- Email: [email protected]
Growth hormone secretion and insulin-like growth factor-1 are related to hyperandrogenism in nonobese patients with polycystic ovary syndrome
Growth hormone secretion and insulin-like growth factor-1 are related to hyperandrogenism in nonobese patients with polycystic ovary syndrome Women of normal weight with polycystic ovary syndrome (PCOS) and hyperinsulinemia presented high growth hormone (GH) levels in response to the L-dopa test, suggesting that the action of GH and insulin-like growth factor-1 (IGF-1) might be responsible for the elevation in LH and the consequent hyperandrogenic anovulation observed in normal weight women with PCOS. Insulin resistance and obesity are related to a reduction in GH secretion in obese women with PCOS. (Fertil Steril威 2005;83:1852–5. ©2005 by American Society for Reproductive Medicine.)
Polycystic ovary syndrome (PCOS) occurs in 6%–10% of women of reproductive age (1). Endocrine disorders such as insulin resistance and compensatory hyperinsulinemia are currently two of the main factors involved in the etiopathogenesis of PCOS (2). Obesity is associated with PCOS in about 50% of patients (3, 4). Obesity itself increases insulin resistance or acts as an aggravating factor of this disease. However, normal weight women with PCOS show a lower incidence of insulin resistance and compensatory hyperinsulinemia (5). Other factors such as the somatotrophic axis, growth hormone (GH), and insulin-like growth factors (IGF) play a role in the etiopathogenesis of this syndrome (6, 7). A lower GH response has been observed in women with PCOS compared to women with ovulatory cycles (8), whereas other investigators have not detected a difference in women with PCOS and have attributed the lower GH response to obesity (4). Therefore, both obese and normal weight women with PCOS present disturbances in GH secretion, but these alterations do not follow the same pattern. The exact mechanisms underlying GH secretion disorders in women with PCOS and the reason for the difference in the response of GH release to stimulation tests between obese and normal weight women with PCOS are still not completely understood. The objective of the present study was to determine GH and IGF-1 secretion after a levodopa (L-dopa) test and serum insulin and glucose levels after an oral glucose tolerance test (GTT) with 75 g glucose in obese and nonobese women with PCOS compared to women with ovulatory cycles. A prospective study was conducted on 40 women divided into four groups: group I, 11 obese women with Received June 11, 2004; revised and accepted October 28, 2004. Reprints requests: Rosana Maria dos Reis, M.D., Department of Gynecology and Obstetrics, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Av. Bandeirantes, 3900, 14049-900 Ribeirão Preto, Brazil (FAX: 16-6330946; E-mail: [email protected]).
1852
PCOS; group II, 9 normal weight women with PCOS; group III, 10 obese women with ovulatory cycles; and group IV, 10 normal weight women with ovulatory cycles. The diagnosis of PCOS was based on presence of oligomenorrhea/amenorrhea and clinical or biochemical signs of hyperandrogenism (9). The exclusion criteria were Cushing’s syndrome, 21-hydroxylase deficiency, thyroid dysfunction, hyperprolactinemia, and androgen-secreting tumors. Women with regular menstrual cycles with an interval of 28 –32 days and normal serum progesterone (ⱖ3 ng/mL) on the 22nd day of the cycle were considered to be ovulatory. Obesity was defined as a body mass index (BMI) ⱖ27 kg/m2 . None of the patients used medications during the 90 days preceding the study. The study was approved by the Ethics Committee of the institution and the patients gave written informed consent to participate. After an overnight fast, all patients were submitted to an GTT with 75 g of dextrose and fasting blood samples were collected every 30 minutes for 2 hours for the determination of insulin and glucose. The L-dopa test was performed 2–3 days after the GTT following the same criteria. A dose of 500 mg of L-dopa was administered orally for the measurement of GH and IGF-1. Any increase in GH levels over baseline (fasting) was considered to be a positive response. The area under the curve (AUC) for glucose (in milligrams per deciliter times 240 minutes), insulin (in micro IU per milliliter times 240 minutes), and GH (in nanograms per deciliter times 240 minutes) were calculated using the trapezoidal method. Plasma glucose was measured by the enzymatic glucose oxidase method. Serum insulin was determined by RIA using a solid-phase kit (DPC, Los Angeles, CA). Plasma GH was measured by a doubleantibody RIA and serum IGF-1 was determined by a radioimmunometric method using a double antibody (DSL 2800 kit without extraction; DSL, Webster, TX). Serum LH, FSH, and PRL were measured by chemiluminescence using DPC kits. Serum testosterone (T), androstenedione (A), 17␣-hydroxyprogesterone (17-OHP), and dehydroepiandrosterone sulfate were measured by RIA. insulin-like
Fertility and Sterility姞 Vol. 83, No. 6, June 2005 Copyright ©2005 American Society for Reproductive Medicine, Published by Elsevier Inc.
