Introduction
Polycystic Ovary Syndrome (PCOS) is one of the most common gynecologic-endocrine disorders of reproductive age women, affecting approximately 5–10% of women, with some estimating the condition to be impacting upwards of 20% (Mastorakos, Koliopoulos & Creatsas, 2002; Pehlivanov & Mitkov, 2007; Haydardedeoglu, Simsek, Kilicdag & Bagis, 2009; Guido et al., 2010; Kahraman et al., 2014; Lenart-Lipińska et al., 2014; Adeniji, Essah, Nestler & Cheang, 2016). Infertile women who exhibited polycystic ovaries were first described by Stein and Leventhal in 1935; however, what was initially described as a primarily gynecological disorder of dysfunctional uterine bleeding, has evolved into a condition encompassing a constellation of metabolic dysfunction with long term health consequences such as obesity, dyslipidemia, hypertension, and cardiovascular disease manifesting by the fourth decade of life (Mastorakos et al., 2002; Bremer, 2010; Nybacka, Carlström, Fabri, Hellström & Hirschberg, 2013; Ganie et al., 2013; Kamboj, & Bonny, 2017; de Melo, dos Reis, Ferriani & Vieira, 2017). PCOS is currently characterized by menstrual dysfunction (chronic oligo- and/or anovulation), hyperandrogenism exhibited by: hirsutism, acne, and alopecia; obesity, disordered gonadotropin [luteinizing hormone (LH) and follicle stimulating hormone (FSH)] secretion as well as polycystic ovaries visualized on ultrasound; however, manifestations vary considerably (Bremer, 2010).
Although a wide range of symptoms are recognized or associated with PCOS, the etiology of PCOS is still unknown, with most therapeutic treatment strategies focused on symptom alleviation. Historically, treatments have revolved around inhibiting and/or decreasing androgen production via luteinizing hormone releasing hormone (LHRH) analogs, aromatase inhibitors, anti-androgens, and combined oral contraceptive pills (OCPs), which are the preferred first line treatment for all women with PCOS not currently seeking to become pregnant (Ganie et al., 2013; Kahraman et al., 2014; de Medeiros, 2017). Combination OCPs contain estrogen and progestin which work to regulate the menstrual cycle through suppression of the hypothalamic-pituitary-gonadal (HPG) axis which improves both clinical and biochemical hyperandrogenism characterized by PCOS (Yildizhan, Gokce, Yildizhan, & Cim, 2015; Amiri, Ramezani, Nahidi, Kabir & Azizi, 2018). OCPs function to suppress the HPG axis through a variety of actions such as the inhibition of follicuologenesis by suppressing the secretion of gonadotrophs, decreasing ovarian and adrenal androgen synthesis, inhibiting 5-reductase and increasing serum sex hormone–binding globulin (SHBG) (Mathur, Levin, & Azziz, 2008; Amiri et al., 2018). The estrogen component of OCPs increases SHBGs which binds free testosterone and results in decreased levels of active testosterone in the serum, while the progestin component inhibits 5α-reductase activity by acting as an antagonist at androgen receptors. The synergistic effects of estrogen and progesterone in reducing hyperandrogenism provides the basis for the first-line monotherapy treatment regimen of OCPs currently utilized in women with PCOS associated menstrual irregularities, hirsutism, and acne (Kahraman et al., 2014).
However good at reducing androgen levels in women with PCOS, OCPs, especially those with high estrogen, may negatively impact lipid, glucose, and insulin metabolism and have been associated with an increased risk of cardiovascular disease, particularly with long-term use. (Guido et al., 2010). Insulin resistance and subsequent hyperinsulinemia have been identified as significant contributors to the pathogenesis of PCOS occurring in up to 80% of women with PCOS and in almost all obese women with PCOS (Haydardedeoglu et al., 2009; Lenart-Lipińska et al., 2014). Consequently, insulin sensitizers, particularly metformin, have been introduced to target insulin resistance in addition to other aspects of the syndrome as long-term metformin treatment has been shown to increase ovulation, improve menstrual cyclicity, and reduce serum androgen levels in these patients (Palomba, Falbo, Zullo, & Orio, 2008; Nestler, 2008).
