Is There Any Relation Between Insulin Resistance and Menstrual Irregularity in Obese Women?

Objectives: Menstrual irregularity is a common issue in obese women that can cause anovulation and infertility. Insulin resistance is also common in obese women. We aimed to nd the relation between menstrual pattern and insulin resistance in obese women. Results: The samples were 233 obese women 18-36 years old coming to obesity clinic for treatment of obesity. We checked Homeostatic Model Assessment for Insulin Resistance (HOMA – IR) and lipid prole in the clients and also we lled a form including the age, educational status, marital status, menstrual pattern and severity of hirsutism. Sample size was 233. The mean age of women was 25.6±8 years. The mean Body mass index (BMI) was 44.6±4.8 kg/m2. 84.55% of women had insulin resistance. There was signicant relation between menstrual pattern and fasting insulin level (p=0.05). There was no signicant relation between presence of Insulin resistance and menstrual pattern (p=0.28). There was signicant relation between age and menstrual pattern (p=0.007). Based on the results, there was no signicant relation between HOMA- IR and menstrual pattern (p=0.28). Perhaps type of obesity (android) may have relation with menstrual pattern and also we need to a standard IR denition to explain the relation between different ndings. HOMA-IR: Homeostatic Model Assessment for Insulin Resistance; BMI: Body mass index PCO: PolyCystic Ovarian; OGTT: oral glucose tolerance test; FBG: fasting blood glucose HDL: high-density lipoproteins; LDL: low-density lipoproteins; HOMA: Homeostatic model assessment; DCI: D – Chiro Inositol; PCOS: Polycystic ovary syndrome; IH: idiopathic hirsutism; QUICKI: quantitative insulin sensitivity check index; TYG index: Triglyceride and glucose index


Introduction
Obesity is epidemic and even pandemic in the world and its prevalence is growing in adolescent. In 2013-2014 in USA, its prevalence in adolescent was 20.6% and in comparison in 1988-1994 in this age group was 10.5% [1][2][3]. The majority of overweight adolescence will become overweight adults. Obesity in girls will make advanced puberty and age at menarche has adverse relation with girls Body mass index (BMI) [4]. One third of women in USA are obese [5]. Obesity results to many complications like: Diabetes, Coronary diseases, hypertension,… and infertility.
Infertility in obese women is due to anovulation and menstrual irregularity. Insulin resistance (IR) is a common nding in obese women but it is not clear that the menstrual irregularity in obese women is due to IR or obesity by itself.
A survey was done on 36 women with Poly Cystic Ovarian (PCO) and 25 age and weight control women.
Fasting insulin and oral glucose tolerance test (OGTT) and ovarian morphology by (trans-vaginal ultrasound) were assessed. There was no signi cant association between insulin sensitivity and ovarian morphology or between insulin sensitivity and menstrual frequency. They concluded that PCO is associated with insulin resistance (IR) but BMI, hyperandrogenism and hyperandrogenemia are independent predictors of insulin sensitivity [6].
In this study on obese women we assessed the IR prevalence and the pattern of menstrual cycles to answer this question that either this is obesity that makes irregular menstrual pattern or combination of obesity with IR?

Methods
This is a prospective cross sectional study on obese women that came to Minimally Invasive Surgery Research Center for treatment of obesity. After consultation and taking inform consent, the women with BMI > 30 that had the inclusion criteria entered the study.

Inclusion Criteria
Eighteen to 36 years old obese women that did not use exogenous insulin were included in the study.
A data collection form was completed by the general practitioner including age, menarche age, marital status, parity, educational status, job, use of metformin, the severity of hirsutism, BMI and menstrual pattern. After 10 hours of fasting, the patients returned to laboratory of Minimally Invasive Surgery Research Center for following blood tests: Fasting blood glucose (FBS), Fasting Insulin, Cholesterol, triglyceride, low-density lipoproteins (LDL), high-density lipoproteins (HDL).
We described normal pattern of menstruation as the regular menses with interval 21-35 days. Abnormal menses were intervals less than 21 or more than 35 days.
The severity of hirsutism was estimated by general physician according to the modi ed Ferriman -Gallwey Scoring System for Hirsutism [7].
Insulin resistance was estimated by calculation of HOMA -IR.
Homeostatic model assessment (HOMA) is a method for assessing β cells function and insulin resistance (IR) can be detected from fasting blood sugar and insulin concentration, that was rst described in 1985 [8]. HOMA -IR calculation was by this formula: Fasting blood sugar (mg/dl) × fasting insulin (mU/lit)/405 . In our study If HOMA -IR was more than 2.5, it was described as insulin resistance [9].

Data analysis:
All analysis was performed using SPSS software version 25. Quantitative variables are presented by mean ± standard deviation (SD).
Data distribution normality was assessed with Kolmogorov -Smirnov test. If a parameter was not normally distributed we used Mann -Whitney test. Independent sample T -test and one way analysis of variance were used for comparing measurements in two groups. P value of < 0.05 was considered signi cant.

