Food habits and physical activities of women with hirsute-polycystic ovary syndrome

Polycystic ovary syndrome (PCOS) is a multisystem disease affecting females of childbearing age. This disease imposes a huge economic, social and psychological burden on the patient. The first line management of PCOS is lifestyle modification. However, there are no Sri Lankan data about the diet and physical activity of these women. This is a case-control study to compare the dietary habits and physical activity between 99 hirsute-PCOS women and 100 age-matched controls. We compared the dietary intake and physical activity of the two groups using a validated food frequency chart and a physical activity questionnaire. The Student t test and Mann-Whitney U test were used for statistical analysis. Food consumption pattern was similar in both groups except for a significantly higher consumption of eggs, refined carbohydrates, pulses and grains in controls. PCOS patents in the BMI category of 18.5 to 24.9 kg/m2 consumed significantly less amounts of fish, eggs, dairy products, pulses and grains. With regard to physical activity, light, moderate and heavy exercise was significantly higher in controls. It is concluded that hirsute PCOS women seem to have different dietary habit and reduced physical activity than controls.


INTRODUCTION
Polycystic ovary syndrome (PCOS) is one of the commonest endocrine disorders affecting women of childbearing age. It comprises endocrine, metabolic and reproductive abnormalities and is associated with a higher incidence of impaired glucose tolerance, type 2 diabetes, hyperinsulinemia, unfavorable lipid profile, increased risk of cardiovascular diseases, subfertility, obesity, fatty liver disease and mood disorders (1). About 80% of the PCOS population consists of a phenotype associated with a high body mass index (BMI) and increased waist to hip ratio (2). This phenotype has also been shown to have worse metabolic features (3). Additionally, there is evidence showing a difference between the diet and physical activity of PCOS patients and controls (4,5).
Lower basal metabolic rate, higher tendency to eat more due to a bulimia-like disorder driven by androgens, and lower preceded satiety and increased appetite, when compared to controls, may be the contributory factors for the higher BMI (6,7). Lower preceded satiety and increased appetite are thought to be due to higher orexigenic hormones (ghrelin) and lower anorexigenic hormones (cholecystokinin) (8). This phenomenon is also linked to the increased serum leptin and leptin resistance, which is associated with abdominal obesity (9). The adolescents with PCOS have been shown to consume a surplus of calories than controls with comparable BMI (10). Therefore, lifestyle changes which consist of dietary modifications and increasing physical activity are aimed at changing the body weight and reducing the complications of this complex disease.
A hypocaloric diet has been shown to improve the hyperandrogenism and insulin resistance of obese PCOS patients. Weight loss has been shown to induce spontaneous ovulation and improve metabolic and hormonal derangements in these patients (11,12). Although the effect of weight loss on reproductive health has not been studied in controlled trials, an uncontrolled study consisting of 33 anovulatory PCOS patients has shown that reducing 5-10% of weight by diet and physical exercise leads to reduced ovarian volume and micro follicle number and spontaneous pregnancy (13). However, the data on the effects of dietary modification in PCOS patients with normal BMI are limited.
Lifestyle modification can be arranged once the baseline dietary habits and physical activity are well documented. Studies examining the pattern of food intake and physical activity of PCOS patients in Sri Lanka are lacking. Therefore, this case-control study was done to compare the dietary habits and physical activity of PCOS patients with controls. April 2015 and ended after 2 years. The convenience sampling method was used. The exclusion criteria were secondary causes for hyperandrogenism (21-hydroxylase-deficient nonclassic congenital adrenal hyperplasia, classic congenital adrenal hyperplasia, androgensecreting neoplasm, side effects from medication, hypothyroidism, hyperprolactinemia, Cushing's syndrome), pregnancy or lactation, presence of chronic systemic disease or taking medications affecting appetite or weight. The exclusion criteria were determined based on the history and examination findings.

SUBJECTS AND METHODS
A validated, interviewer-administered food frequency questionnaire was used to collect dietary data. The participants were asked to provide answers based on their food consumption patterns and frequency of different foods in a daily and a seven-day food recall (14,15)

