Effect of Lifestyle Modification and Metformin Therapy on Emerging Cardiovascular Risk Factors in Overweight Indian Women with Polycystic Ovary Syndrome

Abstract

Introduction:

Polycystic ovarian syndrome (PCOS) is common among women of reproductive age. Although traditional cardiac risk factors are known to be altered and improved with short-term metformin therapy, not much is known about novel cardiac risk factors.

Objective:

The aim of this study was to evaluate the effects of lifestyle modification and short-term metformin therapy on the fasting serum lipids, homeostasis model assessment of insulin resistance (HOMA-IR), serum high-sensitivity C-reactive protein (hsCRP), and serum homocysteine.

Methods:

Native overweight [body mass index (BMI) >23 kg/m²] Indian women diagnosed with PCOS were evaluated and subjected to an oral glucose tolerance test and determination of insulin, homocysteine, hsCRP, and fasting lipids levels. They were started on maximally tolerated doses of metformin along with lifestyle modification. Following 3 months of therapy, they were resampled.

Results:

Out of 36 consecutive patients included, 25 women completed 3 months of metformin treatment and were eligible for repeat evaluation. The age of study group was 22.2±5 years. Twenty-two (61%) women were obese (BMI >25 kg/m²). Improvement was seen in body weight, BMI, serum total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDL-C), hsCRP, and serum testosterone on metformin therapy. However, no improvement was seen in serum fasting insulin, HOMA-IR, or homocysteine.

Conclusion:

Serum hsCRP improved with lifestyle modification and metformin therapy for 3 months in overweight subjects from India with PCOS, along with serum total cholesterol, triglycerides, and HDL-C. However, markers of insulin resistance and serum homocysteine did not improve.

Introduction

P olycystic ovary syndrome (PCOS) is one of the commonest endocrine diseases of women in the reproductive age group, affecting up to 10% of women in that age group.¹ It is characterized by hyperandrogenism, chronic anovulatory cycles, and oligomenorrhea or amenorrhea¹. Moreover, it is often associated with selective insulin resistance, leading to metabolic alterations.² Insulin resistance occurs in approximately 60%–80% of women with PCOS, increasing to 95% among obese women with PCOS.³ Selective insulin resistance and obesity act synergistically to cause PCOS.² Up to 40% of women with classic PCOS develop impaired glucose tolerance (IGT) or type 2 diabetes mellitus (T2DM) by the fourth decade of life,⁴ with age and weight gain worsening glycemic control.^4,5 Metabolic complications associated with insulin resistance are also increased in PCOS independent of obesity, including an adverse cardiovascular risk profile (impaired fibrinolysis, endothelial dysfunction, dyslipidemia, and subclinical atherosclerosis). Indeed, studies point to an increased incidence of cardiovascular complications in PCOS.^6

–10 High-sensitivity C-reactive protein (hsCRP) and homocysteine, the newly described "nonclassical coronary risk factors," are known to be elevated in PCOS, and such elevation has been linked to coronary artery disease.^{11

–18}

Metformin is an insulin sensitizer with known beneficial effects on insulin sensitivity and atherogenic dyslipidemia, in addition to having beneficial effects on ovulation and regularization of cycles.^19
–21 However, its effect on nonclassical coronary risk factors (homocysteine and hsCRP) has not been adequately documented, particularly in Indians. Indeed, some reports suggest that whereas hsCRP levels were lowered on metformin monotherapy,^16,22,23 there is concern that homocysteine levels may be slightly elevated.^24
–26 Therefore, it was considered necessary to study multiple classical and nonclassical coronary disease risk factors to understand the interaction and changes in these parameters before and after treatment with metformin. With this in mind, we carried out a study to document changes in classical and nonclassical coronary disease risk factors in Indian PCOS women before and after treatment with metformin. Our objective was to study the effect of lifestyle modification and short-term metformin therapy on the fasting serum lipids, homeostasis model assessment of insulin resistance (HOMA-IR), serum hsCRP, and serum homocysteine in normoglycemic overweight women from India with PCOS.

Materials and Methods

Women aged 18–40 years, having a body mass index (BMI) >23 kg/m² and diagnosed to be having PCOS as defined by National Institutes of Health (NIH) criteria,¹ were included in this prospective study conducted in the Department of Endocrinology, Sri Venkateswara Institute of Medical Sciences, Tirupati, India, after obtaining their informed consent. The study was approved by institute's ethics committee. PCOS was diagnosed in women with oligomenorrhea (eight or less menstrual periods per year) or amenorrhea (no menstrual periods for 3 or more months) having clinical (modified Ferriman–Gallway score²⁷ of more than 8) or biochemical hyperandrogenism (serum testosterone levels above 1.5 ng/mL) after exclusion of related disorders with similar presentation. Women with virilization, pregnancy, hypothyroidism [thyroid-stimulating hormone (TSH) >5 μIU/dL], or hyperprolactinemia, and those who had a history of the use of oral contraceptive, glucocorticoid, antiandrogen, ovulation induction agent, antidiabetic, or antiobesity drug within the previous 6 months were excluded from the study. Women with impaired fasting glucose (>110 mg/dL) or impaired glucose tolerance (post oral glucose tolerance test, 2-h plasma glucose >140 mg/dL) or cases of overt diabetes as defined by the World Health Organization (WHO)²⁸ were also excluded from the study.

