| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Medical Microbiology, University Hospital Maastricht, Maastricht, The Netherlands (M.K., C.A.B.); the Departments of Internal Medicine (A.J.v.d.V.) and Chemical Endocrinology (F.C.G.J.S.), Radboud University Nijmegen Medical Center Nijmegen, Nijmegen, The Netherlands (A.J.v.d.V.); the Departments of Psychiatry and Neuropsychology (R.v.D.), Cardiology (F.W.H.M.B.), and Medical Psychology (A.A.), Maastricht University, Maastricht, The Netherlands; and the Department of Internal Medicine, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland (T.C.).
Address correspondence and reprint requests to Martijn Kwaijtaal, MSc, Department of Medical Microbiology, University Hospital Maastricht, P. Debeyeplein 25, P.O. Box 5800, 6202 AZ, Maastricht, The Netherlands. E-mail: martijnkwaijtaal{at}gmail.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: Participants were 194 patients who had been recently treated by percutaneous coronary intervention (PCI) and who were exhausted at the start of the study. Half entered a behavioral intervention program. MIF, C-reactive protein, interleukin (IL)-6, IL-1 receptor antagonist, and neopterin were measured in blood collected 6 weeks after PCI (baseline) and 6 and 18 months after baseline. A single measurement of MIF was also available for 129 age- and sex-matched healthy individuals (reference group).
Results: At baseline, MIF in patients undergoing PCI was significantly lower than in the reference group (p < .01). New cardiac events occurred twice as often in the lowest quartile than in the highest quartile of MIF concentrations. However, the association was not significant (
2 = 2.27; df = 3; p = .52). During follow up, MIF concentrations increased significantly in patients undergoing PCI (p < .001). At 18 months, MIF concentrations were significantly lower in the exhausted patients than in the nonexhausted patients (p = .02). hsCRP, IL-1ra, IL-6, and neopter in concentrations did not change over this time period.
Conclusions: The data are suggestive of a negative association of MIF with CAD and of MIF with exhaustion. The observation that those patients who remained exhausted had lower concentrations of MIF fits into earlier observations that suggested that exhausted coronary patients may be characterized by a hypoactivity of the hypothalamic–pituitary–adrenocortical axis.
Key Words: angioplasty • inflammation • MIF • exhaustion • depression
Abbreviations: MIF = macrophage migration inhibitory factor; CAD = coronary artery disease; IL = interleukin; TNF = tumor necrosis factor; ACTH = adenocorticotrope hormone; HPA = hypothalamic–pituitary–adrenocortical; PCI = percutaneous coronary intervention; EXIT = Exhaustion Intervention Trial; MQ = Maastricht Questionnaire; MIVE = Maastricht Interview for Vital Exhaustion; EDTA = ethylenediaminetetraacetic-treated; ELISA = enzyme-linked immunosorbent assay; Ab = antibody; CV = coefficient of variation; TMB = tetramethylbenzidine solution; CABG = coronary artery bypass graft surgery.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
). MIF is also expressed by endocrine organs involved in the stress response, especially by the pituitary gland (3). Pituitary-derived MIF has important roles in the periphery such as antagonizing the effects of glucocorticoids (4). Because MIF is produced by T cells and by the pituitary, it can be classified as a proinflammatory cytokine as well as a hormone. Current knowledge regarding the association of MIF with coronary artery disease (CAD) is ambiguous. On the one hand, MIF induces expression of the intercellular adhesion molecule-1 by vascular endothelial cells, thus promoting atherosclerosis (5). Inhibition of MIF results in a shift of neointimal atherosclerotic plaques toward a more stabilized plaque (6). An upregulation of MIF during the progression of atherosclerosis toward inflammatory stages has been reported in humans (7). Finally, MIF was found to have a positive but weak association with future CAD in the EPIC study (8). These arguments suggest a positive association of MIF with CAD. On the other hand, high concentrations of MIF have been shown to promote neovascularization, especially during hypoxic stress, suggesting that MIF is cardioprotective (9–12).
To our knowledge, the association of MIF with the mental state of coronary patients has not yet been investigated. We have approached the depressive symptomatology that may be observed in more than half of all coronary patients as a state of exhaustion caused by prolonged exposure to stress. Psychophysiological studies have found that exhausted subjects are characterized by higher levels of serological markers of inflammation (e.g., C-reactive protein (CRP), IL-6, TNF-) (13,14). These observations suggest a positive association of MIF with exhaustion. Exhausted subjects are also characterized by lower levels of adenocorticotrope hormone (ACTH) and cortisol (15,16). This suggests that exhaustion is characterized by a decreased activity of the hypothalamic–pituitary–adrenocortical (HPA) axis. Hypoactivity of the HPA axis results in reduced inhibition of immune-mediated inflammation (17,18). Because MIF is also a hormone secreted by the pituitary gland, these observations suggest a negative association of MIF with exhaustion.
