ارزیابی کنترل شده تصادفی از اثرات مدیریت استرس شناختی رفتاری بر پاسخ کورتیزول به استرس حاد در افراد سالم

Psychosocial stress is a potent activator of the hypothalamus–pituitary–adrenal (HPA) axis. While neuroendocrine stress responses are essential for the maintenance of homeostasis, evidence suggests that excessive activation of the HPA axis constitutes a risk for disease and psychopathology. The purpose of the present study was to assess the effect of cognitive–behavioral stress management training on endocrine stress responses and cognitive appraisal under acute psychosocial stress among healthy young subjects. Forty-eight healthy, non-smoking male students without acute or chronic medical or psychiatric disorder on self report were randomly assigned to receive group-based cognitive–behavioral stress management training either before or after a standardized psychosocial stress test (Trier Social Stress Test, TSST). Endocrine and psychological stress responses were assessed with salivary free cortisol response and cognitive appraisal processes to the TSST. In comparison with the control group, subjects in the treatment group showed an attenuated endocrine response (F (2.55/117.41) = 3.81; P = 0.02; effect size f2 = 0.35) to the TSST. In addition, subjects in the SIT group had lower stress appraisal and higher control expectancies (F (2/45) = 6.56; P = 0.003, effect size f2 = 0.29) compared to controls. Short group-based cognitive–behavioral stress management training reduces the neuroendocrine stress response to an acute stressor in healthy subjects. Therefore, stress management training may prove useful in preventing detrimental effects of stress-induced neuroendocrine activation

Psychosocial stress leads to the activation of several physiological stress responses, which in the short term are essential for the maintenance of homeostasis. Being the principal endocrine component of the stress response, the hypothalamus–pituitary–adrenal (HPA) axis mediates a multitude of adaptive physiological and psychological processes. It influences cardiovascular function, fluid volume and hemorrhage, immunity and inflammation, metabolism, neurobiology, and reproductive physiology (Sapolsky et al., 2000). Besides physiological influences (Kirschbaum et al., 1999), the induction of a HPA axis response is regulated by psychological factors. For example, individuals with low self-esteem and negative self-concept failed to habituate to a repeated standardized stressor (Kirschbaum et al., 1995). Furthermore, experimental variation of verbal comments prior to an experimental stressor significantly influenced the subsequent cortisol responses (Rohrmann et al., 1999). Thus, personality factors and cognitive appraisal processes not only play an important part in determining what is stressful (Lazarus and Folkman, 1984), but also modulate the extent and the habituation of the HPA axis response to stress. With regard to possible underlying central nervous system stress circuits involved in the neuroendocrine stress responses, the distinction between ‘systemic’ and ‘processive’ stress pathways has been proposed (Herman and Cullinan, 1997). In contrast to ‘systemic’ stress pathways, which are activated during direct threat of survival, such as hemorrhage and hypoglycemia, and thus do not require integrative processing of higher-order brain structures, psychosocial stressors usually involve the processing of multiple sensory inputs on cortical and limbic levels, including the cerebral cortex, hippocampus, and amygdala. These regions are known to be involved in the cognitive and emotional processing of potentially threatening stimuli and are most likely the primary integrators of the anticipatory stress response, leading to the modulation of neuroendocrine paraventricular output on the hypothalamic level (Herman et al., 2002). The psychological equivalent of these processes could be seen in the stress appraisal processes proposed by Lazarus and Folkman (1984). Consequently, stress-reducing psychosocial interventions aimed at modifying cognitive appraisal are a possible means of influencing HPA axis activity under stress. A number of studies have demonstrated the effectiveness of cognitive–behavioral stress management in influencing psychological and physiological parameters and health outcomes in health and disease. For example, in symptomatic HIV-positive gay men, ten-week long, group-based cognitive–behavioral stress management training has been shown to reduce symptoms of distress and urinary free cortisol output (Antoni et al., 2000) and to prevent increases in the cortisol/dehydroepiandrosterone-sulfate ratio (Cruess et al., 1999). In women with breast cancer, similar cognitive–behavioral stress management training after surgery led to a reduction in evening serum cortisol levels (Cruess et al., 2000). Given that cognitive–behavioral stress management interventions have been shown to influence stress-relevant physiological parameters, we evaluated the effects of short-term, group-based cognitive–behavioral stress management training on endocrine responses and cognitive appraisal under acute stress in a population of healthy young male students.