Hyperinsulinemia stimulates both ovarian and adrenal androgen production and decreases the synthesis of SHBG by the liver resulting in increased ovarian androgens which may lead to premature follicular atresia and anovulation in addition to: metabolic syndrome, type 2 diabetes mellitus (DM), cardiovascular disease and endometrial hyperplasia (Kahraman et al., 2014). Earlier studies which evaluated OCPs in terms of their effects on carbohydrate metabolism in overweight/obese women with PCOS have so far been conflicting with results indicating that OCPs containing the cyproterone acetate have either no effect on insulin sensitivity or act to aggravate insulin resistance by decreasing glucose tolerance while OCPs containing desogestrel have shown a decrease in glucose tolerance and insulin resistance (Minozzi, Costantino, Guaraldi, & Unfer, 2011; Bhattacharya, & Jha, 2012). While more recent studies investigating the association between the use of OCPs and metabolic changes reported no association when types of progestins were varied, studies showed significant heterogeneity, with body mass index (BMI) frequently implicated as a contributory factor to OCPs effects on fasting glucose and insulin resistance (Kahraman et al., 2014; Yildizhan, et al., 2015; Adeniji et al., 2016).
Currently, treatment regimens are implemented to control symptoms of PCOS that commonly manifest and become distressing in adolescence or young adulthood when outward appearances and the desire for menstrual regularity often overshadow the potential future implications of any interventions. OCPs are the drug of choice for women with PCOS seeking to manage their symptoms, and who are not looking to become pregnant in the near future, even though evidence exists which suggests that treatment with OCPs may not always be advisable over concern that OCPs could exacerbate underlying metabolic dysfunction (Adeniji et al., 2016). It is important that treatment regimens be designed to improve oligomenorrhea and phenotypic androgen excess, but also remain mindful of long-term risks associated with metabolic dysfunction, such as obesity, cardiovascular disease, DM and infertility. OCPs should, therefore, be used with caution in women with PCOS, especially when these women are overweight or obese for which even a slight worsening of insulin sensitivity with OCPs may hold clinical relevance (Haydardedeoglu et al., 2009; Sakumoto et al., 2010; Cheang, Essah, Sharma, Wickham & Nestler, 2011). This paper aims to evaluate if the addition of insulin sensitizers to oral contraceptives can reduce the adverse effects of long term oral contraceptives on metabolic dysfunction in the treatment of polycystic ovary syndrome.
Methods
Search Terms and Databases
A literature search was performed using PubMed and Google Scholar. The searches were limited to English language, peer reviewed journal articles published from 2008-present. Literature that did meet these requirements were used as substantial background information such as in the Introduction of this paper. With the help of the abstracts, articles were selected using searches with specific parameters related to the topic. Google Scholar was a secondary aid, and articles were found by using the following keywords: “polycystic ovary syndrome”, “insulin sensitizers”, and “oral contraceptives”.
PubMed searches include:
Inclusion criteria:
¥ Publication dates: 2008 – Wednesday May 2nd, 2018
¥ Study population: reproductive age females, human
¥ Study design: Randomized controlled trials, prospective randomized clinical studies, prospective open-label clinical studies, case control studies, and retrospective study designs
Exclusion criteria:
¥ Publication dates: older than 2008
¥ Study design: Reviews
((insulin resistance) OR obesity) AND "Contraceptives, Oral"[Mesh] AND "Polycystic Ovary Syndrome"[Mesh]
Filters
Studies obtained using the above search strategies were then filtered such that they met the following criteria: 1. English language; 2. human participants; 3. published within the last 10 years. Articles were further reduced by filtering for 1. clinical studies; 2. clinical trials; 3. comparative study; 4. randomized controlled trial to produce the articles for which the hypothesis was tested. Search strategy can be further visualized in Appendix 2.
Study Selection
The PubMed search using the specified search strategies, found in Appendix 2, returned a collective total 77 search results meeting all specified criteria, the process of selection can be seen in Appendix 3. Through the removal of irrelevant articles based on title and subsequently on abstract, a total of fifteen full text articles were accessed and ten were chosen to support the hypothesis as presented in the Evidence Table in Appendix 1. The most common reason for study exclusion was that the study focused on conception, fertility outcomes and immune markers as opposed to therapeutic intervention strategies. Additionally, a significant number of articles obtained using the outlined search parameters produced focused on cardiovascular health and endothelial function, these studies were also removed.
Results
Study characteristics
The studies selected contain results that could provide support or the need for further research into the treatment of PCOS. Three randomized clinical trials, two prospective randomized clinical trials, one retrospective clinical trial and one case control randomized clinical trial were analyzed. All studies were compared and contrasted by their methods used and results published, which is further discussed in the Discussion section. Altogether, the same sizes for the seven studies varied from 41 to 155 women with PCOS. Outcome assessments differed based on focus of each study.