The aim of the current study was to explore the direction of the association of MIF with CAD and of the direction of the association of MIF with a state of exhaustion using blood samples of patients undergoing percutaneous coronary intervention (PCI). To investigate whether MIF is a potential risk factor or a potential protective factor of CAD, we posed the following questions: a) Are MIF concentrations in the blood of patients undergoing PCI different from the MIF concentrations in a reference group? b) What is the association of MIF concentrations with occurrence of new cardiac events in patients undergoing PCI? c) Is the expression pattern of MIF different from the expression pattern of other proinflammatory cytokines during a period of 18 months after PCI?
To investigate the association of MIF with exhaustion, we posed the following question: Are MIF concentrations in exhausted patients undergoing PCI different from MIF concentrations in nonexhausted patients undergoing PCI?
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
14 to enter the second stage. In this stage, the Maastricht Interview for Vital Exhaustion (MIVE) was administered to establish whether patients also met a MIVE cutoff score of 7 to be included in EXIT. The MIVE better predicts future cardiac events than the MQ and consists of 23 questions (range, 0–23) (20). Further details of inclusion and exclusion criteria, cutoff values, and applied behavioral intervention techniques are presented elsewhere (19).
Collection of Blood Samples
In one participating center (Maastricht), blood samples were collected on three occasions (Fig. 1). A medical specialist screened the medical records of the patients undergoing PCI for the use of immunosuppressive medication. This resulted in the exclusion of 18 patients. Therefore, the first blood sample (baseline) was available for 194 patients. The median interval between PCI and baseline was 42 days. The second blood sample was obtained 6 months after baseline and available from 185 patients (i.e., data of nine patients are missing as a result of either laboratory processing errors or because patients did not show up, refused to give blood, or had died). The last blood sample was obtained 18 months after baseline and available of 181 patients (i.e., data of 13 patients are missing as a result of loss to follow-up). In summary, all three blood samples were available of 172 of the initial 194 patients.
|
To compare MIF concentrations of patients undergoing PCI with values in a reference group, a single blood sample was also available for 129 age- and sex-matched individuals who denied having any cardiovascular symptoms and who were apparently healthy. This group was not tested for exhaustion. Of all blood samples, 90% was collected in the morning (range 8:30 AM–12:00 PM). The Institutional Review Board of the participating center approved the study protocol, and all participants gave written informed consent. PCI treatments were conducted between 1996 and 2000; follow up of patients was completed in 2002.
Measurement of Macrophage Migration Inhibitory Factor
Plasma samples were collected from ethylenediaminetetraacetic-treated (EDTA) blood and stored at –20°C until further processing. An enzyme-linked immunosorbent assay (ELISA) for human MIF has been developed using the four-span approach earlier described by Grebenschikov et al. (21). Antibodies were raised in chicken and rabbits using rhMIF as immunogen. The sandwich structure used includes four different antibodies (Abs), viz. a coating Ab (duck anti-chicken), a capture Ab (chicken anti-hMIF), a trapping Ab (rabbit anti-hMIF), and finally a detection Ab (horseradish peroxidase-labeled goat anti-rabbit). The procedure started with treating the microtiter plates with coating antibody (duck–chicken IgY, overnight (i.e., 16 hours) at 4°C), in which after the plates were blocked with BSA (2 hours at 37°C). The next step was the incubation with capture antibody (2 hours at 37°C). The incubation with the unknowns, reference samples, and the standards took place overnight at 4°C (approximately 16 hours). The incubation with trapping antibody as well as the subsequent incubation with detection antibody was performed for 2 hours at ambient temperature. The incubation with substrate solution was performed in darkness for 30 minutes at ambient temperature. Color reaction was stopped by the addition of H2SO4 and the optical density was measured at 492 nm within 30 minutes. In each run, a reference preparation was run to check interassay variability and to monitor overall performance (22). The analytical sensitivity of the assay is 39 pg/mL. The precision profile showed a coefficient of variation (CV) of 20% at 45 pg/mL (i.e., functional sensitivity) decreasing to 7% at higher levels. For estimation of the accuracy of the method, a lyophilized reference preparation (marked 140799) is used. The mean hMIF concentration in 140799 was 20.7 ng/mL; the intraassay and the interassay CV amounted to 6.0% (n = 8) and 12.0% (n = 11 over a period of 13 months), respectively.
