BACKGROUND: The impact of metformin medication in pregnant women with polycystic ovary syndrome on weight gain during pregnancy and after delivery and the impact on growth of the offspring are essentially unexplored.
METHODS: This is a follow-up study of a randomized controlled trial (The Metformin treatment in pregnant PCOS women study), conducted in 11 secondary care centers. Women with PCOS were randomized to metformin (2000 mg daily) or placebo from first trimester to delivery. Questionnaires were sent to 256 participants 1 year postpartum. Maternal weight development in pregnancy and the first year after delivery and offspring anthropometry at birth and weight 1 year postpartum were registered.
RESULTS: Women randomized to metformin gained less weight during pregnancy compared with those in the placebo group. In the newborns, there was no difference between the 2 groups in weight or length. One year postpartum, women who used metformin in pregnancy lost less weight and their infants were heavier than those in the placebo group (10.2 ± 1.2 kg vs 9.7 ± 1.1 kg, P = .003).
CONCLUSIONS: Women randomized to metformin were heavier in the first trimester, gained less weight in pregnancy, and lost less weight in the first year postpartum compared with women randomized to placebo. Children exposed to metformin weighed more at 1 year of age.
- PCOS —
- polycystic ovary syndrome
- PregMet —
- The Metformin treatment in pregnant PCOS women study
- RCT —
- randomized controlled trial
What’s Known on This Subject:
The use of metformin in pregnancy is increasing in the treatment of both gestational diabetes and polycystic ovary syndrome. Metformin crosses the placenta. Teratogenicity is not reported. Possible long-term effects are undetermined.
What This Study Adds:
Intrauterine metformin exposure seems to have long-term effects on infant weight. At 1 year of age, infants born to women and exposed to metformin weigh more than those exposed to placebo in utero.
The role of metformin treatment in pregnant women with polycystic ovary syndrome (PCOS) is not yet determined. Nonrandomized and retrospective studies and 1 small randomized controlled trial (RCT) indicate positive effects of metformin on pregnancy complications.1–7 A large RCT did not support these results.8
Although not approved in pregnancy, metformin is widely used. Metformin crosses the placenta and is present in fetal circulation in therapeutic concentrations.9 So far, no negative effects of metformin have been reported in the mother or in the offspring. Infants born to mothers with PCOS who used metformin in pregnancy did not have any adverse effect on birth length and weight, growth, or motor-social development in the first 18 months of life compared with a background population.10
In an RCT for women with gestational diabetes, randomized to metformin or insulin, 2-year-old children exposed to metformin in utero had more subcutaneous fat, but overall body fat was the same as in children whose mothers were treated with insulin alone.11 It is important to establish the possible long-term impact and safety of intrauterine metformin exposure in the offspring, and this can only be done in RCTs.
To investigate the possible effect of fetal metformin exposure in utero we performed a follow-up investigation of offspring and mothers from a previous RCT, in which women with PCOS were treated with metformin in pregnancy (The Metformin treatment in pregnant PCOS women [PregMet] study).8 We hypothesized that 1 year postpartum, (1) mothers in the metformin group would weigh less (as they did during pregnancy) compared with those in the placebo group and (2) infants exposed to metformin in utero would weigh less compared with those exposed to placebo.
The current study is a follow-up of The PregMet study. The PregMet study was a prospective, randomized, double-blind, multicenter trial that compared metformin 2000 mg daily with placebo from the first trimester to delivery.8
In the PregMet study the inclusion criteria were (1) PCOS diagnosed according to The Rotterdam Criteria,12 (2) age 18 to 45 years, (3) gestational age between 5 and 12 weeks, and (4) a singleton viable fetus shown on ultrasonography. The exclusion criteria were alanine aminotransferase level >90 IU/L, serum creatinine concentration >130 µmol/L, known alcohol abuse, previously diagnosed diabetes mellitus or fasting serum glucose >7.0 mmol/L at the time point of inclusion, treatment with oral glucocorticoids, or use of drugs known to interfere with metformin.
