Hypothalamicpituitaryadrenal axis dysfunction in chronic fatigue syndrome

[kaffiend]

I haven't gone through this yet, but it's free access so everyone should be able to read it. At first glance, a statement in the abstract is inconsistent with previous results by authors (low cortisol predicts CBT non-responders).

They're clearly shilling for CBT, but there might be some good information for people who want to understand the HPA axis. The existing evidence is incomplete without studies that use a physical stressor challenge and measure endocrine responses at 24 and 48 hours post-challenge.


Hypothalamicpituitaryadrenal axis dysfunction in chronic fatigue syndrome

 

[adreno]

SSRIs also increase cortisol levels, in those who are low. So does SJW.

 

[kaffiend]

Key points [pasted directly from the paper]

• The bulk of evidence points to modest reductions in cortisol levels in some cohorts of patients with chronic fatigue syndrome (CFS), and these changes are more apparent in women than in men
• Underlying the reduction in cortisol levels is a hypothalamicpituitaryadrenal (HPA) axis with attenuated diurnal variation, enhanced negative feedback and blunted response to challenges
• Low cortisol levels have clinical relevance as they might contribute to symptomsalong with other factorsand are associated with a worsened outcome of currently recommended treatments for CFS
• A multidimensional etiological model remains most probable, with low cortisol levels occurring at a later stage of the illness, moderated by factors such as activity levels, depression, early-life stress and psychotropic medication
• Cortisol levels can be increased by treatment with cognitive behavioral therapy, potentially because of reversal of some moderating factors
• Further improvements in research designs remain necessary to fully understand HPA axis dysfunction in CFS

----

The evidence cited in the paper for a multidimensional etiological model is extremely weak. This could have been a good review if it didn't try to push the flimsy support for a psychological model. There is a single study cited in support of an association with early life stress and other statements making such claims are made without any citations whatsoever.

 

[oceanblue]

Thanks Kaffiend

Given what is now a fairly consistent pattern of findings for the type of HPA axis changes found in CFS, we recommend that future work focuses on improving our understanding of the cause and relevance of these observed changes.
(my italics)

Further improvements in research designs remain necessary to fully understand HPA axis dysfunction in CFS

Maybe they need to nail the nature of the changes first; walk before they can run. Did they cite this paper by Wessely and Cleare that found reduced urinary levels of cortisol were not corroborated by urinary cortisol metabolite levels, undermining the low cortisol finding?

This could have been a good review if it didn't try to push the flimsy support for a psychological model.

Story of CFS research in general. A recent review of immune and neurological dysfunction by Komaroff made a great job of proposing an interesting hypothesis while being clear about the limitations of the evidence.

 

[Marco]

Key points [pasted directly from the paper]

• The bulk of evidence points to modest reductions in cortisol levels in some cohorts of patients with chronic fatigue syndrome (CFS), and these changes are more apparent in women than in men
• Underlying the reduction in cortisol levels is a hypothalamicpituitaryadrenal (HPA) axis with attenuated diurnal variation, enhanced negative feedback and blunted response to challenges
• Low cortisol levels have clinical relevance as they might contribute to symptomsalong with other factorsand are associated with a worsened outcome of currently recommended treatments for CFS
• A multidimensional etiological model remains most probable, with low cortisol levels occurring at a later stage of the illness, moderated by factors such as activity levels, depression, early-life stress and psychotropic medication
• Cortisol levels can be increased by treatment with cognitive behavioral therapy, potentially because of reversal of some moderating factors
• Further improvements in research designs remain necessary to fully understand HPA axis dysfunction in CFS

----

The evidence cited in the paper for a multidimensional etiological model is extremely weak. This could have been a good review if it didn't try to push the flimsy support for a psychological model. There is a single study cited in support of an association with early life stress and other statements making such claims are made without any citations whatsoever.

Couldn't agree more kaffiend.

The review of the evidence for hypocortisolism is very careful and they make valuable proposals for improving and standardising test methods.

They then conclude however that it is most likely an epiphenomenon and they then go completely off the rails with their risk factors and 'accepted treatment'.