A detailed history was taken, and a physical examination including height, weight (BMI calculated accordingly), and blood pressure was performed. At baseline, blood samples of all patients were collected for measurement of serum testosterone, prolactin, thyrotropin, lipid profile, fasting glucose, insulin, hsCRP, and plasma homocysteine. All blood samples were obtained in the morning between 0800 and 0900 h after an overnight fast. Each subject underwent a WHO standard oral glucose tolerance test, and blood samples for plasma glucose were obtained before and after 120 min of glucose administration.²⁸

All blood samples that were not being analyzed immediately were stored at −20°C until assayed. Venous blood samples for homocysteine were immediately packed in ice and centrifuged within 30 min to avoid release of homocysteine from red blood cells. Plasma samples were then stored at −20°C.

Women were advised to take oral metformin for 3 months. The dose of the drug was increased stepwise, from 500 mg once daily in first week to 500 mg twice daily (b.i.d.) the next week, and then to 1,000 mg b.i.d. Those who were not able to tolerate this dose were kept at the maximally tolerated dose. Patients were evaluated by a qualified dietician and were advised a 500-kcal deficit diet. Patients were also advised to walk for 30–45 min daily for at least 5 days per week. Repeat samples for fasting glucose, lipid profile, insulin, hsCRP, and homocysteine were collected after 3 months of metformin therapy.

Biochemical assays

Thyrotropin (TSH) was measured using immunoradiometric kits (IRMAK-9, Bhabha Atomic Research Center, Mumbai). The minimum and maximum detectable limits were 0.15–50 mU/L. Serum for prolactin was collected as a pooled sample, and levels were determined by an immunoluminometric sandwich assay using a Liaison Prolactin (code 312151) kit (DiaSorin, USA). Serum total testosterone was measured using radioimmunoassay (RIA) kits (RIAK-16 Testosterone kit, BRIT, BARC, Mumbai). Normal values for women were <1.5 ng/mL, as mentioned in the kit insert. The sensitivity of the assay was 0.19 ng/mL and avidity was 1.0×10¹⁰L/M. Intraassay coefficient of variability (CV) was 7.7% and 7.1%, whereas interassay CV was 9.7% and 9.4% for pool I and pool II sera, respectively.

Cortisol levels were measured by the Clinical Assays^™ GammaCoat^™ Cortisol ¹²⁵I RIA Kit (DiaSorin, USA). Serum insulin levels were measured by enzyme-linked immunosorbent assay (ELISA) using a DIAsource Ins-Easia kit (DIAsource ImmunoAssays S.A., Nivelles, Belgium). The detection limit of the test is 0.17 μIU/mL. Intraassay CV of the kit was 6%, whereas the interassay CV was 9%. Plasma glucose levels were measured by the glucose oxidase–peroxidase method. Impaired glucose tolerance and diabetes were diagnosed using WHO criteria.²⁸

Serum total cholesterol was measured by the cholesterol oxidase–peroxidase method, triglycerides were measured by the glycerol phosphate oxidase–peroxidase method, and high-density lipoprotein cholesterol (HDL-C) was measured by the immuno-inhibition method. Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedwald formula.²⁹ Intraassay CV was 1.8%, 4.6%, and 3.9%, whereas interassay CV was 2.4%, 1.4%, and 4.2% for serum total cholesterol, triglycerides, and HDL-C, respectively.

Insulin resistance was assessed by the HOMA, and the following index was calculated. HOMA IR=Fasting immunoreactive insulin (μU/mL)×fasting blood sugar (mg/dL)/405.³⁰ Serum hsCRP was determined by turbidimetric immunoassay using a CRP U2A-000 kit (APTEC Diagnostics nv, Belgium) with a CX9 autoanalyzer. The measuring range of the kit was 0–14 mg/L, and the detection limit was 0.013 mg/dL with no hook effect. The intrarun CV was 3.63%, 3.15%, 1.61%, and 1.66%, whereas the interrun CV was 4.23%, 3.84%, 2.41%, and 2.08% for ultralow, low, medium, and high controls, respectively.

Homocysteine levels were determined by an enzymatic assay using a commercial kit from CATCH Diagnostics Pvt. Ltd, London. The measuring range of the kit was from 2 to 50 μmol/L. The limit of quantification (CV <20%) was 1.0 μmol/L. Within-run precision was 8%, 7%, and 8%, whereas total precision was 10%, 9%, and 10% for low, medium, and high controls, respectively. We defined hyperhomocysteinemia as plasma homocysteine of more than 15 μmol/L.^31,32

All RIA estimations were performed using a Wallac 1480 automated gamma counter. The insulin ELISA was performed on a Chemwell automated analyzer (Awareness Technologies). All other parameters were assayed on Cx9 fully automated random access clinical chemistry analyzer from Beckman Coulter.

