1 Department of Obstetrics and Gynaecology and 2 Department of Endocrinology, UCSC, Rome and 3 Oasi Institute for Research, Troina, Italy
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: body weight/growth hormone/insulin/meal/polycystic ovarian syndrome
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Recent studies have well documented an impairment of GH response to several stimuli in women with polycystic ovarian syndrome (PCOS) (Lee et al., 1993; Piaditis et al., 1995
). This syndrome is characterized by anovulation, hyperandrogenism and in a considerable percentage of subjects by obesity and hyperinsulinism. The blunted GH response to GHRH in PCOS women could be related only partially to obesity since lean hyperinsulinaemic subjects showed a similar reduction of GH secretion after GHRH administration (Lanzone et al., 1995
), thus indicating that several other factors may interfere with GH secretion.
The aim of the present study was to investigate the plasma GH response to direct stimulation with GHRH before and after a standard meal in PCOS subjects in order to study the influence of body weight and insulin on GH secretion.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
PCOS was diagnosed by clinical findings (presence of amenorrhoea or oligoamenorrhoea and hirsutism), plasma androgen concentrations at the upper limit or above the normal range (androstenedione 1.985.58 nmol/l, testosterone 0.582.01 nmol/l), and bilateral normal or enlarged ovaries with the presence of at least 10 microcysts (28 mm diameter) (from the inner margin to the outer margin in longitudinal cross-sections) associated with an increase in ovarian stroma (Polson et al., 1988) at the time of ultrasonography. A normal LH to FSH ratio was not considered an exclusion criterion. Data were compared with those obtained from 14 health normovulatory patients, matched for age and body mass index (BMI), seven simply obese patients (i.e. not PCOS) were recruited to our endocrinology department and seven lean controls to our sterility centre. Obesity was defined as a BMI > 27 kg/m2 (normal range 1925 kg/m2). All subjects underwent the same clinical protocol. Informed consent was obtained from each patient. The study was approved by our ethical committee. In the follicular phase, women were hospitalized the day before the beginning of the basal study: this last one was performed after a 3 day standard carbohydrate diet (300 g) and fasting overnight. At 0700 h blood samples were taken for evaluation of LH, FSH, 17ß-oestradiol, 17-hydroxyprogesterone (17-OHP), sex hormone-binding globulin (SHBG), dehydroepiandrosterone sulphate (DHEA-S), androstenedione, testosterone, growth hormone (GH), insulin, glucose, and free fatty acid (FFA) plasma concentrations.
Then all women had an oral glucose tolerance test (OGTT). Blood samples were collected every 30 min for 4 h after ingestion of 75 g of glucose; insulin and glucose were assayed in each sample. On different days all subjects underwent GHRH (50 µg GHRH; Geref, Serono, Italy) tests both before and after lunch in randomized order.
Pre-prandial tests
After an overnight fast the women had their usual breakfast at 0800 h and they ate nothing thereafter until the last test. At 1245 h, while the women were resting in a quiet room, an indwelling venous cannula was inserted in one arm, through which normal saline (0.9%) was given slowly. At time 0 (1300 h) a 50 µg bolus dose of GHRH was injected i.v. Blood samples were collected 15 min and just before and 30, 60, 90 and 120 min after GHRH administration.
Post-prandial tests
Fasting throughout the night the women had their usual breakfast at 0800 h and then ate nothing until the last test. At 1200 h the women ate a 800 kcal meal composed of 55% carbohydrate, 30.6% lipid, and 13.6% protein, as previously described (De Marinis et al., 1991). A saline infusion was started at 1245 h and GHRH (50 µg) was administered 15 min later (time 1300 h). Blood samples were collected at 15, 0, 30, 60, 90 and 120 min.
Assays
All blood samples were promptly centrifuged (1500 r.p.m.) and stored at 20°C until assayed. LH, FSH, oestradiol, testosterone, 17-OHP, androstenedione, DHEA-S and SHBG were determined in the basal condition. These hormones were measured in duplicate by radioimmunoassay methods using commercial kits (Radim, Pomezia, Italy). The immunoradiometric assay (IRMA) on solid phase (coated tube), based on monoclonal double-antibody technique, was used for LH, FSH and SHBG. Steroids were assayed by a radioimmunoassay direct method in human serum or plasma. Insulin was assayed using radioimmunoassay methods and glucose concentrations were determined by the glucose oxidase technique. Intra-assay and inter-assay coefficients of variation were as follows: LH, 5.6% and 9.1%; FSH, 6.9% and 8.4%; oestradiol, 2.3% and 3.5%; testosterone and androstenedione, 6.1% and 9.3%; insulin, 5.1% and 6.2%; SHBG, 6.9% and 8.5%. GH was determined by the IRMA method using commercial kits from Radim. Intra-assay and inter-assay coefficients of variation were respectively 2.5% and 5.8%. The lowest amount of GH detected was 0.04 µg/l. Free fatty acids were determined by an acyl-coenzyme-A oxidase based colorimetric method (Okabe et al., 1980).
