Assessment of early vascular changes in adult females with polycystic ovary syndrome: correlation with insulin resistance

Abstract Polycystic ovary syndrome (PCOS) is the commonest endocrine disorder in females in the reproductive period with estimated prevalence of 5% to 18% [1]. It contributes to the mortality and morbidity in patients with PCOS due to the increased risk of different metabolic and cardiovascular (CV) complications [2]. Despite the presence of obesity in 40-60% of cases [3], the disease may occur in non-obese women. The occurrence of metabolic disorders in non-obese PCOS patients, suggests that the syndrome itself may play a role in the development of metabolic and CV co-morbidities [4]. The identification of early stages of atherosclerosis in patients with PCOS might be useful in the development of new strategies to control modifiable CV risk factors [5]. Assessment of vascular endothelial function (ED) as an initial reversible step in atherosclerosis development, may serve as an integral index for CV risk factor burden [6]. In addition, carotid intima media thickness (CIMT) is a helpful marker for atherosclerosis and for the identification of increased risk of CV disease [7]. Our study assessed the early vascular changes in Egyptian women with PCOS both physically and functionally by looking at the CIMT using high resolution Doppler ultrasound and by measuring ED using brachial artery flow–mediated vasodilatation (FMD). Our results indicate that patients with PCOS have significant ED and premature atherosclerosis which is, to a great extent, independent of obesity and IR. This suggests that PCOS patients are at increased risk for premature CVD and may benefit from early detection and management.


Introduction
Polycystic ovary syndrome (PCOS) is the commonest endocrine disorder in females in the reproductive period with estimated prevalence of 5% to 18% [1].It contributes to the mortality and morbidity in patients with PCOS due to the increased risk of different metabolic and cardiovascular (CV) complications [2].Despite the presence of obesity in 40-60% of cases [3], the disease may occur in non-obese women.The occurrence of metabolic disorders in non-obese PCOS patients, suggests that the syndrome itself may play a role in the development of metabolic and CV co-morbidities [4].
The identification of early stages of atherosclerosis in patients with PCOS might be useful in the development of new strategies to control modifiable CV risk factors [5].Assessment of vascular endothelial function (ED) as an initial reversible step in atherosclerosis development, may serve as an integral index for CV risk factor burden [6].In addition, carotid intima media thickness (CIMT) is a helpful marker for atherosclerosis and for the identification of increased risk of CV disease [7].According to previous research, patients with PCOS may present with ED, endothelial cell proliferation, endothelial inflammation and coagulation disorders.Therefore, early detection of ED might be useful in atherosclerosis prevention [8].Our study's objective was to analyze the early vascular changes in Egyptian women with PCOS, both physically and functionally, by looking at the CIMT using high resolution Doppler ultrasound and by measuring ED using brachial artery flow-mediated dilatation (FMD).

Patients and methods
Our study included seventy Egyptian females aged 18 to 40 years.The patient group included 50 women who were diagnosed as PCOS using Rotterdam Criteria (PCOS group) whereas the control group included 20 healthy subjects.In the years 2019 and 2020, all patients were enrolled from the General Gynecology Clinics and the Endocrinology & Metabolic Diseases Outpatient Clinic at Cairo University hospitals.The study was approved by the Ethics Committee of the Faculty of Medicine, Cairo University.
Patients known to have atherosclerotic cardiovascular disease, hypertension, diabetes type 1 or type 2 and history of any chronic medical illness e.g.chronic kidney disease, chronic liver disease as well as patients younger than 18 years or older than 40 years were excluded from the study.Informed consent was obtained from all subjects after explanation of the study goals and techniques.
All subjects underwent detailed history taking, thorough clinical examination including assessment of weight, height, body mass index (BMI), waist circumference, systolic and diastolic blood pressure measurement, laboratory investigations in the form of: fasting insulin, fasting blood glucose, HbA1C, lipid profile (serum cholesterol, LDL-C, HDL-C and serum triglycerides), TSH and free testosterone.HOMA-IR index was calculated according to the formula: HOMA-IR = {fasting glucose (mg/dl) × fasting insulin (uIU/ml)}/405.
Measurement of CIMT was performed using colored Doppler using a high-resolution linear probe with frequency of 7-4.All of the patients were examined while lying flat on their backs with their heads hyperextended and turned to the side being scanned.The common carotid arteries on the right and left, as well as the carotid bifurcation-bulb region, were all sonographically studied (lateral, anterior oblique, and posterior oblique).Images were taken 1-2 cm from the carotid bulb and just next to the bifurcation's origin from the distal section of both common carotid arteries.From the B-mode screen, the CIMT was calculated at end-diastole.The average of six electronic caliber measurements of the right and left carotid arteries was used to determine the mean CIMT for each side.Endothelial function was assessed by using brachial artery flow-mediated vasodilatation (FMD) and calculating the percentage of change of flow using the duplex machine and imaging the brachial artery caliber before and after inflation of sphygmomanometer cuff just above the antecubital fossa.Doppler examination for all patients and control subjects was performed by the same skilled sonographer.