Statistical analysis

The mean and standard deviation (SD) for various anthropometric variables, blood pressure, and biochemical and hormonal parameters were calculated both at baseline and after 3 months of treatment, and changes in these continuous variables at the two points in time were tested for significance using the paired t-test on SPSS 13.0 for Windows (release 13.0, SPSS Inc., Chicago, IL). Changes in the proportion of abnormal results at the two time points were likewise tested for significance using the chi-square test. Correlations between various baseline parameters were assessed by measuring the Pearson coefficient of correlation (r). Variables that were not distributed normally (testosterone, HOMA, hsCRP, and homocysteine) were log transformed. The level of significance for all these determinations was set at a P value<0.05.

Results

Thirty-six consecutive patients with confirmed PCOS were included in the study. Twenty-five women completed 3 months of metformin treatment and were eligible for repeat evaluation. Of the remaining 11 patients, 5 women required oral contraceptive pills (OCP) for regularization of menstrual periods, 2 women asked for hirsutism management and were consequently started on OCP and spironolactone, and 4 women were lost to follow-up.

Baseline clinical features

The age (mean±SD) of the study group was 22.2±5 years (min-max 18–36 years) with a BMI of 28.9±6.2 kg/m². Twenty-two (61%) women were obese by Asian criteria (BMI >25 kg/m²). Mean systolic blood pressure was 112.2±8.3 mmHg, whereas diastolic blood pressure was 73.6±4.9 mmHg. All 36 women had menstrual irregularities in the form of either oligomenorrhea (78%, n=28) or amenorrhea (22%, n=8). Twenty-eight percent (n=10) of women had a history of acne, 17% (n=6) had hirsutism, 53% (n=19) had family history of T2DM in a first-degree relative, and 83% (n=30) had acanthosis nigricans.

Results of biochemical investigations are shown in Table 1. Forty-two percent (n=15) of patients had triglyceride levels more than 150 mg/dL, 17% (n=6) had total cholesterol levels more than 200 mg/dL, all patients had HDL-C less than 50 mg/dL, and 17% (n=6) had LDL-C more than 130 mg/dL. Two-hour post glucose load plasma glucose correlated positively with total cholesterol (r=0.476, P=0.003) and triglycerides (r=0.479, P=0.03). Mean HOMA-IR was 4.31±2.25, and 29 (81%) women were insulin resistant (HOMA-IR >2.4). HOMA-IR correlated positively with serum triglycerides (r=0.367, P=0.027). There was no significant difference in the mean testosterone, HOMA-IR, hsCRP, and homocysteine at baseline between women who had completed study and those who did not (Table 2).

Table 1.

Biochemical Parameters at Baseline

Parameter	Total group (n=36)
FBS (mg/dL)	86.6±7.9
PGBS (mg/dL)	120.9±15.9
Cholesterol (mg/dL)	171.9±34.5
TGL (mg/dL)	152.3±68.3
VLDL-C (mg/dL)	30.5±13.6
HDL-C (mg/dL)	36.6±3.9
LDL-C (mg/dL)	103±29.9
Testosterone (ng/mL)	2±0.75
TSH (μIU/mL)	2.9±1.7
Prolactin (ng/mL)	9.9±3.4
Insulin (μIU/mL)	20±10.5
HOMA-IR	4.3±2.2
hsCRP (mg/L)	4.7±2.4
Homocysteine (μmol/L)	19.5±7.6

According to National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) criteria normal values: FBS <100 mg/dL, PGBS <140 mg/dL, total cholesterol <200 mg/dL, TGL <150 mg/dL, HDL-C >50 mg/dL, and LDL-C <130 mg/dL. Normal values for TSH: 0.1–5 μIU/mL, prolactin <25 ng/mL, HOMA-IR <2.4, hsCRP <1mg/L, and homocysteine <15 μmol/L.

FBS, fasting blood sugar; PGBS, postglucose blood sugar; cholesterol, total cholesterol; TGL, triglycerides; VLDL-C, very-low-density lipoprotein; HDL-C, high-density lipoprotein; LDL-C, low-density lipoprotein; testosterone, serum testosterone; TSH, serum thyroid-stimulating hormone; HOMA-IR, homeostatis model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein.

Table 2.

Comparison of Baseline Data

Parameter	Group 1 (n=25)	Group 2 (n=11)	P value
Age (years)	22.4±5.35	21.7±4.43	NS
BMI (kg/m²)	28±4.6	31±8.5	NS
SBP (mmHg)	112±7	112.7±11	NS
DBP (mmHg)	74±5	72.7±4.6	NS
FBS (mg/dL)	86.6±7.4	86.7±9.5	NS
PGBS (mg/dL)	120.8±16.6	121.2±9.5	NS
Total cholesterol(mg/dL)	173.5±33	168.3±15	NS
Triglycerides (mg/dL)	165.8±75.8	121.8±39	NS
HDL-C (mg/dL)	36.2±3.9	37.8±3.9	NS
LDL-C (mg/dL)	102.7±25.6	104±41	NS
Testosterone (ng/mL)	2.1±0.8	2.1±0.6	NS
HOMA-IR	3.9±1.8	5.2±2.3	NS
hsCRP (mg/L)	4.7±2.4	4.7±2.2	NS
Homocysteine (μmol/L)	20±7.2	18±8.6	NS

Group 1, those who had followed up; group 2, those women who lost to follow-up.

