روابط بین پریشانی روانی، سبک های مقابله ای و واکنش پذیری محور HPA در افراد سالم

Abstract Psychological distress and coping styles have been suggested to relate to altered function in the hypothalamic-pituitary-adrenal (HPA) axis, although there remains much to be understood about their relationships. High and low cortisol levels (or reactivity) both represent HPA axis dysfunction, with accumulated evidence suggesting that they are linked to different types of psychopathology. The dexamethasone (DEX)/corticotropin-releasing hormone (CRH) test has been extensively used to identify HPA axis abnormalities in various psychiatric conditions including mood disorders; however, the possible associations of psychological distress and coping styles with HPA axis function have not been well documented using this test. Here, we examined the relationships of HPA axis reactivity as measured by the DEX/CRH test with subjectively perceived psychological distress and coping styles, both of which were assessed with self-report questionnaires, in 121 healthy volunteers. Subjects were divided into three groups by the cortisol suppression pattern, namely the incomplete-suppressors (DST-Cortisol ≥ 5 μg/dL or DEX/CRH-Cortisol ≥ 5 μg/dL), moderate-suppressors (DST-Cortisol < 5 μg/dL and 1 μg/dL ≤ DEX/CRH -Cortisol < 5 μg/dL), and enhanced-suppressors (DST-Cortisol < 5 μg/dL and DEX/CRH-Cortisol < 1 μg/dL). The enhanced-suppressors showed significantly higher scores in obsessive-compulsive, interpersonal sensitivity and anxiety symptoms and significantly more frequent use of avoidant coping strategy, compared to the other two groups. These results point to the important role of enhanced suppression of cortisol, or blunted cortisol reactivity, in non-clinical psychopathology such as avoidant coping strategy and greater psychological distress.

Introduction A wide variety of stress is associated with alteration in the hypothalamic-pituitary-adrenal (HPA) axis function. Studies looking at cortisol as the main output substance of the HPA axis have thus been critical to advancing our understanding of psychobiological underpinnings of various stress-related conditions (de Kloet et al., 2005 and Heim et al., 2000). For instance, perceived stress in everyday life (Pruessner et al., 1999), stressful situations such as academic examinations and seafaring (Droogleever Fortuyn et al., 2004 and Liberzon et al., 2008), self-reported symptoms (Van den Bergh et al., 2008), psychological coping styles (Nicolson, 1992 and O'Donnell et al., 2008), rejection sensitivity (Tops et al., 2008), sleep status (Backhaus et al., 2004, Lasikiewicz et al., 2008 and Wright et al., 2007) and personality profile (Tyrka et al., 2007) have been reported to be associated with alteration in cortisol levels. These studies in healthy subjects have investigated HPA axis function using several different cortisol measures including diurnal cortisol profiles, cortisol awakening response, and cortisol reactivity to psychosocial challenge tests such as Trier Social Stress Test (Kirschbaum et al., 1993). On the other hand, HPA axis function in clinical populations, particularly in patients with major depression, has been investigated with pharmacological challenge tests including dexamethasone (DEX) suppression test (DST, Carroll et al., 1976) and DEX/corticotropin-releasing hormone (CRH) test (Heuser et al., 1994a and Holsboer et al., 1987). The DEX/CRH test is an integrated challenge test for HPA axis function that combines DEX-pretreatment with CRH administration on the following day; thus, it is essentially a DST followed by CRH challenge. The merit of this combined test is that at the moment of CRH infusion the HPA axis is downregulated due to negative feedback induced by the DEX. In the DEX/CRH test a relatively high dose (i.e., 1.5 mg) of DEX is usually used, whereas DST studies, in particular those which examine the HPA function of post-traumatic stress disorder (PTSD), have used a lower dose (i.e., 0.5 mg or 1 mg) of DEX (e.g., Grossman et al., 2003 and Yehuda et al., 2004). Sensitivity of the DEX/CRH test in depressed patients has been shown to be high in prior studies including ours (Heuser et al., 1994a, Kunugi et al., 2004, Kunugi et al., 2006 and Watson et al., 2006b). Moreover, this test has revealed altered HPA axis function in those individuals with specific characteristics: dampened cortisol reactivity in healthy adults reporting childhood emotional abuse (Carpenter et al., 2009), increased cortisol responses in healthy adults reporting childhood parental loss with the exception of attenuated cortisol responses in those with parental desertion and low levels of care (Tyrka et al., 2008b), increased cortisol responses in healthy adults with certain personality traits (Tyrka et al., 2006 and Tyrka et al., 2008a), and attenuated cortisol responses in depressed women on