Clinical Research Concerning The Glucose And Lipid Metabolism In Polycystic Ovary Syndrome(PCOS)

Background and Objective:Polycystic ovary syndrome (PCOS) is the most common endocrine disease in women of reproductive age, whose incidence is4%-12%. The clinical manifestations of PCOS are various, for example abnormal menstruation (oligomenorrhea or amenorrhea), hirsutism, acne, obesity, infertility and so on. Long term complications include hypertension, hyperlipidemia, diabetes and cardiovascular disease. PCOS starts from adolescence, not only affects the body endocrine and metabolism, but also brought psychological problems to the patients. Previous studies have shown that the incidence of dyslipidemia and impaired glucose tolerance (IGT) is higher in patients with PCOS than normal people. The cause has not yet fully elucidated, and may be associated with obesity, insulin resistance and hyperandrogenism. Because of the differences in race, life style and dietary habit, the incidence of IGT and dyslipidemia is varied in different countries. Basically, western countries’is higher while Asian countries’is lower. A growing number of studies support that insulin resistance is the key pathogenesis of PCOS, metabolic abnormalities are associated with it. But there are also studies consider that metabolic abnormalities have a more close relationship with obesity or hyperandrogenism. In this study, we further explore the incidence and type of dyslipidemia and abnormal glucose tolerance in Chinese PCOS patients through, and the related factors of glucose and dyslipidemia, hoping to provide better guidance for the treatment of PCOS.In recent years, metabolic syndrome (MS) is getting more and more concerning. Most countries have found that the incidence of MS in PCOS patients increased significantly than normal healthy people. MS is characterized by the central obesity, along with abnormal blood glucose, blood lipid and blood pressure. People diagnosed with this syndrome are at a high risk of getting diabetes, atherosclerosis related diseases. Some studys have found that apolipoprotein (apo) has a closer relation with atherosclerosis disease than triglyceride(TG) or total cholesterol (TC), especially the apoB/apoA I ratio is better than any other lipid components in the prediction of cardiovascular disease. This study aims to explore the incidence of MS and the value of apoB/apoA I ratio in diagnosting MS in PCOS patients.Chapter I The glucose and lipid metabolism in patients with polycystic ovary syndrome and their relationship with insulin resistanceObjects:185patients were included in this study, who visited the outpatient of obstetrics and gynecology department in Zhujiang Hospital of Southern Medical University for the first time in November2008to November2012, and had not taking any drugs that affected the hormones, plasma glucose and lipid in the recent3monthes before the diagnosis, and conformed to the2003Rotterdam diagnostic criteria of PCOS.Objectives:to explore the incidence of abnormal glucose and lipid metabolism in PCOS patients and their relationships with insulin resistance.Methods:1. All included patients were gathered the medical history (menarche age, menstrual history, past marital and fertile history, incentives of menstrual disorders, etc.), and family history.2. Measured the height, weight, blood pressure, waist circumference, hip circumference, and calculated the BMI (body mass index, BMI) and waist-to-hip ratio. 3. Fasting venous blood was collected on the second or third day of menstrual cycle or withdrawal bleeding, being used to detect the concentration of estradiol (E2), follicle-stimulating hormone (FSH), luteinizing hormone (LH), prolactin (PRL), testosterone (T) and blood lipid, including total cholesterol (TC), triglyceride (TG), low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol (HDL-C), apolipoprotein B (apoB) and apolipoprotein A I (apoA I).4. B-ultrasonic examination was performed on the fourth or fifth day of the menstrual cycle or withdrawal bleeding by specific ultrasound doctors.5. Oral glucose tolerance test (OGTT) and insulin release test (IRT):every included object fasted for8-12hours, and fasting blood was collected in the next morning. Then solubled83g normal glucose powder (containing a H2O molecule) in250-300ml water and drinked it off in5minutes. Timing started when the first bite of sugar water was taked. Blood samples were taken on the first hour and the second hour after taking the sugar water. The serum glucose and insulin levels were detected in all the blood samples. The homeostasis model assessment-insulin resistance (HOMA-IR) index was calculated using the values of fasting glucose (FPG) and fasting insulin (FINS).6. Statistical