Key words: GESTATIONAL DIABETES MELLITUS, HYPERTENSIVE DISORDER IN PREGNANCY, POLYCYSIC OVARY SYNDROME, PREMATURE DELIVERY
The aim of this study was to determine the prevalence of gestational diabetes mellitus (GDM) and the pregnancy outcomes in Asian women with polycystic ovary syndrome (PCOS). The retrospective cohort study was performed to compare pregnancy outcomes of 47 pregnancies in 41 PCOS women with 264 pregnancies in 222 women with normal menstruation. Logistic regression was used to assess the risk of PCOS on GDM, hypertensive disorder in pregnancy (HDP) and premature delivery. The mean age of both groups was 31 years. The mean body mass index (BMI) and proportion of BMI of > 25 kg/m^sup 2^ were significantly higher in the PCOS than in the control group. There was no difference in the prevalence of GDM between the PCOS women and the high-risk group of the controls. The prevalence of HDP and premature delivery was significantly greater in PCOS women (21.3 and 13.3%) than in the controls (6.4 and 5.4%), respectively. PCOS was demonstrated as a risk factor for GDM ami HDP with borderline statistical significance, but not for premature birth. The Cesarean section rate was higher in the PCOS than in the control group. In conclusion, the prevalence of GDM in Asian women with PCOS is high and comparable to those of a high-risk group.
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in women, characterized by menstrual irregularities and hyperandrogenism1. It is well accepted that insulin resistance (IR) is an important pathogenesis of PCOS2. IR is the cause of abnormal glucose tolerance”, therefore type 2 diabetes mellitus (I)M) and impaired glucose tolerance (IGT) can be found in PCOS women. IGT in 25-31% and 7.5-15.2% of type 2 DM in these women has been reported recently4-6.
Because pregnancy induces IR , pregnant women with PCOS may be at increased risk for gestational DM (GDM) and IGT. However, there have been contradictory results in the prevalence of GDM when compared between PCOS and control women8-12.
In addition, there is evidence that IR is associated with hypertension and hypertensive disorder in pregnancy (HDP)13. PCOS women may also be at risk for HDP. The prevalence of HDP and pre- eclampsia in PCOS has been reported to be higher than in controls8,9,14. By contrast, there have been studies that did not show this increased risk10,12.
Recently, some studies showed that the IR of Asian ethnic women who lived in Western countries was not lower than that of Caucasian women15. The fact that the prevalence of IGT and type 2 DM in Asian women ‘ was reported to be comparable to those in Caucasians4,5 provided evidence that there may be no difference in IR between the two ethnic groups. Like other ethnic populations, Asian pregnant women with PCOS would be at risk for GDM as well. However, there is no report on GDM in Asian women with PCOS. Therefore, the aim of this study was to determine the prevalence of GDM and pregnancy outcome in Asian women with PCOS.
Between 3 June 1996 and 31 May 2002, 41 PCOS women with 47 pregnancies who were treated in the Reproductive Endocrinology and Infertility Unit were enrolled in the study. The criteria for the diagnosis of PCOS according to Homburg16 were as follows: menstrual irregularities; clinical hyperandrogenism such as acne, seborrhea and hirsutism; and typical ultrasound presentation of bilateral polycystic ovaries with ten or more follicles of 2-8 mm in diameter, as described by Adams and colleagues17. Of the 47 pregnancies, six were conceived spontaneously and 17 were conceived after ovulation induction with clomiphene citrate (CC). The remaining were CC- resistant PCOS women, and pregnancies occurred after laparoscopic ovarian drilling, ovulation induction with metformin, and in vitro fertilization and embryo transfer in 15, eight and one women, respectively.
Starting from October 1997, the protocol for GDM screening in all pregnant women with PCOS was established. These pregnant women had a 3-h, 100 g oral glucose tolerance test (OGTT) performed at 24-28 weeks of gestation, according to the one-step method of the American Diabetic Association (ADA)18.
GDM was diagnosed if two or more plasma concentrations were at or exceeded the levels according to the ADA18 as follows: fasting, ≥ 95 mg/dl; 1 h, ≥ 180 mg/dl; 2 h, ≥ 155 mg/ dl; and 3 h, ≥ 140 mg/dl.