0015-0282/05/$30.00 doi:10.1016/j.fertnstert.2004.10.057
growth factor binding protein-3 (IGFBP-3) was determined by immunoradiometric assay (IRMA) using the DSL-6600 kit. Data were analyzed statistically by the nonparametric Kruskal-Wallis test and Spearman’s correlation coefficient. Multiple comparisons were made by the Dunn test. Normality of the data was evaluated by the nonparametric Kolmogorov-Smirnov test. The level of significance was set at 5%. The mean serum T values were higher in group I than in the group IV (group I, 61.25 ⫾ 17.80 ng/dL vs. group IV, 40.39 ⫾ 10.20 ng/dL, P⫽.025). The mean plasma AUC values for glucose were lower in group IV (9,467 ⫾ 1,673 mg/dL) than those observed for the groups I and III (group I, 12,544 ⫾ 2,515 mg/dL and group III, 13,528 ⫾ 2,667 mg/dL, Pⱕ.01). Group I (21.55 ⫾ 10.46 IU/mL) had higher fasting serum insulin levels than group IV (9.76 ⫾ 5.74 IU/mL), with Pⱕ.05. The mean AUC for insulin was higher in the groups I and II (group I, 17,479 ⫾ 12,403 IU/mL ⫻ 240 minutes and group II, 17,150 ⫾ 12,312 IU/mL ⫻ 240 minutes) than in the group IV (5,057 ⫾ 2,375 IU/mL ⫻ 240 minutes), P⫽.0046. Mean fasting plasma IGF-1 values were higher in group II (264.50 ⫾ 69.21 ng/mL) than in the others groups (group I, 154.58 ⫾ 60.99 ng/mL; group III, 165.17 ⫾ 100.65 ng/mL; and group IV, 141.30 ⫾ 49.31 ng/mL, Pⱕ.01). Regarding GH, all groups responded to the L-dopa test; however, a tendency toward a greater response was observed among normal weight women with and without PCOS. Obese women with PCOS had lower mean AUC–GH values (group I, 140.00 ⫾ 77.00 ng/mL ⫻ 240 minutes) than group IV (320.60 ⫾ 138.38 ng/mL ⫻ 240 minutes), P⫽.016. Analysis of the data for the four groups as a whole regarding the AUC–insulin, BMI, and AUC–GH revealed a negative correlation between AUC–GH and AUC–insulin (r ⫽ ⫺0.44, ⫺0.6633 to ⫺0.1342, P⫽.005) and between AUC–GH and BMI (r ⫽ ⫺0.56, ⫺0.7445 to ⫺0.2885, P⫽.0002). A positive correlation was observed between AUC–insulin and BMI (r ⫽ 0.32, 0.002958 to 0.5827, P⫽.042), and a negative correlation between fasting GH and T (r ⫽ ⫺0.4822, ⫺06951 to ⫺0.1917, P⫽.0016) (Table 1). An abnormal GH response to stimulation tests such as and GH-releasing hormone (GHRH) has been observed in women with PCOS (8). Thus, our findings suggest that the regulation of GH secretion is influenced by androgens, but other factors such as obesity/insulin resistance can interfere with the regulation of the somatotrophic axis in PCOS. Hyperandrogenism can contribute to a reduction in GH secretion after the L-dopa test because T directly stimulates somatostatin secretion (10). Wu et al. (11) detected in nonobese women with PCOS a decline in androgen levels (testosterone/androstenedione) and an increase in GH and the GH/IGF-1 ratio after the L-dopa test 6 months after ovarian wedge resection. L-dopa
Fertility and Sterility姞
Women with PCOS show an increase in the bioactivity of IGF-1 resulting from an increase in IGF-1 levels or from a reduction in IGFBP-1 concentrations (12). In the present study we observed higher IGF-1 values in normal weight women with PCOS and no variations in IGFBP-3 values compared to the other groups. The IGF-1 interferes with the regulation of GH secretion in women with PCOS, probably through the inhibition of hypothalamic GH secretion by somatostatin. In our study, obese women showed a lower response to the L-dopa test, as well as a negative correlation between obesity and the AUC for GH, in agreement with data reported by other investigators (4, 13). It is important to emphasize that obese women show an increase in hypothalamic somatostatin secretion (14), with the imbalance between opioid secretion and somatostatin being responsible for the low GH values observed in these patients. Some evidence suggests that normal weight women with PCOS also present an imbalance in opioid tone interfering with GH secretion, with