Semin Reprod Med 2023; 41(01/02): 045-058
DOI: 10.1055/s-0043-1777724
Review Article

Beyond Counting Sheep: Exploring the Link between Polycystic Ovary Syndrome and Sleep Health

Authors

  • Nur K. Abdul Jafar

    1   Monash Centre for Health Research and Implementation, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia
  • Christie J. Bennett

    2   Be Active Sleep and Eat (BASE) Facility, Department of Nutrition and Dietetics, School of Clinical Sciences, Faculty of Medicine, Nursing and Health Sciences, Monash University, Clayton, Victoria, Australia
  • Lisa J. Moran

    1   Monash Centre for Health Research and Implementation, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia
  • Darren R. Mansfield

    3   Monash Lung and Sleep, Monash Health, Clayton, Victoria, Australia
    4   Faculty of Medicine, Nursing and Health Sciences, School of Psychological Sciences, Monash University, Clayton, Victoria, Australia
 

Abstract

Polycystic ovary syndrome (PCOS) is a common hormonal condition with reproductive, metabolic, and psychological sequelae that affects 8 to 13% of reproductive-aged women and 3 to 11% of adolescent girls. Sleep is often compromised in women with PCOS due to increased rates of sleep problems, with the most established problem being obstructive sleep apnea (OSA). OSA is highly prevalent in reproductive-aged adult women with PCOS, but not so in adolescence. The international evidence-based PCOS guideline to improve health outcomes in women with PCOS indicated routine screening to identify and alleviate symptoms of OSA. The guidelines, however, did not weigh other multidimensional constructs of sleep health such as sleep disturbances (e.g., sleep quality and quantity), beyond OSA. This is perhaps due to the lack of research and existing mixed findings in the area of PCOS and sleep health. This narrative review summarizes the current knowledge about OSA and expands further to include the limited knowledge about other sleep problems in PCOS among reproductive-aged women and adolescent girls. We broadly cover the prevalence, risk factors, and mechanisms of sleep problems in PCOS and their relationship with cardiometabolic and psychological health. A brief summary on treatment and intervention strategies for sleep problems in PCOS and future recommendations will be deliberated.


Polycystic ovary syndrome (PCOS) is a chronic complex multisystem disorder, affecting 8 to 13% of reproductive-aged women and 3 to 11% of adolescent girls.[1] [2] This varies depending on the type of population and diagnostic criteria used.[3] Current PCOS diagnostic criteria is based on the 2023 International Evidence-Based PCOS Guideline criteria,[3] which has evolved from the internationally endorsed European Society for Human Reproduction and Endocrinology/American Society for Reproductive Medicine (ESHRE/ASRM) consensus Rotterdam Criteria.[4] [5] [6] In adults, diagnosis of PCOS is based on the presence of two or three features of oligo/amenorrhea (OA), clinical/biochemical hyperandrogenism (HA), and polycystic ovary morphology (PCOM) on ultrasound,[5] while in adolescence, PCOS diagnosis can be confirmed from both OA and HA diagnosis.[7] [8] Serum anti-Müllerian hormone (AMH) could also be used as an alternative over PCOM for diagnosing PCOS in adults.[3] Affected women often experience significant metabolic, reproductive, psychological, and dermatological sequelae.[9] [10] [11] [12] [13] [14] [15] Insulin resistance (IR) is a key pathophysiological contributor to clinical features of PCOS that presents in a form mechanistically distinct from obesity-associated IR.[16] Excess adipose tissue further worsens the presentation of PCOS likely through the contribution of extrinsic IR associated with weight gain.[17] [18] Another key pathophysiological driver of PCOS is hyperandrogenism, clinically manifested as hirsutism, acne, and alopecia.[19] Hormonal imbalances such as increased luteinizing hormone (LH) and decreased follicle-stimulating hormone (FSH) stimulate excess androgen production, contributing to hyperandrogenism in women with PCOS.[19] IR accompanied by compensatory hyperinsulinemia as well as hyperandrogenism drive the pathogenesis of PCOS, contributing to metabolic and reproductive complications in women with PCOS.[20]

Sleep health is a multidimensional pattern of sleep–wakefulness, adapted to individual, social, and environmental demands.[21] Sleep is a prerequisite for all physiological systems and is critical for optimal cognitive functioning, performance, and overall well-being.[22] Among the most well-studied sleep disorders in women with PCOS is obstructive sleep apnea (OSA). OSA is characterized by repetitive collapse of the upper airway during sleep.[23] Beyond OSA, knowledge about other sleep problems in PCOS is less studied. The few studies in this area have reported that women with PCOS tend to sleep lesser (<5–6 hours) than the recommended sleep duration of 7 to 9 hours of sleep per day for adults,[24] in addition to poor sleep quality and difficulty sleeping.[25] [26] [27] [28] [29] In a subset of women with PCOS, sleeping >9 hours may be indicative of a comorbid condition.[30]

