Obstructive sleep apnoea (OSA) is a common disorder that is associated with serious comorbidities with a negative impact on quality of life, life expectancy and health costs. As OSA is related to obesity and is associated with sleep disruption, increased inflammation and oxidative stress, it is not surprising that OSA has an impact on the secretion of multiple hormones and is implicated in the development of many endocrine conditions. On the other hand, many endocrine conditions that can affect obesity and/or upper airways anatomy and stability have been implicated in the development or worsening of OSA. This bidirectional relationship between OSA and the endocrine system has been increasingly recognised in experimental and epidemiological studies and there are an increasing number of studies examining the effects of OSA treatment on endocrine conditions and vice versa. In this review article, we will critically appraise and describe the impact of OSA on the endocrine system including obesity, dysglycaemia, the pituitary, the thyroid, the adrenals, the reproductive system and the bones. In each section, we will assess whether a bidirectional relationship exists, and we will describe the potential underlying mechanisms. We have focused more on recent studies and randomised controlled trials where available and attempted to provide the information within clinical context and relevance.
Invited Author’s profile
Abd A Tahrani is a National Institute for Health Research (NIHR) Clinician Scientist at the University of Birmingham (UoB), an Honorary Consultant Endocrinologist at the UniversityHospitals of Birmingham NHS Foundation Trust, the lead for weight management research and diabetic neuropathy services at Birmingham Heartlands Hospital, the lead for translational research in the Centre of Endocrinology, Diabetes and Metabolism (CEDAM), Birmingham Health Partners and the communication co-lead in the Institute of Metabolism and Systems Research (IMSR) at the UoB. He obtained his PhD in 2013 from UoB, which was focused on the impact of obstructive sleep apnoea (OSA) in patients with type 2 diabetes and was awarded a SCOPE National Fellowship in 2014 from the World Obesity Federation. Abd’s research has the following themes: the metabolic consequences of sleep-related disorders, obesity management and complications, diabetes-related neuropathy and the pharmacology of type 2 diabetes and obesity.
Obstructive sleep apnoea (OSA) is a common disorder that affects 13–33% of men and 6–19% of women (1). OSA is characteried by instability in the upper airways (UAs) leading to recurrent episodes of the UA obstruction, particularly during the transition to sleep and rapid eye movement (REM) sleep (characterised by low-amplitude, mixed-frequency theta EEG waves, pronounced eye activity and low muscle tone (2)) (Supplementary data, see section on supplementary data given at the end of this article) (3, 4, 5, 6). These repeated obstructions are associated with recurrent episodes of oxygen desaturation/re-saturation, cyclical changes in blood pressure (BP), heart rate, sympathetic activity and intrathoracic pressure, brief microarousals and changes to sleep architecture, such as the loss of REM and slow wave sleep (SWS or deep sleep is stage N3 of NREM sleep characterised by high-amplitude slow waves, further decrease in muscle tone, possible eye movement cessation and is a restorative sleep stage decreasing though with age (2)) (Fig. 1 and Supplementary data) (3, 5, 7).
The interactions between OSA and the endocrine system have attracted much attention and they often can be bidirectional, which is not surprising considering the diurnal secretion pattern of many hormones. In addition, OSA treatment (namely continuous positive airway pressure CPAP) has an impact on the endocrine system (such as insulin resistance, cortisol secretion) while treating endocrine disorders (such as obesity, hypothyroidism or acromegaly) can also improve OSA. Moreover, the well-established higher OSA risk in men vs women also emphasises the potential relationship between sex hormones and OSA pathogenesis. Hence, it is important to understand the links between OSA and the endocrine/metabolic system in order to improve our understanding of the pathogenesis and the comorbidities and mortality associated with OSA and a variety of endocrine disorders (8).
In this article, we will review the interactions between OSA and the endocrine system, and we will highlight the underlying mechanisms underpinning this bidirectional relationship when exists, as well as explore the potential impact of OSA treatment on the endocrine disorders and vice versa. Some aspects of this article require some understanding of the pathogenesis of OSA; hence, we have provided an overview of OSA and its pathogenesis in the Supplementary data.
OSA and obesity interplay
Obesity is a major risk factor for the development of OSA (9, 10, 11), which is driving the increase in OSA prevalence (1, 12). Obesity prevalence in patients with OSA (approx. 70%) is also higher than that of the general population (13).
The impact of weight change on OSA
Weight changes have significant impact on OSA and its severity. In a longitudinal study of randomly selected patients from Wisconsin, a 10% weight gain over 4 years was associated with 32% (95% CI 20–45%) increase in the apnoea–hypopnoea index (AHI: the average number of apnoea and hypopnea events per hour of sleep) and six-fold higher risk of developing moderate-to-severe OSA (95% CI 2.2–17) compared to weight stability (11). On the other hand, 10% weight loss was associated with 26% (95% CI 18–34%) decrease in the AHI compared to weight stability (11), partly due to a reduction in UAs collapsibility observed with weight loss (14). The favourable impact of weight loss on OSA and its severity seems to be evident regardless of the method of losing weight such as lifestyle interventions, pharmacotherapy or bariatric surgery as has been shown by several studies among them and randomised controlled trials (RCTs) (14, 15, 16, 17, 18).
In a RCT, of 60 patients with obesity and moderate-to-severe OSA, laparoscopic adjustable gastric banding (LAGB) resulted in greater weight loss (5.1 vs 27.8 kg) and greater reductions in AHI (based on PSG) (−14.0 vs −25.5 events/h; between-group difference was −11.5 events/h 95% CI −28.3 to 5.3; P = 0.18) over 2 years compared to lifestyle intervention (dietary, physical activity and behavioural conventional programme) (15). In a recent post hoc analysis of this RCT, patients who achieved a normal supine AHI (i.e. AHI <5/h) lost significantly more weight than those who had persistently elevated AHI (weight change −23.0 (−21.0 to −31.6)% vs −6.9 (−1.9 to −17.4)%, P = 0.001) (19). Other studies also showed significant improvements in the AHI and a high proportion of OSA resolution following sleeve gastrectomy and gastric bypass (16, 17). A meta-analysis confirmed the positive impact of bariatric surgery on OSA severity, by showing a significant reduction of AHI post surgery (by 38.2 events/h, 95% CI: 31.9–44.4) (20). A more recent systematic review and meta-analysis by Wong et al. showed that bariatric surgery was associated with a reduction in the AHI (WMD −25.1 events/h (95% CI −29.9, −20.2)); with the pooled mean pre- and post-surgery AHI of 39.3 ± 15.1 and 12.5 ± 5.6 events/h respectively; however, OSA persisted in most patients and there was high between-studies heterogeneity mostly due to baseline AHIO and duration of follow-up (21). Hence, RCTs remain needed to address the impact of bariatric surgery on OSA, although these might be challenging to conduct. In another RCT, liraglutide 3 mg daily combined with lifestyle intervention resulted in greater reductions in weight (−5.7% vs −1.6%, P < 0.0001) and AHI (−12.2 vs −6.1 events/h, estimated treatment difference: −6.1 events/h; 95% CI −11.0 to −1.2, P = 0.015) compared to lifestyle intervention only over 32 weeks (18). The degree of weight loss correlated significantly with improvements in OSA in this trial (18).
The impact of OSA on weight
The impact of OSA on obesity is controversial. One possibility is that OSA could lead to worsening obesity via multiple mechanisms such as increased excessive daytime sleepiness (EDS) leading to a reduction in physical activity, sleep disruption leading to changes in hunger and satiety hormones (37, 38, 39) (leptin resistance, increased ghrelin, increased orexin and neuropeptide Y levels), changes to sleep duration and architecture (40, 41, 42, 43, 44). Sleep restriction was associated with increased activation of the brain regions related to emotional response to stimuli and motivation and reward system based on functional MRI, which was similar to what observed following energy deprivation resulting in corrective behaviour of seeking food (45, 46). This is supported by cross-sectional studies showing that the AHI was significantly associated with increased preference of calorie-dense foods independent of the severity of obesity in adolescents and children (47, 48) and that visceral obesity was increased in patients with OSA and short sleep duration (<5 h/night) (OR, 4.40, 95% CI 1.80–10.77), compared to those who slept ≥7 h/night (49). In addition, disruption of sleep architecture (suppression of SWS as happens in OSA) without affecting sleep duration in young healthy men, increased hunger for high-calorie food in the afternoon and evening (50). OSA could also contribute to increased fat mass by activation of the HPA axis and increased cortisol secretion and by hypercapnia-induced adipogenesis (51, 52). However, despite the above-mentioned plausible mechanisms, epidemiological evidence for an impact of OSA on weight longitudinally is lacking. One small (n = 53) prospective study of patients with newly diagnosed OSA showed 7.4 ± 1.5 kg weight gain over 12 months, but these patients had also a history of weight gain in the year preceding OSA diagnosis (53), hence quantifying the impact of OSA is difficult without an appropriate control group (Fig. 2B).
Nonetheless, if OSA is a cause of obesity, then it would be expected that OSA treatment will lead to weight loss. However, a systematic review of 3181 patients from 25 RCTs showed that CPAP resulted in a modest but statistically significant increase in BMI and weight compared to control (BMI change: −0.018 ± 0.243 kg/m2 for controls vs 0.134 ± 0.273 kg/m2 for CPAP; weight change: −0.096 ± 0.718 kg for controls vs 0.417 ± 0.718 kg for CPAP) (54). The mechanisms behind the weight gain after CPAP are not fully elucidated. However, CPAP reduces leptin (satiety hormone), intermittent hypoxia and sympathetic activity leading to reductions in lipolysis and energy expenditure and hence can cause weight gain (55, 56, 57, 58, 59, 60, 61).
Furthermore, it is plausible that OSA can lead to weight loss via increased sympathetic activity leading to increased energy expenditure and lipolysis via lipoprotein lipase inhibition and sympathetic activation (62, 63). The net effects of the above-mentioned opposing mechanisms/impacts of weight gain and weight loss is potentially weight maintenance in patients with OSA. CPAP treatment tilts the balance between these opposing mechanisms towards weight gain by inhibiting sympathetic activity (Fig. 2), but this might be opposed to a certain degree by the impact of CPAP on increasing growth hormone (GH) levels leading to lipolysis (64). The above, however, is only a hypothesis that requires further investigations.
OSA and dysglycaemia
As obesity is a major risk factor for OSA, much of the research in this field has focused on pre-diabetes/T2D. However, it is now increasingly recognised that OSA is common in patients with T1D as well. In this section, we will focus mostly on pre-diabetes/T2D but we will also summarise the evidence regarding T1D.
In general population studies, OSA has been shown to be associated with various comorbidities, including T2D (9), which is not surprising since obesity is a common risk factor for OSA and T2D (7, 65). Several cross-sectional studies showed a high prevalence of OSA (mild: 5 ≤ AHI < 15; moderate: 15 ≤ AHI < 30; severe: AHI ≥30) in patients with T2D (8.5–86%, 23.8–70% moderate-to-severe OSA), and a high prevalence of T2D in patients with OSA (15–30%) (7, 66). This variation in prevalence estimates is due to different diagnostic methods and criteria used to define OSA and differences in study populations (67, 68, 69, 70, 71).