Measurement of Other Proinflammatory Markers
Blood was allowed to clot at room temperature and centrifuged. Serum samples were stored at –20°C until further processing. hsCRP, IL-1ra, and IL-6 were measured in serum using quantitative ELISAs (DiaMed Eurogen, Turnhout, Belgium). Serum was diluted 1000x for hsCRP, whereas IL-1ra and IL-6 were quantified in undiluted serum. Test samples and standard solutions were incubated on precoated ELISA plates. A biotinylated secondary antibody against the relevant inflammatory marker was used. Peroxidase conjugated streptavidin was then applied to bind to the biotinylated antibody and after washing, a tetramethylbenzidine-solution (TMB) was used to stain the remainder of the bound streptavidin; the reaction was stopped by adding sulfuric acid. Optical densities were read using a PowerWaveX Reader (MWG Biotech, Ebersberg, Germany) at 450 nm; data were calculated using KC4 software (MWG Biotech, Ebersberg, Germany). Neopterin was measured using a competitive ELISA (IBL-Hamburg, Hamburg, Germany) in accordance with the manufacturer’s recommendations. All serum samples were analyzed in duplicate. Sensitivity of all tests was calculated by the mean of six zero values + 3 standard deviations extrapolated on the standard curve. Intraassay and interassay CV for hsCRP was 5.1% (n = 10) and 14.3% (n = 7), respectively. Intraassay and interassay CV for IL-1ra, IL-6, and neopterin were 6.1% (n = 10) and 9.2% (n = 7), 5.5% (n = 10) and 6.8% (n = 7), and 3.6% (n = 11) and 7.6% (n = 6), respectively.
Assessment of New Cardiac Events
To investigate whether MIF is associated with new cardiac events in patients undergoing PCI, a research assistant and a cardiologist (F.W.H.M.B.) assessed the occurrence of new cardiac events (defined as pre-PCI, coronary artery bypass graft surgery (CABG), myocardial infarction, or cardiac death) through inspection of the medical records. Hospitalizations for unstable angina were not counted as new cardiac events. The mean cardiac follow up was 24 months. To ensure that no deaths were missed, the family physicians received a letter asking whether a patient was still alive and, if the patient had died, what had been the cause of death. A 100% cardiac follow up and cause of death was achieved, so no loss to follow up or loss of events occurred. Cardiac events were defined as events occurring at least 1 month after PCI (Fig. 1). Cardiac events occurring within 1 month after PCI were defined as early events and excluded from the present analyses, because a number of these events were missed as a result of the design of the study. Furthermore, early cardiac events are not a reflection of the progression of atherosclerosis but are usually the result of inadequate intervention, recoil of the vessel wall, or neointimal hyperplasia (an effect of the vascular damage resulting from PCI) (23).
Macrophage Migration Inhibitory Factor and Exhaustion
The question whether MIF is associated with exhaustion could not be studied with data obtained at the start of EXIT, because only exhausted patients undergoing PCI were included in EXIT. The MIVE was readministered 18 months after inclusion and allowed us to investigate whether exhausted and nonexhausted patients undergoing PCI displayed similar MIF concentrations. Many scientists approach the symptoms of fatigue, loss of energy, and increased irritability as clinical depression. Therefore, we also analyzed the association of MIF with major depression as assessed at baseline and 18 months by the SCID.
Statistical Analysis
Nonparametric testing was used as concentrations of MIF and inflammatory markers were not normally distributed (evidenced by Kolmogorov-Smirnov goodness of fit testing) even after log transformation. To test whether patients undergoing PCI have MIF concentrations that are different from those of the reference group, the Mann-Whitney U test was applied. To test whether MIF concentrations are associated with occurrence of new cardiac events; Mann-Whitney U test was applied to test for differences between baseline MIF concentrations in the group with late cardiac events versus the group without late cardiac events. In addition, baseline MIF concentrations were categorized into quartiles. Categorization of skewed distributions of inflammatory markers is common in risk prediction (24,25). A crosstab analysis with 2 significance testing was applied to explore the direction of the association of MIF with new cardiac events, if any. In addition, a 2 test for trend was performed because it is more informative with respect to whether the risk increases from lower to higher quartiles. To test whether MIF increased or decreased during follow up, the median concentrations observed at baseline, 6 months, and 18 months were compared using the Friedman rank test for related samples. The same procedure was used to investigate the changes in the expression of other inflammatory markers.
