Menstrual irregularity is common in women with acromegaly, occurring in 40-
84%. Although it has been attributed to gonadotropin deficiency and/or PRL
excess, it has not been evaluated in detail, and its pathogenesis is not we
ll understood. To explore the various possible pathogenic mechanisms, we ha
ve analyzed the clinical, endocrinological, and radiological characteristic
s of 47 women with active acromegaly within the reproductive age range (15-
41 yr) with respect to their menstrual pattern; 9 patients (19%) had normal
cycles, 7 (15%) had oligomenorrhea, 29 (62%) had amenorrhea, and 2 (4%) ha
d polymenorrhea. Compared to patients with normal cycles (n = 9), patients
with menstrual irregularity (oligo/polymenorrhea or amenorrhea; n = 38) wer
e more hirsute, had lower serum estradiol(normal: median, 76.5 pmol/L; rang
e, 20-570; menstrual irregularity: median, 283; range, 140-431; P < 0.01),
and sex hormone-binding globulin (SHBG; normal: median, 19.6 nmol/L; range,
5-52; menstrual irregularity: median, 48; range, 18-60; P < 0.01), but sim
ilar testosterone levels; in addition, patients with amenorrhea had higher
serum GH (normal: median, 100 mU/L; range, 8.8-400; amenorrhea: median, 30;
range, 19.7-120; P < 0.05). PRL levels in excess of 1000 mU/L were found i
n 16 of the 38 patients with menstrual irregularity compared to only 1 of t
he 9 patients with normal cycles. Patients with menstrual irregularity had
a greater impairment of anterior pituitary function than patients with norm
al cycles. Acromegalic patients who were defined as estrogen sufficient (es
tradiol, >140 pmol/L) had clinical baseline endocrine profiles and LH respo
nses to GnRH stimulation similar to those in patients with polycystic ovari
an disease. There was a positive correlation between GH levels and tumor si
ze (r = 0.35; P < 0.05) and an independent inverse correlation between GH a
nd SHBG levels (r = -0.6; P < 0.01), which persisted even in patients who w
ere estrogen sufficient, but there was no correlation between GH and estrad
iol levels; in addition, there was a negative correlation between estradiol
levels and tumor size (r = -0.42; P < 0.05). Thirty-five of the patients w
ith menstrual irregularity had meso- or macroadenomas and 3 had microadenom
a, whereas 6 of the 9 patients with normal cycles had microadenomas. In con
clusion, menstrual irregularity is common in women with acromegaly (81% of
our patients). Amenorrheic patients have higher GH levels, are mainly estro
gen deficient, and tend to have larger tumors than patients with normal cyc
les. However, the independent negative correlation between GH and SHBG leve
ls suggests that GH may, directly or indirectly, lead to a fall in SHBG, po
ssibly determined by the hyperinsulinemia known to occur in acromegaly. Low
SHBG levels may contribute to the menstrual disturbance seen in acromegaly
in addition to any gonadotropin deficiency or hyperprolactinemia and may a
ccount for hirsutism in the presence of normal testosterone levels.