this derangement differing from that observed in obese women with PCOS (13). Confirming previous data reported by our group (3), in the present study we observed high AUC–insulin values in women with PCOS irrespective of obesity and also in obese women with ovulatory cycles. Various studies have analyzed the role of insulin in GH secretion. In vivo studies have demonstrated that insulin blocks the peripheral action of GH (14). In healthy individuals, 24-hour GH secretion is higher after fasting (15). Euglycemic clamp studies have shown that the GH response to hypoglycemia is inversely proportional to the degree of hyperinsulinemia (16). No difference in AUC–GH values was observed between normal weight women with PCOS and the normal weight control group, confirming that the GH response to L-dopa in the fasting state was normal, although the women with PCOS were hyperinsulinemic. On the other hand, obese women with PCOS and hyperinsulinemia had lower AUC–GH values. This finding suggests a role of insulin feedback in the somatotrophic axis and indicates that in obese women with PCOS the excess weight and high circulating insulin concentrations interfere with the GH response to stimulation tests (13). Based on the present findings, we suggest that the presence of PCOS in normal weight women associated with high GH and IGF-1 values indicates the involvement of two different mechanisms in the physiopathology of this syndrome according to BMI. Thus, the combined action of GH and IGF-1 might be responsible for the elevation of LH and the consequent hyperandrogenism in nonobese women with PCOS, whereas in obese patients with PCOS obesity and insulin resistance promote a reduction in GH secretion. 1853
TABLE 1 Clinical characteristics and hormone levels of women with polycystic ovary syndrome (obese/normal weight) and women with ovulatory cycles (obese/normal weight). Women with PCOS
Age (y) BMI (kg/m2) LH/FSH Testosterone (ng/dL) Androstenedione (ng/dL) IGF-1 (ng/mL) basal IGF-1 (ng/mL) 30 minutes Fasting glucose (mg/dL) Fasting insulin (IU/mL) Fasting GH (ng/mL) AUC for glucose (mg/dL ⫻ 240 minutes) AUC for insulin (IU/mL ⫻ 240 minutes) AUC for GH (ng/mL ⫻ 240 minutes) GH response to L-dopa IGFBP-3 (ng/mL)
Women with ovulatory cycles
Obese (group I) (n ⴝ 11)
Normal weight (group II) (n ⴝ 9)
Obese (group III) (n ⴝ 10)
Normal weight (group IV) (n ⴝ 10)
23.73 ⫾ 2.90a 35.70 ⫾ 4.94c 2.53 ⫾ 0.98e 61.25 ⫾ 17.18g 180.01 ⫾ 92.38 154.58 ⫾ 60.99 168.73 ⫾ 72.43 79.45 ⫾ 12.10 21.55 ⫾ 10.46i 0.37 ⫾ 0.47 12,544 ⫾ 2,515
21.67 ⫾ 3.84b 22.31 ⫾ 2.73 2.54 ⫾ 0.81f 56.13 ⫾ 18.55 217.53 ⫾ 98.03 264.50 ⫾ 69.21h 264.53 ⫾ 80.28 80.11 ⫾ 7.85 13.16 ⫾ 6.72 1.24 ⫾ 2.51 11,988 ⫾ 2,434
32.00 ⫾ 7.66 36.09 ⫾ 6.22d 0.58 ⫾ 0.44 43.07 ⫾ 20.97 175.10 ⫾ 101.92 165.17 ⫾ 100.65 176.20 ⫾ 112.87 87.30 ⫾ 4.00 19.97 ⫾ 9.89 0.39 ⫾ 0.27 13,528 ⫾ 2,667
31.60 ⫾ 3.34 22.26 ⫾ 2.51 1.03 ⫾ 0.83 40.39 ⫾ 10.20 145.40 ⫾ 58.65 141.30 ⫾ 49.31 197.50 ⫾ 83.36 82.20 ⫾ 9.05 9.76 ⫾ 5.74 1.42 ⫾ 1.85 9,467 ⫾ 1,673j
17,479 ⫾ 12,403
17,151 ⫾ 12,312
11,538 ⫾ 6,914
5,057 ⫾ 2,375k
140.00 ⫾ 77.00l
334.77 ⫾ 326.68
142.34 ⫾ 106.21
320.60 ⫾ 138.38
4.83 ⫾ 5.72 3,081 ⫾ 502,40
2.33 ⫾ 1.91 2,923 ⫾ 479,20
4.53 ⫾ 3.06 2,683 ⫾ 696,50
2.07 ⫾ 1.42 2,749 ⫾ 599,60
Note: Data are reported as mean ⫾ SD. PCOS ⫽ polycystic ovary syndrome; BMI ⫽ body mass index; IGF-1 ⫽ insulin-like growth factor-1; GH ⫽ growth hormone; AUC ⫽ area under the curve. a P⬍.01, group I vs. groups III and IV. b P⬍.01, group II vs. groups III and IV. c P⬍.01, group I vs. groups II and IV. d P⬍.01, group III vs. groups II and IV. e P⬍.01, group I vs. groups III and IV. f P⬍.01, group II vs. groups III and IV. g P⬍.05, group I vs. group IV. h P⬍.01, group II vs. groups I, III, and IV. i P⬍.05, group I vs. group IV. j P⬍.01, group IV vs. groups I and III. k P⬍.01, group IV vs. groups I and group II. l P⬍.05, group I vs. group IV. Premoli. Growth hormone and PCOS. Fertil Steril 2005.