This study demonstrates that short, group-based, cognitive–behavioral stress management training reduces the salivary free cortisol stress response to an acute stressor in healthy male subjects. As indicated by analysis of covariance, these endocrine response differences were influenced by the observed differences in the cognitive appraisal of the situation. Subjects in the treatment group appraised the situation as less stressful and displayed more competence in coping with the situation. According to reported conventions, all reported effect sizes for significant endocrine and psychometric group differences in the TSST were large, whereas the pre–post changes in perceived stress were of medium effect size (Buchner et al., 1997). To guarantee internal validity, we randomized subjects and controlled for stress-relevant physiological and psychological factors. A comparison of demographic and psychometric parameters confirmed the randomization of subjects. Furthermore, since both groups had inconspicuous scores in the personality and stress scales, it is unlikely that participants were particularly stressed or stress prone. Thus, we assume that the reported results are not influenced by pre-existing differences between the groups or selection bias. The salivary free cortisol responses observed in our sample is somewhat higher than those published by other groups using the TSST. This could be a consequence of the altered TSST protocol we used in order to obtain data concerning the anticipatory appraisal processes. However, this difference in the response magnitude does not seem to be a result of group differences, since groups did not differ significantly in the basal cortisol levels and the respective psychometric scales. We excluded female subjects because their inclusion would have required the control for menstrual cycle phase and for the use of oral contraceptives (Kirschbaum et al., 1999). Although the exclusion of female subjects affects the external validity of the study, we have decided against the inclusion of women in order to enhance the internal validity. However, since gender differences of HPA axis stress responses seem to be mediated through differences in sex hormone levels and the observed response differences in our study were mediated through differences in the cognitive appraisal, we are confident that the stress inoculation training has similar neuroendocrine effects in women. Because a repeated exposure to the TSST leads to a habituation of the neuroendocrine stress response in a majority of subjects (Kirschbaum et al., 1995 and Pruessner et al., 1997), we have decided against a repeated, i.e., pre- and post-treatment/control condition assessment of neuroendocrine stress reactivity. The employed study design can not rule out that the HPA axis response differences observed could be a result of pre-existing group differences in stress reactivity, independent of the stress inoculation training. However, the results of the psychometric personality and stress scales, both in pre- and in pre–post-training comparison between the groups, speak against this possibility. Also, because the treatment group received their training before the TSST and thus could have been more acquainted with testing circumstances, it could have been possible that groups differed with regard to their perception of novelty in the testing situation, which could consequently influence the neuroendocrine stress response. However, groups did not differ significantly in their perception of novelty nor did this factor have a significant influence on the salivary free cortisol response. With regard to the fact that the salivary free cortisol response differences between the treatment and control group were mediated through differences in the cognitive anticipatory appraisal processes, it seems appropriate to assume that the stress inoculation training exerted its effects through the modulation of ‘processive stress pathways’, thus influencing the cognitive and affective processing of stressful stimuli (Herman and Cullinan, 1997). From a psychotherapeutical perspective, a prerequisite of cognitive, emotional, and consequently behavioral changes induced by psychotherapy is the activation and self-awareness of relevant cognitive schemes and the alteration of dysfunctional aspects by ‘corrective emotional experience’ (Grawe, 2002). It is possible that through the use of experiential exercises, role-plays, and the encouragement to practice the stress-reducing techniques at home, these mediators of psychotherapeutical change were effectively employed by the stress inoculation training. This is the first study to report that short, group-based, cognitive–behavioral stress management training attenuates the endocrine and psychological response to acute stress in healthy subjects. Alterations of HPA axis functioning have been linked to the development and maintenance of psychosomatic and psychiatric disorder (Ehlert et al., 2001) and somatic illness (McEwen, 1998). According to the concept of allostatic load, which represents a marker of cumulative biological burden exacted on the body through attempts to adapt to life’s demands, several conditions of how stress leads to alterations of the HPA axis can be distinguished. These include repeated activation during chronic stress and failure to habituate to repeated stressors (McEwen, 2000). With the observed attenuation of the neuroendocrine stress response and the changes in cognitive appraisal of the stress situation, it is possible that group-based, cognitive–behavioral stress management training could prove useful in preventing detrimental consequences of stress-induced neuroendocrine responses, such as the risk of developing hypertension (al’Absi and Arnett, 2000 and al’Absi et al., 1998) and metabolic syndrome X (Bjorntorp and Rosmond, 1999). However, it is important to note that we have not assessed the effects of short, group-based, cognitive–behavioral stress management training on markers of allostatic load, but rather on mechanisms that have been discussed to lead to the development of allostatic load. There is consensus that the relation between HPA axis parameters and health is not linear, thus both too much and too little HPA axis activity and reactivity can be linked to disease and health complaints (McEwen, 1998). As a consequence, our finding of a reduced neuroendocrine stress response should not be considered to be protective per se, but rather with regard to its possible role in the development of stress-related health complaints. Since cortisol has been considered a primary mediator in the development of allostatic load (Seeman et al., 2001), further studies are needed to evaluate possible long-term effects of the neuroendocrine response differences we observed. Recently, increased pituitary sensitivity to psychosocial stress, possibly due to severe early life stress, has been described as a biological risk factor for psychopathological conditions in adulthood (Heim et al., 2000). As we have employed the identical psychosocial stress protocol, our findings indicate that similar cognitive–behavioral interventions could be a useful non-pharmacological approach for the prevention of psychopathological conditions related to early-life stress. However, before definite conclusions are drawn, it needs to be shown that the observed neuroendocrine effects persist over a longer period of time and generalize across different stress situations.