106
VOLUME 8 NUMBER 3 • SEPTEMBER 2011
REVIEW
SA JOURNAL OF DIABETES & VASCULAR DISEASE
Metformin decreases thyrotropin in overweight women with
polycystic ovarian syndrome and hypothyroidism
SM Taghavi, H Rokni, S Fatemi
Abstract
O
bjective:
To assess the effect of metformin
administration on thyroid function in overweight
women with polycystic ovarian syndrome (PCOS).
Methods:
Twenty-seven overweight women with PCOS and
hypothyroidism were selected. Fifteen patients (group I)
were treated with metformin 1 500 mg/day for six months
and 12 patients (group II) with placebo. Serum thyrotropin
(TSH), free T3 and free T4 were measured at baseline and six
months after the beginning of the study.
Results:
A significant decrease (
p
<
0.001) in TSH levels was
observed in group I but not in group II subjects after six
months of metformin treatment. No significant change in
free T3 and free T4 was observed throughout the study in
any group.
Conclusion:
In obese PCOS patients with primary
hypothyroidism, metformin results in a significant fall and
sometimes normalisation of TSH, without causing any
reciprocal changes in other thyroid function parameters.
Keywords:
hypothyroidism, metformin, polycystic ovarian syndrome
(PCOS), thyrotropin (TSH)
Introduction
Polycystic ovarian syndrome (PCOS) is a common endocrinopathy
occurring in five to seven per cent of women of reproductive age
and primary hypothyroidism is also very prevalent in women and
occurs more commonly in PCOS.
1
Metformin has been used for the
treatment of type 2 diabetes and PCOS for many years. Recently
there have been some reports that metformin is able to influence
thyroid function tests, mainly by a decrease in serum levels of
thyrotropin (TSH).
2
We evaluated the interplay between metformin
and thyroid tests by following thyroid hormone levels in overweight
PCOS patients with hypothyroidism after treatment with metformin
or placebo.
Patients and methods
This study was performed in 27 overweight women with PCOS
and primary hypothyroidism. PCOS was diagnosed in accordance
with the Rotterdam consensus diagnostic criteria. Hypothyroidism
in all patients was subclinical and was diagnosed for the first time.
Fifteen patients (group I) were treated with metformin 1 500 mg/
day for six months and 12 patients (group II) with placebo. The
patients did not receive any other medication for treatment of
PCOS. Serum TSH, free T4 (FT
4
) and free T3 (FT
3
) were measured
once at baseline and six months after the beginning of the study.
All statistical analysis was performed using SPSS 14. Data are shown
as mean ± standard deviation. Paired
t
-tests were used to analyse
changes in thyroid function tests and BMI.
P
-values less than 0.05
were considered significant.
Results
Clinical characteristics and thyroid function tests in the two groups
of patients are summarised in Table 1. There was no significant
difference between mean age and BMI in the two groups. Mean
basal TSH levels decreased significantly in group I after six months
of metformin treatment (
p
< 0.001). In five patients in this group
(30%) serum TSH levels reached normal range. Mean basal TSH
levels in patients in group II did not change significantly after
placebo treatment. In one patient in this group, serum TSH levels
reached normal range, which can be seen in the natural course
of subclinical hypothyroidism.
3
Serum FT4 and FT3 levels did not
change significantly during the study in any group (Table 1).
Discussion
These data show for the first time that metformin administration in
overweight PCOS patients with primary subclinical hypothyroidism
results in a significant fall in TSH levels. This is an important finding
clinically, because hypothyroidism occurs in more than 10% of
PCOS patients
1
and metformin is a common prescription for them.
There are several possible explanations for the TSH-reducing effect
of metformin but they are speculative at the present time. A subtle
increase in the gastrointestinal absorption of levothyroxin is the first
possible mechanism, but a reduction in thyroid hormone levels was
not reciprocal with TSH in patients with hypothyroidism treated
with levothyroxin in initial case reports by Vigersky
et al
.
4
This
finding was also shown in another study by Cappelli
et al
. He also
Correspondence to: S Morteza Taghavi
Endocrine Research Center, Mashhad Medical University, Ahmad Abad Street,
Ghaem Hospital, Mashhad, Iran.
Email:
From:
Diabetes Vasc Dis Res
2011;
8
(1): 47–48
S Afr J Diabetes Vasc Dis
2011;
8
: 106–107.
Table 1.
Clinical characteristics and thyroid function tests in two groups of
patients (mean
±
SD)
Group I
before
treatment
Group I
after
treatment
Group II
before
treatment
Group II
after
treatment
BMI (kg/m
2
) 29.7
±
2.5 28.59
±
2.56 28.2
±
2.2 27.96
±
1.73
TSH (mIU/l)
7.78
±
1.74 6.14
±
2.47 8.02
±
2.21
8.82
±
2.89
FT4 (pg/ml)
16
±
2.26 15.6
±
2.45 15.8
±
2.2
16.5
±
3.8
FT3 (pg/ml)
4.8
±
0.99
4.83
±
1.26
4.6
±
1.1 4.67
±
1.28
Group 1: Fifteen patients treated with metformin 1 500 mg/day for six
months
Group 2: Twelve patients treated with placebo