Longitudinal studies have also shown that OSA is an independent risk factor for the development of T2D. A recent meta-analysis of eight studies (63,647 participants) showed that OSA was an independent risk factor for T2D after adjustment for age, sex and BMI (adjusted RR: 1.49, 95% CI: 1.27–1.75), which remained significant even in studies that defined OSA as AHI ≥5 (adjusted RR: 1.42; 95% CI 1.02–1.99) (72). A small RCT of 12 weeks in 80 patients with obesity (BMI >45 kg/m2 and mostly with metabolic syndrome) suggested that CPAP resulted in improvements in impaired glucose tolerance status compared to no CPAP and that CPAP lowered the 2-h glucose levels following OGTT (73). However, there remains a need for large RCTs of long duration to assess the impact of CPAP, on its own or in combination with lifestyle intervention, on T2D prevention.
OSA and insulin resistance and β-cell function
The impact of OSA on incident T2D is likely to be mediated by the effects of OSA on insulin resistance (IR) and β-cell dysfunction (7). Studies that examined the relationship between OSA and IR had conflicting results, due to variations in the definitions of OSA and IR, but most of the studies showed an association (65). The association between OSA and IR was present in lean men, suggesting that the relationship is not dependant on obesity (74, 75). Variation in EDS might contribute to the variation in the associations between IR and OSA observed in the different studies as Barcelo et al. showed that the association between OSA and IR was only evident in patients with EDS vs without EDS despite being matched for BMI (76). In support of the relationship between OSA and IR, a recent meta-analysis of six RCTs of adults without diabetes showed a favourable effect of CPAP on IR vs no CPAP (mean difference in HOMA-IR: −0.43; 95% CI: −0.75 to −0.11, P = 0.008) (77).
The impact of OSA on β-cell function is much less examined in the literature. In one study of patients without diabetes, patients with moderate-to-severe OSA had a lower β-cell function (measured using the disposition index during frequent sampling intravenous glucose tolerance test (IVGTT)) compared to healthy controls; and higher AHI was associated with lower β-cell function despite adjustment for obesity (78). Similar results were found in a more recent study (79) and in another study in patients with T2D (80). Similar to IR, CPAP improved β-cell function in compliant patients with moderate-to-severe OSA without diabetes (uncontrolled trial) (81) or with pre-diabetes (RCT) (82).
Mechanisms: OSA leading to dysglycaemia and T2D
In rodent models, IH has been shown to increase β-cell death (83) and impair β-cell function (84). Results from experimental studies in healthy adults showed that 5 h of IH (24.3 events/h, average oxygen saturation 90.6%, range 75.4–98%) resulted in blunted, rather than increased, insulin secretion despite reductions in insulin sensitivity (based on IVGTT) (85). Chronic IH can lead to β-cell dysfunction and IR via increased oxidative stress (86), which pancreatic β-cells are less able to handle compared to other tissues (87, 88, 89), and increased inflammation (increased CD8+ cytotoxic T-cells recruitment, shift to M1-proinflammatory macrophages in crown-like structures, IL and TNFA) (90, 91). In addition, chronic IH can increase free fatty acid (FFA) release leading to ectopic fat deposition in the liver and muscle rusting in IR (90). The impacts of chronic IH and oxidative stress on IR could also be mediated by hypoxia-inducible factor (HIF) tissue effects (92). In rodents, 35 days of chronic IH decreased insulin receptor expression and phosphorylation in skeletal muscle and adipose tissue, but not in the liver which was accompanied by upregulation of HIF1a in the liver and downregulation of HIF1a and HIF2a in skeletal muscle (93).
Changes in sleep architecture can also contribute to the effects of OSA on glucose metabolism (94). In an experimental study of young healthy adults, all-night suppression of SWS (without awakening the subjects, changing sleep duration or REM sleep) was achieved via acoustic stimuli of varying intensity and frequency for three nights (94). This resulted in a reduction in insulin sensitivity (by 25%, which is similar to a weight gain of 8–13 kg) without a compensatory increase in insulin release (based on IVGTT) (94). These changes in insulin sensitivity and β-cell function were associated with increased sympathetic activity and in some cases changes in cortisol levels (94, 95). In addition, several other neurohormonal mechanisms are involved in the links between OSA and T2D, which are summarised in Fig. 3 (30, 39, 51, 65, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113).
The impact of dysglycaemia on OSA
While the impact of OSA on glucose metabolism has been widely studied, the impact of T2D and dysglycaemia on OSA has not received much attention. Many cross-sectional studies showed a high prevalence of OSA in patients with T2D as we detailed above, but whether this prevalence is higher than an age- and obesity-matched population without T2D remains unclear. Recently, a population-based study of 151,194 participants with T2D showed a hazard risk of incident OSA 1.53 (95% CI: 1.32–1.77) and further patients treated with insulin had higher risk of OSA, especially if they were women (1.43; 95% CI: 1.11–1.83) (114). In addition, the incidence and natural history of OSA in patients with T2D are currently unknown. One longitudinal study assessed the relationship between IR and possible OSA prospectively and showed that HOMA-IR was an independent predictor for incident witnessed sleep apnoea (not formally diagnosed OSA) over 6 years (OR: 1.31; 95% CI 1.13–1.51) (115).
Several possible mechanisms make it plausible that dysglycaemia/diabetes can lead to the development or worsening of OSA as summarised in Fig. 3 (7, 11, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132).
OSA in patients with T2D
OSA and glycaemic control in T2D
Several cross-sectional studies in patients with T2D showed that patients with OSA had worse fasting plasma glucose, glycaemic variability and HbA1c compared to patients without OSA despite adjustment for confounders (difference in HbA1c between patients with and without OSA 0.7–3.7%) (7, 133, 134, 135). In addition, OSA severity is correlated with worse glycaemic measures (7). Interestingly, one study showed that the relationship between AHI and HbA1c was only evident for the AHI during REM sleep and not during NREM sleep (after adjustment for confounders) (136). This raised the possibility that OSA treatment might improve glycaemic parameters in patients with T2D.
Several uncontrolled trials showed that CPAP improved glycaemic variability, postprandial glucose levels and HbA1c over the short term (65, 137). However, three RCTs showed conflicting results. Two of these RCTs showed that CPAP had no impact on HbA1c (138, 139), while another RCT showed that CPAP for 6 months lowered HbA1c by −0.4% (95% CI: −0.7% to −0.04%; P = 0.029) while there was no change in HbA1c in the control group (140). These conflicting results could be due to differences in studies population (β-cell reserve), baseline glycaemic control (for example one of the negative RCTs had a baseline HbA1c of 7.3%, while the RCT that showed positive effects of CPAP had baseline HbA1c of 7.6%) (139) or study duration (3 vs 6 months) (138). There were no significant changes in weight or anthropometrics measures in these RCTs between the CPAP and the control arm to explain the conflicting results. However, an important difference between these RCTs was compliance with CPAP; the positive RCT showed CPAP usage of 5.2 h per night compared to below 4 h/night in the trial by West et al. (138, 140). Longer CPAP duration per night might have an important impact on glycaemic control as REM tends to occur later during sleep and the AHI during REM correlated with HbA1c better than the AHI during NREM (82, 136). Hence, there is still a need for well-designed RCTs of longer CPAP duration to answer the question whether CPAP can (or cannot) improve glycaemic control in patients with T2D.
OSA and vascular complications in patients with T2D
Several plausible mechanisms have led to the hypothesis that OSA could lead to the development or progression of macro- and microvascular complications in patients with T2D as shown in Fig. 4 (141, 142, 143, 144, 145, 146).
The relationship between OSA and CVD in patients with T2D has not been studied widely. A retrospective observational study showed that in patients with T2D and newly diagnosed OSA, CPAP for 9–12 months lowered systolic (mean change: −6.81, 95% CI: −9.94 to −3.67) and diastolic (−3.69, −5.53 to −1.85) BP (147). Similar reductions in BP levels were observed after 3 months of CPAP in an RCT in which patients with T2D and OSA were randomised to early (<1 week) or late (1–2 months) CPAP (148). The sleep AHEAD study showed an association between AHI and a history of stroke (adjusted OR: 2.57; 95% CI: 1.03–6.42) but not with coronary artery disease (149). In a longitudinal study in 132 patients with T2D and a normal baseline exercise echocardiography test, OSA predicted incident coronary artery disease (adjusted HR: 2.2; 95% CI: 1.2–3.9; P = 0.01) and heart failure (3.5; 1.4–9.0; P < 0.01) over a median follow-up of 4.9 years (150). In another recent study of 1311 patients who had percutaneous coronary intervention (PCI), OSA was associated with increased risk of major adverse cardiac and cerebrovascular events (MACCEs) over 3 years in patients with diabetes mellitus (adjusted HR: 2.03, 95% CI: 1.10–3.74, P = 0.023) after adjustment for age, sex, ethnicity, BMI and hypertension (151). There is no interventional RCT published regarding the impact of CPAP on CVD in patients with T2D.
OSA has been shown to be associated with diabetes-related microvascular complications including peripheral neuropathy, chronic kidney disease (CKD), retinopathy and autonomic neuropathy (71). Most of these studies were cross-sectional and no interventional studies have been published although several are ongoing.
A recent systematic review of 15 cross-sectional studies concluded that there was no convincing evidence that OSA was associated with diabetic retinopathy (DR), but that there was some evidence to suggest that OSA was associated with greater DR severity (152). The systematic review also suggested that OSA was associated with maculopathy (152). It is plausible that the impact of OSA on DR is more related to disease progression rather than the development of disease (which is a function of hyperglycaemia) (7). The increased retinal oxygen demands overnight will make the retina particularly vulnerable to the effects of the IH that occur in patients with T2D and OSA. This is supported by a recent longitudinal study in patients with T2D in which OSA was not associated with the development of DR but was associated with progression to pre-proliferative and proliferative DR (153). In this longitudinal study, OSA was associated with sight threatening DR (STDR) (adjusted OR: 2.3; 95% CI: 1.1–4.9; P = 0.035) and maculopathy (adjusted OR: 2.7, 95% CI: 1.2–5.9, P = 0.01) at baseline (153). After a median follow-up of 43.0 (IQR: 37.0–51.0) months, patients with OSA were more likely than patients without OSA to develop pre-proliferative/proliferative DR (18.4% vs 6.1%; P = 0.02), which remained significant after adjustment for potential confounders (adjusted OR: 5.2; 95% CI: 1.2–23.0; P = 0.03) (153). Interestingly in this study, patients with moderate-to-severe OSA who were compliant with CPAP were significantly less likely to develop pre-proliferative/proliferative DR compared to non-compliant patients (153). This finding was supported by another proof-of-concept study that showed that CPAP treatment ≥2.5 h/night CPAP over 6 months in individuals with OSA and significant macular oedema was associated with improvement in visual acuity but without improvement in the oedema (154). Currently, RCTs assessing the impact of CPAP on DR are ongoing.
In a systematic review of two longitudinal and ten cross-sectional studies, there was an association between OSA and CKD in patients with T2D (pooled OR: 1.73, 95% CI: 1.13–2.64) (155). In a longitudinal study in patients with T2D, CKD prevalence was higher in patients with OSA vs without OSA (49.3 vs 23.8%, P < 0.001), which remained significant after adjustment for confounders (adjusted OR: 2.64, 95% CI: 1.13–6.16, P = 0.02). OSA was also associated with lower eGFR and more micro- and macro-albuminuria (156). After an average follow-up of 2.5 (0.7) years, eGFR decline was greater in patients with vs without OSA (median: −6.8%; IQR: −16.1 to 2.2 vs −1.6%;−7.7 to 5.3%), P = 0.002) (156). After adjustment, having OSA (B = −3.8, P = 0.044) and higher AHI (B = −4.6, P = 0.02) were predictors of lower study-end eGFR (156).
