- Email: [email protected]
Polycystic ovary syndrome in type 2 diabetes: does it predict a more severe phenotype?
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY
Polycystic ovary syndrome in type 2 diabetes: does it predict a more severe phenotype? Stephanie Y. T. Sim, M.B.B.S.,a Sian L. Chin, M.B.B.S.,a Jocelyn L. K. Tan, B.Sc., M.Sc.,b Suzanne J. Brown, B.Sc.,c Andrea J. Cussons, M.B.B.S., Ph.D.,c and Bronwyn G. A. Stuckey, M.B.B.S.a,b,c a c
School of Medicine and Pharmacology, University of Western Australia; b Keogh Institute for Medical Research; and Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands, Western Australia, Australia
Objective: To examine the prevalence of a history of polycystic ovary syndrome (PCOS) in women with type 2 diabetes (DM2) and to compare metabolic and reproductive outcomes between women with and without PCOS. Design: Cross-sectional study. Setting: Tertiary hospital. Patient(s): Female inpatients age 18–75 years with DM2. Intervention(s): A face-to-face questionnaire was administered. Main Outcome Measure(s): Age at diagnosis of diabetes, history of gestational diabetes, family history of diabetes, and reproductive history, fertility history, number of miscarriages, and morbidity in pregnancy. Result(s): One hundred seventy-one inpatients with DM2 participated. The prevalence of a history of PCOS was 37%. Women with PCOS had an earlier mean age of diagnosis of DM2 (44.2 vs. 48.8 years), higher recalled peak body mass index (BMI; 43.1 kg/m2 vs. 36.8 kg/m2), higher rate of gestational diabetes (28% vs. 18%), and higher rate of hypertension in pregnancy (40% vs. 22%). Women with PCOS were less likely to have a family history of DM2 than those without PCOS (45% vs. 67%). Conclusion(s): A history of PCOS in women with DM2 is associated with earlier onset of DM2, higher BMI, and a more severe phenotype. Since PCOS subjects were less likely to have a family history of DM2, lack of a family history of DM2 in women with PCOS is not reassuring for DM2 risk. We recommend identifying PCOS in early life and intervening to reduce the risk of diabetes and its comorbidities and suboptimal reproductive outcomes. (Fertil SterilÒ 2016;-:-–-. Ó2016 by American Society for Reproductive Medicine.) Key Words: PCOS, type 2 diabetes, gestational diabetes, cardiovascular disease Discuss: You can discuss this article with its authors and with other ASRM members at
P
olycystic ovarian syndrome (PCOS) and type 2 diabetes mellitus (DM2) are both associated with insulin resistance. In Australia the prevalence of DM2 in the population is 3.8%, increasing from 0.1% of those aged 0–34 years to 14.7% of 65– 69 year olds, while PCOS affects approximately 6%–8% of women of reproductive age (1, 2). The pathogenesis of DM2 involves a combination of insulin
resistance and relative impairment of insulin secretion leading to hyperglycemia and is often accompanied by hypertension and dyslipidemia, commonly referred to as the metabolic syndrome. PCOS is a heterogeneous disorder characterized by ovulatory dysfunction and hyperandrogenism. While the pathogenesis of PCOS is still not completely understood, insulin resistance is present in 65%–80% of
Received April 3, 2016; revised and accepted June 28, 2016. S.Y.T.S. has nothing to disclose. S.L.C. has nothing to disclose. J.L.K.T. has nothing to disclose. S.J.B. has nothing to disclose. A.J.C. has nothing to disclose. B.G.A.S. has nothing to disclose. S.Y.T.S. and S.L.C. should be considered similar in author order. Reprint requests: Professor Bronwyn G. A. Stuckey, M.B.B.S., Keogh Institute for Medical Research, 1st floor, C Block Sir Charles Gairdner Hospital, Nedlands, Western Australia 6009, Australia (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2016 0015-0282/$36.00 Crown Copyright ©2016 Published by Elsevier Inc. on behalf of the American Society for Reproductive Medicine http://dx.doi.org/10.1016/j.fertnstert.2016.06.040 VOL. - NO. - / - 2016
women with PCOS and plays a significant role in its etiology (3). The prevalence of impaired glucose tolerance and diabetes in women with PCOS has been widely studied and reviewed (4). However, the converse, that is, the prevalence of PCOS in women with diabetes, has received less attention. Those studies that have examined the prevalence of PCOS in women with DM2 have all confined their study population to ambulatory premenopausal women (5–10). Since age is a risk factor for developing DM2, with the average age at diagnosis being 46–52, and since diabetes is associated, with time, with significant comorbidity, we have studied women with DM2 in a wider age group, including postmenopausal women and in the inpatient setting (11). 1
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY Our study therefore aimed to compare the history and metabolic profile of women with DM2, with and without a history consistent with PCOS, to examine whether there was a difference in phenotypic features. For this study we surveyed women with DM2 who were inpatients in two tertiary hospitals.
MATERIALS AND METHODS Subjects
This was a cross-sectional study of women with DM2 who were inpatients. Women between the ages of 18 and 75 years with DM2 who were admitted to Sir Charles Gairdner Hospital or Royal Perth Hospital in Perth, Australia, were invited to participate. Pregnant women and women with type 1 diabetes were excluded from the study. The study was approved by the Human Research Ethics Committees of both hospitals (EC 2012/197), all participants gave informed written consent, and the study was conducted in the first 6 months of 2013.
Questionnaire A questionnaire was administered in person by two researchers (S.Y.T.S. and S.L.C.) during the hospital admission. In the questionnaire, women were asked about age at onset of diabetes, history of gestational diabetes, family history of diabetes, previous diagnosis of PCOS, menstrual history, use of the oral contraceptive pill, number of pregnancies, time to conception, previous hysterectomy, history of hirsutism during the reproductive years as selfrated with a Ferriman-Gallwey score (12), current body weight, and recalled highest and lowest body weight in adulthood. Body mass index (BMI) was calculated as weight divided by the square of the height in meters (kg/m2). A sample of 20 responders was contacted after hospital discharge by two other researchers (A.J.C. and B.G.A.S.) to check reproducibility of answers to all the questions on the questionnaire. Hospital electronic records for all participants were checked by two researchers (B.G.A.S. and J.L.K.T.).
Statistical Analysis Participants were classified as having a history of PCOS if they had either a previous diagnosis of PCOS or a history of irregular menses and a self-rated Ferriman-Gallwey score of 8 or more during reproductive years. Subjects were classified as PCOS or non-PCOS for analysis of reproductive and metabolic parameters. Power calculation based on the expected outcome of gestational diabetes predicted a sample size of 159 would be sufficient to detect a medium effect size (in this case 0.45) with 80% power and at a 5% significance level. Demographic variables for the whole cohort and for the PCOS and non-PCOS subgroups are summarized with mean and standard deviation (SD) for continuous variables, while count and percentage are given for categorical variables (Table 1). Means in PCOS and non-PCOS groups were compared via independent samples t-tests, or, in nonnormal data, a nonparametric Mann-Whitney U-test. We tested for association between PCOS and categorical variables with a c2-test. Where category numbers were too low (<5), Fisher's exact test was used. Multiple logistic regression analysis was performed to evaluate the significance of explanatory variables for PCOS and reproductive and metabolic outcomes.
RESULTS One hundred seventy-one women were included in the study. The participants were ages 23–75 years, with a mean age of 62 years. The mean BMI within the group was 34.0 kg/m2. Age at diagnosis of diabetes ranged from 14 to 73 years, with a mean age of 47 years. Age of menarche was between 8 and 17 years, with a mean age of 12.8 years. Of the women surveyed, 153 had had one or more pregnancies. In this study, 64 (37%) women met the criteria for hyperandrogenism and ovulatory dysfunction and were classified as having a history consistent with PCOS. Of these 64, 21 women had been previously diagnosed as having PCOS. There was a significant age difference between those who reported a previous diagnosis—53 years (14)—and those who did not—63 years (9) (P¼ .001). Table 1 compares the metabolic profiles of the PCOS and the non-PCOS groups. PCOS was associated with an earlier age of diagnosis of DM2 (44.2 vs.
