Plasma interleukin-18 levels are increased in the
polycystic ovary syndrome: relationship of carotid
intima-media wall thickness and cardiovascular risk
factors
Cemil Kaya, M.D.,aRecai Pabuccu, M.D.,aB€ulent Berker, M.D.,band Hakan Satıroglu, M.D.b aDepartment of Obstetrics and Gynecology, Ufuk University, andbDepartment of Obstetrics and Gynecology, Ankara
University, Ankara, Turkey
Objective: To determine serum interleukin (IL)-18 levels and to find out whether IL-18 is associated with carotid intima-media wall thickness (IMT) and various cardiovascular risk factors in women with polycystic ovary syn-drome (PCOS).
Design: A prospective, controlled study. Setting: University hospital.
Patient(s): Sixty women with PCOS and 60 healthy women were included this study.
Intervention(s): Serum levels IL-18, homocysteine (Hcy), C-reactive protein (CRP), IL-6, malonyldialdehyde (MDA), lipid and hormone profiles were measured. Carotid IMT was evaluated for both common carotid arteries. Main Outcome Measure(s): Serum IL-18, carotid IMT, Hcy, CRP, IL-6, MDA, and homeostasis model assessment of insulin resistance.
Result(s): The evaluation, which was made without the obesity influence taken into consideration, revealed that patients with PCOS have increased serum IL-18 levels than that of the control group (214 102 vs. 170 78 pg/mL). The interaction between PCOS and obesity was seen to have statistical significance (F ¼ 67.8). Body mass index (BMI), waist to-hip ratio, Hcy, and homeostasis model assessment of insulin resistance are in-dependent determinants of plasma IL-18 in patients with PCOS. Elevated serum IL-18 levels were positively and significantly correlated with a greater carotid IMT. For Hcy and carotid IMT, the interaction between PCOS and obesity was found in a two-way ANOVA variation analysis (F¼ 48.5 and F ¼ 81.5, respectively). Conclusion(s): Elevated serum IL-18 levels were associated with cardiovascular risk factors and carotid IMT in patients with PCOS. (Fertil Steril2010;93:1200–7.2010 by American Society for Reproductive Medicine.)
Key Words: PCOS, IL-18, IL-6, BMI, insulin resistance, Hcy, MDA, carotid IMT
Polycystic ovary syndrome (PCOS) is considered as a clear-put plurimetabolic syndrome, being associated with obesity, insulin resistance (IR), dyslipidemia, endothelial dysfunc-tion, reduced vascular compliance, and premature carotid atherosclerosis(1–5). Obesity and IR are associated with ad-ipose tissue secretion of proinflammatory cytokines(6). Se-rum markers of inflammation are being increasingly recognized as predictors of atherosclerosis and cardiovascu-lar disease(6–8).
Serum interleukin (IL)-18 levels in women with PCOS are elevated and correlated with obesity and IR(9, 10). El-evated IL-6 and C-reactive protein (CRP) levels are asso-ciated with obesity, inflammation, and coronary heart disease(10–13). Serum levels of CRP are significantly in-creased and may be involved in the atherogenic process in
women with PCOS (14). Homocysteine (Hcy) has been reported to promote atherosclerosis by inducing endothe-lial dysfunction (15). There is some evidence that IL-18 concentrations may be linked with hyperhomocysteinemia in various diseases such as diabetes mellitus or systemic lupus erythematosus (SLE) (16, 17). Malonyldialdehyde (MDA) is a marker of lipid peroxidation and increases in oxidative stress states(18). The complex interaction be-tween inflammatory state and oxidative stress may result in the accelerated atherosclerosis seen in patients with PCOS.
Recent studies reveal that elevated plasma levels of IL-18 have adverse effects on the cardiovascular system. Carotid in-tima-media wall thickness (IMT) is a commonly used clinical marker of atherosclerosis(19). In addition, increased carotid IMT has been a predictor of future coronary events and stroke (20).
However, at present, the association of IL-18 with carotid IMT and various cardiovascular risk factors, such as Hcy, CRP, IL-6, and MDA, in patients with PCOS have not been specifically established. Therefore, the aim of the study is to determine serum IL-18 levels and to determine whether IL-18 is associated with cardiovascular risk factors (obesity,
Received May 31, 2008; revised October 29, 2008; accepted October 30, 2008; published online January 7, 2009.
C.K. has nothing to disclose. R.P. has nothing to disclose. B.B. has noth-ing to disclose. H.S. has nothnoth-ing to disclose.
Selected for poster presentation during ASRM 2008, San Fransisco, Cal-ifornia, November 8–12, 2008.
Reprint requests: Cemil Kaya, M.D., Department of Obstetrics and Gyne-cology, Ufuk University, Balgat, 312Ankara, Ankara, Turkey (FAX: 90-312-2851158; E-mail:kayacemil000@yahoo.com).
waist-to-hip ratio [WHR], IR, Hcy, IL-6, CRP, and MDA) and carotid IMT in patients with PCOS.
