Arterial hypertension and metabolic profile in patients with polycystic ovary syndrome

PURPOSE: To evaluate parameters related with arterial pressure and metabolic profile in women with polycystic ovary syndrome (POS). METHODS: This monocentric study at the University Hospital Endocrinology Section included 60 women aged 18-45 years, 42 being diagnosed with POS and acting as 18 controls. All women were subjected to transvaginal ultrasound and monitored for arterial pressure for 24 h in the ambulatory (MAP). Venous blood samples were taken between 07.00 and 09.00, after 12 h fasting. Basal (BG) and fasting glucose concentrations, total cholesterol and its fractions, triglycerides and insulin (to calculate the homeostatic assay insulin-resistance, HOMA-IR) were measured. Collected data were the mean arterial blood pressure (24-h awake/sleep cycle), arterial pressure nocturnal descensus, glycemia and fasting glucose for HOMA-IR, and lipid profile. The Student’s t test was used to compare homogeneous variables; the Mann-Whitney test was used to compare non-homogeneous variables; the Pearson’s correlation coefficient was used to search for correlation between the variables. The χ2 test was used for comparison of the absence of nocturnal descensus. Significance was taken as p<0.05. RESULTS: The mean age of the patients with POS was 27.4±5.5 (18–45 years, n=42) and the body mass index (BMI) was 30.2±6.5 kg/m2 (18.3–54.9). In the Control Group, the mean age was 31.4±6.1 (18–45 years) and the BMI was 27.1±6.2 kg/m2 (18.3–54.9, n=18). No difference in the metabolic parameters and insulin resistance was observed between the two groups. Comparison between these parameters and MAP showed that the only parameter with a correlation was the BMI, independent of the POS diagnosis. This was not seen in nocturnal descensus, which was uncorrelated with POS and any of the other studied parameters. CONCLUSION: POS women do not show higher arterial blood pressure, glycemia, HDL-col, TG, HOMA-IR and BMI compared to non-POS women. However, POS patients showed correlation between arterial pressure and BMI, suggesting that obesity is a primary factor involved in arterial pressure changes in these patients.


Introduction
Polycystic ovary syndrome (POS) is a heterogeneous clinical condition affecting 1 in 15 women in the world 1 , and is the most common endocrine disorder at a reproductive age 2,3 , varying from 6-10%, according to diagnosing criteria 4,5 .In spite of its physiopathology being not fully understood, it is of multifactorial genesis, with genetical, metabolic and neuroendocrine influences 6,7 , and a fundamental insulin-resistance background 8 .
In addition to classic chronic anovulation and hyperandrogenism, POS leads to higher risk of cardiovascular problems related to insulin resistance 9 , such as systemic arterial hypertension (SAH), glucose intolerance and type 2 diabetes mellitus, dyslipidemia, obesity and metabolic syndrome [10][11][12][13] .
One of the risk factors for cardiovascular diseases is SAH, diagnosed in one third of the population worldwide 14,15 .Given its high prevalence 16 and high morbidmortality rates, it is of socio-economic importance.
An association between SAH and POS is not completely established 9 .Insulin resistance and hyperinsulinemia are predisposing factors for SAH in POS women 17 , but no study has shown a direct correlation between the arterial pressure and the insulin resistance parameters in these patients.In spite of the comorbidity related to insulin resistance in POS patients influencing arterial pressure 18 , it is not known if the POS patients are characterized by differential arterial pressure (AP) compared to other women, and whether its parameters would be correlated with other cardiovascular risk factors.Mean arterial pressure (MAP) studies to evaluate arterial pressure in POS patients are scarce.
Thus we decided to study the changes in arterial pressure and metabolic profile in POS patients.

