This patient web page highlights some of the important health issues involved with polycystic ovarian syndrome. This condition affects women throughout their entire life. The information contained herein is from several research publications. Researchers from all over the world are performing studies which have led to an explosion in our knowledge of PCOS. This will certainly lead to improved treatments. I feel that our patients strongly desire information concerning their health. I hope that this information a good adjunct to my personal discussions with them. I can be contacted at anytime to clarify what is presented here.
Depending on the diagnostic criteria used, PCOS in its "classical" form may affect 5 %- 7% of women. The ‘classical’ form includes elevated androgens and anovulation. Mild forms appear to exist characterized by elevated androgens but normal ovulation. Many women, about 16 %- 25 % of all, have polycystic appearing ovaries on ultrasound exam but are otherwise normal. Interestingly, a recent report showed that about 1/3 of these patients have abnormal serum lipid profiles and increased androgen production from the ovaries.
In 1990, a conference at the National Institutes of Health was convened to establish the diagnostic criteria for PCOS. Not all experts agreed 100 %. Definite criteria were increased androgens (64%), exclusion other causes (60%), exclusion of congenital adrenal hyperplasia (59%), menstrual dysfunction (52%), and clinical hyperandrogenism (45%). Possible criteria included insulin resistance (69%), increased LH/FSH (55%), ultrasound evidence of polycystic ovaries (52 %), clinical hyperandrogenism (52%), and menstrual dysfunction (45%). In 2003, the ESHRE and ASRM PCOS Consensus Workshop determined that patients need 2 out of the following 3: oligo- or anovulation ; clinical and/or biochemical hyperandrogenism; and/or PCO appearance on pelvic ultrasound exam. All other causes should be excluded.
In summary, it is best to consider PCOS as increased androgens clinically (acne, excessive hair on face, abdomen, or thinning of scalp hair) or in the blood (total or free testosterone, DHEAS), with oligo-ovulation (cycles greater the every 35 days, low mid-luteal progesterone, monophasic basal body temperature chart). It is important to exclude other disorders of the thyroid (hyper or hypothyroidism), pituitary (hyperprolactinemia), and adrenal gland (Cushing’s Syndrome, late onset congenital adrenal hyperplasia). A pelvic ultrasound may show the classic ‘necklace sign’. The European Criteria includes more than 12 follicles 2mm – 9 mm in each ovary or a volume of greater than 10 ml.
1. The Ovary - A Source of Androgens in PCOS
The ovaries are involved in a vicious cycle where they are stimulated to produce excessive androgens which leads to abnormal ovarian follicular growth, lack of ovulation, and the skin manifestations of the syndrome. The ovary has two functional compartments. The theca cells surround the follicles. These produce androgens which act as precursors of the follicular estrogen production by granulosa cells. Elevated insulin, or tonic elevations of luteinizing hormone (LH) may lead to excessive androgen production.
2. Insulin Resistance and PCOS
Our understanding of the metabolic factors that play a role in PCOS has advanced dramatically over the past few years. In now appears that a metabolic state exists where the patient is resistant to some of the effects of insulin resulting in elevated insulin levels which appear to alter ovarian function. The metabolic picture is somewhat variable. In hirsute women that ovulate, slight increases in insulin and total and free testosterone is noted. The glucose to insulin ratio may be decreased which is consistent with a metabolic dysfunction. In non-hirsute women that ovulate but have polycystic ovaries on ultrasound slight increases in free testosterone, DHEAS and insulin are seen which is also consistent with a metabolic condition. No real long term studies have been done on these more mild forms of PCOS at present.
One way to think about hormone resistance is that the body does not see the effects of the hormonal signal it sends out. Insulin is normally secreted by the pancreas into the blood stream where it travels to distant organs (i.e., liver, muscle, fat) and binds a receptor on a cell to transmit a chemical signal. It is kind of like turning on a light switch. It seems that something in the switch is not working well so that the signal is not completely turned on. Thus, it begins to over-produce the hormone in order to compensate for the perceived lack of effects. The ovaries are excessively stimulated by insulin to produce abnormal amounts of androgen and fail to ovulate normally.