In 2009, Costantino, D., Minozzi, G., Minozzi, E., & Guaraldi, C. evaluated the effects of treatment with Myo-inositol on serum androgen concentrations, circulating insulin levels, glucose tolerance and ovulation in women with PCOS. This was a double blind trial. There were 42 women with PCOS who were randomized to receive Myo-inositol with folic acid or folic acid alone (placebo). Inclusion criteria included age from 18 to 40 years, oligomenorrhea, elevated serum free testosterone and/or hirsutism, polycystic ovaries were observed via pelvic ultrasonography. Serum studies including total testosterone, free testosterone, triglycerides, and insulin were evaluated in addition to clinical evaluation of BMI, systolic and diastolic blood pressures and waist circumference. Clinical and serological evaluations were conducted following 3-4 months of intervention and compared to baseline. In conclusion of the study, reductions in serum levels of total testosterone, free testosterone, triglycerides, cholesterol, and insulin in Myo-inositol treated participants reached significance when compared to the placebo group (P=0.003, P=0.01, P=0.001, P=0.001, P=0.03, respectively). Clinical reductions in systolic and diastolic blood pressures reached significance in Myo-inositol treated participants when compared to placebo group (P=0.002, P=0.001). Clinical changes in BMI and waist circumference did not differ significantly between placebo and Myo-inositol treated participants.
In 2011, Minozzi, M., Costantino, D., Guaraldi, C., & Unfer, V. compared the effects of OCPs when combined with myo-inositol on endocrine and metabolic parameters in women with PCOS. This was a prospective open-label clinical study. There were 155 women who were randomized to receive OCPs monotherapy or OCPs in combination with myo-inositol for 12 months. Inclusion criteria included the fulfillment of two out of three diagnostic criteria for PCOS according to the 2003 Rotterdam Consensus conference. Clinical evaluations included BMI and hirsutism score via modified Ferriman-Gallwey score while serum studies included total cholesterol, HDL, LDL, TGs, apolipoprotein B, apolipoprotein A, fasting glucose, fasting insulin, insulin resistance via HOMA-IR, testosterone, SHBG, D4-androstenedione, dehydroepiandrosterone (DHEAS) and LH levels. Clinical and serological evaluations were conducted following 12 months of intervention and compared to baseline. In conclusion of the study, OCPs plus metformin therapy resulted in a greater reduction in FG score when compared to treatment with OCPs alone (P=<0.001). Combination therapy significantly decreased hyperinsulinemia through the reduction of serum insulin and HOMA-IR (P=<0.001, P=<0.001, respectively) while significance was not reached when OCPs were used alone. Testosterone, DHEAS and D4-androstenedione decreased in both treatment groups; however, more significant reductions were observed in the combination therapy group (P=<0.001, P=<0.01, P=<0.001, respectively). Serum lipid profiles improved in the combination therapy group through the reduction of LDL, and increase in mean HDL (P=<0.05, P=,0.001, respectively). Free androgen levels did not differ significantly between treatment groups nor did the increase in total cholesterol.
In 2015, Iwata, M. C., Porquere, L., Sorpreso, I. C. E., Baracat, E. C., & Soares, J. M. compared clinical and laboratory parameters in when with PCOS using metformin or OCPs after 6 months of treatment. This was a retrospective study which analyzed the records of women with PCOS using the Androgen Excess and Polycystic Ovary Syndrome (AE-PCOS) Society criteria. There were 41 women with PCOS, 16 of which were treated with OCPs, 16 treated with metformin and 9 treated with a combination of metformin and OCPs. Clinical evaluation including BMI, acne, modified Ferriman-Gallwey index, menstrual cycle index in addition to serum concentrations of LH, FSH, total testosterone, andosternedione and homeostasis model assessment: insulin resistance (HOMA-IR) index. Clinical and serological evaluations were conducted following 6 months of intervention and compared to baseline. In conclusion of the study, monotherapy OCPs was superior to metformin in regards to acne, Ferriman index, menstrual cycle index, LH, total testosterone and andosternedione levels while metformin monotherapy was superior in HOMA-IR index (P=0.0007). OCPs plus metformin, when compared to metformin alone, showed continued improvement of acne, Ferriman index, menstrual cycle index and testosterone levels while the HOMA-IR index remained lower in metformin monotherapy group (P=0.046). However, OCPs plus metformin vs. OCPs alone shows no significant difference in acne improvement, Ferriman index, menstrual cycle index, LH, free testosterone and andosternedione levels which indicates that the addition of metformin does not lead to additional benefits in these parameters. The HOMA-IR index did not differ significantly between both treatment groups (P=0.75) therefore, indicating that using metformin with OCPs may not improve insulin resistance as much as it does when metformin is used alone.
