN. Fatigue in cardiopulmonary disease. Phys Med Rehabil Clin N Am 2009;20(2):389–404.
[18] Als´en P, Brink E. Fatigue after myocardial infarction – a two-year follow-up study. J Clin Nurs 2013;22(11− 12):1647–52.
[19] Bunevicius A, Stankus A, Brozaitiene J, Girdler SS, Bunevicius R. Relationship of fatigue and exercise capacity with emotional and physical state in patients with coronary artery disease admitted for rehabilitation program. Am Heart J 2011;162 (2):310–6.
[20] van Geffen ME, Hoeve NT, Sunamura M, Stam HJ, van Domburg RT, van den Berg- Emons RJG. Fatigue during and after cardiac rehabilitation. J Rehabil Med 2015; 47(6):569–74.
[21] Eckhardt AL. Fatigue in the presence of coronary heart disease. Nurs Res 2014;63 (2):83.
[22] Staniute M. Relationship of health-related quality of life with fatigue and exercise capacity in patients with coronary artery disease. Eur J Cardiovasc Nurs 2014;13 (4):338.
[23] Strahler J, Skoluda N, Rohleder N, Nater UM. Dysregulated stress signal sensitivity and inflammatory disinhibition as a pathophysiological mechanism of stress- related chronic fatigue. Neurosci Biobehav Rev 2016;68:298–318. [24] Cho HJ, Bower JE, Kiefe CI, Seeman TE, Irwin MR. Early life stress and
inflammatory mechanisms of fatigue in the Coronary Artery Risk Development in Young Adults (CARDIA) study. Brain Behav Immun 2012;26(6):859–65. [25] Doerr JM, Ditzen B, Strahler J, Linnemann A, Ziemek J, Skoluda N, et al.
Reciprocal relationship between acute stress and acute fatigue in everyday life in a sample of university students. Biol Psychol 2015;110:42–9.
[26] Coventry LL. Sex differences in symptom presentation in acute myocardial infarction: a systematic review and meta-analysis. Heart Lung 2011;40(6):477. [27] Tamres LK. Sex differences in coping behavior: a meta-analytic review and an examination of relative coping. Personal Soc Psychol Rev (Lawrence Erlbaum Associates) 2002;6(1):2.
[28] Brichant JF. The Helsinki declaration on patient safety in anaesthesiology. Acta Anesthesiol Belg 2010;61(2):49.
[29] Rozanski A, Blumenthal JA, Davidson KW, Saab PG, Kubzansky L. The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice. Emerg Field Behav Cardiol 2005;45(5):637–51.
[30] McEwen BS, Stellar E. Stress and the individual: mechanisms leading to disease. Arch Intern Med 1993;153(18):2093–101.
[31] Siegrist J, Li J, Montano D. Psychometric properties of the effort-reward imbalance questionnaire. Germany: Duesseldorf University; 2014.
[32] Stanhope J. EffortReward imbalance questionnaire. Occup Med 2017;67(4):314. [33] Arrindell WA. On the psychometric properties of the Maudsley Marital
Questionnaire (MMQ): evaluation of self-ratings in distressed and normalvolunteer couples based on the Dutch version. Personal Individ Differ 1983;4(3):293. [34] Garnefski N, Kraaij V, Spinhoven P. Negative life events, cognitive emotion
regulation and emotional problems. Personal Individ Differ 2001;30(8):1311–27. [35] Keller A, Litzelman K, Wisk LE, Maddox T, Cheng ER, Creswell PD, et al. Does the
perception that stress affects health matter? The association with health and mortality. Health Psychol 2012;31(5):677–84.
[36] Denollet J. Health complaints and outcome assessment in coronary heart disease. Psychosom Med 1994;56(5):463.
[37] Vermunt JK, Magidson J. Upgrade manual for latent GOLD 5.1. Belmont, MA: Statistical Innovations Inc.; 2016.
[38] IBM SPSS Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp; 2020. [39] Green P, MacLeod CJ. SIMR: an R package for power analysis of generalized linear
mixed models by simulation. Methods Ecol Evol 2016;7(4):493–8.
[40] Kimble LP, Dunbar SB, Weintraub WS, McGuire DB, Manzo SF, Strickland OL. Symptom clusters and health-related quality of life in people with chronic stable angina. J Adv Nurs 2011;67(5):1000–11.
[41] Ryan CJ, DeVon HA, Horne R, King KB, Milner K, Moser DK, et al. Symptom clusters in acute myocardial infarction: a secondary data analysis. Nurs Res 2007; 56(2):72–81.