TABLE 1 Demographics and family history of the participants. Demographics Subjects, n (%) Weight, kg Age DM2 diagnosed, y BMI, kg/m2 Highest recalled BMI, kg/m2 Lowest recalled BMI, kg/m2 Age at menarche, y Relative DM2, n (%) Father DM2, n (%) Mother DM2, n (%) Mother or father DM2, n (%)
All
PCOS
Non-PCOS
PCOS vs. non-PCOS P value
171 87.7 (27.2) 47.1 (13.7) 34.0 (9.9) 39.2 (11.8) 23.9 (6.3) 12.8 (1.7) 101 (59) 26 (15) 53 (31) 73 (43)
64 (37) 90.9 (30.2) 44.2 (12.7) 35.3 (11.0) 43.1 (14.0) 24.8 (7.9) 13.0 (2.0) 29 (45) 7 (11) 14 (22) 20 (31)
107 (63) 85.8 (25.2) 48.8 (14.1) 33.3 (9.1) 36.8 (9.6) 23.3 (5.0) 12.7 (1.5) 72 (67) 19 (18) 39 (36) 53 (50)
.286 .028 .310 .003 .840 .218 .007 .460 .078 .025
Note: Values presented as mean (SD) unless stated otherwise. PCOS is defined as a participant with a history of hyperandrogenic ovulatory dysfunction. Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
2
VOL. - NO. - / - 2016
Fertility and Sterility® 48.8 years; P¼ .028) and higher recalled peak BMI (43.1 kg/m2 vs. 36.8 kg/m2; P¼ .003). Importantly, the presence of a family history of DM2 was significantly lower among PCOS subjects than among non-PCOS subjects (45 vs. 67%; P¼ .007). Similarly, a history of DM2 in both parents was less common with PCOS than with non-PCOS (31 vs. 50%; P¼ .025). The reproductive history of PCOS and non-PCOS subjects is compared in Table 2. Of the 153 women who had had a pregnancy, 33 (22%) had a history of gestational diabetes. After adjusting for family history of DM2, a history of gestational diabetes was more common in women with PCOS than in women without PCOS (28% vs. 18% P¼ .043). Likewise, a history of hypertension in pregnancy was more common with PCOS than without (40% vs. 22%; P¼ .017). Of women with PCOS, 44% had taken more than 12 months to conceive, compared with 21% of non-PCOS women (P¼ .003). Women with PCOS had experienced more miscarriages on average (1.3 vs. 0.5; P< .001), although PCOS and nonPCOS women had a similar number of live births. Previous use of oral contraceptive pill (OCP) was reported by 77% of PCOS participants versus 54% of non-PCOS participants (P¼ .0058). In both PCOS and non-PCOS women, prior use of OCP was associated with a later mean onset of DM2 (P¼ .025)—in PCOS women, 45.1 years versus 41.2 years; and in non-PCOS women, 50.4 years versus 46.9 years. Using a multivariate model, highest BMI and age of menarche were positive predictors of having PCOS, while a family history of DM2 was a negative predictor. Each unit increase in highest recalled BMI increased the odds of PCOS by 5.1% (odds ratio [OR] ¼ 1.05; 95% confidence interval [CI], 1.02–1.09). When menarche was later than 14, the odds of PCOS were greater than threefold (OR ¼ 3.78; 95% CI, 1.48–10.2), compared with women for whom age of menarche was 14 years or earlier. A family history of DM2 in at least one first-degree relative reduced the odds of PCOS (OR ¼ 0.37; 95% CI, 0.18–0.76) compared with women with no family history.
Both a family history of DM2 and PCOS increased the risk of gestational diabetes. In women classified as having PCOS, the odds of gestational diabetes were greater than twofold (OR ¼ 2.35; 95% CI, 1.03–5.45), compared with women without PCOS. With a family history of DM2, the odds of gestational diabetes were greater than threefold (OR ¼ 3.48; 95% CI, 1.41–9.63), compared with women with no family history. As the participants were hospital inpatients, the admitting diagnoses and comorbidities for the participants are shown in Table 3. Admissions for surgical procedures, for cardiovascular events, and for treatment of infection were the most common reasons for admission. The commonest comorbidity in both the PCOS and non-PCOS groups was cardiovascular disease including hypertension and renal failure. There were 12 women in the PCOS group with a history of breast cancer (19%) versus 9 in the non-PCOS group (8%; P¼ .06).
DISCUSSION Our study found, as expected, a high prevalence of history of PCOS in women with DM2, although only 32% had previously been diagnosed. This lack of prior diagnosis is entirely consistent with the data from March et al., who estimated that up to 68% of young women with PCOS go undiagnosed (2). Those women who reported a diagnosis of PCOS were younger than those who did not. This may reflect the greater awareness of the diagnosis in recent decades. Because DM2 is generally a disease of later onset, it is important to conduct a study that includes postmenopausal women. There have been previous studies of the prevalence of PCOS in women with DM2, which are summarized in Table 4 (5–10). All of these studies have been confined to premenopausal women with an average age of 38 years with the stated prevalence of PCOS in women with DM2, diagnosed by varying criteria, ranging from 4.3% to 82%. However, although Conn et al. found that 82% of women with DM2 had polycystic ovaries on ultrasound, only 26% had oligoamenorrhoea and only 32% had hirsutism (5).
