|Ahead of print publication
Effect of metformin therapy on thyroid-stimulating hormone levels in women with polycystic ovarian syndrome
Vinay Dhanpal1, Mala Dharmalingam2, Pramila Kalra3
1 Senior Resident, Department of Endocrinology, M S Ramaiah Medical College, Bengaluru, Karnataka, India
2 Professor, Department of Endocrinology, Ramaiah Medical College and Hospitals, Bengaluru, Karnataka, India
3 Department of Endocrinology, Professor and Consultant, M S Ramaiah Medical College and Hospitals, Bengaluru, Karnataka, India
|Date of Submission||23-Aug-2020|
|Date of Acceptance||14-Dec-2020|
|Date of Web Publication||19-Apr-2021|
M S Ramaiah Medical College and Hospitals, New Bel Road, MSRIT Post, Bengaluru - 560 054, Karnataka
Source of Support: None, Conflict of Interest: None
Introduction: Metformin has been shown to decrease thyroid-stimulating hormone (TSH) levels without effect on total T4 and total T3 levels, especially in patients with underlying thyroid dysfunction.
The Aim of the Study: To evaluate the effect of metformin therapy on TSH levels in polycystic ovarian syndrome (PCOS) patients who were euthyroid with or without treatment.
Design of the Study Nonrandomized prospective intervention trial.
Materials and Methods: The study included all euthyroid PCOS patients of the reproductive age group diagnosed according to the modified Rotterdam criteria and the patients were divided into two groups. The first group was put on lifestyle modification alone (Group-I), whereas the second group was put on lifestyle modification and metformin therapy (Group-II). In both groups of patients, TSH, total T4, and total T3 levels were done at baseline and after 3 months of follow-up.
Results: A total of 105 patients with PCOS were nonrandomly assigned to Group I (n = 53) and Group II (n = 52). The baseline parameters (age, body mass index, TSH, and Homeostatic Model Assessment for Insulin Resistance) were similar. Thirty-six patients in Group I and 39 in Group II were followed up for 3 months. The change in TSH levels in both groups was not significant at follow-up (Group I [2.56 ± 0.87 vs. 3.01 ± 1.54; P = 0.102] and Group II [2.90 ± 0.81 vs. 2.76 ± 1.26;P = 0.503]). In a subgroup analysis in patients who had thyroid dysfunction, there was a significant decrease in TSH levels in Group II (3.10 ± 0.54 vs. 2.57 ± 0.50;P = 0.031).
Conclusion: Metformin significantly decreased TSH levels in women with PCOS with underlying thyroid dysfunction, while it did not show any effect on women without underlying thyroid dysfunction.
Keywords: Metformin, polycystic ovarian syndrome, thyroid-stimulating hormone
|How to cite this URL:|
Dhanpal V, Dharmalingam M, Kalra P. Effect of metformin therapy on thyroid-stimulating hormone levels in women with polycystic ovarian syndrome. Thyroid Res Pract [Epub ahead of print] [cited 2021 Jun 22]. Available from: https://www.thetrp.net/preprintarticle.asp?id=314163
Metformin is the first-line recommended and commonly used drug for type 2 diabetes mellitus (T2DM). The Endocrine Society recommends metformin in women with polycystic ovarian syndrome (PCOS) who have T2DM or impaired glucose tolerance (IGT) who fail lifestyle modification and also as a second-line therapy in women with menstrual irregularities who cannot take or tolerate hormonal contraceptives. Metformin may directly decrease ovarian androgen production. Metformin has been shown to improve ovulation rates when combined with clomiphene citrate in women resistant to it. South Asians are more insulin resistant and have significantly lower insulin sensitivity compared to Caucasians and Europeans. with the same body mass index (BMI). Metformin monotherapy for 6 months resulted in regular menses within 4 months of treatment, but a consistent reversal toward pretreatment conditions was observed within 3 months of metformin withdrawal. Recently, there have been some reports that metformin can influence thyroid function tests, mainly by a decrease in serum levels of thyrotropin (thyroid-stimulating hormone [TSH]). In contrast, in other studies, metformin was not associated with changes in TSH levels in euthyroid patients without underlying thyroid dysfunction.,
| Introduction|| |
| Materials And Methods|| |
This was a nonrandomized prospective observational study conducted at the Department of Endocrinology for 1 year in a tertiary center in Bengaluru, India, after getting approval from the institutional ethics review board.
PCOS patients who were not on drug treatment for the past 6 months or were drug naïve and were euthyroid were recruited in the study. Patients who were overt hypothyroid or subclinical but were euthyroid with treatment at the time of inclusion into the study, without any requirement of change of levothyroxine dose, were also recruited in the study. The levothyroxine brand on which the patient was already on at the time of recruitment was continued.