At 18 months, the MIVE was readministered. To establish whether patients undergoing PCI were exhausted or not at that time, the same cutoff was applied (i.e., MIVE 7) that was used to include patients in EXIT. Mann-Whitney U test was applied to test whether the MIF concentrations in the exhausted group were different from those in the nonexhausted group. The association of MIF with exhaustion at 18 months was also calculated using bivariate correlations of the continuous variables (Spearman’s correlation). We also investigated whether the intervention influenced MIF concentrations using multivariate analysis of variance. In this analysis, MIF data obtained at baseline, at 6 months, and at 18 months were included. Data processing and statistical analyses were performed using SPSS for Windows software, version 11.0 (SPSS, Chicago, IL). Significance levels were based on two-tailed tests with level <0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Macrophage Migration Inhibitory Factor Concentrations and Occurrence of New Cardiac Events in Patients Undergoing Percutaneous Coronary Intervention
Of the 194 patients, 2 (1%) experienced an early cardiac event and were excluded from the analysis. Of the remaining 192 patients, 31 (16%) experienced a late cardiac event. The "late cardiac event" group did not differ significantly from the "no event" group with respect to demographics (e.g., age, gender, smoking) and medical characteristics (e.g., blood pressure, diabetes, major depression), except that angiotensin converting enzyme inhibitor intake was more frequent in the "late cardiac event" group (2 = 7.1; p < .01). Mann-Whitney U showed that the MIF concentrations of the late cardiac event group was lower than those in the group without late cardiac events, although this association was not significant (Mann-Whitney U Z = –1.08; p = .28). To further investigate the association, MIF concentrations were categorized into quartiles (Q1 to Q4). MIF concentrations in Q1 ranged from 2.51 to 22.97 ng/mL, and 10 (21%) patients in this subgroup suffered from a late cardiac event. In Q2 (MIF: 22.98–39.59 ng/mL), seven (14%) patients experienced a late cardiac event. In Q3 (39.60–90.29 ng/mL) and Q4 (90.30–618.08 ng/mL), late cardiac events were observed in nine (18%) patients and five (10%) patients, respectively (2 = 2.27; df = 3; p = .52). Thus, a trend toward a decrease in cardiac events over the quartiles could be observed in the data. However, the 2 test for trend did not show a significant trend for new cardiac events (2 = 1.32; df = 1; p = .25). However, this trend was not significant. To control for the possible influence of angiotensin converting enzyme inhibitors, the data were also analyzed by Cox regression analysis controlling for angiotensin converting enzyme inhibitors. Results showed that the negative association could not be attributed to the use of angiotensin converting enzyme inhibitors (data not shown).
Expression Pattern of Macrophage Migration Inhibitory Factor and Other Proinflammatory Markers
The expression patterns of MIF, hsCRP, IL-1ra, IL-6, and neopterin over the 18 months of follow up are shown in Figure 2. The MIF concentration (Fig. 2A) increased significantly over 18 months of follow up (baseline: median 39.9 ng/mL; 95% range, 6.7–347.6; 6 months: median 53.4 ng/mL; 95% range, 4.9–321.3; 18 months: median 69.6; 95% range, 19.9–386.4; Friedman rank 2 = 18,7; p < .001). There were no significant increases or decreases in the expression patterns of hsCRP, IL-1ra, IL-6, or neopterin.
|
Macrophage Migration Inhibitory Factor Concentrations in Exhausted and Nonexhausted Patients Undergoing Percutaneous Coronary Intervention
At 18 months, 92 patients (51%) still felt exhausted (MIVE: mean 12.3; standard deviation (SD) ± 4.3), whereas 89 patients (49%) were no longer exhausted (MIVE: mean 2.7; SD ± 1.9). The median MIF concentration in the exhausted PCI group (63.4 ng/mL; 95% range, 19.9–440.5 ng/mL) was significantly lower than the median MIF concentration in the nonexhausted group (79.1 ng/mL; 95% range, 17.5–386.2 ng/mL; Mann-Whitney U Z = –2.3; p = .02; Fig. 3). Those who still fulfilled the inclusion criterion for exhaustion at 18 months were characterized by lower concentrations of MIF. However, when bivariate correlations were used to test the association between MIF and exhaustion, there was no significant association (r = –0.04; p = .56).
|
At baseline, 30 (16%) patients experienced major depression. Major depression was not associated with MIF (Mann-Whitney U Z = –0.73; p = .46). At 18 months, 11 (6%) patients experienced major depression. The median concentration of MIF in the depressed group (50.4 ng/mL; 95% range, 30.45–65.40 ng/mL) was significantly lower than the median concentration of the nondepressed group (74.7; 95% range, 19.9–429.4 ng/mL; Mann-Whitney U Z = –2.51; p = .01).