In 2008, Ozgurtas, T., Oktenli, C., Dede, M., Tapan, S., Kenar, L., Sanisoglu, S. Y., … & Baser, I. evaluated the correlation between circulating asymmetric dimethylarginine (ADMA) concentrations and insulin resistance, gonadotrophins and androgen secretion in women with PCOS and then investigated the effects of metformin and OCPs treatments on serum ADMA concentrations. This was a case control randomized clinical trial. There were 44 non-obese women with PCOS with irregular menstrual cycles or hyperandrogenism and exhibited polycystic ovaries on ultrasound who were age and BMI matched to 22 controls with regular menses and normal appearing ovaries on ultrasound. Women were randomly assigned to receive metformin or OCPs for 3 months. Inclusion criteria included non-smoking with regular daily activity, normotensive (<120/80mmHg in two measurements) and were not regular consumers of alcohol. For at least 3 months all subjects refrained from using steroid hormones or any other medications likely to interfere with ovarian function, insulin sensitivity, or lipid metabolism. Clinical evaluation including serum ADMA concentrations, total cholesterol, TG, HDL, LDL, glucose levels, VLDL, insulin, LH, FSH, total testosterone, estradiol, dehydroepiandrosterone sulphate (DHEA-S), SHBG and HOMA-IR. Clinical and serological evaluations were conducted following 3 months of intervention and compared to baseline. In conclusion of the study, participants with PCOS showed significant increases in serum ADMA, HOMA-IR, TGs, VLDL, total testosterone, free testosterone, DHEA-S with a decreased SHBG compared to healthy controls. Following 3 months of treatment with metformin participants showed significant reductions in BMI (P=0.001), Waist-Hip Ratio (P=<0.001) and serologic reductions in TGs, ADMA, HOMA-IR, total cholesterol, VLDL and free testosterone (P=<0.001, P=0.004, P=<0.001, P=<0.001, P=<0.001, P=0.002, respectively) while participants treated with OCPs showed significant reductions in serum levels of ADMA, LH, FSH, total testosterone, free testosterone and androgens (P=0.006, P=0.001, P=<0.001, P=<0.001, P=<0.001, P=<0.001, respectively). Participants treated with OCPs also exhibited significant elevations in serum SHBG, VLDL and TGs (P=<0.001, P=<0.001, P=<0.001). Linear regression conducted to evaluate the results showed a positive correlation between ADMA and Waist to Hip Ratio, free testosterone, total testosterone, TGs and VLDL (all clinically significant with P=<0.001). Additionally, HOMA-IR was found to be positively correlated to Waist-Hip Ratio (P=0.001). There was no clinical significant change in HDL, LDL, LH, FSH, total testosterone, Androstenedione, DHEA-S and SHBG following 3 months of treatment.
In 2014, Glintborg, D., Altinok, M. L., Mumm, H., Hermann, A. P., Ravn, P., & Andersen, M. evaluated whether treatment with metformin or metformin combined with OCPs resulted in a more advantageous body composition than would result from treatment with OCPs alone in women with PCOS. This was a randomized, controlled clinical trial. There were 90 participants enrolled in the study, with 65 completing the 12-month treatment with metformin, metformin and OCPs or OCPs. Inclusion criteria were 2 of 3: irregular periods for >1 year in combination with a cycle length >35 days; total or free testosterone levels above the reference interval (Total >1.8nmol/L, free >0.035 nmol/L) and/or hirsutism; transvaginal ultrasound with polycystic ovaries. Clinical evaluations including weight, BMI and whole body dual energy x-ray absorptiometry scans in addition to serum concentrations of free testosterone, SHBG, insulin, C-peptide and HOMA were assessed at study completion following 12 months of intervention and compared to baseline levels. In conclusion of the study, the metformin and metformin and OCP treatments were superior to OCP in regards to weight and regional fat mass (P=<0.05) while OCP and metformin and OCPs were superior to metformin in regards to reducing free testosterone levels (P=<0.05). There was not a significant change in study parameters when OCPs were utilized as monotherapy.