Page 4/10
This is a prospective cross sectional study between Jan 2018 -March 2019. The samples were 233 obese women 18-36 years old that came to Minimally Invasive Surgery Research Center for treatment of obesity. The mean age of women was 25.6 ± 8 years. The mean BMI was 44.6 ± 4.8. Table 1 shows quantitative variables.  Table 2 shows baseline variables.  There was signi cant relation between age and menstrual pattern (p = 0.007) (Additional le: Table S1).
There was signi cant relation between menstrual pattern and fasting insulin level (p = 0.05) (Additional le: Table S2).
There was no signi cant relation between BMI and menstrual pattern (p = 0.08) (Additional le: Table S3).
There was not signi cant relation between HOMA -IR and menstrual pattern (p = 0.28) (Additional le: Table S4).
There was not signi cant relation between HOMA -IR and severity of hirsutism (p = 0.18) (Additional le: Table S5).

Discussion
In this study there was no signi cant relation between HOMA-IR and menstrual pattern (p = 0.28).
In a survey on 57 morbid obese women the samples were in two groups. In group one the morbid obese women had not hirsutism or menstrual irregularity and group 2 had menstrual dysfunctions with or without hirsutism. Thyroid function tests, FBS and serum lipid pro le in two groups had not signi cant difference but in group two HOMA -IR was signi cantly higher than group 1. They concluded that IR in morbid obese women can make menstrual irregularity even without elevated androgens [10]. But in our study in 82% of obese women with normal menstrual pattern and in 85.28% of obese women with abnormal menstrual pattern HOMA -IR was more than 2.5. On the other hand in some studies IR was not related with menstrual regularity [6].
The effectiveness of treatment with D -Chiro Inositol (DCI) in improving IR in patients with PCOS had been con rmed in several reports. In one retrospective study 47 women with PCOS were treated with DCI for 15 months. In 51.6% the serum AMH levels and IR indexes signi cantly decreased during the treatment. They concluded that DCI for treatment of PCOS makes improvement of IR and menstrual cycle irregularity [11].
In one study on biochemical characteristics of women with menstrual disturbances, the samples were 257 women with oligomenorrhea. Between women with oligomenorrhea 149 were hyper androgenic and 108 non hyper androgenic. They had 213 controls with normal cycles. Endocrine and metabolic parameters and IR were compared among different menstrual patterns. Hyper androgenic women with amenorrhea had higher levels of androgens and more lipid pro le disorders than hyper -androgenic women with oligomenorreha. However in non -androgenic women with amenorrhea the degree of IR and metabolic disturbances were similar to hyper androgenic women with oligomenorrhea. They resulted that in women without excess levels of androgens menstrual disturbance does not correlate with severity of IR and metabolic disturbances [12].
In one survey on 137 women with PCOS 84.7% had hyperandrogenism, 84.7% had oligo anovulation and 89% had PCO. 69.4% had classic phenotype of PCOS, 15.3% had ovulatory phenotype, 15.3% had normo androgenic phenotype. 71.4% of subjects were IR and IR frequency differed among phenotypes (p < 0.001). In normo androgenic phenotype there was not IR case and resulted that in normo androgenic PCOS women there is not IR and are different from other PCOS cases [13].
There is a survey on 70 non obese patients with hirsutism, to nd the role of IR on hirsutism. The samples were in three groups. 30 women with idiopathic hirsutism (IH) and 20 with hirsutism due to PCO and 20 healthy controls were assessed.
16 with IH and 17 with PCOS had IR. There was signi cant difference in fasting lnsulin levels and HOMA -IR in three groups, but there was no signi cant difference between IH and PCOS patients. Women with android fat deposition had higher fasting insulin levels and HOMA -IR. They concluded that IR occurs in non obese patients with IH, and may be related to android obesity [14].
By a research on 976 Korean people the cut of points for describing insulin resistance was calculated. The cut off point for fasting insulin was 12.94 mU/ml, for the quantitative insulin sensitivity check index (QUICKI) was 0.32 and for HOMA -IR was 3.04 [15]. In our research we described the IR as HOMA -IR > 2.5. There is heterogeneity for IR de nitions and in different studies HOMA -IR index cut off values to de ne IR was different. The IR cut off range in different studies was between 2 to 3.5 [16], and was dependent to gender, race, age, obesity, smoking and etc.
Triglyceride and glucose index (TYG index) also have been suggested for diagnosis of insulin resistance.
In a review article with 69922 participants the TYG index was evaluated by HOMA as reference test. The highest speci city was 99% with HOMA -IR with a cut off value of 4.68 [16]. In one study optimal insulin resistance cut off was 2 [17]. In another study IR cut off was 2.8 [18]. Considering various normal IR cut off de nitions, the relation between different variables is not reliable.
We concluded that the role of IR in obese women, on menstrual pattern is not clear.
Not only obesity by itself but also obesity with combination with hirsutism or IR does not have signi cant relation with menstrual pattern.

Limitations
Perhaps type of obesity (android) may have relation with menstrual pattern. But this is just half of the picture and we need to a standard IR de nition to explain the relation between different ndings.