Approach the Patient With Obstructive Sleep Apnea and Obesity

Abstract

Obstructive sleep apnea (OSA) and obesity are highly prevalent and bidirectionally associated. OSA is underrecognized, however, particularly in women. By mechanisms that overlap with those of obesity, OSA increases the risk of developing, or having poor outcomes from, comorbid chronic disorders and impairs quality of life. Using 2 illustrative cases, we discuss the relationships between OSA and obesity with type 2 diabetes, dyslipidemia, cardiovascular disease, cognitive disturbance, mood disorders, lower urinary tract symptoms, sexual function, and reproductive disorders. The differences in OSA between men and women, the phenotypic variability of OSA, and comorbid sleep disorders are highlighted. When the probability of OSA is high due to consistent symptoms, comorbidities, or both, a diagnostic sleep study is advisable. Continuous positive airway pressure or mandibular advancement splints improve symptoms. Benefits for comorbidities are variable depending on nightly duration of use. By contrast, weight loss and optimization of lifestyle behaviors are consistently beneficial.

Obesity, the prevalence of which in adults is more than 42%, 31%, and 28% in the United States, Australia, and the United Kingdom, respectively, is a major global public health issue. The rapid rise in obesity prevalence parallels a significant reduction in restorative sleep and an increase in the prevalence of obstructive sleep apnea (OSA) (1).

OSA is characterized by repeated episodes of partial or complete upper airway obstruction during sleep, resulting in breathing cessation (apnea) or a significant reduction in airflow for at least 10 seconds (hypopnea) despite ongoing respiratory effort, leading to hypoxia and sleep fragmentation. The number of apneas and hypopneas per hour, or the Apnea Hypopnea Index (AHI) is used to characterize the severity of OSA. Symptoms may include 1 or more of loud snoring, choking, or gasping during sleep, witnessed apneas, fatigue, excessive daytime sleepiness (EDS), morning headaches, difficulty concentrating, mood changes, and irritability. OSA is causally associated with multiple comorbidities (2).

Obesity and OSA share common risk factors including genetic predisposition, sex, age, and lifestyle factors such as diet and physical inactivity (3-5). In addition, there is a bidirectional causal relationship between these 2 conditions. The Wisconsin Sleep Cohort Study showed that a 10% weight gain predicted a corresponding 32% increase in the AHI and a 6-fold increase in the risk of developing moderate to severe OSA. Conversely, a 10% weight loss predicted a 26% decrease in the AHI (6). Obesity contributes to OSA by the accumulation of adipose tissue around the pharyngeal airway, reducing the airway lumen size and increasing the propensity for collapse of the airway during sleep (7). Conversely, OSA may contribute to obesity by disrupting the balance of appetite-regulating hormones, leading to increased hunger and weight gain (8, 9). OSA-induced fatigue and mood changes may result in reduced physical activity, thereby adding to the risk of weight gain (10). Both obesity and OSA increase the risks of type 2 diabetes (T2D), dyslipidemia, hypertension, cardiovascular disease (CVD), depression, reproductive disorders, and other comorbidities described later.

Using 2 illustrative cases, we describe an approach to the diagnosis and management of OSA and highlight the differences in clinical presentation and nature of the disorder in males and females.

Case 1

A 54-year-old nonsmoking male accountant presented with obesity and T2D requesting optimization of management. His weight, which he considered normal through his early 20s, gradually increased after he got married, reached a peak of 120 kg a year ago. He had lost approximately 5 kg since then. T2D was diagnosed 2 years before his presentation. He had poorly controlled hypertension, dyslipidemia, and significant generalized and visceral obesity. Medications included metformin, linagliptin, dapagliflozin, rosuvastatin, and irbesartan. He was irritable and had symptoms of depression, sexual dysfunction, and troublesome nocturia that woke him 3 to 4 times per night. He drank 2 beers each night and was largely sedentary. His wife slept in a different room because of his snoring. His body mass index (BMI) was 45 kg/m² and his waist circumference was 118 cm (46.5 inches). Seated blood pressure (BP) was 165/110 mm Hg. His hemoglobin was 160 g/L. HbA1c was 10% (86 mmol/mol), had been as high as 12% (108 mmol/mol) and not below 9% (75 mmol/mol). Serum triglycerides, gamma-glutamyl transferase, alanine aminotransferase, and uric acid were elevated. Serum testosterone was 317 ng/dL (11 nmol/L), SHBG 16 nmol/L, LH, FSH, prolactin, and thyroid hormones were in the mid-normal range.