In 2008, Hoeger, K., Davidson, K., Kochman, L., Cherry, T., Kopin, L., & Guzick, D. S. investigated the effects of metformin, combined oral contraceptives (OCPs) and/or lifestyle modification in obese adolescent females with PCOS. This evaluated through two small randomized placebo-controlled clinical trials. 79 obese adolescents with PCOS participated in this study, and were randomly assigned to receive metformin, placebo, lifestyle modification or OCP while the combined trial participants all received lifestyle modification and OCPs and were randomized to either a placebo or metformin. Inclusion criteria included postmenarchal women between the ages of 12 and 18 years with a BMI greater than the 95th percentile and evidence of menstrual irregularity in addition to clinical or biochemical hyperandrogenism. Serum concentrations of androgens and lipid were assessed following 6 months of intervention and then compared to baseline levels. Serum studies including total testosterone, SHBG, total cholesterol, HDL, LDL, TGs and fasting glucose were evaluated in addition to clinical evaluation of BMI, waist circumference and Ferriman-Gallwey score. In conclusion of the study, both lifestyle modification and OCPs significantly reduced androgens and increased SHBG in obese adolescents with PCOS. Lifestyle modification reduced free androgen index by 59% (P=<0.05) while increasing SHBG by 122% (P=<0.05); OCPs significantly reduced total testosterone by 44% (P=<0.01) and free androgen index by 86% (P=<0.01). Combination of lifestyle modification, OCPs and metformin produced a 55% decrease in total testosterone (P=<0.05) vs. 33% (P=<0.05) in the placebo group, a 4% reduction in total waist circumference (P=<0.05) and a 46% increase in HDL (P=<0.05). There was not a significant reduction in overall weight of participants.
In 2011, Kilic, S., Yilmaz, N., Zulfikaroglu, E., Erdogan, G., Aydin, M., & Batioglu, S. evaluated the optimal treatment strategy to address the cardiovascular risk in obese and nonobese women with PCOS. This was a prospective randomized clinical trial. There was 96 normoandrogenemic and oligoamenorrheic women with PCOS in addition to impaired glucose tolerance which were divided into obese and non-obese treatment groups before being further randomized to receive metformin or metformin and OCPs. Clinical evaluation including serum concentrations of ADMA, homocysteine, high sensitive C-reactive protein (hs-CRP) and HOMA-IR were conducted following 6 months of intervention and compared to baseline. In conclusion of the study, a reduction in serum ADMA, homocysteine and HOMA-IR levels was observed in participants receiving metformin while an increase in ADMA and hs-CRP in participants receiving OCPs. Changes observed reached significance in obese patients, obese and metformin reduced ADMA, homocysteine, hs-CRP and HOMA-IR (P=0.001, P=0.001, P=0.006, P=0.001, respectively) obese participants treated with OCPs showed increased ADMA and hs-CRP (P=0.032, P=0.001, respectively). Serological changes did not reach significance in either non-obese treatment group. This study showed improvement in hormonal and metabolic parameters evaluated while decreasing ADMA and homocysteine levels potentially independent of participants BMI while the use of OCPs in participants with PCOS regardless of BMI produced impaired glucose tolerance and increased ADMA and hs-CRP potentiating an increase in metabolic risk to these patients.
Discussion
Seven studies focused on the interaction of oral contraceptive pills and insulin sensitizers on the metabolic profiles of women with Polycystic Ovary Syndrome (PCOS). Mixed results were found mostly showing the beneficial effects of insulin sensitizers but on different serological and clinical parameters. The studies by Minozzi et al (2011) and Costantino et al (2009) both experimented with myo-inositol as an alternative to metformin as an insulin sensitizer, and focused on endocrine and metabolic parameters in women with PCOS. Costantino et al. (2009) studied myo-inositol monotherapy for 3-4 months and found a significant reduction in serum triglycerides and cholesterol levels when compared to the placebo group while clinical changes in BMI and waist circumference did not differ significantly between placebo and myo-inositol treated participants; which has been observed in women with PCOS when treated with metformin (Hoeger et al., 2008; Glintborg et al., 2014). In 2011 Minozzi et al. compared the effects of myo-inositol when combined with OCPs for 12 months demonstrated that reductions in serum testosterone, DHEAS, D4-androstenedione were seen in participants treated with OCPs as well as myo-inosital but these reductions were greater and reached significance in addition to improving serum lipid profiles in the participants who were treated with a combination of OCPs and myo-inositol. Similarly, Ozgurtas et al. (2008) investigated the effects of metformin and OCP treatments in women with PCOS for a period of 3 months and observed that OCP monotherapy elevated serum triglyceride and VLDL levels while treatment including metformin significantly reduced total cholesterol and VLDL levels; there was no significant change in serum HDL and LDL levels when measured independently. These three studies share experimental groups treated with insulin sensitizers as monotherapy or in combination with OCPs and concluded independently significant lipid panel alterations between experimental groups. All studies used serum lipid levels as a serologic parameter at baseline and upon study completion to evaluate the effects of each treatment arm. Currently, myo-inositol is not generally considered a first line treatment for PCOS but is gaining some popularity as an alternative insulin sensitizer (Kamboj & Bonny, 2017; Zeng & Yang, 2017). These studies provide evidence to support the use of myo-inositol as an insulin sensitizer in women with PCOS due to the similar effects as metformin; therefore, future studies should concentrate on the effects of each option as compared to one another to clarify the roles, benefits, and consequences of both interventions to initiate treatments that maximize effectiveness.
