Aims. Despite the very clear association between polycystic ovary syndrome (PCOS) and dysglycemia, few studies have explored the continuum of glycemic alterations leading from minor glucose abnormalities to overt diabetes. The purpose of this review is to trace the natural history of glycemic alteration in women with PCOS. Methods. We performed a literature review without time limit until August 2019. Inclusion criteria were studies addressing the association between impaired glucose tolerance or impaired fasting glucose or type 2 diabetes (T2D) and PCOS with at least an English abstract. The exclusion criteria were no PCOS or impaired glucose tolerance or impaired fasting glucose or T2D as outcome. The outcomes of interest were the onset of impaired glucose tolerance, impaired fasting glucose, T2D, and the progression from impaired glucose tolerance or impaired fasting glucose to T2D. Results. Healthy diet and physical activity are the first-line therapy for PCOS. Treatment with metformin was associated with significant lower 2-hour postload glucose levels and with reduction in fasting glucose when compared to placebo. Thiazolidinediones were more effective in reducing fasting glucose levels compared to placebo. Metformin and pioglitazone treatments showed similar effects on fasting glucose levels. The sodium-glucose cotransporter-2 inhibitor empagliflozin did not show differences in metabolic parameters when compared to metformin. The combination therapy with metformin plus the glucagon-like peptide-1 receptor agonist liraglutide was associated with significant improvements in basal and postload glucose levels compared with only liraglutide. Likewise, a combination therapy with the dipeptidyl peptidase-4 inhibitor saxagliptin and metformin demonstrated superiority versus metformin in fasting glucose and oral glucose tolerance test normalization. Myo-inositol supplementation was associated with lower insulin levels, glucose levels, and insulin resistance when compared with placebo, metformin, or estrogen treatments. Conclusions. The use of insulin-sensitizing agents, such as metformin and inositols, along with lifestyle interventions may improve the metabolic profile in PCOS women.
1. Introduction
Diabetes mellitus is a worldwide epidemic. Its prevalence and incidence are steeply growing with an estimated 425 million of people currently having diabetes [1]. The great social burden of the disease is worsened by the huge number of people with prediabetes (i.e., impaired fasting glucose and/or impaired glucose tolerance) who are at high risk of developing it. In addition, one in two adults with diabetes (about 212 million of people) is undiagnosed. Several risk factors for the development of the disease have been well recognized. Some risk factors are gender specific, such as gestational diabetes mellitus (GDM) and polycystic ovary syndrome (PCOS). PCOS is defined by its reproductive features of hyperandrogenism, chronic oligoanovulation, and/or polycystic ovarian morphology [2]. Its prevalence is 5–15%, depending on the diagnostic criteria applied [3]. PCOS is associated with metabolic abnormalities, including insulin resistance (IR) and β-cell dysfunction [2]. The result of IR is hyperinsulinemia, which has a central role in the pathogenesis of androgen excess in PCOS. Indeed, insulin acts as a cogonadotropin to increase luteinizing hormone (LH)-induced androgen synthesis in theca cells [4] and can enhance gonadotropin-releasing hormone (GnRH)-mediated gonadotropin secretion [5]. Insulin also reduces hepatic sex hormone binding globulin (SHBG) synthesis, thereby increasing the levels of bioavailable androgens [6].
The defect of insulin action was quantified in PCOS using the euglycemic clamp [7]. Insulin action was reduced by 35–40% in both lean and obese women with PCOS compared to control women of similar age and body composition [8].
Recent studies in daughters of women affected by PCOS have found evidence for pancreatic β-cell dysfunction prior to menarche [9]. Genetic analyses showed that metabolic abnormalities such as obesity and IR contribute to the pathogenesis of PCOS [10]. Women with PCOS have a higher cardiometabolic risk compared with women without ovarian problems [11].
In women with PCOS, dysglycemia typically consists of impaired glucose tolerance [12], its prevalence being of almost 30% in both adult women [13] and affected adolescents [14]. For this reason, PCOS is associated with a two times increased risk for type 2 diabetes (T2D) [15].
Despite the very clear association between PCOS and dysglycemia, few studies have explored the continuum of glycemic alterations leading from minor glucose abnormalities to overt diabetes. The purpose of this review is to summarize the effect of lifestyle and pharmacological management on glycemic alterations of women with PCOS.
2. Materials and Methods
We searched the online databases of PubMed/Medline, Scopus, Web of Science, Science Direct, Embase, CINAHL, EBSCO, and Google Scholar search engine using MeSH keywords of “impaired glucose tolerance” or “IGT,” “impaired fasting glucose” or “IFG,” “type 2 diabetes” or “T2D,” and “polycystic ovary syndrome” or “PCOS” or “PCO syndrome” without time limit until August 2019. All the reference lists of the retrieved articles were reviewed.
All steps of the study were independently performed by two researchers and any disagreement between researchers was resolved by a third researcher.
Inclusion criteria were studies addressing the association between impaired glucose tolerance or impaired fasting glucose or T2D and PCOS with at least an English abstract with a special focus on the available PCOS treatment.
The exclusion criteria were as follows: no PCOS or impaired glucose tolerance or impaired fasting glucose or T2D as outcome, conference presentations, and letters to the editors.