Two hundred seventy-four pregnancies (in 258 women) were randomly assigned to either metformin or placebo treatment (16 women participated twice). Randomization, blinding, and performed measurements are described in detail elsewhere.8
All participants received written and individual verbal counseling on diet and lifestyle at inclusion. Thereafter treatment with metformin hydrochloride 500 mg (Metformin; Weifa AS, Oslo, Norway) or identically coated placebo tablets was initiated. The participants took 1 tablet twice daily during the first week and thereafter 2 tablets twice daily until delivery, when study medication was stopped. To counteract a possible adverse effect of metformin on vitamin B levels, patients were advised to take 0.8 mg of folic acid daily and 1 daily multivitamin tablet containing both vitamin B6 and B12.
Standardized interviewer-administered questionnaires were used to obtain self-reported data on education, smoking habits, and study medication. Height was recorded at inclusion and weight at each prescheduled visit. Body weight was recorded with light clothes on and without shoes. Gestational age was determined by mid-pregnancy ultrasound examination, measuring biparietal diameter, femur length, and mean abdominal diameter of the fetus.
The Committee for Medical Research Ethics of Health Region IV, Norway, and The Norwegian Medicines Agency approved the study. Written informed consent was obtained from each patient before inclusion, and the Declaration of Helsinki was followed throughout the study. The study was conducted according to principles of “Good Clinical Practice,” and the trial is registered at www.clinicaltrials.gov as NCT00159536.
The Follow-up Study
The participants in The PregMet Study gave their written consent to be contacted after the end of the original study. Of the 274 included pregnancies (in 258 women) in The PregMet Study, 3 patients had miscarriages, 12 dropped out, 1 was excluded due to misdiagnosis, and 2 infants died perinatally. Two hundred forty women with 256 pregnancies were invited to participate in the follow-up study. One year after delivery, a questionnaire and prepaid envelope was sent by mail. A reminder was sent about 2 to 3 weeks later to nonrespondents. At this time point, the participants were not aware of whether they had been randomized to metformin or to placebo.
The participants were asked about their own weight and the infant’s weight (registered at the child`s weight card) at 12 months’ postpartum. In Norway, newborns and older infants are closely followed up in a public health care system free of charge. The mothers carry a “weight card” where the infant’s weights are regularly registered at different time points after birth by a public health nurse, also at 12 months of age.
All data entry, data management, and data analyses were performed at the Institute of Laboratory Medicine, Children’s and Women’s, Norwegian University of Science and Technology. The data were analyzed according to the intention-to-treat principle. PASW statistics version 18.0 for Windows (IBM SPSS Inc USA, Chicago, IL) was used. The differences between the study groups were compared with 2-tailed t tests for independent samples. Values are reported as means (SD) or absolute numbers. A χ2 test was used to test differences between the groups. If the smallest expected value in a cell was <5, we used the Fisher exact test. Associations were investigated with univariate and multivariate linear regression analyses. Two-tailed tests were used throughout, and P < .05 was considered significant. No adjustments for multiple testing were performed.
Role of the Funding Source
The Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology funded the study. Weifa AS (Oslo, Norway) supplied the study drug free of charge. None of the funding sources had a role in the collection, analysis, and interpretation of the data or in writing and deciding to submit the report.
Of the 256 (78%) women with PCOS who participated in The PregMet Study, 199 responded to the questionnaire, 1 year postpartum. Except for a higher BMI at inclusion (in the first trimester of pregnancy, before randomization), no differences were found in baseline data between those who were randomized to metformin or placebo treatment in pregnancy (Table 1).
Maternal Weight Development
Women in the metformin group gained less weight in pregnancy than did those in the placebo group. However, after delivery, the women in the placebo group lost more weight during the first year and had a lower BMI than did those in the metformin group 1 year after delivery (Fig 1). The change in BMI from the first trimester of pregnancy to 1 year postpartum was +1.0 ± 2.9 kg/m2 in the metformin group vs +0.2 ± 2.0 kg/m2 in the placebo group (P = .03) (Table 1).
Offspring Anthropometry at Birth
There were no differences in birth weight, birth length, and ponderal index between newborns who were exposed to metformin and those who were exposed to placebo in utero. Boys in the metformin group had higher birth weight, were longer, and had larger head circumference at birth compared with the placebo group (Table 1). However, when adjusted for gestational age, maternal smoking, maternal BMI, and maternal height, these differences disappeared (data not shown).
