Diet and exercise in the management of PCOS: Starting from the basics

Insulin resistance

Insulin resistance (IR) and concomitant hyperinsulinemia represent an inherent attribute of PCOS: 60%–80% of the patients are insulin resistant and this is independent of obesity, although IR is more common in overweight/obese women with PCOS . In PCOS, apart from a defect in insulin action, there is also a defect in insulin production (beta-pancreatic cell dysfunction, leading to impaired first-phase insulin release) . Consequently, the prevalence of impaired glucose tolerance (IGT) and type 2 diabetes is markedly increased in PCOS, while the evolution from IGT to frank diabetes is 5–10 times faster than in controls .

The origins of IR in PCOS are not fully understood. It seems that there is a postreceptor defect in the intracellular signaling pathway of insulin involving phosphatidylinositol 3-kinase (PI3-kinase). The PI3-kinase pathway mediates metabolic actions of insulin, such as glucose transportation within target-tissues . The steroidogenic and cell growth effects of insulin are mediated through separate pathways, such as the Ras—Raf—MEK and the inositol glycan pathway, which are intact in PCOS . Hence, the presence of selective resistance to the metabolic actions of insulin leads to compensatory hyperinsulinemia, which induces hyperandrogenemia by acting on multiple levels. In the pituitary, it enhances LH production ; in the ovary, it promotes steroidogenesis via activation of CYP17 and IGF-1 mediated actions while it also promotes follicular cell dysfunction ; in the adrenals, it stimulates CYP17 leading to increased androgen production ; and in the liver, it suppresses sex hormone-binding globulin (SHBG) production, leading to increased levels of biologically active androgens .

The relationship between increased body weight and IR is well-defined. Undoubtedly, in predisposed individuals, weight gain and obesity aggravate IR . Similarly, in patients with PCOS, excess body weight worsens IR and hyperinsulinemia, as well as other metabolic and reproductive features of the syndrome . In contrast, weight loss is associated with significant improvement in insulin sensitivity, menstrual pattern, hirsutism and fertility . Besides, pharmacologic attenuation of IR (not involving weight loss) leads to reduced androgen levels, improved metabolic profile, menstrual cyclicity and fertility . These data prove the cardinal role of IR in the pathogenesis of PCOS and highlight the significant contribution of obesity in its pathogenesis.

However, IR in PCOS cannot be fully attributed to obesity. Indeed, there is a very high prevalence of IR in lean PCOS women, whereas weight loss improves but does not completely restore insulin sensitivity . Recent studies imply that obesity-related and PCOS-related IR might be explained by defects in different components of the intracellular insulin signaling pathways (more proximal to the insulin receptor in obesity and more distal in PCOS) . The co-existence of obesity with PCOS has an additive deleterious effect on the intrinsic IR characterizing the syndrome.

Central adiposity

The majority of women with PCOS (even lean individuals) have increased visceral fat deposition, which manifests clinically as increased waist circumference and increased waist-to-hip ratio . Visceral fat is less sensitive to insulin and more sensitive to catecholamines, while it also contributes to metabolic dysfunction and IR by altered secretion of adipokines and fatty acids . Androgens promote visceral fat deposition and hyperandrogenemia may therefore represent a link between PCOS status (cause) and central adiposity (effect). However, there are studies questioning the preponderance of visceral fat in women with PCOS .

Metabolic syndrome

MS is characterized by the clustering of several risk factors for cardiovascular disease (central adiposity, dyslipidaemia, hypertension, impaired glucose tolerance/diabetes). It is associated with long-term adverse outcomes such as cardiovascular disease (CVD), type 2 diabetes, obstructive sleep apnoea, certain cancers, etc. .