MATERIALS AND METHODS Patients
Patients with PCOS presented with irregular menstrual cy-cles. Control group subjects were healthy volunteers with normal menstrual cycles and who had no clinical or biochem-ical features of hyperandrogenism. The majority of the con-trol group consisted of students and hospital staff. The diagnosis of PCOS was made when R2 of the following 3 criteria existed, as proposed at the Rotterdam Consensus Meeting: oligomenorrhea or amenorrhea, clinical hyperan-drogenism or hyperandrogenemia, and polycystic ovaries (PCO)(21). The presence of polycystic ovarian appearance was determined ultrasonographically (22). Oligomenorrhea (cycle intervals >35 days), amenorrhea (absence of menstru-ation for 3 consecutive months), and luteal phase P measure-ments less than 4 ng/mL in regular menstrual cycles were considered indicative of oligoovulation. Hirsutism was deter-mined by a modified Ferriman-Gallwey score above 7(23). Nonclassic adrenal 21-hyroxylase deficiency, hyperprolacti-nemia, and androgen-secreting tumors were excluded by ap-propiate tests before the diagnosis of PCOS was made.
Neither the PCOS patients nor the control subjects had a thyroid disorder, hyperprolactinemia, infectious diseases, diabetes mellitus, hypertension, congenital adrenal hyperpla-sia, androgen-secreting tumors, signs or symptoms of other androgen-secreting tumors, or other endocrinopathies. Con-founding medications, including oral contraceptive agents, antilipidemic drugs, hypertensive medications, and insulin-sensitizing drugs, which may affect the metabolic criteria, were questioned. Subjects taking vitamins or drugs, which in-crease Hcy levels, within the previous 6 months and those containing folic acid, vitamin B12, or vitamin B6or with de-ficiencies were excluded from the study. The women were in good general health and none had been taking hormonal med-ication or oral contraceptives during the past 3 months before the investigation. Patients and controls who had a history of smoking were excluded as well. This study was conducted at Ufuk University and at Ankara University. All subjects were asked to give a written consent and the institutional re-view boards (IRB) of the hospitals approved the study.
Clinical and antropometric variables, including a modified hirsutism score, body mass index (BMI), and WHR were de-termined by a single investigator (C.K.) in all the subjects. The BMI was calculated as weight (kilograms) divided by height (meters) squared (kg/m2). The BMI values of R30 kg/m2 were considered as obese. Weight and height were measured in light clothing without shoes. Waist circumfer-ence was measured at the narrowest level between the costal margin and the iliac crest, and the hip circumference was measured at the widest level over the buttocks while the sub-jects was standing and breathing normally. The WHR was calculated. A WHR >0.72 was considered abnormal(24).
Carotid IMT was derived from a noninvasive ultrasound of the common carotid arteries, using a high resolution ultra-sound machine (Logic Q7; General Electric, Tokyo, Japan) with a 7.5-MHz mechanical sector transducer. The carotid IMT was defined as the distance between the blood–intima and media–adventitia boundaries on B-mode imaging. Ca-rotid IMT was defined as the distance from the leading edge of the first echonegic line in the sonographic image. Measurements of the carotid IMT were made at each of three sites of the greatest thickness on both sides. Carotid IMT was defined as the mean of these maximal carotid IMT measure-ments. Ultrasonographic measurements were performed by an experienced radiologist(25, 26).
Venous blood samples were obtained in the follicular phase of a spontaneous cycle. After a 3-day 300-g carbohydrate diet and a 12-hour overnight fast, samples were obtained to measure the levels of serum IL-18, Hcy, CRP, IL-6, MDA, hormonal parameters (FSH, LH, PRL, total T, DHEAS, 17a-hydroxyprogesterone [17-OHP], and TSH), and lipid profile (total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, very low-density lipoprotein cholesterol, triglyceride). The Hcy concentrations and basal insulin levels were assessed. Then, a standart 75-g oral glucose tolerance test (OGTT) was per-formed and glucose tolerance state was evaluated by using the criteria of the American Diabetes Association (27). Plasma glucose was determined with the glucose hexokinase (Cobas Integra 400 Plus; Roche Diagnostics, Mannheim, Germany). The presence of an insulin sensitivity index was investigated by using basal insulin levels, fasting glucose, homeostasis model assessment of insulin resistance (HOMA-IR). The HOMA-IR calculated by using the following formula: Fasting glucose (mg/dL) Fasting insulin (mU/mL) 0.055/22.5(28, 29).
Serum levels of FSH, LH, PRL, DHEAS, insulin, and TSH were measured with specific chemiluminescence assays from Roche Diagnostic (Elecys 2010 Hitachi). Serum levels of 17-OHP were measured by RIA. Serum sex hormone-binding globulin (SHBG) was measured by electrochemilumines-cence (Elecsys 2010; Roche Diagnostics). The free androgen index (FAI) was calculated according to the equation: FAI (%)¼ T (ng/mL) 3.47 100/SHBG (nmol/L). Levels of total cholesterol, HDL cholesterol, LDL cholesterol, and tri-glycerides were determined with enzymatic colorimetric as-says (Roche Diagnostic). Samples were immediately centrifuged, and serum was separated and frozen at -20C un-til assayed. The intra-assay and interassay coefficients of var-iation (CV) were <5% for all assays performed.
