PCOS

Gonadotropin pulsatility in Cushing’s syndrome compared withpolycystic ovary syndrome

ZORANA PENEZIC, MILOS ZARKOVIC, SVETLANA VUJOVIC, JASMINA CIRIC,

BILJANA BELESLIN, MIOMIRA IVOVIC, ANA POKRAJAC, & MILKA DREZGIC

Institute of Endocrinology, University of Belgrade, School of Medicine, Belgrade, Serbia and Montenegro

AbstractMany of the presenting features in women with Cushing’s syndrome (CS) are similar to those observed for patients withpolycystic ovary syndrome (PCOS). The aim of this study was to compare gonadotropin pulsatility characteristics in CS andPCOS. We evaluated 32 females divided into three groups. The first group comprised 12 females with clinically andbiochemically proven CS, subsequently confirmed by histology (seven with Cushing’s syndrome, five with adrenal adenoma).The second group comprised ten females with clinical, endocrine and ultrasonographic parameters for PCOS, while the thirdgroup comprised ten healthy females with regular menstrual cycles to serve as controls. Blood samples were taken at 15-minintervals for 6 h in the follicular phase, for determination of luteinizing hormone (LH) and follicle-stimulation hormone(FSH). Pulse analysis was carried out using the PulsDetekt program, and statistical analysis was done using the Kruskal–Wallis test. The following data, presented as median (minimum–maximum), were found for the three groups respectively.Number of LH pulses: 0 (0–5), 7 (3–8) and 3 (2–7); LH pulse amplitude: 2.29 (1.98–3.49), 2.27 (1.15–5.90) and 2.03 (1.02–4.46) mU/l; LH pulse mass: 17.81 (14.82–26.20), 29.85 (8.59–185.82) and 27.57 (7.63–66.69) mU/l6min. Number ofFSH pulses: 3 (0–3), 2 (0–5) and 3 (1–5); FSH pulse amplitude: 1.62 (1.29–1.94), 1.49 (1.19–4.40) and 2.02 (1.37–2.52)mU/l; FSH pulse mass: 12.17 (9.64–41.69), 11.18 (8.92–33.02) and 15.16 (10.31–18.93) mU/l6min. Only the number ofpulses was compared because other parameters of pulsatile secretion cannot be estimated when no pulses are detected. Thedifference in number of LH pulses between groups was statistically significant (p5 0.05); however, there was no difference inthe number of detected FSH pulses between groups (p4 0.05). Attenuation of pulsatile LH secretion indicatinggonadotropin deficiency in the majority of women with CS is mostly due to alterations in serum cortisol levels. Our data alsosuggest that different mechanisms alter LH pulsatile secretion in CS and PCOS.

Keywords: Cushing’s syndrome, polycystic ovary syndrome, pulsatile secretion, gonadotropins

Introduction

Menstrual irregularity and gonadal dysfunction are

common characteristics of women with Cushing’s

syndrome (CS) [1,2]. The precise causes are not

clear, although two possible mechanisms, acting

separately or in concern, have been proposed that

lead to dysfunction of gonadotropin-releasing hor-

mone (GnRH) secretion and increased androgen

and estrogen production [3,4]. The GnRH pulse

generator serves as the final processor by which the

hypothalamus and extrahypothalamic tissues con-

trol gonadotropin secretion, integrating neuronal

and hormonal inputs and transducing them to the

release of GnRH [5]. The episodic release of

gonadotropins is under negative and positive feed-

back control of ovarian steroids, resulting in

dynamic changes in pulse patterns in different

phases of the menstrual cycle [6]. Also, there is

accumulating evidence to support a critical role of

paracrine factors at both hypothalamic and pituitary

level [7]. Among numerous receptors GnRH

neurons are known to express glucocorticoid

receptors [8], indicating a possible mechanism by

which hypercortisolemia leads to menstrual disor-

ders in CS.