Fries et al (2005) propose a different explanation for this 'epiphenomenon' :


A new view on hypocortisolism
Eva Fries, Judith Hesse, Juliane Hellhammer, Dirk H. Hellhammer

Sickness behavior at the behavioral level appears to be the expression of a central motivational state that reorganizes the organisms priority to cope with infectious pathogens (Hart, 1988). Therefore, Van Hoof et al. (2003) argue that the state of atypical depression or chronic fatigue is essentially characterized by a decrease in energy consumption during periods of sickness or injury to promote subsequent recuperation. Thus, the disease may be adaptive to conserve energy during threats beyond the organisms ability to cope and thereby serving an important function for survival.

All these examples demonstrate that reduced HPA axis reactivity with low cortisol levels is not only maladaptive but, contrarily, may have even beneficial and advantageous effects for the organisms survival.

Based on these results we propose that hypocortisolism is a protective response dampening chronic HPA axis activity and thereby reducing the damaging effects of the glucocorticoid response to daily hassles at the expense of symptoms such as high stress sensitivity, pain, and fatigue.

http://cfids-cab.org/cfs-inform/Hypotheses/fries.etal05.pdf

Of course I could be accused of cherry picking.

 

[August59]

My HPA is ery abnormal and my cortisol rythm is not even close in resemblence. My cortisol is close to zero in the a.m. then goes to a peak at around 4 p.m. with the same level of decline from 4 p.m back to 8 a.m. Done a 24 hour urine for cortisol to look at total output and I was only clearing 10 (whatever the units are). I've got to go back and re-read it all because it was a little confusing. I'm either not making quit what the saliva test indicates by the curve supllied or I'm not making enough at the high part of the curve.

My testosterone and thyroid is abnormal to. I finally went to an integrative doctor tha chaecked my free T4 and free T3. They have raised my synthroid 3 times and my TSH drops very little. The free T4 was real good, but my free T3 was right at the edge of low range. He might add Isocort in the a.m. and start a low dose of Cytomel, but will not know that till next week.

 

[Snow Leopard]

The problem with hypocortisolism is that it is nonspecific. It can be caused by simple distruption of our bio/sleeping rhythms, it can be caused due leading a low-stress lifestyle. Oh and gluticorticoids have important metabolic effects in the body, so dietary patterns can also change the pattern of expression.

Fries et al. as well as authors of other articles have often speculated that hypocortisolism results from overresponsiveness of the rest of the gluticorticoid pathways. The problem with this hypothesis is that when gluticorticoid receptor expression (eg mRNA) or activity is measured we find that it too is damped/low (in most studies, the remainder found normal levels). So the suppression of the HPA axis seems to be either further downstream or further upstream than hypothesised.
In any event, this means that the simplistic chronic stress->ongoing dysregulation of HPA axis assumption is almost certainly wrong. While chronic stress may indeed be a risk factor for onset, it does not seem to be a key maintaining factor in many cases.

 

[Dolphin]

Maybe they need to nail the nature of the changes first; walk before they can run. Did they cite this paper by Wessely and Cleare that found reduced urinary levels of cortisol were not corroborated by urinary cortisol metabolite levels, undermining the low cortisol finding?

They did actually

Basal cortisol in urine

Of the three studies that reported 24 h urinary free cortisol
(UFC) excretion (Table 2), two measured cortisol output
in 24 h specimens and found no differences,13,23 whilst in
the third study Jerjes et al.24 compared cortisol excretion
in patients and in controls in five 3 h urine collections over
a 15 h period (06002100 h). All five 3 h collections from
patients had significantly lower UFC levels compared with
controls, except the last one (18002100 h).

Because it is often argued that the 24 h UFC value might
not be an accurate indicator of basal HPA function,6,25
Jerjes et al.25 also measured total urinary cortisol metabolite
(TUCM) excretion, finding no significant differences
between patients and controls. As the only study to have
measured TUCM, this finding is of uncertain significance.

24. Jerjes, W. K. et al. Diurnal excretion of urinary cortisol, cortisone, and cortisol metabolites in chronic fatigue syndrome. J. Psychosom. Res. 60, 145153 (2006).
25. Jerjes, W. K., Taylor, N. F., Wood, P. J. & Cleare, A. J. Enhanced feedback sensitivity to prednisolone in chronic fatigue syndrome. Psychoneuroendocrinology 32, 192198 (2007).