Serum CRP was measured by latex immunoturbidometric methodology on an automated clinical analyzer system (Co-bas Integra, Roche Diagnostic). Plasma Hcy levels were mea-sured as total Hcy technique by high-performance liquid chromatography (HPLC) using chromosystems kits with fluorescence detector (Mannheim, Germany). The intra-as-say and interasintra-as-say CVs were <2%. Serum MDA levels were determined in plasma by the 2-thiobarbituric acid
reactive substance method. 1,1,3,3 Tetraethoxypropane (Sigma Aldrich, St. Louis, MO) was used as a standard.
Serum 18 levels were measured by ELISA (Human IL-18 ELISA Kit; Medical & Biological Laboratories Co. Ltd., Nagoya, Japan) with a lower limit of detection of 12.5 pg/mL and mean intra-assay and interassay CVs of 7.3% and 7.5%, respectively.
Serum IL-6 levels were measured by ELISA (Human IL-6 ELISA Kit; Medical & Biological Laboratories Co. Ltd.) and mean intra-assay and interassay CVs of 3.5% and 8.7%, respectively.
Statistical Analysis
Data analysis was performed by using SPSS for Windows, version 11.5 (SPSS Inc., Chicago, IL). The Shapiro Wilk test was used to determine whether continuous variables were normally distributed. Data are presented as mean SD, median and interquartile range, and P<.05 was consid-ered statistically significant. Mean ages were compared by Student’s t test. The differences between PCOS and control groups regarding homogenously distributed variables were evaluated by the Mann-Whitney U test. Two-way analysis of variance (ANOVA) was performed to evaluate the influ-ence of PCOS or control status and of being nonobese and obese (BMI R30 as cutoff) and to determine the possible in-teraction between these independent variables, on serum IL-18. To clarify the influence of PCOS and obesity on other CV risk factors such as IL-6, CRP, MDA, Hcy, and carotid IMT, we applied the two-way ANOVA once more. The differences between PCOS and control groups were evaluated by using Student’s t or Mann-Whitney U tests, as appropriate. Degree of association between continuous variables and serum IL-18 were calculated by Spearman’s rho correlation coefficient. Stepwise multiple linear regression was used to find the ma-jor determinants of serum IL-18 among those variables show-ing significant correlations. The stepwise multiple linear regression method was used to determine the independent predictors that mostly affected carotid IMT, among serum in-flammatory markers, serum androgens, lipid profiles, and IR. Logarithmic (log) transformations were applied for IL-18 in both two-way ANOVA and lineer regression analysis because of the non-normally distributed data.
RESULTS
After completion of log transformation for IL-18, a two-way ANOVA variation analysis was conducted to study the influ-ence of PCOS and obesity on IL-18. We found a significant ef-fect for PCOS, and efef-fect for obesity. The interaction between PCOS and obesity was seen to have statistical significance (F ¼ 67.8, P<.05). The evaluation made without the obesity in-fluence taken into consideration revealed that patients with PCOS have increased serum IL-18 levels than that of the con-trol group (214 102 vs. 170 78 pg/mL, P<.05) (Table 1). The main effect of PCOS on IL-18 was statistically significant (F¼ 78.4, P<.05). Serum IL-18 levels of the obese patients,
when compared to that of nonobese patients, were higher both in PCOS and control groups (254 93 vs. 54 31.4 pg/mL, P<.001 and 245 79 vs. 28 15.2 pg/mL, P<.01, respec-tively) (Fig. 1). The main effect of obesity on IL-18 was sta-tistically meaningful (F ¼ 177.6, P<.001). However, when obesity was taken into account, PCOS had no effect on the obese group. When PCOS and control groups were compared within the obese group, there was no difference in terms of se-rum Il-18 levels (254 93 pg/mL vs. 245 79 pg/mL, P>.05). When the obesity effect was taken into account, the PCOS effect completely disappeared. When obese and non-obese subjects were compared without the PCOS factor being predetermined, the serum IL-18 level in obese group was sta-tistically significantly higher (275 89 vs. 36.5 19 pg/mL, P<.001) (Fig. 1).
After completion of the log transformation for IL-6, CRP, MDA, Hcy, and carotid IMT, a two-way ANOVA variation analysis was conducted to study the influence of PCOS and obesity on these parameters. For IL-6, CRP, MDA, and ca-rotid IMT, no interaction between PCOS and obesity was
TABLE 1
Correlations coefficient of plasma IL-18 with cardiovascular risk factors in patients with PCOS. Variable Spearman rho coefficient P value Age (y) 0.19 NS BMI (kg/m2) 0.84 <.001 WHR 0.71 <.001 Glucose (mg/dL) 0.13 NS
Fasting insulin (mIU/min/mL) 0.33 <.05
HOMA-IR 0.45 <.01 Total cholesterol (mg/dL) 0.002 NS LDL cholesterol (mg/dL) 0.016 NS HDL cholesterol (mg/dL) 0.05 NS TG (mg/dL) 0.11 NS FAI (%) 0.17 NS IL-6 (mg/dL) 0.18 NS Hcy (mmol/L) 0.68 <.01 CRP (mg/dL) 0.26 <.05 MDA (nmol/L) 0.10 NS
Mean carotid IMT (mm) 0.74 <.001
Note: Statistical significance was defined as P<0.05. PCOS¼ polycystic ovary syndrome; NS ¼ not
signifi-cant; BMI¼ body mass index; WHR ¼ waist-to-hip ratio; HOMA-IR ¼ homeostasis model assessment of insulin resistance; LDL¼ low-density lipoprotein; HDL ¼ high-density lipoprotein; TG ¼ triglyceride; FAI¼ free androgen index; IL ¼ interleukin; Hcy ¼ ho-mocysteine; CRP¼ C-reactive protein; MDA ¼ malo-nyldialdehyde; IMT¼ intima-media wall thickness. Kaya. IL-18 levels in PCOS. Fertil Steril 2010.