Acknowledgments: The authors thank Elettra Greene for translating the Portuguese original. Research was supported by grants from Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP).
Ana Cristina Premoli, M.D. Laura Ferreira Santana, M.D. Rui Alberto Ferriani, M.D. Marcos Dias de Moura, M.D. Marcos Felipe Silva De Sá, M.D. Rosana Maria dos Reis, M.D. Sector of Human Reproduction, Department of 1854
Premoli et al.
Correspondence
Gynecology and Obstetrics, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, Brazil
REFERENCES 1. Asuncion M, Calvo RM, San Millan JL, Sancho J, Avila S, EscobarMorreale HF. A prospective study of the prevalence of the polycystic ovary syndrome in unselected Caucasian women from Spain. J Clin Endocrinol Metab 2000;85:2434 – 8. 2. Dunaif A, Segal KR, Futterweit W, Dobrjansky A. Profound periph-
Vol. 83, No. 6, June 2005
3.
4.
5.
6.
7.
8. 9.
10.
eral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 1989;38:1165–74. Reis RM, Foss MC, de Moura MD, Ferriani RA, Silva de Sa MF. Insulin secretion in obese and non-obese women with polycystic ovary syndrome and its relationship with hyperandrogenism. Gynecol Endocrinol 1995;9:45–50. Morales AJ, Laughlin GA, Butzow T, Maheshwari H, Baumann G, Yen SS. Insulin, somatotropic, and luteinizing hormone axes in lean and obese women with polycystic ovary syndrome: common and distinct features. J Clin Endocrinol Metab 1996;81:2854 – 64. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Tapanainen JS. Insulin sensitivity, insulin secretion, and metabolic and hormonal parameters in healthy women and women with polycystic ovarian syndrome. Hum Reprod 2000;15:1266 –74. Aquino Carlos P, Hernandez Valencia M, Hicks Gomez JJ, Fonseca Yerena ME. The role of insulin-like growth factor in polycystic ovary syndrome. Ginecol Obstet Mex 1999;67:267–71. de Boer JA, Lambalk CB, Hendriks HH, van Aken C, van der Veen EA, Schoemaker J. Growth hormone secretion is impaired but not related to insulin sensitivity in non-obese patients with polycystic ovary syndrome. Hum Reprod 2004;19:504 –9. Acar B, Kadanali S. Diminished growth hormone responses to L-dopa in polycystic ovarian disease. Fertil Steril 1993;60:984 –7. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocrinol Rev 1997; 18:774 – 800. Devesa J, Lois N, Arce V, Diaz MJ, Lima L, Tresguerres JA. The role
Fertility and Sterility姞
11.
12.
13.
14.
15.
16.
of sexual steroids in the modulation of growth hormone (GH) secretion in humans. Steroid Biochem Mol Biol 1991;40:165–73. Wu X, Sallinen K, Zhou S, Su Y, Pollanen P, Erkkola R. Androgen excess contributes to altered growth hormone/insulin-like growth factor-1 axis in nonobese women with polycystic ovary syndrome. Fertil Steril 2000;73:730 – 4. Iwashita M, Mimuro T, Watanabe M, Setoyama T, Matsuo A, et al. Plasma levels of insulin-like growth factor-I and its binding protein in polycystic ovary syndrome. Horm Res 1990;33 (Suppl) 2:21– 6. Guido M, Romualdi D, Mancini A, Lattanzi F, Villa P, Barini A, et al. Effect of the opioid blockade on the feeding-induced growth hormone response to growth hormone-releasing hormone in women with polycystic ovary syndrome. Fertil Steril 2002;78: 994 –1000. Lanzi R, Manzoni MF, Andreotti AC, Malighetti ME, Bianchi E, Sereni LP, et al. Evidence for an inhibitory effect of physiological levels of insulin on the growth hormone (GH) response to GHreleasing hormone in healthy subjects. J Clin Endocrinol Metab 1997;82:2239 – 43. Ho KY, Veldhuis JD, Johnson ML, Furlanetto R, Evans WS, Alberti KG, et al. Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. J Clin Invest 1988;81:968 –75. Diamond MP, Hallarman L, Starick-Zych K, Jones TW, ConnollyHoward M, Tamborlane WV, et al. Suppression of counterregulatory hormone response to hypoglycemia by insulin per se. J Clin Endocrinol Metab 1991;72:1388 –9.