TABLE 2 Reproductive history of the participants. History Subjects, n (%) Delayed conception Used ART In women with R1 pregnancy Delayed conception Used assisted reproductive technology Gestational diabetes Hypertension in pregnancy In women over 40 y Pregnancies, R1 Pregnancies, mean (SD) Miscarriages, mean (SD) Live births, mean (SD)
All
PCOS
Non-PCOS
171 48 (28) 8 (5) 153 45 (29) 7 (5)
64 (37) 26 (41) 5 (8) 57 (37) 25 (44) 5 (9)
107 (63) 22 (21) 3 (3) 96 (63) 20 (21) 2 (2)
33 (22) 44 (29) 167 151 (90) 3.7 (2.3) 0.8 (1.4) 2.9 (1.9)
16 (28) 23 (40) 61 (37) 56 (92) 4.0 (2.7) 1.3 (2.0) 2.7 (1.8)
17 (18) 21 (22) 106 (63) 95 (90) 3.5 (2.1) 0.5 (0.8) 3.1 (1.9)
PCOS vs. non-PCOS P value .008 .152 .003 .103 .156 .017 .788 .382 .0002 .208
Note: Values presented as n (%), unless stated otherwise. Delayed conception is defined as ever having taken more than 12 months to conceive while actively trying for a pregnancy. Data in parentheses are percentages. Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
VOL. - NO. - / - 2016
3
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY
TABLE 3 Admitting diagnosis and the comorbidities for the participants in the study, expressed as percentage of the entire group and of the PCOS and nonPCOS group. Admitting diagnosis Diagnoses Surgical admission Cardiovascular disease Cardiovascular event Hypertension Peripheral vascular disease Peripheral vascular Diabetic ulcer Amputation Renal disease Renal failure Dialysis Infection Urosepsis Other infection Thrombosis: VTE Neurological Stroke Peripheral neuropathy Glucose control Hyperglycemia Hypoglycemia Cancer Breast cancer Endometrial cancer Other Fracture Medical other Respiratory Diagnosis not recorded
Comorbidities
Whole group
PCOS
Non-PCOS
Whole group
PCOS
Non-PCOS
18
20
16
–
–
–
16 1
16 0
17 1
43 53
38 48
46 56
3 0 1
2 0 0
4 0 1
9 4 4
9 2 2
8 5 6
1 1
0 0
2 2
22 5
14 0
26 8
4 19 1
6 16 2
2 21 1
6 25 8
11 23 6
4 25 8
4 1
2 0
5 2
9 3
8 2
10 4
2 1
3 2
2 1
3 1
3 2
3 1
4 1 8 2 8 3 1
9 2 8 2 9 2 0
0 1 9 3 7 4 2
12 1 15 3 8 9 –
19 2 13 3 9 11 –
8 1 16 3 8 8 –
Note: Data presented as percentages. All admitting diagnoses are included in the co-morbidities table except for surgical admissions. VTE ¼ venous thromboembolic disease, including venous thrombosis and pulmonary embolus. Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
Four studies have used National Institutes of Health (NIH) criteria to classify PCOS (6, 8–10). There was a wide range of age of onset of diabetes in our study (14–73 years). In those studies that have reported it, the age of onset of diabetes is earlier in women with PCOS compared with those without (6, 9). Our study found a similar significant difference, although, because of the inclusion of postmenopausal women, the mean age of onset in our study was older. This implies that insulin resistance is manifest at an earlier age or is more marked in women with PCOS than in those without. This has clinical significance since a longer duration of DM2 is a significant predictor of cardiovascular risk and is associated with macrovascular and microvascular complications (13–15). An important, and surprising, finding in our study is that a family history of diabetes is less common in diabetic women with PCOS than in those without (45% vs. 67%; P¼ .007). A similar but nonsignificant trend was found in the only other study that examined this (10). Earlier studies have identified a higher prevalence of a positive family history of diabetes in PCOS compared with population controls (16–18). However, later genetic studies have failed to find an association between genes known to influence the heritability of DM2 and the presence of PCOS (19–21). Together these latter 4
studies imply that the development of DM2 with and without a prior history of PCOS is under different genetic control. Our study also implies that a lack of family history of DM2 in women with PCOS is not reassuring with regard to diabetes risk. Obesity is associated with both PCOS and DM2 and is associated with poorer reproductive outcomes and metabolic profile both in the general population and in women with PCOS. We compared body mass in two ways—from measured weight on admission and from recalled highest body weight. The measured BMI on admission was not statistically different between women with a history of PCOS and those without (35.3 vs. 33.3). However a recalled highest adult body weight was higher in women with PCOS (BMI, 43.1 vs. 36.8; P¼ .003). It might be considered that recall is inaccurate, but these results are consistent with most previous studies examining premenopausal women where the BMI in women with PCOS and DM2 was generally higher than in those with nonPCOS DM2 (5, 6, 9, 10). It is apparent from Table 1 that the BMI for both groups on admission was lower than the recalled peak BMI. When considering the reasons for this, one must take in account the comorbidities with which these patients were admitted. Prior OCP use was higher in women with PCOS than without, often reportedly for correction of menstrual VOL. - NO. - / - 2016
NA 0.63 vs. 1.2 NA 1.2 vs. 0.5 NA NA 1.3 vs. 0.5c 55 vs. 29 NA NA NA NA 33 vs. 21 28 vs. 18
Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
2000 2001 2005 2006 2008 2008 2014 Conn et al. Peppard et al. Zargar et al. Kelestimur et al. Amini et al. Mirzaei and Kazemi Sim and Chin
VOL. - NO. - / - 2016
Note: Age is reported as mean SD, except for Conn et al., where it is reported as median and range. NA signifies that the clinical outcome was not reported in the study publication. a Polycystic ultrasound appearances only. b P< .05. c P< .005.
NA NA NA NA NA 44 vs. 46 45 vs. 67b NA 26 vs. 29.4 NA NA 29.3 vs. 33.7 NA 44.2 vs. 48.8b NA 11.4 vs. 12.4 13.5 vs. 13.5 NA 13.5 vs. 13.2 NA 13.0 vs. 12.7 31.4 vs. 30.9 40.3 vs. 34.8 25.4 vs. 25.7 NA 32.8 vs. 30.0b 28.2 vs. 26.1b 35.3 vs. 33.3 82 27 37 4.3 8.3 19.5 37.0 US NIH NIH þ US NIH NIH NIH NIH 41 (24–52) 35.3 5.6 36.2 4.4 38.9 0.5 39.3 19.2 38.8 5.9 62.0 10.1
BMI, kg/m PCOS, % Criteria Age, y n Year Author
38 30 125 92 157 92 171
GDM, % Family DM2, % Age at DM2, y Age at menarche, y
PCOS vs. non-PCOS
2
Comparison between this and previously published studies.
TABLE 4
a
Miscarriage
Fertility and Sterility® irregularity. Despite common concerns about an adverse influence of OCP use on insulin resistance, the use of OCP in this study was found to be associated with a significantly later onset of DM2 in both PCOS and non-PCOS women. A recent Cochrane meta-analysis has shown no deleterious effect on carbohydrate metabolism with steroid contraception (22). Further, our data suggest that the effect of the OCP may be similar to the reduction in the incidence of DM2 seen with exogenous estrogen and progestin in postmenopausal women in randomized controlled trials (23). Gestational diabetes mellitus (GDM) and preeclampsia are both established risk factors for DM2 and also associated with obesity (24, 25). Therefore, it is not surprising that a history of these two disorders was reported at a much higher rate in our study compared with background population figures (26, 27). In addition, the presence of PCOS further increased the risk of GDM and preeclampsia. A recent epidemiological study reported a prevalence of gestational diabetes of 9.6% using the Australasian criteria (28), while in our study the prevalence of a history of gestational diabetes in PCOS and non-PCOS women was 28% and 18%, respectively. Mirzaei and Kazemi found a similarly increased history of gestational diabetes in women with DM2 and a further increase in PCOS women compared with in non-PCOS women (10). In the wider population, Bjercke et al. found that women with PCOS have a higher prevalence of GDM and preeclampsia than controls, which is consistent with preconception insulin resistance playing a role in these two disorders (29). Gestational diabetes is associated with an increased risk of premature rupture of membranes, cesarean section, and preterm delivery (27). It also increases the risk of macrosomia in the infant and subsequently the development of adolescent obesity and DM2 (27, 30). The preconception presence of PCOS should alert the physician to these possible consequences. Additional findings of the current study included higher rates of miscarriage and delayed conception in PCOS subjects, which may be expected in a group defined by reproductive dysfunction. The admitting diagnoses and comorbidities were examined for all participants (Table 3). Consistent with the diagnosis of DM2, cardiovascular disease and renal disease are the most common comorbidities. There is a trend to excess cases of breast cancer in the PCOS group, which does not reach statistical significance. If this is real, possible contributors are higher BMI, greater hyperinsulinemia, and probable later parity. This study is the largest survey of PCOS conducted specifically in women with DM2. It is the first one to include a wide age range that included postmenopausal women and the first to be conducted in an inpatient population. We have relied on a recalled history of menstrual disturbance and hirsutism for the diagnosis of PCOS. As with any questionnaire-based study, our data relied predominantly on the participants' ability to recall and therefore may have been subject to bias and inaccuracies. However, we believe that most women can recall age at menarche and reproductive history, and we did find this to be the case in our quality control check. Moreover, any effect of recall would have been seen in both PCOS and non-PCOS women. Because our study included mostly 5
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY postmenopausal women, ovarian biochemistry and ultrasound would have been unhelpful. We have verified comorbidities from hospital records for all patients. It is possible that some comorbidities relevant to this study were not recorded. Finally, our study was conducted in an inpatient population and our findings will not be the same as in the ambulatory population.