found by two-way ANOVA. This finding suggests that the in-fluence of PCOS on serum IL-6, CRP, MDA, and carotid IMT was not different in lean women compared with obese women, and that the influence of obesity on serum IL-6, CRP, and MDA concentrations was the same in the PCOS and control groups (F ¼ 0.19, P¼.6; F ¼ 1.0, P¼.2; F ¼ 0.07, P¼.7, respectively).
For Hcy, the interaction between PCOS and obesity was found in the two-way ANOVA variation analysis (F¼ 48.5, P<.001). Although there were no statistically significant dif-ferences in terms of the Hcy levels between the obese and nonobese control subjects (P¼.12), serum Hcy levels were higher in obese women with PCOS compared with in nonob-ese women with PCOS (P¼.01). Although there were no statistically significant differences in terms of Hcy levels be-tween nonobese PCOS and nonobese control groups (P¼.8), Hcy levels were higher in obese women with PCOS com-pared with obese control subjects (P<.001).
For carotid IMT, the interaction between PCOS and obe-sity was found in the two-way ANOVA variation analysis (F¼ 81.5, P<.001). Although there were no statistically sig-nificant differences in terms of the carotid IMT between
obese and nonobese control subjects (P¼.12), carotid IMT were higher in obese women with PCOS compared with the nonobese women with PCOS (P<.001). When the PCOS and control groups were compared within the obese and nonobese groups, carotid IMT was higher in patients with PCOS compared with controls (P<.001).
We then compared clinical, biochemical, and endocrine variables between patients with PCOS and control subjects. There were no statistically significant differences in the age, BMI, and serum FSH, LH, lipid profile, and serum MDA levels between the groups. Serum IL-18 levels, WHR, fasting insulin, HOMA-IR, Hcy, CRP, IL-6, total T, FAI, and carotid IMT were significantly higher in patients with PCOS compared with controls (for each parameter P<.05).
In the Spearman’s correlation analysis of the relationship between the parameters of cardiovascular risk and IL-18 levels in patients with PCOS, increased serum IL-18 levels is positively and significantly correlated, after log trans-formation with mean carotid IMT, BMI, WHR, fasting insulin, HOMA-IR, Hcy, and CRP (Table 2). No correla-tion was found among IL-18 levels, androgens, and MDA (Table 1).
The stepwise multiple linear regression method was used to determine the independent predictors that mostly affected IL-18. All of the procedures of the stepwise multiple linear regression analysis were completed when the fourth stage was reached. The study revealed that BMI, WHR, Hcy, and HOMA-IR are independent determinants of plasma IL-18 in patients with PCOS (Table 2). This model explains 65% of variation of plasma IL-18 (Table 2). The multiple linear re-gression stepwise method was used to determine the indepen-dent predictors, mostly carotid IMT, among inflammatory factors, androgens, lipid concentrations, and HOMA-IR. By multiple linear regression analyses, serum IL-18 levels are in-dependent determinants of carotid IMT in PCOS (b coeffi-cient 0.075, P<.001, 95% confidence interval [CI] 0.00– 0.095).
DISCUSSION
In this study it was found that increased serum IL-18 levels in patients with PCOS are positively and significantly corre-lated, after log transformation, with BMI, carotid IMT, WHR, fasting insulin, HOMA-IR, Hcy, and CRP. Stepwise multiple linear regression analysis revealed that BMI, WHR, Hcy, and HOMA-IR are independent determinants of plasma IL-18 in patients with PCOS. This model explains 65% of variation of plasma IL-18.
In this study, we confirmed that plasma concentrations of IL-18 are elevated in patients with PCOS when compared with healthy controls (214 102 vs. 170 78 mg/dL, P<.05). Increased risk of atherosclerotic heart disease has been reported in patients with PCOS in comparison with healthy controls(3, 30–32). Polycystic ovary syndrome is as-sociated with endothelial dysfunction, reduced vascular
FIGURE 1
Serum interleukin (IL)-18 levels in nonobese and obese patients with polycystic ovary syndrome (PCOS) and controls. Box and whisker plots depicting the IL-18 serum levels betwen PCOS and control groups. Solid lines inside boxes depict the median IL-18 level, whereas the upper and lower limits of the boxes and whiskers indicate 75th, 25th, 95th, and 5th percentiles. BMI¼ body mass index.