1855
Polycystic ovary syndrome (PCOS) occurs in 6%–10% of women of reproductive age (1). Endocrine disorders such as insulin resistance and compensatory hyperinsulinemia are currently two of the main factors involved in the etiopathogenesis of PCOS (2). Obesity is associated with PCOS in about 50% of patients (3, 4). Obesity itself increases insulin resistance or acts as an aggravating factor of this disease. However, normal weight women with PCOS show a lower incidence of insulin resistance and compensatory hyperinsulinemia (5). Other factors such as the somatotrophic axis, growth hormone (GH), and insulin-like growth factors (IGF) play a role in the etiopathogenesis of this syndrome (6, 7). A lower GH response has been observed in women with PCOS compared to women with ovulatory cycles (8), whereas other investigators have not detected a difference in women with PCOS and have attributed the lower GH response to obesity (4). Therefore, both obese and normal weight women with PCOS present disturbances in GH secretion, but these alterations do not follow the same pattern. The exact mechanisms underlying GH secretion disorders in women with PCOS and the reason for the difference in the response of GH release to stimulation tests between obese and normal weight women with PCOS are still not completely understood. The objective of the present study was to determine GH and IGF-1 secretion after a levodopa (L-dopa) test and serum insulin and glucose levels after an oral glucose tolerance test (GTT) with 75 g glucose in obese and nonobese women with PCOS compared to women with ovulatory cycles. A prospective study was conducted on 40 women divided into four groups: group I, 11 obese women with Received June 11, 2004; revised and accepted October 28, 2004. Reprints requests: Rosana Maria dos Reis, M.D., Department of Gynecology and Obstetrics, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Av. Bandeirantes, 3900, 14049-900 Ribeirão Preto, Brazil (FAX: 16-6330946; E-mail: [email protected]).
1852
PCOS; group II, 9 normal weight women with PCOS; group III, 10 obese women with ovulatory cycles; and group IV, 10 normal weight women with ovulatory cycles. The diagnosis of PCOS was based on presence of oligomenorrhea/amenorrhea and clinical or biochemical signs of hyperandrogenism (9). The exclusion criteria were Cushing’s syndrome, 21-hydroxylase deficiency, thyroid dysfunction, hyperprolactinemia, and androgen-secreting tumors. Women with regular menstrual cycles with an interval of 28 –32 days and normal serum progesterone (ⱖ3 ng/mL) on the 22nd day of the cycle were considered to be ovulatory. Obesity was defined as a body mass index (BMI) ⱖ27 kg/m2 . None of the patients used medications during the 90 days preceding the study. The study was approved by the Ethics Committee of the institution and the patients gave written informed consent to participate. After an overnight fast, all patients were submitted to an GTT with 75 g of dextrose and fasting blood samples were collected every 30 minutes for 2 hours for the determination of insulin and glucose. The L-dopa test was performed 2–3 days after the GTT following the same criteria. A dose of 500 mg of L-dopa was administered orally for the measurement of GH and IGF-1. Any increase in GH levels over baseline (fasting) was considered to be a positive response. The area under the curve (AUC) for glucose (in milligrams per deciliter times 240 minutes), insulin (in micro IU per milliliter times 240 minutes), and GH (in nanograms per deciliter times 240 minutes) were calculated using the trapezoidal method. Plasma glucose was measured by the enzymatic glucose oxidase method. Serum insulin was determined by RIA using a solid-phase kit (DPC, Los Angeles, CA). Plasma GH was measured by a doubleantibody RIA and serum IGF-1 was determined by a radioimmunometric method using a double antibody (DSL 2800 kit without extraction; DSL, Webster, TX). Serum LH, FSH, and PRL were measured by chemiluminescence using DPC kits. Serum testosterone (T), androstenedione (A), 17␣-hydroxyprogesterone (17-OHP), and dehydroepiandrosterone sulfate were measured by RIA. insulin-like
Fertility and Sterility姞 Vol. 83, No. 6, June 2005 Copyright ©2005 American Society for Reproductive Medicine, Published by Elsevier Inc.
