Abstract
The aim of the study was to investigate whether altered adipose tissue secretion of various adipokines is secondary to obesity, hyperandrogenism, and hyperinsulinemia or intrinsic to polycystic ovary syndrome (PCOS). This cross-sectional study included 151 women diagnosed with PCOS by the Rotterdam criteria and 95 healthy women matched by age, body mass index (BMI), and waist-to-hip ratio (WHR). Clinical, biochemical, and hormonal characteristics were assessed. Serum concentrations of ghrelin and adiponectin were found to be significantly lower and concentrations of leptin and resistin significantly higher in women with PCOS than in healthy women matched by age, BMI, and WHR. A PCOS diagnosis made the largest contribution to predicting serum levels of leptin, adiponectin, resistin, and ghrelin in all stepwise multiple regression models, which included PCOS diagnosis, BMI, WHR, luteinizing hormone, total testosterone, free testosterone and homeostatic model assessment of insulin resistance as independent predictors. Leptin, adiponectin, ghrelin and resistin levels may serve as independent biomarkers for the diagnosis of PCOS.
摘要
本研究的目的是探讨各种脂肪组织分泌的脂肪因子的改变是继发于肥胖、高雄激素血症和高胰岛素血症, 还是多囊卵巢综合征（PCOS）的内在原因。这项横断面研究包括151名根据鹿特丹标准诊断为多囊卵巢综合征的妇女和95名在年龄、体重指数（BMI）和腰臀比（WHR）与之相匹配的健康妇女。评估临床、生化和激素特征。研究发现, 与年龄、体重指数和腰臀比相匹配的健康女性相比, 多囊卵巢综合征患者的血清中饥饿素和脂联素的浓度显著降低, 瘦素和抵抗素的浓度显著升高。在逐步多元回归模型中, PCOS诊断对预测血清瘦素、脂联素、抵抗素和饥饿素水平的贡献最大, 其中, PCOS诊断、BMI、WHR、促黄体生成素、总睾酮、游离睾酮和胰岛素抵抗的稳态模型评估作为独立预测因子。瘦素、脂联素、胃饥饿素和抵抗素水平可作为诊断多囊卵巢综合征的独立生物标志物。
The Chinese abstracts are translated by Prof. Dr. Xiangyan Ruan and her team: Beijing Obstetrics and Gynecology Hospital, Capital Medical University, Beijing 100026, China.
Introduction
Polycystic ovary syndrome (PCOS) is a multisystem, endocrinological, reproductive and metabolic disorder characterized by oligo-/anovulation, hyperandrogenism and polycystic ovaries [Citation1,Citation2]. Many of the metabolic abnormalities that manifest in PCOS, mainly including insulin resistance (IR), impaired glucose tolerance, type 2 diabetes mellitus (DM2), and dyslipidemia, are worsened by concurrent obesity [Citation3]. However, some of these metabolic perturbations occur even in lean women with PCOS and therefore are rightfully recognized as intrinsic to PCOS [Citation4–6]. Although many of these findings can be largely explained by the increased prevalence of abdominal obesity, even in normal-weight PCOS patients [Citation7], some data suggest that disrupted secretion of adipose tissue-derived hormones (adipokines) and the gut hormone ghrelin precede the signs of metabolic syndrome observed in PCOS patients [Citation6]. Whether adipokine and ghrelin dysfunction is a consequence of the interaction among obesity, visceral fat distribution, hyperandrogenemia, and hyperinsulinemia or is an intrinsic feature of PCOS is yet to be determined.
Therefore, the aim of this study was to investigate whether altered adipose tissue secretion of various adipokines is secondary to obesity, hyperandrogenism, and hyperinsulinemia or intrinsic to PCOS.
Subjects and methods
Study population
A portion of the data from this cohort was published previously [Citation8]. The patients were recruited from the University of Zagreb Clinical Hospital Center Zagreb, Croatia, from 2009 to 2011. A total of 151 PCOS patients and 95 healthy control subjects were enrolled. The diagnosis of PCOS was confirmed according to the Rotterdam consensus criteria [Citation1,Citation2].
The control group consisted 95 healthy volunteers before entering an in vitro fertilization (IVF) program due to male-factor infertility. For all women included in the control group, ovulation was confirmed by a progesterone level ≥22 nmol/L obtained during the luteal phase of two consecutive menstrual cycles. All patients in the control group had no clinical or biochemical signs of hyperandrogenism and had normal ultrasound imaging results of the ovaries. The control group and the PCOS group were matched by age, body mass index (BMI), and waist-to-hip ratio (WHR).
No participants used any medication that could influence androgen, glucose, insulin or lipid levels for at least six months prior entering to study. All participants were recruited during the early follicular phase of a spontaneous or progesterone-induced menstrual cycle (day 3–5). After overnight fasting (>12 h), blood samples for biochemical and hormonal analysis were drawn, transvaginal ultrasound was performed, and the BMI and WHR were calculated.