Case 2

A 58-year-old female nurse presented with fatigue, brain fog, getting up to urinate 3 times per night, morning headaches, sexual dysfunction, and hot flushes that had persisted since the age of 49 years. Only the hot flushes were relieved by hormone replacement therapy. She has longstanding obesity and had previously been diagnosed with polycystic ovarian syndrome. She required assisted reproduction to conceive. Her pregnancy was complicated by gestational diabetes and severe hypertension. She recently had a cardioversion for atrial fibrillation. Her BMI was 33 kg/m² and waist circumference was 110 cm (43.3 inches). A recent blood test showed impaired fasting glucose and elevated serum triglycerides and alanine aminotransferase. Besides her symptoms, she was concerned about difficulty losing weight and the risk of progressing to T2D.

Prevalence of OSA in Males and Females and by age, sex, and Obesity

The prevalence of OSA in men has been reported to be 38% (11) overall and 17% for moderate or severe OSA (AHI ≥ 15). The prevalence in women before menopause is about half that of men, but after menopause it approaches that of men (12). OSA prevalence increases with age and increasing adiposity (11, 13). For every 10% increase in body weight, the risk of OSA increases 6-fold (6). The prevalence of OSA is also increased in a variety of medical conditions as outlined next.

Pathophysiology and Associations of OSA With Other Disorders

The effects of OSA are the consequence of 2 broad mechanisms, intermittent hypoxia, and variations in intrathoracic pressure, thereby inducing oxidative stress, inflammation, hypercoagulability, and changes in various neurohormonal systems. Obesity has similar effects, the magnitude of which are greater with increasing visceral obesity.

Effects of obesity and OSA in common include the following: increased IL-6 and TNF-α and reduced nitric oxide, leading to endothelial dysfunction; activation of the renin-angiotensin-aldosterone system increases hypertension and cardiovascular risk; hypercoagulability increases the risk of occlusive vascular disease; and autonomic dysfunction contributes to cardiac and cerebrovascular disease, atrial fibrillation, T2D, and dyslipidemia (14).

OSA-related factors associated with insulin resistance and elevated blood glucose include the degree and duration of hypoxia (15), particularly during REM sleep (16), sleep (17), and circadian clock disruptions (18), inflammation (18, 19), increased sympathetic nervous system activity (20), increased morning cortisol and lipolysis (21), and altered GLUT4 translocation (22). Conversely, diabetes-mediated nerve damage may affect reflexes controlling the upper airway and the central control of respiration, increasing the propensity to develop, or severity of, OSA.

OSA has been shown to induce changes in appetite control mechanisms, favoring an increase in food intake. These include a decrease in serum leptin and increase in serum ghrelin concentrations, altered cognitive control, and reward pathways, thus increasing preference for energy-density foods with high hedonic value (23). Such changes lead to weight gain or at the very least make it more difficult to lose weight.

A small but significant weight gain has been reported in people with OSA in response to treatment with continuous positive airways pressure (CPAP) (24), an effect particularly seen in those with T2D (25) that seems counterintuitive to the relationship between obesity and OSA described previously. This is resolved by interrogation of the effect of OSA and CPAP on body composition. Recent metanalyses have shown no relationship between the use of CPAP and either subcutaneous (26, 27) or visceral adipose tissue (28, 29) volume. OSA is associated with reduced bone mass (30) but not skeletal muscle despite morphological and functional changes (31, 32). CPAP has been shown to increase both lean body mass (33, 34) and IGF-1 (33), suggesting that CPAP-induced increase in weight may reflect positive change in body composition.

Endocrine and Metabolic

Type 2 diabetes

The prevalence of OSA in adults with impaired glucose tolerance or T2D has been reported to be up to 80% (35-37).

OSA causes insulin resistance and abnormal glucose metabolism independent of adiposity (38), and there is a positive relationship between AHI and glycemic variability (39). A recent meta-analysis showed that OSA was an independent risk factor for prediabetes, and both prevalent and incident T2D in men and women, with the risk increased with increasing OSA severity (40).

The combination of OSA and T2D increases the risk of diabetes complications, including diabetic kidney disease (41-44) and proliferative diabetic retinopathy (45, 46), and confers a greater risk of CVD than either condition alone (47).

Accumulating data indicate that the deleterious effect of OSA on glycemia is greater with REM-OSA than non-REM (nREM)-OSA (16), as detailed in the section on REM OSA. The risk of diabetes-related renal disease is also greatest in those with REM-related OSA (42).

Meta-analyses of randomized controlled trials comparing CPAP with sham CPAP found improvements in insulin resistance but no significant change in fasting glucose (48) or HbA1c (49). This may reflect inadequacy of CPAP use, particularly during the second half of the night when REM-OSA occurs. There are 2 lines of evidence in support of this notion. First, 1 week of nightly CPAP for 8 hours per night has been shown to improve glycemic control (50). Second, in a propensity-matched cohort treated either with upper airway surgery (n = 2132) or CPAP (n = 1712) and followed for at least 5 years, the risk of new-onset diabetes was lower in those who had upper airway surgery compared with those treated with CPAP, presumably by overcoming the problem of CPAP compliance and adequate duration of use each night (51).