Costantino et al. (2009) evaluated the effects of myo-inositol in women with PCOS for 3-4 months and determined that myo-inositol significantly reduced serum levels of total testosterone, and free testosterone when compared to the placebo group. When myo-inositol was combined with OCPs in the study by Minozzi et al. (2011) the reductions in serologic parameters were enhanced; however, the free androgen levels did not differ significantly between OCP monotherapy and combination therapy. Similar effects were found in the study conducted by Iwata et al. (2015) where metformin and OCP monotherapy were compared to a combination of the two interventions. Following 6 months of intervention monotherapy OCP produced superior reductions in acne, Ferriman index, menstrual cycling, LH and total testosterone. When OCP therapy was combined with metformin continued improvement in the aforementioned parameters were observed; however, the HOMA-IR index was found to be superior in metformin monotherapy, suggesting that OCPs may further enhance insulin resistance which may be of clinical significance in a woman with pre-existing insulin resistance receiving OCP monotherapy. Interestingly, in 2008 Hoeger et al. conducted a study to evaluate the effects of metformin, OCPs and lifestyle modification and determined that following 6 months of intervention both lifestyle modification and OCP monotherapy significantly reduced androgens and increased SHBG; however, through the combination of treatments and the addition of metformin the greatest reductions in metabolic and endocrine parameters were observed without a significant reduction in participant’s overall body weight. Length of study varied between the studies, from 12 months, Minozzi et al. (2011), 6 months, Hoeger et al. (2008) to 3-4 months Ozgurtas et al. (2008) and Costantino et al. (2009), respectively. Length of study likely resulted in greater participant retention in the study overall, but generalizability of results to support long-term use of insulin sensitizers and OCPs in the women with PCOS is diminished by the condensed timeframe for which the interventions are studied. A study in 2015 by Yildizhan, R., Gokce, A. I., Yildizhan, B., & Cim, N. found OCP composition could alter the metabolic and cardiovascular outcomes when the interventions were studied for 24 months of use. Due to the fact that women with PCOS generally begin to receive medical interventions in late teens and early twenties and maintain their regimen unless pregnancy is sought it is unlikely that studies evaluating the effects of OCPs, insulin sensitizers or lifestyle modifications accurately reflect the long-term consequences of said interventions on this population. Future studies designed to address this timeframe to provide a more complete analysis of the potential benefits or risks of continued treatments is recommended in order to provide women being treated for PCOS the best evidence in order to base their treatment options on.
The study by Hoeger et al. (2008) evaluated the effects of metformin, OCPs, and/or lifestyle modification in obese adolescent females with PCOS for 6 months of treatment and found that there was no significant reduction in overall weight of participants; however, body composition was altered due to evidence of a significant reduction in total waist circumference. The study by Glintborg et al. (2014) evaluated the effect of treating women with PCOS with metformin alone or in combination with OCPs on body composition. There was a significant difference between the two treatment groups in weight and regional fat mass following 12 months of intervention. The study by Ozgurtas et al. (2008) evaluated the effect of metformin versus OCPs on serum ADMA levels in women with PCOS. BMI was a clinical parameter utilized in assessing treatment effects and following 3 months of treatment the metformin treated participants showed significant reductions in BMI in addition to Waist-Hip Ratio. These three studies are in common in that the experimental groups were treated with OCPs and metformin either as monotherapy or in combination and they share similar results of significant body changes between groups. All studies used BMI as a clinical parameter at baseline and upon study completion as one measure to evaluate the effects of each treatment arm; however, study participants were carefully selected in the Hoeger et al. (2008) study to target obese adolescents while Glintborg et al. (2014) and Ozgurtas et al. (2008) did not provide an inclusion criteria of BMI or stratify the participants to account for variations in BMI at the onset of the study. The specific inclusion criteria of Hoeger et al. (2008) and the non-specific stratification of participants in either Glintborg et al. (2014) and Ozgurtas et al. (2008) reduce the generalizability of either study to women with PCOS in the general population.