job-stress-related longterm sickleave (Rydmark et al., 2006 and Wahlberg et al., 2009). On the other hand, the possible associations of more commonly presented psychopathology such as perceived distress in everyday life and coping styles with HPA axis function have not been well documented using the DEX/CRH test. However, these psychological measures are suggested to relate to altered cortisol level (Heim et al., 2000, Heim et al., 2002, Nicolson, 1992, O'Donnell et al., 2008, Pruessner et al., 1999 and Van den Bergh et al., 2008). For instance, severity of daily hassles in the past month was negatively related to cortisol concentrations (Heim et al., 2002). Perceived stress was positively, and burnout was negatively, associated with cortisol levels after DEX administration (Pruessner et al., 1999). Passive coping is suggested to relate to hypocortisolism (Heim et al., 2000). Healthy adults scoring high in either problem engagement or seeking social support showed lower cortisol levels (O'Donnell et al., 2008). Given these findings, it would be of interest to examine HPA axis function in relation to psychopathology at a non-clinical level such as psychological distress and coping styles by using the DEX/CRH test. Various kinds of psychiatric disorders have been shown to be associated with HPA axis hyperactivity as reflected by the high cortisol levels and impaired negative feedback inhibition due to an impaired corticosteroid receptor function (Holsboer, 2000). On the other hand, a number of psychoneuroendocrinological studies have demonstrated that a variety of conditions are associated with hypocortisolism, including low basal cortisol levels, enhanced sensitivity to the negative feedback, and blunted reactivity of provoked cortisol. Examples of psychiatric conditions characterized by hypocortisolism include PTSD, chronic fatigue syndrome, fibromyalgia and atypical depression (Fries et al., 2005 and Heim et al., 2000). Together, while both of these two extremes of cortisol activity can represent HPA axis dysfunction, they are likely to be linked to different types of psychopathology. Concerning hypocortisolism, there remains much to be clarified as to its natural course and meaning. Although hypocortisolism is considered to represent the result of prolonged stress exposure (Fries et al., 2005, Heim et al., 2000 and Ising et al., 2005), a condition so-called “allostasis” (McEwen, 2003), there also exists some evidence suggesting that this state could be a preexisting vulnerability to stress-related disorders (Delahanty et al., 2000, Wahlberg et al., 2009 and Yehuda et al., 2000). Arginine vasopressin (AVP), in addition to CRH, is an HPA axis secretagogue. AVP produced in parvocellular neurons of hypothalamic paraventricular nucleus (PVN) and secreted into pituitary portal vein system plays an important role in stress response (Herman, 1995 and Romero and Sapolsky, 1996). It is reported that, in chronic stress paradigms, the expression of AVP in parvocellular neurons increases and pituitary V1b receptor, through which AVP stimulates the ACTH secretion, up-regulates (Aguilera et al., 1994 and Aguilera and Rabadan-Diehl, 2000). There also exist clinical studies that support this notion. For example, de Kloet et al. (2008) have recently reported elevated plasma AVP levels in veterans with PTSD. Watson et al. (2006a) measured plasma AVP levels after pre-treatment of DEX in patients with chronic depression and those with bipolar disorder, and found significantly higher post-DEX AVP levels in the patient groups than in healthy controls, suggesting that post-DEX AVP levels could be more sensitive than baseline AVP levels in detecting HPA axis abnormalities. These findings raise the possibility that the post-DEX AVP measure may help understand whether the hypocortisolism, if present, is a result of chronic HPA axis overactivity or a preexisting vulnerability factor for psychopathology. In this context, the present study sought to examine the relationships between subjectively perceived psychological distress, psychological coping styles and the cortisol suppression pattern to the DEX/CRH test in non-clinical volunteers. We also examined the relationships of these psychological measures with the post-DEX AVP level to see whether the possible low cortisol levels would reflect allostatic shift or preexisting vulnerability. The study hypothesis was that the higher cortisol levels (or less suppression of cortisol) and/or lower cortisol levels (or more suppression of cortisol) would be related to greater distress and a unique pattern of coping strategies. If the low cortisol, together with elevation of AVP, is related to these psychological measures, it would indicate allostatic shift; while if the low cortisol, together with no elevation of AVP, is related to such psychological measures, it may indicate preexisting vulnerability.