Conclusion In this study, a history consistent with PCOS in the setting of DM2 was associated with a more severe phenotype—earlier onset of DM2, higher adult BMI, and higher rates of gestational diabetes and preeclampsia. Prior use of OCPs was associated with later onset of DM2 in both PCOS and non-PCOS women. In this group of DM2 patients, a family history of DM2 was less common in women with PCOS than in those without. If it is reasonable to assume that nondiabetic PCOS patients are unlikely to have a greater prevalence of family history of DM2 than diabetic PCOS patients, then this result may suggest that relying on family history to identify DM2 risk among PCOS patients is not valid and a lack of family history of DM2 is not reassuring. This study demonstrates that within a population of women with DM2, a history of PCOS is associated with a more unfavorable phenotype. It also demonstrates the accumulation of comorbidities in women with DM2, both with and without PCOS. Early identification of women with PCOS should allow earlier intervention and modification of lifestyle factors and the use of targeted medical therapies to improve long-term prognosis from both the cardiovascular and reproductive perspectives. Acknowledgments: The authors thank the women who participated in this survey; and Gloria Daniels, Clinical Nurse Consultant, for her assistance with recruitment.
REFERENCES 1.
2.
3.
4.
5. 6.
7.
6
Australian Institute of Health and Welfare. Diabetes prevalence in Australia: detailed estimates for 2007–08. In: Diabetes series no. 17, Cat. No. CVD 56. Canberra: AIHW; 2011. March WA, Moore VM, Willson KJ, Phillips DIW, Norman RJ, Davies MJ. The prevalence of polycystic ovary syndrome in a community sample assessed under contrasting diagnostic criteria. Hum Reprod 2010;25:544–51. DeUgarte CM, Bartolucci AA, Azziz R. Prevalence of insulin resistance in the polycystic ovary syndrome using the homeostasis model assessment. Fertil Steril 2005;83:1454–60. Moran LJ, Misso ML, Wild RA, Norman RJ. Impaired glucose tolerance, type 2 diabetes and metabolic syndrome in polycystic ovary syndrome: a systematic review and meta-analysis. Hum Reprod Update 2010;16:347–63. Conn JJ, Jacobs HS, Conway GS. The prevalence of polycystic ovaries in women with type 2 diabetes mellitus. Clin Endocrinol (Oxf) 2000;52:81–6. Peppard HR, Marfori J, Iuorno MJ, Nestler JE. Prevalence of polycystic ovary syndrome among premenopausal women with type 2 diabetes. Diabetes Care 2001;24:1050–2. Zargar AH, Gupta VK, Wani AI, Masoodi SR, Bashir MI, Laway BA, et al. Prevalence of ultrasonography proved polycystic ovaries in North Indian women with type 2 diabetes mellitus. Reprod Biol Endocrinol 2005;3:35.
8.
Kelestimur F, Unluhizarci K, Baybuga H, Atmaca H, Bayram F, Sahin Y. Prevalence of polycystic ovarian changes and polycystic ovary syndrome in premenopausal women with treated type 2 diabetes mellitus. Fertil Steril 2006;86:405–10. 9. Amini M, Horri N, Farmani M, Haghighi S, Sattari G, Pornaghshband Z, et al. Prevalence of polycystic ovary syndrome in reproductive-aged women with type 2 diabetes. Gynecol Endocrinol 2008;24:423–7. 10. Mirzaei F, Kazemi N. Prevalence of polycystic ovary syndrome in women with type 2 diabetes in Kerman, Iran. Metab Syndr Relat Disord 2008;6:215–7. 11. Koopman RJ, Mainous AG 3rd, Diaz VA, Geesey ME. Changes in age at diagnosis of type 2 diabetes mellitus in the United States, 1988 to 2000. Ann Fam Med 2005;3:60–3. 12. Yildiz BO, Bolour S, Woods K, Moore A, Azziz R. Visually scoring hirsutism. Hum Reprod Update 2010;16:51–64. 13. Hillier TA, Pedula KL. Complications in young adults with early-onset type 2 diabetes: losing the relative protection of youth. Diabetes Care 2003;26: 2999–3005. 14. Quah JHM, Liu YP, Luo N, How CH, Tay EG. Younger adult type 2 diabetic patients have poorer glycaemic control: a cross-sectional study in a primary care setting in Singapore. BMC Endocr Disord 2013;13:18. 15. Song SH, Hardisty CA. Early-onset type 2 diabetes mellitus: an increasing phenomenon of elevated cardiovascular risk. Expert Rev Cardiovasc Ther 2008;6:315–22. 16. Fox R. Prevalence of a positive family history of type 2 diabetes in women with polycystic ovarian disease. Gynecol Endocrinol 1999;13:390–3. 17. Sir-Petermann T, Angel B, Maliqueo M, Carvajal F, Santos JL, Perez-Bravo F. Prevalence of type II diabetes mellitus and insulin resistance in parents of women with polycystic ovary syndrome. Diabetologia 2002;45:959–64. 18. Lerchbaum E, Schwetz V, Giuliani A, Obermayer-Pietsch B. Influence of a positive family history of both type 2 diabetes and PCOS on metabolic and endocrine parameters in a large cohort of PCOS women. Eur J Endocrinol 2014;170:727–39. 19. Mendoza N. Common genetic aspects between polycystic ovary syndrome and diabetes mellitus. Curr Diabetes Rev 2011;7:377–91. 20. Ewens KG, Jones MR, Ankener W, Stewart DR, Urbanek M, Dunaif A, et al. Type 2 diabetes susceptibility single-nucleotide polymorphisms are not associated with polycystic ovary syndrome. Fertil Steril 2011;95:2538–41. e1–6. 21. Saxena R, Welt CK. Polycystic ovary syndrome is not associated with genetic variants that mark risk of type 2 diabetes. Acta Diabetol 2013;50:451–7. 22. Lopez LM, Grimes DA, Schulz KF. Steroidal contraceptives: effect on carbohydrate metabolism in women without diabetes mellitus. Cochrane Database Syst Rev 2014:CD006133. 23. Margolis KL, Bonds DE, Rodabough RJ, Tinker L, Phillips LS, Allen C, et al. Effect of oestrogen plus progestin on the incidence of diabetes in postmenopausal women: results from the Women's Health Initiative Hormone Trial. Diabetologia 2004;47:1175–87. 24. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care 2002;25:1862–8. 25. Lykke JA, Langhoff-Roos J, Sibai BM, Funai EF, Triche EW, Paidas MJ. Hypertensive pregnancy disorders and subsequent cardiovascular morbidity and type 2 diabetes mellitus in the mother. Hypertension 2009;53:944–51. 26. Steegers EAP, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. Lancet 2010;376:631–44. 27. Xiong X, Saunders LD, Wang FL, Demianczuk NN. Gestational diabetes mellitus: prevalence, risk factors, maternal and infant outcomes. Int J Gynaecol Obstet 2001;75:221–8. 28. Moses RG, Morris GJ, Petocz P, San Gil F, Garg D. The impact of potential new diagnostic criteria on the prevalence of gestational diabetes mellitus in Australia. Med J Aust 2011;194:338–40. 29. Bjercke S, Dale PO, Tanbo T, Storeng R, Ertzeid G, Abyholm T. Impact of insulin resistance on pregnancy complications and outcome in women with polycystic ovary syndrome. Gynecol Obstet Invest 2002;54:94–8. 30. Catalano PM, Kirwan JP, Haugel-de Mouzon S, King J. Gestational diabetes and insulin resistance: role in short- and long-term implications for mother and fetus. J Nutr 2003;133:1674S–83S.