Biochemical measurements
Fasting blood samples were obtained for measurements of luteinizing hormone (LH), follicle-stimulating hormone (FSH), thyrotrophic hormone (TSH), prolactin (PRL), total testosterone (TT), sex hormone-binding globulin (SHBG), free testosterone (FT), dehydroepiandrosterone sulfate (DHEAS), androstenedione (A), 17-OH-progesterone (17-OHP), glucose, insulin, leptin, adiponectin, resistin and ghrelin. Serum LH, FSH, TSH, PRL and TT concentrations were determined by chemiluminescent immunometric assays (Ortho-Clinical Diagnostics, Johnson & Johnson, Rochester, NY). Serum SHBG, DHEAS, and A levels were measured using chemiluminescent immunometric assays (Siemens Healthcare Diagnostics Inc., Deerfield, IL). The concentration of 17-OHP was determined by a solid-phase enzyme-linked immunosorbent assay (ELISA) based on the principle of the competitive binding (DRG-diagnostics, Marburg, Germany). FT was calculated from the TT and SHBG levels as previously described [Citation9] using a web-based calculator (http://www.issam.ch/freetesto.htm). Plasma glucose levels were measured by the UV-photometric hexokinase method, and serum insulin levels were measured by chemiluminescent immunometric assay (Siemens Healthcare Diagnostics Inc., Deerfield, IL). IR was quantified using the homeostatic model assessment of IR (HOMA-IR) (fasting insulin (mU/L)×fasting glucose (mmol/L))/22.5) [Citation10]. Serum leptin levels were measured by radioimmunoassay (RIA, DRG International, NJ). Serum adiponectin levels were measured by an ELISA (Bio Vendor, Czech Republic). Serum resistin levels were measured by an enzyme immunoassay (DRG International, NJ). Serum ghrelin levels were measured with an RIA (DRG International, NJ).
Statistical analysis
Values are expressed in means ± standard deviations. An independent Student’s t-test was used to compare the mean values between patients and controls. Pearson’s correlation coefficient (r) was used to assess the linear associations of different clinical, biochemical and hormonal parameters with serum leptin, adiponectin, resistin and ghrelin levels in both patients and controls. To explore the effects of PCOS, BMI, WHR, HOMA-IR, LH, TT and FT on serum leptin, adiponectin, resistin, and ghrelin levels, stepwise multiple regression models were constructed with leptin, adiponectin, resistin and ghrelin as dependent variables and all others as independent variables. All statistical analyses were performed using SPSS for Windows (version 15.0; SPSS, Inc., Chicago, IL). A value of p < .05 was considered to indicate a statistically significant difference.
The Ethics Committee of the Medical School, University of Zagreb, approved the study (protocol No. 04–1116-2006). Informed written consent was obtained from all participants enrolled in the study.
Results
The baseline characteristics of PCOS patients and control subjects are presented in . Associations of different clinical, hormonal and metabolic parameters with serum leptin, adiponectin, resistin, and ghrelin levels in PCOS patients and controls are presented in .
Table 1. Clinical, hormonal and metabolic characteristics of women with PCOS and the healthy control group.
Table 2. Associations of different clinical, hormonal and metabolic parameters with serum leptin, adiponectin, resistin, and ghrelin levels in PCOS patients and controls.
To determine whether adipokine secretion is possibly disrupted due to the interactions among obesity, abdominal fat distribution, LH, hyperandrogenism and IR or is an intrinsic feature of PCOS, we tested the predictive values of a PCOS diagnosis and the BMI, WHR, LH, TT, FT and HOMA-IR values for the serum concentrations of adipokines in all subjects included in the study (). All variables were entered into stepwise regression models with leptin, adiponectin, ghrelin, and resistin as dependent variables. A model using PCOS, BMI, WHR, and TT as predictors explained 70.5% (adjusted R2) of the variability in serum leptin levels, whereas 77.9% (adjusted R2) of the variability of serum adiponectin could be explained by PCOS and BMI. A model using BMI, PCOS, WHR, and HOMA-IR as predictors explained 79.6% (adjusted R2) of the variability of serum ghrelin levels, while 59% (adjusted R2) of the variability of serum resistin levels could be explained using PCOS, BMI, and FT as predictors. In all stepwise multiple regression models, a PCOS diagnosis was the most significant predictor for each adipokine tested ().
Table 3. Stepwise multiple regression models with leptin, adiponectin, ghrelin, and resistin as dependent variables.
Discussion
Although adipokine dysregulation is common in PCOS, it is still unknown whether these changes are secondary to obesity, hyperandrogenism, and hyperinsulinemia or intrinsic to PCOS. This study aimed to clarify the inconsistent results reported by previous studies.