Patients who were already on drug treatment for PCOS during the past 6 months of recruitment into the study, patients who were overt or subclinical hypothyroid, and had abnormal TSH on therapy at the time of presentation were excluded from the study. Patients with conditions altering TSH levels such as acute or chronic inflammatory processes of other tissues, malignancy, myocardial infarction, congestive heart failure, impaired renal or hepatic function, diabetes, disorders of the central nervous system, trauma, pregnancy or breastfeeding, and BMI above 35 kg/m were excluded from the study.
The study included all subsequent PCOS women of the reproductive age group (18–44 years). PCOS was diagnosed according to the modified Rotterdam criteria. All patients who were willing for enrollment and met inclusion and exclusion criteria were enrolled in the study after obtaining written informed consent. PCOS patients were divided into two groups. Group I was on lifestyle modification, whereas Group II was on lifestyle modification and metformin therapy (1000 mg OD). The decision to give metformin was made by the treating physician. Each group was further subdivided into two subgroups, subgroup A with thyroid dysfunction (overt hypothyroid and subclinical hypothyroid patients on treatment but euthyroid at the time of recruitment) and subgroup B with euthyroid patients (without underlying thyroid dysfunction). BMI was calculated by weight (kg)/height (m). In both groups of patients, TSH, total T3, total T4, and Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) levels were done at baseline and after 3 months of follow-up. Antithyroid peroxidase (TPO) antibodies were done at baseline.
Sample size and selection of patients
Based on the study conducted by Morteza Taghavi et al. and expecting similar results with 80% power, 95% confidence level, and 1.5 mIU/L as a minimum detectable difference between two groups, the sample size of a minimum of 50 patients in each group was calculated.
Serum TSH, total T4, total T3, and insulin analysis were done by the ELISA method (Calbiotech, USA). Anti-TPO was analyzed by the ELISA method (Hycor, USA). The reference ranges for TSH, total T4, and total T3 are 0.4–4.2 uIU/ml, 4.8–11.6 ug/dl, and 0.52–1.85 ng/ml respectively. Fasting blood glucose was analyzed by the hexokinase method. HOMA-IR was calculated by fasting blood glucose (mg/dl) × fasting serum insulin (uIU/ml)/405. Anti-TPO levels (U/ml) were considered to be negative if they were <35, equivocal between 35 and 50, and positive more than 50.
Descriptive statistics of TSH, total T3, total T4, and BMI were analyzed and presented in terms of mean and standard deviation. Paired t-test was used to compare the mean difference between baseline and follow-up at 3 months. Independent t-test was used to compare the mean TSH, T3, T4, and BMI between the two groups. All statistical analyses were performed using SPSS software 18.0 version (All the statistical analyses were performed using Statistical Package for Social Sciences version 18.0, SPSS, Inc.).
A total of 105 patients were recruited in the study (Group I [n = 53] and Group II [n = 52]). Age, BMI, HOMA-IR, and TSH were matched between the two groups at baseline (age [years] [25.13 ± 3.19 vs. 26.19 ± 2.64; P = 0.067], BMI kg/m [25.71 ± 1.78 vs. 25.87 ± 2.27; P = 0.67], HOMA-IR [4.08 ± 1.05 vs. 4.51 ± 1.15; P = 0.149], and TSH [2.50 ± 0.88 uIU/ml vs. 2.83 ± 0.88uIU/ml; P = 0.06]) [Table 1].
| Results|| |
Thirty-six patients in Group I and 39 patients in Group II were followed up for 3 months. The difference in TSH levels at follow-up compared to baseline was not significant in both groups (Group I [2.56 ± 0.87 vs. 3.01 ± 1.54; P = 0.102] and Group II [2.90 ± 0.81 vs. 2.76 ± 1.26; P = 0.503]) [Table 2].
On further subanalysis, TSH levels in subgroup A in Group II were found to be significantly low at follow-up (3.10 ± 0.54 vs. 2.57 ± 0.50; P = 0.031), whereas it was not significant in Group I (2.55 ± 0.65 vs. 2.95 ± 0.80; P = 0.199). TSH levels in subgroup B in both groups were not low at follow-up (Group I – 2.56 ± 0.95 vs. 3.03 ± 1.75; 0.201 and Group II – 2.82 ± 0.89 vs. 2.83 ± 1.45; P = 0.969) [Table 3]. The mean difference in TSH between both groups at follow-up was also not significant (0.45 ± 1.60 vs. −0.14 ± 1.30; P = 0.083).