Dyslipidemia

Intermittent hypoxia increases serum and liver triglyceride concentrations in lean mice. In obese mice, these effects were absent, likely because of preexisting fatty liver disease (52). Intermittent hypoxia also slows triglyceride clearance from the bloodstream and amplifies free fatty acid mobilization from adipose tissue (53).

Severe OSA, assessed by the oxygen desaturation index (ODI), is associated with elevated serum triglyceride and reduced high-density lipoprotein concentrations, independent of obesity and other confounding factors (54). In a community study of men, a positive correlation was found between serum triglyceride levels and severity of OSA in those with a waist circumference smaller than 95 cm (37.4 inches) (55), a finding that aligns with the animal studies discussed previously.

Nonalcoholic fatty liver disease

The prevalence of nonalcoholic fatty liver disease (NAFLD) is increased in patients with OSA even in those without obesity. The severity of OSA and NAFLD are associated and severe OSA increases the risk of fibrosis (56). CPAP alone does not appear sufficient to improve NAFLD (57). It may augment the effects of weight loss and lifestyle interventions (58). Because of the implications for optimal approach to management, patients with OSA should be screened for NAFLD and vice versa.

Renin-angiotensin-aldosterone system

There is a bidirectional relationship between activation of the renin-angiotensin-aldosterone system and OSA that is independent of the degree of obesity (59). Intermittent hypoxia increases serum concentration of renin and aldosterone in animal models (60), and hormones of the renin-angiotensin-aldosterone system are elevated in people with OSA independent of resistant hypertension and obesity (59). Patients with primary aldosteronism have an increased frequency of OSA (61), which is more severe in those with higher serum aldosterone concentrations (62) and that improves with treatment of the primary aldosteronism (63). The likely mechanism is an increase in fluid shifts at night affecting the mass and function of upper airway muscles (63).

Other endocrine and metabolic effects

The prevalence of OSA in people with overt hypothyroidism is approximately 30% (64), and the relationship between hypothyroidism and OSA is independent of obesity (65). A large goiter may obstruct the airway at night, causing OSA independent of thyroid function (66).

OSA is also increased in patients with acromegaly (67), Cushing syndrome (68), and hypothalamic obesity (69, 70) disproportionately to the degree of obesity.

A recent meta-analysis showed that OSA is associated with osteoporosis and lower bone mineral density of the lumbar spine (71).

Cardiovascular

There is a high prevalence of OSA, and associated comorbidity, in patients with CVD (Table 2), including hypertension, atrial fibrillation (AF), heart failure, coronary artery disease, and stroke. The severity of these disorders and the severity of OSA are positively associated. OSA abrogates or abolishes the physiological nocturnal dip in BP where nighttime systolic and diastolic BP are typically 10% to 20% lower than corresponding daytime values (72). Loss of this dipping on 4-hour ambulatory BP monitoring points to OSA regardless of symptoms (73). Heart failure symptom progression, hospitalization, and mortality are increased by OSA (74). OSA doubles the risk of cardiovascular events (75) and increases the risk of cardiovascular complications after percutaneous coronary intervention (76) and of death after myocardial infarction (77).

Table 2.

Common chronic disorders associated with OSA

Associated disorder	Prevalence of OSA
Resistant hypertension	73%-82% (165, 166)
Atrial fibrillation	76%-85% (167, 168)
Type 2 diabetes	65%-85% (169)
Stroke	71% (170)
Depression	20% (121)
Heart failure	16%-36% OSA (171, 172) 37-40% CSA (171, 172)
Coronary heart disease	48% (173)

Abbreviations: CSA, central sleep apnea; OSA, obstructive sleep apnea.

OSA is a risk factor for incident stroke (78), recurrent stroke (79), stroke mortality, and adverse functional and cognitive outcomes following a stroke (80).

CPAP lowers BP in moderate to severe OSA and when used continuously reduces both systolic and diastolic BP, augments the effect of antihypertensive therapy, and is pivotal in the management of resistant hypertension (81). CPAP prevents AF and reduces AF burden after ablation or cardioversion (82, 83). The SAVE trial enrolled 2717 patients aged 45 to 75 years with moderate-to-severe OSA without excessive sleepiness and with coronary or cerebrovascular disease. After 3.7 years and average CPAP use of 3.3 hours per night, there was no cardiovascular benefit (84). A subsequent metanalysis suggested that use of CPAP for more than 4 hours per night reduced major adverse cardiovascular events and stroke (85). Observational data show that increasing number of hours use of CPAP per night was inversely associated with major adverse cardiovascular events in OSA with the greatest effect in a sleepy subgroup (86); in another study in which the median use of CPAP was 6.4 hours per night, there was a decreased risk of fatal and nonfatal CVD events (87).