VOL. - NO. - / - 2016
Polycystic ovary syndrome in type 2 diabetes: does it predict a more severe phenotype? Stephanie Y. T. Sim, M.B.B.S.,a Sian L. Chin, M.B.B.S.,a Jocelyn L. K. Tan, B.Sc., M.Sc.,b Suzanne J. Brown, B.Sc.,c Andrea J. Cussons, M.B.B.S., Ph.D.,c and Bronwyn G. A. Stuckey, M.B.B.S.a,b,c a c
School of Medicine and Pharmacology, University of Western Australia; b Keogh Institute for Medical Research; and Department of Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Nedlands, Western Australia, Australia
Objective: To examine the prevalence of a history of polycystic ovary syndrome (PCOS) in women with type 2 diabetes (DM2) and to compare metabolic and reproductive outcomes between women with and without PCOS. Design: Cross-sectional study. Setting: Tertiary hospital. Patient(s): Female inpatients age 18–75 years with DM2. Intervention(s): A face-to-face questionnaire was administered. Main Outcome Measure(s): Age at diagnosis of diabetes, history of gestational diabetes, family history of diabetes, and reproductive history, fertility history, number of miscarriages, and morbidity in pregnancy. Result(s): One hundred seventy-one inpatients with DM2 participated. The prevalence of a history of PCOS was 37%. Women with PCOS had an earlier mean age of diagnosis of DM2 (44.2 vs. 48.8 years), higher recalled peak body mass index (BMI; 43.1 kg/m2 vs. 36.8 kg/m2), higher rate of gestational diabetes (28% vs. 18%), and higher rate of hypertension in pregnancy (40% vs. 22%). Women with PCOS were less likely to have a family history of DM2 than those without PCOS (45% vs. 67%). Conclusion(s): A history of PCOS in women with DM2 is associated with earlier onset of DM2, higher BMI, and a more severe phenotype. Since PCOS subjects were less likely to have a family history of DM2, lack of a family history of DM2 in women with PCOS is not reassuring for DM2 risk. We recommend identifying PCOS in early life and intervening to reduce the risk of diabetes and its comorbidities and suboptimal reproductive outcomes. (Fertil SterilÒ 2016;-:-–-. Ó2016 by American Society for Reproductive Medicine.) Key Words: PCOS, type 2 diabetes, gestational diabetes, cardiovascular disease Discuss: You can discuss this article with its authors and with other ASRM members at
P
olycystic ovarian syndrome (PCOS) and type 2 diabetes mellitus (DM2) are both associated with insulin resistance. In Australia the prevalence of DM2 in the population is 3.8%, increasing from 0.1% of those aged 0–34 years to 14.7% of 65– 69 year olds, while PCOS affects approximately 6%–8% of women of reproductive age (1, 2). The pathogenesis of DM2 involves a combination of insulin
resistance and relative impairment of insulin secretion leading to hyperglycemia and is often accompanied by hypertension and dyslipidemia, commonly referred to as the metabolic syndrome. PCOS is a heterogeneous disorder characterized by ovulatory dysfunction and hyperandrogenism. While the pathogenesis of PCOS is still not completely understood, insulin resistance is present in 65%–80% of
Received April 3, 2016; revised and accepted June 28, 2016. S.Y.T.S. has nothing to disclose. S.L.C. has nothing to disclose. J.L.K.T. has nothing to disclose. S.J.B. has nothing to disclose. A.J.C. has nothing to disclose. B.G.A.S. has nothing to disclose. S.Y.T.S. and S.L.C. should be considered similar in author order. Reprint requests: Professor Bronwyn G. A. Stuckey, M.B.B.S., Keogh Institute for Medical Research, 1st floor, C Block Sir Charles Gairdner Hospital, Nedlands, Western Australia 6009, Australia (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2016 0015-0282/$36.00 Crown Copyright ©2016 Published by Elsevier Inc. on behalf of the American Society for Reproductive Medicine http://dx.doi.org/10.1016/j.fertnstert.2016.06.040 VOL. - NO. - / - 2016
women with PCOS and plays a significant role in its etiology (3). The prevalence of impaired glucose tolerance and diabetes in women with PCOS has been widely studied and reviewed (4). However, the converse, that is, the prevalence of PCOS in women with diabetes, has received less attention. Those studies that have examined the prevalence of PCOS in women with DM2 have all confined their study population to ambulatory premenopausal women (5–10). Since age is a risk factor for developing DM2, with the average age at diagnosis being 46–52, and since diabetes is associated, with time, with significant comorbidity, we have studied women with DM2 in a wider age group, including postmenopausal women and in the inpatient setting (11). 1
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY Our study therefore aimed to compare the history and metabolic profile of women with DM2, with and without a history consistent with PCOS, to examine whether there was a difference in phenotypic features. For this study we surveyed women with DM2 who were inpatients in two tertiary hospitals.
MATERIALS AND METHODS Subjects
This was a cross-sectional study of women with DM2 who were inpatients. Women between the ages of 18 and 75 years with DM2 who were admitted to Sir Charles Gairdner Hospital or Royal Perth Hospital in Perth, Australia, were invited to participate. Pregnant women and women with type 1 diabetes were excluded from the study. The study was approved by the Human Research Ethics Committees of both hospitals (EC 2012/197), all participants gave informed written consent, and the study was conducted in the first 6 months of 2013.
Questionnaire A questionnaire was administered in person by two researchers (S.Y.T.S. and S.L.C.) during the hospital admission. In the questionnaire, women were asked about age at onset of diabetes, history of gestational diabetes, family history of diabetes, previous diagnosis of PCOS, menstrual history, use of the oral contraceptive pill, number of pregnancies, time to conception, previous hysterectomy, history of hirsutism during the reproductive years as selfrated with a Ferriman-Gallwey score (12), current body weight, and recalled highest and lowest body weight in adulthood. Body mass index (BMI) was calculated as weight divided by the square of the height in meters (kg/m2). A sample of 20 responders was contacted after hospital discharge by two other researchers (A.J.C. and B.G.A.S.) to check reproducibility of answers to all the questions on the questionnaire. Hospital electronic records for all participants were checked by two researchers (B.G.A.S. and J.L.K.T.).
Statistical Analysis Participants were classified as having a history of PCOS if they had either a previous diagnosis of PCOS or a history of irregular menses and a self-rated Ferriman-Gallwey score of 8 or more during reproductive years. Subjects were classified as PCOS or non-PCOS for analysis of reproductive and metabolic parameters. Power calculation based on the expected outcome of gestational diabetes predicted a sample size of 159 would be sufficient to detect a medium effect size (in this case 0.45) with 80% power and at a 5% significance level. Demographic variables for the whole cohort and for the PCOS and non-PCOS subgroups are summarized with mean and standard deviation (SD) for continuous variables, while count and percentage are given for categorical variables (Table 1). Means in PCOS and non-PCOS groups were compared via independent samples t-tests, or, in nonnormal data, a nonparametric Mann-Whitney U-test. We tested for association between PCOS and categorical variables with a c2-test. Where category numbers were too low (<5), Fisher's exact test was used. Multiple logistic regression analysis was performed to evaluate the significance of explanatory variables for PCOS and reproductive and metabolic outcomes.
RESULTS One hundred seventy-one women were included in the study. The participants were ages 23–75 years, with a mean age of 62 years. The mean BMI within the group was 34.0 kg/m2. Age at diagnosis of diabetes ranged from 14 to 73 years, with a mean age of 47 years. Age of menarche was between 8 and 17 years, with a mean age of 12.8 years. Of the women surveyed, 153 had had one or more pregnancies. In this study, 64 (37%) women met the criteria for hyperandrogenism and ovulatory dysfunction and were classified as having a history consistent with PCOS. Of these 64, 21 women had been previously diagnosed as having PCOS. There was a significant age difference between those who reported a previous diagnosis—53 years (14)—and those who did not—63 years (9) (P¼ .001). Table 1 compares the metabolic profiles of the PCOS and the non-PCOS groups. PCOS was associated with an earlier age of diagnosis of DM2 (44.2 vs.
