Phenotype and metabolic profile of South Asian women with polycystic ovary syndrome (PCOS): results of a large database from a specialist Endocrine Clinic

Chandrika N. Wijeyaratne 2 Ruwanthi de A. Seneviratne 2 Shamalka Dahanayake 2 Vindya Kumarapeli 1 Ethusha Palipane 2 Nadeera Kuruppu 0 Chandrika Yapa 0 Rohini de A. Seneviratne 4 Adam H. Balen 3 0 Professorial Unit in Obstetrics and Gynaecology, De Soysa Hospital for Women , Colombo, Sri Lanka 1 Non Communicable Disease Unit, Ministry of Health , Colombo 10, Sri Lanka 2 Department of Obstetrics and Gynaecology, Faculty of Medicine, University of Colombo , PO Box 271, Kynsey Road, Colombo 08, Sri Lanka 3 The Leeds Centre for Reproductive Medicine, Leeds Teaching Hospitals , Leeds, UK 4 Department of Community Medicine, Faculty of Medicine, University of Colombo , Colombo 08, Sri Lanka background: Compared with other populations, South Asians have a greater propensity to insulin resistance and the metabolic syndrome (MetS). This is the first study to determine the distribution of phenotypes of polycystic ovary syndrome (PCOS) and their relationship to the MetS among indigenous South Asians. method: An evaluation of the phenotype and metabolic characteristics of PCOS was conducted by recruiting consecutive women diagnosed by Rotterdam consensus criteria from an Endocrine clinic in Colombo, Sri Lanka. Prevalence of MetS was determined, in relation to the phenotypic subgroup of PCOS and compared with ethnically matched, BMI- and age-adjusted controls (n 231). results: Acanthosis nigricans (AN) occurred in 64.6% of women with PCOS (n 469). MetS occurred in 30.6% of the PCOS group compared with 6.34% of controls (P 0.0001). Those with PCOS and MetS had significantly higher median BMI, blood pressure (BP), fasting plasma glucose, insulin and triglycerides and lower high-density lipoprotein and sex hormone-binding globulin (SHBG), but similar testosterone concentrations compared with those with PCOS alone. Prevalence of MetS was similar in the four PCOS phenotypes, although oligomenorrhoeic women were more obese compared with the normal cycling hyperandrogenic group. Multivariate logistic regression confirmed age 35 years, BMI 25 kg/m2 and AN as significant predictors of MetS in PCOS. Case - control comparisons showed that the presence of PCOS results in higher odds of having the MetS, a high waist circumference, elevated diastolic BP, abnormal fasting lipids and high fasting insulin and plasma testosterone concentrations. conclusions: Young indigenous South Asians with PCOS have greater odds of being centrally obese, with a third having the MetS that bears no relationship to the androgenic phenotype. Significant predictors for MetS within the PCOS cohort are advancing age, obesity determined by the Asian cut off (BMI .25 kg/m2) and AN, while family history of diabetes, hyperandrogenism and elevated SHBG have no predictive value. - & The Author 2010. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: Introduction Polycystic ovary syndrome (PCOS) is associated with insulin resistance (IR) and the metabolic syndrome (MetS) (Apridonidze et al., 2005; Moran et al., 2010). We previously demonstrated that migrant South Asians with anovulatory PCOS have more severe symptoms at a younger age, with greater IR than white Europeans (Wijeyaratne et al., 2002, 2004). A recent community survey in Sri Lanka found the prevalence of PCOS as 6.3% among women aged 15 39 (Kumarapeli et al., 2008). The manifestations of PCOS are heterogeneous, can change during a womans lifetime (Franks et al., 2006), and can impact on health-related quality of life (Ching et al., 2007), as well as on long-term cardiovascular health (Teede et al., 2010). Although the revised Rotterdam diagnostic criteria led to some controversy about over-diagnosis of PCOS (Azziz, 2006), there are four possible phenotypes based on the presence of oligo-/anovulation (O), hyperandrogenism (H) and PCO (P): (i) classical (O + H + P), (ii) hyperandrogenic (O + H), (iii) ovulatory (H + P) and (iv) nonhyperandrogenic (O + P) (Shroff et al., 2007). Most reports on MetS are from studies carried out in the West (Apridonidze et al., 2005; Shroff et al., 2007; Goverde et al., 2009) and encompass cardiovascular risks such as hyperlipidaemia, hypertension and diabetes mellitus/pre-diabetes that cause endothelial dysfunction and atherosclerosis (Kressel et al., 2009). To the best of our knowledge, there are no reports on the prevalence of MetS among South Asian women with PCOS, nor the distribution of the differing PCOS phenotypes. This study aimed for the first time to identify phenotypic subgroups of PCOS and their relationship to the MetS among a large cohort of indigenous women of South Asian descent seeking specialist endocrine care at a single centre in Sri Lanka. PCOS is reported from many regions of the world (Franks, 1989; Diamanti-Kandarakis et al., 1999; Asuncion et al., 2000; Azziz et al., 2004) with well recognized ethnic variations in its manifestation. Much of our current knowledge is based on evidence from predominantly white European women. Studies of mixed populations residing in the USA, attempted to identify the Asian ethnic subgroup (Ehrmann et al., 2006; Legro et al., 2006; Lo et al., 2006; Welt et al., 2006), although migrant Asians comprised ,10% of each cohort without detail of their exact geographic origin from Asia, which is the largest continent and has wide variation between its subgroups e.g. East Asians (Chinese/Japanese), South East Asians (Thai/Malay people) and South Asians (Indians/Sri Lankans). Critical appraisal of the expression of PCOS within a distinct ethnic group requires careful and uniform phenotyping of a sufficiently large and representative sample, to evaluate age of manifestation, metabolic characteristics and link(s) between the androgenic, reproductive and metabolic phenotypes. The link between IR and PCOS is strong and has been identified as an important determinant of its phenotypic manifestations and response to treatment (Salley et al., 2007; Shroff et al., 2007). Recent data on type 2 diabetes mellitus (T2DM) confirms its greater propensity among young adults of South Asia, which is linked to the epidemiological transition and rapid urbanization occurring in the region (Eapen et al., 2009; Misra and Khurana, 2009). In view of our previous comparison of migrant South Asians in the UK with indigenous white Europeans, demonstrating a disparity in the phenotypic expression of PCOS (Wijeyaratne et al., 2002), an in-depth study of affected indigenous South Asians could shed better light on this subgroup from Asia, and help disentangle possible confounding effects of migration. Finally, it would enable developing evidence-based and ethnically appropriate management guidelines. Materials and Methods Objective and study design Our aim was to determine the distribution of phenotypes of PCOS and their relationship to the MetS among indigenous South Asians. A descriptive cross-sect (...truncated)