Lower Urinary Tract

OSA is independently linked with lower urinary tract symptoms, of which nocturia is the most frequent manifestation (88-91). Nocturia affects nearly 70% of women and 52% of men with OSA (91, 92). In severe OSA cases, the prevalence rises to 82% (92). Frequent nocturia is a strong predictor of severe OSA (90, 93). With nocturia 3 or more times per night, the likelihood of OSA is .71 (93). In mice, both metabolic syndrome and intermittent hypoxia independently and jointly contribute to nocturia (94). CPAP effectively reduces nocturia frequency in men and women (95, 96), thus improving sleep quality.

Reproductive and Sexual Function

Males: Low serum testosterone concentrations in men with OSA are the consequence of obesity rather than OSA (97) and improve with weight loss (98) but not CPAP (99). Men with an elevated BMI, being treated with testosterone replacement who develop polycythemia on testosterone replacement, should be screened for underlying OSA (100).

There is an independent correlation between OSA and erectile dysfunction (ED) (101-103). In men aged younger than 65 years, only severe OSA is linked with ED; however, for men aged 65 years and older, ED is associated with both moderate and severe OSA (102). It is important to consider the presence of OSA in men with ED, obesity, and hypertension (104) and/or symptoms of depression (105). Treatment with CPAP has been reported to variably have no effect (106) or improve (107-109) erectile function, a benefit enhanced by PDE5 inhibitors (109).

OSA is associated with decreased sexual desire, independent of obesity, serum testosterone, and other confounding factors (110).

In mice, intermittent hypoxia reduces sperm motility and fertility (111). Changes in sperm shape and count occur in men with OSA (13), but the relative contributions of OSA, obesity, and metabolic factors are unresolved. A large-scale case-control study involving 4607 men with infertility and 18 428 controls, sourced from Taiwan's National Health Insurance Research Database, has indicated a significant link between OSA and infertility in men (112).

Females: Concurrent OSA and insomnia increase the prevalence of hypoactive sexual desire disorder, female sexual arousal disorder, and female orgasmic disorder compared with those without such sleep issues (113). However, the independent nature of this relationship is ambiguous, and obesity or depression may be more important (114).

In a Taiwanese retrospective case-control study, women aged 20 years and older with infertility had twice the prevalence of OSA compared with controls (115). The prevalence of OSA in women with polycystic ovary syndrome (PCOS) approaches that of men. Even after adjustment for obesity, there is a significantly higher prevalence of OSA in women with PCOS. Women with PCOS and OSA are more likely to have cardiovascular disease, T2D, and mood disturbance than those with either PCOS or OSA alone (116, 117). The nature of the interactions between PCOS and OSA remain unclear (117).

OSA during pregnancy increases the risk of preeclampsia, postpartum bleeding, blood clots, mortality, premature births. and infants with birth defects (118).

Depression and Anxiety

OSA is associated with a higher prevalence of major depressive disorder (MDD) (119), and individuals with MDD are more likely to suffer from OSA (120, 121). In a 2019 meta-analysis of 7 studies, 23% of patients with OSA had clinical depression (119). A 2016 meta-analysis showed that 36.3% of individuals with MDD, and 19.8% in clinical and population cohort studies, had OSA, with no significant sex difference (121).

People with both MDD and insomnia have a higher likelihood of having OSA, whereas neither condition alone significantly predicts OSA (122). In men, the combination of insomnia and OSA increases the prevalence and severity of depressive symptoms (123). Men with sleepiness should be screened for depression, regardless of the severity of OSA (124). CPAP therapy improves clinical depression (125) and mood (84) in individuals with OSA.

Intermittent hypoxia triggers anxiety behaviors in mice (126). In men, OSA has been linked to increased severity of anxiety (127). OSA is more likely in men aged 18 to 64 years with anxiety and hypertension than those with only hypertension (128). Anxiety decreased after 6 months of CPAP treatment in 1 small study (129).

Cognitive Function

OSA is an independent risk factor for cognitive decline and accelerates the progression of mild cognitive impairment to dementia (130). There is a modest elevation of brain amyloid in people with severe OSA compared with controls (131). CPAP appears to slow the rate of cognitive decline and may reduce the risk of dementia (130, 132) and reverse white matter damage over 12 months in people with severe OSA (133).