Data analysis
An abnormal glycaemic response to the OGTT was defined according to the criteria of National Diabetes Data Group (National Diabetes Group, 1994). No patient showed impaired glucose tolerance. All results were presented as mean ± SD. Insulin and GH plasma concentrations were also expressed as area under the curve (AUC) after OGTT or GHRH test respectively, calculated by the trapezoidal rule and expressed as µIU/mlx240 min for insulin (AUC-I) and µg/lx120 min for GH. Incremental area under curve for GH (AUCi-GH) was calculated by trapezoidal rule after subtracting basal hormone concentration. The difference (AUCd-GH) between the GHRH elicited incremental pre-prandial and post-prandial area (AUCd-GH = AUCi post-prandial AUCi-GH pre-prandial) was considered as indicator of GH secretion capacity in relation to a meal. The ratio of testosteronex100/SHBG was used to calculate the free androgen index (FAI). The distribution of the data was tested by KolmogorovSmirnov test to verify whether the samples came from a specified distribution and it was found that the data were not normally distributed. The significance of differences between the same tests performed before and after a meal was assessed by the non-parametric Wilcoxon rank-sum test. The comparison between different study groups was performed by the non-parametrical MannWhitney U-test. Linear regression analysis was used to analyse possible correlation between endocrine findings. The level of statistical significance was set at P < 0.05.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Fasting insulin concentrations were significantly higher in the obese PCOS group and in obese controls with respect to those found in lean controls (P < 0.01). Moreover, AUC-I concentrations in obese PCOS patients were significantly higher than in the other groups (P < 0.001). No differences were found among groups for fasting glycaemia and AUC-glucose values.
Obese patients (both PCOS and controls) had lower baseline GH values versus lean PCOS and control subjects (P < 0.01).
Figure 1 shows plasma GH concentrations after GHRH administration before and after a meal both in lean and obese control and PCOS subjects.
|
In lean PCOS subjects GH peak reached 12.3 ± 9.2 µg/l values significantly lower than lean controls (P < 0.05). However, GH concentrations remained significantly higher than those observed in PCOS obese patients at 60, 90 and 120 min (P < 0.05).
Post-prandial test
At 1300 h, 1 h after a meal, lean control subjects showed GHRH-induced GH peak concentrations significantly lower (10.7 ± 3.84 µg/l; P < 0.05) than that observed in pre-prandial tests. In obese patients plasma GH response to GHRH injection was significantly increased at 30 min after GHRH injection and the peak GH response was significantly higher in comparison with the one registered without meal administration (14.84 ± 4.95 µg/l; P < 0.05).
The plasma GH response to GHRH in lean PCOS subjects increased in respect to pre-prandial response but not significantly, otherwise the GH post-prandial peak in these subjects was significantly higher than the GH post-prandial peak in normal subjects (23.6 ± 11 µg/l, P < 0.05). In the obese PCOS subjects the GH plasma peak was significantly lower than in lean PCOS women and GH plasma concentrations at 30, 60, 90 and 120 min were not significantly different from those before lunch.
To explore the relative influence of BMI and insulin on GHRH-induced GH secretion the correlation of these parameters with the difference in GH incremental area after and before a meal (AUCd-GH) was examined. A significantly negative linear correlation between AUCd-GH values and BMI (P < 0.001; r = 0.87) as well as AUC-I values (P < 0.001; r = 0.813) was found in the whole PCOS group. However in controls a correlation was found between BMI and AUCd-GH (P < 0.05; r = 0.55) but no correlation between AUC-I and AUCd-GH.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Several authors indicated that obesity is associated with an abnormal elevation in both fasting and post-prandial FFA concentration (Dieguez and Casanueva, 1995; Cordido et al., 1998
). Previous studies demonstrated that a glucose infusion or a rise in plasma FFA concentrations impaired GH responsiveness to GHRH (Sharp et al., 1984
; Imaki et al., 1985
). Such negative feedback of FFA on GH release may occur both at the pituitary and hypothalamic level (Pontiroli et al., 1986
). In the current study, a primary role of FFA seems to be excluded. Obese controls as well as lean PCOS subjects showed an augmented GH secretion after a meal despite higher basal plasma FFA.