Methods
Sixty women who attended the Hospital Clementino Fraga Filho (HUCFF) Endocrinology Section voluntarily between 2010 and 2012 were enrolled in this study.The study was approved by the Hospital's Ethics and Research Committee under the number 085/09.All volunteers agreeing to participate in the study signed an informed consent form, according to the National Health Council (Resolution 196/96).
Women aged 18 to 45 included in the study presented symptoms related to hyperandrogenism (hirsutism, acne, alopecia), anovulation (amenorrhea, oligomenorrhea, infertility) and insulin resistance (obesity, acantose nigricans).The POS diagnosing criteria were those established by the Rotterdam consensus 19 , and included two of the three following criteria: oligomenorrhea or anovulation; clinical or biochemical signs of hyperandrogenism, polycystic ovaries confirmed by ultrasonography (presence of 12 or more 2-9 mm diameter follicles in each ovary and/or ovarian volume >10 mm 3 ).Women included in the Control Group were in the same age range, had similar complaints as the ones from the POS group, but did not fulfill the diagnostic criteria.
Women showing other correlated diseases resulting in hyperandrogenism and/or anovulation, such as thyroid disfunction, hyperprolactinemia, classical and non-classical congenital adrenal hyperplasia, Cushing syndrome and androgen-producing tumors of the adrenal gland and ovaries, were not included in the study.Women administered with medicine contributing to false-positive results (corticoids, anticoagulants, oral contraceptives, anti-lipidemic or hypoglycemic drugs etc.) were also excluded.Some patients interrupted contraceptive treatment to take part in the study.
Evaluated parameters were age, symptoms associated with hyperandrogenism and/or anovulation, comorbidities, medications in use, physiological and familiar risk factors.Patients were given a clinical examination to determine body mass (kg), height (m), waist circumference (WC), and arterial pressure.WC was measured at the mean point between the costal margin and the anterosuperior iliac crest.BMI was determined by the Quetelet's 20 rate between body mass (kg) divided by height in m 2 .A diagnosis of overweight/obesity was determined according to the WHO criteria 20 : normal (18.5-24.9kg/m 2 ), overweight (25-29.9kg/m 2 ), and obese (>30 kg/m 2 ).In addition, insulin resistance signs, such as acantose nigricans in skin folding region and signs of hyperandrogenism, such as hirsutism, acne and alopecia.Hirsutism was diagnosed according to the modified semiquantitative Ferriman-Gallwey score, using a cut off of 8 21 .
All women were submitted to transvaginal ultrasound examination to determine the ovarian topography to determine ultrasound criteria established by the Rotterdam consensus.
MAP was measured for 24 h, using a Dynamapa monitor that involved the oscilometric method.MAP along the awake and sleep cycle and nocturnal descensus were determined.The results were interpreted according to the V Brazilian Directions for Ambulatory Monitoring of Arterial Pressure 22 .
Criteria established by the American Diabetes Association 23 were to determine glucose value and those from NCEP, ATP III 24 , for lipid profiles.Venous blood samples were collected between 07.00 and 09.00 after 12 h fasting.Fasting and basal (BG) glucose levels, total cholesterol and fractions, triglycerides and insulin (to calculate HOMA-IR) levels were measured.Serum glucose was measured enzymatically, with a level lower than 99 mg/dL as reference.Total cholesterol was determined by a colorimetric enzymatic method, with a reference level <200 mg/dL; LDL-col was measured by the Friedwald formula; triglycerides were determined by a colorimetric enzymatic method, using a reference level lower than 150 mg/dL; insulin was determined by a electrochemoluminescence method, using reference levels between 2 and 13 mcU/mL (for BMI up to 25), 2 and 19 mcU/mL (for BMIs of 25-30), and 2 and 23 mcU/mL (for BMI>30).
Student's t-test was used to compare the mean of the variables in arterial pressure, glucose, HDL-col, triglycerides and HOMA-IR between the obese and non-obese POS patients.To compare the distribution of the means of variables arterial pressure, glucose, HDLcol, triglycerides, and HOMA-IR between the control and POS groups, and within the POS group those with and without nocturnal descensus, the non-parametric Mann Whitney test.To establish any correlation between the metabolic paramenters and insulin resistance with the mean arterial pressure within the POS group, the Pearson correlation coefficient was used (p>0=positive correlation; p<0=negative correlation; p=0: no correlation; p: 0-3.0=weak correlation; p: 3.0-7.0=moderatecorrelation; p>7,0=strong correlation).For those parameters showing some correlation, a simple linear regression helped to quantify the influence of these parameters in MAP.To compare the proportion of the absence of nocturnal descensus between the control and POS groups, and between the obese and non-obese POS, the χ 2 test was employed, the values being expressed as percentages.
The results have been expressed as the mean ± standard deviation, and p<0.05 was taken as statistically significant.Statistical Analysis System (SAS) software for Windows, version 9.1, was used throughout this work.
The mean values of systolic and diastolic arterial pressures in the 24 h awake/sleep cycle are given in Table 2. Sixteen POS patients (36.8%) and six controls (33.3%) did not show nocturnal descensus, with no difference between the two groups (p=0.95).There was no significant difference between POS and control with respect to the metabolic parameters, insulin resistence and MAP parameters (mean values for systolic and diastolic arterial pressure, and nocturnal descensus).This was also observed when the obese women in the two groups were compared.