PCOS Control IL-18 (pg/ml) 1200 1000 800 600 400 200 0 -200 BMI Non-obese Obese
compliance, and premature carotid atherosclerosis. Higher serum IL-18 levels appear to be associated with greater ca-rotid IMT, suggesting the link between IL-18 and atheroscle-rosis (33). Serum IL-18 levels are associated with cardiovascular death in patients with coronary artery disease and with coronary events in apparently healthy men(13, 34). The risk of myocardial infarction has been reported to in-crease by 7.4-fold in women with PCOS(35). It is possible that these findings are due in part to increased IL-18 levels.
In the present study the serum IL-18 levels of patients with PCOS were higher than that of the control group (P<.05). The main effect of PCOS on IL-18 was statistically meaning-ful. In addition, the main effect of obesity on IL-18 was also statistically meaningful. Serum IL-18 levels of obese pa-tients, when compared with that of nonobese papa-tients, were higher both in PCOS and control groups (P<.01). Obesity had the same significance and meaningful effect on IL-18 levels in PCOS and control groups. Obesity was a factor to increase serum IL-18. However, in the cases where the BMI was at R30, when PCOS and control groups were com-pared, there was no difference among the groups in term of serum IL-18 levels. This fact leads to the possibility that the PCOS effect on serum IL-18 levels could be different in obese and nonobese women. This indicates that the meaning-ful effect of PCOS on serum IL-18 levels disappeared when obesity was taken into account. More recently, Escobar-Mor-reale et al.(9)reported that the PCOS effect on serum IL-18 levels in obese and lean groups was similar. However, in our study, although the PCOS effect on IL-18 disappeared after correction was made in accordance with BMI, the significant effect of obesity on IL-18 continued. This situation suggests that it is the obesity factor that has the actual effect on serum IL-18 levels rather than the PCOS. Thus, serum IL-18 levels
appear to be determined both by overall and by visceral adi-posity. Patients with PCOS are likely to have increased vis-ceral fat mass relative to BMI-matched controls, as the WHR is often greater in such women. Increased IL-18 levels may be produced by adipose tissue, especially by omental fat in patients with PCOS(11, 12). The data we obtained from this study supported this hypothesis.
Previously, Zhang et al.(36)show that serum IL-18 levels were significantly increased in women with PCOS and were associated with IR displayed by the euglycemic hyperinsuli-nemic clamp test. The PCOS women presented some de-creased insulin sensitivity (according to HOMA-IR) when compared with control subjects. Increased serum IL-18 levels in patients with PCOS is positively and significantly corre-lated, after log transformation, fasting insulin, and HOMA-IR. Therefore, the increase in IL-18 levels in patients with PCOS are probably related to IR.
At present an association between Hcy and IL-18 has not been investigated in patients with PCOS. We found a positive correlation between plasma concentration of IL-18 and Hcy in patients with PCOS. Hyperhomocysteinemia has been shown to be an independent risk factor for women with PCOS(36). In addition, an increase of plasma Hcy is associ-ated with IR in patients with PCOS(38). In that study, serum Hcy levels of patients with PCOS were higher than that of the control group (12.8 1.7 vs. 9.3 1.3 mmol/L, P<.01). In the study, based on the results of two-way ANOVA, the influ-ence of PCOS on serum Hcy levels was different in lean women compared with obese women. Hyperhomocysteine-mia has been shown to be an independent risk factor for women with PCOS(37, 39). Both Hcy and IL-18 seem to be involved in vascular inflammation and coronary artery disease.
TABLE 2
Stepwise MLR analysis of relationship between plasma IL-18 concentrations and selected variables in patients with PCOS.
95% Confidence interval for B MLR model Independent variables Coefficient of regression (B) P value Lower bound Upper bound Adjusted R2 Step 1 BMI 0.098 <.001 0.083 0.113 50.6% Step 2 BMI 0.073 <.001 0.058 0.088 61.7% WHR 0.083 <.001 0.067 0.099 Step 3 BMI 0.082 <.001 0.066 0.098 64.3% WHR 0.091 .005 0.077 0.105 HOMA-IR 0.001 .01 0 0.002 Step 4 BMI 0.101 <.001 0.163 0.04 65.0% WHR 0.066 <.001 0.052 0.081 HOMA 0.091 <.001 0.077 0.105 Hcy 0.077 <.001 0.062 0.093
Note: MLR¼ multiple linear regression; PCOS ¼ polycystic ovary syndrome; BMI ¼ body mass index; WHR ¼ waist-to-hip ratio; HOMA-IR¼ homeostasis model assessment of insulin resistance; Hcy ¼ homocysteine.
McLachlan et al. (40) demonstrated that plasma Hcy are positively associated with plasma IL-18 in coronary artery by-pass surgery patients with high plasma IL-18 concentration. Another study by Aso et al. (16)revealed that plasma Hcy levels were significantly higher in type 2 diabetic patients with high plasma IL-18 concentration than in those with nor-mal plasma IL-18 concentration. In that study, Hcy was an in-dependent determinant of plasma IL-18 in patients with PCOS. This situation suggests that hyperhomocysteine-mia—other than obesity and IR—could increase serum IL-18 levels in patients with PCOS. The possible contributing factor to cardiovascular risk increased due to hyperhomocys-teinemia could be the serum IL-18 increase that is caused by Hcy. With the previously performed studies (16, 40)from cited literature that support this hypothesis, the same possibil-ity for a contributing factor could be valid for PCOS as well. However, there is an absolute need for more specific, prospective studies focusing on the possible relation between Hcy and serum IL-18 in patients with PCOS.