There was a significant decrease in BMI in both the groups at follow-up (Group I – 25.71 ± 1.78 vs. 25.37 ± 1.86; P = 0.028 and Group II – 25.87 ± 2.27 vs. 25.23 ± 1.92; P = 0.001), but there was no significant change in the mean difference in BMI between the groups (0.337 ± 1.086 vs. 0.642 ± 1.289; P = 0.192). There was a significant change in HOMA-IR in Group II (n = 20) (3.33 ± 0.91 vs. 2.75 ± 0.86; P = 0.000) at follow-up, whereas there was no difference in Group I (n = 16) (3.22 ± 1.14 vs. 3.08 ± 0.81; P = 0.435).
Anti-TPO antibodies were positive in 9.5% of patients (Group I – n = 4/53 [7.5%] and Group II – n = 6/52 [11.5%]). There was no significant change in TSH levels in either groups in anti-TPO-positive women (Group I [2.95 ± 0.39 vs. 2.35 ± 1.18; P = 0.186] and Group II [2.62 ± 0.91 vs. 2.63 ± 0.50; P = 0.94]).
There was no significant change in total T4 at follow-up in both groups (Group I – 6.80 ± 1.21 vs. 6.75 ± 0.94; P = 0.804 and Group II – 6.40 ± 1.61 vs. 6.49 ± 1.35; P = 0.508). Neither there was any significant change in total T3 in both groups at follow-up (Group I – 1.28 ± 0.43 vs. 1.35 ± 0.47; P = 0.515 and Group II – 1.24 ± 0.45 vs. 1.38 ± 0.97; P = 0.700) [Table 2].
The data suggest that in PCOS women who have thyroid dysfunction, i.e., either overt or subclinical, metformin use was associated with a significant change in TSH levels. At the same time, it was not the case in patients who did not have thyroid dysfunction.
| Discussion|| |
The mean age of patients in this study was 25.66 ± 2.96 years, which is the same as the study conducted by Ibraheem (24.3 ± 4.9 years), whereas it was lower than the study done by Krysiak and Okopien (38 ± 3 years).
Euthyroid patients (with or without treatment) were recruited in the present study, and this is in contrast to other studies where they have recruited subclinical hypothyroid also., The dose of metformin used in the current study was 1000 mg OD, and this is similar to the study conducted by Ibraheem et al. (1000 mg), whereas the other studies conducted by Morteza Taghavi et al.,Krysiak and Okopien, and Rotondi et al., had used 1500 mg, 2.25 g, and 1.38 ± 0.53 g, respectively. Patients were followed up for 3 months in our study; this is similar to the study conducted by Ibraheem et al. where they had followed the patients for three months, but in contrast to other studies where follow-up was 6 months] and 4 months, respectively.,
The significant effect of metformin on TSH in patients with thyroid dysfunction was in concurrence with other studies conducted by Rotondi et al., in which they observed no significant change in the euthyroid group, while TSH-lowering effect was seen in hypothyroid patients. In contrast, a study conducted by Morteza Taghavi et al. and Ibraheem. showed a significant decrease in serum TSH levels in obese PCOS and subclinical hypothyroid women treated with metformin. In insulin-resistant PCOS patients, 6 months of treatment with insulin sensitizers was found to reduce TSH levels significantly.
In a study done by Dimic et al., it was shown that metformin not only has TSH-lowering effect in patients with type 2 DM and hypothyroidism but also in euthyroid TPOab-positive and levothyroxine-naive patients. They also concluded that TSH-lowering effect of metformin is not dependent on long-term metformin therapy.
There was no significant change in total T4 and total T3 levels in either group, and these results were similar to the studies conducted by Morteza Taghavi et al. for free T4 and free T3, Ibraheem for total T4 and total T3, and Rotondi et al. for free T4.
The mechanism of metformin action on the thyroid axis is complex and multifactorial. The affinity or the number of thyroid hormone receptors may be changed by metformin. It may increase the central dopaminergic tone or may directly act on TSH regulation. This enhances the effect of thyroid hormones on the pituitary gland. Metformin inhibits AMPK activity in the hypothalamus and enhances the inhibitory modulation of thyroid hormones on TSH secretion. In patients with impaired thyroid–hypophyseal feedback, the TSH reduction may happen, but not in patients with preserved feedback system. Thus, our findings of the significant reduction in the TSH levels after metformin therapy in the group, which was hypothyroid on treatment, can be explained. A meta-analysis performed by Lupoli et al. showed that metformin induced a reduction in TSH levels both in overt and in subclinical hypothyroidism. At the same time, there was no change in TSH levels in euthyroid patients.