The relationship between OSA and peripheral neuropathy in patients with T2D was examined in a cross-sectional study, which showed that OSA is associated with peripheral neuropathy based on the Michigan Neuropathy Screening Instrument (MNSI) vs patients without OSA (60 vs 27%, P < 0.001), which remained significant after adjustment (OR: 2.82; 95% CI: 1.44–5.52; P = 0.003) (143). In addition, OSA was associated with lower intra-epidermal nerve fibre density (based on skin biopsies), and a history of foot ulceration in patients with T2D (141). These studies suggest that OSA was associated with both large and small fibre neuropathy in patients with T2D. Cohort studies and RCTs assessing the relationship between OSA and CPAP on diabetes-related neuropathy and its complications are ongoing.
OSA and T1D
As patients with T1D tend to be lean or leaner than patients with T2D, examining OSA in T1D received much less attention than in T2D (157). However, there is increasing interest in OSA in patients with T1D, particularly that some recent studies suggest that OSA in T1D might be more related to autonomic neuropathy rather than obesity (158). In addition, epidemiological studies suggest that obesity prevalence is increasing in patients with T1D which might further increase their risk of developing OSA (159).
In a systematic review of four studies (n = 186 patients), the prevalence of OSA (defined as AHI ≥5) was 51.9% among adult patients with T1D, but the 95% CI was wide (31.2–72.6) reflecting the small sample size the variation between studies (160). The prevalence of moderate-to-severe OSA (AHI ≥15) in the same meta-analysis was 16.7% (95% CI: 1.1, 34.5) (160).
Autonomic neuropathy was suggested as one potential mechanism for the high prevalence of OSA in T1D as shown in a cross-sectional study of 199 patients with T1D in which OSA was present in 32% of the patients with normal BMI (161). And another study showed a higher prevalence of OSA in patients with T1D and cardiac autonomic neuropathy compared to patients with T1D but without neuropathy (67 vs 23%) (162). Other factors might contribute to the high prevalence of OSA in children and adolescents with T1D including lower mean lung volumes (FVC, PEF, MMEF) (163, 164) and impaired gas exchange with lower diffusing capacity for carbon monoxide (165). There are similar findings of impaired pulmonary function in adult patients with T1D (166, 167, 168). The natural history, impact and pathogenesis of OSA in patients with T1D remain poorly explored and large well-designed studies are needed.
OSA and the renin–angiotensin–aldosterone system (RAAS)
The links between OSA and RAAS activation are potentially bidirectional (Fig. 5). Hyperaldosteronism might also play an important role in the well-established links between OSA and hypertension (particularly resistant hypertension-RH) (Fig. 5) (9, 169, 170, 171, 172, 173).
The pathophysiology of hyperaldosteronism in patients with OSA is mainly attributed to the activation of the RAAS due to cyclical/IH (172). In addition, some studies suggested a higher prevalence of primary aldosteronism (PA) in patients with OSA compared to patients without OSA (173).
A recent meta-analysis has examined the relationship between OSA and RAAS activation (174). The meta-analysis included 14 studies, all but one were case–control studies and they included a relatively small sample size (mostly <100, range: 12–120) (174). The studies generally included middle age men and eight of them included patients with hypertension (174). The meta-analysis found no significant relationship between OSA and plasma renin activity (PRA) (mean difference: 0.17 ng/mL per hour (95% CI: −0.22 to 0.55, P = 0.40)) or plasma renin concentration (PRC) (mean difference 0.95 ng/mL (95% CI: −0.58 to 2.48, P = 0.23) (174). However, angiotensin II levels were significantly higher in patients with OSA compared to those without OSA (mean difference of 3.39 ng/L; 95% CI: 2.00–4.79, P < 0.00001) (174). There was a trend towards higher plasma aldosterone concentration (PAC) in patients with OSA vs no OSA (mean difference: 0.95 ng/dL; 95% CI: −0.16 to 2.07, P = 0.09) (174). However, when examined in patients with and without hypertension separately, patients with hypertension and OSA had significantly higher PAC vs patients with hypertension but without OSA (mean difference: 1.32 ng/dL; 95% CI: 0.58–2.07, P = 0.0005) (174).
The above-mentioned meta-analyses had high heterogeneity, which could be due to variations in the definition of OSA (174). The heterogeneity can also be attributed to the medication used prior to RAAS measurements; however, a meta-regression showed that anti-hypertensives did not affect the relationship between OSA and PAC (174). Supporting the findings of this meta-analysis, another study showing that the AHI correlated significantly with PAC and urinary aldosterone levels (r = 0.568, P = 0.0009; r = 0.533, P = 0.002, respectively) in patients with RH and hyperaldosteronism (175).
Several uncontrolled studies in patients with hypertension (mostly RH) showed that CPAP lowered angiotensin II and aldosterone levels (176, 177, 178, 179). One RCT in which 117 patients with RH were randomised to CPAP (n = 57) vs no CPAP (n = 60) showed that 6 months of CPAP resulted in greater reduction in aldosterone excretion (based on 24 h urine) compared to the control group in the per-protocol analysis (mean difference: −3.3 μg/24 h; 95% CI: −6.1 to −0.4 μg/24 h; P = 0.027) (180). However, the intention-to-treat analysis showed only a trend (P = 0.07). The impact of CPAP on lowering aldosterone was particularly evident in those with uncontrolled hypertension, nondipping in nocturnal BP, not using spironolactone and with patients with worse hypoxia (180). A recent meta-analysis of three observational studies and two RCTs (did not include the above-mentioned RCT) showed that CPAP lowered aldosterone levels compared to no/sham CPAP (mean difference: −0.236, 95% CI: −0.45 to −0.02, P = 0.034) (181).
On the other hand, RAAS activation and hyperaldosteronism might lead to or worsen OSA via multiple mechanisms as detailed in Fig. 5. In a retrospective cohort registry-based study, the risk of developing OSA was higher in patients with hypertension and hyperaldosteronism compared to those without hyperaldosteronism after adjustment for age, sex, BMI, diabetes mellitus and heart failure (adjusted OR: 1.8; 95% CI: 1.3–2.6) (186). Moreover, in a cross-sectional study of patients with RH, spironolactone treatment was associated with lower AHI (187). In another uncontrolled study in patients with RH, spironolactone (25–50 mg daily for 8 weeks) improved OSA severity (based on PSG) (AHI: 39.8 ± 19.5 vs 22.0 ± 6.8 events/h; P < 0.05) (188). A recent systematic review and meta-analysis found three studies (one RCT) and concluded that spironolactone reduced the AHI by a mean of −21.12 (95% CI: −27.47 to −14.77, P < 0.00001) (175). Furthermore, in a small study of 20 patients with PA who had PSGs, having MR antagonists (n = 13) or adrenalectomy (n = 7) resulted in AHI reduction from 22.5 (14.7) to 12.3 (12.1) (P = 0.02) (185). These studies support the notion that hyperaldosteronism could worsen OSA and suggest that aldosterone antagonists can be useful in patients with hypertension or PA and OSA.
Finally, due to the links between OSA and PA the recent guidelines of the Endocrine Society on the management of PA recommend that patients with hypertension and OSA are screened for PA (173). Furthermore, well-designed RCTs assessing the impact of MR antagonists on OSA are needed, particularly that OSA is associated with increased CVD risk and that CPAP compliance is often not optimal.
Although not directly related to RAS activation, it is important to note that patients with OSA can present with hypertension and the clinical and biochemical features of phaeochromocytoma without the presence of a catecholamine secreting tumour (i.e. pseudo-phaeochromocytoma) (110, 189, 190, 191). These cases are rare but have been reported in multiple case reports and series, and the clinical and biochemical features usually resolve with CPAP treatment or weight loss (110, 189, 190, 191).
OSA and hypothalamic–pituitary–adrenal (HPA) axis
Cortisol secretion has a well-described circadian rhythm and is closely related to sleep stages (192, 193). Sleep onset and SWS are associated with a decline in cortisol levels followed by increased cortisol secretion in late sleep (which is consistent with the rise in early morning) (194). On the other hand, cortisol might impact on sleep architecture, for example, HPA axis hyperactivity inhibits SWS and promotes nocturnal awakening (193).
OSA and HPA axis activation
The impact of OSA on HPA axis is controversial with conflicting results due to the confounding effects of obesity, the sampling frequency (single time point vs 24-h profile), variability in matching between patients with and without OSA, small sample sizes and short CPAP duration with variability in compliance. Some studies showed no relationship between OSA and the HPA axis, while some even suggested that OSA might inhibit the HPA axis (AHI and ODI correlated negatively with morning cortisol levels: r = −0.444, P = 0.002 and r = −0.381, P = 0.011 respectively) (195, 196, 197, 198). In a systematic review of studies that compared cortisol levels in patients with OSA to either obese or lean control, there was no evidence of HPA activation in patients with OSA in 6/7 studies (199). However, only two of these studies had plasma cortisol measurements over 24 h, while the rest had single time point measurements (199). The two studies that measured 24-h cortisol profile reported contradicting results as one showed no difference in mean 24-h plasma cortisol between patients with OSA and obese controls (200), while the other showed that OSA was associated with HPA activation compared to obese controls (201).
However, the impact of OSA on HPA axis may not necessarily be consistent over the 24-h period, as the study by Vgontzas et al. showed that mean plasma cortisol levels between 23:00 h and 07:00 h were higher in patients with OSA and obesity vs obese controls, consistent with nocturnal HPA activation when there is IH and disruption of the sleep architecture (201) (Fig. 6). Another important aspect is that the impact of OSA on HPA axis may not be simply related to basal or 24-h cortisol profiles but might be related to the dynamic responses to HPA inhibition or stimulation. Carneiro et al. showed that although basal salivary cortisol was not different between patients with OSA vs obese controls, the salivary cortisol inhibition following overnight dexamethasone suppression test (ONDST) was significantly less pronounced in patients with OSA compared to obese controls (196). Interestingly, this deficit was corrected after 3 months of CPAP (196). Another study also showed that ACTH responses to CRH stimulation were higher in patients with OSA compared to obese and lean controls (202).
In the same above-mentioned systematic review, eight uncontrolled studies assessed the impact of CPAP on cortisol levels (blood or salivary) (199). Five studies showed no impact (76, 196, 203, 204, 205), while three studies showed that CPAP lowered cortisol levels (blood and salivary) (201, 206, 207). The studies that showed favourable impacts of CPA measured cortisol more frequently during the 24 h compared to the negative studies (199). However, a recent in-laboratory study showed that 8 h of CPAP per night did not have any effect on 24-h cortisol profile (208). Nonetheless, this study was over a 1-week period, unlike the studies that showed positive impact of CPAP on cortisol which were over 3-month period. A slightly longer study of 14 days, showed that CPAP can lower morning salivary cortisol in men and women with obesity and OSA (209). The confounding effects of obesity and gender on the relationship between OSA and HPA axis were addressed in a recent study of nonobese men and postmenopausal women which showed that OSA patients had higher 24-h blood cortisol levels compared to controls, which were lowered after 2 months of CPAP (51).
Overall, while the studies showed conflicting results there is evidence that OSA is associated with HPA activation particularly nocturnally and that CPAP (14 days to 3 months) can lower cortisol 24-h profile rather than cortisol levels at single time points. The effects of OSA on the HPA axis can be mediated via mechanisms related to night awakenings (even when brief), sleep restriction and IH (51, 210, 211, 212, 213, 214, 215, 216) as shown in Fig. 6.