CS and polycystic ovary syndrome (PCOS) share

many presenting features, such as menstrual

abnormalities, signs of hyperandrogenism, obesity

and insulin resistance, as well as some pathophy-

siological aspects [1,9,10]. Increased luteinizing

hormone (LH) pulse frequency and augmented

LH response to GnRH are characteristic of PCOS,

irrespective of obesity [11,12]. On the other hand,

the majority of women with CS have gonadotropin

deficiency with a relatively low estradiol level,

consistent with hypogonadotropic hypogonadism

[4]. We have evaluated gonadotropin pulsatility

among 12 patients with CS compared with PCOS

sufferers and healthy controls.

Correspondence: Z. Penezic, Institute of Endocrinology, Dr Subotica 13, 11000 Belgrade, Serbia and Montenegro. Tel: 381 11 361 63 17. Fax: 381 11 685 357.

E-mail: zpenezic@eunet.yu

Gynecological Endocrinology, March 2005; 20(3): 150 – 154

ISSN 0951-3590 print/ISSN 1473-0766 online # 2005 Taylor & Francis Group Ltd

DOI: 10.1080/09513590400027190

Gyn

ecol

End

ocri

nol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

DL

-UC

San

ta C

ruz

on 1

0/30

/14

For

pers

onal

use

onl

y.

Methods

Subjects

The study was performed among 12 women of

reproductive age with active CS, seven of them with

adrenocorticotropic hormone (ACTH)-secreting pi-

tuitary microadenoma and five with cortisol-secreting

adrenal adenoma. Mean age was 35.4+ 7.1 years,

mean body mass index (BMI) was 29.8+ 5.5 kg/m2.

Dynamic studies including circadian rhythm of

cortisol, low-dose dexamethasone test, ACTH level

and high-dose dexamethasone test all led to the

diagnosis of CS. Trans-sphenoidal surgery was

performed in seven patients and unilateral adrena-

lectomy in the remaining five. Histological

examination and postoperative testing confirmed the

initial diagnosis. One of them had regular menstrual

cycles (30 days), ten had oligomenorrhea (35–120

days) and one had amenorrhea (4 120 days).

The second group consisted of ten patients with

PCOS; mean age 26.5+ 3.7 years, mean BMI

31.3+ 4.1 kg/m2. PCOS was confirmed by clinical

(symptoms/signs of hyperandrogenism), endocrine

(LH/follicle-stimulating hormone (FSH) 4 2 or

elevated testosterone) and ultrasonographic para-

meters (ten or more subcapsular microcysts,

increased amount of stroma). Two of them had

regular menstrual cycles, seven had oligomenorrhoea

and one had amenorrhoea. Patients had not pre-

viously been treated with estrogens or progesterone,

and they did not use contraceptive pills during the

previous year. Hypercorticism was excluded by

adequate cortisol suppression after administration

of 1 mg dexamethasone at 23.00 hours [10]. All

patients from the PCOS group were monitored for at

least 1 year, and there was no progression of clinical

characteristics or changes in blood pressure or blood

glucose level.

There were ten healthy females in the control

group, mean age 28.3+ 4.4 years and mean BMI

23.8+ 2.2 kg/m2, with regular menstrual cycles of

28+ 6 days. All had normal blood pressure, normal

oral glucose tolerance test, normal liver and renal

function tests, and no significant psychiatric distur-

bances. They were not taking any drugs nor did they

have a history of alcohol abuse.

Study protocol

All patients with regular menstrual cycles and

oligomenorrhea were studied in the follicular phase,

from day 7 to day 10 of the cycle. Two females with

amenorrhea were considered to be in follicular phase,

still amenorrheic 1 month after testing. For the

assessment of gonadotropin pulsatility, blood was

drawn via an in-dwelling catheter for 6 h, starting at

14.00 hours, at intervals of 15 min. Informed con-

sent was obtained from all subjects. The local ethical

committee approved the study.

All blood samples were immediately separated and

kept frozen at – 208C until assayed. Plasma LH and

FSH concentrations were determined using immu-

noradiometric assay (INEP Zemun, Belgrade,

Serbia). Normal follicular-phase level of LH for the

method is 1–10 IU/l. The procedure has an intra-

assay coefficient of variation of 3.6%. Normal level of

FSH is 0–15 IU/l, with an intra-assay coefficient of

variation of 5.6%.

