Hypothalamic-Pituitary-Adrenal Axis Function in Chronic Fatigue Syndrome

There is evidence for a hypofunction of the hypothalamicpituitary-adrenal (HPA) axis in a proportion of the patients with chronic fatigue syndrome (CFS), despite the negative studies and methodological difficulties. In this review, we focus on challenge studies and on the role of the HPA axis in the pathogenesis of CFS. Mild hypocortisolism, blunted adrenocorticotropin response to stressors and enhanced negative feedback sensitivity to glucocorticoids are the main findings. Several underlying mechanisms have been proposed. Currently, it is a matter of debate whether these disturbances have a primary role in the pathogenesis of CFS. However, even if the HPA axis dysfunctions are secondary to other factors, they are probably a relevant factor in symptom propagation in CFS. Copyright © 2007 S. Karger AG, Basel Received: August 29, 2006 Accepted after revision: March 17, 2007 Published online: June 27, 2007 Filip Van Den Eede, PhD Department of Psychiatry, University Hospital Antwerp (UZA) Wilrijkstraat 10 BE–2650 Edegem (Belgium) Tel. +32 3 821 49 11, Fax +32 3 825 16 41, E-Mail [email protected] © 2007 S. Karger AG, Basel 0302–282X/07/0552–0112$23.50/0 Accessible online at: www.karger.com/nps HPA Axis Function in Chronic Fatigue Syndrome Neuropsychobiology 2007;55:112–120 113 this review, we will mainly focus on HPA axis-related challenge studies and on possible pathophysiological mechanisms in CFS. HPA Axis Disturbances in CFS Basal Hormonal Changes Studies on basal plasma cortisol (single or serial measures), free salivary cortisol and urinary free cortisol have been reviewed extensively by Parker et al. [15] and by Cleare [2] . In summary, in about half of the investigations there was evidence for lowered cortisol levels in CFS. There is only 1 report of elevated salivary cortisol levels in CFS [16] . In all the other studies, no differences were found between CFS patients and control individuals. More recently, the study by Jerjes et al. [17] provided further evidence for reduced basal HPA axis function in CFS. The group of 15 CFS patients without psychiatric comorbidity showed lower urinary free cortisol and corticosterone concentrations than the group of 20 healthy control individuals, whereas diurnal rhythm was normal in CFS patients. Furthermore, Roberts et al. [18] reported a lower salivary cortisol response to awakening in 56 CFS patients compared with 35 control individuals. In contrast, Di Giorgio et al. [19] found no abnormalities in the levels of plasma cortisol in a sample of 15 CFS patients, although they reported reduced levels of adrenocorticotropic hormone (ACTH) over a full circadian cycle and during the physiological morning peak. Finally, regarding basal hormone concentrations in CSF, Demitrack et al. [20] measured cerebrospinal fluid levels of corticotropin-releasing factor (CRF) and ACTH in 19 CFS patients and 26 control individuals, but no differences were apparent. The authors considered this finding to be ‘inappropriately normal’ in CFS, given the reduction in glucocorticoid secretion in the periphery (and thus reduced negative feedback) [20, 21] . Challenge Tests Hypothalamic and Pituitary Function Table 1 presents an overview of HPA axis-related challenge tests that have been undertaken in CFS. The insulin tolerance test (ITT) is a standardized method for assessing the entire HPA axis [22] . To the best of our knowledge, only 1 study demonstrated reduced ACTH responses in the ITT in CFS [23] . In 2 relatively large studies [24, 25] and 1 small study [26] , there were no significant differences in ACTH responses in the ITT between CFS patients and healthy control individuals. These data do not provide strong evidence for a disturbed HPA axis regulation in CFS, although the ITT may be too blunt to be useful in detecting more subtle changes [2] . Demitrack et al. [20] and Scott et al. [27] reported significantly lower ACTH responses after CRF in CFS patients in comparison with healthy individuals, whereas two other research groups did not find any differences in the CRF challenge [24, 28] . In addition, Altemus et al. [29] argued that the ACTH response to arginine vasopressin (AVP) acts as an indirect index of hypothalamic CRF levels. They found a trend towards a reduced ACTH response to AVP in CFS, which they attributed to a lower ambient level of hypothalamic CRF. Alternatively, Scott et al. [30] hypothesized that a deficit in endogenous AVP could contribute to the attenuated ACTH response in the CRF test. Using desmopressin (DDAVP), an AVP analog, they found that coadministration of DDAVP with CRF normalized the blunted ACTH response to CRF. Consequently, they suggested that CFS may be associated with AVP deficiency and upregulated AVP receptors on the pituitary. Furthermore, blunted ACTH responses to other stressors have been reported in CFS: naloxone [31] , exercise and social stress [23] . Finally, in the study by Gaab et al. [32] , patients with CFS showed an enhanced and prolonged suppression of salivary free cortisol after the administration of a low-dose of dexamethasone (0.5 mg), pointing to enhanced sensitivity to the negative feedback of glucocorticoids at the pituitary level. In accordance with these findings, we observed lower salivary free cortisol responses in the combined low-dose dexamethasone/CRF test in CFS patients than in control individuals [Van Den Eede, et al., unpubl. data]. Adrenal Function In the above-mentioned investigations, an inconsistency can be observed in the relationship between ACTH and cortisol responses in challenge tests. More specifically, a reduced ACTH response in CFS patients has been associated with: (1) lower cortisol responses [27] (but without any correlation between the two variables); (2) no difference in cortisol responses [20, 23, 31] ; (3) higher salivary cortisol [23] ( table 1 ). In order to clarify this issue and to examine adrenal function in CFS, several ACTH challenge studies have been performed. In 2 studies with 1 g of ACTH or more, there were no differences in cortisol responses between CFS patients and healthy individuals [33, 34] , suggesting that adrenal insufficiency is unlikely to play a significant role in CFS. In contrast, Scott et al. [35] and Demitrack et al. [20] reported reduced cortisol responses in the ACTH test. The findings of chal-