Assessment of the level of physical activity
The short version of the International Physical Activity Questionnaire (IPAQ) was employed to assess physical activity during a day and seven-day period (16).
Physical activity data were recorded by selfreported recall with regard to frequency and duration of three categories of activity lasting 10 minutes or more, i.e. light exercise (like walking briskly for recreation or exercise or to get to or from places), moderate exercise (like sweeping, gardening, lifting moderately heavy goods, moderate exercise classes, recreational swimming, dancing) and heavy exercise that makes a person breathe harder or puff and pant (like cycling fast, chopping firewood, lifting heavy goods, farming). Physical activity for the whole week was converted to metabolic equivalent of task (MET) minutes in each of the three categories of physical activity by multiplying with the metabolic equivalent value assigned to each category: light exercise minutes × 3.3 METs, moderate exercise minutes × 4.0 METs and heavy exercise minutes × 8 METs (17).
One MET is defined as the amount of oxygen consumed while sitting at rest and is equal to 3.5 ml oxygen per kg body weight. The MET minute is used as a method to indicate and compare the absolute intensity and energy expenditure of different physical activities.
Anthropometric measurements -The body weight and height were measured by standard techniques. BMI was calculated by dividing weight in kilograms by the square of height in meters Based on previous studies, the sample size was calculated using the formula: Where is the standard deviation and is the sample mean. The value of α was taken at 95% confidence level and β at 80% confidence level According to the calculation, a sample size of 100 patients was needed for the study

Statistical analysis
All statistical analyses were carried out using SPSS (version 20) and p values < 0.05 were considered statistically significant. For data with a normal distribution (BMI data), student t-test was used to identify differences between the two groups. Mann Whitney U test was used for food frequency data and physical activity data that were not normally distributed.

RESULTS
There were 99 hirsute-PCOS patients and 100 control subjects. They were between 19-43 years of age. The mean BMI of PCOS group was 27.48 (±5.13) kg/m 2 and controls was 20.86 (±4.71) kg/m 2 and this difference was statistically significant (p < 0.001). Table 1 shows the number of patients in different BMI categories. For the analysis, the dietary questionnaire was categorized to 8 main food categories (meat, fish, egg, pulses and grains, vegetables and fruits, refined carbohydrates, fried foods and dairy products).
The consumption of meat, fish, vegetables and fruits, fried foods and dairy products was equal between the two groups. Eggs, refined carbohydrate, pulses and grains consumption was statistically significantly higher in controls (

DISCUSSION
In this study, consumption of eggs, refined carbohydrates, pulses and grains was significantly higher in controls. Although refined carbohydrates are considered to promote weight gain and impaired glucose tolerance, our cohort of hirsute-PCOS patients had consumed significantly low amounts of refined carbohydrates than the controls and it is a favorable trend. A similar study showed PCOS group consumed significantly more white bread (7.9 +/-4.4 vs. 5.5 +/-2.9 servings over 4 days) and tended to consume more fried potatoes than did the control group which is in contrast to our findings (18).
The consumption of meat, fish, dairy products, fried foods, vegetables and fruits was equal in both groups. In both PCOS patients and controls, daily fruit and vegetable consumption was 3.7 and 3.2 times per day respectively. The recommended daily consumption of fruits and vegetables of 5 servings was not achieved by both groups. Therefore, in general, the fruits and vegetable consumption was not satisfactory in our cohorts of people. The reason could be the higher price of fruits and vegetables compared to starchy foods.
Finding of this study suggest that non-PCOS individuals with normal BMI (18.5 -24.9 kg/m 2 ) consume more healthy foods which contain fibers and protein than people with PCOS and same BMI. The reasons for PCOS patients with normal BMI consuming less amount of fiber and protein need further evaluation.
A considerable amount of evidence supports the fact that the patients with PCOS consume an equal amount of macro and micronutrients to non-PCOS women regardless of their BMI (19,20). But some studies show that women with PCOS consume more calories and less fiber than women without PCOS (4,21). Surprisingly, some studies show that women with PCOS eat a healthier diet than women without PCOS (17).
In our study, all categories of physical activity were less in hirsute-PCOS patients. In both groups, all were mostly engaged in light exercises. The difference in the amount of heavy and moderate exercise was not statistically significant. However, the control group had engaged in significantly more light exercise than PCOS patients.
However, some PCOS women with equal amount of energy intake to normal controls show higher body weight than controls suggesting that diet alone may not be the contributory factor for higher BMI (17)(22). These findings suggest that it may be the type of specific food rather than the total energy intake and lack of physical activity that contribute to higher BMI (17).
Food habits differ considerably from country to country and the choice depends on the local availability and cultural preferences. There is no Sri Lankan study that has examined the dietary habits and physical activity of hirsute-PCOS patients. Lifestyle modification could be achieved better if data on the dietary habits and the physical activity of our PCOS patients are available.
This study suggests that hirsute-PCOS patients did not differ significantly from controls concerning diet but their physical activity of all intensities is significantly lower. When the physical activity was compared according to BMI categories, patients in BMI category of 18.5 -24.9 had a significantly lower amount of physical activity than the controls. However, the number of patients in the other three BMI categories was not adequate to compare the true difference.
The limitations of our study were that the baseline BMI did not match and data on the economic status and the social background were not collected (since they may affect the food consumption and physical activity). However, it is difficult to get a BMI matched control group as PCOS patients tend to have higher BMI.