Statistical analysis
Data analysis was done using the computerized SPSS version 26 (IBM Corp., Armonk, NY, USA).For quantitative variables, the mean, standard deviation, median, minimum, and maximum; and for categorical variables, frequencies (number of instances) and relative frequencies (percentages).For regularly distributed quantitative data, the unpaired t test was used to compare groups.For quantitative variables with non-normal distribution, the Mann-Whitney test was applied.An analysis using the Chi square [2] test was done to compare categorical data.Correlations between quantitative variables were done using Pearson correlation coefficient.Each of the key end goals (FMD, IMT) was tested individually using stepwise multiple linear regression analysis against the independent variables of age, BMI, waist circumference, free testosterone, lipid concentrations, systolic and diastolic blood pressure.Linear regression analysis was done to detect if PCOS acts as an independent predictor of CIMT and ED.P values less than 0.05 were considered as statistically significant.Data was checked for normality using normality plot and Shapiro Wilk test.
Total cholesterol and LDL-C were significantly higher (185.76 ± 37.56, 118.84 ± 36.63 mg/dl respectively) in the patient group than in the control group (161.65 ± 31.53, 100.39 ± 29.31 mg/ dl respectively) with p value of 0.014, 0.049 respectively.Triglycerides and HDL-C were not significantly different between the two groups.Also there was no statistically significant difference in the fasting blood glucose, fasting insulin, HOMA-IR and HBA1C between both groups.
Hormonal profile of the studied groups showed that FSH was significantly higher in control group compared to the PCOS group (p = 0.001), and LH was significantly higher in the PCOS group compared to control group (p = 0.001).LH/FSH was significantly higher in PCOS group than in control group (p = 0.001).There was no statistically significant difference in TSH & free testosterone levels between the two groups.CIMT was significantly higher in the PCOS group (0.06 ± 0.01 cm) than in the control group (0.05 ± 0.01 cm) (p < 0.001) (Table 1).
PCOS patients with insulin resistance showed higher levels of total cholesterol and LDL-C than those without insulin resistance (Table 2).
There was a significant positive correlation between HOMA-IR and BMI, waist circumference, systolic blood pressure, diastolic blood pressure, triglycerides and total Cholesterol.There was significant negative correlation between HOMA-IR and HDL-C (Table 3).After reactive hyperemia, a more intense vasodilatation was observed in controls than in polycystic ovary syndrome (PCOS) patients (p < 0.001) (Table 4).
Correlation analysis of LH/FSH ratio and CIMT &ED in the patient group revealed that LH/FSH ratio correlates negatively with FMD (r=-0.298,p value 0.012).
Multivariate regression analysis showed that systolic blood pressure was an independent determinant of CIMT in patients with PCOS, and triglycerides level was an independent determinants of ED with positive correlation between ED and triglycerides level.Linear regression analysis was done and showed that PCOS acts as a strong independent predictor of CIMT and ED (Table 5).