The findings in this study were like other studies in both hypothyroid and euthyroid patients. However, a long-term follow-up study may be required to see a discernible effect in euthyroid patients also as they may not show a change in 3 months. The lowering effect of metformin on TSH seems to be independent of change in BMI. A lower dose of metformin compared to other studies probably may not have shown a significant change of TSH levels in the euthyroid group in our study. The unique factors in our study are the younger age group and the Asian population. Although there was no significant change in TSH levels in either group with anti-TPO positivity women, it could be due to the smaller number of patients in the anti-TPO positivity group.
Metformin significantly decreased TSH levels in women with PCOS with underlying thyroid dysfunction, whereas it did not show any effect on women without underlying thyroid dysfunction. The effect of the change on TSH levels was seen independent of change in BMI, so it is prudent to repeat TSH levels in patients with underlying thyroid dysfunction at follow-up who are initiated for metformin therapy. It is especially important in patients with TSH levels in low normal range who are planned for metformin therapy.
| Conclusion|| |
Financial support and sponsorship
This study was funded by the Endocrine Society of India (called as young researcher award for DM students).
Conflicts of interest
There are no conflicts of interest.
| References|| |
Legro RS, Arslanian SA, Ehrmann DA, Hoeger KM, Murad MH, Pasquali R, et al
. Diagnosis and treatment of polycystic ovary syndrome: An Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2013;98:4565-92.
Attia GR, Rainey WE, Carr BR. Metformin directly inhibits androgen production in human thecal cells. Fertil Steril 2001;76:517-24.
Kar S, Sanchita S. Clomiphene citrate, metformin or a combination of both as the first line ovulation induction drug for Asian Indian women with polycystic ovarian syndrome: A randomized controlled trial. J Hum Reprod Sci 2015;8:197-201.
] [Full text]
Raji A, Seely EW, Arky RA, Simonson DC. Body fat distribution and insulin resistance in healthy Asian Indians and Caucasians. J Clin Endocrinol Metab 2001;86:5366-71.
Sharp PS, Mohan V, Levy JC, Mather HM, Kohner EM. Insulin resistance in patients of Asian Indian and European origin with non-insulin dependent diabetes. Horm Metab Res 1987;19:84-5.
Ibáñez L, Valls C, Potau N, Marcos MV, de Zegher F. Sensitization to insulin in adolescent girls to normalize hirsutism, hyperandrogenism, oligomenorrhea, dyslipidemia, and hyperinsulinism after precocious pubarche. J Clin Endocrinol Metab 2000;85:3526-30.
Isidro ML, Penín MA, Nemiña R, Cordido F. Metformin reduces thyrotropin levels in obese, diabetic women with primary hypothyroidism on thyroxine replacement therapy. Endocrine 2007;32:79-82.
Lupoli R, Di Minno A, Tortora A, Ambrosino P, Lupoli GA, Di Minno MN. Effects of treatment with metformin on TSH levels: A meta-analysis of literature studies. J Clin Endocrinol Metab 2014;99:E143-8.
Dimic D, Golubovic MV, Radenkovic S, Radojkovic D, Pesic M. The effect of metformin on TSH levels in euthyroid and hypothyroid newly diagnosed diabetes mellitus type 2 patients. Bratisl Lek Listy 2016;117:433-5.
Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril 2004;81:19-25.
Morteza Taghavi S, Rokni H, Fatemi S. Metformin decreases thyrotropin in overweight women with polycystic ovarian syndrome and hypothyroidism. Diab Vasc Dis Res 2011;8:47-8.
Ibraheem QA. Influence of metformin administration on a modification of TSH, T3 and T4 level in women with polycystic ovarian syndrome. Al-Mustansiriyah J Sci 2012;23:4.
Krysiak R, Okopien B. The effect of metformin on the hypothalamic-pituitary-thyroid axis in women with polycystic ovary syndrome and subclinical hypothyroidism. J Clin Pharmacol 2015;55:45-9.
Rotondi M, Cappelli C, Magri F, Botta R, Dionisio R, Iacobello C, et al
. Thyroidal effect of metformin treatment in patients with polycystic ovary syndrome. Clin Endocrinol (Oxf) 2011;75:378-81.
Morgante G, Musacchio MC, Orvieto R, Massaro MG, De Leo V. Alterations in thyroid function among the different polycystic ovary syndrome phenotypes. Gynecol Endocrinol 2013;29:967-9.
Vigersky RA, Filmore-Nassar A, Glass AR. Thyrotropin suppression by metformin. J Clin Endocrinol Metab 2006;91:225-7.
López M, Varela L, Vázquez MJ, Rodríguez-Cuenca S, González CR, Velagapudi VR, et al
. Hypothalamic AMPK and fatty acid metabolism mediate thyroid regulation of energy balance. Nat Med 2010;16:1001-8.
[Table 1], [Table 2], [Table 3]