The CRP may be directly involved in the atherogenic pro-cess by promoting endothelial dysfunction and complement activation (16, 41, 42). Kelly et al. (5) have shown that CRP concentration is significantly increased in women with PCOS. The IL-18 is a proinflammatory cytokine that induces the production of CRP. Serum CRP levels were higher in pa-tients with PCOS than in control subjects (4.4 1.6 vs. 1.5 0.6 mg/dL, P<.01). Based on the two-way ANOVA variation analysis, it is suggested that the influence of PCOS on serum CRP levels was not different in lean women compared with obese women and that the influence of obesity on serum CRP levels was the same in the PCOS and control groups. In the present study WHR and IR were associated with ele-vated CRP levels and eleele-vated CRP levels were positively correlated with serum IL-18. It is likely that an increase of CRP by IL-18 in patients with PCOS promotes inflammation and this enhances the development and progression of CRP-related atherosclerosis. In addition, CRP and IL-18 might have a synergic effect in women with PCOS.
An in vitro study has shown that IL-18 signaling mediated by IL-18 receptors on endothelial cells or on smooth muscle cells induced atherogenesis by the release of IL-6(7). Omen-tal adipose tissue produces 3-fold more IL-6 than subcutane-ous adipose tissue (43). Previously, Vgontzas et al. (44) showed that IL-6 levels had increased in patients with PCOS. Likewise, IL-6 levels were found to be higher in our study (5.1 1.4 vs. 2.7 1.1 mg/dL, P<.01). Serum IL-6 levels were positively and significantly correlated with IR. In our study, the increase in IL-6 levels found in PCOS is most probably related to IR. However, we could not determine a correlation between serum IL-18 and IL-6 levels by univar-iate and multivarunivar-iate analyses. This situation suggests a differ-ent interaction between serum IL-18 and IL-6 in patidiffer-ents with PCOS. Based on the two-way ANOVA variation analysis, no interaction was found between PCOS and obesity, suggesting that the influence of PCOS on serum IL-6 levels was not dif-ferent in lean women compared with obese women.
Serum FAI levels of patients with PCOS were higher than that of the control group (14.7% vs. 8.9%, P<.01). No corre-lation was found between IL-18 and FAI in patients with PCOS. Thus, elevation of plasma FAI levels may not enhance the serum levels of IL-18 in patients with PCOS. Patients with PCOS have been reported to have higher triglyceride levels and lower HDL cholesterol values (44, 45). The MDA is a marker of lipid peroxidation and it increases with oxidative stress states (20, 46, 47). Various studies re-ported an increased oxidative and decreased antioxidative status in women with PCOS(48, 49). We found lipid profile and MDA to be comparable and among PCOS IL-18 quartile subgroups. In the two-way ANOVA variation analysis, the in-fluence of PCOS on serum MDA levels was not different in lean women compared with obese women. In addition, no correlation was found between IL-18 and MDA in patients with PCOS. This suggests that elevation of plasma IL-18 levels may not enhance lipid peroxidation in patients with PCOS.
Although elevated IL-18 levels can predict the develop-ment of cardiovascular disease, their association with carotid IMT remains to be examined in patients with PCOS(13, 50). In the present study, the mean carotid IMT were higher in pa-tients with PCOS than control women (0.61 0.17 vs. 0.44 0.11 mm, P<.001). Elevated serum IL-18 levels were posi-tively and significantly correlated with a greater carotid IMT. By multiple linear regression analyses, serum IL-18 levels are independent determinants of carotid IMT in PCOS when adjusting traditional atherosclerotic risk factors. These findings support the association between IL-18 and ca-rotid atherosclerosis in patients with PCOS. In the two-way ANOVA variation analysis, the influence of PCOS on carotid IMT was different in lean women compared with obese women. Considering all factors, practioners treating PCOS cases should be aware of the risk of carotid atherosclerosis in lean patients with PCOS. This study was cross-sectionally designed and the causal relationships cannot be proven by cross-sectional data. A prospective, large, longitudinal study will be necessary to clarify between IL-18 and carotid athero-sclerosis.
In conclusion, the present study indicates that serum IL-18 levels are elevated and the synergistic effects of BMI and WHR, hyperinsulinemia, IR, and hyperhomocysteinemia most likely contribute to the elevation of plasma IL-18 con-centrations. We have demonstrated an association between elevated serum IL-18 levels and greater carotid IMT. These findings support the link between IL-18 and carotid athero-sclerosis in women with PCOS.
REFERENCES
1. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mech-anism and implications for pathogenesis. Endocrinol Rev 1997;18: 774–800.
2. Lobo RA, Carmina E. The importance of diagnosing the polycystic ovary syndrome. Ann Intern Med 2000;132:989–93.
3. Talbott EO, Guzick DS, Sutton-Tyrrell K, McHugh-Pemu KP, Zborowski JV, Remsberg KE, et al. Evidence for association between
polycystic ovary syndrome and premature carotid atherosclerosis in mid-dle-aged women. Arterioscler Thromb Vasc Biol 2000;20:2414–21. 4. Paradisi G, Steinberg HO, Hempfling A, Cronin J, Hook G, Shepard MK,
et al. Polycystic ovary syndrome is associated with endothelial dysfunc-tion. Circulation 2001;103:1410–5.