However, if ref 25 found it also, then they omitted to mention it with regard to ref 24.
My guess is that it's an error and they meant ref 24 (somebody could have written "Jerjes et al" and it got translated to the wrong reference by the end.

In general, the findings are pretty transparent - although I would have liked more info on the reduced activity point from the Tak et al. systematic review.

I'm afraid my head is not good at remembering all the different findings so the review has a value at collating them e.g. the tables and indeed the accompanying text.

 

[Dolphin]

The problem with hypocortisolism is that it is nonspecific. It can be caused by simple distruption of our bio/sleeping rhythms, it can be caused due leading a low-stress lifestyle. Oh and gluticorticoids have important metabolic effects in the body, so dietary patterns can also change the pattern of expression.

Fries et al. as well as authors of other articles have often speculated that hypocortisolism results from overresponsiveness of the rest of the gluticorticoid pathways. The problem with this hypothesis is that when gluticorticoid receptor expression (eg mRNA) or activity is measured we find that it too is damped/low (in most studies, the remainder found normal levels). So the suppression of the HPA axis seems to be either further downstream or further upstream than hypothesised.
In any event, this means that the simplistic chronic stress->ongoing dysregulation of HPA axis assumption is almost certainly wrong. While chronic stress may indeed be a risk factor for onset, it does not seem to be a key maintaining factor in many cases.

I'm no expert on this but sometimes expression doesn't tell one output levels e.g. increased gene expression doesn't necessarily mean increased protein levels.

The paper talks about the results from suppression tests along with stimulation tests.

I would not be happy if research concentrates too much on the HPA axis as it tends to go around in circles I think and "blame the patient".

However, I think it is interesting that even emotional stressors can cause setbacks (and I don't simply mean a change in mood but more objective changes like a sort throat, swollen glands/similar). So I think there can be an abnormal response to stress in this condition. Where it is, seems to be unclear. For example, if our mitochondria were affected, maybe a shortage of fuel could lead to what seems to the HPA axis, a rather extreme stress. Or perhaps we are just about keeping viruses under control but some sort of small stress can mean they reactivate, again putting the HPA axis under stress. In both scenarios, maybe the HPA axis is actually behaving relatively normally. Hopefully further research will answer some of these questions.

 

[Dolphin]

They suggest using hair and/or nail samples to get information over longer periods

They suggest using hair and/or nail samples to get information over longer periods. Thought it was interesting e.g. hydrocortisolism should be more easy to spot.

We also suggest that methods able to measure cortisol levels over a period of days, weeks or months, for example, in hair and nails, could be useful. Cortisol and other steroid hormones passively diffuse from blood into the hair and nail matrix, where they accumulate over a period of time. Measurement of hair cortisol levels might, therefore, provide a window into the past and give an indication of the extent of cortisol exposure over the previous 6 months. This approach has so far produced encouraging results in certain conditions of long-term exposure to elevated levels of cortisol (hypercortisolemia), such as in the late stages of pregnancy,78 Cushing syndrome79 or in the long-term unemployed.80 In a pilot study, Young et al.81 measured cortisol and DHEA in finger nails as a measure of stress in medical students undertaking exams. Another study focused on hair cortisol in patients with generalized anxiety disorder.82 Although salivary cortisol in these patients did not differ from healthy controls, hair cortisol concentration was reduced by approximately half, which suggests the presence of significant hypocortisolism over a long period and in a naturalistic setting. Such an approach could be used to investigate hypocortisolism in patients with CFS, although several practical difficulties of this novel approach must first be overcome.83 In addition, cross-validation of this new technique with other established methods will be needed. One study has found that hair cortisol measures are able to retrospectively detect changes in cortisol levels after treatment for Cushing disease or hydrocortisone replacement for Addison disease, further supporting the validity of the method.84 Should such methodology stand up to scrutiny, it might aid in designing the next generation of studies into the neuroendocrinology of CFS.

 

[richvank]

Hi, all.

I continue to believe that the dysfunction of the HPA axis in ME/CFS is due to glutathione depletion in the pituitary. This causes problems with the synthesis and regulated secretion of ACTH, which in turn causes problems with adequate secretion and proper diurnal variation of cortisol. I think this is also responsible for the low aldosterone that has been found in two studies of ME/CFS.