The international evidence-based guidelines for the assessment and management of PCOS recommended lifestyle management interventions (defined as improving dietary or physical activity via behavioral modification) as the primary initial treatment strategy.[3] Although the PCOS guidelines mentioned routine screening to identify and alleviate symptoms of OSA,[3] other multidimensional constructs of sleep health are not fully considered. Improving sleep health by targeting sleep problems may promote adherence to healthy lifestyle behaviors in women with PCOS.[26] [31] In this review, sleep problems encompass both sleep disorders and sleep disturbances. Here, we defined sleep disorders as clinical sleep conditions such as OSA, sleep-disordered breathing (SDB), insomnia, and circadian rhythm (sleep–wake) disorders. Sleep disturbances are defined as a spectrum of sleep conditions that encompass both clinical sleep symptoms or features of a sleep disorder (e.g., excessive daytime sleepiness [EDS] or hypersomnia, restless legs syndrome [RLS]) as well as other sleep measures that are not necessarily associated with clinical sleep symptoms that identify as poor sleep quality, disrupted sleep patterns, sleep difficulties, sleep loss, and abnormalities in sleep architecture (see [Table 1]). We narratively synthesized the literature for studies that examined the associations between OSA and other sleep problems in women with PCOS.

Table 1

Categories of sleep problems in women with PCOS

Sleep problem categories

Definition

Sleep disorders

Clinical sleep conditions such as OSA, SDB, insomnia, and circadian rhythm (sleep–wake) disorder

Sleep disturbances

Spectrum of sleep conditions that encompass both clinical sleep symptoms or features of a sleep disorder (e.g., EDS or hypersomnia, RLS) as well as other sleep measures that are not necessarily associated with clinical sleep symptoms, such as poor sleep quality, disrupted sleep patterns, sleep difficulties, sleep loss, and abnormalities in sleep architecture

Abbreviations: EDS, excessive daytime sleepiness; OSA, obstructive sleep apnea; PCOS, polycystic ovary syndrome; RLS, restless legs syndrome; SDB, sleep-disordered breathing.


Role of Sleep in Health

Sleep plays an essential role in both body and mind recovery, by restoring and supporting various biological functions like memory consolidation, physiological and psychological development, modulation of immune response, and removal of brain waste.[22] Sleep is divided into two main categories: non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep that occur in 45-minute stages and alternates between each other forming a 90-minute cycle. NREM sleep consists of stages N1, N2, and N3 (i.e., slow-wave sleep [SWS] or stages 3 and 4).[32] A higher proportion of REM sleep, important for memory consolidation, appears in the latter half of the sleep period at the expense of NREM sleep, where the body regenerates and immune system strengthened.[32] Unhealthy sleep such as reduced sleep duration can decrease sleep cycles and time spent in NREM sleep (see [Fig. 1]). Sleep problems can also occur in chronic conditions such as PCOS and vice versa.[33] From [Table 2], beyond the presence of sleep deficiency, sleep health incorporates other dimensions of sleep[21] such as satisfaction/quality of sleep, sleep duration, sleep timing, alertness/sleepiness, and sleep continuity or efficiency. Taken together, healthy sleep is important for maintaining both body and psychological health.

Table 2

Dimensions of sleep

Sleep dimensions

Definition

Satisfaction/quality of sleep

The subjective assessment of “good” or “poor” sleep

Sleep duration

The total amount of sleep obtained per 24 h

Sleep timing

The placement of sleep within the 24-h day

Alertness/sleepiness

The ability to maintain attentive wakefulness

Sleep continuity or efficiency

The ease of falling asleep and returning to sleep

Note: Dimensions of sleep adapted from Buysse.[21]


Zoom
Fig. 1 The sleep cycle and respective functions. Sleep is divided into two main categories: non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep that occur in 45-minute stages, and alternates between each other forming a 90-minute cycle. NREM sleep consists of stages N1, N2, and N3 (i.e., slow-wave sleep [SWS] or stages 3 and 4). Unhealthy sleep such as reduced sleep duration can decrease sleep cycles and time spent in NREM sleep.

Measurement of Sleep in PCOS

The sleep measures that are commonly used to detect sleep problems in general populations, including reproductive-aged women and adolescent girls with PCOS, are summarized in [Supplementary Table S1] (available in the online version only). A gold standard of sleep measurement, overnight polysomnography (PSG) monitors sleep and respiration, including sleep–wake states and sleep stages (i.e., NREM and REM) as well as other parameters such as nasal air pressure, oxygen desaturation, and abdominal and thoracic respiratory efforts.[34] Previous studies used PSG to detect the presence and/or severity of OSA or SDB in women with PCOS based on derived index scores of apnea–hypopnea index (AHI), respiratory disturbance index (RDI), or oxygen desaturation index (ODI).[34] [35] [36] Some of these studies also used PSG to examine abnormalities in sleep architecture, referred to as atypical patterns of sleep stages across the night, sleep latency, or sleep efficiency.[37] [38] [39] [40] [41] [42] [43] [44] [45] However, the PSG is expensive, time-consuming, and first-night results may be biased to the new environment.[34] A simplified PSG testing using cardiorespiratory polygraphy (level 3 sleep testing) to diagnose OSA may be preferred given its utility, cost-effectiveness, and portability for home-based sleep monitoring.[46] Another time-saving and cost-effective proxy measure for sleep is actigraphy, usually worn on the wrist to record gross motor activity and heart rate.[47] In adult and pediatric populations, actigraphy may be used to diagnose people with insomnia as well as circadian rhythm disorders using a combination of actigraphic parameters such as total sleep time (total amount of time spent actually sleeping during a planned sleep episode), sleep onset latency (time taken to transit from wake to sleep), and number of wake episodes > 5 minutes.[48] Both objective and subjective methods measure different aspects of sleep.[49] For example, self-perceived sleep problems in women with PCOS may differ from their biological sleep (objectively measured), considering that other biases and confounding factors may also influence sleep as a result of impaired daily functioning.[20] Performance of sleep questionnaires commonly used to detect sleep disorders or sleep disturbances tend to differ in adult women relative to adolescent girls with PCOS due to variations in epidemiology and pathophysiology of SDB as well as differential sleep–wake patterns between the two populations.[50] Some of these questionnaires include the Berlin Questionnaire (BQ) to assess risk for OSA in adults,[36] pediatric sleep questionnaire–sleep related disordered breathing scale (PSQ-SRDB),[51] and Pittsburgh Sleep Quality Index (PSQI).[52] The reliability and validity of screening questionnaires relies on their specificity and sensitivity performance. Questionnaires can be effective negative screening tools. Positive results from screening questionnaires will render further diagnostic testing such as a follow-up PSG examination to confirm the diagnosis.[53] With good internal consistency and construct validity,[54] sleep questionnaires can be a feasible method to capture common sleep disorders or sleep disturbances in both reproductive-aged adult women and adolescent girls with PCOS.