BMI, body mass index; NS, not significant; SBP, systolic blood pressure; DBP, diastolic blood pressure; PGBS, postglucose blood sugar; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; HOMA-IR, homeostasis model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein.

Serum hsCRP

Of 36 women, 4 (11%), 5 (14%), and 27 (75%) had hsCRP levels of <1, 1–3, and more than 3 mg/L, respectively. Serum hsCRP correlated positively with BMI (r=0.479, P=0.003) and TSH (r=0.422, P=0.01) but not with HOMA-IR (P=0.73).

Serum homocysteine

Mean homocysteine at baseline was 19.5±7.6 μmol/L. A majority (24 out of 36; i.e., 67%) of patients had homocysteine levels more than 15 μmol/L. Serum homocysteine did not correlate with either weight, BMI, or HOMA-IR. There was no significant difference in the baseline biochemical parameters among overweight (BMI 23–24.9 kg/m²) and obese (BMI ≥25 kg/m²) women.

Changes after metformin treatment

Out of the 36 women included, 25 women completed 3 months of metformin therapy and were eligible for repeat evaluation. All women tolerated the drug well except for mild gastric irritation in some. Most of them used 2 grams of metformin daily, mean dose being 1.6±0.5 gram/day. Twelve women took 1 gram of metformin per day, another 12 took 2 grams per day, and only 1 woman took 1.5 grams of metformin per day. Menstrual cycles became regular in 14 out of 25 patients (56%). There was improvement in BMI (27.9±4.6 kg/m² at baseline vs. 27.4±4.8 kg/m² on follow up, P=0.02) and weight (67.8±12 kg at baseline vs. 66.5±12 kg on follow up, P=0.016).

Changes in biochemical parameters before and after use of metformin are given in Table 3. Following treatment, there was a reduction in fasting glucose, total cholesterol, very-low-density lipoprotein cholesterol (VLDL-C) and triglycerides along with an increase in HDL-C. No change was observed in LDL-C. Testosterone fell significantly after treatment. Serum hsCRP decreased with metformin therapy (P<0.001). Although homocysteine did not fall with treatment, the proportion of patients with homocysteine more than 15 μmol/L was reduced (76% before treatment vs. 64% after treatment, P<0.001). There was no difference in the any of the measured biochemical indices, either at baseline or after treatment, between the women who had regular menstrual cycles after treatment and those who did not (data not shown).

Table 3.

Comparison of Biochemical Parameters (n=25)

Parameter	Before treatment	After treatment	P value
FBS (mg/dL)	86.6±7.4	83±6.9	0.006^a
Cholesterol (mg/dL)	173.5±33	162±30.5	0.026^b
TGL (mg/dL)	165.8±75.8	125.6±45.5	<0.001^a
VLDL-C (mg/dL)	33±15	24.9±8.9	<0.001^a
HDL-C (mg/dL)	36±3.9	39.5±5.7	0.007^a
LDL-C (mg/dL)	102.6±24.6	96.8±25.8	NS
Testosterone (ng/mL)	2.2±0.8	1.4±0.7	<0.001^a
Insulin (μIU/mL)	18.4±8.7	19.9±9.4	NS
HOMA-IR	3.9±1.8	4.0±2.0	NS
hsCRP (mg/L)	4.6±2.5	2.5±1.7	<0.001^a
Homocysteine, (μmol/L)	20±7.3	18.2±8.9	NS

According to National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) criteria total cholesterol <200 mg/dL, TGL <150 mg/dL, HDL-C >50 mg/dL, and LDL-C <130 mg/dL. Normal values for TSH: 0.1–5 μIU/mL, prolactin <25 ng/mL, HOMA-IR <2.4, hsCRP <1mg/L, and homocysteine <15 μmol/L.

Significance P<0.01.

Significance P<0.05.

FBS, fasting blood sugar; PGBS, postglucose blood sugar; cholesterol, total cholesterol; TGL, triglycerides; VLDL-C, very-low-density lipoprotein; HDL-C, high-density lipoprotein; LDL-C, low-density lipoprotein; NS, not significant; testosterone, serum testosterone; TSH, serum thyroid-stimulating hormone; HOMA-IR, homeostasis model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein.

Discussion

Literature suggests that women with PCOS without any other apparent chronic disease may be at increased risk for cardiovascular disease (CVD) compared with normal cycling women of similar age and BMI.^7
–9 The present study was a prospective study conducted on 36 consecutive patients with a BMI more than 23 kg/m², who had newly diagnosed PCOS. It was planned to document the baseline prevalence of various classical and novel cardiac risk factors and the changes with metformin therapy.