The studies by Hoeger et al. (2008) and Kilic et al. (2011) contain a significant factor that need to be emphasized. Hoeger et al. (2008) specifically chose obese adolescents with PCOS. The researchers sought to select a sample size that would best represent the spectrum of PCOS patients who would benefit from their treatment interventions and were also not representative of the study participants previously studied. This study saw a lack of published literature and designed a study to best answer the questions asked. The inclusion of lifestyle modification as a treatment intervention is novel in that many published studies emphasize the education given to participants during studies informing patients of the importance of lifestyle changes, but this was the only paper retrieved that chose to implement a strategy as part of the trial. This study focused on common serologic markers easily obtained through basic serology and routines included in more specific studies published. Clinical evaluations were also simplistic and commonly associated with PCOS and therefore regularly measured resulting in the data collected being readily compared to other studies available. Kilic et al. (2011) did not specifically recruit for obese participants with PCOS but rather chose to account for the confounding variable of obesity through the division of participants by BMI status prior to randomization. BMI has repeatedly been shown to amplify the clinical severity of PCOS and further increase the risk of metabolic dysfunction in this subset of the population; therefore, it is likely a confounding variable that needs to be addressed yet most published literature fail to do so (Bremer, 2010). The case control-randomized clinical trial by Ozgurtas et al. (2008) selected for participants who were of normal BMI and performed regular daily activities resulting in a lack of generalizability of the studies findings to the general PCOS population, which occurs frequently across most published literature. Future studies would benefit greatly through the application of previous research findings to obese populations such that the findings could be more broadly applied to the population as a whole, but also to remove the confounding variable, or at least reduce its effects.
Would the addition of insulin sensitizers, such as metformin, to OCPs reduce metabolic dysfunction associated with PCOS as well as the long term health consequences such as obesity, dyslipidemia, hypertension, and cardiovascular disease? Following the analysis of some studies, there is still more research needed in this area, especially in regards to confounding variables, such as BMI. Although there is not enough evidence to outright state that combination therapy of OCPs and insulin sensitizers directly promotes favorable endocrine and metabolic parameters, studies have shown that a combination intervention does provide beneficial outcomes when compared to monotherapy of OCPs treatment groups, specifically in regards to lipids, sugar metabolism and body composition; established risk factors for metabolic syndrome and cardiovascular disease. Positive outcomes from combined interventions include reduced serum total cholesterol, triglycerides, very-low-density lipoprotein, Asymmetric dimethylarginine, Homeostatic Model Assessment for Insulin Resistance, high sensitivity C-reactive Protein, insulin, free testosterone, free androgens, improved body composition, increased serum sex-hormone binding globulin, reduced hirsutism and improved menstrual regularity. These improvements along with the stabilization of other metabolic parameters could then affect the women’s body composition and overall health reducing the risk of metabolic syndrome, obesity, type 2 diabetes mellitus and cardiovascular disease. In addition to these beneficial outcomes, there are minimal adverse reactions to long term insulin sensitizer use, most often gastrointestinal distress that can be minimized through progressive dosing and decrease with longer use.
PCOS currently has no clear pathophysiological process leaving much treatment regimens centered around symptom management, typically initiated in adolescence or young adulthood. OCPs are routinely prescribed due to their ease of use, effectiveness in treating androgen excess and reversibility if desired. It is important that treatment regimens be designed to improve oligomenorrhea, hirsutism, and androgen excess, primary concerns of women with PCOS, but it is also important to remain mindful of the long-term risks associated with OCPs such as cardiovascular disease, insulin resistance, and diabetes mellitus. For such reasons, OCPs may not always be advisable over concern that OCPs could exacerbate underlying metabolic dysfunction, especially if these women are obese or have underlying metabolic dysfunction (Adeniji et al., 2016). OCPs should, therefore, be used with caution in women with PCOS, especially when these women are overweight or obese, for which even a slight worsening of insulin sensitivity with OCPs may hold clinical relevance (Haydardedeoglu et al., 2009; Sakumoto et al., 2010; Cheang et al., 2011). OCPs may still be treatment of choice in these women, but the combination of insulin sensitizers and lifestyle modification may be necessary to mediate the adverse effects of OCP monotherapy on metabolic profiles.