Leptin
We found significantly increased serum leptin levels in PCOS patients compared to control subjects, even when stratified by BMI, suggesting that the elevated serum leptin level in PCOS patients is not only a result of increased body mass but is also due to specific factors related to PCOS. Our results, however, showed no association of serum leptin levels with TT or FT serum levels in women with PCOS, which is consistent with previously published studies [Citation11,Citation12]. In contrast, when these associations were controlled for BMI and WHR, a significant negative association between serum leptin and TT levels was observed. TT was retained as a significant negative predictor of the serum leptin level when it was included as an independent variable in our stepwise regression model, further confirming the association.
The HOMA-IR value was not independently associated with serum leptin levels. We probably failed to detect this association because 2/3 of our PCOS patients were of normal weight. We speculate that if a higher number of overweight PCOS patients had been included in the study, the more pronounced role of IR associated with a higher BMI on serum leptin levels would have been identified.
When we constructed a stepwise regression model employing the serum leptin level as an objective variable and adjusted for various confounding factors as explanatory variables, a PCOS diagnosis had the greatest predictive value for serum leptin levels, followed by BMI. It is therefore, obvious that the serum leptin level is nearly equally affected by obesity and the diagnosis of PCOS per se and that among all examined adipokines in our study, leptin is perhaps the least specific for the diagnosis of PCOS.
Adiponectin
Many studies, including ours, have demonstrated hypoadiponectinemia in PCOS patients irrespective of obesity; however, these studies highlighted the importance of abdominal fat distribution. We failed to establish a link between serum levels of adiponectin and WHR, contrary to the results of other studies [Citation13–16]. The lack of association between WHR and adiponectin in our population may be partly explained by the higher proportion of normal-weight PCOS patients included in our study, as well as possible differences in body fat distribution among Croatian women compared to women of other European countries.
We found no association between serum adiponectin levels and serum concentrations of TT or FT in PCOS patients, which is consistent with the work of several other authors [Citation17]. We observed a significant negative association between DHEAS and adiponectin levels in our PCOS group, but this was lost after controlling for BMI and WHR.
Several reports have described associations of adiponectin levels with indicators of IR [Citation17–19] which were further supported by a meta-analysis published in 2009 [Citation17]. We were unable to demonstrate any association between serum adiponectin levels and IR measured by the HOMA-IR in our population, similar to studies by Sarray et al. and O’Connor et al. [Citation20] and the meta-analysis published by Li et al. [Citation21]. These conflicting results are likely attributable to the differences in genetic predisposition for obesity and IR in distinct populations.
We demonstrated that a PCOS diagnosis served as the strongest independent predictor of serum adiponectin levels in stepwise multiple regression models. Adiponectin was the most specific factor for a diagnosis of PCOS among all adipokines examined.
Resistin
In our study, serum resistin levels were significantly elevated in PCOS patients compared to controls, irrespective of BMI, which is consistent with the results of Yilmaz et al. [Citation22]. We found significant inverse correlations between serum resistin level and BMI as well as WHR in both the PCOS and control group, suggesting that obesity parameters negatively influence serum resistin levels.
We did not establish a link between resistin levels and HOMA-IR values or between resistin and the levels of androgen hormones in women with PCOS. Surprisingly, FT remained a significant independent predictor of serum resistin levels when we controlled for PCOS, BMI, WHR, HOMA-IR, LH, and TT in a stepwise regression model.
In stepwise multiple regression analysis, when a diagnosis of PCOS, BMI, WHR, HOMA-IR, LH, TT, and FT were included as independent explanatory variables of the concentration of serum resistin, a PCOS diagnosis was the strongest predictor among them.
Ghrelin
We found significantly lower serum ghrelin levels in PCOS patients than in healthy women, irrespective of BMI, which is consistent with data published by Barber et al. [Citation23]. We found a significant negative association between serum ghrelin levels and HOMA-IR in the PCOS group, but this association was lost after adjustment for BMI and WHR. Furthermore, we did not establish a link between the concentrations of serum ghrelin and TT or FT in women with PCOS, in contrast with previously published studies [Citation14,Citation24].
The stepwise multiple regression model demonstrated that the variability of serum ghrelin could be explained by a PCOS diagnosis, BMI and HOMA-IR as confounding factors. A PCOS diagnosis made the largest unique contribution to predicting serum ghrelin levels.
Conclusion
The results of our study clearly demonstrate significantly altered secretion of leptin, adiponectin, resistin, and ghrelin in PCOS patients compared to healthy subjects irrespective of obesity parameters. In all stepwise multiple regression models, a PCOS diagnosis made the largest unique contribution to predicting serum leptin, adiponectin, resistin, and ghrelin levels. Therefore, it is likely that other factors independent of obesity, IR and increased production of androgens, yet unique to the diagnosis of PCOS, affect serum levels of adipokines and ghrelin in women with PCOS.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
- Rotterdam ESHRE/ASRM‐sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod. 2004;19:41–47.