Thirty-nine percent (n=14) of women in our study were overweight (BMI 23–25 kg/m²) and 61% (n=22) were obese (BMI >25 kg/m²). Obesity is observed in approximately 60% among patients with PCOS.³³ After metformin therapy, coupled with instructions on calorie restriction and exercise, weight and BMI decreased in our study. Our results are similar to those obtained by Tan et al., who prospectively assessed the effect of 6 months of metformin treatment in 188 PCOS patients and 102 controls, divided into three groups according to BMI (lean, BMI <25 kg/m²; overweight, BMI 25–29 kg/m²; and obese, BMI >30 kg/m²).³⁴ In the overweight and obese PCOS groups, metformin therapy showed improvements in BMI. In another trial by Harborne et al. specifically designed to evaluate the effects of metformin therapy on body weight, BMI decreased by approximately 4% at 8 months (P<0.0001) in both obese and morbidly obese women with PCOS after metformin therapy.³³ Furthermore, in a large multicenter trial of more than 600 women with PCOS receiving metformin alone or in combination with combination oral contraceptive pills, a significant reduction in BMI was noted following metformin treatment.³⁵

HOMA-IR

A HOMA-IR cutoff of 2.4 (mean+2 SD) was used based upon data published by Viswanathan et al., in which the normal cutoff value for insulin resistance was derived from normoglycemic control subjects from a similar geographic area and population. Subjects having a HOMA-IR value of above 2.4 were considered to have insulin resistance.³⁶ Eighty percent of our patients (both overweight and obese) had evidence of insulin resistance with a HOMA-IR value more than 2.4, which is in accordance with the high prevalence of insulin resistance (60%–80%) described in PCOS.³ In an Indian study, insulin resistance was seen in 76.9% of women suffering with PCOS, similar to our study.³⁷

It has been shown that metformin therapy reduces serum insulin and HOMA-IR in overweight and obese women with PCOS having insulin resistance.^38,39 However, in our study, there was no significant improvement in HOMA-IR with metformin treatment. Acbay et al. also found that metformin does not decrease insulin resistance in PCOS and suggested that the cellular mechanism of insulin resistance is different from other common insulin-resistant states, such as type 2 diabetes mellitus and obesity.⁴⁰ Likewise, in another study by Ehrmann et al., 14 obese, nondiabetic women with PCOS (BMI pretreatment, 39.0±7.7 kg/m²; posttreatment, 39.1±7.9 kg/m²) were studied for effect of metformin therapy (850 mg, orally, three times daily for 12 weeks) on insulin levels and insulin sensitivity.⁴¹ These investigators found that insulin sensitivity was not improved by metformin, and they concluded that hyperinsulinemia and androgen excess in obese nondiabetic women with PCOS are not improved by the administration of metformin. The reasons for conflicting results are not clear. A possible explanation is that only about 50% of women with PCOS respond to metformin,⁴² and this could be why some studies (including ours) did not report benefit from metformin therapy. Tissue specificity of insulin resistance in PCOS may be another explanation, and HOMA-IR is not a sensitive marker of tissue-specific insulin resistance.⁴²

Lipid abnormalities

In our study, 42% patients had triglyceride levels >150 mg/dL, all patients had HDL-C <50 mg/dL, and 17 % had LDL-C >130 mg/dL. Thus, an atherogenic lipid profile was present in the majority of patients at baseline. There were significant reductions in serum total cholesterol, serum triglycerides, and elevation in serum HDL-C with metformin therapy. However, no change was seen in serum LDL-C with metformin therapy.

Different beneficial effects are reported on dyslipidemia in PCOS women when treated with metformin.^19
–21 Metformin therapy has been shown to decrease plasma total cholesterol, LDL-C, and triglycerides,^19
–21 as well as to increase serum HDL-C concentrations.^19,20 The results of our study were similar to reported effects except for LDL-C. Our patients did show a trend toward reduction in LDL-C, which did not reach significance. This might be due to small sample size. The prevalence of increased LDL-C in PCOS is generally not so high as that of other components of atherogenic lipoprotein phenotype, ranging from 24% to 40%, and is less dependent on body weight.⁴³

Serum hsCRP

CRP seems to be a more potent independent predictor of cardiovascular events than LDL-C, and it adds prognostic information at all levels of calculated Framingham risk and at all levels of the metabolic syndrome.¹² CRP was a strong predictor of CVD, even several years after the initial testing.¹⁴ We classified patients into three groups: <1, 1–3, and >3 mg/L, because these represent approximately the tertile values of hsCRP in the population.²⁸ The highest tertile has been shown to be having a two-fold higher risk of cardiovascular events than the lowest.¹³ Out of 36 women in our study, 11%, 14%, and 75% had hsCRP levels of <1, 1–3, and more than 3 mg/L, respectively. There was a significant fall in hsCRP levels after metformin treatment. Previous studies by Kelly et al. on serum hsCRP in PCOS women demonstrated an increased prevalence of high hsCRP in patients with PCOS, compared with controls (geometric means, 2.12 and 0.67 mg/L, respectively; P=0.016).¹⁵ Verit studied 52 normoinsulinemic women with PCOS who did not have metabolic syndrome, and they demonstrated increased hsCRP in the study group compared to the control group.⁴⁴ In a study by Boulman et al., among 116 PCOS patients and 94 controls, elevated serum CRP levels were seen in patients with PCOS compared to controls. The mean was 5.46±7.0 mg/L in the PCOS group versus 2.04±1.9 mg/L in the control. They used the cutoff of CRP >3 mg/L as a risk factor for CVD and suggested that CRP levels may be considered among the battery of tests for PCOS patients as a possible risk factor analysis for future cardiovascular events.⁴⁵ In another study by Guzelmeric et al. that included 44 PCOS women and 26 healthy controls matched by age and BMI, serum CRP was elevated in women with PCOS compared with controls.⁴⁶