methods:SPSS19.0statistical software was used to analyze the datas. The homogeneity of variance was tested with Leneve. The measurement data was described with x±s, and the enumeration data was described with absolute terms (percentage). Comparison between two groups was done with independent sample T-test and chi-square test. Correlation analysis was done with Pearson or Spearman. Difference was statistically significant with P<0.050.7. Groups:(1) According to the WHO diagnostic criteria for diabetes in2006,185patients with PCOS were divided into normal glucose tolerance group (NGT) and the abnormal glucose tolerance group (AGT). Cases of the two groups were148and37respectively. The differences of related factors between the two groups were compared, and stratified analysis of the incidences of abnormal glucose tolerance according to BMI and age were also done. (2) According to a dyslipidemia prevention guideline for Chinese adults in2007,185patients with PCOS were divided into normal blood lipids group and dyslipidemia group. Cases of the two groups were154and31respectively. The differences of related factors between the two groups were compared, and stratified analysis of the incidences of dyslipidemia according to BMI and age were also done.(3) According to the data of the national diabetes cooperation investigation, insulin resistance was defined as the HOMA-IR greater than or equal to2.69.185PCOS patients were divided into non-insulin resistant group and insulin resistant group. Cases of the two groups were135and50respectively. The incidences of dyslipidemia and abnormal glucose tolerance of the two groups were compared, and the correlations of dyslipidemia and abnormal glucose tolerance with insulin resistance were also analyzed.Results:1. In this study, we found that80.00%(148/185) of all the185PCOS patients had normal glucose tolerance (NGT), while20%of them had IGT and DM:IGT17.84%(33/185) and DM2.16%(4/185).2. The mean values of age, BMI, waist hip ratio, HOMA IR, fasting insulin (FINS),2h-INS were higher in abnormal glucose tolerance(AGT) group (37cases) than in NGT group (148cases), the difference was statistically significant (P<0.05); the mean values of LH, T, systolic pressure, diastolic blood pressure and blood lipid levels had no obvious difference.3. Stratified analysis of abnormal glucose tolerance (AGT) according to BMI were performed and we found that overweight people (23≤BMI<25kg/m2) accounted for17.30%(32/185) in all the patients, while overweight people (BMI>25kg/m2) accounted for20.54%(38/381). Along with the increase of BMI, the incidence of increased significantly, from20.83%in the BMI<18kg/m2layer to55.56%in the BMI≥28kg/m2layer.4. Stratified analysis of abnormal glucose tolerance (AGT) according to age were also performed and we found that And we found that, as the growth of the age, the incidence of AGT increased significantly, form0.00%in the16～20years-old layer to33.33%in the≥36years-old layer.5. In this study, the total incidence of dyslipidemia was16.76%(31/185), in which the TC abnormality was3.24%(6/185), TG abnormality was4.32%(8/185), LDL-C abnormal accounted for2.16%(4/185), HDL-C accounted for9.73%(18/185). The incidences from high to low were:low HDL-C, high TG, high TC, and high LDL-C.6. The mean values of BMI, waist-to-hip ratio, systolic blood pressure, diastolic blood pressure, fasting insulin, HOMA-IR and apoB/apoA I in dyslipidemia group (31cases) were significantly elevated than that in the nomal blood lipid group, while the mean values of age, LH, T, fasting plasma glucose, fasting insulin,2h-glucose and2h-insulin had no obvious difference between the two groups.7. Stratified analysis of dyslipidemia according to BMI were performed, and we found that along with the increase of BMI, the incidence of dyslipidemia PCOS patients increased significantly, from0.00%in the BMI<18kg/m2layer to33.33%in the BMI≥28kg/m2layer.8. Stratified analysis of dyslipidemia according to age was also performed, and we found that the incidence of dyslipidemia in the16-20years-old layer was the same with that in the>36years-old layer (33.33%).9. The incidence of insulin resistance was27.03%(50/185), the incidences of dyslipidemia and abnormal glucose tolerance of the insulin resistant group were30.00%(15/50) and38.00%(19/50) respectively, while those of the non-insulin resistant group were11.85%(16/135) and13.33%(18/135) respectively. The total incidence of dyslipidemia and abnormal glucose tolerance of the insulin resistant group increased significantly than the non-insulin resistant group (chi-square values were8.615and8.615respectively; P values all were0.000). The incidence of higher TG in the insulin resistant group increased significantly than that in the non-insulin resistant group (chi-square value was5.335, P value was0.021), while the differences of incidences of higher TC, lower HDL-C and higher LDL-C between the two groups had no statistical significance (chi-square values were0.125,3.067and0.125respectively; P values were0.724,0.080and0.296respectively). 