The following paragraphs show some of the reasons why we think insulin over-secretion and resistance is present in PCOS. Some women may show signs of insulin resistance by careful examination of the skin. Darkened patches around the neck and under the breasts or arm. This is called acanthosis nigricans. The photo below shows this.
Other women may have elevated serum insulin levels, impaired glucose tolerance or diabetes mellitus. Some investigators have measured insulin production over several hours and found that it is increased. This is shown below.
Others have shown that both lean and obese patients with PCOS have an exaggerated response of serum insulin in response to a glucose challenge test. In addition , the obese patients have abnormal glucose levels (i.e. diabetes). This study is very important because it shows that all patients with PCOS have insulin resistance which is unique and intrinsic to the condition and that obese patients with PCOS have a form of insulin resistance that is related to obesity which is superimposed.
Another study examined PCOS patients that had a positive family history of Diabetes. It was interesting that there was an uncoupling of the insulin and glucose. This indicated that the pancreas is not working normally in the patients with PCOS and may be an early manifestation of future problems of diabetes. Normally insulin and glucose are tightly linked. When glucose levels rise after a meal, insulin secretion from the pancreas increases in like fashion. They are coupled. In PCOS patients with a family history of Diabetes but not Diabetes themselves, the glucose and insulin were not tightly linked. At times glucose would be low and insulin high and vice versa.
In general, obese patients have impaired glucose tolerance (31%) or diabetes (7.5%) and lean PCOS patients have impaired glucose tolerance (10 %) and diabetes (1.5%). These are significantly increased compared to the normal population.
Taken together, all of these studies point to insulin resistance as a crucial metabolic derangement in PCOS. This has lead to novel treatments which will be discussed later.
3. Obesity and PCOS
Obesity is very common in patients with PCOS . Patients with obesity have significant health related risks. Much of the US population is overweight. There is a J-shaped curve of mortality related to obesity. Dr. Donahue can calculate your Body Mass Index (BMI).
Obviously reducing weight has many health benefits. It is of interest that the distribution of fat also plays a role in PCOS. Patients with “central” or ‘android’ obesity (apple shape) have more severe insulin resistance compared to people with ‘gynoid’ obesity (pear shape). This contributes to the syndrome. Several physical measurements have been assessed in patients with PCOS. The waist circumference correlates well with the hormonal abnormalities. A waist-hip-ratio (WHR) greater than 0.8 is associated with higher androgen levels and insulin resistance. Body mass index does not take into account patient habitus. This may be a less sensitive indicator.
4. Genetics and Family History
The difficulty in studying the genetics of PCOS is due to the variable expression of the condition, poor agreement of diagnostic criteria, and lack of a male PCOS patient. Additionally, factors such as environment (fat/carbohydrate intake), exercise, peri-pubertal stress are important but difficult to reliably quantify.
In a study using strict diagnostic criteria, investigators from Birmingham, Alabama studied 218 patients. 168 (77%) had another female relative with PCOS, ½ of the mothers and sisters, ¼ maternal and paternal aunts had PCOS. This was consistent with an autosomal dominant with variable phenotype inheritance. In the next few years, it is hoped that the actual gene defects will be discovered. Studies have shown with age and weight matched controls that women with PCOS and without PCOS had similar nutrient intake (calories, carbs, protein) and physical activity (light, moderate, vigorous). Lean women with PCOS did, however, have a lower caloric intake. Thus, the obesity in PCOS is not simply because the patients are eating more and exercising less. Clearly, individual metabolism plays a role.
5. Puberty and PCOS
Many patients are able to date their symptoms to their puberty. Several studies support that for many patients there are hormonal imbalances at this time. Normally, the progression through puberty is characterized by irregular menstrual cycles, most of which are not associated with ovulation. During the 1st year after menarche (time of 1st menses) only 15 % of the cycles are ovulatory, by the 3rd year 40 % , and by the 6th year 70 % are ovulatory.