OSA in patients with Cushing’s syndrome
Several studies have shown that OSA is common in patients with Cushing’s syndrome (CS) (whether endogenous or exogenous) (217). The prevalence of OSA (based on PSG) was higher in women with active CS (n = 35) compared to age-, gender- and BMI-matched controls (n = 30) (50% vs 23%, P = 0.003) (218). After controlling for BMI and HOMA score, serum cortisol remained independently associated with AHI (R 2: 77.8%, P < 0.001), suggesting that the relationship between CS and OSA are not only related to obesity (218). A recent Taiwanese population-based cohort study showed that patients with CS (n = 53) were at increased risk of developing OSA compared to matched controlled (matched for age, sex and comorbidities including obesity, T2D and hypertension) (4.11 vs 1.70 per thousand person/year; HR 2.82, 95% CI: 1.67–4.77), with slightly higher risk in men vs women (219). Interestingly in this study, the survival curves for OSA development starting separating clearly from the first year after the diagnosis of CS (219). Similarly, in patients without OSA (n = 17) who had PSG before and after 3 months of prednisolone (10 mg daily or more), AHI worsened by 56% compared to controls (with mild OSA but no steroid treatment) (220). This increase in AHI did not correlate with changes in weight and neck circumference suggesting mechanisms other than adiposity responsible for the worsening in AHI (220).
While obesity might play an important role in the relationship between CS and OSA, it is clear from the above-mentioned studies that obesity is not the only factor. In addition to obesity, hyperglycaemia, IR and ectopic fat (in the peritoneum, mediastinum and parapharyngeal spaces) may also play a role in the increased risk of OSA in patients with CS (217, 221). Moreover, hypercortisolism can induce UA myopathy leading to compromised UAs (Fig. 6) (217, 219, 222).
Future studies need to assess the impact of CS treatment on the incidence and severity of OSA and to examine whether the increased OSA risk in patients with CS is lifelong or simply related to the period where CS is active. In addition, endocrinologists, surgeons and anaesthetists need to be aware of the high risk of OSA in patients with CS when considering surgical treatment (both pituitary and adrenal) in order to ensure the safety of the surgical intervention.
OSA and GH/IGF axis
Summary of OSA impact on GH/IGF axis as well as the relation of GH excess and deficiency to OSA development or worsening can be found in Fig. 7.
OSA and the dysregulation of GH/IGF axis
OSA-associated chronic IH and disruption of sleep architecture can lead to dysregulation of the GH/IGF axis as GH secretion is increased after sleep onset and during SWS (both of which are disrupted in patients with OSA) (223, 224). Overall, studies in rodents and humans suggest that OSA is associated with suppression of basal and stimulated GH and IGF1 levels, which are improved by CPAP (225).
In rodents, IH was shown to cause a recoverable dose-dependent suppression of GH release and GH mRNA expression, possibly due to modulation of somatostatin activity (226). In humans, OSA was shown to be associated with a marked reduction in GH blood levels, which increased following one night of CPAP (64). In addition, fasting IGF1 levels correlated negatively with the ODI in men with OSA, but increased following 3 months of CPAP (195). Sleep disruption also plays a role in the relationship between OSA and the GH/IGF1 axis. In an experimental study of patients with OSA who were examined for one night without CPAP and one night with CPAP, GH plasma levels and secretion rate (bloods were collected every 10 min overnight) were reduced and increased after CPAP treatment; this improvement correlated with the improvement in SWS (227).
In support of the impact of OSA on the GH/IGF axis, a recent RCT in 65 middle-aged men with moderate-to-severe OSA showed that CPAP vs sham CPAP increased IGF1 levels, total and pulsatile GH secretion, mean GH concentration, mass of GH secreted per pulse and pulse frequency after 12 weeks of treatment with further increases in IGF1 levels and a decrease in IGFBP1 levels by week 24 (228). Furthermore, other treatments that can improve OSA, such as adenotonsilectomy in children, have also been shown to improve IGF1 and IGFBP3 levels (229).
Obesity is a potential confounder for the relationship between OSA and GH/IGF1 dysregulation as obesity (particularly visceral) is linked to a reduction in GH secretion, IGF1 levels and peripheral GH sensitivity, which can recover with weight loss (230). However, IGF1 levels were lower in patients with OSA compared to the weight-matched control despite that both these groups had lower IGF1 levels compared to the lean control (96).
OSA and acromegaly
Many cross-sectional studies showed that OSA is highly prevalent in patients with active acromegaly (45–80%) (231), with an average prevalence of 69% in PSG-based studies (232). Although lowering GH/IGF1 improves OSA, up to 40% (range: 21–58% (231)) of those with controlled acromegaly have persistent OSA that required evaluation and the consideration of CPAP (233, 234). ‘Although clinicians seem to be aware of the links between acromegaly and OSA (as shown by a survey in Italy), only few patients undergo PSG in clinical practice (235).
In addition, OSA contributed to the adverse outcomes of acromegaly, despite that there were no differences in GH or IGF1 levels between patients with OSA + acromegaly vs acromegaly alone (236). The presence of impaired glucose tolerance or T2D was higher in patients with acromegaly and OSA vs acromegaly only (n: 10/17 vs 5/19) (236), although this was not adjusted for obesity. In addition, OSA contributed to IR in patients with acromegaly, which improved by CPAP in a RCT (237). Furthermore, OSA might play an important role in other acromegaly-related comorbidities such as hypertension and heart failure/cardiomyopathy (238).
As a result of the high prevalence of OSA and its impact on acromegaly-related comorbidities, the 2014 Endocrine Society Clinical Practice Guideline for acromegaly recommended evaluating all patients for OSA (234). In addition, the guidelines recommended that patients with severe pharyngeal thickness and OSA should be treated with somatostatin receptor ligands preoperatively to reduce the OSA-related surgical risks (234).
On the other hand, a recent study of 507 patients with OSA showed that ten patients (1.97%) had elevated IGF1 levels, of which nine patients suppressed GH levels on OGTT giving an acromegaly prevalence of 0.2% (1/507) (239). These findings suggest that screening for acromegaly in OSA should not be routinely performed. However, if in addition to OSA, there are other features of acromegaly or acromegaly-associated conditions (such as T2D, debilitating arthritis, carpal tunnel syndrome, hyperhidrosis and hypertension), then measurement of IGF1 levels is recommended as per the Endocrine Society Clinical Practice Guideline for acromegaly (239). Finally, although we have focused here on OSA, central sleep apnoea (SA) can also occur in the context of acromegaly (240), but far less common than OSA (236).
The impact of acromegaly treatment on OSA
Considering that OSA is driven by the excess of GH/IGF1 in patients with acromegaly, it is not surprising that treating acromegaly can improve OSA, but it is also common for OSA to persist or even worsen after acromegaly is brought under control (234). In a small study of six patients with SA syndrome (obstructive or central with EDS) and acromegaly, trans-sphenoidal adenomectomy resulted in resolution of the SA syndrome in all patients regardless of whether acromegaly was cured or not (255). In another study of 24 patients with acromegaly (20 with OSA) who had remission following trans-sphenoidal surgery, at 1 month post surgery, the tongue area declined while the airway volume increased significantly, accompanied with improved OSA (256). The prevalence of severe OSA was reduced from 45.8 to 28% by 6 months with significant improvements in AHI but the average AHI remained in the moderate OSA range (256). Similar results were observed in patients with acromegaly following treatment with somatostatin analogues (246, 249, 257, 258, 259, 260) and pegvisomant (261, 262).
The above-mentioned studies clearly show that curing acromegaly or significant improvements in GH/IGF1 levels can improve OSA, but many patients with acromegaly have persistent moderate-to-severe OSA that might require CPAP. In fact, OSA might occur in patients with acromegaly following achieving normal IGF1 levels even when OSA was not present at baseline as shown by Chemla et al. (OSA cured in 57%, new OSA that was not present at baseline 22%) (263). Similarly, Castellani et al. showed that AHI increased in 55.5% of patients with acromegaly after complete/partial biochemical control (either after surgery, radiotherapy and/or medical therapy) (231). OSA persistence following acromegaly treatment is probably due to multiple factors including increased BMI and/or irreversible craniofacial-skeletal deformities/fibrosis (231). Hence, OSA evaluation is needed post acromegaly treatment regardless of the normalisation of GH/IGF1 (264).
OSA in adults with GH deficiency (GHD)
OSA is much less examined in GHD in comparison to acromegaly. OSA is very common in adults with GHD with a prevalence of 63%, which is mainly due to the increased obesity either due to GHD or hypothalamic obesity as a result of surgical or radiotherapy treatment delivered to the underlying pituitary or hypothalamic pathology (265).
GH replacement and OSA
GH replacement in patients with GHD might improve OSA due to a reduction in adiposity (strong lipolytic properties of GH (266, 267)) or it could worsen OSA if the replacement was excessive. The studies in the literature show a mixed picture. In a small study of five men who received GH replacement (median dose: 2 U/day; median serum IGFI: 351 µg/L) for 1–2 years post pituitary surgery GHD showed that 6 months after stopping GH treatment the median obstructive AHI decreased significantly from 4.4 to 0.1 (P = 0.03), whereas the central AHI increased from 6.3 to 14.6 (P = 0.03), suggesting that GH replacement worsened the OSA but improved central SA (268). However, another study of 19 patients with GHD showed that GH replacement for 6 months had no impact on AHI (pre vs post treatment: 28.2/h vs 28/h), regardless of baseline OSA status (265). Still, in a large observational longitudinal study of GH-treated (n = 1988) and- untreated (n = 442) patients with GHD showed that after a mean follow-up of 2.3 years the SA incidence was greater in the group that received GH replacement (3.3 vs 0.9%, P < 0.05), despite that the GH-treated vs -untreated groups had similar BMI at baseline and the GH-treated group were younger (269). However, the GH-treated group had higher baseline IGF1 levels (108 ± 61 vs 90 ± 51 µg/L, P < 0.001) and serum IGFBP3 levels (2.4 ± 0.9 vs 2.1 ± 1.0 µg/L, P < 0.001) (269). In a 12-month double-blind RCT of 40 men with obesity and dysglycaemia who were randomised to either GH or placebo, GH treatment increased IGF1 from 168 ± 72 to 292 ± 117 µg/L, the AHI from 31 ± 20 to 43 ± 25 and the ODI from 18 ± 14 to 29 ± 21 (all P values ≤0.001) (270). Interestingly, GH treatment in this study increased neck transverse diameter, circumference and total cross-sectional area, while reduced abdominal visceral adipose tissue (based on CT) (270).
Hence, more data are required to assess the impact of GH replacement on pre-existing OSA and the development of new OSA. However, GH replacement might result in the development or worsening of pre-existing OSA via increasing IGF1 levels or via affecting adipose tissue distribution (increasing neck circumference).
OSA in Prader–Willi syndrome
Children and patients with Prader–Willi syndrome (PWS) are also at high risk of having OSA (prevalence: 1:10,000–25,000 live children), and as a result, screening for OSA in this population has been recommended (271). The high prevalence of OSA in patients with PWS is likely to be multifactorial due to GH deficiency, increased viscosity of UAs secretions, craniofacial abnormalities with small airways, UAs muscles hypotonia and secondary alveolar hypoventilation (obesity and scoliosis causing lung volume restriction) all leading to airway collapsibility (271).
