Characterization of pituitary function with emphasis on GH secretion in the chronic fatigue syndrome, 2000, Moorkens et al

ScoutB

ScoutB

Senior Member (Voting Rights)
Characterization of pituitary function with emphasis on [Growth Hormone] secretion in the chronic fatigue syndrome

Moorkens, Greta; Berwaerts, Joris; Wynants, Herlindis; Abs, Roger

Abstract
OBJECTIVE
Previous studies have revealed that hormonal disturbances may accompany the chronic fatigue syndrome (CFS). Changes in the secretion of the pituitary–adrenal axis have been demonstrated, as well as abnormalities in the GH‐IGF‐I axis. However, data have not always been well characterized and were sometimes conflicting. The small number of CFS patients investigated in earlier studies may have played a role in the interpretation of the results.

SUBJECTS AND DESIGN
Hormonal testing was performed in 73 nonobese CFS patients and nonobese 21 age‐and gender‐matched healthy controls. We investigated GH, ACTH and cortisol responses to insulin‐induced hypoglycaemia. In a subgroup of patients arginine and clonidine stimulation for GH was also performed. Nocturnal secretion of GH, ACTH and cortisol were determined. Serum levels of IGF‐I, prolactin, TSH, and free thyroxine were also measured. Visceral fat mass was assessed by CT scanning.

RESULTS
GH response to insulin induced hypoglycaemia assessed by peak value (17.0 ± 13.1 μg/l vs. 22.1 ± 9.8 μg/l; P = 0.01) and by AUC (450.0 ± 361.3 μg/l vs. 672.3 ± 393.0 μg/l; P = 0.002) was significantly decreased in CFS patients vs. controls. Nocturnal GH secretion assessed by GH peak value (5.4 ± 3.7 vs. 9.0 ± 5.1 μg/l; P = 0.44) and by AUC (34.4 ± 20.2 vs. 67.4 ± 43.1; P = 0.045) was also significantly impaired in CFS patients. Arginine and clonidine administration showed no differences in GH secretion between CFS patients and controls. In the CFS group, GH peak values were significantly higher after ITT than after arginine (P = 0.017) or clonidine (P = 0.001). No differences in serum IGF‐I levels were found between CFS patients and controls. Except for a significantly lower nocturnal cortisol peak value, no differences were found in ACTH and cortisol secretion between CFS patients and controls. Significantly higher serum prolactin levels (7.4 ± 4.7 μg/l vs. 4.4 ± 1.3 μg/l; P = 0.004) and significantly higher serum TSH levels (1.6 ± 1.0 mU/l vs. 1.0 ± 0.4 mU/l; P = 0.011) were found in CFS patients. Serum free thyroxine was comparable in both groups. Visceral fat mass was significantly higher in CFS patients (86.6 ± 34.9 cm2vs. 51.5 ± 15.7 cm2; P < 0.001).

CONCLUSIONS
We observed a significant impairment of GH response during insulin‐induced hypoglycaemia and a low nocturnal GH secretion in CFS patients. These changes did, however, not lead to different concentrations in serum IGF‐I. The clinical expression of this inadequate GH secretion can thus be questioned, although the alteration in body composition may be related to this relative GH deficiency. Significantly increased prolactin and TSH levels were found when compared to controls. These findings give support to the hypothesis of a decreased dopaminergic tone in CFS. Further investigations are required in order to identify specific adaptations within the neurotransmitter system in CFS and to determine the clinical importance of the impaired GH homeostasis.

Web | DOI | PDF | Clinical Endocrinology | Paywalled | Sci-hub
(Old paper, continuing the cortisol round-up)
 
Some notes:
  • They say: the purpose of the present study was to examine the hormonal characteristics in a large group of CFS patients.
  • Hormonal testing was performed in 73 CFS patients and 21 age-and gender-matched healthy controls. (This was the total number of people involved in the study, the individual tests had smaller numbers.)
  • Fukuda criteria
  • All patients had ‘insidious onset of complaints’, no patient had post-viral onset. (This surprised me at first, but then I wonder if CFS was not discussed as a post-viral disease as much as it is today? And if that could swing the responses to a question like this.)
  • Co-morbid psychiatric disorder was an exclusion criterion. None of the patients included were taken prescribed medication in the 2 months before study entry.
  • The endocrine investigations were performed in the early follicular phase of menstrual cycle.

I have a general question about the statistics. They say:
For the comparisons between two independent groups, the Student's t-test or the Mann±Whitney test were employed where appropriate. Shapiro-Wilk test was used to assess normality. Comparison between three or more paired groups was done using the Friedman test with the Wilcoxon Signed Rank test to identify the differences (making due allowance for multiple testing). The proportion among groups were compared by the Chi Squared test or Fisher's Exact test. A P-value < 0.05 (two-sided) was considered to as statistical significant.
Click to expand...
My understanding of multiple comparison correction is worth about 1 skim of the wiki page on bonferroni corrections. In this paper they test a whole bunch of different hormones and report a few that have P values a bit below 0.05. Would we (today) expect them to do some kind of correction for this, or are these different hormones considered different datasets and therefore not in need of correction?


This could be useful information to have on hand:
Normal value for serum cortisol is 190±660 nmol/l at 0800 and < 140 nmol/l at 2400 h.
Click to expand...
(Though I wish they’d given a citation.)
 