A study of hyperandrogenic girls between the ages of 11 and 18 years when compared to age matched normally cycling girls were found to have increased secretion of Luteinizing Hormone, increased LH/FSH ratio, increased insulin, and decreased insulin-like growth factor binding protein-1. Thus, there appears to be hormonal differences in these young patients with PCOS.
A study by Apter and Vihko in 1990 examined long term 200 girls between the ages of 7 and 17 in 1974. In 1992, patients that had not yet conceived had elevated testosterone levels. There was an inverse relationship between the testosterone levels as a teen and the likelihood to become pregnant.
In a study by Reichart and Southam, patients were followed between 1941 and 1964. If they had irregular cycles for more than 2 years, 2/3 still had abnormal menstruation as they got older. Dhalgren, in 1992, followed women that had PCOS and a wedge resection between 1956 and 1965. Many still had irregular cycles after the age of 40. Post-menopausal patients had elevated testosterone, LH, FSH, and estrone. They had more hysterectomies, entered menopause later, increased central obesity, insulin levels, diabetes, and hypertension.
Thus, it appears that patients with PCOS experience distinct hormonal abnormalities that, with other symptoms, persist throughout the reproductive years and beyond.
6. Cardiovascular Risk Factors and PCOS
Patients with PCOS have abnormal lipid profile including elevated triglyceride, LDL-Cholesterol, VLDL-Cholesterol, apoB, and decreased HDL. These have been linked to cardiovascular disease, in general.
In a study of over 200 patients with PCOS, Talbott found increased BMI, insulin, triglyceride, cholesterol, LDL, and blood pressure. The elevated insulin levels were found to correlate with the increased cardiovascular risk independently in PCOS patients.
In a very interesting study by Birdsall and collegues examining 143 women who had cardiac catheterization at less than 60 years of age 42% had PCOS. These patients had more extensive disease.
A study of over 1400 Swedish patients found that the myocardial infarction risk rate was 4 fold higher in patients between 40 and 49 years of age and 11 fold higher in patients between 50 and 61 years of age.
Overall, these studies suggest that PCOS patients have increased risk for cardiovascular disease. Ultimately, more large prospective population based studies will be completed to lend more support to the assertion that cardiovascular risk is increased.
7. Cancer and PCOS
It has been known that patients with PCOS are at increased risk for Endometrial cancer for many years. The unopposed estrogen (i.e. lack of ovulation and subsequent menses) is thought to continuously stimulate the uterine lining eventually leading to cancer. Most patients with endometrial cancer are older than 40. The risk factors the have include obesity, hypertension, anovulation, nulli-parity.
In a study of patients less than the age of 40, 25 % were found to have PCOS. Breast cancer was found to be decreased 50 % in a study of over 4700 women. More studies are needed to confirm these findings. Ovarian cancer was found to be 2-3 fold increased in PCOS patients, and this was greatest in non-obese women without prior birth control pill use. Obviously, we are learning more about these risks and the causal relationships have yet to be determined.
8. Pregnancy complications and PCOS
Pregnancy complications are increased in patients with PCOS. In one study, 5/13 patients with PCOS developed gestational diabetes by the 3rd trimester. Additionally, increases in pre-eclampsia, pre-term labor, and stillborns have been reported. The risk of early 1st trimester pregnancy loss is 30 %, compared to about 15 % in the general population. This high miscarriage rate may be due to low progesterone levels, tonic elevations of Luteinizing Hormone, endometrial dysfunction (luteal phase defect), or poor eggs derives from degenerating follicles. It is very possible that the abnormalities discussed above play a role in these problems. Recently, insulin resistance has been shown to be associated with decreased uterine vascularity, decreased IGF-Binding Protein 1 (a protein that helps implantation), and decreased glycodelin (a protein that protects the fetus from attack by the maternal immune system). Studies have shown that on-going pregnancies are increased when the insulin resistance is treated.