Both gestational hypertension and pre-eclampsia are components of HDP and were considered in this study. Gestational hypertension was defined as blood pressure of ≥ 140/90 mniHg with no proteinuria, and pre-eclampsia was diagnosed when hypertension was accompanied by proteinuria ( ≥ 300 mg/24 h)19. Delivery at
The controls had normal menstruation, were matched for age ( 2 years) and registered on the same day. The women who were at risk for GDM also had an OGTT performed at 24-28 weeks of gestation according to the one-step method of the ADA18. The pregnant women who delivered elsewhere or who did not receive antenatal care in this hospital or suffered a miscarriage before 20 weeks of gestation were excluded from the study.
All parameters of pregnancy outcomes were collected from medical records of both the pregnant women and the neonates. This study was approved by the ethical committee of our institute.
Unpaired t test, χ^sup 2^ test and Fisher’s exact test were used to compare the groups, where indicated. Univariate and multivariate logistic regressions were used to analyze risk factors for GDM, HDP and premature delivery. A value of p
There were 288 pregnancies in 246 women who were used as the controls. Of these, 24 pregnancies in 24 women were excluded from the study because 21 delivered elsewhere or did not receive antenatal care in this institute and three miscarried. The remaining 264 pregnancies of the control group and 47 of the study group were analyzed.
The mean age of both groups was 31 years. The proportion with age > 35 years was not different between the groups. The mean pre- pregnancy body weight, body mass index (BMI) and prevalence of obesity (BMI > 25 kg/m^sup 2^) were significantly higher in the PCOS than in the control group (p = 0.000). However, gestational weight gain was not different between the groups. Nulliparity was more prevalent in the PCOS women than in the controls (p = 0.004), whereas the multiple pregnancy rate was not different. All multiple pregnancies in this study were twin pregnancies. Four out of five twin pregnancies were premature. The Cesarean section rate was high in both groups, and significantly greater in the PCOS than in the control group (p = 0.004, Table 1).
Thirty-six (76.6%) of PCOS women and 100 (37.9%) of the controls received an OGTT. Eight and 18 women were diagnosed as GDM in the PCOS and in the control group, respectively. The prevalence of GDM did not differ between the PCOS women and the controls who had an OGTT (Table 2). After analysis by logistic regression, PCOS and the family history of DM were independent risk factors for GDM, with an odds ratio (OR) of 3.3 (95% CI 1.1-9.7) and 9.3 (95% CI 3.2-26.9), respectively, whereas age, BMI and parity were not in the singleton pregnancies (Table 3).
Ten (21.3%) of PCOS women and 17 (6.4%) of the controls were diagnosed with HDP. Of ten in the PCOS group, eight (17%) had gestational hypertension and two (4.3%) had pre-eclampsia. One case of pre-eclampsia was detected in the control group (Table 2). Using logistic regression, PCOS, age and GDM were the risk factors with borderline statistical significance in the singleton pregnancies (Table 4).
Table 1 Patients’ characteristics and maternal pregnancy outcomes
Table 5 shows the comparison of the neonatal outcomes of singleton pregnancies between the groups. Gestational age was significantly lower (p = 0.027) and premature rate was higher (p = 0.046) in women with PCOS than in the controls. Analysis of risk factors for premature delivery of the singleton pregnancies indicated that HDP was an independent risk factor (Table 6).
Two (4.3%) perinatal deaths occurred in PCOS women. One was a dead fetus in utero at 27 weeks of gestation. The cause of death could not be established in this case after the investigation. The other was a preterm neonate of a twin pregnancy at 22 weeks of gestation. One died of respiratory distress syndrome (RDS) 5 days after birth, and the another also had RDS but survived after intensive treatment.
Table 2 Prevalence of gestational diabetes mellitus (GDM) and hypertensive disorder in pregnancy (HDP) between the group with polycystic ovary syndrome (PCOS) and the control group
Table 3 Risk factors for gestational diabetes mellitus in the singleton pregnancies
Table 4 Risk factors for hypertensive disorder in pregnancy in the singleton pregnancie\s
This study has demonstrated that GDM was more frequently found in the PCOS group than in the control group, as has been found by others8,10. More than one decade ago, Gjonnaess14 reported that the incidence of GDM in PCOS women who conceived after laparoscopic ovarian drilling was 8.1%, which was more than that of the general population at that time. Subsequent studies showed that the prevalence of GDM in PCOS women was around 20%8,10, whereas it was 3- 9% in controls.
Although most studies have shown a higher prevalence of GDM in PCOS than in controls, a few studies, in contrast, did not demonstrate a difference in GDM rate between PCOS and controls11,12. In the study of Wortsman and colleagues , there was a difference of GDM rate between PCOS and the control women (16.7% in 22 and 6.7% in 44), but this did not reach statistical significance. With the small number of these populations studied, the power of testing was only 0.159, which could certainly not show a difference. Moreover, some authors demonstrated recently that the rate of GDM was not statistically different between PCOS and control women .