Offspring Weight Development
At 1 year of age, infants exposed to metformin in utero were 5% heavier compared with those exposed to placebo (10.2 ± 1.2 kg vs 9.7 ± 1.1 kg; P = .003) (Table 1). The difference remained significant in a multivariate regression analysis, where we adjusted for gestational age, birth weight, maternal smoking in pregnancy, maternal BMI, maternal height, and duration of breastfeeding (P = .001) (Table 2). Both boys and girls exposed to metformin tended to be heavier at 1 year of age (Table 3).
The most important findings of the current study are that (1) maternal BMI is higher at 1 year after delivery in participants who were randomized to metformin in pregnancy and stopped medication at delivery than in those randomized to placebo and (2) infants exposed to metformin in utero had higher body weight at 1 year of age compared with those exposed to placebo.
We have previously reported that metformin treatment in women with PCOS reduced weight gain in pregnancy.8 Contrary to our hypothesis, the current study shows that weight reduction after delivery is less in mothers who were randomized to metformin compared with those randomized to placebo during pregnancy. It could reflect that women in the metformin group at baseline were more overweight and gained more weigh after a pregnancy and postpartum period. However, we have adjusted for maternal baseline BMI, and the difference persists between the groups. We believe that higher BMI 1 year after delivery can be attributed to a rebound effect after ceased metformin medication at delivery.
At birth, there were no differences in weight or length between the 2 groups. Interestingly, at 1 year of age, metformin-exposed infants of each gender are heavier than placebo-exposed ones. This weight difference persisted also after adjustment for factors known to influence weight development and cannot be attributed to a “big mothers–big infants” phenomenon.
Unfortunately, we have no data on body composition of these infants. Accordingly we do not know whether the weight difference represents increased lean body mass, increased fat mass, or both. The probability that metformin may have lasting effects in children, as seen in the current study, is supported by data from small-for-gestational age girls with premature adrenarche.13 In these girls, treatment with metformin delayed premature menarche and prevented excessive weight gain. The weight effect persisted also after metformin treatment had been stopped.14 Taken together with our data, this indicates that metformin, when used during a critical time window, might induce long-term endocrine and/or metabolic changes. Imprinting of genes may be the mechanism involved. It has been shown that metformin has the potential to affect transcription of genes.15
This is the first report providing evidence on metformin influence on intrauterine development. Interestingly, this effect persists at least 1 year after birth, indicating that metformin may have long-term metabolic or endocrine effects in the offspring.
Although there were no differences in birth weight and length, at 1 year of age, both boys and girls exposed to metformin had higher weight compared with placebo-exposed boys and girls. Additional studies are needed to confirm and explain our findings and to establish the safety of intrauterine metformin exposure.
- Accepted June 26, 2012.
- Correspondence to Eszter Vanky, Department of Obstetrics and Gynecology, St Olavs Hospital, University Hospital of Trondheim, Olav Kyrres gt 16, 7006 Trondheim, Norway. E-mail:
Dr Carlsen made substantial contributions to the conception and design, analysis, and interpretation of data, in addition to writing the article and approving the version to be published. Dr Martinussen provided analysis and interpretation of data, in addition to drafting the article or revising it critically for important intellectual content and providing final approval of the version to be published. Dr Vanky made substantial contributions to the conception and design, acquisition of data, and analysis and interpretation of data, in addition to drafting the article or revising it critically for important intellectual content and providing final approval of the version to be published.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: The Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology (NTNU) funded the study. Weifa A/S (Oslo, Norway) supplied metformin and placebo tablets free of charge.
- Glueck CJ,
- Wang P,
- Goldenberg N,
- Sieve-Smith L
- Vanky E,
- Salvesen KA,
- Heimstad R,
- Fougner KJ,
- Romundstad P,
- Carlsen SM
- Glueck CJ,
- Goldenberg N,
- Pranikoff J,
- Loftspring M,
- Sieve L,
- Wang P
- Rowan JA,
- Rush EC,
- Obolonkin V,
- Battin M,
- Wouldes T,
- Hague WM
- Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group
- Ibáñez L,
- Valls C,
- Ong K,
- Dunger DB,
- de Zegher F
- Copyright © 2012 by the American Academy of Pediatrics