Hopefully, by better pre-conceptual treatment we can decrease some of these complications. Studies have shown that the use of metformin to reduce insulin levels lowers the miscarriage rate.
9. Fertility Treatments - New options
The recognition that insulin plays a key role in PCOS, as shown above, has led to new treatment options involving insulin sensitizing and lowering drugs. In the past, clinicians would often begin their treatment regimen with several months of clomid. If the patients failed to conceive on clomid , they would often begin several cycles of gonadotropins (i.e. Pergonal, Metrodin). Often the patients would hyperstimulate or have high order multiple gestations or have their cycle cancelled. When all failed, the use of IVF or GIFT was considered next. All of these treatment strategies were geared towards stimulation of the ovaries either indirectly with clomid (it works at the level of the brain to stimulate the ovaries), or directly with gonadotropins (this basically purified follicle stimulating hormone which stimulates the ovaries). None of them directly treat the underlying mechanism of the disease (i.e. insulin resistance). Over the past few years, the additions of metformin (glucophage) or thiazolidinediones (Actos, Avandia) have been used to directly treat the elevated insulin which is present in PCOS patients. Aromatase Inhibitors (i.e. Letrozole) are non-steroidal suppressors of estrogen biosynthesis that have been used in ovulation induction.
Metformin (glucophage) has been used in the treatment of diabetes for about 40 years. The effects of the drug are therefore well known. It is a category B drug in pregnancy (safe, no known fetal malformations in humans). In the liver it suppresses gluconeogenesis (production of glucose), and in muscle and fat cells it enhanced glucose uptake and utilization. This effectively lowers glucose and insulin levels. In studies performed in 1998, 34 % of patients taking metformin alone ovulated (compared to 4 % taking a placebo); and when metformin was combined with clomid, 90 % of the patients ovulated compared to only 8 % that took the placebo. Thus, the combination of metformin and clomid markedly improved ovulation in PCOS patients by correcting the underlying metabolic problem.
Thiazolidinediones (TZDs - Actos, Avandia) effectively lower insulin levels by effects of target cells that occur downstream from the actual binding of Insulin to the insulin receptor. Insulin resistance is therefore decreased.
Aromatase Inhibitors (Letrozole). Please refer to the other section of our web site for information on ovulation induction with aromatase inhibitors in PCOS. At present, the drug is not FDA approved for this indication and the manufacturer states it should not be used in women that may conceive.
10. Surgical Treatment of PCOS (Ovarian Drilling, Wedge Resection)
Stein and Leventhal performed ovarian wedge resection over 80 years ago and noted ‘astonishingly good results’ of regular menses and progress in some patients. In 1935, they reported on 7 patients that regained regular menstruation after surgery. Many studies have been performed over the years, some with impressive results. Unfortunately, the majority have not been well controlled from a research methodology standpoint. That is, other factors than surgery may have actually caused the onset of menstruation and the statistical analysis may have been flawed. Post operatively, some of the hormonal changes were reported to be transitory. Decreases in testosterone and estradiol were seen, but no changes in FSH or LH. Often the studies did not show these changes so it is hard to be sure of the actual hormonal changes of their significance.
The issue of post-operative adhesion formation has concerned fertility specialists for many years. One study showed that all patients that had laparotomy (bikini abdominal incision) developed adhesion. In one study, the researchers expected to see a 75% cumulative pregnancy rate but actually saw only 48%, possibly due to adhesion formation after surgery. Laparoscopy (belly button incision) may also cause adhesions.
The major problem with many of these studies is lack of uniformity and in general small study sizes. One study reviewed a total of 706 patients, 83% ovulated and 55% achieved pregnancy. Still, a major concern is adhesion formation which appears to be seen in approximately 80%-100%. If severe adhesions form, patients may develop pelvic pain or need IVF to conceive.