The impact of GH replacement on OSA in children with PWS is debatable. Salvatoni et al. showed that short-term treatment with rhGH (6 weeks) did not worsen the AHI, and there was no difference in AHI between the treatment and control group at baseline or study end (272). Nonetheless, in this study, the AHI increased (i.e. OSA worsened) in 50% of the cases following GH replacement (272). Similar results were shown in another study suggesting that the AHI worsens in a subgroup of patients following GH replacement over the short run (273), which in part could be due to the development of adenotonsillar hyperthophy following GH treatment (273). However, longer term follow-up (2 years) showed that GH replacement did not worsen AHI during the follow-up except in those who worsened shortly after GH initiation (274, 275). As a result, the 2013 consensus guidelines considered untreated severe OSA as an exclusion criteria for rhGH initiation, till the patient is treated with CPAP (276, 277). This is particularly important considering that sudden death early in the course of GH replacement in patients with PWS, associated with sleep-disordered breathing/OSA, have been reported in the literature (278, 279, 280).
OSA and hypothalamic–pituitary–thyroid (HPT) axis
OSA in patients with hypothyroidism
A recent systematic review of one observational and five interventional studies (501 patients in their 4th–5th decade of life) found that 25–50% of patients with overt hypothyroidism (OH) had nocturnal breathing abnormalities (snoring, choking, apnoea periods), which improved with levothyroxine 4 (LT4) treatment (281). In one study, 30% of patients with recently diagnosed OH had evidence of OSA (AHI ≥5 based on PSG) and LT4 improved the AHI (from a median of 14.3 (7.4–33.6) to 2.1 (0.8–4.6)) (282). In addition, in the later study LT4 treatment improved hypoxaemia and sleep architecture (TpO2 sat <90%: 14% (2.2–19.9) vs 0.2% (0–1.7), P < 0.05; SWS%: 18.4 (7.2–25.2) vs 28.2 (15–33.4), P < 0.05) (282). This suggests that hypothyroidism can lead to/worsen OSA, which improves with LT4 treatment. However, larger studies including RCTs are needed before confirming this relationship.
There is lack of good-quality data regarding the relationship between OSA and subclinical hypothyroidism (SH); one small observational study (n = 108) showed that 53% of patients with untreated SH had OSA (based on PSG) (283). However, these results are likely to represent selection bias as the prevalence of OSA in healthy controls with normal thyroid functions was higher (75%) than that in patients with untreated SH despite that SH patients were heavier and the patients are recruited from the respiratory department (283). Hence, currently we cannot be certain about the relationship between OSA and SH.
Hypothyroidism in patients with OSA
While studies are not consistent, overall, there is no evidence that hypothyroidism is more common in patients with OSA compared to patients without OSA (284, 285). A recent study also supported this conclusion as it showed that the prevalence of raised TSH in 813 patients with severe OSA was 4.7% which is similar to the general population (286). Some studies showed that the prevalence of SH was higher in OSA vs control, but these studies have potential selection bias as the population was recruited from sleep clinics and the control group was younger and leaner (287, 288, 289). Other studies did not show a high prevalence of SH in patients with OSA (290). In a study of 245 euthyroid patients with suspected OSA, the prevalence of Hashimoto’s thyroiditis was 32.2% in patients without OSA vs 46.8% in patients with OSA (based on PSG) (P = 0.03) (291). The prevalence of Hashimoto’s increased with worsening severity of OSA (291).
Mechanisms linking OSA and thyroid disorders
OSA and non-thyroidal illness syndrome (NTIS)
A recent cross-sectional study showed that patients with moderate-to-severe OSA (n = 125) had a higher prevalence of NTIS (defined as normal TSH and low FT3) compared to controls (n = 60) (10.4 vs 0%), but the control group was lean, and there were more men in the OSA group (301). Within the OSA group, patients with NTIS had worse nocturnal hypoxemia compared to patients without NTIS (301). This suggests that IH could play a role in the high prevalence of NTIS in patients with OSA, possibly via downregulation of deiodinase 1 and enhancing deiodinase 3-inactivating T3 and T4 (302). In addition, oxidative stress and low-grade inflammation, resulting from OSA, can also contribute to the association between OSA and NTIS (303, 304). CPAP for 5 months has been shown to improve FT3 levels in patients with NTIS supporting the notion that OSA might lead to NTIS, but this study was not controlled (301). However, it is important the clinicians take into account the possibility of NTIS when interpreting thyroid function results in patients with OSA.
In summary, SA and thyroid specialists need to have a low threshold to test for thyroid disorders if indicated clinically. In addition, OSA can be associated with NTIS and clinicians interpreting the thyroid function results need to take the presence of OSA into consideration. However, cohort studies with well-matched control groups and RCTs are needed to enable us to understand the complex relationship between OSA and HPT axis and the impact of treating one or the other.
OSA and the hypothalamic–pituitary–gonadal (HPG) axis
The interaction between sex hormones and OSA was initially brought to attention by the consistently reporting a higher prevalence of OSA in men vs women. This relationship was further emphasised by several observations including that testosterone replacement in men worsens/increases the risk of having OSA, the prevalence of OSA in postmenopausal women was higher than that in premenopausal women; hormone replacement therapy reduced the risk of OSA in postmenopausal women and oral contraceptives were associated with lowered OSA risk in women with polycystic ovarian syndrome (PCOS) (65, 305).
OSA is associated with hypogonadotropic hypogonadism due to altered gonadotropin synthesis and release (306). In a cross-sectional analysis of a prospective study of healthy older men (n = 1312, ≥65 years old), lower testosterone levels (based on quartiles) were associated with significantly less SWS, higher AHI (based on PSG) and more sleep time spent with O2 sat <90% after adjustment for age and race (307). However, adjustment for BMI made these associations non-significant (307). Other studies showed that patients with OSA had lower area under the curve and mean levels for LH (24.9 ± 10.2 IU/L vs 43.4 ± 9.5 IU/L, P < 0.005) and testosterone (67.2 ± 11.5 nmol/L vs 113.3 ± 26.8 nmol/L, P = 0.003) compared to healthy controls, but the control group was leaner numerically (308). Similar findings were found in other studies (309, 310, 311).
Testosterone replacement and OSA
Patients receiving testosterone replacement are at increased risk of developing OSA. In a cohort study, 3422 of US military service members, aged 40–64 years, who were free of OSA at baseline and received testosterone replacement, were matched based on age and comorbidities to men who did not receive testosterone treatment (312). The absolute 2-year risk of incident OSA was greater in patients who received testosterone replacement vs those who did not (16.5% (95% CI: 15.1–18.1) vs 12.7% (95% CI: 11.4–14.2), P < 0.001) (312). Interestingly, the increased risk of OSA was greater for those who used injectable vs topical testosterone (312). This is also supported by a small RCT in which healthy ambulatory men aged >60 years were randomised to receive three injections of weekly intramuscular testosterone esters (500, 250 and 250 mg) or matching oil-based placebo and then crossed over to the other treatment after 8-week washout. Testosterone replacement in this RCT resulted in worsening RDI (approximately by seven events per hour), mainly during non-REM (NREM) sleep and worsened nocturnal hyoxaemia measures, while placebo had minimal effects on RDI and hypoxia parameters (313). Several other studies suggested a link between testosterone replacement and incident or worsening OSA (314, 315, 316, 317). As a result, the Endocrine Society clinical practice guidelines recommended against the use of testosterone replacement in men with untreated severe OSA (318). It is unclear whether different methods of testosterone replacement have a differential impact on the risk of developing or worsening OSA due to the variations in the pharmacokinetics profiles of these agents.
The effects of testosterone can be time limited as shown in a RCT of 67 men who received hypocaloric diet and were randomised to intramuscular injections of 1000 mg testosterone undecanoate or placebo (319), in which testosterone replacement worsened the ODI by 10.3 events/h (95% CI, 0.8–19.8 events/h; P = 0.03) and on nocturnal hypoxaemia at 7 weeks but not at 18 weeks (319). This time-dependent effects might be as a result of time-dependent changes in hyperoxic ventilatory recruitment threshold following testosterone replacement (320).
Low testosterone in men can lead to loss of muscle mass and increased visceral adiposity, which can contribute to the increased/worsening OSA in men with hypogonadism (321, 322). It is unclear how testosterone replacement leads to OSA, but postulated mechanisms include altered ventilator responses such as increased response to hypoxaemia (leading to CO2 levels below apnoea threshold), reduced sensitivity to hypercapnia or anabolic effects (leading to UA narrowing) and an effect on the neuromuscular control of UA (323, 324). However, these mechanisms are not well proven with multiple studies showing conflicting results. In one interesting mechanistic study, androgen blockade with flutamide did not influence chemo-responsiveness to hypoxia/hypercapnia (325).
In addition, OSA can impact the HPG axis via several mechanisms including IH, sleep fragmentation and obesity (306, 310, 326). Testosterone levels peak during REM (fewer REM sleep episodes and REM sleep latency are related to lower testosterone concentrations (323)); hence, the disruption of sleep architecture in OSA (loss of REM) might explain the link between OSA and low testosterone (193).
The impact of OSA treatment on the HPG axis
CPAP effects on the HPG axis in men remains controversial with a limited number of studies in the literature. A meta-analysis in 2014 found only two RCTs and five observational studies with a total sample size of 232 men showing the paucity of available data (327). In this meta-analysis, an average of 6 months of CPAP treatment had no effects on testosterone levels despite good CPAP compliance (standardised mean difference (SMD) = −0.14, 95% CI: −0.63 to 0.34) (327). CPAP also had no effects on free testosterone or SHBG levels (327).
Summary of the trials assessing the impact of OSA treatment (CPAP and surgical) on HPG axis can be found in Table 1 (195, 205, 328, 329, 330, 331, 332, 333, 334, 335). The two RCTs showed no effect of CPAP on testosterone levels, but the study participants did not have hypogonadism at baseline and the CPAP duration was short. The uncontrolled studies mostly showed no effects of CPAP on testosterone levels except two studies, which showed that CPAP increased testosterone levels (Table 2). In one of these studies, the increase in total testosterone was associated with increased SHBG, which suggests that the impact of free testosterone was rather limited. In the other study, patients had hypogonadism at baseline and CPAP improved testosterone levels along with LH, but the impact on SHBG was not reported (Table 1). Hence, the impact of CPAP on HPG axis in men remains unclear but future trials need to consider the potential difference in response between men with and without hypogonadism and need to ensure adequate CPAP treatment duration and the impact on free testosterone.
Non-randomised trials examining the effects of OSA treatment (CPAP/surgery) on the hypothalamic–pituitary–gonadal (HPG) axis (229, 239, 350, 351, 352, 353, 354, 355, 356, 357). Data are presented as mean or mean ± s.d.
|Study||Intervention group||Controls||OSA severity (AHI baseline)||CPAP duration (months)||HG at baseline||LH (IU/L)||Total testosterone (nmol/L)||Free testosterone (nmol/L)||SHBG (nmol/L)|
|(195)||43||NA||NA||NA||Severe||3||No||NA||NA||14.4 ± 1.2||17.0 ± 1.1*||0.512 ± 0.047||0.509 ± 0.035||19.8 ± 1.9||22.7 ± 2.0*|
|(328)||16||51.3||32||32a||Severe (43.3 ± 4.7)||7||No||3.0 (1–5)||4.0* (1–7)||17.7 ± 1.8||16.7 ± 2.1||0.5 ± 0.07 (n = 3)||0.43 ± 0.04 (n = 7)||13.7 ± 1.9 (n = 3)||27.1 ± 6.5 (n = 7)|
|(329)||5||49.5||31.7||NA||Severe||9||Yes||2.1 ± 0.8||2.6 ± 0.2||2.9 ± 6.7||2.6 ± 11.3*||NA||NA||NA||NA|
|(330)||67||56.2||33.7||NA||Severe||6||No||NA||NA||20.7 ± 7.45||20.1 ± 7.32||NA||NA||NA||NA|
|(331)||27||65.4||32||NA||Severe (median: 43)||3||No||NA||NA||12.7 ± 4.5||11.9 ± 3.5||0.26 ± 0.07||1 month: 0.23 ± 0.063 months: 0.25 ± 0.05||32.3 ± 11.6||1 months: 31.1 ± 13.23 months: 31.1 ± 12.5|
|(333)||53||43.87||28.71||NA||Severe||3||No||5.99 ± 2.50||5.82 ± 2.38||13.45 ± 5.48||14.05 ± 5.18||NA||NA||NA||NA|
|(334)||32||47.4||31.3||60b||Severe (55.9 ± 12.9)||1||Yes||4.1 ± 1.8||6.3 ± 2.1*||4.5 ± 1.4||12.1 ± 3.5*||NA||NA||NA||NA|
|(335)||12||9c||Moderate- severe||UPP, 3 months post surgery||No||NA||NA||13.31 ± 1.07||16.59 ± 0.72*||NA||NA||NA||NA|
aBMI and age-matched; bBMI matched, but were leaner; cBMI matched but were younger; *statistically significant.