Discussion
There was no difference between the metabolic parameters (RI, glucose, HDL and triglycerides) between the women diagnosed with POS and Control Group.However, the prevalence of this metabolic alteration was significantly higher in obese and non-obese POS women, revealing the relevant role of obesity in the metabolic profile of POS patients.
In spite of the prevalence of metabolic disorders in POS patients 25,26 , no difference was seen in the glucose, HDL, triglycerides, HOMA-IR and BMI levels between the POS and control women.This might be correlated with the fact that both groups showed metabolic syndrome symptoms, hyperandrogenism and anovulation.This similarity between the clinical presentations in the two groups could perhaps result from equivalent metabolic characteristics.However, when obese and non-obese POS patients were compared, significantly different levels of glucose, triglycerides and HOMA-IR were found.Studies on the BMI as a risk factor for POS are conflicting.Some authors stress that BMI influences the prevalence of metabolic disorders in POS patients, while other claim that a poorer metabolic profile of POS patients is independent of BMI 27,28 .
SAH is a risk factor for different diseases and, as such, it is of high socio-economic concerns.The association between POS and SAH is not well established.The actual mechanism responsible for SAH in POS patients is a matter of great controversy.In addition to a crucial role in SAH in POS patients, other mechanisms may be involved, such as the characteristic pro-inflammatory state in the syndrome, hyperandrogenism, and the influence of both the angiotensin-aldosterone and sympathetic system 10,29,30 .
By comparing the metabolic parameters and the HOMA-IR with mean arterial pressure, the only variable with a positive correlation was with BMI.In POS patients, a positive correlation between BMI and MAP was apparent.A moderate correlation existed between the BMI and the systolic arterial pressure, and a weak correlation existed between BMI and diastolic arterial pressure (Table 3).
Within the POS group, there were significant differences between the non-obese (BMI<29.9)and obese women (BMI>30) with respect to glycemia, triglycerides, HOMA-IR, and the mean arterial pressure (Table 4).
There was no difference with respect to the metabolic profile, HOMA-IR, and mean arterial pressure when the POS patients with and without nocturnal descensus were compared to each other (Table 5).
Finally, the presence or absence of nocturnal descensus did not show statistically significant differences with respect to POS (p=0.9).The same was observed for the BMI, meaning that the nocturnal descensus was not correlated with obesity (p=0.2).