Phenotypic Variation in OSA

Studies of clinic-based populations may have created a potentially problematic image of a stereotypical patient with OSA. More recent studies of community-based populations combining measures of AHI and ODI with symptoms and comorbidities and with consideration of stage of sleep, sex, and response to treatment have begun to capture the heterogeneous nature of OSA allowing personalization of approach to investigation and management. For example:

Symptom clusters.

Four main symptom cluster subtypes have been described: (1) disturbed sleep and insomnia; (2) minimally symptomatic, (3) EDS; and (4) moderately sleepy. Approximately 40% to 60% of people with OSA report EDS. EDS is a predictor of comorbidities including hypertension, CVD, and heart failure (134, 135).

Residual EDS has been reported in 34% of patients after 3 months of CPAP treatment, decreasing to 22% in those who used CPAP for 6 hours per night or more (136). In a large 6-month randomized controlled trial, the overall proportion of patients with the EDS after 6-month CPAP treatment was 22%, but was 18% and 31% with use of CPAP for more than 4, or, less than 4 hours per night, respectively (137). Of patients with a high score on the Epworth Sleepiness Scale (ESS), young otherwise healthy males and males with psychiatric disorders benefit most from CPAP. Males with obesity and systemic hypertension, and elderly males with multimorbid obesity, benefit least (138).

Causes of EDS independent of OSA include poor sleep hygiene, circadian misalignment, depression, and other sleep disorders such as narcolepsy and idiopathic insomnia or restless leg syndrome (RLS), infections or postinfectious syndrome, hypothyroidism, poorly controlled diabetes, obesity, physical inactivity, substance abuse, and psychotropic medication. Non-OSA-related EDS may also be associated with an increased cardiovascular risk; therefore, EDS in OSA may be overlapping pathophysiological processes (139, 140).

It is important to recognize that the ESS, although commonly used, does not sufficiently define the presence of EDS (134), which may be present in those with an ESS score below 11 (141).

 

REM-related OSA

In REM-related OSA, apneas or hypopneas occur predominantly during REM sleep. The overall AHI may not be significantly elevated because REM sleep accounts for only approximately 20% of total sleep time. Severe REM OSA (AHI ≥ 30 during REM sleep) has an independent association with hypertension (142, 143), and may be a more important predictor of hypertension then AHI during nREM sleep (144). Mood disturbance is more severe with REM OSA than nREM OSA (145). REM-OSA is independently associated with cognitive dysfunction (146) but not sleepiness (147). OSA during REM sleep has been shown to be associated with increased insulin resistance after controlling for nREM OSA (148) and to abrogate the decline in blood glucose concentration that otherwise normally occurs during REM sleep (149). In people with T2D, increasing REM-AHI, but not nREM-AHI is associated with increased HbA1c (16). Because REM sleep predominantly occurs in the second half of the night, extended use of CPAP is required for REM OSA-related comorbidities, a notion supported by the observation that one week of nightly CPAP for 8 hours improves glycemic control (50).

OSA in women:

Compared with men, women more commonly report morning headache, fatigue, mood disturbance, brain fog, nightmares, and insomnia symptoms and are less likely to report snoring or witnessed apneas. In addition, significant OSA occurs at lower BMIs and neck circumference. On polysomnography (PSG), women are more likely than men to have a REM-predominant OSA, arousals rather than oxygen desaturation, and more frequent partial airway obstruction that may not be classified as a hypopnea. After menopause (natural or surgical) and with increasing age, the differences in the characteristics of OSA on PSG between women and men diminish (150, 151).

Concomitant sleep disorders

The sleep disorders most commonly comorbid with OSA are insomnia and restless legs, and in those with obesity, obesity hypoventilation syndrome (OHS). A circadian rhythm disorder may also be present and OSA may induce abnormal patterns of circadian gene expression and disordered circadian rhythmicity (152).

The combination of OSA and insomnia is referred to as COMISA, and these conditions are associated. Troublesome insomnia is reported by 30% to 50% of people with OSA, and approximately the third of people with chronic insomnia have OSA. The combination of conditions increases fatigue, sleepiness, irritability, depression, headaches, cognitive dysfunction, cardiovascular and metabolic comorbidities, and sensitivity to pain (153). Although CPAP is often beneficial for insomnia, insomnia may limit patient compliance with CPAP and should be specifically managed (154).

RLS characterized by paresthesia-like sensations associated with periodic leg movements affects about 10% of adults, severely in 2% to 3%, disrupting sleep. A combination of OSA and RLS increases the likelihood of a high score on the ESS, T2D, CVD, high BP, stroke, asthma, and insomnia. CPAP may make RLS worse, resulting in treatment cessation. RLS may be caused by low iron stores, and it is generally recommended to maintain serum ferritin above 75 mcg/mL (155).