AHI, apnoea–hypopnea index; BMI, body mass index; CPAP, continuous positive airway pressure; HG, hypogonadism; LH, luteinizing hormone; NA, not applicable; SHBG, sex hormone binding globulin; UPP, uvulopalatopharyngo-plasty.
|Study||CPAP arm||Control arm (n)||CPAP duration (MONTHS)||CPAP compliance||HG at baseline||Total testosterone (nmol/L)||LH (IU/L)|
|n||Age||BMI||Exposed cohort||Control group||Active NCPAP||Sham CPAP|
|(205)||52||NA||NA||49a||1||Yesc||No||13.5 ± 5.8||13.2 ± 4.7||14.4 ± 5.0||12.9 ± 4.4*||5.99 ± 2.55||5.81 ± 2.42||6.07 ± 2.70||5.34 ± 3.00*|
|(332)||27||51 ± 9||32 ± 3||20b||2.5||Yesd||No||16.0 ± 4.7||16.2 ± 5.2||19.1 ± 6.9||17.9 ± 5.8||NA||NA||NC||NA|
Data are presented as mean or mean ± s.d.
aSham CPAP; boral appliance; cfor both cohorts (5.4 and 4.6 h); d25/27 effective treatment; *statistically significant.
AHI, apnoea–hypopnea index; BMI, body mass index; CPAP, continuous positive airway pressure; HG, hypogonadism; LH, luteinizing hormone; NA, not applicable; s.d., standard deviation; SHBG, sex hormone binding globulin; UPP, uvulopalatopharyngo-plasty.
It is important to note that CPAP might still have beneficial impacts on scores for sexual and erection function despite the lack of impact of hormonal measurements (332, 333). However, in two RCTs sildenafil was superior to CPAP in regards to ED (336, 337).
OSA impact on the HPG axis in women is less well studied compared to men. Based on animal studies sex hormones can influence breathing not only via androgens but also via the effects of progesterone and oestradiol on CB and the brainstem (338). In addition, lack of progesterone receptor in rodent led to reduced hypoxic ventilator response (339) and lower UA resistance was found in the luteal phase in healthy premenopausal women with the peak in progesterone secretion (340). On the other hand, OSA has a negative effect on female sex hormones and on sexual function and is associated with PCOS.
In a cohort of 53 women (24–72 years old), AHI >10/h was associated with lower morning levels of 17-OH-progesterone, progesterone and oestradiol (341). However, hormone replacement therapy (HRT) in postmenopausal women was associated with lower prevalence of moderate-to-severe OSA compared to women not taking HRT and less time spent in oxygen saturations <90%, particularly in women who received combined oestrogen–progesterone vs oestrogen alone (342). The impact of CPAP on the HPG axis in women remains to be explored in large studies, and since one small uncontrolled study showed no effect (330) RCTs in this area are needed.
Similar to men, OSA has been associated with sexual dysfunction (FSFI score: desire, arousal, lubrication, orgasm, satisfaction and pain) in pre- and postmenopausal women compared to matched controls (343, 344) (345). Unlike in men, evidence for CPAP’s impact on sexual dysfunction in women is lacking (346). In this review we did not discuss the impact of OSA on pregnancy.
OSA and polycystic ovarian syndrome (PCOS)
OSA is highly prevalent in women of reproductive age with PCOS. A recent systematic review and meta-analysis from our group (15 studies, n = 568) showed that 36.1% (95% CI: 22.4–51.0) of women with PCOS had OSA regardless of the PCOS definition used (347). In addition, OSA prevalence was significantly higher in obese women with PCOS compared to lean (OR: 3.96, 95% CI: 1.29–12.13) and in adult women compared to adolescents, both of which are expected since obesity and age are main risk factors of OSA, and thus, PCOS precedes OSA development (347). However, in this meta-analysis, there was significant heterogeneity among studies, most studies came from the United States in women with obesity (class II), and there is a high level of selection bias since controls came from general population while exposed cohorts were recruited from specialised clinics (347). It is plausible that in some cases the OSA could precede PCOS development as detailed in a recent study showing that 1/3 of adolescent girls with PCOS had previous tonsillar enlargement/tonsillectomy (348).
It is also interesting that although androgens are considered to impact OSA pathogenesis, contributing to the higher OSA prevalence in women with PCOS, three studies showed that women with PCOS and increased androgens did not have higher prevalence of OSA compared to controls, and the relationship between OSA severity and hyper-androgonaemia were not consistent across the studies (347). This could be due to the low circulating androgen levels in women with PCOS compared to men.
In another meta-analysis from our group comparing women with PCOS and OSA vs women with PCOS only showed that the earlier group had higher BMI (mean difference: 6.01 kg/m2, 95% CI: 4.69–7.33), waist circumference (MD: 10.93 cm, 95% CI: 8.03–13.83), IR (HOMA-IR: MD = 2.23, 95% CI: 1.41–3.06; I 2 = 0%), systolic BP (10.8 mmHg 95% CI: 6.21–15.39), diastolic BP (4.63 mmHg 95% CI: 1.06–8.21), impaired glucose tolerance (2 h plasma glucose on OGTT: MD = 2.23, 95% CI: 0.67–2.11, I 2 = 0%) and worse lipids profile (higher total cholesterol, LDL and triglycerides and lower HDL) compared to the alter group (349). The androgen levels were not different between the two groups but hirsutism was worse in the OSA group (349). However, these studies included were relatively small, at high risk of selection bias, and did not account for important potential confounders such as obesity (349).
OSA and bone metabolism
Although cross-sectional studies assessing the relationship between OSA and bone mass density (BMD) showed conflicting results (351, 352, 353, 354); longitudinal studies showed an increased risk of osteoporosis in patients with OSA (355, 356). In a large retrospective cohort study of 1377 patients with newly diagnosed OSA and 22,655 matched controls (age, sex and index date), the risk of osteoporosis was greater in patients with OSA vs control in both men and women (incidence rate: 2.52/1000 person-years vs 1.00/1000 person-years, adjusted HR: 2.74, 95% CI: 1.69–4.44) over the 6-year follow-up (355). The HR in this study was adjusted for age, gender, diabetes status, obesity, CVD risk factors, CKD, CVD, gout and social demographics.
Consistent with the increased risk of osteoporosis in patients with OSA, several studies suggested that OSA might increase the risk of fractures, although these studies examined conditions that are related to OSA rather than OSA per se. In a study of 2911 men older than 67 years, men who spent ≥10% of their sleep time with O2 saturations <90% had increased risk of incident non-spinal fractures compared to men spent <1% of sleep time with O2 saturation <90% over 7 years follow-up (adjusted relative hazards: 1.42, 95% CI: 0.94–2.15, P = 0.047) (357). In the same study, the relative risk of having ≥1 fall was also higher in the group with nocturnal hypoxaemia (relative risk: 1.25, 95% CI: 1.04–1.51) (357). Another longitudinal study that followed up 8101 women aged 69 years or older for 6 years found that self-reported daily napping was associated with increased risk of incident hip fractures compared to women who did not nap daily (age-adjusted HR: 1.29, 95% CI: 1.02–1.65; fully-adjusted HR: 1.33, 95% CI: 0.99–1.78) and similar to the previous study there was an increased risk of falls in women who napped daily (358). In a recent cohort study, women (n = 3220) and men (n = 2969) aged 40 years and older, severe snoring (a common OSA symptom) was associated with increased risk of fractures over 10-year follow-up in women (adjusted HR: 1.68, 95% CI: 1.16–2.43, P = 0.006), with similar non-significant trend in men (359).
Consistent with the increased risk of osteoporosis and fractures in patients with OSA, bone resorption markers (such as serum C-terminal telopeptide of type I collagen CTX) has been shown to be higher in patients with OSA compared to controls in men and the AHI was independently associated with urinary CTX independently of age, BMI and other variables (352, 360). Furthermore, CPAP for 3 months lowered the creatinine-adjusted urinary CTX levels significantly (211 ± 107 vs 128 ± 59 μg/mmol/creatinine; P < 0.01) (360).
Summary and conclusion
In this review we have demonstrated that there are multiple bidirectional interactions between OSA and the endocrine system although the observed relationships varied depending on the endocrine system examined. The impact of OSA on the endocrine system was mostly mediated by intermittent hypoxaemia, sympathetic activation, the elevated BP and the increased inflammation and oxidative stress. While the impact of the endocrine system on OSA was mostly mediated via increased upper body adiposity, narrowing of the UAs, weakening of upper airway muscles, changes to chemosensitivity and ventilatory drive as well as autonomic dysfunction.
Our review also shows that there are multiple knowledge gaps in the field at a mechanistic level and also due to the lack of well-designed cohort and interventional studies in many areas. This is further complicated by the difficulty in achieving good compliance with CPAP in clinical studies, the diurnal nature of the endocrine system and the interaction between OSA and other sleep disorders such as short sleep duration and misalignment in the circadian rhythm. In particular, our review found the following need to be explored in future studies due to either no, minimal or inconsistent evidence currently available: the impact of OSA and CPAP on weight, the impact of diabetes treatment on OSA as well as the impact of OSA on diabetes-related outcomes, the impact of PA treatment on OSA, the effects of OSA on the HPA axis and the natural history of OSA and its response to treatment in patients with CS, the long-term impact of GH replacement on OSA as well as central SA, the impact of thyroxine replacement on OSA in patients with hypothyroidism, the relationship between OSA and SH, the impact of long-term testosterone replacement and the different methods of replacement on OSA, the impact of OSA and CPAP in women with PCOS and men with hypogonadism and the impact of CPAP on bone metabolism.
Finally, clinicians treating patients with endocrine conditions should not assume that OSA would recover by curing the underlying endocrine disorder (such as Cushing’s, acromegaly or hypothyroidism) and that OSA status need to be clarified by formal testing following the successful treatment of the endocrine condition. Furthermore, clinicians, surgeons and anaesthetists involved in the treatment of the endocrine conditions that are associated with OSA need to be aware of this association and treat the OSA in order to improve the safety of the general anaesthesia and surgical procedures.
This is linked to the online version of the paper at https://doi.org/10.1530/EJE-18-0411.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector. But A A Tahrani is a NIHR Clinician Scientist receiving funding from the National Institute for Health Research in the UK (CS-2013-13-029; 2013).
A A Tahrani is a clinician scientist supported by the National Institute for Health Research in the UK. The views expressed in this publication are those of the author(s) and not necessarily those of the National Health Service, the National Institute for Health Research or the Department of Health.