5. Kelly CCJ, Lyall H, Petrie JR, Gould GW, Connell JMC, Satar N. Low grade chronic inflammation in women with polycystic ovarian syn-drome. J Clin Endocrinol Metab 2001;86:2453–5.
6. Frishman WH. Biologic markers as predictors of cardiovascular disease. Am J Med 1998;104:18S–27S.
7. Gerdes N, Sukhova GK, Libby P, Reynolds RS, Young JL, Schonbeck U. Expression of interleukin (IL)-18 and functional IL-18 receptor on hu-man vascular endothelial cell, smooth muscle cells, and macrophages: implications for atherogenesis. J Exp Med 2002;195:245–57. 8. Bastard J-P, Jardel C, Delattre J, Hainque B, Bruckert E, Oberlin F.
Ev-idence for a link between adipose tissue IL-6 content and serum C-reac-tive protein concentrations in obese subjects. Circulation 1999;99: 2219–22.
9. Escobar-Morreale HF, Botella-Carretero JI, Villuendas G, Sancho J, San Millan JL. Serum interleukin-18 concentrations are increased in the polycystic ovary syndrome: relationship to insulin resistance and to obe-sity. J Clin Endocrinol Metab 2004;89:806–11.
10. Esposito K, Pontillo A, Ciotola M, Di Palo C, Grella E, Nicoletti G, et al. Weight loss reduces interleukin-18 levels in obese women. J Clin Endo-crinol Metab 2002;87:3864–6.
11. Blankenberg S, Tiret L, Bickel C, Peetz D, Cambien F, Meyer J, et al. In-terleukin-18 is a strong predictor of cardiovascular death in stable and unstable angina. Circulation 2002;106:24–30.
12. Tedgui A, Mallat Z. Antiinflammatory mechanism in the vascular wall. Circulation Res 2001;88:877–87.
13. Yudkin JS, Kumari M, Humphries SE, Mohamed-Ali V. Inflammation, obesity, stress and coronary heart disease: is interleukin-6 the link? Ath-erosclerosis 2000;148:209–14.
14. Boulman N, Levy Y, Leiba R, Schachar S, Linn R, Zinder O, et al. Increased C-reactive protein levels in the polycystic ovary syndrome: a marker of car-diovascular disease. J Clin Endocrinol Metab 2004;89:2160–5. 15. Clarke R, Daly L, Robinson K, Naughten E, Cahalane S, Fowler B, et al.
Hyperhomocysteinemia: an independent risk factor for vascular disease. N Engl J Med 1991;324:1149–55.
16. Aso Y, Okumura K, Takebayashı K, Wakabayashı S, Inukaı T. Relation-ship of plasma interleukin-18 concentrations to hyperhomocysteinaemia and carotid intimal-media wall thickness in patients with type 2 diabetes. Diabetes Care 2003;26:2622–7.
17. Tso TK, Huang WN, Huang HY, Chang CK. Relationship of plasma in-terleukin-18 concentrations to traditional and non-tradiotional cardiovas-cular risk factors in patients with systemic lupus erythematosus. Rheumatology 2006;45:1148–53.
18. Gutteridge JM, Halliwell B. The measurement and mechanism of lipid peroxidation in biological system. Trend Biochem Sci 1989;15:129–35. 19. Chambless LE, Heiss G, Folsom AR, Rosamond W, Szklo M, Sharett AR, et al. Association of coronary heart disease incidence with carotid arterial wall thickness and major risk factos: the Atherosclerosis Risk in Communities (ARIC) Study, 1987–1993. Am J Epidemiol 1997;146:483–94.
20. Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid intima-media thickness and risk of stroke and myocardial infarc-tion: the Rotterdam Study. Circulation 1997;96:1432–7.
21. Rotterdam ESHRE/ASRM Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term healthy risks related to polycystic ovary syndrome. Fertil Steril 2004;81:19–25.
22. Balen AH, Laven JS, Tan SL, Dewailly D. Ultrasound assessment of the polycystic ovary: international consensus definitions. Hum Reprod Up-date 2003;9:505–1.
23. Ferriman D, Gallway JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21:1440–7.
24. Ashwell M, Chinn S, Stalley S, Garrow JS. Female fat distribution— a simple classification based on two circumference measurements. Int J Obes 1982;6:143–52.
25. Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medial thickness of the arterial wall: a direct measurements with ultrasound im-aging. Circulating 1986;74:1399–406.
26. Wofford J, JKahl F, Howard G, McKinney W, Toole J, Crouse J. Relation of extent of extracranial carotid artery atherosclerosis as measured by B-mode ultrasound to the extent of coranary atherosclerosis. Arterio Thromb Vasc Biol 1991;11:1786–94.
27. Expert Committee on the Diagnosis and Classification. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care 2003;22:141–6. 28. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Tracher DF,
Turner RI. Homeostasis model assessment: insulin resistance and b cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9.