In our study, serum hsCRP correlated with BMI and TSH but not with HOMA-IR. The positive correlation with BMI is consistent across several studies.^44,46 Metformin treatment significantly reduced hsCRP levels in our study. Our results are in accordance with clinical data, confirming the effects of metformin on chronic inflammatory markers. Metformin administration has been shown to improve CRP, leukocyte count,^16,22 as well as serum hsCRP and soluble vascular cell adhesion molecule levels in PCOS patients.¹⁶ Diamanti-Kandarakis et al. studied 62 women with PCOS and 45 normal women of similar age, BMI, and waist-to-hip ratio (WHR). Twenty-two women with PCOS received 1,700 mg of metformin daily for 6 months. They found higher plasma hsCRP in the PCOS group compared with controls and a significant reduction in the same after 6 months of metformin administration (pretreatment 1.92±0.60 mg/L vs. posttreatment 0.52±0.26 mg/L, P=0.005).¹⁶ The decrease in serum CRP during metformin therapy is in agreement with the known beneficial metabolic effects of this drug, and it suggests that CRP or other inflammatory parameters could be used as markers for the efficiency of metformin therapy in PCOS.¹⁶ The reduction in circulating levels of CRP with metformin therapy has also been detected in obese women with PCOS.²³ Our study replicates these findings in the Indian population.

Serum homocysteine

Elevated plasma homocysteine has adverse effects on the cardiovascular system, including the enhanced oxidation of LDL, proliferation of smooth muscle cells, increased platelet adhesiveness, and endothelial cytotoxicity.⁴⁷

In our patients a large number of women had homocysteine levels >15 μmol/L. There was no correlation of homocysteine with HOMA-IR or insulin levels in our patients. There was a small reduction in homocysteine levels with metformin therapy that, however, did not reach significance in our study; but proportion of women with hyperhomocysteinemia decreased with metformin therapy.

In the study by Heutling et al., metformin treatment likewise decreased homocysteine levels (decreased from 9.3±1.7 μmol/L to 7.2±1.4 μmol/L, P=0.001), despite there being no difference in the baseline homocysteine levels between women with PCOS and controls.⁴⁸ In contrast, other authors have found elevated homocysteine levels in women with PCOS, but no effect of metformin treatment.⁴⁹ Stefano Palomba et al. showed that metformin exerts a slight but significant deleterious effect on serum homocysteine in patients with PCOS, and supplementation with folate is useful to increase the beneficial effect of metformin on the vascular endothelium.²⁴ This was corroborated by Vrbikova et al., who reported that metformin treatment (1,000 mg/day) in 9 women with PCOS increased homocysteine levels (10.1±2.6 to 13.4±5.1 μmol/L, P<0.05).²⁵ Kilicdag et al. found the same results with metformin (850 mg/day for 3 months) treatment in a similar study conducted on 15 women with PCOS. Serum homocysteine levels were found to be significantly elevated in women with PCOS compared with healthy women (P<0.005), and homocysteine levels increased from 8.93±0.49 to 11.26±0.86 μmol/L, P=0.002.²⁶ Dhanalakshmi et al. studied the effect of metformin therapy in women with PCOS on serum lipids and homocysteine. They found a significant reduction in BMI, testosterone, total cholesterol, and triglycerides and a significant increase in homocysteine in the metformin-treated group compared to untreated group with PCOS.⁵⁰ In our study, the absence of any rise in homocysteine levels in our Indian patients following metformin treatment was reassuring.

We also found that obese women with PCOS usually have a strong family history of T2DM and have increased cardiac risk as evidenced by abnormal lipid profile, high insulin resistance, and elevated nonclassical cardiac risk factors like hsCRP and homocysteine. Administration of metformin in tolerable doses to these women causes improvement in body weight, androgen levels, lipid abnormilities, CRP levels, and the proportion of women with hyperhomocysteinemia. However, it has no effect on insulin resistance in the short term. Long-term randomized outcome studies in large groups of women with PCOS are required to determine whether these improvements in risk factors translate into reductions in cardiovascular events or not.

Our study included relatively young subjects (22.2±5 years). With the natural decline in insulin sensitivity with age, it is possible that our findings would have been different in an older cohort of women with PCOS. However, PCOS is more of a problem in young women who are concerned about ovulatory and cosmetic issues rather than older women. Thus, our patient population reflects the usual age group of women seeking treatment for PCOS. Moreover, it is more meaningful to study effects of treatment on cardiac risk factors in young women without established atherosclerosis (so as to prevent the same) rather than older women who may already be having well established lesions.

One limitation of our study is the assay used to measure testosterone and cortisol, which was RIA. RIA for testosterone is not specific, especially at lower concentrations, as seen in women and children.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Zawadski

, Dunaif

. Diagnostic criteria for polycystic ovary syndrome: Towards a rational approach. Dunaif

, Givens

, Haseltine

, Merriam

. Polycystic Ovary Syndrome. Boston: Blackwell Scientific Publications, 1992; 377–384.