TABLE 1 Demographics and family history of the participants. Demographics Subjects, n (%) Weight, kg Age DM2 diagnosed, y BMI, kg/m2 Highest recalled BMI, kg/m2 Lowest recalled BMI, kg/m2 Age at menarche, y Relative DM2, n (%) Father DM2, n (%) Mother DM2, n (%) Mother or father DM2, n (%)
All
PCOS
Non-PCOS
PCOS vs. non-PCOS P value
171 87.7 (27.2) 47.1 (13.7) 34.0 (9.9) 39.2 (11.8) 23.9 (6.3) 12.8 (1.7) 101 (59) 26 (15) 53 (31) 73 (43)
64 (37) 90.9 (30.2) 44.2 (12.7) 35.3 (11.0) 43.1 (14.0) 24.8 (7.9) 13.0 (2.0) 29 (45) 7 (11) 14 (22) 20 (31)
107 (63) 85.8 (25.2) 48.8 (14.1) 33.3 (9.1) 36.8 (9.6) 23.3 (5.0) 12.7 (1.5) 72 (67) 19 (18) 39 (36) 53 (50)
.286 .028 .310 .003 .840 .218 .007 .460 .078 .025
Note: Values presented as mean (SD) unless stated otherwise. PCOS is defined as a participant with a history of hyperandrogenic ovulatory dysfunction. Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
2
VOL. - NO. - / - 2016
Fertility and Sterility® 48.8 years; P¼ .028) and higher recalled peak BMI (43.1 kg/m2 vs. 36.8 kg/m2; P¼ .003). Importantly, the presence of a family history of DM2 was significantly lower among PCOS subjects than among non-PCOS subjects (45 vs. 67%; P¼ .007). Similarly, a history of DM2 in both parents was less common with PCOS than with non-PCOS (31 vs. 50%; P¼ .025). The reproductive history of PCOS and non-PCOS subjects is compared in Table 2. Of the 153 women who had had a pregnancy, 33 (22%) had a history of gestational diabetes. After adjusting for family history of DM2, a history of gestational diabetes was more common in women with PCOS than in women without PCOS (28% vs. 18% P¼ .043). Likewise, a history of hypertension in pregnancy was more common with PCOS than without (40% vs. 22%; P¼ .017). Of women with PCOS, 44% had taken more than 12 months to conceive, compared with 21% of non-PCOS women (P¼ .003). Women with PCOS had experienced more miscarriages on average (1.3 vs. 0.5; P< .001), although PCOS and nonPCOS women had a similar number of live births. Previous use of oral contraceptive pill (OCP) was reported by 77% of PCOS participants versus 54% of non-PCOS participants (P¼ .0058). In both PCOS and non-PCOS women, prior use of OCP was associated with a later mean onset of DM2 (P¼ .025)—in PCOS women, 45.1 years versus 41.2 years; and in non-PCOS women, 50.4 years versus 46.9 years. Using a multivariate model, highest BMI and age of menarche were positive predictors of having PCOS, while a family history of DM2 was a negative predictor. Each unit increase in highest recalled BMI increased the odds of PCOS by 5.1% (odds ratio [OR] ¼ 1.05; 95% confidence interval [CI], 1.02–1.09). When menarche was later than 14, the odds of PCOS were greater than threefold (OR ¼ 3.78; 95% CI, 1.48–10.2), compared with women for whom age of menarche was 14 years or earlier. A family history of DM2 in at least one first-degree relative reduced the odds of PCOS (OR ¼ 0.37; 95% CI, 0.18–0.76) compared with women with no family history.
Both a family history of DM2 and PCOS increased the risk of gestational diabetes. In women classified as having PCOS, the odds of gestational diabetes were greater than twofold (OR ¼ 2.35; 95% CI, 1.03–5.45), compared with women without PCOS. With a family history of DM2, the odds of gestational diabetes were greater than threefold (OR ¼ 3.48; 95% CI, 1.41–9.63), compared with women with no family history. As the participants were hospital inpatients, the admitting diagnoses and comorbidities for the participants are shown in Table 3. Admissions for surgical procedures, for cardiovascular events, and for treatment of infection were the most common reasons for admission. The commonest comorbidity in both the PCOS and non-PCOS groups was cardiovascular disease including hypertension and renal failure. There were 12 women in the PCOS group with a history of breast cancer (19%) versus 9 in the non-PCOS group (8%; P¼ .06).
DISCUSSION Our study found, as expected, a high prevalence of history of PCOS in women with DM2, although only 32% had previously been diagnosed. This lack of prior diagnosis is entirely consistent with the data from March et al., who estimated that up to 68% of young women with PCOS go undiagnosed (2). Those women who reported a diagnosis of PCOS were younger than those who did not. This may reflect the greater awareness of the diagnosis in recent decades. Because DM2 is generally a disease of later onset, it is important to conduct a study that includes postmenopausal women. There have been previous studies of the prevalence of PCOS in women with DM2, which are summarized in Table 4 (5–10). All of these studies have been confined to premenopausal women with an average age of 38 years with the stated prevalence of PCOS in women with DM2, diagnosed by varying criteria, ranging from 4.3% to 82%. However, although Conn et al. found that 82% of women with DM2 had polycystic ovaries on ultrasound, only 26% had oligoamenorrhoea and only 32% had hirsutism (5).
