The trusted source for
healthcare information and
Abstract & Commentary
Synopsis: Women with polycystic ovaries had nearly a 3-fold increase in gestational diabetes compared to normal women.
Source: Mikola M, et al. Hum Reprod. 2001;16:226-229.
Mikola and colleagues from helsinki evaluated the obstetrical outcome in 99 women with polycystic ovaries compared with 712 normal women. Women with polycystic ovaries were heavier and older, hyperandrogenic, and had more multiple births because of ovulation induction. Comparing singleton pregnancies, there were no differences in birth weight. However, the c-section rate was 2-fold higher in the women with polycystic ovaries. The most important observed difference was nearly a 3-fold increase in gestational diabetes in the polycystic ovary group that was still a significant 2-fold increase after adjustment for weight, age, nulliparity, and multiple pregnancy. There was no difference in the incidence of preeclampsia.
COMMENT BY LEON SPEROFF, MD
Obstetricians-gynecologists now recognize two common conditions associated with an increase in insulin resistance resulting in hyperinsulinemia: 1) the hormonal changes associated with pregnancy; and 2) anovulation with polycystic ovaries, especially in overweight women. Combining the two produces a challenge to the pancreas of sufficient degree to make gestational diabetes a likely consequence. However, reports in the literature have not been uniform. An early report indicated that obesity was the major risk factor, not the polycystic state, and another found no increase in gestational diabetes.1,2 One small study did report an increase in gestational diabetes.3 The reason for these differences, it seems to me, is that the studies have been retrospective in nature, and most important, screening for gestational diabetes was not uniformly applied to the subjects. I would expect that a careful, prospective study with glucose challenge screening applied to every patient would reveal an increased incidence of gestational diabetes in women with polycystic ovaries, although largely concentrated in overweight women. Until then, the combination of significant hyperinsulinemia and the high risk of developing adult onset diabetes mellitus at an early age makes it reasonable (indeed, mandatory) to provide appropriate screening and surveillance for gestational diabetes in every overweight pregnant woman who previously displayed all the characteristics of anovulation and polycystic ovaries.
Clinicians have rapidly learned that women with polycystic ovaries (especially overweight women) who are resistant to induction of ovulation with clomiphene successfully achieve pregnancy at a high rate with the use of metformin, either alone or in combination with clomiphene. This raises the question of whether it is appropriate to treat these pregnant women with an oral hypoglycemic agent. Sulfonylurea drugs have not been recommended because of their potential to produce hypoglycemia in the fetus and the possibility of teratogencity. Glyburide (a sulfonylurea drug), however, has been demonstrated in laboratory studies to cross the human placenta in insignificant amounts. A clinical trial randomized 404 women with gestational diabetes to glyburide or insulin, and the outcomes (birth weight, macrosomia, respiratory distress, hypoglycemia, fetal anomalies) were the same.4 Glyburide was not detected in the cord serum.
What information is available for metformin? Coetzee and colleagues in South Africa have advocated for more than 20 years the use of metformin during pregnancy, even in the first trimester, reporting no evidence of teratogenicity or neonatal hypoglycemia.5 A report from Copenhagen, however, indicated that women treated with metformin had a higher prevalence of preeclampsia and perinatal mortality compared with women treated with sufonylurea or insulin.6 It is not clear to me, however, whether this was due to metformin or the influence of other risk factors; only a randomized trial would give the answer. In the laboratory, metformin does not affect human placental glucose uptake or transport.7 In a small study of 19 women, metformin treatment continued through the first trimester reduced the rate of early spontaneous abortion, and there was no evidence of teratogenicity.8
Therefore, there is a growing story that treatment of pregnant women who were previously anovulatory and hyperinsulinemic with polycystic ovaries is worthwhile. At this point in time, it seems to me that treatment should not be initiated before 28 weeks of pregnancy, both because of the lingering concern of teratogenicity and because hypoglycemic treatment is not necessary any earlier. In an anovulatory, hyperinsulinemic woman with repetitive early pregnancy losses, I would offer the option of metformin treatment without interruption after pregnancy is achieved.
1. Gjonnaess H. Br J Obstet Gynaecol. 1989;96:714-719.
2. Wortsman J, et al. J Reprod Med. 1991;36:659-661.
3. Radon PA, et al. Obstet Gynecol. 1999;94:194-197.
4. Langer O, et al. N Engl J Med. 2000;343:1134-1138.
5. Coetzee EJ. S Afr Med J. 1984;65:635-637.
6. Hellmuth E, et al. Diabet Med. 2000;17:507-511.
7. Elliott BD, et al. Am J Obstet Gynecol. 1997;176:527-530.
8. Glueck CJ, et al. Fertil Steril. 2001;75:46-52.