Pulse analysis

Data are presented as median, minimum–maximum.

Statistical analysis was done using the Kruskal–

Wallis test. Bonferroni correction was used to correct

significance for multiple comparisons. Pulse analysis

was carried out using the PulsDetekt program

[13,14].

Results

LH pulses were detected for four of the 12 patients

with CS. For the remaining eight patients, no LH

pulses were detected over the 6 h of observation. LH

pulsatility data are presented in Table I. Only the

number of pulses was compared because other

parameters of pulsatile secretion cannot be estimated

when no pulses are detected. The difference in the

number of LH pulses was statistically significant

between the CS and PCOS group (p=0.003),

between the CS and control group (p=0.012) and

between the PCOS and control group (p=0.023).

FSH pulsatility data are presented in Table II.

There was no significant difference in the number of

FSH pulses detected, or in any other parameter of

FSH pulsatile secretion, between groups (p4 0.05).

Mean LH level in CS was 1.99+ 1.96 IU/l; in

PCOS was 5.87+ 3.90 IU/l; and in controls

2.72+ 2.20 IU/l. Mean FSH level in CS was

2.62+ 1.38 IU/l; in PCOS was 4.32+ 1.97 IU/l;

and in controls 4.90+ 1.96 IU/l.

LH pulsatility profiles for three representative

subjects from each group are shown in Figure 1.

Table I. Pulsatility data for luteinizing hormone, presented as median and range (minimum–maximum).

Variable Cushing’s syndrome Polycystic ovary syndrome Controls

Number of pulses in 6 h 0 (0–5) 7 (3–8) 3 (2–7)

Interpulse interval (min) 360 (75–360) 54 (39–98) 98 (58–270)

Pulse amplitude (IU/l) 2.29 (1.98–3.49) 2.27 (1.15–5.90) 2.03 (1.02–4.46)

Pulse mass (mU/l6min) 17.81 (14.82–26.20) 29.85 (8.59–185.82) 27.57 (7.63–66.69)

Gonadotropin pulsatility in hypercorticism 151

Gyn

ecol

End

ocri

nol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

DL

-UC

San

ta C

ruz

on 1

0/30

/14

For

pers

onal

use

onl

y.

Discussion

The occurrence of regular ovulatory cycles is

dependent on a complex set of highly integrated

interactions between the central nervous system,

hypothalamus, pituitary and ovary. Loss of appro-

priately timed signals from any of these components

results in loss of cyclic functions, chronic anovulation

and amenorrhoea. The hypothalamic–pituitary–adre-

nal axis exerts profound, multilevel inhibitory effects

on the female reproductive system including: sup-

pression of hypothalamic GnRH secretion by

corticotropin-releasing hormone (CRH); inhibition

of GnRH, pituitary LH and ovarian estradiol secre-

tion by cortisol; and cortisol-induced target tissue

resistance to estradiol [15]. CS is associated with

decreased production of CRH, excluding very rare

ectopic CRH production, so the first mechanism is

not operative. Although glucocorticoids have long

been known to suppress gonadotropin secretion, the

mechanisms of this inhibition still remain unclear

[3,4,15–20].

Glucocorticoid effects depend on species, gonadal

status, and type, dose and duration of steroid excess.

Short-term cortisol excess such as 24-h hydrocorti-

sone infusion did not change mean or pulsatile LH or

FSH secretion [21], thus the effects of stress or

hypercortisolism on the gonadal axis may require

higher cortisol levels, more prolonged exposure or

some other mediator(s). Short-term moderate excess

achieved with dexamethasone was without influence

on episodic release of LH in normal men [22].

Women with regular cycles before and during

hydrocortisone administration demonstrated re-

duced LH pulse frequency and prolonged LH

interpulse intervals without altering pulse amplitude

[23]. A study of patients with Addison’s disease as an

in vivo model [24] suggested that the attenuation of

pulsatile LH secretion in man during hypo- and

hypercortisolism is due to variations in hypothalamic

Table II. Pulsatility data for follicle-stimulating hormone, presented as median and range (minimum–maximum).