Sleep Problems in Women with PCOS

Obstructive Sleep Apnea

Prevalence

OSA is characterized by a repetitive collapse (complete [apnea] or partial [hypopnea]) of the upper airway during sleep for at least 10 seconds and normally associated with oxygen desaturation and/or arousal from sleep.[23] Individuals with OSA symptoms tend to exhibit non-restorative sleep and daytime sleepiness, resulting in otherwise reduced performance, productivity, and higher rates of accidents.[23] It is possible that symptoms of OSA specific to women with PCOS differ in comparison to the general population. For example, in the clinical setting, women with PCOS and OSA often present with insomnia, fatigue, or mood disturbances, rather than snoring or daytime sleepiness.[23] [36] [55] In the general population, an OSA prevalence of 9 to 28% is reported in reproductive-aged women compared with 13 to 37% in older women and men.[23] [36] [56] In women with PCOS, OSA prevalence tends to increase significantly and ranges from 17 to 75%.[57] [58] [59] In adolescent with PCOS, OSA prevalence is generally variable (ranges from 0 to 57% as reviewed in the study by Helvaci et al[57]). Some literature also reported a lower OSA prevalence in adolescents compared with adults with PCOS (8 vs. 32% in one older meta-analysis).[57]

OSA is known to be more apparent in higher weight PCOS patients,[60] and more severe during REM sleep.[37] [61] A meta-analysis of clinic-based studies (n = 13 studies) reported the risk of OSA was higher in adult patients with PCOS compared with those without PCOS (odds ratio [OR] = 9.74, 95% confidence interval [CI]: 2.76–34.41), but not in adolescents (OR = 4.54, 95% CI: 0.56–36.43).[57] In another meta-analysis, Kahal and colleagues (2020) reported 35.0% of women with PCOS had OSA (diagnosed using PSG and/or level 3 sleep testing) and this was markedly higher in obese (33.33–40.91%) than in lean (0%) women with PCOS (n = 2 studies) and in women with PCOS than in controls (n = 8 studies, OR = 3.83, 95% CI: 1.43–10.24).[59] While being of a higher weight may increase the risk of OSA in those with PCOS compared with healthy controls,[61] [62] the correlation between having obesity and severity of OSA symptoms in PCOS is not usually apparent.[63] Several population-based studies have demonstrated increased risk of OSA in women with PCOS above and beyond potential risk factors such as age and body mass index (BMI).[64] [65] [66] [67] For example, a population-based study from the United Kingdom reported significantly higher incidence rate for OSA in PCOS than controls, regardless of age and BMI.[67] Increased OSA prevalence in women with PCOS may still persist beyond the influence of weight. In a study using the Taiwan National Health Insurance Database, OSA prevalence increased significantly in women with PCOS (n = 4,595, 1.71 vs. 0.63 1,000 person-years) compared with age-matched controls (n = 4,595) and persisted even after adjustment for demographic data and medical comorbidities (i.e., hypertension, dyslipidemia, diabetes mellitus, and obesity).[65] Notably, the risks of OSA reported among women with PCOS compared with controls were relatively higher based on clinic-based studies (ORs = 3.83–11.24)[57] [59] [68] [69] compared with population-based studies (HRs = 1.50–2.63).[64] [65] [67] One explanation for this finding is that clinic patients may result in a potential selection bias of more severe PCOS. Taken together, both clinic- and population-based studies showed elevated risk of OSA among women with PCOS. Excess adiposity and/or increasing age may have partly contributed to, but not solely explained, the relationship between PCOS and OSA.

To date, only n = 19 reported prevalence studies have included a PSG diagnosis.[37] [38] [39] [40] [41] [42] [43] [44] [45] [61] [62] [63] [70] [71] [72] [73] [74] [75] [76] Moreover, the seemingly lower prevalence of OSA (ranges from 0.4 to 2%) reported from population-based studies are likely to be underestimated given that these were based on diagnostic coding in the medical records.[64] [65] [66] [67] In some studies, however, undiagnosed OSA syndrome as well as underestimation of true prevalence of OSA in women with PCOS may occur when ruling out the presence of OSA based on self-reported absence of snoring and daytime sleepiness.[42] [53] [77] [78] [79] Indeed, the wide variations in prevalence of OSA in previous studies were partly related to clinic-based participants with relatively small sample sizes, differences in study populations, and inclusion of mostly higher weight women with PCOS. Together, these limitations from prevalence studies affect the ability to establish cause–effect relationships between PCOS and OSA.