DixonJBSchachterLMO'BrienPEJonesKGrimaMLambertGBrownWBaileyMNaughtonMT. Surgical vs conventional therapy for weight loss treatment of obstructive sleep apnea: a randomized controlled trial. JAMA 2012 308 1142–1149. (https://doi.org/10.1001/2012.jama.11580)
SharplesAJCharalampakisVDaskalakisMTahraniAASinghalR. Systematic review and meta-analysis of outcomes after revisional bariatric surgery following a failed adjustable gastric band. Obesity Surgery 2017 27 2522–2536. (https://doi.org/10.1007/s11695-017-2677-7)
BlackmanAFosterGDZammitGRosenbergRAronneLWaddenTClaudiusBJensenCBMignotE. Effect of liraglutide 3.0 mg in individuals with obesity and moderate or severe obstructive sleep apnea: the SCALE Sleep Apnea randomized clinical trial. International Journal of Obesity 2016 40 1310–1319. (https://doi.org/10.1038/ijo.2016.52)
WongAMBarnesHNJoostenSALandrySADabscheckEMansfieldDRDharmageSCSenaratnaCVEdwardsBAHamiltonGS. The effect of surgical weight loss on obstructive sleep apnoea: a systematic review and meta-analysis. Sleep Medicine Reviews 2018 42 85–99. (https://doi.org/10.1016/j.smrv.2018.06.001)
SchwabRJGuptaKBGefterWBMetzgerLJHoffmanEAPackAI. Upper airway and soft tissue anatomy in normal subjects and patients with sleep-disordered breathing. Significance of the lateral pharyngeal walls. American Journal of Respiratory and Critical Care Medicine 1995 152 1673–1689. (https://doi.org/10.1164/ajrccm.152.5.7582313)
HeinzerRCStanchinaMLMalhotraAFogelRBPatelSRJordanASSchoryKWhiteDP. Lung volume and continuous positive airway pressure requirements in obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine 2005 172 114–117. (https://doi.org/10.1164/rccm.200404-552OC)
ArnardottirESMaislinGSchwabRJStaleyBBenediktsdottirBOlafssonIJuliussonSRomerMGislasonTPackAI. The interaction of obstructive sleep apnea and obesity on the inflammatory markers C-reactive protein and interleukin-6: the Icelandic Sleep Apnea Cohort. Sleep 2012 35 921–932. (https://doi.org/10.5665/sleep.1952)
PhillipsBGKatoMNarkiewiczKChoeISomersVK. Increases in leptin levels, sympathetic drive, and weight gain in obstructive sleep apnea. American Journal of Physiology: Heart and Circulatory Physiology 2000 279 H234–H237. (https://doi.org/10.1152/ajpheart.2000.279.1.H234)
VgontzasANPapanicolaouDABixlerEOHopperKLotsikasALinHMKalesAChrousosGP. Sleep apnea and daytime sleepiness and fatigue: relation to visceral obesity, insulin resistance, and hypercytokinemia. Journal of Clinical Endocrinology and Metabolism 2000 85 1151–1158. (https://doi.org/10.1210/jcem.85.3.6484)
BenedictCBrooksSJO'dalyOGAlmènMSMorellAÅbergKGingnellMSchultesBHallschmidMBromanJ-E. Acute sleep deprivation enhances the brain’s response to hedonic food stimuli: an fMRI study. Journal of Clinical Endocrinology and Metabolism 2012 97 E443–E447. (https://doi.org/10.1210/jc.2011-2759)
KritikouIBastaMVgontzasANPejovicSFernandez-MendozaJLiaoDBixlerEOGainesJChrousosGP. Sleep apnoea and the hypothalamic-pituitary-adrenal axis in men and women: effects of continuous positive airway pressure. European Respiratory Journal 2016 47 531–540. (https://doi.org/10.1183/13993003.00319-2015)
KikuchiRTsujiTWatanabeOYamaguchiKFurukawaKNakamuraHAoshibaK. Hypercapnia accelerates adipogenesis: a novel role of high CO2 in exacerbating obesity. American Journal of Respiratory Cell and Molecular Biology 2017 57 570–580. (https://doi.org/10.1165/rcmb.2016-0278OC)
ChopraSRathoreAYounasHPhamLVGuCBeselmanAKimIYWolfeRRPerinJPolotskyVY et al. Obstructive sleep apnea dynamically increases nocturnal plasma free fatty acids, glucose, and cortisol during sleep. Journal of Clinical Endocrinology and Metabolism 2017 102 3172–3181. (https://doi.org/10.1210/jc.2017-00619)
HendersonLAFatoulehRHLundbladLCMcKenzieDKMacefieldVG. Effects of 12 months continuous positive airway pressure on sympathetic activity related brainstem function and structure in obstructive sleep apnea. Frontiers in Neuroscience 2016 10 90. (https://doi.org/10.3389/fnins.2016.00090)
HarschIKonturekPKoebnickCKuehnleinPFuchsFSchahinSPWiestGHahnELohmannTFickerJ. Leptin and ghrelin levels in patients with obstructive sleep apnoea: effect of CPAP treatment. European Respiratory Journal 2003 22 251–257. (https://doi.org/10.1183/09031936.03.00010103)
CooperBGWhiteJEAshworthLAAlbertiKGGibsonGJ. Hormonal and metabolic profiles in subjects with obstructive sleep apnea syndrome and the acute effects of nasal continuous positive airway pressure (CPAP) treatment. Sleep 1995 18 172–179. (https://doi.org/10.1093/sleep/18.3.172)
AminAAliAAltafQAPiyaMKBarnettAHRaymondNTTahraniAA. Prevalence and associations of obstructive sleep apnea in south asians and white europeans with type 2 diabetes: a cross-sectional study. Journal of Clinical Sleep Medicine 2017 13 583–589. (https://doi.org/10.5664/jcsm.6548)
AnothaisintaweeTReutrakulSVan CauterEThakkinstianA. Sleep disturbances compared to traditional risk factors for diabetes development: systematic review and meta-analysis. Sleep Medicine Reviews 2016 30 11–24. (https://doi.org/10.1016/j.smrv.2015.10.002)
SalordNFortunaAMMonasterioCGasaMPerezABonsignoreMRVilarrasaNMontserratJMMayosM. A randomized controlled trial of continuous positive airway pressure on glucose tolerance in obese patients with obstructive sleep apnea. Sleep 2016 39 35–41. (https://doi.org/10.5665/sleep.5312)
IftikharIHHoyosCMPhillipsCLMagalangUJ. Meta-analyses of the association of sleep apnea with insulin resistance, and the effects of CPAP on HOMA-IR, adiponectin, and visceral adipose fat. Journal of Clinical Sleep Medicine 2015 11 475–485. (https://doi.org/10.5664/jcsm.4610)
HermansMPAhnSAMahadebYPRousseauMF. Sleep apnoea syndrome and 10‐year cardiovascular risk in females with type 2 diabetes: relationship with insulin secretion and insulin resistance. Diabetes/Metabolism Research and Reviews 2013 29 227–234. (https://doi.org/10.1002/dmrr.2387)
PamidiSWroblewskiKStepienMSharif-SidiKKilkusJWhitmoreHTasaliE. Eight hours of nightly continuous positive airway pressure treatment of obstructive sleep apnea improves glucose metabolism in patients with prediabetes. A randomized controlled trial. American Journal of Respiratory and Critical Care Medicine 2015 192 96–105. (https://doi.org/10.1164/rccm.201408-1564OC)
WelshNMargulisBBorgLWiklundHJSaldeenJFlodstrÃM. Differences in the expression of heat-shock proteins and antioxidant enzymes between human and rodent pancreatic islets: implications for the pathogenesis of insulin-dependent diabetes mellitus. Molecular Medicine 1995 1 806. (https://doi.org/10.1007/BF03401895)
MurphyAMThomasACrinionSJKentBDTambuwalaMMFabreAPepinJ-LRocheHMArnaudCRyanS. Intermittent hypoxia in obstructive sleep apnoea mediates insulin resistance through adipose tissue inflammation. European Respiratory Journal 2017 49 1601731. (https://doi.org/10.1183/13993003.01731-2016)
SacramentoJFRibeiroMJRodriguesTGuarinoMPDiogoLNSeicaRMonteiroECMatafomePCondeSV. Insulin resistance is associated with tissue-specific regulation of HIF-1alpha and HIF-2alpha during mild chronic intermittent hypoxia. Respiratory Physiology and Neurobiology 2016 228 30–38. (https://doi.org/10.1016/j.resp.2016.03.007)
GianottiLPivettiSLanfrancoFTassoneFNavoneFVittoriERossettoRGaunaCDestefanisSGrottoliS et al. Concomitant impairment of growth hormone secretion and peripheral sensitivity in obese patients with obstructive sleep apnea syndrome. Journal of Clinical Endocrinology and Metabolism 2002 87 5052–5057. (https://doi.org/10.1210/jc.2001-011441)
HobzovaMSalzmanRStejskalDZapletalovaJKolekV. Serum adiponectin level in obstructive sleep apnea: relation of adiponectin to obesity and long-term continuous positive airway pressure therapy. Advances in Medical Sciences 2016 61 130–134. (https://doi.org/10.1016/j.advms.2015.10.004)
NgSSLiuEKMaRCChanTOToKWChanKKNgaiJYipWHKoFWWongCK et al. Effects of CPAP therapy on visceral fat thickness, carotid intima-media thickness and adipokines in patients with obstructive sleep apnoea. Respirology 2017 22 786–792. (https://doi.org/10.1111/resp.12963)
KritikouIBastaMVgontzasANPejovicSLiaoDTsaoussoglouMBixlerEOStefanakisZChrousosGP. Sleep apnoea, sleepiness, inflammation and insulin resistance in middle-aged males and females. European Respiratory Journal 2014 43 145–155. (https://doi.org/10.1183/09031936.00126712)
HoyosCMKillickRYeeBJPhillipsCLGrunsteinRRLiuPY. Cardiometabolic changes after continuous positive airway pressure for obstructive sleep apnoea: a randomised sham-controlled study. Thorax 2012 67 1081–1089. (https://doi.org/10.1136/thoraxjnl-2011-201420)
MishraPNugentCAfendyABaiCBhatiaPAfendyMFangYElarinyHGoodmanZYounossiZM. Apnoeic–hypopnoeic episodes during obstructive sleep apnoea are associated with histological nonalcoholic steatohepatitis. Liver International 2008 28 1080–1086. (https://doi.org/10.1111/j.1478-3231.2008.01822.x)
HuangTLinBMStampferMJTworogerSSHuFBRedlineS. A population-based study of the bidirectional association between obstructive sleep apnea and type 2 diabetes in three prospective U.S. cohorts. Diabetes Care 2018 41 2111–2119. (https://doi.org/10.2337/dc18-0675)
TantucciCSciontiLBottiniPDottoriniMLPuxedduECasucciGSorbiniCA. Influence of autonomic neuropathy of different severities on the hypercapnic drive to breathing in diabetic patients. Chest 1997 112 145–153. (https://doi.org/10.1378/chest.112.1.145)
KleinOLAviles-SantaLCaiJCollardHRKanayaAMKaplanRCKinneyGLMendesESmithLTalaveraG. Hispanics/Latinos with type 2 diabetes have functional and symptomatic pulmonary impairment mirroring kidney microangiopathy: findings from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). Diabetes Care 2016 39 2051–2057. (https://doi.org/10.2337/dc16-1170)