Limitations and Strengths of Review
Current literature in regards to treatment options for the symptomatic management of PCOS tend to utilize healthy women as participants. It is difficult to generalize the data obtained in these studies and draw conclusions to the long term effects of using OCPs to treat PCOS when there are limited studies as a whole on this subset of the population. Of the seven studies evaluated, only Hoeger et al. (2008) focused on interventions for obese women with PCOS, specifically adolescents, and Kilic et al. (2011) studied women with PCOS and a metabolic disorder but divided the participants by BMI before intervention randomization.
Additionally, the long term consequences of OCP use may not be fully investigated in any of the studies presented as two studies took place over 12 months, three over 6 months, and two for 3 months. The length of these studies may show alterations to androgen levels, lipid panels and metabolic profiles but it is difficult to determine whether these levels may equilibrate or further adjust with length of intervention. The management of PCOS symptoms through the use of OCPs is simply just that, a management, it is not rectifying the condition or allowing the body to adapt, but rather masking the symptoms presented; therefore, it is unreasonable to speculate as to the adverse effects of OCPs used for prolonged periods of times such as is often the case in women with PCOS.
Initial studies conducted to establish the safety of OCP use in women as a form of birth control were conducted using normal weight and ovulating women preventing effective generalizations to be made in terms for the relative safety when used in obese women, or anovulatory women, characteristic of women with PCOS (Cheang et al., 2011). So far, studies that have evaluated OCPs and their effects in overweight/obese women with PCOS have been conflicting, likely owing to that being overweight or obese can amplify the clinical severity of PCOS and further increase the risk of metabolic dysfunction (Bremer, 2010).
Future Directions
To best test the efficacy of long term OCP use in the management of PCOS it would be advisable to establish a study utilizing women representative of this subset of the population. Due to the variability in symptoms experienced it would be advisable for the study participants to be subdivided to establish the effect of OCPs in terms of length of treatment received as well as body composition and BMI. Due to the association of weight to symptom severity in PCOS patients, it should be a study parameter that is controlled for when assigning participants to treatment interventions. Further, the ability to evaluate the effectiveness of OCP use in adolescence with PCOS versus combination therapeutic interventions could also allow for the study of BMI and body composition alterations that may or may not occur if interventions are initiated prior to the establishment of metabolic dysfunction.
It would be incredibly expensive, and ambitious, but a thorough longitudinal study following participants for 5 to 10 years would provide a more complete account of the adverse effects that may be elicited by OCPs for that length of time. Al-Sahab, Ardern, Hamadeh, & Tamim estimated in 2010 that the mean and median age of menarche in Canadian females was 12.72 years, while Bichell reported in 2016 that the mean age of a women’s first pregnancy to be 26.3 years of age in the USA. If one considers that adolescent females typically exhibit irregular menstrual cycles and acne for the first 2 years following menarche, it would not be unreasonable to assume that if these symptoms persisted that a adolescent female may be prescribed OCPs by the age of 18 and maintain utilizing them until a planned pregnancy by 25 years of age, meaning a potential 7 year history of use is possible.
Conclusions
OCPs have been recognized for more than 50 years as convenient, reversible, and reliable contraceptives, but OCPs are so much more than contraceptives. The therapeutic uses for OCPs is staggering: menstrual irregularities, endometriosis, premenstrual syndrome, pelvic inflammatory disease, endometrial and ovarian cancers, bone mineral density, menstrual migraine, even rheumatoid arthritis and multiple sclerosis (Caserta et al., 2014). Importantly, however, is the ability for OCPs to treat hyperandrogenism as is seen in PCOS. Without an established pathophysiologic mechanism, the treatment of choice for PCOS has long been considered symptomatic management without much thought to the potential adverse effects of OCPs. The efficacy of OCPs was established 50 years ago, with the sole goal of providing contraception; however, the multitude of uses for OCPs has prompted more studies to assess the adverse effects of OCPs particularly when used for the long term management of chronic conditions.
After analyzing the studies included in this paper, there is no complete conclusion that combining insulin sensitizers with OCPs will reduce the adverse effects of long term OCP use on metabolic dysfunction in the treatment of PCOS. There is data to support combining therapeutic interventions can favorably effect a women’s endocrine and metabolic parameters; however, whether these alterations apply to all women with PCOS or can be maintained long-term has not been established. With further research, this potentially can have beneficial implications like reducing metabolic syndrome, obesity, type 2 diabetes mellitus, and cardiovascular disease in this subset of the population.