TABLE 2 Reproductive history of the participants. History Subjects, n (%) Delayed conception Used ART In women with R1 pregnancy Delayed conception Used assisted reproductive technology Gestational diabetes Hypertension in pregnancy In women over 40 y Pregnancies, R1 Pregnancies, mean (SD) Miscarriages, mean (SD) Live births, mean (SD)
All
PCOS
Non-PCOS
171 48 (28) 8 (5) 153 45 (29) 7 (5)
64 (37) 26 (41) 5 (8) 57 (37) 25 (44) 5 (9)
107 (63) 22 (21) 3 (3) 96 (63) 20 (21) 2 (2)
33 (22) 44 (29) 167 151 (90) 3.7 (2.3) 0.8 (1.4) 2.9 (1.9)
16 (28) 23 (40) 61 (37) 56 (92) 4.0 (2.7) 1.3 (2.0) 2.7 (1.8)
17 (18) 21 (22) 106 (63) 95 (90) 3.5 (2.1) 0.5 (0.8) 3.1 (1.9)
PCOS vs. non-PCOS P value .008 .152 .003 .103 .156 .017 .788 .382 .0002 .208
Note: Values presented as n (%), unless stated otherwise. Delayed conception is defined as ever having taken more than 12 months to conceive while actively trying for a pregnancy. Data in parentheses are percentages. Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
VOL. - NO. - / - 2016
3
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY
TABLE 3 Admitting diagnosis and the comorbidities for the participants in the study, expressed as percentage of the entire group and of the PCOS and nonPCOS group. Admitting diagnosis Diagnoses Surgical admission Cardiovascular disease Cardiovascular event Hypertension Peripheral vascular disease Peripheral vascular Diabetic ulcer Amputation Renal disease Renal failure Dialysis Infection Urosepsis Other infection Thrombosis: VTE Neurological Stroke Peripheral neuropathy Glucose control Hyperglycemia Hypoglycemia Cancer Breast cancer Endometrial cancer Other Fracture Medical other Respiratory Diagnosis not recorded
Comorbidities
Whole group
PCOS
Non-PCOS
Whole group
PCOS
Non-PCOS
18
20
16
–
–
–
16 1
16 0
17 1
43 53
38 48
46 56
3 0 1
2 0 0
4 0 1
9 4 4
9 2 2
8 5 6
1 1
0 0
2 2
22 5
14 0
26 8
4 19 1
6 16 2
2 21 1
6 25 8
11 23 6
4 25 8
4 1
2 0
5 2
9 3
8 2
10 4
2 1
3 2
2 1
3 1
3 2
3 1
4 1 8 2 8 3 1
9 2 8 2 9 2 0
0 1 9 3 7 4 2
12 1 15 3 8 9 –
19 2 13 3 9 11 –
8 1 16 3 8 8 –
Note: Data presented as percentages. All admitting diagnoses are included in the co-morbidities table except for surgical admissions. VTE ¼ venous thromboembolic disease, including venous thrombosis and pulmonary embolus. Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
Four studies have used National Institutes of Health (NIH) criteria to classify PCOS (6, 8–10). There was a wide range of age of onset of diabetes in our study (14–73 years). In those studies that have reported it, the age of onset of diabetes is earlier in women with PCOS compared with those without (6, 9). Our study found a similar significant difference, although, because of the inclusion of postmenopausal women, the mean age of onset in our study was older. This implies that insulin resistance is manifest at an earlier age or is more marked in women with PCOS than in those without. This has clinical significance since a longer duration of DM2 is a significant predictor of cardiovascular risk and is associated with macrovascular and microvascular complications (13–15). An important, and surprising, finding in our study is that a family history of diabetes is less common in diabetic women with PCOS than in those without (45% vs. 67%; P¼ .007). A similar but nonsignificant trend was found in the only other study that examined this (10). Earlier studies have identified a higher prevalence of a positive family history of diabetes in PCOS compared with population controls (16–18). However, later genetic studies have failed to find an association between genes known to influence the heritability of DM2 and the presence of PCOS (19–21). Together these latter 4
studies imply that the development of DM2 with and without a prior history of PCOS is under different genetic control. Our study also implies that a lack of family history of DM2 in women with PCOS is not reassuring with regard to diabetes risk. Obesity is associated with both PCOS and DM2 and is associated with poorer reproductive outcomes and metabolic profile both in the general population and in women with PCOS. We compared body mass in two ways—from measured weight on admission and from recalled highest body weight. The measured BMI on admission was not statistically different between women with a history of PCOS and those without (35.3 vs. 33.3). However a recalled highest adult body weight was higher in women with PCOS (BMI, 43.1 vs. 36.8; P¼ .003). It might be considered that recall is inaccurate, but these results are consistent with most previous studies examining premenopausal women where the BMI in women with PCOS and DM2 was generally higher than in those with nonPCOS DM2 (5, 6, 9, 10). It is apparent from Table 1 that the BMI for both groups on admission was lower than the recalled peak BMI. When considering the reasons for this, one must take in account the comorbidities with which these patients were admitted. Prior OCP use was higher in women with PCOS than without, often reportedly for correction of menstrual VOL. - NO. - / - 2016
NA 0.63 vs. 1.2 NA 1.2 vs. 0.5 NA NA 1.3 vs. 0.5c 55 vs. 29 NA NA NA NA 33 vs. 21 28 vs. 18
Sim. Polycystic ovary syndrome in type 2 diabetes. Fertil Steril 2016.
2000 2001 2005 2006 2008 2008 2014 Conn et al. Peppard et al. Zargar et al. Kelestimur et al. Amini et al. Mirzaei and Kazemi Sim and Chin
VOL. - NO. - / - 2016
Note: Age is reported as mean SD, except for Conn et al., where it is reported as median and range. NA signifies that the clinical outcome was not reported in the study publication. a Polycystic ultrasound appearances only. b P< .05. c P< .005.
NA NA NA NA NA 44 vs. 46 45 vs. 67b NA 26 vs. 29.4 NA NA 29.3 vs. 33.7 NA 44.2 vs. 48.8b NA 11.4 vs. 12.4 13.5 vs. 13.5 NA 13.5 vs. 13.2 NA 13.0 vs. 12.7 31.4 vs. 30.9 40.3 vs. 34.8 25.4 vs. 25.7 NA 32.8 vs. 30.0b 28.2 vs. 26.1b 35.3 vs. 33.3 82 27 37 4.3 8.3 19.5 37.0 US NIH NIH þ US NIH NIH NIH NIH 41 (24–52) 35.3 5.6 36.2 4.4 38.9 0.5 39.3 19.2 38.8 5.9 62.0 10.1
BMI, kg/m PCOS, % Criteria Age, y n Year Author
38 30 125 92 157 92 171
GDM, % Family DM2, % Age at DM2, y Age at menarche, y
PCOS vs. non-PCOS
2
Comparison between this and previously published studies.
TABLE 4
a
Miscarriage
Fertility and Sterility® irregularity. Despite common concerns about an adverse influence of OCP use on insulin resistance, the use of OCP in this study was found to be associated with a significantly later onset of DM2 in both PCOS and non-PCOS women. A recent Cochrane meta-analysis has shown no deleterious effect on carbohydrate metabolism with steroid contraception (22). Further, our data suggest that the effect of the OCP may be similar to the reduction in the incidence of DM2 seen with exogenous estrogen and progestin in postmenopausal women in randomized controlled trials (23). Gestational diabetes mellitus (GDM) and preeclampsia are both established risk factors for DM2 and also associated with obesity (24, 25). Therefore, it is not surprising that a history of these two disorders was reported at a much higher rate in our study compared with background population figures (26, 27). In addition, the presence of PCOS further increased the risk of GDM and preeclampsia. A recent epidemiological study reported a prevalence of gestational diabetes of 9.6% using the Australasian criteria (28), while in our study the prevalence of a history of gestational diabetes in PCOS and non-PCOS women was 28% and 18%, respectively. Mirzaei and Kazemi found a similarly increased history of gestational diabetes in women with DM2 and a further increase in PCOS women compared with in non-PCOS women (10). In the wider population, Bjercke et al. found that women with PCOS have a higher prevalence of GDM and preeclampsia than controls, which is consistent with preconception insulin resistance playing a role in these two disorders (29). Gestational diabetes is associated with an increased risk of premature rupture of membranes, cesarean section, and preterm delivery (27). It also increases the risk of macrosomia in the infant and subsequently the development of adolescent obesity and DM2 (27, 30). The preconception presence of PCOS should alert the physician to these possible consequences. Additional findings of the current study included higher rates of miscarriage and delayed conception in PCOS subjects, which may be expected in a group defined by reproductive dysfunction. The admitting diagnoses and comorbidities were examined for all participants (Table 3). Consistent with the diagnosis of DM2, cardiovascular disease and renal disease are the most common comorbidities. There is a trend to excess cases of breast cancer in the PCOS group, which does not reach statistical significance. If this is real, possible contributors are higher BMI, greater hyperinsulinemia, and probable later parity. This study is the largest survey of PCOS conducted specifically in women with DM2. It is the first one to include a wide age range that included postmenopausal women and the first to be conducted in an inpatient population. We have relied on a recalled history of menstrual disturbance and hirsutism for the diagnosis of PCOS. As with any questionnaire-based study, our data relied predominantly on the participants' ability to recall and therefore may have been subject to bias and inaccuracies. However, we believe that most women can recall age at menarche and reproductive history, and we did find this to be the case in our quality control check. Moreover, any effect of recall would have been seen in both PCOS and non-PCOS women. Because our study included mostly 5
ORIGINAL ARTICLE: REPRODUCTIVE ENDOCRINOLOGY postmenopausal women, ovarian biochemistry and ultrasound would have been unhelpful. We have verified comorbidities from hospital records for all patients. It is possible that some comorbidities relevant to this study were not recorded. Finally, our study was conducted in an inpatient population and our findings will not be the same as in the ambulatory population.