OHS is the presence of alveolar hypoventilation, sleep disordered breathing, and daytime hypercapnia, unrelated to obstructive events or other causes of hypoventilation in people with obesity. Those with OHS and OSA tend to be more obese and have a greater neck circumference. Approximately 10% of those with OSA and severe obesity have OHS, and the prevalence is similar in men and women. Compared with OSA alone, OSA and OHS result in higher pulmonary artery pressures, hemoglobin, and hematocrit, and morbidity and mortality are increased (156).

Approach to Workup and Diagnosis of OSA

A careful clinical assessment of symptoms, comorbidities, medication use, health-related behaviors, together with a sleep and family history, and findings on examination as outlined here, selects those who should undergo overnight PSG, which is required for the diagnosis of OSA. Diagnosis and severity are based on the AHI, which measures the frequency of apneas (complete cessation of airflow) and hypopneas (partial reduction in airflow) per hour of sleep. OSA is diagnosed if the AHI is 15 or greater, regardless of symptoms, or if the AHI is between 5 and 15 (mild OSA) with a supportive clinical presentation. An AHI ≥ 15 and <30 and AHI ≥ 30 are classified as moderate and severe OSA, respectively (157).

A detailed approach to a sleep history can be found here: https://aasm.org/resources/medsleep/{harding}questions.pdf

Clinical Presentation and Assessment

Symptoms

The symptoms of OSA are documented in a recent consensus statement (157) and summarized in Table 1. There are no pathognomonic symptoms. Regular snoring, witnessed breathing pauses, and regular nocturnal gasping episodes are most strongly associated with OSA. However, patients with severe OSA may report few, minor, or no symptoms and each of the symptoms apart from those listed have multiple causes other than OSA. Stress is likely a major factor in the presence and severity of symptoms attributable OSA (164).

Table 1.

OSA symptoms—population and OSA prevalence

Symptom	Population prevalence	OSA prevalence
Snoring	35% with male predominance Men: 34%-44% Women: 19%	38%-80% with male predominance, nonspecific symptom (158)
Witnessed breathing pauses during sleep	6%-29%	10%-67% with male predominance (159)
Choking or gasping during sleep	1.5%	11%, highly specific symptom (158)
Excessive daytime sleepiness.	18%-28%	47% (160)
Insomnia	30%	39% with female predominance (161) Females: 35% Males: 20%
Fatigue	25%-30%	57% (160)
Morning headaches	5%-8%	12%-18% (162)
Motor vehicle accidents	—	17% increased risk (163)
Nocturia	40%, higher in older age	25%-37% (90)
Mood disorder	20%	23% (119)

Abbreviation: OSA, obstructive sleep apnea.

Presence of OSA-associated comorbidities.

The comorbidities that occur most frequently with OSA are shown in Table 2. A careful clinical evaluation should be undertaken to determine whether these conditions are present and, if so, their severity.

Medication use

Treatment of hypogonadal men with testosterone transiently worsens OSA (174). Medication-induced weight gain increases OSA severity. The effects of atypical antipsychotics are disproportionate to their weight-increasing effects (175). Benzodiazepines increase the risk and severity of OSA, an effect related to proximity of use and cumulative dose (176). Opioids cause or exacerbate central sleep apnea but their effects on OSA are unresolved (177).

Alcohol and Smoking

Alcohol relaxes airway muscles, reduces sensitivity to apnea, and causes weight gain. A systematic review and meta-analysis showed that consumption of any alcohol increases the risk of OSA by 25% (any compared with none or high compared with low intake), independent of obesity (178).

Smoking causes disordered sleep structure, high arousal index, hypoxemia during sleep, increased nasal resistance, and airway collapsibility and increases the incidence and severity of OSA. Smokers also have higher ESS scores (179).

Family History

First-degree relatives of people with OSA are more likely to have OSA independent of age, obesity, and sex. The heritability estimates for the severity of OSA based on AHI is 40%, and for insomnia and EDS are 25% to 45% and 70%, respectively (180).

Examination and laboratory investigations:

The risk and severity of OSA increases with increasing obesity. BMI does not account for the effects of fat distribution. A waist circumference of greater than 95 cm in females or 100 cm in males or neck circumference exceeding 42 cm in males or 39 cm in females are better associated with risk. Nevertheless, obesity alone, unless very severe, has a low positive predictive value for OSA (181), and approximately 25% of the adult population with OSA have a normal BMI (182). Nasal obstruction, tonsillar hypertrophy, and midfacial structural factors such as retrognathia, micrognathia, maxillary hypoplasia, and high arched palate should be looked for. Further examination, tailored by sex and age of the individual, should focus on the OSA-associated comorbidities previously listed.