10. HOMA-IR index had a positive correlation with BMI, waist circumference, fasting plasma glucose,1h-glucose,2h-glucose, TG, LDL-C, abnormal glucose tolerance and dyslipidemia (correlation coefficients r values were0.590,0.547,0.356,0.279,0.422,0.281,0.286,0.146and0.239respectively, and P values were0.000,0.000,0.000,0.000,0.000,0.000,0.000,0.000and0.001respectively), while a negative correlation with HDL-C (correlation coefficient r values were0.268, and P value was0.000). The correlation with TC had no statistical significance (correlation coefficient r value was0.022, and P value was0.764).Conclusion:1. In this study, the incidence of abnormal glucose tolerance patients with PCOS was20.00%, and increased significantly as BMI and age rose. All the PCOS patients in the clinical should have an OGTT experiment, and early intervention should carry out in patients with abnormal glucose tolerance (for example, weight loss or using hypoglycemic drugs) to prevent or delay the development of diabetes.2. The incidence of dyslipidemia was16.76%, the incidence of each type from high to low were:lower HDL-C, higher TG, higher TC, and higher LDL-C.3. The incidence of dyslipidemia in PCOS patients increased significantly along with the rise of BMI. Obesity might worsen the lipid metabolism in PCOS patients, and weight control was he primary goal of the treatment of dyslipidemia in PCOS patients.4. In our study, the incidence of dyslipidemia in adolescent PCOS patients was the same with that of the middle-aged patients, all were33.33%. That may be related to high incidence of overweight and obesity in adolescent patients. Attentions should be pay to screening of dyslipidemia in adolescent PCOS patients, especially overweight or obese patients.5. Our study found out that insulin resistance had a positive correlation with abnormal glucose tolerance and dyslipidemia. Improvement of insulin resistance may be helpful in correcting the abnormal glucose and lipid metabolism in PCOS patients.Chapter II Diagnostic value of apoB/apoA I ratio for metabolic syndrome in patients with polycystic ovary syndrome Objects:185patients were included in this study, who visited the outpatient of obstetrics and gynecology department in Zhujiang Hospital of Southern Medical University for the first time in November2008to November2012, and had not taking any drugs that affected the hormones, plasma glucose and lipid in the recent3monthes before the diagnosis, and conformed to the2003Rotterdam diagnostic criteria of PCOS.(The same with the first chapter)Objectives:to explore the incidence of metabolic syndrome (MS) and the diagnostic value of apoB/apoAⅠ ratio for metabolic syndrome in patients with polycystic ovary syndrome (PCOS).Methods:1. All included patients were gathered the medical history (menarche age, menstrual history, past marital and fertile history, incentives of menstrual disorders, etc.), and family history. Measured the height, weight, blood pressure, waist circumference, hip circumference, and calculated the BMI (body mass index, BMI) and waist-to-hip ratio.2. Fasting venous blood was collected on the second or third day of menstrual cycle or withdrawal bleeding, being used to detect the concentration of sex hormones and metabolism indexes.3. B-ultrasonic examination was performed on the fourth or fifth day of the menstrual cycle or withdrawal bleeding by specific ultrasound doctors.4. Statistical methods:SPSS19.0statistical software was used to analyze the datas. The homogeneity of variance was tested with Leneve. The measurement data was described with x±s, and the enumeration data was described with absolute terms (percentage). Comparison between two groups was done with independent sample T-test and chi-square test, and multiple comparisons with one-way ANOVA. Correlation analysis was done with Pearson or Spearman. Using ROC curve analyzed the diagnostic value of apoB/apoAⅠ ratio for metabolic syndrome. Difference was statistically significant with P<0.050.5. Groups:(1) According to the international diabetes federation (IDF) diagnostic criteria of for metabolic syndrome in2005,185patients were divided into the metabolic syndrome and non-metabolic syndrome group. Cases of the two groups were148and37respectively. The differences of related factors between the two groups were compared.