For more details on the mechanism I have proposed, please see the video and/or slides at

http://iaomt.media.fnf.nu/2/skovde_2011_me_kroniskt_trotthetssyndrom/$%7Bweburl%7D

Best regards,

Rich

 

[Dolphin]

Review contains some empiric criteria studies

It is frustrating that the (so-called) empiric criteria (Reeves et al., 2005) studies are not highlighted as such in the review, esp. when it points out the problems that can occur from using patients with depression:

22. Nater, U. M. et al. Alterations in diurnal salivary cortisol rhythm in a population-based sample of cases with chronic fatigue syndrome. Psychosom. Med. 70, 298305 (2008).

36. Nater, U. M. et al. Attenuated morning salivary cortisol concentrations in a population-based study of persons with chronic fatigue syndrome and well controls. J. Clin. Endocrinol. Metab. 93, 703709 (2008).

37. Heim, C. et al. Childhood trauma and risk for chronic fatigue syndrome: association with neuroendocrine dysfunction. Arch. Gen. Psychiatry 66, 7280 (2009).

 

[oceanblue]

They did actually [re normal urinary metabolites]

Because it is often argued that the 24 h UFC value might
not be an accurate indicator of basal HPA function,6,25
Jerjes et al.25 also measured total urinary cortisol metabolite
(TUCM) excretion, finding no significant differences
between patients and controls. As the only study to have
measured TUCM, this finding is of uncertain significance.

In general, the findings are pretty transparent - although I would have liked more info on the reduced activity point from the Tak et al. systematic review.

Thanks for digging that out. I like their idea of using nail/hair to measure cortisol over time too. But reading that section is seems there is doubt over the validity of this technique, and indeed over all the evidence about cortisol levels and whether or not physiologically important levels of cortisol differ between healthy people and CFS patients. Which is kind of a familiar story.

 

[Dolphin]

Have there been studies looking at effect of exercise stress test on the HPA axis?

Have there been any studies looking at the effect of an exercise stress test on the HPA axis?? I thought there were (they're not mentioned in the paper, IIRC) - but perhaps it was just a physician discussing unpublished data?

 

[biophile]

Dolphin wrote: Have there been any studies looking at the effect of an exercise stress test on the HPA axis?

This one comes to mind: "Ottenweller et al. (2001) showed that in CFS patients, plasma cortisol was lower than in controls only in response to exercise and not to the stable level of fatigue." (http://www.umcutrecht.nl/NR/rdonlyr...terWolbeek_Cortisolandseverefatigue_along.pdf) Ottenweller et al 2001 = http://www.ncbi.nlm.nih.gov/pubmed/11150897 .

 

[Dolphin]

This one comes to mind: "Ottenweller et al. (2001) showed that in CFS patients, plasma cortisol was lower than in controls only in response to exercise and not to the stable level of fatigue." (http://www.umcutrecht.nl/NR/rdonlyr...terWolbeek_Cortisolandseverefatigue_along.pdf) Ottenweller et al 2001 = http://www.ncbi.nlm.nih.gov/pubmed/11150897 .

Cheers. I never read the full paper of that (don't have it).
I suppose the current review has an excuse to exclude that as the review only covers from 2003 on, although some papers from before that are then quoted (I was just thinking that in general, setting a time period like that would allow somebody to ignore findings from a certain period, while they could still selectively quote findings from that period).

That paper was discussed in his 2003 review (Cleare The Neuroendocrinology of CFS Endocrine Reviews, April 2003, 24(2):236252 243):

3. Other challenges to the HPA axis. Ottenweller et al. (31) used
a maximal exercise challenge to the HPA, clearly of relevance
given the exercise intolerance reported by CFS subjects. Seventeen
CFS subjects and 14 controls (see Table 1 for details)
were compared, with measures before, 4 min after, and 1 d
after a period of exercise to exhaustion. CFS subjects showed
a lower ACTH response immediately after this stressor, although
cortisol responses did not differ. Interestingly, the
time to exhaustion on the test was related to basal cortisol
levels: lower plasma cortisol was associated with a shorter
duration of exercise. There were no differences the following
day, suggesting that endocrine disturbance is not related to
the prolonged fatigue after exercise complained of by CFS
subjects.