Metformin use during pregnancy: Is it really safe?

Metformin is the first-line drug for the treatment of patients with type 2 diabetes mellitus (T2DM). It is used as an adjunct to diet and physical activity, especially in obese individuals. Although there are suggestions from the UKPDS (UK Prospective Diabetes Study) that metformin lowers the incidence of macrovascular disease,1 the extent of this effect has been challenged thereafter.2, 3 Nevertheless, because of its blood glucose-lowering effects, reduction of appetite, and low price, metformin plays an important role in the T2DM treatment strategy.4 Metformin has been also used extensively for the treatment of women with polycystic ovary syndrome (PCOS), in order to improve insulin sensitivity, reduce androgen overproduction, and improve fertility.5 The metabolic features of metformin make the it an attractive option for the treatment of women with gestational diabetes mellitus (GDM), the type of diabetes that presents during pregnancy. In this situation, insulin resistance develops due to the effects of placental hormones, and GDM ensues in those women whose β-cells cannot increase their insulin production to keep up with the higher demands. Indeed, several studies have shown that treatment with metformin may efficiently regulate blood glucose levels in GDM and may delay the use of insulin.6-8 However, a recent study reported that women randomized to metformin had higher blood glucose levels throughout pregnancy, and a higher percentage of large-for-gestational age (LGA) babies (21% vs 6%) compared with those on insulin therapy.9 These results highlight not only the effective use of insulin, but also the reluctance to escalate pharmacologic treatment when blood glucose control is not sufficient. A 2011 study had reported higher subcutaneous fat in children exposed to metformin.10 In addition, metformin use in PCOS pregnancies is associated with a higher risk for the child of being overweight at the age of 4 years.11 In obese women, metformin had no significant effect on birth weight percentile.12 A study in 25 children exposed to metformin during the prenatal period reported higher fasting blood glucose and systolic blood pressure, but lower low-density lipoprotein cholesterol than children not exposed to metformin.13 The question of the safety of metformin for pregnant women and their children then arises, because only few long-term studies have assessed its metabolic effects. However, there are other issues that need our attention too. Briones et al.14 extensively described the fate of metformin after it leaves our body and how it appears in wastewater and the environment. In this occasion, metformin may act as a real endocrine disruptor. There is evidence that metformin, in a concentration found in the environment, can induce a type of intersex in fish.15-17 Metformin can upregulate a number of genes associated with an endocrine-disrupting profile, including those for the androgen receptor, 3β-hydroxysteroid dehydrogenase (HSD) and 17β-HSD in the male gonads.18 Further, metformin reduced testicular weight and testosterone production in prepubertal chickens.19 Metformin treatment during pregnancy caused changes in the sexual behavior of and reduced the sperm count in the male offspring of female Wistar rats.20 It must be acknowledged that there is still a lot of debate as to whether animal models do predict the effects of endocrine disruptors in humans.21 However, the number of reports in which the possible endocrine-disrupting effects of