(2) According to whether abdominal obesity and insulin resistance (IR) or not,185patients were divided into four groups:l=not abdominal obesity not IR group,2=not abdominal obesity and IR group,3=abdominal obesity not IR group,4=abdominal obesity IR group. Cases of the four groups were89,10,46and40respectively. The mean values of apoB/apoA I ratio were compared among the for groups, and the correlation between the apoB/apoA I ratio and some indexes of PCOS was exploded.(3) According to the numbers of abnormal metabolic syndrome components,185patients were divided into five groups:0group,1group,2group,3group,4group and5group. Cases of the five groups were80,58,35,9and3respectively. The mean values of apoB/apoA I ratio were compared among the for groups, and the diagnostic value of apoB/apoA I ratio for metabolic syndrome was exploded by using ROC curve.Results:1. In this study, we found that the morbidity of MS was19.46%in all the185PCOS patients, and56.76%of them had at least one abnormal components of MS. The mean values of BMI, waist circumference, waist-to-hip ratio, systolic blood pressure and diastolic blood pressure, fasting glucose, fasting insulin,1h glucose,1h insulin,2h glucose,2h insulin, HOMA-IR, TG, LDL-C, apoB and apoB/apoA I of MS group were significantly higher than those of the non-MS group (P values were0.000,0.000,0.000,0.000,0.000,0.000,0.000,0.012,0.014,0.000,0.040,0.000,0.000,0.010,0.000and0.000respectively); while HDL-C and apoA I decreased significantly (P values were0.000and0.000respectively). The mean values of age and TC between the two groups had no statistical significance (P values were0.398and0.393respectively).2. ApoB/apoA I ratio was positively correlated with BMI, waist circumference, waist-to-hip ratio, blood pressure, fasting glucose, HOMA IR, TG and LDL-C (correlation coefficients r values were0.469,0.415,0.411,0.317,0.219,0.285and0.292respectively, and P values all were0.000), and negatively correlated with HDL-C (correlation coefficients r values was-0.465, and P values was0.000), all statistically significant (P<0.05). But the correlation of age and ovarian volume had no statistical significance (correlation coefficients r values were0.010and-0.062respectively, and P values were0.889and0.398respectively).3. All patients were divided into four groups according to whether they had abdominal obesity and IR. Levene test showed that the homogeneity of variances of had no statistical significance (F=0.544, P=0.544), and we found that the differences of averages of apoB/apoA I ratio between group1and3, group1and4, group2and4were all statistically significant (P values were0.000,0.000,0.049and0.015), while the differences of averages between group1and2, group3and4had no statistical significance (P values were0.918and0.428respectively).4. According to the numbers of abnormal metabolic syndrome components,185patients were divided into five groups, Levene test showed that the homogeneity of variances of apoB/apoA I ratio had no statistical significance (F=0.544, P=0.544), multiple comparisons among five groups were done with one-way ANOVA. We found that the differences of averages of apoB/apoA I ratio between group0and2, group0and3, group0and4, group1and2, group1and3, group2and3, group2and4, group3and4were all statistically significant (P values were0.000,0.000,0.000,0.000,0.000,0.000,0.009,0.000and0.007respectively), while the differences of averages between group0and1had no statistical significance (P values was0.250).5. Using the receiver operating characteristics (ROC) curve to analysis the diagnostic value of apoB/apoA I ratio for MS in PCOS patients, we found that when the cutoff value of apoB/apoA I was0.66, the area under the ROC curve was the biggest (0.885), with a sensitivity of0.917and specificity of0.700.Conclusion:1. In this study, the incidence of metabolic syndrome in PCOS patients was 19.46%, and more than half of the patients had at least one abnormal components of the metabolic syndrome, who were likely to develop metabolic syndrome in the future. Patients diagnosed with metabolic syndrome were at a higher risk of developing diabetes and cardiovascular disease. Therefore, it has important significance to attach great importance in screening of metabolic abnormalities in patients with PCOS and carrying out timely intervention soon afterwards.2. The average of apoB/apoA I ratio rose as the number of abnormal metabolic syndrome components increased, that means it had a close relationship with metabolic syndrome.3. ApoB/apoA I ratio had a good diagnostic value for metabolic syndrome. Those PCOS patients, who did not meet the diagnostic criteria for metabolic syndrome but the apoB/apoA I ratio was more than0.66, should be alert to the occurrence of the metabolic syndrome, and should take appropriate intervention measures in a timely manner.