29. Belli SH, Graffigna MN, Oneta A, Otero P, Schurman L, Levalle OA. Ef-fect of rosiglitazone on insulin resistance, growht factors, and reproduc-tive disturbances in women with polycystic ovary syndrome. Fertil Steril 2004;81:624–9.
30. Guzick DS, Talbott EO, Sutton-Tyrrell K, Herzog HC, Kuller LH, Wolfson SK. Carotid atherosclerosis in women with polycystic ovary syndrome: initial results from a case-control study. Am J Obstet Gynecol 1996;174:1224–9.
31. Wild RA. Polycystic ovary syndrome: a risk for coronary artery disease? Am J Obstet Gynecol 2002;180:35–43.
32. Meyer C, McGrath BP, Teede HJ. Overweight women with polycystic ovary syndrome have evidence of subclinical cardiovascular disease. J Clin Endocrinol Metab 2005;90:5711–6.
33. Yamagami H, Kitagawa K, Hoshi T, Furukado S, Hougaku H, Nagai Y, et al. Associations of serum IL-18 levels with carotid intima-media thick-ness. Arterioscler Thromb Vasc Biol 2005;25:1458–62.
34. Mallat Z, Corbaz A, Scoazec A, Besnard S, Leseche G, Chvatchka Y, et al. Expression of interleukin-18 in human atherosclerotic plaques and relation to plaque instability. Circulation 2001;104:1598–603. 35. Dahlgren E, Janson PO, Johansson S, Lapidus L, Oden A. Polycystic
ovary syndrome and risk for myocardial infarction. Evaluated from a risk factor model based on a prospective population study of women. Acta Obstet Gynecol Scand 1992;71:599–604.
36. Zhang YF, Yang YS, Hong J, Gu WQ, Shen CF, Xu M, et al. Elevated serum levels of interleukin-18 are associated with insulin resistance in women with polycystic ovary syndrome. Endocrine 2006;29:419–23. 37. Yaralı H, Yıldırır A, Aybar F, Kabakcı G, B€uk€ulmez O, Akg€ul E, et al.
Diastolic dysfunction and increased serum homocysteine concentrations may contribute to increased cardiovascular risk in patients with polycys-tic ovary syndrome. Fertil Steril 2001;76:511–6.
38. Schacter M, Raziel A, Friedler S, Starssburger D, Bern O, Ron-El R. In-sulin resistance in patients with polycystic ovary syndrome is associated with elevated plasma homocysteine. Hum Reprod 2003;18:721–7. 39. Nygard O, Nordrehaug JE, Refsum H, Ueland PM, Farstad M,
Vollset SE. Plasma homocysteine and mortality in patients with coronary artery disease. N Engl J Med 1997;337:230–6.
40. McLachlan CS, Chua WC, Wong PT, Kah TL, Chen C, El Oakley RM. Homocysteine is positively associated with cytokine IL-18 plasma levels in coronary artery bypass surgery patients. Biofactors 2005;23: 69–73.
41. Lagrand WK, Visser CA, Hermens WT, Niessen HW, Verheugt FW, Wolbink GJ, et al. C-reactive protein as a cardiovascular risk factor: more than an epiphenomenon? Circulation 1999;100:96–102. 42. Pasceri V, Willerson JT, Yeh ETH. Direct proinflammatory effect of
C-reactive protein on human endothelial cells. Circulation 2000;102: 2165–8.
43. Fernandez-Real JM, Vayreda M, Richart C, Gutierrez C, Broch M, Vendrell J, et al. Circulating interleukin 6 levels, blood pressure, and in-sulin sensitivity in apparently healthy men and women. J Clin Endocrinol Metab 2001;86:1154–115.
44. Vgontzas AN, Trakada G, Bixler EO, Lin HM, Pejoric S. Plasma inter-leukin-6 levels are elevated in polycystic ovary syndrome independently of obesity or sleep apnea. Metabolism 2006;55:1076–8.
45. Wild RA. Obesity, lipids, cardiovascular risk and androgen excess. Am J Med 1995;98:27S–32S.
46. Bozlan AD, Bianchi MS, Bianchi NO. Superoxide dismutase and gluta-thione peroxidase activities in human blood: influence of sex, age and cigarette smoking. Clin Biochem 1997;30:449–54.
47. Betteridge DJ. What is the oxidative stress? Metabolism 2000;49(Suppl 1):13–8.
48. Sabuncu T, Vural H, Harma M, Harma M. Oxidative stress in polycystic ovary syndrome and its contribution to the risk of cardiovascular disease. Clin Biochem 2001;34:407.
49. Fenkci V, Fenkci S, Yilmazer M, Serteser M. Decreased total antioxidant status and increased oxidative stress in women with polycystic ovary syndrome may contribute to the risk of cardiovascular disease. Fertil Steril 2003;80:123–7.
50. Blankenberg S, Luc G, Ducimetiere P, Arveiler D, Ferrieres J, Amouyel P, et al. for the PRIME Study Group. IL-18 and the risk of coronary heart disease in European men: the Prospective Epidemiological Study of Myo-cardial Infarction (PRIME). Circulation 2003;108:2453–9.