PCOS has been demonstrated as a risk factor for GDM with borderline statistical significance10. This could be explained, in part, by the increased IR and hyperinsulinemia in the pregnant PCOS women compared with the control women7,22 . Likewise, PCOS was a predictive factor for GDM in this study; IR in Asian women with PCOS may not be different from the other ethnic groups15.
Family history of DM is one of the indications for an OGTT according to the ADA18 and was more frequently found in PCOS women12. However, it was not considered as a risk factor in previous studies10,12. In this study, it was demonstrated as an independent risk factor for GDM
Mikola and colleagues10 reported that obesity was an independent predictor for GDM, but this was not shown in our study. There were some controversial reports about obesity as risk factors for GDM in women with PCOS. Some studies showed that the prevalence of GDM was increased by BMI10,14. In contrast, a study from Turkey demonstrated not only that the prevalence of GDM was higher in PCOS than in controls, but also that the differences occurred merely between the lean PCOS and the lean control women when compared between subdivided groups8.
Table 5 Comparison of neonatal outcomes of singleton pregnancies between the group with polycystic ovary syndrome (PCOS) and the control group
Table 6 Risk factors for pretcrm delivery in the singleton pregnancies
There have been reports of the risk of PCOS on HDP8,9,14. Some studies have shown a risk of PCOS on pre-eclampsia9,14, and some on gestational hypertension8. By contrast, Mikola and colleagues did not show the risk of PCOS on pre-eclampsia. In agreement with most of the previous studies, HDP and gestational hypertension were significantly more frequent m PCOS women than in the controls in this study. The risk of PCOS on HDP was more than three times that with respect to the controls. The risk of PCOS on HDP can be explained by the existence of IR. Previous studies showed that the pregnant women who developed HDP had hyperinsulinemia22. Moreover, there was a report of an increased risk for HDP among women with GDM23, supporting the finding in this study.
Although Urman and colleagues demonstrated that a low birth weight was more frequent in PCOS than in controls, other studies showed no differences in any parameters of neonatal outcomes comparing the singleton pregnancy between both groups10,12. On the other hand, this study showed the gestational age at birth was lower and the rate of prematurity was higher in the PCOS than in the control group. However, PCOS was not indicated as a risk factor for premature delivery. This finding was consistent with that of others8,10. It is well documented that pre-eclampsia is an important risk factor for preterm delivery . Therefore, HDP may, in part, be a risk factor that accounts for premature delivery in pregnant women with PCOS.
There was a report of one neonatal death with respiratory syndrome 20 h after a Cesarean section for abruptio placentae14. Urman and colleagues also reported one perinatal death in a PCOS woman with abruptio placentae, and one in the control group. In the current study, there were two perinatal deaths: one fetal death in utero and one of the twins with severe pre-eclampsia that died of respiratory distress syndrome. It can be seen that perinatal deaths in PCOS women were associated with severe pre-eclampsia and prematurity. Therefore, we cannot assume that PCOS is a risk for poor neonatal outcome.
The Cesarean section rate was high, both in the PCOS group and in the control group, and higher with respect to the other reports . This may be due, first, to the fact that there was no policy or protocol for vaginal birch after Cesareaii section (VBAC). It is known that VBAC policy can reduce the rate of Cesarean delivery. In this study, one-fourth of PCOS women and 40% of the controls had had a previous Cesarean delivery. This finding was consistent with a previous survey24. second, the high rate of Cesarean delivery might be due to the attitude, as in some individuals, that Cesarean delivery is better for the baby25. In addition, the difference in Cesarean section rates between the PCOS group and the control group could not be explained by the maternal and fetal complications. Instead, infertility may be the factor that, in part, explains this difference.
In conclusion, it was shown that the prevalence of GDM, HDP and premature delivery was higher in Asian women with PCOS than in the control women. PCOS was a risk factor for GDM and HDP, but not a potential risk factor for premature birth. However, it cannot be concluded that PCOS poses a risk on neonatal outcome.