Growth Hormone Investigations
  • Nocturnal GH secretion was assessed in 29 CFS patients and nine age- and BMI-matched controls.
  • Age matching looks ok: 39.1 ± 7.4 (CFS) vs 32.4 ± 10.7 (controls)
  • However gender had a 9:20 male-to-female ratio for patients, versus a 4:5 ratio for controls
  • Actually, apparently women typically make much more GH than men? So if anything this would make their low-GH-in-CFS-patients results even more surprising?
  • Nocturnal GH peak (P < 0.044) and nocturnal GH AUC (P < 0.045) were significantly impaired in CFS patients compared to controls (Fig. 1).
  • Hypoglycaemia induced GH peak (P < 0.01) and GH AUC (P < 0.002) were significantly lower in CFS patients than in controls (Fig. 2).
  • GH stimulation by arginine or clonidine did not reveal a significant difference in GH peak and GH AUC between CFS patients and controls.

Some caveats about the GH results:
  • In the places where they checked if CFS patients were in normal reference ranges (i.e. the insulin tolerance test), the majority of CFS patients were still in the normal range.
  • They also checked IGF-1, which is produced in response to GH and is supposed to be a more stable indicator than GH itself of GH production. There was no differences in serum IGF-I level between CFS patients and controls.

Still, the growth hormone graphs are a bit interesting:
1781051531579.webp

1781051545932.webp

I know GH was an area of interest at the time. I wonder if anything ever came of it?
 
ACTH and Cortisol Investigations
  • Same patients and controls were used for these tests, and so the same issue of women being underrepresented amid the control population applies.
  • No statistical differences in basal plasma ACTH and basal serum cortisol value were found between CFS patients and controls.
  • Nocturnal ACTH peak value in CFS patients was not different from controls.
  • Nocturnal cortisol peak value was significantly lower in CFS patients than in controls (P < 0.021). They say this, but the table says the opposite?
  • ACTH AUC and cortisol AUC during nocturnal sampling was identical in the CFS group and the control group.
  • Both ACTH peak and cortisol peak during [insulin tolerance test] were similar between CFS patients and controls.
  • ACTH AUC and cortisol AUC during [insulin tolerance test] were comparable for both groups.

Now enjoy this table which reports the opposite of what they say throughout the paper:
1781051611565.webp
Maybe I'm just completely foggy, but that nocturnal cortisol peak looks higher in the CFS group. Do we assume they just flipped the columns here? I’m also confused about how the ‘nocturnal peak’ is so high in the first place — I’m used to seeing nocturnal cortisol < 200 nmol/L. Maybe there’s something going on with how they measured it, but that wouldn’t explain how the nocturnal measure is higher than their morning cortisol measure.

In other studies, cortisol is doing this at night:
1781051690833.webp
(source for the above graph)

So the nocturnal ‘peak’ would just be at one of the two ends, I think? They got the nocturnal cortisol measurements by taking 5 samples from 10pm to 6am (i.e. every 2 hours). As you can see, cortisol is generally super low in that period, with its highest point probably being at 6am in most people. So I am a bit mystified.
 
Anyway, the last group of results is:

Prolactin, TSH and Thyroxine Investigations
  • For this measure the male-to-female ratio was 18:55 for patients and only 9:12 for controls. So, as we often see, the female-predominance of the CFS sample was not that well matched.
  • Serum prolactin level was significantly higher in the CFS group compared to the control group: 7.4 ± 4.7 mg/l vs. 4.4 ± 1.3 mg/l (P < 0.004). (IIRC other studies of this era did not find elevated prolactin so that’s interesting, but maybe this is due to the skewed gender ratio.)
  • A modest but significant increase in serum TSH level was also present in CFS patients: 1.6 ± 1.0 vs. 1.0 ± 0.4 mU/l (P < 0.011).
  • Serum free thyroxine level was not different between CFS patients and controls.

And finally, some concluding remarks that might be relevant to our current interest in dopamine:
The main findings in our study fit into the theory of a reduced dopaminergic tone in CFS (Bruno et al., 1998 - a paper on post polio fatigue). The impaired GH secretion can be explained accordingly since dopamine acts through reducing the somatostatinergic tone (Vance et al., 1987). Both prolactin and TSH secretion are directly inhibited by dopamine, but are also controlled by somatostatin. However, a decreased somatostatin tone in CFS can not account for the impaired GH secretion. Our data provide a rationale for further study of the role played by neurotransmitters in the symptomatology of CFS. Although interaction between the different systems may confound definite conclusions, impairment of the dopaminergic neurotransmission may underlie the endocrine disturbances of CFS.
Click to expand...
 
ScoutB said:
So the nocturnal ‘peak’ would just be at one of the two ends, I think? They got the nocturnal cortisol measurements by taking 5 samples from 10pm to 6am (i.e. every 2 hours). As you can see, cortisol is generally super low in that period, with its highest point probably being at 6am in most people. So I am a bit mystified.
Click to expand...

Daily studies of cortisol in ME/CFS patients need to be normalised to their regular daily cycle, not merely time of day. Changing their sleep/wake times to match the sampling for a study will bias the results substantially. If ME/CFS patients normally sleep longer, or sleep to a later time, this will be a problem. Keep this in mind when reading the methodologies of studies like this.
 
You must log in or register to reply here.
Back
Top Bottom