Risk Factors and Mechanisms

Several potential risk factors have been identified in past studies that may contribute to the increased prevalence of OSA in women with PCOS. These include central adiposity, IR, hyperandrogenemia, low progesterone, and low oestradiol levels, which all have been implicated in OSA pathophysiology ([Fig. 2]).[55] [58] [71] [80] [81] Central adiposity is the accumulation of excess adipose visceral fat that is common in PCOS, even among women without overweight or obesity. Fat accumulated in structures surrounding the upper airway increases the likelihood for airway collapse. Upper-body adiposity may also reduce lung volume and adversely impact respiratory control. Together, these physiological changes may contribute to OSA development.[55] [58] [71] [81] Increased risk for OSA in women with PCOS may still persist even after adjustment for BMI,[37] [61] suggesting that body fat distribution rather than total body fat (indicated by body size) may be more useful in determining actual OSA risk in PCOS. OSA is also strongly associated with IR,[37] [82] another common risk factor in PCOS.[20] [61] [83] [84] IR can be influenced by central adiposity through activation of inflammatory pathways, reduction in insulin-sensitizing adipokines, as well as directly between visceral depots and liver through portal circulation.[85] The increased amount of central adiposity accompanied with IR and insufficient sleep in PCOS may be a key factor related to the development of OSA.[61] [84] [86]

Zoom
Fig. 2 Schematic pathways of the interrelating connections between polycystic ovary syndrome (PCOS), obstructive sleep apnea (OSA), sleep disturbances, and cardiometabolic risk. The hallmark characteristics of PCOS such as low progesterone and low estrogen levels (characterized by oligo-anovulation), hyperandrogenemia, central adiposity, and insulin resistance (IR) can promote the development of sleep disturbances and sleep disorders, especially OSA. Hyperandrogenemia and IR favor the development of excess adipose tissue. In turn, both excess adiposity and IR increase the severity of PCOS. OSA enhances sympathetic tone and oxidative stress, which also increases IR. In turn, sleep disturbances related to OSA and recurrent hypoxia can increase weight and contribute to IR, glucose intolerance, and β-cell dysfunction, leading to type 2 diabetes, and cardiovascular diseases.

Other potential mechanisms that might increase the risk of OSA in PCOS include hyperandrogenemia, low progesterone, and low estrogen levels. Hyperandrogenemia or excess androgen levels have been associated with increased OSA risk in women with PCOS in some[38] [62] [70] but not all studies.[37] [58] [61] In PCOS patients, concentration of sex hormone-binding globulin (SHBG) is typically low since women with this condition tend to have elevated androgen levels and potentially compensated hyperandrogenemia and IR.[12] [87] Higher dehydroepiandrosterone-sulfate (DHEAS) levels are also associated with overactive adrenal glands and PCOS symptoms. Hyperandrogenemia predisposes to OSA via effects on neural control of breathing[88] and upper airway mechanics.[89] Due to anovulation, both low progesterone and estrogen concentrations can potentially contribute to increase risk of OSA via alterations in upper airway muscle activity[90] as well as reduce clearance of norepinephrine, an adrenal hormone which impairs the body's ability to initiate and maintain sleep.

The severity of OSA attributed from other risk factors remains unclear and varies across studies depending on different cutoff points and methods to diagnose OSA. Furthermore, the temporality of PCOS and OSA remains equivocal. PCOS may either precede or succeed the development of OSA given that OSA may develop over time as features of PCOS worsen and that previous history of enlarged tonsils in adolescence with PCOS can also cause OSA during childhood.[42] [91] [92] Overall, previous findings suggest that IR was a stronger risk factor than age, weight, or hyperandrogenism for OSA,[37] [61] [73] and that the risk for sleep problems in PCOS is likely to be multifactorial, owing also to genetic and environmental determinants.[11] [19] More research is needed to understand the interplay of risk factors and comorbidities that mediate the associations between PCOS and OSA.


Impact on Cardiometabolic Health

OSA is linked to detrimental metabolic derangements that lead to metabolic syndrome (MetS) (∼60% MetS prevalence occurring in OSA patients) and IR,[93] [94] [95] [96] as well as type 2 diabetes (T2DM) and cardiovascular disease development.[97] Sequentially, metabolic abnormalities associated with MetS and IR may exacerbate sleep disorders such as OSA,[98] although evidence demonstrating this latter association remains uncertain. In women with PCOS, greater OSA presence and severity have shown to precipitate or exacerbate IR, glucose intolerance, and hypertension.[55] [80] [81] [83] OSA has also been linked with non-alcoholic fatty liver disease in PCOS, independent of BMI.[99] Women with PCOS who had OSA or sleep disturbances are also susceptible to worsened cardiovascular risk factors (i.e., higher levels of BMI, systolic blood pressure, diastolic blood pressure, low-density lipoprotein cholesterol, fasting glucose, 2-hour glucose, waist circumference, and lower levels of high-density lipoprotein cholesterol), compared with controls.[69] From [Fig. 2], OSA enhances sympathetic tone and oxidative stress, which also increases IR.[100] [101] In turn, symptoms of OSA (e.g., sleep loss and/or fragmentation, less deep sleep) and recurrent hypoxia can increase central adiposity (e.g., by increasing energy intake and adipose tissue deposition) and contribute to IR, glucose intolerance, and β-cell dysfunction, leading to T2DM and cardiovascular diseases. These metabolic disturbances eventually culminate in a vicious cycle and worsen PCOS phenotypic presentation.[55] [80] [81] [83] Regardless of weight, women with PCOS and OSA may still represent a higher metabolic risk population relative to those without OSA. Several studies have observed increased risk for metabolic disorders in those women with PCOS and OSA to persist, even after adjustment for BMI.[61] [70] Conversely, Tasali and colleagues found this association to be insignificant after adjusting for BMI,[82] suggesting that adiposity could be a potential confounder.[42] [45] [102]