Table 3. Pearson correlation coefficient between Body Mass Index with the mean arterial pressures
AP: arterial pressure; ASP: arterial systolic pressure; ADP: arterial diastolic pressure.The Pearson correlation coefficient was used (p>0= positive correlation; p<0= negative correlation; p=0: no correlation; p: 0-3.0=weak correlation; p: 3.0-7.0=moderatecorrelation; p>7.0= strong correlation).MAP is an appropriate way to inspect cardiovascular status in these patients 22 .By keeping indirect and intermittent registers of arterial pressure for 24 h, MAP could detect circadian cycle variations in arterial pressure, which had considerable prognostic implications.Among the parameters studied, the most indicative were the mean arterial pressure and the absence of nocturnal descensus.The mean arterial systolic pressure (ASP) and the mean arterial diastolic pressure (ADP) obtained in a 24 h period, including the awake/sleep cycle, show consistent correlations with lesions in the target organs and cardiovascular morbidity and mortality 31,32 .When comparing the mean arterial pressure detected by MAP in POS and control patients, no difference was found.These results are difficult to consider, since most of the results in the literature are controversial and use casual ambulatorial arterial pressure measurements, and MAP studies are scarce.

AP (mmHg) Correlation (P)
The only cardiovascular risk factor analysed in our investigation that showed a correlation with the arterial systolic and diastolic pressures was BMI.Two interesting results were a moderate correlation between the mean arterial systolic pressure, and a weak correlation between the mean arterial diastolic pressure, and POS.The same influence of BMI occurred between the mean arterial pressure and obesity in POS patients.A 5% significance was found in the comparison between these parameters, confirming previous reports indicating an association between obesity and arterial pressure 33,34 .Nonetheless, the differences between the levels of glycemia, triglycerides and HOMA-IR in the obese and non-obese groups was significant, confirming an interaction between the adipose tissue and the comorbidities related to the insulin resistance.
The adipose tissue is recognized as a complex endocrine organ, capable of influencing the sympathetic system and producing chemical signal that, together with insulin resistance, influence arterial pressure in POS women.Overstimulation of sympathetic system in obesity promotes an expansion of the extracellular volume and regional flux, causing increase cardiac debit 35 .Obese patients show higher plasma rennin activity, increasing the plasma levels of angiotensinogen, increased activity of tissue converting enzyme and higher plasma levels of aldosterone, implicating the renin-angiotensinaldosterone system in ovary physiology and pathology 36 .Another important aspect of adipose tissue physiology is hyperinsulinemia that compensates in the insulin resistance, which exercises a trophic effect on blood vessel smooth muscle, causing increased vascular resistance and arterial pressure 37 .POS might share a common genetic background with arterial hypertension and obesity; finding this trait could elucidate the mechanisms involved in these diseases and other comorbidities due to insulin resistance 38 .However, it is not possible to conclude that POS patients are at greater risk of cardiovascular diseases, and the validity of this association needs testing.
The correlation between nocturnal descensus and risk factors for cardiovascular disease remains debatable since the results are controversial 22,31,39 .In the present study, the risk of presenting nocturnal descensus was similar between the POS, the controls and their BMIs.No correlation existed between nocturnal descensus and the mean values of glycemia, HDL-col, triglycerides and HOMA-IR.Similarly, no correlation was found between MAP.These findings, in general, contradict those in the literature.This might be attributed to the fact that sleep disorders, such as obstructive dyspnea, were not excluded and could influence the prevalence of nocturnal descensus and present a relevant bias in our study 40 .
One advantage of our study was the uniform profile of the patients seeking treatment in our institution by presenting suspected POS, all of whom had the same diagnosis.In both groups -POS and Control -the complaints were related to hyperandrogenism, anovulation and obesity.POS diagnosis was careful and followed the Rotterdam consensus; other causes of hyperandrogenism were excluded.These criteria were fundamental to the reliability of the study.However, the small number of patients in the Control Group might have introduced some bias to the analyses.

Table 2 .
Mean values of systolic and diastolic arterial pressures in the 24 h/ awake/sleep cycle AP: arterial pressure; ASP: arterial systolic pressure; ADP: arterial diastolic pressure.Note: *p significant <0.05 using the the non-parametric Mann Whitney test.

Table 5 .
Clinical and biochemical characteristic of the polycystic ovary syndrome patients with and without nocturnal descensus