We advocate an approach to investigation and management of a patient with obesity and OSA that considers the commonly associated comorbidities. Accordingly, we recommend that initial laboratory investigations include a full blood count to access for polycythemia, liver function tests (LFTs) to screen for fatty liver disease, HbA1c to screen for T2D or assess glycemic control in those with comorbid T2D, thyroid function to assess for hypothyroidism, and renal function and fasting urine albumin creatinine ratio in the setting comorbid T2D or hypertension. Other investigations should be dictated by the clinical presentation.

Screening Questionnaires

In nonclinical populations, self-administered questionnaires such as the ESS, Berlin, and STOP-Bang, fall short in identifying a requirement for PSG (183). In high OSA prevalence populations, these screening tools offer limited additional diagnostic value. An adequate clinical assessment can identify patients with high probability of significant OSA and good potential for treatment response. In such cases, clinicians should skip these questionnaires and directly refer for PSG (184).

Polysomnography

The gold standard is the overnight PSG, performed in a laboratory with supervision (157). Multiple channels record brain activity, eye movements, muscle activity, heart rhythm, airflow, respiratory effort, oxygen saturation, limb movements, snoring, and body position. The report details sleep metrics, limb movements, oxygen levels, and AHI. Home sleep studies are a practical and cost-effective alternative for patients with a high probability of OSA (157). If OSA is identified on a home sleep study, no further testing is required. Because of significant night-to-night variability in OSA severity, equivocal studies or suspicion of other sleep disorders necessitates additional sleep testing (185).

AHI, as an OSA severity metric, overlooks the intensity of desaturation and arousal events and inconsistently correlates with OSA-related complications (186). The ODI, a hypoxic burden indicator, independently correlates with multiple comorbidities and better predicts cardiovascular mortality than AHI (186-189).

Approach to Management

Specialist sleep physicians generally undertake management. However, the bidirectional interrelationships between obesity and OSA and metabolic and endocrine comorbidities require the input of an endocrinologist or metabolic physician.

Devices and Procedures to Treat OSA

The standard treatment approach for OSA is the use of CPAP, the benefit of which improves symptoms and lowers BP that is incremental with more than 4 hours of use on most nights (136). For REM OSA, CPAP should ideally be used throughout the night. A high proportion of patients either refuse to initiate, or choose to discontinue, CPAP use, most frequently because symptoms are not improved, or the device cannot be tolerated. The CPAP nonadherence rate based on a 7-hour/night sleep time has remained 34.1% over the past 20 years, and on average, behavioral interventions increase adherence by 1 hour per night (190). A mandibular advancement splint may be a satisfactory alternative (191).

Severe tonsillar hypertrophy or nasal obstruction should be referred for surgical opinion. More generally, surgical procedures targeted at the upper airway may be beneficial and safe for selected patients with severe and symptomatic OSA intolerant of CPAP (192).

Gastroesophageal Reflux

Treatment of gastroesophageal reflux with a proton pump inhibitor reduces the AHI, snoring, and daytime sleepiness (193).

Optimization of Health-related Behaviors and Weight Loss

A program for smoking cessation and counseling regarding alcohol consumption, if relevant, are required. Physical activity reduces the risk (194) and severity (195) of OSA and protects against the development of T2D in those with OSA (194) independent of change in BMI.

A Mediterranean diet improves OSA indices and symptoms more than standard care alone regardless of CPAP use or weight loss (196).

Longitudinal data from the Wisconsin Sleep Study showed that a 10% weight loss predicted a 26% (95% CI, 18-34) decrease in the AHI (6). In the Sleep-Ahead study, participants were randomized to either intensive lifestyle intervention (ILI) or usual care for 1 year. After 10 years of follow-up, weight loss and reduction in AHI in the ILI compared with the usual care group were 3.6 kg and 4.0 events/hour, respectively. For every kilogram of weight lost from baseline AHI decreased by .68 events/hour and .54 events/hour in the ILI and usual care groups, respectively (197). It may be more difficult to achieve weight loss if the OSA is not diagnosed and treated, and the ongoing presence of OSA may increase obesity related morbidity. Optimization of lifestyle behaviors are a critical component of managing OSA with well-established benefits to reduce the severity of OSA and independently improve comorbidities. This is particularly important given the limitations and uncertain benefits of CPAP for cardiometabolic comorbidities (198).

For patients with severe obesity, bariatric surgery is the most effective approach to manage both severe obesity and OSA, resulting in remission of OSA in 65% of cases (199).