1. Frank S. Polycystic ovary syndrome. N Engl J Med 1995;333:853- 61
2. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanisms and implications for pathogenesis. Endocr Rev 1997; 18:774-800
3. Weyer C, Tataranni PA, Bogardus C, et al. Insulin resistance and insulin secretory dysfunction are independent predictors of worsening of glucose tolerance during each stage of type 2 diabetes development. Diabetes Care 2001;24:89-94
4. Legro RS, Kunselman AR, Dodson WC, et al. Prevalence and predictors of risk for type 2 diabetes mellitus and impaired glucose tolerance in polycystic ovary syndrome: a prospective, controlled study in 254 affected women. J Clin Endomnol Metab 1999; 84:165-9
5. Ehrmann DA, Barnes RB, Rosenfield RL, et al. Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care 1999;22:14l-6
6. Weerakiet S, Srisombut C, Bunnag P, et al Prevalence of type 2 DM and impaired glucose tolerance in Asian women with polycystic ovary syndrome, hit J Gynecol Obstet 2001 ;75:177-84
7. Knopp RH, Montes A, Childs M, et al. Metabolic adjustments in normal and diabetic pregnancy. Clin Obstet Gynecol 1981;24:21-49
8. Urman B, Sarac E, Dogan L, et al. Pregnancy in infertile PCOD patients: complications and outcome. J Reprod Med 1997;42:501-5
9. Radon PA, McMahon MJ, Meyer WR. Impaired glucose tolerance in pregnant women with polycystic ovary syndrome. Obstet Gynecol 1999;94: 194-7
10. Mikola M, Hiilesmaa V, Halttunen H, et al. Obstetric outcome in women with polycystic ovarian syndrome. Hum Reprod 2001; 16:226- 9
11. Wortsman J, de Angeles S, Futterweit W, et al. Gestational diabetes and neonatal macrasomia in the poly cystic ovary syndrome. J Rcp rod Med 1991;36:659-61
12. Turhan NO, seckin NC, Aybar F, et al. Assessment of glucose tolerance and pregnancy outcome of polycystic ovary patients, Int J Gynecol Obstet 2003;81:163-8
13. Roberts R. Hypertension in women with gestational diabetes. Diabetes Care 1998;21 (Suppl 2):B27-1332
14. Gjonnaess H. The course and outcome of pregnancy after ovarian electrocautery in women with polycystic ovarian syndrome: the influence of bodyweight. Br J Obstet Gynaecol 1989;96:714-19
15. Wijeyaratne CN, Balen AH, Barth JH, et al. Clinical manifestations and insulin resistance (IR) in polycystic ovary syndrome (PCOS) among South Asians and Caucasians: is there a difference? CHn Endocrinol (Oxf) 2002;57:343-50
16. Homburg S. Polycystic ovary syndrome from gynaecological curiosity to multisystem endocrinopathy. Hum Reprod 1996; 11:29-39
17. Adams J, Poison DW, Franks S. Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br Med J 1986;293:355-9
18. American Diabetes Association. Gestational diabetes mellitus. Diabetes Care 2001;24(Suppl 1):s77-s79
19. National high blood pressure education program: working group report on high blood pressure in pregnancy. Am J Obstet Gynecol 2000;183:sl-s22
20. Creasy RK, lams JD. Preterm labor and delivery. In: Creasy RK, Resnik R, eds. Maternal- Fetal Medicine, 4th edn. Philadelphia: WB Saunders, 1999:498-531
21. Casey BM, Mclntire DD, Leveno KJ. The continuing value of the Apgar score for the assessment of newborn infants. N Engl J Med 2001 ;344:467-71
22. Hamasaki T, Yasuhi I, Hirai M, et al. Hyperinsulinemia increases the risk of gestational hypertension. Int J Gynaecol Obstet 1996;55:141-5
23. Suhonen L, Teranio K. Hypertension and preeclampsia in women with gestational glucose intolerance. Acta Obstet Gynecol Scand 1993;72: 269-72
24. Chanrachakul B, Herabutya Y, Udomsubpayakul U. Epidemic of cesarean section at the general, private and university hospitals in Thailand. J Obstet Gynaecol Res 2000;26:357-61
25. YangX, Hsu-HageB, ZhangH, tal. Women with impaired glucose tolerance during pregnancy have significantly poor pregnancy outcomes. Diabetes Care 2002;25:1619-24
S. Weerakiet, C. Srisombut, A. Rojanasakul, P. Panburana, A. Thakkinstian* and Y. Herabutya
Department of Obstetrics and Gynecology; *Clinical Epidemiology Unit; Faculty of Medicine, Ramathibodi Hospital, Mahidol U\niversity, Bangkok, Thailand
Correspondence: Dr S. Weerakiet, Department of Obstetrics and Gynecology, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand 10400
Copyright CRC Press Sep 2004