Dunaif

, Segal

, Futterweit

et al. Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes, 1989; 38:1165–1174.

DeUgarte

, Bartolucci

, Azziz

. Prevalence of insulin resistance in the polycystic ovary syndrome using the homeostasis model assessment. Fertil Steril, 2005; 83:1454–1460.

Legro

, Gnatuk

, Kunselman

et al. Changes in glucose tolerance over time in women with polycystic ovary syndrome: A controlled study. J Clin Endocrinol Metab, 2005; 90:3236–3242.

Boudreaux

, Talbott

, Kip

et al. Risk of T2DM and impaired fasting glucose among PCOS subjects: Results of an 8-year follow-up. Curr Diab Rep, 2006; 6:77–83.

Birdsall

, Farquhar

, White

. Association between polycystic ovaries and extent of coronary artery disease in women having cardiac catheterization. Ann Intern Med, 1997; 126:32–35.

Guzick

, Talbott

, Sutton-Tyrrell

et al. Carotid atherosclerosis in women with polycystic ovary syndrome: Initial results from a case-control study. Am J Obstet Gynecol, 1996; 174:1224–1232.

Pierpoint

, McKeigue

, Isaacs

et al. Mortality of women with polycystic ovary syndrome at long-term follow-up. J Clin Epidemiol, 1998; 51:581–586.

Wild

, Pierpoint

, McKeigue

et al. Cardiovascular disease in women with polycystic ovary syndrome at long-term follow-up: A retrospective cohort study. Clin Endocrinol (Oxf), 2000; 52:595–600.

10.

Shaw

, Bairey Merz

, Azziz

et al. Postmenopausal women with a history of irregular menses and elevated androgen measurements at high risk for worsening cardiovascular event-free survival: Results from the National Institutes of Health–National Heart, Lung, and Blood Institute sponsored Women's Ischemia Syndrome Evaluation. J Clin Endocrinol Metab, 2008; 93:1276–1284.

11.

Jialal

, Devaraj

, Venugopal

. C-reactive protein: Risk marker or mediator in atherothrombosis? Hypertension, 2004; 44:6–11.

12.

Ridker

, Hennekens

, Buring

et al. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med, 2000; 342:836–843.

13.

Pearson

, Mensah

, Alexander

et al. Centers for Disease Control and Prevention; American Heart Association. Markers of inflammation and cardiovascular disease: Application to clinical and public health practice: A statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation, 2003; 107:499–511.

14.

Ridker

, Buring

, Cook

et al. C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: An 8-year follow-up of 14,719 initially healthy American women. Circulation, 2003; 107:391–397.

15.

Kelly

, Lyall

, Petrie

et al. Low grade chronic inflammation in women with polycystic ovarian syndrome. J Clin Endocrinol Metab, 2001; 86:2453–2455.

16.

Diamanti-Kandarakis

, Paterakis

, Alexandraki

et al. Indices of low-grade chronic inflammation in polycystic ovary syndrome and the beneficial effect of metformin. Hum Reprod, 2006; 21:1426–1431.

17.

Schachter

, Raziel

, Friedler

et al. Insulin resistance in patients with polycystic ovary syndrome is associated with elevated plasma homocysteine. Hum Reprod, 2003; 18:721–727.

18.

Nygård

, Nordrehaug

, Refsum

et al. Plasma homocysteine levels and mortality in patients with coronary artery disease. N Engl J Med, 1997; 337:230–236.

19.

, Wat

, Ho

. Effects of metformin on ovulation rate, hormonal and metabolic profiles in women with clomiphene-resistant polycystic ovaries: A randomized, double-blinded placebo-controlled trial. Hum Reprod, 2001; 16:1625–1631.

20.

Fleming

, Hopkinson

, Wallace

et al. Ovarian function and metabolic factors in women with oligomenorrhea treated with metformin in a randomized double blind placebo-controlled trial. J Clin Endocrinol Metab, 2002; 87:569–574.

21.

Moghetti

, Castello

, Negri

et al. Metformin effects on clinical features, endocrine and metabolic profiles, and insulin sensitivity in polycystic ovary syndrome: A randomized, double-blind, placebo-controlled 6-month trial, followed by open, long-term clinical evaluation. J Clin Endocrinol Metab, 2000; 85:139–146.

22.

Orio

, Manguso

, Di Biase

et al. Metformin administration improves leukocyte count in women with polycystic ovary syndrome: A 6-month prospective study. Eur J Endocrinol, 2007; 157:69–73.

23.

Morin-Papunen

, Rautio

, Ruokonen

et al. Metformin reduces serum C-reactive protein levels in women with polycystic ovary syndrome. J Clin Endocrinol Metab, 2003; 88:4649–4654.

24.

Palomba

, Falbo

, Giallauria

et al. Effects of metformin with or without supplementation with folate on homocysteine levels and vascular endothelium of women with polycystic ovary syndrome. Diabetes Care, 2010; 33:246–251.

25.

Vrbíková

, Bicíková

, Tallová

et al. Homocysteine and steroids levels in metformin treated women with polycystic ovary syndrome. Exp Clin Endocrinol Diabetes, 2002; 110:74–76.

26.