The outcomes of interest were the onset of impaired glucose tolerance, impaired fasting glucose, T2D, and the progression from impaired glucose tolerance or impaired fasting glucose to T2D with a special focus on the available PCOS treatment.
3. Results and Discussion
3.1. Pathogenesis of Prediabetes and Type 2 Diabetes
Several risk factors have been hypothesized having a causal role in the pathogenesis of prediabetes and T2D of women with PCOS. Classical risk factors such as genetic background, obesity, and familiarity for diabetes and PCOS-specific risk factors have been described. The role of obesity does not impact independently on the onset of prediabetes and T2D because it was shown that also lean women with PCOS have a high risk of glucose alterations. However, obesity surely represents a strong additional risk factor. Among the PCOS-specific factors, IR and hyperandrogenism have been reported. Subjects with hyperandrogenism have a higher level of IR compared with those without hyperandrogenism [16]. Treatment with antiandrogens improves insulin sensitivity; therefore hyperandrogenism could contribute to the pathogenesis of prediabetes and T2D through mechanism of sustaining higher level of IR. However, hyperandrogenism in PCOS could worsen glucose tolerance by stimulating low-grade inflammation [17, 18].
Even if IR is not a feature always present in patients with PCOS, it is considered one of the key elements underlying the pathogenesis of the syndrome. It is the main factor associated with the development of T2D in those women. Reported prevalence of IR in women with PCOS varies from 44% to 70% [19] mostly because of different methods used to assess it [20]. IR is the typical condition of subjects with T2D. Women with PCOS share with people with T2D the same impaired glucose pattern consisting of a prevalent disturbance of fasting blood glucose. Higher levels of IR stress the pancreatic beta cell function, resulting in earlier functional depletion of insulin secretion capacity and higher risk of developing prediabetes and T2D. The other key element characterizing PCOS is hyperandrogenism.
The relation between hyperandrogenism and IR is a chicken and egg problem. On one side, androgens may have a direct role in the inhibition of hepatic and peripheral insulin action by reducing amount and efficiency of the glucose transporter type 4 (GLUT4), especially in adipose tissue and muscles [21]. Moreover, androgens in association with increased free fatty acids (FFA) levels (a common feature in women with PCOS) reduce insulin excretion in the liver and insulin-dependent glucose uptake in skeletal muscles contributing to IR and compensatory hyperinsulinemia [22]. On the other side, it is well demonstrated that insulin has a direct role in ovarian steroidogenesis and in the ovulation control. Insulin in fact directly stimulates androgen production from ovaries, enhancing the activity of CYP17α and other steroidogenic enzymes with increased androgen production. It also inhibits the hepatic synthesis of SHBG and the secretion of the insulin-like growth factor-binding protein (IGFBP-1) resulting in elevated levels of free IGF and androgens. Furthermore, at pituitary level, insulin stimulates LH secretion, which, in association with insulin itself, acts synergistically on theca cells, increasing androgen biosynthesis [4, 23].
As shown by Dunaif et al., IR associated with PCOS, seems to be determined also by alterations at the level of the insulin receptor signalling. They observed that insulin receptors isolated from fibroblast cultured and skeletal muscle of women with PCOS presented increased insulin-independent autophosphorylation which is associated with reduced receptor activity [24, 25] and can lead to IR. However, other abnormalities have been observed also in downstream pathways, such as the serine phosphorylation of the insulin receptor substrates which inhibits its binding with PI3K and prevents the propagation of the signal downstream [26, 27]. Because of this central role of IR and compensatory hyperinsulinemia in the pathogenesis of the disease, it appears clear how insulin sensitizers represent possible therapeutic agents.
Alternative mechanisms of the pathogenesis of prediabetes and T2D in subjects with PCOS are muscle mitochondrial dysfunction [28] and the gut microbiome [29]. The latter hypothesis is sustained by the presence of specific taxa of gut microorganisms that are associated with lower androgen levels.
The following sections and Table 1 report existing evidence of the effect of different treatments on impaired glucose tolerance and impaired fasting glucose in women with PCOS. No treatment was effective in restoring normal glucose tolerance or reverse to impaired glucose tolerance or to impaired fasting glucose when T2D was already diagnosed.
Intervention
Impaired glucose tolerance (IGT)
Impaired fasting glucose (IFG)
Lifestyle modifications
Improved (28, 29, 33, 34)
Improved (28, 29, 33, 34)
Metformin
(i) Improvement in 2-hour glucose levels at the OGTT (46, 47)
Reduction in fasting glucose levels compared to placebo (42)
(ii) Reversion from IGT to normal glucose tolerance (47)
Thiazolidinediones
Reversion from IGT to normal glucose tolerance (60)
Reduction in fasting glucose levels compared to placebo (58)
SGLT2i
Neutral effect (61)
Neutral effect (61)
GLP-1 RA
Improved with the combination of liraglutide and metformin (62)
Improved with the combination of liraglutide and metformin (62)
DPP4i
Improved with the combination of saxagliptin and metformin (63)
Improved with the combination of saxagliptin and metformin (63)
Inositol
Improved (78)
Improved (78)
SGLT2i, sodium-glucose cotransporter-2 inhibitors; GLP-1 RA, glucagon-like peptide-1 receptor agonists; DPP4i, dipeptidyl peptidase-4 inhibitors; OGTT, oral glucose tolerance test.