PCOS AND INSULIN LOWERING THERAPEUTIC MODALITIES
- To summarize the relationship between insulin and androgen excess
- To summarize the role of insulin secretion in PCOS
- To summarize the pharmacologic interventions designed to treat hyperinsulinemia and its role in the treatment of PCOS in fertility and non-infertility patients
- Polycystic Ovary Syndrome (PCOS) and Insulin Resistance (IR)
- The first recognition of an association between diabetes (glucose intolerance) and hirsutism was in 1921 when Archard and Theirs reported a case of a bearded diabetic woman. It was not until 60 years later that the association between PCOS and IR was clarified.
- The familial clustering of anovulation and PCOS suggests an underlying genetic basis. Family members with PCOS have an increased incidence of hyperinsulinemia in females and premature baldness in males. In one study hyperinsulinemia was found in 69% of the siblings/parents of PCOS subjects and 56% also had lipid abnormalities. 79% of the female relatives also had PCOS like ovaries on ultrasound and 88% of the male relatives had premature baldness.
- Hyperandrogenism and insulin resistance is often clinically associated with acanthosis nigricans (AN). The presence of AN in hyperandrogenic women suggests insulin resistance. The mechanism which causes AN to develop is unclear. AN can be found in normal women and its presence is not an absolute marker for PCOS/insulin resistance.
- PCOS is characterized by chronic anovulation and hyperandrogenism. IR is a reduced glucose response to a given amount of insulin sometimes referred to as Syndrome X.
- Initial evidence for the association between PCOS, IR, and glucose intolerance/diabetes was based on studies evaluating basal and postprandial insulin responses in women with PCOS relative to weight matched controls. Although glucose concentrations between groups were similar (normal) the women with PCOS had higher basal and 24-hour insulin levels. This was verified with euglycemic clamp studies. The euglycemic clamp study establishes a steady state of hyperinsulinemia with normal glucose levels. Adding insulin measures the glucose uptake rate. The more insulin used, the greater the resistance.
- Remember that puberty is characterized by temporary insulin resistance and insulin levels alone should not be utilized to make the diagnosis of PCOS or IR in pubertal women. A clinical picture should suggest PCOS and follow up insulin levels should be drawn at age 20.
- The relationship between Hyperandrogenism and IR
- Obesity: A large portion of PCO women are obese. It appeared that the IR may be explained by obesity alone. Studies by Dunaif and colleagues firmly established that the magnitude of IR is greater than what can be explained by obesity alone.1,2
- Defects in the insulin receptor number and/or receptor affinity: Careful binding assays have excluded this as a cause in studies by Dunaif et al and Peters et al. 3,4
- Post receptor defects: Fibroblasts isolated from women with PCOS exhibit decreased insulin receptor autophosphorylation both basally and in response to insulin. There was decreased tyrosine phosphorylation (an essential component needed in insulin action) and increased insulin dependent serine phosphorylation. Serine phosphorylation increases 17,20 lyase activity, which is involved in androgen production, establishing a connection between androgen excess and IR. 5
- A popular hypothesis in the 1990's was PCOS represented a deregulation of the P450c17, the enzyme responsible for both 17 alpha hydroxylase and the 17,20 lyase activities. This hypothesis was not proved.
- PCOS and the development of glucose intolerance
- Given the above findings it is not surprising that women with PCOS are at risk for the development of glucose intolerance.
- Long term follow up suggests women with PCOS have an increased risk of developing Diabetes Mellitus (DM) when compared to controls (15% vs. 2.3%). If you include glucose intolerance, the numbers increase significantly with 30-40% of patients with PCOS demonstrating glucose intolerance.
- Studies have generally evaluated women with PCOS and investigated the incidence of AODM in PCOS women. However for the reverse, there was one study that evaluated women at a diabetic clinic for evidence of PCOS. Diabetic women previously given the diagnosis of PCOS were excluded from the study. In one diabetic clinic 52% of the diabetic women had clinical evidence of hyperandrogenism and menstrual disturbance. 82% of the women had PCOS like ovaries on ultrasound. It was uncertain if the patients had PCOS and then developed glucose intolerance or if the glucose intolerance brought on a PCOS like picture. However, not all women with hyperinsulinemia in the diabetic clinic developed PCOS. The moral of this story is look at all your diabetic patients and make sure they don't have PCOS.
- Insulin resistance and heart disease
- Hyperinsulinemic women with PCOS have more CV events than normoinsulinemic PCO women. All PCO patients regardless of insulin resistance status have more cardiovascular risk than control women. Many feel this increased risk may be due to lipid changes only.