0015-0282/05/$30.00 doi:10.1016/j.fertnstert.2004.10.057
growth factor binding protein-3 (IGFBP-3) was determined by immunoradiometric assay (IRMA) using the DSL-6600 kit. Data were analyzed statistically by the nonparametric Kruskal-Wallis test and Spearman’s correlation coefficient. Multiple comparisons were made by the Dunn test. Normality of the data was evaluated by the nonparametric Kolmogorov-Smirnov test. The level of significance was set at 5%. The mean serum T values were higher in group I than in the group IV (group I, 61.25 ⫾ 17.80 ng/dL vs. group IV, 40.39 ⫾ 10.20 ng/dL, P⫽.025). The mean plasma AUC values for glucose were lower in group IV (9,467 ⫾ 1,673 mg/dL) than those observed for the groups I and III (group I, 12,544 ⫾ 2,515 mg/dL and group III, 13,528 ⫾ 2,667 mg/dL, Pⱕ.01). Group I (21.55 ⫾ 10.46 IU/mL) had higher fasting serum insulin levels than group IV (9.76 ⫾ 5.74 IU/mL), with Pⱕ.05. The mean AUC for insulin was higher in the groups I and II (group I, 17,479 ⫾ 12,403 IU/mL ⫻ 240 minutes and group II, 17,150 ⫾ 12,312 IU/mL ⫻ 240 minutes) than in the group IV (5,057 ⫾ 2,375 IU/mL ⫻ 240 minutes), P⫽.0046. Mean fasting plasma IGF-1 values were higher in group II (264.50 ⫾ 69.21 ng/mL) than in the others groups (group I, 154.58 ⫾ 60.99 ng/mL; group III, 165.17 ⫾ 100.65 ng/mL; and group IV, 141.30 ⫾ 49.31 ng/mL, Pⱕ.01). Regarding GH, all groups responded to the L-dopa test; however, a tendency toward a greater response was observed among normal weight women with and without PCOS. Obese women with PCOS had lower mean AUC–GH values (group I, 140.00 ⫾ 77.00 ng/mL ⫻ 240 minutes) than group IV (320.60 ⫾ 138.38 ng/mL ⫻ 240 minutes), P⫽.016. Analysis of the data for the four groups as a whole regarding the AUC–insulin, BMI, and AUC–GH revealed a negative correlation between AUC–GH and AUC–insulin (r ⫽ ⫺0.44, ⫺0.6633 to ⫺0.1342, P⫽.005) and between AUC–GH and BMI (r ⫽ ⫺0.56, ⫺0.7445 to ⫺0.2885, P⫽.0002). A positive correlation was observed between AUC–insulin and BMI (r ⫽ 0.32, 0.002958 to 0.5827, P⫽.042), and a negative correlation between fasting GH and T (r ⫽ ⫺0.4822, ⫺06951 to ⫺0.1917, P⫽.0016) (Table 1). An abnormal GH response to stimulation tests such as and GH-releasing hormone (GHRH) has been observed in women with PCOS (8). Thus, our findings suggest that the regulation of GH secretion is influenced by androgens, but other factors such as obesity/insulin resistance can interfere with the regulation of the somatotrophic axis in PCOS. Hyperandrogenism can contribute to a reduction in GH secretion after the L-dopa test because T directly stimulates somatostatin secretion (10). Wu et al. (11) detected in nonobese women with PCOS a decline in androgen levels (testosterone/androstenedione) and an increase in GH and the GH/IGF-1 ratio after the L-dopa test 6 months after ovarian wedge resection. L-dopa
Fertility and Sterility姞
Women with PCOS show an increase in the bioactivity of IGF-1 resulting from an increase in IGF-1 levels or from a reduction in IGFBP-1 concentrations (12). In the present study we observed higher IGF-1 values in normal weight women with PCOS and no variations in IGFBP-3 values compared to the other groups. The IGF-1 interferes with the regulation of GH secretion in women with PCOS, probably through the inhibition of hypothalamic GH secretion by somatostatin. In our study, obese women showed a lower response to the L-dopa test, as well as a negative correlation between obesity and the AUC for GH, in agreement with data reported by other investigators (4, 13). It is important to emphasize that obese women show an increase in hypothalamic somatostatin secretion (14), with the imbalance between opioid secretion and somatostatin being responsible for the low GH values observed in these patients. Some evidence suggests that normal weight women with PCOS also present an imbalance in opioid tone interfering with GH secretion, with