Variable Cushing’s syndrome Polycystic ovary syndrome Controls

Number of pulses in 6 h 3 (0–3) 2 (0–5) 3 (1–5)

Interpulse interval (min) 135 (68–360) 95 (64–360) 105 (45–360)

Pulse amplitude (IU/l) 1.62 (1.29–1.94) 1.49 (1.19–4.40) 2.02 (1.37–2.52)

Pulse mass (mU/l6min) 12.17 (9.64–41.69) 11.18 (8.92–33.02) 15.16 (10.31–18.93)

Figure 1. Representative pulsatility profiles of luteinizing hormone (LH). Three subjects from each group are presented: top row, controls;

middle row, polycystic ovary syndrome; bottom row, Cushing’s syndrome. *, measured LH concentrations; ^, location of detected pulses;

upper full line, fitted values; lower full line, mathematically obtained secretion.

152 Z. Penezic et al.

Gyn

ecol

End

ocri

nol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

DL

-UC

San

ta C

ruz

on 1

0/30

/14

For

pers

onal

use

onl

y.

opioid activity secondary to alterations in serum

cortisol levels. Gender differences might be ex-

plained by a higher level of opioid receptor activity

in men than in low-estrogen women. Stress-like

concentrations of cortisol block or delay follicular

development and the preovulatory surge of LH in

sheep, while episodic GnRH overrides cortisol-

induced delay in follicular maturation [25]. This

anti-ovulatory effect of cortisol is consistent with the

response to exogenous glucocorticoid reported in

women [23].

Studies of patients with CS as a state of prolonged

hypercortisolemia with suppressed CRH level sug-

gested reduced basal levels of gonadotropins with a

low estrogen level, indicative of hypogonadotropic

hypogonadism, irrespective of coexistent hyperan-

drogenemia [4,16–20]. At the same time, response to

GnRH stimulation was poor, normal or exaggerated.

A recent study demonstrated that, in the ovariecto-

mized ewe, cortisol acutely inhibits the pulsatile

release of LH by suppressing pituitary responsiveness

to GnRH rather than by inhibiting hypothalamic

GnRH release [20].

Our studies were performed in 12 premenopausal

females with active CS. In eight patients no LH

pulses were detected, while in four of them one, two,

four and eight LH pulses were detected during 6 h.

Of the four patients with detected LH pulsatility, one

had regular menstrual cycles and three were oligo-

menorrheic. Three had pituitary microadenoma and

one adrenal adenoma. Compared with healthy

controls, LH pulsatility in CS is markedly sup-

pressed, with mean LH levels suggesting

gonadotropin deficiency. There was no difference

in LH pulsatility between ACTH-dependent and

ACTH-independent CS. It was previously reported

that ACTH alone did not control adrenal androgen

secretion in CS [26], suggesting an influence of

cortisol on LH pulsatility in CS.

We have chosen PCOS for comparison since CS

and PCOS share many presenting features, as well as

some pathophysiological aspects [1,12]. Several

investigations in women with PCOS have resulted

in a consensus that plasma LH is commonly elevated,

whereas FSH is within the lower follicular-phase

range [11,12,27,28]. Sensitive immunoassays and

multiple measurements of serum LH have revealed

marked increases in the amplitude and frequency of

spontaneous LH pulses in women with PCOS. The

increased amplitude is in accordance with enhanced

LH response to GnRH, suggesting gonadotropin-

enhanced sensitivity to GnRH. The increase in LH

pulse frequency suggests a possible abnormality in

the regulation of GnRH secretion. A critical question

is whether the rapid GnRH pulse frequency is a

primary hypothalamic defect or secondary to the

elevated plasma estrogens, androgens or insulin

concentrations present in PCOS [12,28]. Adrenal

androgen excess found in a significant fraction of

PCOS appears to be due to an alteration in the

intrinsic behavior of the adrenal cortex and not to

abnormalities of its hypothalamic–pituitary control

[29]. Circulating ACTH levels are not higher

compared with those in normal women [30].