Kilicdag

, Bagis

, Zeyneloglu

et al. Homocysteine levels in women with polycystic ovary syndrome treated with metformin versus rosiglitazone: A randomized study. Hum Reprod, 2005; 20:894–899.

27.

Ferriman

, Gallwey

. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab, 1961; 21:1440–1447.

28.

World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation. Part 1: Diagnosis and Classification of Diabetes Mellitus. Geneva: World Health Organization, 1999.

29.

Friedewald

, Levy

, Fredrickson

. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 1972; 18:499–502.

30.

Mathews

, Hosker

, Rudenski

et al. Homeostasis model assessment: Insulin resistance and β cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia, 1985; 28:412–419.

31.

Refsum

, Smith

, Ueland

et al. Facts and recommendations about total homocysteine determinations: An expert opinion. Clin Chem, 2004; 50:3–32.

32.

Yajnik

, Deshpande

, Lubree

et al. Vitamin B12 deficiency and hyperhomocysteinemia in rural and urban Indians. J Assoc Physicians India, 2006; 54:775–782.

33.

Harborne

, Sattar

, Norman

et al. Metformin and weight loss in obese women with polycystic ovary syndrome. Comparison of doses. J Clin Endocrinol Metab, 2005; 90:4593–4598.

34.

Tan

, Hahn

, Benson

, Dietz

et al. Metformin improves polycystic ovary syndrome symptoms irrespective of pre-treatment insulin resistance. Eur J Endocrinol, 2007; 157:669–676.

35.

Legro

, Barnhart

, Schlaff

et al. Cooperative Multicenter Reproductive Medicine Network. Clomiphene, metformin, or both for infertility in the polycystic ovary syndrome. N Engl J Med, 2007; 356:551–566.

36.

Viswanathan

, Tilak

, Meerza

et al. Insulin resistance at different stages of diabetic kidney disease in India. J Assoc Physicians India, 2010; 58:612–615.

37.

Kalra

, Nair

, Rai

. Association of obesity and insulin resistance with dyslipidemia in Indian women with polycystic ovarian syndrome. Indian J Med Sci, 2006; 60:447–453.

38.

Diamanti-Kandarakis

, Kouli

, Tsianateli

et al. Therapeutic effects of metformin on insulin resistance and hyperandrogenism in polycystic ovary syndrome. Eur J Endocrinol, 1998; 138:269–274.

39.

Morin-Papunen

, Vauhkonen

, Koivunen

et al. Endocrine and metabolic effects of metformin vs. ethinyl estradiol-cyproterone acetate in obese women with polycystic ovary syndrome: A randomized study. J Clin Endocrinol Metab, 2000; 85:3161–3168.

40.

Acbay

, Gundogdu

. Can metformin reduce insulin resistance in polycystic ovary syndrome? Fertil Steril, 1996; 65:946–949.

41.

Ehrmann

, Cavaghan

, Imperial

et al. Effects of metformin on insulin secretion, insulin action, and ovarian steroidogenesis in women with polycystic ovary syndrome. J Clin Endocrinol Metab, 1997; 82:524–530.

42.

Kahn

, Prigeon

, McCulloch

et al. Quantification of the relationship between insulin sensitivity and β -cell function in human subjects. Evidence for a hyperbolic function. Diabetes, 1993; 42:1663–1672.

43.

Valkenburg

, Steegers-Theunissen

et al. A more atherogenic serum lipoprotein profile is present in women with polycystic ovary syndrome: A case-control study. J Clin Endocrinol Metab, 2008; 93:470–476.

44.

Verit

. High sensitive serum C-reactive protein and its relationship with other cardiovascular risk factors in normoinsulinemic polycystic ovary patients without metabolic syndrome. Arch Gynecol Obstet, 2010; 281:1009–1014.

45.

Boulman

, Levy

, Leiba

et al. Increased C-reactive protein levels in the polycystic ovary syndrome: A marker of cardiovascular disease. J Clin Endocrinol Metab, 2004; 89:2160–2165.

46.

Guzelmeric

, Alkan

, Pirimoglu

et al. Chronic inflammation and elevated homocysteine levels are associated with increased body mass index in women with polycystic ovary syndrome. Gynecol Endocrinol, 2007; 23:505–510.

47.

Fonseca

, Guba

, Fink

. Hyperhomocysteinemia and the endocrine system: Implications for atherosclerosis and thrombosis. Endocr Rev, 1999; 20:738–759.

48.

Heutling

, Schulz

, Nickel

et al. Asymmetrical dimethylarginine, inflammatory and metabolic parameters in women with polycystic ovary syndrome before and after metformin treatment. J Clin Endocrinol Metab, 2008; 93:82–90.

49.

Yilmaz

, Bukan

, Ayvaz

et al. The effects of rosiglitazone and metformin on oxidative stress and homocysteine levels in lean patients with polycystic ovary syndrome. Hum Reprod, 2005; 20:3333–3340.

50.

Dhanalakshmi

, Sumathi

, Pasupathi

et al. Lipid profile and Homocysteine level changes in metformin and metformin-letrozole therapy with South Indian PCOS women. Int J Cur Biomed Phar Res, 2011; 1:102–108.