Certainly BMI is correlated with GH-stimulated secretion after a meal in PCOS and in the control population. Therefore body weight per se is an important but not the only factor responsible for blunting GHRH-induced release.
In this concern the relation between AUC-I and AUCd-GH is present only in the PCOS population, where the PCOS obese patients showed the highest AUC-I values in respect to lean PCOS patients and also to obese control patients.
The role of insulin is less established and needs to be studied further. Several pieces of experimental data suggest a possible direct effect of insulin on GH axis; for instance, insulin receptors are present in rat hypothalamus and insulin binding sites have been demonstrated in normal rat pituitary adenoma cells as well as in human pituitary adenoma cells (Ceda et al., 1985; Schwartz et al., 1991
). In-vitro studies demonstrate that insulin is able to inhibit the peripheral action of GH (Ji et al., 1999
). In normal subjects integrated 24 h GH concentration is elevated during fasting (Ho et al., 1988
); moreover, during euglycaemic insulin clamp, a reduction of GH response to GHRH has been observed, whereas during hypoglycaemic insulin clamp the GH response to hypoglycaemia is inversely related to the degree of hyperinsulinaemia (Diamond et al., 1991
; Press et al., 1992
). Recently it has been demonstrated that insulin, independently of FFA, is able to exert an inhibitory effect on GH release (Lanzi et al., 1997
).
Thus insulin negative feedback may have physiological relevance, because the insulin concentrations, able to reduce GH response to GHRH, are commonly observed during the post-prandial period. Therefore, elevated insulin concentrations can inhibit pituitary GH release and insulin resistance may be involved in the alteration of GH secretion, particularly in obese PCOS patients.
In the attempt to explain the different behaviour between controls and hyperandrogenized PCOS subjects the influence of hormonal milieu on GH secretion could be considered. Some studies have demonstrated that steroid environment may affect GH dynamics and higher androgen concentrations could modify the sensitivity to regulatory stimuli (De Marinis et al., 1997; Kaltsas et al., 1998
).
Moreover since previous work indicated that opioids are involved in the post-prandial GH increase in normal subjects (De Marinis et al., 1989) and it was suggested that derangement of opioid tone may have some relation with the blunted response of GH to GHRH in PCOS (Villa et al., 1997
), it may also be that the difference in opioid tone among controls and PCOS subjects (Lanzone et al., 1995
) may partially explain the results presented here.
In conclusion the data of the present work indicate for the first time a different behaviour of GHRH-induced GH secretion related to food ingestion in PCOS in respect to control subjects. Obese PCOS patients did not exhibit any change of GH response or the paradoxical response observed in matched controls. Several factors apart from BMI may be involved in these pathophysiological events. Further studies are needed in order to elucidate this exciting new field of research.
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Cordido, F., Fernandez, T., Martinez, T. et al (1998) Effect of acute pharmacological reduction of plasma free fatty acids on growth hormone (GH) releasing hormone-induced GH secretion in obese adults with and without hypopituitarism. J. Clin. Endocrinol. Metab., 83, 43504354.
De Marinis, L., Folli, G., D'Amico, C. et al. (1988) Differential effect of feeding on the ultradian variation of the growth hormone (GH) response to GH-releasing hormone in normal subjects and patients with obesity and anorexia nervosa. J. Clin. Endocrinol. Metab., 66, 598604.[Abstract]
De Marinis, L., Mancini, A., Folli, G. et al. (1989) Nalaxone inhibition of postprandial growth hormone releasing hormone-induced growth hormone release in obesity. Neuroendocrinology, 50, 529532.[ISI][Medline]
De Marinis, L., Mancini, A., Zuppi, P. et al. (1991) GH response to GHRH before and after meals at different hours of the day in obese patients. Psychoneuroendocrinology, 16, 361365.[ISI][Medline]
De Marinis, L., Mancini, A., Valle, D. et al. (1997) Physiological role of the opioidcholinergic interaction in growth hormone neuroregulation: effect of sex and food intake. Metabolism, 46, 740744.[ISI][Medline]
Diamond, M.P., Hallarman, L., Starick-Zich, K. et al. (1991) Suppression of counter-regulatory hormone response to hypoglycaemia by insulin per se. J. Clin. Endocrinol. Metab., 72, 13881390.[Abstract]
Dieguez, C. and Casanueva, F.F. (1995) Influence of metabolic substrates and obesity on growth hormone secretion. Trends Endocrinol. Metab., 6, 5559.[ISI]
Gianotti, L., Broglio, F., Ramunni, J. et al. (1998) The activity of GH/IGF-1 axis in anorexia nervosa and in obesity: a comparison with normal subjects and patients with hypopituitarism or critical illness. Eat Weight Disord., 3, 6470.[Medline]
Ho, K.Y., Veldhuis, J.D., Johnson, M.I. et al. (1988) Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. J. Clin. Invest., 81, 968975.[ISI][Medline]
Imaki, T., Shibasaki, T., Shizume, K. et al. (1985) The effects of free fatty acids on growth hormone (GH)-releasing hormone mediated GH secretion in man. J. Clin. Endocrinol. Metab., 60, 290295.[Abstract]
Ji, S., Guan, R., Frank, S.K. et al. (1999) Insulin inhibits growth hormone signaling via the growth hormone receptor /JAK2/STAT5B pathway. J. Biol. Chem., 274, 1343413442.