metformin have been studied in humans are very limited, because most, still few, studies have focused on anthropometric measures of the offspring.9-13 An in vitro study with human testes showed that metformin reduces testosterone production and testicular development.22 A study from Finland reported no difference in testicular size between boys whose mothers had been treated with metformin during pregnancy and boys whose mothers had been treated with insulin.23 However, in that study, the mean age of the boys was only 5 years, and specific data on puberty and the post-pubertal period are lacking. In summary, the use of metformin in GDM is gradually increasing. Arguments in favor of its use are the achievement of stable control, and possibly a reduced need to start insulin injections, which by themselves may mitigate the pregnancy-associated weight gain. The long-term effects of metformin use during pregnancy for the offspring are still unknown, and can only be fully assessed when children exposed to metformin in utero are carefully evaluated once they have reached adulthood. Metformin treatment during pregnancy should still be considered as experimental therapy and, until long-term safety data have been obtained, this must only be performed within the situation of a clinical trial. Providers of research funding should consider financing such long-term studies, or to provide funding for short-term studies with the request for long-term follow-up. It would be a good idea to start an international registry of studies with metformin in GDM in order to collaboratively collect long-term prospective data in the offspring and really answer the question about metformin's long-term safety for both the mother and the offspring. The authors declare no conflicts of interest. 二甲双胍是2型糖尿病(T2DM)患者的一线治疗药物。它作为饮食控制与体育活动的辅助治疗药物, 尤其适用于肥胖的患者。虽然来自UKPDS((UK Prospective Diabetes Study)的研究结果表明二甲双胍可以降低大血管疾病的发病率1, 但是这种效果的幅度却在此后受到了质疑2, 3。尽管如此, 由于它不但具有降低血糖的作用、而且可以减少食欲并且价格低廉, 所以二甲双胍在T2DM的治疗策略中占据了重要的地位4。目前二甲双胍亦用于治疗多囊卵巢综合征(polycystic ovary syndrome, PCOS)妇女的胰岛素敏感性, 降低高雄激素血症, 提高生育率5。 二甲双胍的代谢特征使其成为有吸引力的治疗妊娠期糖尿病(GDM, 妊娠期间存在的糖尿病)的药物。在妊娠状况下, 胎盘激素的影响导致了胰岛素抵抗, 从而在β细胞不能相应增加胰岛素的分泌量以跟上更高需求的妇女中导致了GDM的发生。事实上, 有一些研究已经表明GDM患者使用二甲双胍治疗可以有效地调节血糖水平, 并可以延缓使用胰岛素治疗6-8。然而, 最近有一项研究报告, 随机分配到二甲双胍治疗组的妇女与使用胰岛素治疗的妇女相比前者在整个妊娠期间的血糖水平都较高, 并且生产大胎龄婴儿的比例也更高(分别为21%与6%)9。这些结果不但强调了要有效地使用胰岛素治疗, 而且也发现有一些妇女即便血糖控制不佳也不愿意逐步升级药物治疗方案。据2011年的一项研究报告, 使用二甲双胍治疗后儿童的皮下脂肪含量更高10。另外, PCOS妇女在妊娠期间使用二甲双胍治疗可导致孩子4岁时体重超重的风险增加11。在肥胖的妇女中, 二甲双胍对婴儿出生体重的百分位数没有显著影响12。在一项调查了25名产前接触二甲双胍的儿童研究报告中, 与产前没有接触二甲双胍的儿童相比, 发现前者的空腹血糖与收缩压都更高, 但是低密度脂蛋白胆固醇却更低13。那么二甲双胍对孕妇及其子女的安全性问题就浮现出来了, 因为目前只有很少的长期研究评估了它的代谢作用。 然而, 还有其他问题需要引起我们的注意。Briones等14广泛地描述了二甲双胍离开人体后的去向以及它如何出现在废水与环境中。在这种情况下, 二甲双胍可能是一种真正的内分泌干扰物。有证据表明, 环境中的二甲双胍达到某种浓度后可诱导生成一种兼具有雌雄间性的鱼类15-17。二甲双胍能够使多种与内分泌紊乱状态有关的基因上调, 包括雄激素受体、3β-羟类固醇脱氢酶(HSD)以及17β-HSD在男性性腺中的表达18。此外, 二甲双胍可导致青春期前的鸡睾丸重量以及睾酮生成减少19。雌性Wistar大鼠孕期使用二甲双胍治疗后可导致雄性后代出现性行为改变以及精子数量减少20。我们必须认识到, 关于动物模型是否能够预测内分泌干扰物对人类的影响目前还存在许多争论21。然而, 在人类中研究二甲双胍有可能对内分泌干扰造成影响的报告数量非常有限, 因为大多数的研究都集中在后代的人体测量学指标上, 即使是这样数量仍然很少9-13。一项体外的人体检验结果表明, 二甲双胍可以减少睾酮的生成并且延缓睾丸的发育22。来自芬兰的一项研究表明, 母亲在怀孕期间接受二甲双胍治疗的男孩与母亲在怀孕期间接受胰岛素治疗的男孩相比较睾丸大小没有差异23, 但是在该研究中男孩的平均年龄只有5岁, 并且缺乏青春期与青春期后的特定数据。 总之, 二甲双胍在GDM中的应用正在逐渐增加。赞成使用它的论据是可以获得稳定的血糖控制, 并且有可能减少注射胰岛素的需求, 而且它本身还可以减少妊娠相关的体重增加。在怀孕期间使用二甲双胍治疗对后代的长期影响目前尚未明确, 对于在子宫中接触二甲双胍的儿童来说, 只有当他们成年时接受进一步的仔细评估后, 才能够对二甲双胍的影响进行全面的评估。妊娠期间使用二甲双胍治疗目前仍应被视为实验性治疗, 并且在获得长期的安全性数据之前, 必须仅在临床试验的情况下在妊娠期间使用二甲双胍治疗。研究经费提供者应该考虑为这类长期的研究提供资金, 或者为需要长期随访的短期研究提供资金。为了能够共同合作前瞻性地长期收集后代相关数据, 并真正回答有关二甲双胍对于母亲与后代的长期安全性问题, 针对GDM使用二甲双胍治疗开展国际性的登记研究是一个好办法。