Among adolescents with PCOS and obesity, those with MetS had significantly worse symptoms of OSA or SDB compared with those without MetS.[45] This suggests that in addition to obesity, OSA could also be potentially driven by MetS, leading to increased risk of IR (by homeostatic model assessment [HOMA] index), hypertension, dyslipidemia, and triglycerides (TG).[42] Despite the presence of MetS, other studies have observed that the risk of OSA is not significantly increased in adolescent with PCOS due to the lower AHI levels that did not meet the OSA diagnostic criteria.[39] [41] [43] [44] These authors concluded that the pathological mechanisms leading to OSA syndrome in adolescent patients with PCOS may develop in the later course of the disease.[41] [43] [44] Hence, studies in adolescent PCOS populations may provide insights into early identification of risk groups for the development of OSA as well as MetS. Taken together, sleep problems and metabolic abnormalities occur at increased frequency among higher weight women with PCOS and this could worsen cardiometabolic parameters.


Treatment and Intervention Strategies

The cardiometabolic benefits of treating OSA in both adults and adolescents with PCOS are unclear. A study conducted in reproductive-aged women with PCOS and obesity (n = 19) showed that 8 weeks of home continuous positive airway pressure (CPAP) treatment of OSA may exert beneficial effects on cardiometabolic function in PCOS such as decreased sympathetic output and improved insulin sensitivity.[103] In a case study of an adolescent patient with PCOS and obesity, nasal CPAP (nCPAP) treatment in combination with lifestyle intervention (caloric restriction and exercise program) and hormonal treatment resulted in the resolution of OSA, significant weight loss (from obese [BMI = 34.5 kg/m2] to healthy BMI [BMI = 23.5 kg/m2]), and significant improvements of clinical and biochemical hyperandrogenism.[104] However, it is not known whether CPAP treatment or lifestyle interventions mainly drove changes in OSA symptoms or weight. It is important that treatment of OSA should not be generalized across both adults and adolescents with PCOS as younger individuals could also report lower OSA symptom profiles.[105] Appliance-based treatments including CPAP or mandibular advancement splints may be efficacious,[106] but consideration must be given to adherence and tolerability. Taken together, a forward-looking focus of OSA treatment in PCOS patients should prioritize the management of symptoms,[31] [60] with the benefits of longer-term health risk reduction remaining to be established.




Other Sleep Problems

Insomnia, Circadian Rhythm Disorders, and Sleep Disturbances

Prevalence

Another common sleep disorder in PCOS is clinical insomnia which is characterized by the impairment of the ability to initiate or maintain sleep for at least three nights per week and affects daytime functioning.[107] [108] Insomnia is associated with daytime symptoms of fatigue without sleepiness[109] and is linked to long-term health outcomes, including increased cardiovascular risk and cognitive decline later in life.[110] [111] Insomnia is also relatively more prevalent in women with PCOS compared with healthy controls.[77] [112] For example, Franik et al reported clinically significant insomnia to occur more often in women with PCOS than in controls according to both Athens Insomnia Scale (AIS: 12.6 vs. 3.2%) and Insomnia Severity Index (ISI: 10.5 vs. 1.1%).[112]

Circadian rhythm disorders or circadian misalignment, where one's body's internal clock is misaligned with the environment, can manifest as an advanced, delayed, or non-24 hours (i.e., free running circadian phase).[113] Circadian misalignment impacts social functioning,[114] and is closely related to mental health conditions.[115] Up to 3% of adults in the general population have reported a form of circadian rhythm disorder[116]; however, the prevalence of circadian rhythm disorders in women with PCOS remains unknown. In women with PCOS, levels of melatonin concentrations (from saliva, urine, and follicles) have been used to detect abnormal circadian rhythms.[117] [118] [119] For example, Shreeve et al reported elevated night-time urinary levels of melatonin metabolite 6-sulfatoxymelatonin (aMT6s) and increased oxidative stress marker levels with reduced sleep quality in those with PCOS, compared with controls.[117] The presence of both poor sleep quality and high melatonin metabolite levels among PCOS group may seemed counterintuitive, since melatonin is a hormone that promotes sleep.[120] However, the authors discussed that elevated night-time melatonin levels could potentially act as a free radical scavenger[121] for the increased oxidative stress (oxidants and antioxidants imbalance) often seen in those with PCOS.[122] In another study, obese adolescent girls with PCOS showed later clock-hour of melatonin offset, later melatonin offset relative to sleep timing, and longer duration of melatonin secretion than obese controls.[118] More studies using objective sleep measures together with comprehensive investigation of serum melatonin levels are needed to determine the prevalence of circadian rhythm disorders and understand the associations between sleep and melatonin alterations in PCOS populations.