It is important that we have a discussion concerning the risks and benefits of this treatment before proceeding with this therapy.
11. Non-fertility Treatments - Acne and Hirsutism
Patients with PCOS often have acne or excessive hair growth. The over-production of androgens stimulate the hair follicles which lead to the problem as shown in the diagram above. Androgens (i.e. Testosterone) circulate in the bloodstream largely bound up in a protein called sex-hormone binding globulin (SHBG). This is basically a sponge that sops up the testosterone. It is only free testosterone that is able to leave the blood stream and cause hirsutism. Insulin actually lowers the SHBG, leaving more free testosterone to cause problems. Our treatment is basically 2 pronged. It is well known that birth control pills will increase SHBG, and thus bind up free testosterone and suppress luteinizing hormone (LH stimulates the ovaries to make more androgens). When combined with a medicine that actually acts on the hair follicles to block the effects of testosterone like spironolactone, BCPs improve acne and hirsutism. Spironolactone is structurally very similar to testosterone. Of course, the addition of insulin sensitizers may improve the out-come as mentioned above and a number of studies have shown beneficial effects of adding these medicines. Due to the life cycle of the hair follicle, it may take 3-6 months for the effects of these drugs to be seen on the hair growth and acne. Mechanical or cosmetic treatments may be used in combination with the suppressive therapies mentioned above. It is of note that plucking and waxing probably do little . Electrolysis, lasers, shaving, and bleaching have been used successfully. Over 80% of patients that are compliant with therapy will have improvement.
12. Exercise/Life Style Modification
Studies have shown that long term use of hypocaloric diets will improve the metabolic derangements in patients with PCOS. Some have concerns about a low carbohydrate and high fat diet in PCOS due to the already abnormal lipid profiles seen in patients with PCOS. Many have tried low carbohydrate diets in an attempt to lower insulin levels.
In a study published by Kiddy et al in 1992, twenty-four obese PCOS spent 6 months on a low calorie (1000 kcal), low fat diet. There was a marked improvement in their clinical parameters and lowered insulin levels. A report by Jakubowicz and Nestler showed a reduction in serum testosterone levels using a similar dietary regimen. A very well designed study from Italy examined the long term effects of metformin and hypocaloric diet on PCOS. Metformin improved the hirsutism , menstrual function, visceral adipose tissue, and glucose stimulated insulin secretion. Thus, it appears that diet and some pharmacologic may be helpful in patients with PCOS.
Many patients have attempted to diet all their life with limited success. Some have even attempted gastric bypass surgery to effectively starve themselves. There is some evidence that "life style" modification may be an effective adjunct to our treatment of PCOS. A very interesting study from Sweden by Ek et al, showed that PCOS patients had a marked reduction in the lipolytic (i.e. fat breakdown) effects of noradrenalin due to a decreased number of noradrenalin receptors on fat cells.
Other experimental studies have suggested that the sympathetic nervous system, which innervates the ovary, may be activated in PCOS. Weight reduction has been shown to increase noradrenalin sensitivity in PCOS patients. Thus, there may be a link between the sympathetic nervous system and PCOS which exercise may help. The goal of life style modification should be greater than or equal to 7% loss in weight and maintenance with <25% of calories from fat and a total caloric intake of 1200 to 1800 calories per day. Also, we would like the patient to have more than 2 ½ hours of moderate physical activity per week and be on a low glycemic index diet to improve insulin resistance.
I would like to suggest to our patients that they consider brisk walking 3-5 times a week. Indianapolis is lucky to have numerous "Runs & Walks” where hundreds of people walk anywhere from 5 kilometers (3.2 miles) to the mini-marathon (12.1 miles). These walks are a lot of fun, and are well-organized. You have an "official time" which is posted on the Internet, and will see your time improve with each walk. Websites for the local walks are Tuxedo Brothers, Ken Long & Associates, and The Runners Forum. You can train during the week, and do the walks on weekends. There is a real sense of accomplishment when you complete these events.