this derangement differing from that observed in obese women with PCOS (13). Confirming previous data reported by our group (3), in the present study we observed high AUC–insulin values in women with PCOS irrespective of obesity and also in obese women with ovulatory cycles. Various studies have analyzed the role of insulin in GH secretion. In vivo studies have demonstrated that insulin blocks the peripheral action of GH (14). In healthy individuals, 24-hour GH secretion is higher after fasting (15). Euglycemic clamp studies have shown that the GH response to hypoglycemia is inversely proportional to the degree of hyperinsulinemia (16). No difference in AUC–GH values was observed between normal weight women with PCOS and the normal weight control group, confirming that the GH response to L-dopa in the fasting state was normal, although the women with PCOS were hyperinsulinemic. On the other hand, obese women with PCOS and hyperinsulinemia had lower AUC–GH values. This finding suggests a role of insulin feedback in the somatotrophic axis and indicates that in obese women with PCOS the excess weight and high circulating insulin concentrations interfere with the GH response to stimulation tests (13). Based on the present findings, we suggest that the presence of PCOS in normal weight women associated with high GH and IGF-1 values indicates the involvement of two different mechanisms in the physiopathology of this syndrome according to BMI. Thus, the combined action of GH and IGF-1 might be responsible for the elevation of LH and the consequent hyperandrogenism in nonobese women with PCOS, whereas in obese patients with PCOS obesity and insulin resistance promote a reduction in GH secretion. 1853
TABLE 1 Clinical characteristics and hormone levels of women with polycystic ovary syndrome (obese/normal weight) and women with ovulatory cycles (obese/normal weight). Women with PCOS
Age (y) BMI (kg/m2) LH/FSH Testosterone (ng/dL) Androstenedione (ng/dL) IGF-1 (ng/mL) basal IGF-1 (ng/mL) 30 minutes Fasting glucose (mg/dL) Fasting insulin (IU/mL) Fasting GH (ng/mL) AUC for glucose (mg/dL ⫻ 240 minutes) AUC for insulin (IU/mL ⫻ 240 minutes) AUC for GH (ng/mL ⫻ 240 minutes) GH response to L-dopa IGFBP-3 (ng/mL)
Women with ovulatory cycles
Obese (group I) (n ⴝ 11)
Normal weight (group II) (n ⴝ 9)
Obese (group III) (n ⴝ 10)
Normal weight (group IV) (n ⴝ 10)
23.73 ⫾ 2.90a 35.70 ⫾ 4.94c 2.53 ⫾ 0.98e 61.25 ⫾ 17.18g 180.01 ⫾ 92.38 154.58 ⫾ 60.99 168.73 ⫾ 72.43 79.45 ⫾ 12.10 21.55 ⫾ 10.46i 0.37 ⫾ 0.47 12,544 ⫾ 2,515
21.67 ⫾ 3.84b 22.31 ⫾ 2.73 2.54 ⫾ 0.81f 56.13 ⫾ 18.55 217.53 ⫾ 98.03 264.50 ⫾ 69.21h 264.53 ⫾ 80.28 80.11 ⫾ 7.85 13.16 ⫾ 6.72 1.24 ⫾ 2.51 11,988 ⫾ 2,434
32.00 ⫾ 7.66 36.09 ⫾ 6.22d 0.58 ⫾ 0.44 43.07 ⫾ 20.97 175.10 ⫾ 101.92 165.17 ⫾ 100.65 176.20 ⫾ 112.87 87.30 ⫾ 4.00 19.97 ⫾ 9.89 0.39 ⫾ 0.27 13,528 ⫾ 2,667
31.60 ⫾ 3.34 22.26 ⫾ 2.51 1.03 ⫾ 0.83 40.39 ⫾ 10.20 145.40 ⫾ 58.65 141.30 ⫾ 49.31 197.50 ⫾ 83.36 82.20 ⫾ 9.05 9.76 ⫾ 5.74 1.42 ⫾ 1.85 9,467 ⫾ 1,673j
17,479 ⫾ 12,403
17,151 ⫾ 12,312
11,538 ⫾ 6,914
5,057 ⫾ 2,375k
140.00 ⫾ 77.00l
334.77 ⫾ 326.68
142.34 ⫾ 106.21
320.60 ⫾ 138.38
4.83 ⫾ 5.72 3,081 ⫾ 502,40
2.33 ⫾ 1.91 2,923 ⫾ 479,20
4.53 ⫾ 3.06 2,683 ⫾ 696,50
2.07 ⫾ 1.42 2,749 ⫾ 599,60
Note: Data are reported as mean ⫾ SD. PCOS ⫽ polycystic ovary syndrome; BMI ⫽ body mass index; IGF-1 ⫽ insulin-like growth factor-1; GH ⫽ growth hormone; AUC ⫽ area under the curve. a P⬍.01, group I vs. groups III and IV. b P⬍.01, group II vs. groups III and IV. c P⬍.01, group I vs. groups II and IV. d P⬍.01, group III vs. groups II and IV. e P⬍.01, group I vs. groups III and IV. f P⬍.01, group II vs. groups III and IV. g P⬍.05, group I vs. group IV. h P⬍.01, group II vs. groups I, III, and IV. i P⬍.05, group I vs. group IV. j P⬍.01, group IV vs. groups I and III. k P⬍.01, group IV vs. groups I and group II. l P⬍.05, group I vs. group IV. Premoli. Growth hormone and PCOS. Fertil Steril 2005.