- Mean serum homocysteine and uric acid concentrations were higher in PCOS/IR compared to controls. There was considerably higher incidence of diastolic dysfunction in PCOS/IR compared to controls which may contribute to CV risk. 20
- PCOS women have an increased risk of subclinical atherosclerosis as documented by Intima Media thickness of the carotid artery. This correlated well with insulin resistance status/concentration.
- Insulin resistance and recurrent pregnancy loss (RPL)
An increased risk of RPL is documented in patients with PCOS. Recent studies have shown that Metformin may decrease the risk of pregnancy loss. While well controlled studies evaluating the safety of Metformin in pregnancy are still ongoing, metformin has been administered to a small number of women with diabetes throughout pregnancy and no fetal anomalies have been described to date. Both metformin and the thiazolidenediones are category B drugs.
- Glueck et al22 examined pregnancy loss in PCOS on Metformin therapy and compared these patients to a retrospective control group. All (N=22) patients in the study conceived on Metformin and continued the Metformin throughout pregnancy. 19 continued until delivery, 3 stopped with the positive pregnancy test. Spontaneous loss rates: 73% in controls, 10% on Metformin for entire pregnancy, 30% in those stopping with positive pregnancy test.
- Barken et al studied the incidence of insulin resistance in patients with RPL. This study was not specifically limited to PCOS. Among 73 women with RPL 26% demonstrated IR while only 11% of matched controls had IR. (p=.024). The RPL and control groups had similar fasting glucose. These authors concluded that women with RPL have a significantly increased prevalence of IR when matched to fertile controls. Thus, These authors concluded that fasting glucose and insulin should be added to workup for RPL.
- Insulin sensitizing agents and PCOS
- The role of hyperinsulinemia/diabetes in androgen excess/PCOS has been tested using three different means to lower insulin levels:
- Weight loss
- Metformin (disubstituted biguanide)
- Troglitazone or thiazolidinediones
- Metformin: A disubstituted biguanide (second generation) that reduce fasting glucose concentrations in subjects with type II DM as well as improving glucose tolerance by modest reductions in plasma insulin levels. The primary mechanism of action is to reduce hepatic glucose output by activating glucose transporters that allow passage of glucose into hepatic and muscle cells. Side effects include GI symptoms (which are dose related) and lactic acidosis. The GI effects are dose related and tend to resolve after several weeks. Secondary to the rare effect of lactic acidosis metformin should not be prescribed to patients with renal, hepatic, or cardiovascular disease. Temporary withdrawal of metformin is needed for patients undergoing radiologic procedures using iodinated contrast media or surgery. Recent evaluation of human theca-like tumor cells treated with various concentrations of Metformin (Attia et al) revealed Metformin inhibited androstenedione and T production. Western blot analysis revealed Metformin inhibited STAR (steroidgenic acid regulatory) protein and 17 alpha hydrolase expression. There was no change in 3bHSD. Demonstrating that Metformin may have a direct effect on thecal cells androgen production.
- Studies of insulin and androgen excess
- Velazquez et al6: 26 women with PCOS treated with Metformin, 500 mg TID. Decreases in total and free T, increased SHBG, decreased insulin responses to glucose, and weight loss. Changes in insulin response could not be clearly separated from weight loss.
- Ehrmann et al7: 14 obese nondiabetic women with PCOS treated with Metformin, 500 mg TID. During the study the patient's body weight was controlled. Both glucose and insulin responses in this group were not improved by Metformin. There was also no effect on hyperandrogenemia. It was concluded from this and other studies8,9 that there was no improvement in insulin sensitivity independent of weight loss. This data was corroborated with in vitro studies using theca cells10. These findings were in contrast to Nestler and Jakubowicz11.
- Nestler and Jakubowicz et al: Found a 53% decrease under the serum insulin curve and a reduction in both basal and leuprolide stimulated 17 hydroxyprogesterone concentration. These authors found similar responses in lean and obese PCOS women. Several other studies since this study have shown a small benefit from Metformin independent of weight loss.
- Studies of PCOS/Metformin and fertility
- Pirwany et al12: Evaluated the effects of Metformin, 850 mg BID on ovarian function, follicular growth, and ovulation rate in 20 women with PCOS. Significant increases in the rate of ovulation were observed.
- Velazquez et al13: Metformin, 500 mg TID to 26 PCOS women. Results: 7 regular cycles, 3 conceived during therapy (12%).