[42] Khan NA, Daskalopoulou SS, Karp I, Eisenberg MJ, Pelletier R, Tsadok MA, et al. Sex differences in prodromal symptoms in acute coronary syndrome in patients aged 55 years or younger. Heart. 2017;103(11):863–9.
[43] Lilienfeld SO, Hess TH. Psychopathic personality traits and somatization: sex differences and the mediating role of negative emotionality. J Psychopathol Behav Assess 2001;23(1):11–24.
[44] Tatsuta H, Ebata C, Tatmeichi M. Gender differences in the subjective and objective markers for fatigue associated with shift work: a pilot study. Occup Med Health Aff 2018;06.
[45] Lau ES, Cunningham T, Hardin KM, Liu E, Malhotra R, Nayor M, et al. Sex differences in cardiometabolic traits and determinants of exercise capacity in heart failure with preserved ejection fraction. JAMA Cardiol 2020;5(1):30–7. [46] Cohen S, Janicki-Deverts D, Doyle WJ, Miller GE, Frank E, Rabin BS, et al. Chronic
stress, glucocorticoid receptor resistance, inflammation, and disease risk. Proc Natl Acad Sci 2012;109(16):5995–9.
[47] Dantzer R, Kelley KW. Twenty years of research on cytokine-induced sickness behavior. Brain Behav Immun 2007;21(2):153–60.
[48] Andr´es-Rodríguez L, Borr`as X, Feliu-Soler A, P´erez-Aranda A, Rozadilla-Sacanell A, Montero-Marin J, et al. Immune-inflammatory pathways and clinical changes in fibromyalgia patients treated with Mindfulness-Based Stress Reduction (MBSR): a randomized, controlled clinical trial. Brain Behav Immun 2019;80:109–19. [49] Pascoe MC, Thompson DR, Ski CF. Yoga, mindfulness-based stress reduction and
stress-related physiological measures: a meta-analysis. Psychoneuroendocrinology 2017;86:152–68.
[50] Kearney DJ. Mindfulness-based stress reduction in addition to usual care is associated with improvements in pain, fatigue, and cognitive failures among veterans with gulf war illness. Am J Med 2016;129(2):204.
[51] Lawson CA, Solis-Trapala I, Dahlstrom U, Mamas M, Jaarsma T, Kadam UT, et al. Comorbidity health pathways in heart failure patients: a sequences-of-regressions analysis using cross-sectional data from 10,575 patients in the Swedish Heart Failure Registry. PLoS Med 2018;15(3):e1002540.
[52] Barbareschi G. Educational level and the quality of life of heart failure patients: a longitudinal study. J Card Fail 2011;17(1):47.
[53] Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol 1996;1(1):27.
[54] Lewis G, Wessely S. The epidemiology of fatigue: more questions than answers. J Epidemiol Community Health 1992;46(2):92–7.
[55] Hickie IB, Hooker AW, Hadzi-Pavlovic D, Bennett BK, Wilson AJ, Lloyd AR. Fatigue in selected primary care settings: sociodemographic and psychiatric correlates. Med J Aust 1996;164(10):585–8.
[56] Smith OR, Kupper N, de Jonge P, Denollet J. Distinct trajectories of fatigue in chronic heart failure and their association with prognosis. Eur J Heart Fail 2010;12 (8):841–8.
[57] Kessing D, Denollet J, Widdershoven J, Kupper N. Fatigue and self-care in patients with chronic heart failure. Eur J Cardiovasc Nurs 2016;15(5):337–44.
[58] Holtzer R, Foley F. The relationship between subjective reports of fatigue and executive control in multiple sclerosis. J Neurol Sci 2009;281(1–2):46–50. [59] Petrowski K, Braehler E, Schmalbach B, Hinz A, Bastianon C, Ritz T. Psychometric
properties of an English short version of the Trier Inventory for chronic Stress. BMC Med Res Methodol 2020;20(1):306.
[60] Gallo LC, Jim´enez JA, Shivpuri S, Espinosa de los Monteros K, Mills PJ. Domains of chronic stress, lifestyle factors, and allostatic load in middle-aged Mexican- American women. Ann Behav Med 2011;41(1):21–31.
[61] Slavich GM, Shields GS. Assessing lifetime stress exposure using the Stress and Adversity Inventory for Adults (Adult STRAIN): an overview and initial validation. Psychosom Med 2018;80(1):17–27.