- Azziz R, Carmina E, Dewailly D, et al. Positions statement: criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society guideline. J Clin Endocrinol Metab. 2006;91:4237–4245.
- Teede H, Deeks A, Moran L. Polycystic ovary syndrome: a complex condition with psychological, reproductive and metabolic manifestations that impacts on health across the lifespan. BMC Med. 2010;8:41.
- Diamanti-Kandarakis E, Dunaif A. Insulin resistance and the polycystic ovary syndrome revisited: an update on mechanisms and implications. Endocr Rev. 2012;33:981–1030.
- Moran LJ, Misso ML, Wild RA, et al. Impaired glucose tolerance, type 2 diabetes and metabolic syndrome in polycystic ovary syndrome: a systematic review and meta-analysis. Hum Reprod Update. 2010;16:347–363.
- Randeva HS, Tan BK, Weickert MO, et al. Cardiometabolic aspects of the polycystic ovary syndrome. Endocr Rev. 2012;33:812–841.
- Carmina E, Bucchieri S, Esposito A, et al. Abdominal fat quantity and distribution in women with polycystic ovary syndrome and extent of its relation to insulin resistance. J Clin Endocrinol Metab. 2007;92:2500–2505.
- Baldani DP, Skrgatic L, Goldstajn MS, et al. Clinical and biochemical characteristics of polycystic ovary syndrome in Croatian population. Coll Antropol. 2012;36:1413–1418.
- Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab. 1999;84:3666–3672.
- Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412–419.
- Pehlivanov B, Mitkov M. Serum leptin levels correlate with clinical and biochemical indices of insulin resistance in women with polycystic ovary syndrome. Eur J Contracept Reprod Health Care. 2009;14:153–159.
- Remsberg KE, Talbott EO, Zborowski JV, et al. Evidence for competing effects of body mass, hyperinsulinemia, insulin resistance, and androgens on leptin levels among lean, overweight, and obese women with polycystic ovary syndrome. Fertil Steril. 2002;78:479–486.
- Escobar-Morreale HF, Villuendas G, Botella CJI, et al. Adiponectin and resistin in PCOS: a clinical, biochemical and molecular genetic study. Hum Reprod. 2006;21:2257–2265.
- Glintborg D, Andersen M, Hagen C, et al. Evaluation of metabolic risk markers in polycystic ovary syndrome (PCOS). Adiponectin, ghrelin, leptin and body composition in hirsute PCOS patients and controls. Eur J Endocrinol. 2006;155:337–345.
- Ducluzeau PH, Cousin P, Malvoisin E, et al. Glucose-to-insulin ratio rather than sex hormone-binding globulin and adiponectin levels is the best predictor of insulin resistance in nonobese women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2003;88:3626–3631.
- Sieminska L, Marek B, Kos-Kudla B, et al. Serum adiponectin in women with polycystic ovarian syndrome and its relation to clinical, metabolic and endocrine parameters. J Endocrinol Invest. 2004;27:528–534.
- Toulis KA, Goulis DG, Farmakiotis D, et al. Adiponectin levels in women with polycystic ovary syndrome: a systematic review and a meta-analysis. Hum Reprod Update. 2009;15:297–307.
- Arikan S, Bahceci M, Tuzcu A, et al. Serum resistin and adiponectin levels in young non-obese women with polycystic ovary syndrome. Gynecol Endocrinol. 2010;26:161–166.
- Cassar S, Teede HJ, Harrison CL, et al. Biomarkers and insulin sensitivity in women with polycystic ovary syndrome: characteristics and predictive capacity. Clin Endocrinol. 2015;83:50–58.
- O'Connor A, Phelan N, Tun TK, et al. High-molecular-weight adiponectin is selectively reduced in women with polycystic ovary syndrome independent of body mass index and severity of insulin resistance. J Clin Endocrinol Metab. 2010;95:1378–1385.
- Li S, Huang X, Zhong H, et al. Low circulating adiponectin levels in women with polycystic ovary syndrome: an updated meta-analysis. Tumor Biol. 2014;35:3961–3973.
- Yilmaz M, Bukan N, Demirci H, et al. Serum resistin and adiponectin levels in women with polycystic ovary syndrome. Gynecol Endocrinol. 2009;25:246–252.
- Barber TM, Casanueva FF, Karpe F, et al. Ghrelin levels are suppressed and show a blunted response to oral glucose in women with polycystic ovary syndrome. Eur J Endocrinol. 2008;158:511–516.
- Panidis D, Farmakiotis D, Koliakos G, et al. Comparative study of plasma ghrelin levels in women with polycystic ovary syndrome, in hyperandrogenic women and in normal controls. Hum Reprod. 2005;20:2127–2132.