Acknowledgments: The authors thank Elettra Greene for translating the Portuguese original. Research was supported by grants from Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP).
Ana Cristina Premoli, M.D. Laura Ferreira Santana, M.D. Rui Alberto Ferriani, M.D. Marcos Dias de Moura, M.D. Marcos Felipe Silva De Sá, M.D. Rosana Maria dos Reis, M.D. Sector of Human Reproduction, Department of 1854
Premoli et al.
Correspondence
Gynecology and Obstetrics, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, Brazil
REFERENCES 1. Asuncion M, Calvo RM, San Millan JL, Sancho J, Avila S, EscobarMorreale HF. A prospective study of the prevalence of the polycystic ovary syndrome in unselected Caucasian women from Spain. J Clin Endocrinol Metab 2000;85:2434 – 8. 2. Dunaif A, Segal KR, Futterweit W, Dobrjansky A. Profound periph-
Vol. 83, No. 6, June 2005
3.
4.
5.
6.
7.
8. 9.
10.
eral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 1989;38:1165–74. Reis RM, Foss MC, de Moura MD, Ferriani RA, Silva de Sa MF. Insulin secretion in obese and non-obese women with polycystic ovary syndrome and its relationship with hyperandrogenism. Gynecol Endocrinol 1995;9:45–50. Morales AJ, Laughlin GA, Butzow T, Maheshwari H, Baumann G, Yen SS. Insulin, somatotropic, and luteinizing hormone axes in lean and obese women with polycystic ovary syndrome: common and distinct features. J Clin Endocrinol Metab 1996;81:2854 – 64. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Tapanainen JS. Insulin sensitivity, insulin secretion, and metabolic and hormonal parameters in healthy women and women with polycystic ovarian syndrome. Hum Reprod 2000;15:1266 –74. Aquino Carlos P, Hernandez Valencia M, Hicks Gomez JJ, Fonseca Yerena ME. The role of insulin-like growth factor in polycystic ovary syndrome. Ginecol Obstet Mex 1999;67:267–71. de Boer JA, Lambalk CB, Hendriks HH, van Aken C, van der Veen EA, Schoemaker J. Growth hormone secretion is impaired but not related to insulin sensitivity in non-obese patients with polycystic ovary syndrome. Hum Reprod 2004;19:504 –9. Acar B, Kadanali S. Diminished growth hormone responses to L-dopa in polycystic ovarian disease. Fertil Steril 1993;60:984 –7. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocrinol Rev 1997; 18:774 – 800. Devesa J, Lois N, Arce V, Diaz MJ, Lima L, Tresguerres JA. The role
Fertility and Sterility姞
11.
12.
13.
14.
15.
16.
of sexual steroids in the modulation of growth hormone (GH) secretion in humans. Steroid Biochem Mol Biol 1991;40:165–73. Wu X, Sallinen K, Zhou S, Su Y, Pollanen P, Erkkola R. Androgen excess contributes to altered growth hormone/insulin-like growth factor-1 axis in nonobese women with polycystic ovary syndrome. Fertil Steril 2000;73:730 – 4. Iwashita M, Mimuro T, Watanabe M, Setoyama T, Matsuo A, et al. Plasma levels of insulin-like growth factor-I and its binding protein in polycystic ovary syndrome. Horm Res 1990;33 (Suppl) 2:21– 6. Guido M, Romualdi D, Mancini A, Lattanzi F, Villa P, Barini A, et al. Effect of the opioid blockade on the feeding-induced growth hormone response to growth hormone-releasing hormone in women with polycystic ovary syndrome. Fertil Steril 2002;78: 994 –1000. Lanzi R, Manzoni MF, Andreotti AC, Malighetti ME, Bianchi E, Sereni LP, et al. Evidence for an inhibitory effect of physiological levels of insulin on the growth hormone (GH) response to GHreleasing hormone in healthy subjects. J Clin Endocrinol Metab 1997;82:2239 – 43. Ho KY, Veldhuis JD, Johnson ML, Furlanetto R, Evans WS, Alberti KG, et al. Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. J Clin Invest 1988;81:968 –75. Diamond MP, Hallarman L, Starick-Zych K, Jones TW, ConnollyHoward M, Tamborlane WV, et al. Suppression of counterregulatory hormone response to hypoglycemia by insulin per se. J Clin Endocrinol Metab 1991;72:1388 –9.
1855