- Nestler et al14: 61 obese women with PCOS given Metformin, 500 mg TID or placebo for seven weeks. Of the 35 women given Metformin, 14 started ovulating spontaneously (40%). The 21 who did not ovulate were given Metformin plus CC and 19 women then ovulated (90%). Only one woman in the placebo alone group ovulated spontaneously and two of the 25 women receiving CC plus placebo (8%) were ovulatory.
- Vandermolen et al15: 26 participants received placebo or Metformin, 500 mg TID for seven weeks. CC, 50 mg daily was then begun. 75% of the Metformin/CC and 27% of the placebo/CC group ovulated. 55% conceived on the Metformin/CC and 7% conceived on the placebo/CC.
- Troglitazone= Rezulin (Rosiglitazone= Avandia, Pioglitazone= Actos): Belongs to a class of drugs termed thiazolidinediones which improve the action of insulin in the liver, skeletal muscle, and adipose tissue directly. Troglitazone acts by binding to the nuclear peroxisome proliferator activated receptor gamma (which enhances the transcription of factors that promote glucose breakdown). Troglitazone's major impact is on glucose breakdown rate with only modest changes in hepatic glucose output. The primary concern with thiazolidinediones is liver toxicity. There have been cases of hepatic necrosis with Troglitazone resulting in its removal from the market. Patients receiving these drugs should have LFT's every 2 months for the first year and periodically thereafter. Needless to say these medications should not be used in patients with liver disease.
- Dunaif et al16): Documented improvement in insulin resistance by the administration of Troglitazone. (Decreased fasting insulin and decreased two-hour insulin concentration.)
- Ehrmann DA et al17: Found similar results with decreases in fasting and two-hour insulin concentrations during an OGTT. Beta cell function improved as did total and free T.
- Mitwally et al18: Examined the potential benefit of using pioglitazone for PCOS women who failed Metformin therapy (N=5), 100% responded (ovulation) and 40% conceived.
- PCO/IR and ovulation induction/IVF
- Studies have evaluated patients with CC-resistance. Those resistant to CC have a higher incidence of hyperinsulinemia than those that do not. IR and CC resistance was significally higher in obese PCOS (Int J GYN OB 2001, 75:43-50) although not evaluated directly in this study maybe those patients with hyperinsulinemia should use Metformin or Metformin/CC combination rather than CC alone
- Patient with PCO are typically high responders to gonadotropins. However, a significant number of these oocytes examined at oocyte retrieval are immature compared to controls Stautmauer et al 21
- Stautmauer et al evaluated IVF cycles in CC resistant PCO patients with IR resistance and T levels > 80 ng/ml: Those that took Metformin have fewer follicles &*60;14 mm on the day of hCG and fewer total follicles. Those on Metformin also had lower peak E2 levels. The number of oocytes retrieved was similar, however, the Metformin group had an increased number of mature follicles. This subsequently led to an increased number of embryos. Fertilization rates increased from 43% to 64% in the Metformin group and pregnancy rates increased from 30% to 70% with Metformin use during IVF cycles.
back to top
- Dunaif A et al, JCEM 65:499-507, 1987
- Dunaif A et al, Diabetes 38:1165-1174, 1989
- Dunaif A et al, Ob/Gyn 66:545-552, 1985
- Peters E et al, Metabolism 35:807-813, 1986
- Kruszynska et al, J Invest Med 44:413-428, 1996
- Velazquez E et al, Metabolism 43:647-654, 1994
- Ehrmann D et al, JCEM 82:524-530, 1997
- Fendri S et al, Diabetes Metab 19:245-249, 1993
- Crave JC et al, JCEM 80:2057-2062, 1995
- Duleha et al, Hum Reprod 8:1194-1198, 1993
- Nestler et al, NEJM 338:1876-1880, 1989
- Pirwany et al, Hum Reprod 14:2963-2968, 1999
- Velazquez et al, Metabolism 43:647-654, 1994
- Nestler et al, NEJM 338:1876-1880, 1998
- Vandermolen et al, Fertil Steril 75:310-317
- Dunaif et al, JCEM 81:3299-3306, 1996
- Ehrmann et al, JCEM 82:2108-2116, 1997
- Mitwally MF et al, Fertil Steril 76:5206
- Mather et al, 73:150-156
- Yarah et al, 76:511-516
- Stadtmauer et al, 75:3:505-509
- Glueck et al, 75:46-52