Given that IR has central role in the pathophysiology of both metabolic syndrome (MS) and PCOS, the common co-existence of these two syndromes is not surprising. In fact, MS affects about 1/3 of women with PCOS , with its prevalence in this population being 2–3 times that of healthy women of similar age and BMI . There are differences in MS frequency among PCOS populations from different geographic areas, possibly reflecting environmental and/or genetic effects . Furthermore, MS is more prevalent in women with “classic” PCOS phenotypes .

The frequent presence of one/more components of MS in PCOS patients classifies these women as high risk for future development of CVD and warrants the use of aggressive risk-reducing interventions. It has been shown that even young women with PCOS present chronic low-grade systemic inflammation , hyper-coagulable state , increased plasma viscosity and premature atherosclerosis ; hence, despite the fact that PCOS is usually diagnosed at a young age, the implementation of preventive measures should not delay.

Basal metabolic rate and postprandial thermogenesis

Basal metabolic rate (BMR) is the energy expenditure of a healthy subject at rest and at least 12 h postprandial in a thermally neutral environment. It represents 50%–70% of total daily energy expenditure . Basic metabolic rate is associated with lean body mass, sex, age and climate.

As energy balance is depending upon both energy intake and energy expenditure, low BMR could contribute to the energy imbalance that characterizes obesity. Although most reported values of BMR in obesity, are usually within normal range , BMR has been linked to morbid rather than to moderate obesity .

There is a paucity of data in the literature regarding BMR in PCOS. The few existing studies have yielded conflicting results, with some showing decreased BMR in PCOS women versus controls and others finding no difference .

We have previously shown that adjusted BMR was significantly lower in 91 PCOS women compared to 41 normally menstruating women, matched for age and BMI ( p < 0.001). Adjusted BMR was calculated using the equation: adjusted BMR = (group mean BMR) + (measured BMR − predicted BMR). More precisely, in the control group ( n = 48), the adjusted BMR was 1841.05 ± 44 Kcal/day, while in PCOS women the adjusted BMR was 1445.57 ± 76 Kcal/day . The adjusted BMR was significantly lower in PCOS women both with and without IR compared to the control group ( p < 0.001). When adjusted BMR was adjusted for age and BMI, still the higher values were observed among normal women, and the lowest in the group of PCOS women with IR [ p = 0.007 between noninsulin resistant PCOS (PCOS-NIR), and PCOS-IR, p < 0.001 between PCOS-IR and controls, and p = 0.048 between PCOS-NIR and controls] ( Table 1 ) . It appears therefore that IR constitutes an independent aggravating parameter which contributes to the reduction of BMR in PCOS women.

In another study, comparing 128 PCOS patients and 72 eumenorrheic, nonhirsute women, no significant difference in BMR between PCOS subjects (adjusted mean 5807 kJ/day, 95% CI 5715–5899) and controls (adjusted mean 5916 kJ/day, 95% CI 5786–6046) ( p = 0.193) was noted .

On the contrary, in a recent study concerning 36 women with PCOS compared to 37 healthy women, an increased estimated BMR was noted among PCOS women (1658.7 ± 201.1 kcal vs 1359.2 ± 103.7 kcal; P < 0.0001) .

Differences in methodology, in calculating adjusted BMR and energy intake, as well as different criteria in defining PCOS women included in each study are to be blamed for conflicting results. PCOS is a heterogeneous syndrome with clinical and/or biochemical hyperandrogenemia, chronic anovulation with or without PCOS like ovaries on ultrasound, also characterized by multiple metabolic issues, including obesity, diabetes, hypertension, dyslipidemia and fatty liver. Genetic heterogeneity might also play an additional role. A missense mutation at codon 12 in the PPAR gamma 2 (PPARG2) has been associated with increased BMI and attenuated IR in PCOS . In a case-control association study involving 156 PCOS women with biochemical hyperandrogenism, chronic anovulation and polycystic ovarian morphology in ultrasound and 56 unrelated healthy controls, we have shown that although genotype frequencies of the Pro12Ala polymorphism in PPARG2 did not differ among PCOS women and control subjects, the presence of Pro12Ala polymorphism of PPARG2 was associated with lower BMR (1475.7 ± 678.6 vs 893.2 ± 312.3 kcal/day, p = 0.04) particularly in lean PCOS (BMI < 25 kg/m ² ) . Finally, another possible confounding factor in BMR and energy intake assessment could be IR, as women with PCOS and IR might represent a subgroup of PCOS women with significantly decreased BMR .