Conclusion In this study, a history consistent with PCOS in the setting of DM2 was associated with a more severe phenotype—earlier onset of DM2, higher adult BMI, and higher rates of gestational diabetes and preeclampsia. Prior use of OCPs was associated with later onset of DM2 in both PCOS and non-PCOS women. In this group of DM2 patients, a family history of DM2 was less common in women with PCOS than in those without. If it is reasonable to assume that nondiabetic PCOS patients are unlikely to have a greater prevalence of family history of DM2 than diabetic PCOS patients, then this result may suggest that relying on family history to identify DM2 risk among PCOS patients is not valid and a lack of family history of DM2 is not reassuring. This study demonstrates that within a population of women with DM2, a history of PCOS is associated with a more unfavorable phenotype. It also demonstrates the accumulation of comorbidities in women with DM2, both with and without PCOS. Early identification of women with PCOS should allow earlier intervention and modification of lifestyle factors and the use of targeted medical therapies to improve long-term prognosis from both the cardiovascular and reproductive perspectives. Acknowledgments: The authors thank the women who participated in this survey; and Gloria Daniels, Clinical Nurse Consultant, for her assistance with recruitment.
REFERENCES 1.
2.
3.
4.
5. 6.
7.
6
Australian Institute of Health and Welfare. Diabetes prevalence in Australia: detailed estimates for 2007–08. In: Diabetes series no. 17, Cat. No. CVD 56. Canberra: AIHW; 2011. March WA, Moore VM, Willson KJ, Phillips DIW, Norman RJ, Davies MJ. The prevalence of polycystic ovary syndrome in a community sample assessed under contrasting diagnostic criteria. Hum Reprod 2010;25:544–51. DeUgarte CM, Bartolucci AA, Azziz R. Prevalence of insulin resistance in the polycystic ovary syndrome using the homeostasis model assessment. Fertil Steril 2005;83:1454–60. Moran LJ, Misso ML, Wild RA, Norman RJ. Impaired glucose tolerance, type 2 diabetes and metabolic syndrome in polycystic ovary syndrome: a systematic review and meta-analysis. Hum Reprod Update 2010;16:347–63. Conn JJ, Jacobs HS, Conway GS. The prevalence of polycystic ovaries in women with type 2 diabetes mellitus. Clin Endocrinol (Oxf) 2000;52:81–6. Peppard HR, Marfori J, Iuorno MJ, Nestler JE. Prevalence of polycystic ovary syndrome among premenopausal women with type 2 diabetes. Diabetes Care 2001;24:1050–2. Zargar AH, Gupta VK, Wani AI, Masoodi SR, Bashir MI, Laway BA, et al. Prevalence of ultrasonography proved polycystic ovaries in North Indian women with type 2 diabetes mellitus. Reprod Biol Endocrinol 2005;3:35.
8.
Kelestimur F, Unluhizarci K, Baybuga H, Atmaca H, Bayram F, Sahin Y. Prevalence of polycystic ovarian changes and polycystic ovary syndrome in premenopausal women with treated type 2 diabetes mellitus. Fertil Steril 2006;86:405–10. 9. Amini M, Horri N, Farmani M, Haghighi S, Sattari G, Pornaghshband Z, et al. Prevalence of polycystic ovary syndrome in reproductive-aged women with type 2 diabetes. Gynecol Endocrinol 2008;24:423–7. 10. Mirzaei F, Kazemi N. Prevalence of polycystic ovary syndrome in women with type 2 diabetes in Kerman, Iran. Metab Syndr Relat Disord 2008;6:215–7. 11. Koopman RJ, Mainous AG 3rd, Diaz VA, Geesey ME. Changes in age at diagnosis of type 2 diabetes mellitus in the United States, 1988 to 2000. Ann Fam Med 2005;3:60–3. 12. Yildiz BO, Bolour S, Woods K, Moore A, Azziz R. Visually scoring hirsutism. Hum Reprod Update 2010;16:51–64. 13. Hillier TA, Pedula KL. Complications in young adults with early-onset type 2 diabetes: losing the relative protection of youth. Diabetes Care 2003;26: 2999–3005. 14. Quah JHM, Liu YP, Luo N, How CH, Tay EG. Younger adult type 2 diabetic patients have poorer glycaemic control: a cross-sectional study in a primary care setting in Singapore. BMC Endocr Disord 2013;13:18. 15. Song SH, Hardisty CA. Early-onset type 2 diabetes mellitus: an increasing phenomenon of elevated cardiovascular risk. Expert Rev Cardiovasc Ther 2008;6:315–22. 16. Fox R. Prevalence of a positive family history of type 2 diabetes in women with polycystic ovarian disease. Gynecol Endocrinol 1999;13:390–3. 17. Sir-Petermann T, Angel B, Maliqueo M, Carvajal F, Santos JL, Perez-Bravo F. Prevalence of type II diabetes mellitus and insulin resistance in parents of women with polycystic ovary syndrome. Diabetologia 2002;45:959–64. 18. Lerchbaum E, Schwetz V, Giuliani A, Obermayer-Pietsch B. Influence of a positive family history of both type 2 diabetes and PCOS on metabolic and endocrine parameters in a large cohort of PCOS women. Eur J Endocrinol 2014;170:727–39. 19. Mendoza N. Common genetic aspects between polycystic ovary syndrome and diabetes mellitus. Curr Diabetes Rev 2011;7:377–91. 20. Ewens KG, Jones MR, Ankener W, Stewart DR, Urbanek M, Dunaif A, et al. Type 2 diabetes susceptibility single-nucleotide polymorphisms are not associated with polycystic ovary syndrome. Fertil Steril 2011;95:2538–41. e1–6. 21. Saxena R, Welt CK. Polycystic ovary syndrome is not associated with genetic variants that mark risk of type 2 diabetes. Acta Diabetol 2013;50:451–7. 22. Lopez LM, Grimes DA, Schulz KF. Steroidal contraceptives: effect on carbohydrate metabolism in women without diabetes mellitus. Cochrane Database Syst Rev 2014:CD006133. 23. Margolis KL, Bonds DE, Rodabough RJ, Tinker L, Phillips LS, Allen C, et al. Effect of oestrogen plus progestin on the incidence of diabetes in postmenopausal women: results from the Women's Health Initiative Hormone Trial. Diabetologia 2004;47:1175–87. 24. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care 2002;25:1862–8. 25. Lykke JA, Langhoff-Roos J, Sibai BM, Funai EF, Triche EW, Paidas MJ. Hypertensive pregnancy disorders and subsequent cardiovascular morbidity and type 2 diabetes mellitus in the mother. Hypertension 2009;53:944–51. 26. Steegers EAP, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. Lancet 2010;376:631–44. 27. Xiong X, Saunders LD, Wang FL, Demianczuk NN. Gestational diabetes mellitus: prevalence, risk factors, maternal and infant outcomes. Int J Gynaecol Obstet 2001;75:221–8. 28. Moses RG, Morris GJ, Petocz P, San Gil F, Garg D. The impact of potential new diagnostic criteria on the prevalence of gestational diabetes mellitus in Australia. Med J Aust 2011;194:338–40. 29. Bjercke S, Dale PO, Tanbo T, Storeng R, Ertzeid G, Abyholm T. Impact of insulin resistance on pregnancy complications and outcome in women with polycystic ovary syndrome. Gynecol Obstet Invest 2002;54:94–8. 30. Catalano PM, Kirwan JP, Haugel-de Mouzon S, King J. Gestational diabetes and insulin resistance: role in short- and long-term implications for mother and fetus. J Nutr 2003;133:1674S–83S.
VOL. - NO. - / - 2016