Kaltsas, T., Pontikides, N., Krassas, G.E. et al. (1998) Effect of gonadotrophin-releasing hormone agonist treatment on excretion. J. Clin. Endocrinol. Metab., 80, 35013506.[Abstract]
Lanzi, M., Manzoni, F., Andreotti, C. et al. (1997) Evidence for an inhibitory effect of physiological concentrations of insulin on the growth hormone (GH) response to GH-releasing hormone in healthy subjects. J. Clin. Endocrinol. Metab., 82, 22392243.
Lanzone, A., Villa, P., Fulghesu, A.M. et al. (1995) The growth hormone response to growth hormone-releasing hormone is blunted in polycystic ovary syndrome: relationship with obesity and hyperinsulinaemia. Hum. Reprod., 10, 16531657.[Abstract]
Lee, E.J., Park, K.H., Lee, B.S. et al. (1993) Growth hormone response to L-dopa and pyridostigmine in women with polycystic ovarian syndrome. Fertil. Steril., 60, 5357.[ISI][Medline]
National Diabetes Data Group (1994) Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes, 28, 10391057.[ISI][Medline]
New, M.I., Lorenzen, F., Lerner, A.J. et al. (1983) Genotyping steroid 21-hydroxylase deficiency: hormonal reference data. J. Clin. Endocrinol. Metab., 57, 320326.[Abstract]
Okabe, H., Uji, Y., Nagashima, K. et al. (1980) Enzymic determination of free fatty acids in serum. Clin. Chem., 26, 15401543.
Piaditis, G.P., Kounadi, T.G., Rangon, D.B. et al. (1995) Dysfunction of the growth hormone/insulin-like growth factor-I axis in women with polycystic ovary syndrome. Clin. Endocrinol., 42, 635640.[ISI][Medline]
Polson, D.W., Wadsworth, J., Adams, J. et al. (1988) Polycystic ovaries a common finding in normal women. Lancet, 16, 870872.
Pontiroli, A.E., Lanzi, R., Monti, L.D. et al. (1986) Growth hormone feed back on GH response to GH-releasing hormone. Role of free fatty acids and somatostatin. J. Clin. Endocrinol. Metab., 22, 492495.
Press, M., Caprio, S., Tamborlane, W.M. et al. (1992) Pituitary response to growth hormone releasing hormone in IDDM. Abnormal responses to insulin and hyperglycemia. Diabetes, 41, 1721.[Abstract]
Schwartz, M.W., Figlewicz, D.P., Baskin, D.G. et al. (1991) Insulin in the brain: a hormonal regulator of energy balance. Endocr. Rev., 13, 387414.[ISI][Medline]
Sharp, P.S., Foley, K., Chahal, P. et al. (1984) The effect of plasma glucose on the growth hormone responses to human pancreatic growth hormone releasing factor in normal subjects. Clin. Endocrinol., 20, 497501.[ISI][Medline]
Tanaka, K., Inowe, S., Numata, K. et al. (1990) Very low caloric diet induced weight reduction reverses impaired growth hormone secretion response to arginine, and L-dopa in obesity. Metabolism, 39, 892896.[ISI][Medline]
Villa, P., Fulghesu, A.M., De Marinis, L. et al. (1997) Impact of long-term naltrexone treatment on growth hormone and insulin secretion in hyperandrogenic and normal obese patients. Metabolism, 46, 538543.[ISI][Medline]
Williams, T., Berelowitz, M., Joffe, S. et al. (1984) Impaired growth hormone responses to growth hormone-releasing factor in obesity. N. Engl. J. Med., 311, 14031407.[Abstract]
Submitted on May 30, 2000; accepted on November 30, 2000.