Based on these findings, insulin sensitizers can be a very helpful additional treatment, particularly in obese PCOS women with IR . Administration of insulin sensitizers should always be coupled with low-caloric diet and lifestyle change.

As BMR remains constant throughout the day, representing 60%–70% of the total energy consumption, women with PCOS should decrease their caloric energy intake in order to maintain their body weight within normal limits. The increase in body weight further augments IR creating a vicious circle. Therefore, in order to achieve energy balance, the decrease in energy intake should be coupled with a significant increase in energy expenditure by aerobic exercise. Besides, it has been shown that BMR in women with PCOS was increased after 12 weeks of endurance and resistance exercise .

Finally, there is also controversial data regarding the influence of PCOS status on postprandial thermogenesis or the rise in energy expenditure which normally follows a meal. Postprandial thermogenesis has been found to be markedly lower in patients with PCOS versus healthy women by some researchers , but not by others .

Gut-brain axis

Food intake is finely regulated by the central nervous system; hypothalamic circuits receive various inputs (optical stimuli; concentrations of glucose, amino acids and other nutrients in the bloodstream; levels of orexigenic and anorexigenic peptides secreted in the gut; levels of adipokines) and orchestrate an appropriate response .

The high prevalence of obesity among PCOS individuals has prompted the investigation of an underlying dysfunction in appetite circuits in women with PCOS, leading to impaired satiety and increased food intake. There might be a relation between hyperandrogenemia and increased appetite, as testosterone administration in rodents increases meal frequency , while treatment with antiandrogens decreases hunger in bulimic women, a population with some similarities to PCOS regarding eating behaviour .

Ghrelin is an orexigenic gut hormone which is normally increased in fasting and suppressed with food consumption. In women with PCOS smaller reduction in postprandial ghrelin has been shown compared with BMI-matched controls . Furthermore, ghrelin levels are inversely correlated with androgens and indices of IR . However, other studies did not find any correlation between ghrelin and PCOS status or androgen levels .

Similarly, there is conflicting data regarding cholecystokinin (an anorexigenic gut hormone) levels in PCOS; some researchers report lower postprandial levels in PCOS versus controls , while others found similar basal and meal-stimulated levels .

Glucagon-like peptide-1 (GLP-1) is an incretin peptide, which is secreted in the small intestine following food intake and has insulinotropic effects . GLP-1 has also anorexigenic actions, exerted either directly at the CNS or indirectly via stimulation of the vagus nerve . Regarding its possible role in PCOS, available evidence differ depending on patient BMI: in lean individuals, OGTT-stimulated GLP-1 levels were significantly lower in PCOS than in controls ; in contrast, no difference was shown in basal or stimulated GLP-1 levels between obese PCOS and obese controls .

Mental health and weight loss

PCOS patients are often obese and suffer to various extents from hirsutism, acne, irregular menses and subfertility. In this context, it is not unusual for women with PCOS to present with mental health issues, which affect all age groups. Numerous studies have demonstrated an increased prevalence of anxiety, depression and negative body image in PCOS . Diminished perceived self-control is another consistent finding in this population . All these mental problems can certainly have a negative impact on patients’ attempts to achieve and maintain weight loss, by hindering beneficial lifestyle modifications (healthy eating, engagement in physical activity, adequate sleeping, etc.) . In conclusion, although PCOS status does not seem to directly affect the physical ability of the patients to engage in physical activity, it certainly has an indirect negative effect mediated through emotional barriers .

Pharmaceutical agents and weight in PCOS