In addition to sleep disorders, women with PCOS are also at higher risk of sleep disturbances such as poor sleep quality, disrupted sleep patterns, abnormalities in sleep architecture, sleep difficulties, sleep loss, and sleep-related impairment ([Table 1]). Two recent meta-analyses (n = 9 studies and n = 18 studies, respectively) used data from Asian, Western, and/or Oceania countries and reported higher incidence of poor sleep quality (i.e., higher PSQI scores), shorter duration of sleep, lower sleep efficiency, and higher sleep-onset latency in women with PCOS compared with healthy controls (OR = 11.24, 95% CI: 2.00–63.10 and OR = 6.22; 95% CI: 2.77–13.97, respectively).[68] [69] As early as the adolescent period, girls with PCOS and obesity are already experiencing sleep disturbances such as increased sleep-onset latency and reductions in REM sleep and sleep efficiency, compared with girls without PCOS and obesity.[39] In adult women with PCOS, sleep difficulties (i.e., difficulty initiating sleep, more restless sleep, more severe tiredness) and increased difficulty falling asleep have been reported in numerous epidemiological studies (up to 68%), even after adjustment for BMI and other comorbidities.[27] [28] In some cases, despite similar duration of nocturnal sleep, women with PCOS still report more severe tiredness than healthy controls.[27] Women with PCOS are also prone to sleep disturbances such as EDS or hypersomnia in which an individual repeatedly feels excessively sleepy in the day or sleep longer than usual at night.[123] A recent meta-analysis reported significantly higher Epworth Sleepiness Scale (ESS) scores (i.e., greater daytime sleepiness) in women with PCOS compared with controls (n = 4 studies, mean difference [MD] = 2.49, 95% CI: 0.80–4.18).[68] Although only one case–control study examined EDS with PCOS in adolescence, this study similarly reported significantly higher prevalence for both SDB (45.6 vs. 27.8%) and EDS (54.4 vs. 35.6%) in adolescent girls with PCOS compared with age-, ethnicity-, and BMI-matched controls.[118] However, women with PCOS may experience EDS even without OSA syndrome.[61] RLS is both a sleep and movement disorder characterized by an irresistible urge to move the legs, typically in the evenings and can severely disrupt sleep.[124] Only one study reported higher RLS prevalence in women with PCOS (23.3% out of n = 73) in conjunction with higher PSQI, ESS, ISI, and BQ scores, compared with controls.[77]

Knowledge gaps exist with regard to other types of sleep problems related to PCOS risk. Over the last two decades, the majority of observational studies that investigated the prevalence of sleep problems in reproductive-aged women with PCOS were from cohort or case–control studies and only two were longitudinal studies. Out of these studies, n = 10 examined the presence of OSA, SDB, or sleep apnea only,[63] [65] [66] [67] [72] [73] [78] [79] [82] [125] while others (n = 13) have included sleep disorders such as insomnia, circadian rhythm disorders, and/or sleep disturbances like EDS.[37] [38] [61] [62] [64] [70] [74] [75] [76] [77] [112] [117] [119] Eight studies focused on examining sleep disturbances (i.e., sleep difficulties, short sleep, and/or poor sleep quality), but not clinical sleep disorders in PCOS.[25] [26] [27] [28] [29] [126] [127] [128] To date, nine studies based on three study cohorts have investigated PCOS with sleep problems specific to adolescent girls.[39] [40] [41] [42] [43] [44] [45] [102] [118] There remains a paucity of objective research that examine the prevalence of sleep disturbances in PCOS without clinically diagnosed sleep disorders. This is further compounded with the overlapping terminology of sleep disturbances and sleep disorders in the PCOS literature. It should be acknowledged that sleep instruments such as PSQI not only measures sleep quality but also other sleep disturbances, incorporating features identical to that reported in insomnia.[52] Lastly, the accuracy of sleep instruments to detect sleep problems may vary across the type of study populations, depending on the PCOS diagnostic criteria.[3] [5] [7]


Risk Factors and Mechanisms

Similar to OSA above, studies that reported the link between hyperandrogenism and the risk of disrupted sleep patterns are equally mixed.[74] [77] [117] [126] [127] Melatonin is a neuroendocrine hormone that entrains our 24-hour internal clock and sleep patterns.[129] Alterations in melatonin secretion (i.e., lower overall melatonin and delayed melatonin onset) may influence sleep patterns and/or circadian rhythm in women with PCOS,[130] but studies are limited.[117] [118] [119] For some of these studies, IR and hyperandrogenism were also linked to changes in melatonin secretions.[118] [119] Certain single-nucleotide polymorphisms (SNPs) in the melatonin receptor gene have been associated with PCOS risk and its metabolic complications such as IR.[131] This speculates that genetic predisposition to PCOS may play a role in influencing sleep problems related to the condition. Emerging evidence suggests that women with PCOS are more likely to have evening chronotypes (i.e., circadian preference in behavioral and biological rhythms toward eveningness) than those without the condition.[126] [132] Individuals with evening chronotypes have circadian rhythms that are phase delayed in the 24-hour cycle which may affect the performance of morning activities including work and family life (i.e., social jet-lag).[133] Few studies that examined chronotype in women with PCOS (using questionnaires) reported associations between evening chronotype with increased testosterone levels,[126] [132] poor subjective sleep quality,[126] and worse hormonal and metabolic profile in PCOS.[132]