Discussion
Our study assessed the early vascular changes in Egyptian women with PCOS both physically and functionally by looking at the CIMT using high resolution Doppler ultrasound and by measuring ED using brachial artery FMD.After ruling out other potential causes such congenital adrenal hyperplasia or androgen-secreting neoplasms, the diagnosis must include two of the three Rotterdam Criteria.
Our study showed no significant statistical difference between both groups as regard BMI, waist circumference, SBP, DBP, HBA1C.TC and LDL-C were significantly higher in PCOS patients as compared to the control subjects.This was in agreement with Wekker et al. [9] who found that women with PCOS had higher serum concentration of TC.Also, Ibrahim et al. reported that patients with PCOS had atherogenic lipoprotein profile characterized by high TC, LDL-C and triglycerides which was more marked in obese PCOS patients [10].FSH levels were significantly high in control group (p = 0.001) and LH levels were significantly high in PCOS group (p = 0.001) and LH/FSH was significantly high in PCOS group (p = 0.001).There was negative correlation between the LH/FSH ratio and the brachial artery FMD in the studied participants.High LH levels was a consistent finding in studies that examined ED and CV risk in PCOS patients.Experimental studies showed correlations of LH levels with endothelial functions in other diseases e.g.ulcerative colitis and chronic kidney disease [11].However, there aren't enough studies that explored this correlation in PCOS patients.
We assessed CIMT, as a surrogate marker of atherosclerosis, using high-resolution Doppler ultrasound and it was evident that that CIMT was significantly higher in PCOS patients as compared to the control subjects (p < 0.001).This is in agreement with Jabbour et al. [12] who reported significantly higher CIMT values in patients with PCOS which makes it most likely for PCOS patients to feature signs of premature systemic atherosclerosis.
Assessment of the endothelial function could be a tool with unrealized potential in early diagnosis CVD in the presence or absence of CV threats [13].An ideal method for the direct of endothelial function assessment should be standardized between laboratories [14].FMD, which is considered a safe, noninvasive, reproducible, cost-effective tool, was used to assess ED.After reactive hyperemia, more profound vasodilatation was observed in the control group than in PCOS patients (p < 0.001).Previously described data on endothelial dysfunction in PCOS women have been conflicting.Similar to our results Alexandraki et al. who reviewed 39 studies that assessed FMD showed impaired FMD in PCOS [15].Also in agreement with our results, Kravariti et al. [16] deduced that there is significant ED in women with PCOS at an early age.On the other hand, this was against Krentowska et al. [17] who reported no discrepancy in CIMT and FMD between PCOS patients and control subjects, nor between PCOS phenotypes.
According to a systematic review of 21 studies, mean values of FMD were lower in women with PCOS than control subjects.BMI and age had no effect on FMD.The review concluded that there is evidence of ED in women with PCOS, even if they are young and have normal BMI [18].
There is strong evidence that ED, which is an initial reversible step in atherosclerosis development, is found in most PCOS patients.So, early detection of ED may play an important role in atherosclerosis prevention as it affects apoptosis, coagulation/ fibrinolysis processes and smooth muscle cells proliferation [19].Moreover, studies undertaken in young women with PCOS confirmed overt ED in young, lean patients with PCOS compared to normal women of similar age and BMI [20].
In our study, PCOS patients with insulin resistance (HOMA-IR > 2.5) showed higher levels of TC and LDL-C compared to PCOS patients without insulin resistance.HOMA-IR of PCOS patients correlated positively with BMI, waist circumference, SBP, DBP, TG, TC and negatively with HDL.This is consistent with Wanderley et al. [21] who reported a statistically significant association between insulin resistance and BMI, as well as dyslipidemia and arterial hypertension in a high proportion of PCOS patients.There was no difference between PCOS group and control subjects with regards to plasma insulin concentrations and IR in our study.Additionally, our research concluded no correlation between HOMA-IR and CIMT or ED.This finding may indicate that there are additional mechanisms of cardiovascular risks in PCOS patients other than insulin resistance.Similarly, Lambert et al. [22] reported ED in PCOS patients but failed to find a link between endothelial function and IR. a prospective study Kravariti et al. [16].concluded that young patients with PCOS have ED which correlated with IR.In another study, ED was significantly impaired in obese PCOS patients as the levels of HOMA-IR increased [3].Therefore, even though a greater proportion of women with PCOS have one or more parameter of insulin resistance, this shouldn't be entirely attributed to ED due to inconsistent findings.According to the linear regression analysis, we detected that PCOS acts as a strong independent predictor of CIMT and ED.This agreed with Jabbour et al. [12] who found that even after numerous BMI adjustments, the PCOS diagnosis remained the best indicator of CIMT.Considering the lack of correlation between ED and hemodynamic, metabolic and hormonal parameters, our study is consistent with Lambert et al. [22].On the other hand, Virdis et al. found that endothelial function correlates well with many cardiovascular risk factors including obesity, hypertension and metabolic abnormalities [23].
The small number of the studied population is a limitation that might question the application of the results to all patients with PCOS.A prospective study, analyzing the development of CVD in PCOS with different phenotypes of the syndrome, should yield more information regarding the practicality of CIMT and FMD as predictors of CVD.Getting more information about genetic susceptibility or inflammatory biomarkers of selected PCOS cases that might be the cause or the result of PCOS and its metabolic co-morbidities such as insulin resistance & ED would help in predicting the cardiovascular risk.ED in our study has been assessed by FMD, however, the usage of more than one tool to assess endothelial function could increase the reliability of the results.
Our study didn't use hyperinsulinemic-euglycemic clamp, the gold standard for insulin resistance diagnosis due to that fact that it is expensive and difficult to perform.This was another study limitation.As an alternative, fasting insulin and HOMA-IR, were used as surrogate markers [24].
We can conclude that patients with PCOS have significant ED and premature atherosclerosis which is, to a great extent, independent of obesity and IR.This suggests that PCOS patients are at increased risk for premature CVD and may benefit from early detection and management.We recommend screening for subclinical atherosclerotic disease in PCOS women which is a cause of increased morbidity and mortality especially with co-existing different cardio-metabolic disorders such as hypertension and diabetes mellitus.

Table 2 .
comparison of PcoS cases with insulin resistant and without insulin resistance regarding total cholesterol and ldl-c .

Table 3 .
correlation analysis with HoMa-ir and baseline demographic characteristics, metabolic and hormone profiles in the Polycystic ovary Syndrome (PcoS) patients.

Table 1 .
ciMT of the PcoS group in comparison with control subjects.

Table 4 .
Percentage of brachial artery diameter change 3 min after reactive hyperemia in PcoS group and control group.

Table 5 .
linear regression with ciMT& ed as dependent variable and PcoS as independent predictor.