. Results 3.1. Demographic characteristics of the subjects The numbers of incomplete-suppressors, moderate-suppressors, and enhanced-suppressors were 55, 54, and 12, respectively, indicating that the enhanced-suppressors corresponded to approximately the bottom 10% of total subjects for cortisol levels. The mean ages of incomplete-suppressors, moderate-suppressors, and enhanced-suppressors were 46.8 ± 14.3, 42.7 ± 15.1, and 44.4 ± 14.7, respectively. These three suppression groups did not significantly differ in age [F(2,118) = 1.49, p = 0.23]. Male/female ratios of incomplete-suppressors, moderate-suppressors, and enhanced-suppressors were 6/49, 15/39, and 7/5, respectively. There was a significant difference in sex distribution [χ2(2) = 13.6, p = 0.001]; males demonstrated significantly greater suppression than females. 3.2. Correlations between coping styles and psychological distress Table 1 shows the correlations between coping styles assessed with the WCCL and psychological distress assessed with the HSCL. Significant negative correlations were seen between problem-focused coping strategies (i.e., problem-solving and positive reappraisal) and greater psychological symptoms including interpersonal sensitivity and depression. In contrast, significant positive correlations were observed between emotion-focused coping strategies (i.e., self-blame, wishful thinking and escape-avoidance) and most of the symptom dimensions. Social support was not significantly related to any of the symptom dimensions. Table 1. Correlations between coping styles and psychological distress (Pearson's r). Somatization Obsessive-compulsive Interpersonal sensitivity Anxiety Depression Problem-solving −0.10 −0.22* −0.24* −0.17 −0.25* Positive reappraisal −0.08 −0.16 −0.26** −0.20* −0.32** Social support 0.11 0.14 0.14 0.16 −0.03 Self-blame 0.29** 0.47*** 0.52*** 0.49*** 0.48*** Wishful thinking 0.22* 0.42*** 0.37*** 0.40*** 0.31** Escape-avoidance 0.08 0.30** 0.33*** 0.30** 0.20* *p < 0.05, **p < 0.01, ***p < 0.001. Table options 3.3. Relationships between hormonal measures The cortisol values for the three suppressor groups on the three cortisol indices are provided in Table 2. DST-Cortisol of 64 subjects and DEX/CRH-Cortisol of 12 subjects fell under the detection limit, while DST-AVP did not fall under the detection limit in any subjects. DEX/CRH-Cortisol was significantly higher than DST-Cortisol in the whole sample, as expected (p < 0.001; Wilcoxon signed rank test). There was no significant correlation of DST-AVP with DST-Cortisol (ρ = 0.07, p = 0.50), DEX/CRH-Cortisol (ρ = 0.03, p = 0.75), or ΔCortisol (ρ = 0.02, p = 0.83). The three suppression groups did not significantly differ in DST-AVP [Kruskal–Wallis χ2(2) = 0.14, p = 0.93]. Table 2. Plasma cortisol concentrations (mean ± SD (range)) for the three subject groups, based on the suppression pattern. Incomplete-suppressors (n = 55) d Moderate-suppressors (n = 54) e Enhanced-suppressors (n = 12) f DST-Cortisola 1.4 ± 1.5 (0 ∼ 5.8) 0.4 ± 0.7 (0 ∼ 1.9) 0.1 ± 0.3 (0 ∼ 1.1) DEX/CRH-Cortisolb 10.0 ± 4.6 (5.0 ∼ 25.1) 2.5 ± 1.1 (1.1 ∼ 4.9) 0 ± 0 (0 ∼ 0) ΔCortisolc 8.6 ± 4.4 (2.3 ∼ 20.2) 2.1 ± 1.3 (−0.3 ∼ 4.8) −0.1 ± 0.3(−1.1 ∼ 0) a The concentration of cortisol [μg/dl] at 1500 h (i.e., immediately before the CRH challenge). b The concentration of cortisol [μg/dl] at 1600 h (i.e., 1 h after the CRH challenge). c Defined as "DEX/CRH-Cortisol minus DST-Cortisol". d Defined as “DST-Cortisol ≥ 5 or DEX/CRH-Cortisol ≥ 5”. e Defined as “DST-Cortisol < 5 and 1 ≤ DEX/CRH -Cortisol < 5”. f Defined as “DST-Cortisol < 5 and DEX/CRH-Cortisol < 1”. Table options 3.4. Correlations between hormonal and psychological measures No significant correlations were seen between DST-Cortisol and any measures of the two questionnaires (all p > 0.2). No significant correlations were seen between DEX/CRH-Cortisol and any measures of the two questionnaires (all p > 0.2) except for interpersonal sensitivity (ρ = −0.20, p = 0.03) in the HSCL. Similarly, no significant correlations were observed between ΔCortisol and any measures of the two questionnaires (all p > 0.2) except for interpersonal sensitivity (ρ = −0.21, p = 0.02) in the HSCL. No significant correlation was found between DST-AVP and any of the outcomes of the two questionnaires (all p > 0.1). 