Impact on Cardiometabolic Health

Sleep disturbances such as poor sleep quality, short sleep (<5–6 hours), or long sleep (>9–10 hours) can negatively influence diet, stress, and other lifestyle factors that can contribute to weight gain, hypertension, and obesity, which are risk factors for T2DM and cardiovascular disease.[97] For example, inadequate sleep can affect levels of ghrelin and leptin,[134] lowering adherence to a healthy diet and regular meal patterns,[135] which contributes to weight gain.[136] Insufficient sleep and circadian misalignment also predispose individuals to poor metabolic health and promote weight gain.[137] Poor sleep patterns were also observed in adolescence with PCOS and obesity and were more likely to correlate with metabolic dysfunction and a greater number of MetS symptoms.[45] For example, poorer sleep efficiency measured by actigraphy correlated with higher percentage of liver fat, waist circumference, and higher TG in adolescent girls with PCOS and obesity.[45] Contrary to these findings, studies from another group reported PCOS status, parameters of body weight/body composition, and MetS did not seem to impact significantly on respiratory PSG variables in obese adolescent girls with PCOS.[41] [43] [44] In a cross-sectional study, for example, the authors investigated the differences in PSG variables between adolescents with PCOS and obesity with and without MetS. The authors found no differences among the study groups concerning respiratory PSG variables (i.e., apnea index, hypopnea index, number of obstructive apnea, AHI, and stages 3 and 4), but instead found differences in sleep architecture (i.e., sleep-onset latency and sleep efficiency).[41] Hence, more research is needed to understand the impact of sleep problems on cardiometabolic health in specific PCOS populations.


Treatment and Intervention Strategies

Several studies have explored using other lifestyle-related approaches to improve the sleep quality and/or quantity of women with PCOS, including physical activity,[138] behavioral modification,[139] dietary intake,[140] and psychosocial well-being.[141] Other studies have also explored pharmacological approaches such as magnesium,[142] melatonin,[143] and metformin supplementation.[144] Recently, Oberg and colleagues in a randomized controlled trial compared sleep health variables assessed by actigraphy in women with PCOS (overweight/obese) with controls and also assessed sleep variables after a 4-month behavioral modification intervention (using goal setting, problem solving, stress management, stimulus control, and techniques for avoiding instant gratification to target diet and physical activity and achieve long-term weight control) in comparison with minimal intervention. The authors reported that women with PCOS had normal total sleep duration (7.2 hours), but poorer sleep efficiency than controls (87 vs. 93%) and behavioral modification intervention seemed to reduce the amount of daytime sleep compared with the minimal intervention group, suggesting improved sleep behavior.[139] However, the reduction in daytime napping following behavioral intervention in the PCOS group could also be attributed to the fact that participants were told not to nap during the day. Moreover, collection of sleep data for the control group was partly completed during the COVID-19 pandemic where working from home was encouraged, enabling a wider window of sleep. Given that daytime napping is associated with increased risk of T2DM[145] and death of all causes,[146] behavioral modification interventions that target diet and physical activity could be a feasible strategy to improve the sleep health and cardiometabolic outcomes in women with PCOS.

In non-PCOS adult populations, cognitive behavioral therapy (CBT) for sleep interventions has shown to improve both sleep quality and quantity.[147] [148] CBT for insomnia (CBT-I) is a well-recognized non-pharmacological intervention for individuals with insomnia.[149] CBT-I involves techniques such as stimulus control (e.g., modulating cues for wakefulness), relaxation techniques, sleep restriction, sleep hygiene education (e.g., patient education about healthy sleep habits), and cognitive restructuring (e.g., altering dysfunctional attitudes and beliefs about sleep).[149] [150] However, no studies so far have implemented CBT for sleep disturbance or CBT-I interventions in women with PCOS.

In the general population, non-pharmacological sleep health interventions like behavior change methods, mind–body exercise, sleep education, or relaxation techniques have shown promising evidence in improving both sleep quantity and/or sleep quality.[151] As mentioned above, only few studies in PCOS have used sleep health strategies (involving lifestyle-related approaches).[138] [139] [140] When optimizing sleep health, one should consider other multi-dimensions of sleep[21] ([Table 2]) and factors that shape health, such as genetic, social, environmental, behavioral, and medical care domains.[152] Therefore, more research in PCOS populations is needed to understand the effectiveness of sleep health strategies in enabling better engagement to lifestyle interventions, further supporting their symptoms management efforts.




Sleep and Psychological Health in PCOS

Stress-related mood disorders such as depression and anxiety are prevalent in women with PCOS.[153] Women with PCOS who mostly live in higher weight bodies[13] [154] [155] may experience weight stigma and perceived weight bias that can lead to stress.[156] [157] Furthermore, hyperandrogenemia in women with PCOS can clinically manifest as hirsutism, acne, and androgenic alopecia that impacts body image satisfaction, self-esteem status, leading to psychological symptoms in women with PCOS.[87] [158] Mood disorders have been associated bidirectionally with sleep disruption and reduced sleep.[159] Poor psychological health can influence sleep in women with PCOS as evidenced in a cross-sectional study that reported obesity and depressive symptoms together mediated the increased occurrence of difficulty maintaining sleep in women with PCOS (OR = 1.92, 95% CI: 1.12–3.31).[28] Sleep health, when compromised, can influence individual's adherence to lifestyle management (such as diet). This is shown in a study in which the association between PCOS and improved diet quality may be maintained only if women can obtain enough good quality sleep.[26] Taken together, the assessment and management of both sleep and psychological health problems are integral in the management of PCOS symptoms.