Emerging therapies for obesity are likely to have significant benefit for the management of OSA. SGLT2 inhibitors reduced the relative risk of developing OSA by .35 and .37 in 3 different meta-analyses (200). In a group of patients with severe OSA treated with the glucagon-like peptide-1 receptor agonist (GLP-1RA) liraglutide for 32 weeks, their weights decreased by 5.7% and AHI by all 12 events/hour (201). Newer GLP-1RAs and combined GLP-1 and glucose-dependent insulinotropic polypeptide receptor agonists decrease weight by 15% to 20% and ameliorate the metabolic, cardiovascular, and renal comorbidities associated with obesity (202). Clinical trials to determine their efficacy for the treatment of OSA are under way (203).

Conclusion: The Cases Revisited

Each case represents a patient we would commonly see in clinical practice.

The male described in case 1 has a high pretest probability of OSA by virtue of his age, obesity, poorly controlled T2D, and hypertension. The presence of nocturia, fatty liver disease (suggested by ultrasound), sexual dysfunction, and mood disorder may all, at least in part, relate to underlying OSA. The loud snoring is consistent with OSA. There is no mention of daytime sleepiness, but its absence does not exclude severe OSA. Alcohol at night will significantly worsen OSA.

Semaglutide was added to his treatment and the linagliptin ceased. He agreed to stop drinking alcohol on most nights and reduce his intake of processed foods. He found walking difficult because of fatigue. After 4 months, he was still snoring but not as loudly, his mood and sexual function were a little better, and nocturia was occurring twice per night. His weight was 114 kg, waist circumference 113 cm, and BP 145/95 mm Hg. There was no nocturnal decline of BP on ambulatory monitoring. HbA1c was 8%, and although improved, gamma-glutamyl transferase, ALT, and triglycerides remained elevated. After discussing the options, including waiting for further weight loss with pharmacotherapy or choosing bariatric surgery, he elected to undergo overnight PSG. The overall AHI was 38, with a high hypoxic burden.

Treatment with CPAP was instituted with immediate cessation of snoring and nocturia. After 3 months, he reported walking 30 minutes most nights, and had markedly improved energy, mood, clarity of thought, and sexual function. His weight was 110 kg, waist circumference 108 cm, BP 142/85 mm Hg, and HbA1c 7.2%. It was agreed that if there was further, significant weight loss, the need for CPAP would be reviewed.

In case 2, there should be high suspicion of OSA in a woman who is postmenopausal, with obesity, dysglycemia, dyslipidemia, fatty liver, and a history of PCOS and AF (although currently in sinus rhythm). The persistence of hot flushes, brain fog, early morning headaches, and fatigue are typical of symptoms of OSA in women. As in men, nocturia is frequent. Overnight PSG showed an overall AHI of 15 and REM AHI of 32. Following 3 months of CPAP, which downloads showed she wore for 8 hours per night, there was a complete resolution of symptoms. Although not mentioned previously, she no longer craved sweet snacks in the evenings, and was attending exercise classes 3 times per week and walking 20 minutes on other days. Her weight had not changed. Fasting glucose, lipids, and LFTs were normal and HbA1c was 5.5%. She was in sinus rhythm, a situation we anticipate will endure.

Abbreviations

AF

atrial fibrillation

AHI

Apnea Hypopnea Index

BMI

body mass index

BP

blood pressure

COMISA

combination of obesity and insomnia

CPAP

continuous positive airways pressure

CVD

cardiovascular disease

EDS

excessive daytime sleepiness

ESS

Epworth Sleepiness Scale

FSH

follicle stimulating hormone

GLP-1RA

glucagon-like peptide-1 receptor agonist

GLUT4

glucose transporter type 4

IL-6

interleukin-6

ILI

intensive lifestyle intervention

LFTs

liver function tests

LH

luteinising hormone

MDD

major depressive disorder

NAFLD

non-alcoholic fatty liver disease

nREM

non-rapid eye movement

ODI

oxygen desaturation index

OHS

obesity hypoventilation syndrome

OSA

obstructive sleep apnea

PCOS

polycystic ovary syndrome

PDE5

phosphodiesterase type 5 inhibitor

PSG

polysomnography

REM

rapid eye movement

RLS

restless leg syndrome

SGLT2

sodium-glucose cotransporter-2

SHBG

sex hormone binding globulin

TNF-α

tumor necrosis factor alpha

T2D

type 2 diabetes

Funding

E.J.M. is supported by the Royal Adelaide Hospital Research Fund 2023 Early Career Research Fellowship (MYIP 17236).

Disclosures

E.J.M. declares no conflicts of interest. G.A.W. has received research funding from Lawley pharmaceuticals, Bayer, Weight Watchers, and Lilly; speaker's honoraria from Bayer Pharma and Besins Healthcare; and consulting fees from Bayer Pharma.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.