Journal: Clin Endocrinol (Oxf), 2005 Mar;62(3):343-8.
Authors: Inder WJ, Prickett TC, Mulder RT.
Affiliations: Department of Endocrinology, Christchurch Hospital, and Department of Psychological Medicine, Christchurch School of Medicine, Christchurch, New Zealand.
NLM Citation: PMID: 15730417
Design: Single-blinded case-control study measuring functional and psychological status, basal hormonal parameters and ACTH/cortisol response to naloxone and ovine corticotrophin-releasing hormone (oCRH) vs. placebo in people with CFS and healthy controls.
Patients: Twelve people with CFS and 11 age-matched controls.
Hormonal parameters: basal levels of 09:00 h plasma cortisol, dehydroepiandrosterone sulfate (DHEAS) and IGF-1. 24-h urinary free cortisol. Plasma ACTH and cortisol response to naloxone 125 microg/kg, oCRH 1 microg/kg and placebo (normal saline).
Psychological parameters: SF-36, Hamilton Depression Score, Hospital Anxiety and Depression Scale and Fatigue Scale. Results: There were highly significant differences between the CFS subjects and the controls with respect to the measures of fatigue and physical functioning. However, there were no differences in basal levels of 09:00 h cortisol (367 ± 37 vs. 331 ± 39 nmol/l, P = 0.51), DHEAS (4.2 ± 0.6 vs. 4.0 ± 0.5 micromol/l, P = 0.81), 24-h urinary free cortisol (182 ± 27 vs. 178 ± 21 nmol/24 h, P = 0.91) or IGF-1 (145 ± 19 vs. 130 ± 11 microg/l, P = 0.52) between the CFS group and controls, respectively. There was also no difference between the groups with respect to the ACTH and cortisol response to either oCRH or naloxone.
Conclusions: Our data do not support an aetiological role for deficiency in central opioids or the HPA axis in the symptoms of CFS.
[ . . . ]
This work provides evidence contrary to the notion that significant symptomatic CFS is associated with a blunting of the HPA axis. Some previous studies have uncovered a variety of perturbations, including lower 24-h urinary free cortisol levels, blunted ACTH responses to CRH 4,22 and naloxone 12 and diminished adrenal response to ACTH 1-24. Our results are more in keeping with those of Cleare's group, who have shown no difference in HPA response to low-dose ACTH, insulin hypoglycaemia, d-fenfluramine or human CRH between subjects with CFS and controls, although a reduction in 24-h urinary free cortisol was observed. Similarly, no reduction in salivary cortisol levels was noted by Wood et al.; in fact a small but significantly higher mean salivary cortisol level was observed in the CFS subjects over a 16-h period.
Therefore, there have been a number of studies in subjects with CFS where no evidence for a relative deficiency in HPA axis function has been found. The reasons for the differences are unclear, but do not appear to be related to our CFS subjects being less symptomatic, as determined by the functional questionnaires showing major degrees of physical, emotional and social impairment. CFS is a heterogeneous disorder, and differences in HPA axis function between different studies performed in different countries may simply reflect this. The studies by Scott et al. and Demitrack et al. were performed in the afternoon/evening as opposed to the morning. However, the baseline cortisol levels reported by Scott et al. were similar to those we observed at 10:00 h. The study by Demitrack et al., which showed a reduced integrated ACTH response to oCRH, was characterized by similar peak ACTH levels between the groups and no difference in cortisol response. Therefore, the difference in integrated response could be solely accounted for by higher basal ACTH levels at the start of the test. Performing the oCRH test in the afternoon generally results in a greater ACTH response because of a reduction in negative feedback arising from a lower baseline cortisol, but incremental response to naloxone is similar at different points along the diurnal rhythm. We allowed oestrogen use, as did Cleare et al. 24 This has been demonstrated previously to have no effect on the ACTH and cortisol response to oCRH, 26 but to our knowledge no similar study has examined the effects of oestrogen on the ACTH and cortisol response to naloxone.
It is not known what the effects of low endogenous opioids are on mental and physical well-being. However, a previous study investigating the effects of naloxone infusion during exercise has shown that blockade of the opioids released during exercise resulted in more subjective fatigue and perceived effort. Prolonged infusion of naloxone in normal volunteers results in cognitive impairment and dysphoria. Our a priori hypothesis was that a significant proportion of people with CFS would show evidence of opioid deficiency and that this might partially account for some of the observed symptoms. This would also provide a plausible physiological mechanism for the observed benefits of exercise in people with CFS. The study by Scott et al. had hypothesized that CFS might be associated with an increase in central opioid tone, to account for the previously reported blunting of the HPA axis, but found the reverse. While exogenous opioids are known to suppress ACTH and cortisol, and blockade of endogenous opioids with naloxone results in an increase in ACTH in most normal subjects, there is no convincing evidence that an increase in central opioid tone (as determined by the ACTH response to naloxone) is associated with lower basal ACTH and cortisol levels. In fact our previous work suggests that an increase in central opioid tone may be associated with higher basal ACTH levels. The CFS subjects in our study had no evidence of any reduction in central opioid tone compared to a matched group of healthy controls.
While a type II error is possible as the sample size of our study is small, it is not dissimilar in subject number to that of previously published studies investigating the HPA axis in CFS and also other patient groups by several investigators. Our CFS subjects fulfilled the diagnostic criteria for the condition, and had marked functional impairment demonstrated on standardized tests consistent with previously reported abnormalities. The use of multiple tests that have all produced concordant results is also a strength of the study, and it is generally consistent with Cleare's data, with the exception of the 24-h urinary free cortisol. While this study does not exclude a minor perturbation in the HPA axis of some subjects with CFS, the lack of any significant difference in our patient group suggests that an alteration in either HPA axis function or opioid tone is unlikely to account for the functionally limiting symptoms that the subjects with CFS displayed.
Subjects were blinded to the agent received (naloxone, oCRH or normal saline). The pharmacological agent was administered by a trained endocrine nurse. Assays were performed by T.P., with samples from all three occasions run in the same assay. Subject symptomatology was assessed by self-reported questionnaire, avoiding the potential for observer bias. While the principal investigator (W.I.) was not blinded either to patient group or agent received during the data analysis, this is unlikely to have changed the nature or outcome of the biochemical results.
Investigators have examined the effect of cortisol replacement in CFS. A randomized controlled trial of low-dose hydrocortisone (5-10 mg per day) has demonstrated a statistically significant improvement in fatigue symptoms. However, it is unclear whether the subjects that benefited were those with relative cortisol deficiency. Another study of hydrocortisone replacement therapy in CFS used higher doses (13 mg/m2 in the morning and 3 mg/m2 in the afternoon) and showed only marginal symptomatic benefit and significant adrenal suppression in a number of patients. Recently published guidelines for the management of CFS in Australia do not recommend the routine use of hydrocortisone. Exercise and cognitive behavioural therapy remain the treatments for CFS with the best efficacy as shown by randomized controlled trials. Our study fails to implicate abnormalities of the HPA axis as a significant causative factor for the symptoms of CFS. Further research in this area is required to determine the extent of any neuroendocrine dysfunction in CFS in a larger number of subjects in different populations, and quantify the neuroendocrine response to specific therapies.
Copyright © 2005 Co-Cure
Last Revision: March 6, 2005
Please report any problems with this page to the Webmaster.