In our group of patients with PCOS, LH pulse

amplitude was significantly higher compared with CS

and controls, as well as pulse frequency. This is in

agreement with published data of gonadotropin

pulsatility in PCOS [12,27,28]. These data also

suggest that the pathophysiological mechanisms

altering LH pulsatile secretion in PCOS and CS

are different.

Six hours of sampling might not be enough

because pulsatile secretion of FSH is not easily

detectable in women of reproductive age owing to the

low amplitude as a result of its relatively long half-life

compared with LH, although there is a significant

correlation with LH secretion [6]. Also, none of the

parameters obtained by pulse detection was signifi-

cantly different between groups.

Although CS and PCOS – as apparently distinct

disorders – may either coexist or be interrelated

[9,10], there is different form of LH pulsatility

imbalance: relative LH excess in PCOS compared

with significant suppression of LH pulsatility in CS.

Attention should be paid to the likelihood of CS in

women with menstrual irregularities and signs of

hyperandrogenism who have hypogonadotropic hy-

pogonadism, with or without the morphology of

PCOS.

References

1. Raff H, Findling JW. A physiological approach to diagnosis of

the Cushing’s syndrome. Ann Intern Med 2003;138:980–91.

2. Newell-Price J, Trainer P, Besser M, Grossman A. The

diagnosis and differential diagnosis of Cushing’s syndrome

and pseudo-Cushing’s states. Endocr Rev 1998;19:647–72.

3. Yen SSC. Chronic anovulation caused by peripheral endo-

crine disorders. In: Yen SSC, editor. Reproductive

endocrinology, 3rd ed. Philadelphia (PA): WB Saunders

Co.; 1991. pp 612–3.

4. Lado-Abeal L, Rodriguez-Arnao J, Newell-Price JD, Perry

LA, Grossman AB, Besser GM, Trainer PJ. Menstrual

abnormalities in women with Cushing’s disease are correlated

with hypercortisolemia rather than raised circulating androgen

levels. J Clin Endocrinol Metab 1998;83:3083–8.

5. Stojilkovic SS, Krsmanovic LZ, Spergel DJ, Catt KJ.

Gonadotropin-releasing hormone neurons. Intrinsic pulsatility

and receptor-mediated regulation. Trends Endocrinol Metab

1994;5:201–9.

6. Filicori M, Santoro N, Merriam GR, Crowley WF Jr.

Characterization of the physiological pattern of episodic

gonadotropin secretion through the human menstrual cycle.

J Clin Endocrinol Metab 1986;62:1136–44.

7. Reis FM, Cobellis L, Luisi S, Driul L, Florio P, Faletti A,

Petraglia F. Paracrine/autocrine control of female reproduc-

tion. Gynecol Endocrinol 2000;14:464–75.

8. Ahima RS, Harlan RE. Glucocorticoid receptors in LHRH

neurons. Neuroendocrinology 1992;56:845–50.

9. Kaltsas GA, Korbonits M, Isidori AM, Webb JA, Trainer PJ,

Monson JP, Besser GM, Grossman AB. How common are

polycystic ovaries and the polycystic ovarian syndrome in

women with Cushing’s syndrome? Clin Endocrinol (Oxf)

2000;53:493–500.

Gonadotropin pulsatility in hypercorticism 153

Gyn

ecol

End

ocri

nol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

DL

-UC

San

ta C

ruz

on 1

0/30

/14

For

pers

onal

use

onl

y.

10. Putignano P, Bertolini M, Losa M, Cavagnini F. Screening for

Cushing’s syndrome in obese women with and without

polycystic ovary syndrome. J Endocrinol Invest

2003;26:539–44.

11. 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.

12. Marshal JC, Eagleson CA. Neuroendocrine aspects of poly-

cystic ovary syndrome. Endocrinol Metab Clin North Am

1999;28:295–324.

13. Zarkovic M, Ciric J, Penezic Z, Trbojevic B, Drezgic M.

Temporal coupling of luteinizing hormone and follicle

stimulating hormone secretion in polycystic ovary syndrome.

Gynecol Endocrinol 2001;15:381–8.