3.5. Relationships between psychological measures and DEX/CRH outcomes 3.5.1. Psychological distress and DEX/CRH outcomes Fig. 1 shows the relationships between 5 symptom dimensions of the HSCL and DEX/CRH suppression status. The ANCOVA on 5 symptoms controlling for age and sex showed significant main effects of the suppression status on obsessive-compulsive [F(2,114) = 5.19, p = 0.007], interpersonal sensitivity [F(2,114) = 5.43, p = 0.006], and anxiety [F(2,114) = 5.86, p = 0.004], symptoms. Post-hoc analyses with Bonferroni correction revealed that the enhanced-suppressors, compared to the other two groups or to moderate-suppressors alone, had significantly greater scores on these three symptom dimensions, while no significant differences were seen between incomplete- and moderate-suppressors ( Fig. 1). Comparisons of scores on the 5 dimensions of the Hopkins Symptom Checklist ... Fig. 1. Comparisons of scores on the 5 dimensions of the Hopkins Symptom Checklist (HSCL) between the three suppression groups. Black, white, and borderline bars are incomplete-suppressors (defined as “DST-Cortisol ≥5 or DEX/CRH-Cortisol ≥5”; n = 55), moderate-suppressors (defined as “DST-Cortisol <5 and 1 ≤ DEX/CRH -Cortisol <5”; n = 54), and enhanced-suppressors (defined as “DST-Cortisol <5 and DEX/CRH-Cortisol <1”; n = 12), respectively. *p < 0.05; **p < 0.01. Error bars represent standard errors of the mean. Figure options However, a considerable portion of the subjects fell into the incomplete-suppressors and thus we considered that this group would not necessarily represent those individuals whose cortisol levels were abnormally high. Therefore, to confirm the results obtained by the main analysis, the same ANCOVA was repeated with another grouping criterion as follows: ‘incomplete-suppressors’ to be individuals where either or both of DST- and DEX/CRH-Cortisols were equal to or more than 13 μg/dL, ‘enhanced-suppressors’ to be those whose DST-Cortisol was less than 13 μg/dL and DEX/CRH-Cortisol was less than 1 μg/dL, and ‘moderate-suppressors’ to be the remaining individuals. The reason why we here used the cortisol level of 13 μg/dL, instead of the original 5 μg/dL, as the cut-off value for the ‘incomplete-suppressors’ was that this value corresponded to approximately the top 10% of total subjects for cortisol levels. This 10% derived from the fact that the cortisol value of “enhanced-suppressors” corresponded to approximately the bottom 10% of total subjects. Using this new grouping, additional ANCOVA on the 5 symptoms controlling for age and sex was performed, again showing significant main effects of the suppression status on obsessive-compulsive [F(2,114) = 4.63, p = 0.012], interpersonal sensitivity [F(2,114) = 5.50, p = 0.005] and anxiety [F(2,114) = 5.81, p = 0.004] symptoms. Post-hoc analyses with Bonferroni correction revealed that the enhanced-suppressors, compared to the other two groups or to moderate-suppressors alone, scored significantly higher on these three symptom dimensions, while no significant differences were seen between incomplete- and moderate-suppressors (data not shown). 3.5.2. Coping styles and DEX/CRH outcomes The relations between the 6 different coping styles of WCCL and suppression status are provided in Fig. 2. The ANCOVA on the 6 coping subscales controlling for age and sex showed a significant main effect of the suppression status on escape-avoidance [F(2,95) = 5.26, p = 0.007]. Post-hoc analyses with Bonferroni correction revealed that the enhanced-suppressors, compared to the other two groups, had significantly greater scores on this coping strategy, while no significant differences were found between incomplete- and moderate-suppressors ( Fig. 2). Comparisons of scores on the 6 subscales of the Ways of Coping Checklist (WCCL) ... Fig. 2. Comparisons of scores on the 6 subscales of the Ways of Coping Checklist (WCCL) between the three suppression groups. Black, white, and borderline bars are incomplete-suppressors (defined as “DST-Cortisol ≥5 or DEX/CRH-Cortisol ≥5”; n = 49), moderate-suppressors (defined as “DST-Cortisol <5 and 1 ≤ DEX/CRH-Cortisol <5”; n = 43), and enhanced-suppressors (defined as “DST-Cortisol <5 and DEX/CRH-Cortisol <1”; n = 10), respectively. *p < 0.05; **p < 0.01. Error bars represent standard errors of the mean. Figure options The additional ANCOVA with the other grouping criterion of suppression status on the 6 coping subscales showed significant main effects of the suppression status on wishful thinking [F(2,95) = 3.31, p = 0.041] and escape-avoidance [F(2,95) = 5.56, p = 0.005]. Post-hoc analyses with Bonferroni correction revealed that the enhanced-suppressors, compared to the other two groups or to incomplete-suppressors alone, scored significantly higher on these two coping strategies, while no significant differences were seen between incomplete- and moderate-suppressors (data not shown).