The Effect of the Intervention on Macrophage Migration Inhibitory Factor
There was no effect of the behavioral intervention on the concentrations of MIF (multivariate analysis of variance: F = 0.12; df = 3; p = .95).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
MIF is a proinflammatory cytokine, which can either directly or indirectly promote the production of a large number of other proinflammatory molecules, including cytokines (26). Because MIF is a marker of inflammation and because CAD is an inflammatory disease, one may expect that the median MIF value observed in patients undergoing PCI at baseline would be higher than the median value of a healthy reference group. We observed the opposite. There was a trend toward lower occurrence of late cardiac events in those with higher concentrations of MIF. However, this association was not significant. This may be the result of the limited power of the current study. Together, our data suggest that patients undergoing PCI are characterized by relatively low concentrations of MIF that tend to increase in an 18-month follow up.
Those who remained exhausted had significantly lower concentrations of MIF compared with those who were no longer exhausted at 18 months. These results suggest that MIF and exhaustion were negatively associated. Lower MIF concentrations were also observed among those who were depressed at 18 months. MIF is produced by a variety of sources such as immune cells and organs throughout the body, including the anterior part of the pituitary gland (26). It has been reported that in exhausted subjects, the activity of the HPA axis is decreased (16,27). These observations fit into earlier observations, which suggested that a decreased activity of the HPA axis may underlie the mental state of coronary patients who feel exhausted or depressed (16,27). No data on ACTH and cortisol were collected in the current study. However, Beishuizen et al. observed that MIF in plasma is persistently elevated in septic patients in parallel to plasma cortisol concentrations (28). This gives some additional support to the suggestion of a decreased HPA activity in exhausted, depressed patients. However, MIF and exhaustion were not correlated when exhaustion was assessed as a continuous variable. Furthermore, major depression at baseline was not related to MIF concentrations at baseline. Therefore, the data do not prove that low MIF concentrations are mainly observed in the patients who are exhausted or depressed. Further explorations of changes in MIF concentrations in those who stayed depressed or became depressed during the follow-up period failed because of the small numbers.
It is not likely that these results were influenced by immunosuppressive medication because all patients who used this type of medication (e.g., glucocorticoids, rheumatoid arthritis medicine, i.e., medications produced specifically to reduce inflammation) were excluded from the analyses. Other medications with possible antiinflammatory side effects (i.e., lipid-lowering drugs or statins) were used by 83% of all patients undergoing PCI; however, these medications were also used at 18 months of follow up and did therefore not explain an increase in MIF concentrations over the follow-up period. It is also unlikely that the results presented here were influenced by the behavioral intervention offered to half of the patients, because the intervention had no effect on MIF at all. The use of multivariate analysis of variance is debatable because of the skewed distribution of the data; therefore, we also made pairwise comparisons of changes in MIF in the intervention and the control group using nonparametric tests. These analyses confirmed the negative conclusion resulting from the multivariate analysis of variance analysis.
Several studies indicate that MIF neutralization might be a new target to combat the progression of atherosclerosis (29,30). Inhibition of MIF in vascular injured Apo-E-deficient mice resulted in stabilization of atherosclerotic plaques, which might be attributable to a reduction of monocyte recruitment mediated by endothelial MIF (6). Neutralization of MIF is reported to stop the proinflammatory effects of MIF and therewith the progression of atherosclerosis (30). However, when MIF is neutralized, no biologically active MIF is present. One of the most important actions of MIF is the host’s immunological defense against anomalies (e.g., bacterial infection, damage to vessel wall) (31). Neutralization of MIF will take this function away; furthermore, the activity of MIF in collateral circulation will also be blocked (9). There are several studies investigating the angiogenic properties of MIF supporting a positive effect for MIF in angiogenesis (10,32). One study supports a role for MIF as a therapeutic inducer of neovascularization in the development of collateral circulation in CAD (9). Therefore, trials neutralizing MIF in patients with CAD will have to be considered with extreme caution.
The technical assistance of Anneke Geurts-Moespot and Doorlene van Tienoven is greatly appreciated.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This research was supported by a grant from The Netherlands Heart Foundation (2001B074).
DOI:10.1097/PSY.0b013e31802b8750
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. E. Freedland, E. J.C. de Geus, R. N. Golden, W. J. Kop, G. E. Miller, V. Vaccarino, B. Brumback, M. M. Llabre, V. J. White, and D. S. Sheps What's in a Name? Psychosomatic Medicine and Biobehavioral Medicine Psychosom Med, January 1, 2009; 71(1): 1 - 4. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