14. Zarkovic M, Ciric J, Stojanovic M, Penezic Z, Trbojevic B,

Dresgic M, Nesovic M. Effect of insulin sensitivity on pulsatile

insulin secretion. Eur J Endocrinol 1999;141:494–501.

15. Chrousos GP, Torpy DJ, Gold PW. Interactions between the

hypothalamic–pituitary–adrenal axis and the female reproduc-

tive system. Ann Intern Med 1998;129:229–40.

16. Odagiri E, Yamanaka Y, Ishiwatari N, Jibiki K, Demura R,

Demura H, Suda T, Shizume K. Studies on pituitary–gonadal

function in patients with Cushing’s syndrome. Endocrinol Jpn

1988;35:421–7.

17. Cuerda C, Estrada J, Marazuela M, Vicente A, Astigarraga B,

Bernabeu I, Diez S, Salto L. Anterior pituitary function in

Cushing’s syndrome: study of 36 patients. Endocrinol Jpn

1991;38:559–63.

18. Boccuzzi G, Angeli A, Bisbocci D, Fonzo D, Giadano GP,

Ceresa F. Effect of synthetic luteinizing hormone releasing

hormone (LH-RH) on the release of gonadotropins in

Cushing’s disease. J Clin Endocrinol Metab 1975:40:892–5.

19. White MC, Sanderson J, Mashiter K, Joplin GF. Gonado-

trophin levels in women with Cushing’s syndrome before and

after treatment. Clin Endocrinol (Oxf) 1981;14:23–9.

20. Breen KM, Karsch FJ. Does cortisol inhibit pulsatile luteiniz-

ing hormone secretion at the hypothalamic or pituitary level?

Endocrinology 2004;145:692–8.

21. Samuels MH, Luther M, Henry P, Ridgway EC. Effect of

hydrocortisone on pulsatile pituitary glycoprotein secretion. J

Clin Endocrinol Metab 1994;78:211–15.

22. Veldhuis JD, Lizarradle G, Iramanesh A. Divergent effects of

short term glucocorticoid excess on the gonadotropic and

somatotropic axes in normal men. J Clin Endocrinol Metab

1992;74:96–102.

23. Saketos M, Sharma N, Santono NF. Suppression of the

hypothalamic–pituitary–ovarian axis in normal women by

glucocorticoids. Biol Reprod 1993;49:1270–6.

24. Hangaard J, Andersen M, Grodum E, Koldkjaer O, Hagen C.

Pulsatile luteinizing hormone secretion in patients with

Addison’s disease. Impact of glucocorticoid substitution. J

Clin Endocrinol Metab 1998;83:736–43.

25. Daley CA, Macfarlane MS, Sakurai H, Adams TE. Effect of

stress-like concentrations of cortisol on follicular development

and the preovulatory surge of LH in sheep. J Reprod Fertil

1999;117:11–16.

26. Cunningham SK, McKenna TJ. Dissociation of adrenal

androgen and cortisol secretion in Cushing’s syndrome. Clin

Endocrinol (Oxf) 1994;41:795–800.

27. Franks S. Polycystic ovary syndrome. N Engl J Med

1995;333:853–61.

28. Abbot DH, Dumesic DA, Franks S. Developmental origin of

polycystic ovary syndrome – a hypothesis. J Endocrinol

2002;174:1–5.

29. Azziz R, Black V, Hines GA, Fox LM, Boots LR. Adrenal

androgen excess in the polycystic ovary syndrome: sensitivity

and responsivity of the hypothalamic–pituitary–adrenal axis. J

Clin Endocrinol Metab 1998;83:2317–23.

30. Stewart PM, Penn R, Holder R, Parton A, Ratcliffe JG,

London DR. The hypothalamo–pituitary–adrenal axis across

the normal menstrual cycle and in polycystic ovary syndrome.

Clin Endocrinol (Oxf) 1993;38:387–91.

154 Z. Penezic et al.

Gyn

ecol

End

ocri

nol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

DL

-UC

San

ta C

ruz

on 1

0/30

/14

For

pers

onal

use

onl

y.