Summary of Findings and Research Gaps

This review narratively summarizes the current knowledge on the prevalence, risk factors, and mechanisms of sleep problems and its potential contribution to cardiometabolic health, as well as the link between sleep and psychological health in PCOS. We identified key gaps and challenges that need to be addressed to gain further clarifications on the true prevalence of sleep problems in PCOS and better understand the psycho-cardiometabolic benefits of OSA and sleep problems treatment in PCOS. First, future research in PCOS and sleep problems should include more high-quality studies (e.g., objective sleep measures) to detect sleep disturbances beyond clinical sleep disorders such as OSA. Second, future studies in PCOS should consider defining the terminology of sleep disturbances as a spectrum of sleep conditions that encompass both clinical sleep symptoms as well as findings from objective sleep measures ([Table 1]). Third, the majority of past studies were conducted in Western populations and results may differ depending on the PCOS diagnostic criteria used; therefore, there is a need for more representation of studies across various racial or ethnic groups that may allow for inclusion of broader PCOS phenotypes. Similarly, more research in the pediatric (young person: 13 to <22 years) or adolescent population is warranted,[160] especially since adolescents are more susceptible to living in larger bodies due to poor lifestyle and sleep habits.[161] Adolescence is also a sensitive period where diagnosing PCOS is challenging, for example, due to menstrual irregularity and multi-follicular ovaries.[7] Understanding the role of sleep and how sleep interacts with symptoms of PCOS in the early development of the syndrome may help clinicians identify PCOS tendencies and intervene early to improve metabolic and reproductive outcomes. It will be useful perhaps if the development of sleep behaviors is mirrored together with the life-course approach to PCOS to elucidate causal effects.

Although not mentioned in this review, we also acknowledge the importance of investigating sleep problems during the perinatal life-stage in women with PCOS since physiological and anatomical changes during pregnancy may be conducive to the development of SDB (especially snoring) typically in the last trimester.[162] While key features of PCOS are often no longer present in postmenopausal women, some research suggest that underlying metabolic and hormonal changes due to PCOS may persist into the later years,[163] thus highlighting important research in this area. Finally, investigating the effects of sleep disorders on reproductive health in women with PCOS[164] will provide valuable insights into future treatment options, especially for those women with PCOS who experience fertility issues.


Conclusion

In conclusion, the 2023 International Evidence-Based Guideline for the Assessment and Management of PCOS mentioned routine screening to identify and alleviate symptoms of OSA. This represents a big step to acknowledge the association of sleep health disturbances in PCOS, but other multidimensional constructs of sleep health should be explored further. When clarifying the link between PCOS and sleep health, we should also look beyond just alleviating the presence of sleep disorders, but to also understand how sleep plays a role holistically with other lifestyle aspects to better manage the symptoms of PCOS.



Conflict of Interest

All authors declared no conflict of interests.

Acknowledgments

This work was supported by a Veski Fellowship to L.J.M. and the NHMRC Centres of Research Excellence (CRE) (grant number 1171592). Veski and the NHMRC had no role in the design, analysis, or writing of this article. This research was supported by an Australian Government Research Training Program (RTP) Scholarship.


Address for correspondence

Darren R. Mansfield, MBBS, FRACP, PhD
Monash Lung and Sleep, Monash Health
246 Clayton Road, Clayton, VIC 3168
Australia   

Publication History

Article published online:
19 December 2023

© 2023. Thieme. All rights reserved.

Thieme Medical Publishers, Inc.
333 Seventh Avenue, 18th Floor, New York, NY 10001, USA


Zoom
Fig. 1 The sleep cycle and respective functions. Sleep is divided into two main categories: non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep that occur in 45-minute stages, and alternates between each other forming a 90-minute cycle. NREM sleep consists of stages N1, N2, and N3 (i.e., slow-wave sleep [SWS] or stages 3 and 4). Unhealthy sleep such as reduced sleep duration can decrease sleep cycles and time spent in NREM sleep.
Zoom
Fig. 2 Schematic pathways of the interrelating connections between polycystic ovary syndrome (PCOS), obstructive sleep apnea (OSA), sleep disturbances, and cardiometabolic risk. The hallmark characteristics of PCOS such as low progesterone and low estrogen levels (characterized by oligo-anovulation), hyperandrogenemia, central adiposity, and insulin resistance (IR) can promote the development of sleep disturbances and sleep disorders, especially OSA. Hyperandrogenemia and IR favor the development of excess adipose tissue. In turn, both excess adiposity and IR increase the severity of PCOS. OSA enhances sympathetic tone and oxidative stress, which also increases IR. In turn, sleep disturbances related to OSA and recurrent hypoxia can increase weight and contribute to IR, glucose intolerance, and β-cell dysfunction, leading to type 2 diabetes, and cardiovascular diseases.