OdaEKawaiR. A cross-sectional relationship between vital capacity and metabolic syndrome and between vital capacity and diabetes in a sample Japanese population. Environmental Health and Preventive Medicine 2009 14 284. (https://doi.org/10.1007/s12199-009-0098-5)
KimH-KKimC-HJungYBaeSChoeJParkJLeeK-U. Association of restrictive ventilatory dysfunction with insulin resistance and type 2 diabetes in Koreans. Experimental and Clinical Endocrinology and Diabetes 2011 119 47–52. (https://doi.org/10.1055/s-0030-1268467)
YehH-CPunjabiNMWangN-YPankowJSDuncanBBCoxCESelvinEBrancatiFL. Cross-sectional and prospective study of lung function in adults with type 2 diabetes: the Atherosclerosis Risk in Communities (ARIC) study. Diabetes Care 2008 31 741–746. (https://doi.org/10.2337/dbib7-1464)
NewmanABFosterGGivelberRNietoFJRedlineSYoungT. Progression and regression of sleep-disordered breathing with changes in weight: the Sleep Heart Health Study. Archives of Internal Medicine 2005 165 2408–2413. (https://doi.org/10.1001/archinte.165.20.2408)
PillaiAWarrenGGunathilakeWIdrisI. Effects of sleep apnea severity on glycemic control in patients with type 2 diabetes prior to continuous positive airway pressure treatment. Diabetes Technology and Therapeutics 2011 13 945–949. (https://doi.org/10.1089/dia.2011.0005)
GrimaldiDBeccutiGToumaCVan CauterEMokhlesiB. Association of obstructive sleep apnea in rapid eye movement sleep with reduced glycemic control in type 2 diabetes: therapeutic implications. Diabetes Care 2014 37 355–363. (https://doi.org/10.2337/dc13-0933)
GuoLXZhaoXPanQSunXLiHWangXXZhangLNWangY. Effect of continuous positive airway pressure therapy on glycemic excursions and insulin sensitivity in patients with obstructive sleep apnea-hypopnea syndrome and type 2 diabetes. Chinese Medical Journal 2015 128 2301–2306. (https://doi.org/10.4103/0366-6999.163382)
ShawJEPunjabiNMNaughtonMTWillesLBergenstalRMCistulliPAFulcherGRRichardsGNZimmetPZ. The effect of treatment of obstructive sleep apnea on glycemic control in type 2 diabetes. American Journal of Respiratory and Critical Care Medicine 2016 194 486–492. (https://doi.org/10.1164/rccm.201511-2260OC)
Martinez-CeronEBarquielBBezosAMCasitasRGaleraRGarcia-BenitoCHernanzAAlonso-FernandezAGarcia-RioF. Effect of continuous positive airway pressure on glycemic control in patients with obstructive sleep apnea and type 2 diabetes. A randomized clinical trial. American Journal of Respiratory and Critical Care Medicine 2016 194 476–485. (https://doi.org/10.1164/rccm.201510-1942OC)
AltafQAAliAPiyaMKRaymondNTTahraniAA. The relationship between obstructive sleep apnea and intra-epidermal nerve fiber density, PARP activation and foot ulceration in patients with type 2 diabetes. Journal of Diabetes and its Complications 2016 30 1315–1320. (https://doi.org/10.1016/j.jdiacomp.2016.05.025)
TahraniAAAliARaymondNTBegumSDubbKMughalSJoseBPiyaMKBarnettAHStevensMJ. Obstructive sleep apnea and diabetic neuropathy: a novel association in patients with type 2 diabetes. American Journal of Respiratory and Critical Care Medicine 2012 186 434–441. (https://doi.org/10.1164/rccm.201112-2135OC)
Jullian-DesayesIJoyeux-FaureMTamisierRLaunoisSBorelA-LLevyPPepinJ-L. Impact of obstructive sleep apnea treatment by continuous positive airway pressure on cardiometabolic biomarkers: a systematic review from sham CPAP randomized controlled trials. Sleep Medicine Reviews 2015 21 23–38. (https://doi.org/10.1016/j.smrv.2014.07.004)
MyhillPCDavisWAPetersKEChubbSPHillmanDDavisTM. Effect of continuous positive airway pressure therapy on cardiovascular risk factors in patients with type 2 diabetes and obstructive sleep apnea. Journal of Clinical Endocrinology and Metabolism 2012 97 4212–4218. (https://doi.org/10.1210/jc.2012-2107)
RiceTBFosterGDSandersMHUnruhMReboussinDKunaSTMillmanRZammitGWingRRWaddenTA et al. The relationship between obstructive sleep apnea and self-reported stroke or coronary heart disease in overweight and obese adults with type 2 diabetes mellitus. Sleep 2012 35 1293–1298. (https://doi.org/10.5665/sleep.2090)
SeiceanSStrohlKPSeiceanAGibbyCMarwickTH. Sleep disordered breathing as a risk of cardiac events in subjects with diabetes mellitus and normal exercise echocardiographic findings. American Journal of Cardiology 2013 111 1214–1220. (https://doi.org/10.1016/j.amjcard.2012.12.053)
AltafQADodsonPAliARaymondNTWhartonHFellowsHHampshire-BancroftRShahMShepherdEMiahJ et al. Obstructive sleep apnea and retinopathy in patients with type 2 diabetes. A longitudinal study. American Journal of Respiratory and Critical Care Medicine 2017 196 892–900. (https://doi.org/10.1164/rccm.201701-0175OC)
MasonRHKiireCAGrovesDCLipinskiHJJaycockAWinterBCSmithLBoltonARahmanNMSwaminathanR et al. Visual improvement following continuous positive airway pressure therapy in diabetic subjects with clinically significant macular oedema and obstructive sleep apnoea: proof of principle study. Respiration 2012 84 275–282. (https://doi.org/10.1159/000334090)
ReutrakulSThakkinstianAAnothaisintaweeTChontongSBorelA-LPerfectMMJanovskyCCPSKesslerRSchultesBHarschIA. Sleep characteristics in type 1 diabetes and associations with glycemic control: systematic review and meta-analysis. Sleep Medicine 2016 23 26–45. (https://doi.org/10.1016/j.sleep.2016.03.019)
BanghoejAMNerildHHKristensenPLPedersen-BjergaardUFleischerJJensenAEKLaubMThorsteinssonBTarnowL. Obstructive sleep apnoea is frequent in patients with type 1 diabetes. Journal of Diabetes and its Complications 2017 31 156–161. (https://doi.org/10.1016/j.jdiacomp.2016.10.006)
StubbeBSchipfSSchäperCFelixSBStevelingANauckMVölzkeHWallaschofskiHFriedrichNEwertR et al. The influence of type 1 diabetes mellitus on pulmonary function and exercise capacity – results from the Study of Health in Pomerania (SHIP). Experimental and Clinical Endocrinology and Diabetes 2017 125 64–69. (https://doi.org/10.1055/s-0042-112219)
NietoFJYoungTBLindBKShaharESametJMRedlineSD'agostinoRBNewmanABLebowitzMDPickeringTG. Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. JAMA 2000 283 1829–1836. (https://doi.org/10.1001/jama.283.14.1829)
ZhangWZhangJWuKChenXWangYZhouLWangHChenS. Effect of aldosterone antagonists on obstructive sleep apnea in patients with resistant hypertension: a systematic review and meta-analysis. Journal of Human Hypertension 2017 31 855. (https://doi.org/10.1038/jhh.2017.70)
LloberesPSampolGEspinelESegarraARamonM-ARomeroOFerrerRMartínez-GarciaM-ATovarJ-L. A randomized controlled study of CPAP effect on plasma aldosterone concentration in patients with resistant hypertension and obstructive sleep apnea. Journal of Hypertension 2014 32 1650–1657. (https://doi.org/10.1097/HJH.0000000000000238)
de SouzaFMuxfeldtESMargalloVCortezAFCavalcantiAHSallesGF. Effects of continuous positive airway pressure treatment on aldosterone excretion in patients with obstructive sleep apnoea and resistant hypertension: a randomized controlled trial. Journal of Hypertension 2017 35 837–844. (https://doi.org/10.1097/HJH.0000000000001254)
GaddamKPimentaEThomasSJCofieldSSOparilSHardingSMCalhounDA. Spironolactone reduces severity of obstructive sleep apnoea in patients with resistant hypertension: a preliminary report. Journal of Human Hypertension 2010 24 532. (https://doi.org/10.1038/jhh.2009.96)
CopinschiGChalletE. Chapter 9 – Endocrine rhythms, the sleep-wake cycle, and biological clocks A2 – Jameson, J. Larry. In Endocrinology: Adult and Pediatric7th ed. pp 147.e149–173.e149. Eds LJDGrootDMdKretserLCGiudiceABGrossmanSMelmedJTPotts & GCWeir. Philadelphia, USA : W.B. Saunders2016.
GrunsteinRRHandelsmanDJLawrenceSJBlackwellCCatersonIDSullivanCE. Neuroendocrine dysfunction in sleep apnea: reversal by continuous positive airways pressure therapy. Journal of Clinical Endocrinology and Metabolism 1989 68 352–358. (https://doi.org/10.1210/jcem-68-2-352)
CarneiroGTogeiroSMHayashiLFRibeiro-FilhoFFRibeiroABTufikSZanellaMT. Effect of continuous positive airway pressure therapy on hypothalamic-pituitary-adrenal axis function and 24-h blood pressure profile in obese men with obstructive sleep apnea syndrome. American Journal of Physiology: Endocrinology and Metabolism 2008 295 E380–E384. (https://doi.org/10.1152/ajpendo.00780.2007)
KaracaZIsmailogullariSKorkmazSCakirIAksuMBaydemirRTanriverdiFBayramF. Obstructive sleep apnoea syndrome is associated with relative hypocortisolemia and decreased hypothalamo–pituitary–adrenal axis response to 1 and 250 μg ACTH and glucagon stimulation tests. Sleep Medicine 2013 14 160–164. (https://doi.org/10.1016/j.sleep.2012.10.013)
DadounFDarmonPAchardVBoullu-CioccaSPhilip-JoetFAlessiM-CReyMGrinoMDutourA. Effect of sleep apnea syndrome on the circadian profile of cortisol in obese men. American Journal of Physiology: Endocrinology and Metabolism 2007 293 E466–E474. (https://doi.org/10.1152/ajpendo.00126.2007)
VgontzasAPejovicSZoumakisELinH-MBentleyCBixlerESarrigiannidisABastaMChrousosG. Hypothalamic-pituitary-adrenal axis activity in obese men with and without sleep apnea: effects of continuous positive airway pressure therapy. Journal of Clinical Endocrinology and Metabolism 2007 92 4199–4207. (https://doi.org/10.1210/jc.2007-0774)
LanfrancoFGianottiLPivettiSNavoneFRossettoRTassoneFGaiVGhigoEMaccarioM. Obese patients with obstructive sleep apnoea syndrome show a peculiar alteration of the corticotroph but not of the thyrotroph and lactotroph function. Clinical Endocrinology 2004 60 41–48. (https://doi.org/10.1111/j.1365-2265.2004.01938.x)
ChinKShimizuKNakamuraTNaraiNMasuzakiHOgawaYMishimaMNakamuraTNakaoKOhiM. Changes in intra-abdominal visceral fat and serum leptin levels in patients with obstructive sleep apnea syndrome following nasal continuous positive airway pressure therapy. Circulation 1999 100 706–712. (https://doi.org/10.1161/01.CIR.100.7.706)
MestonNDaviesRMullinsRJenkinsonCWassJStradlingJ. Endocrine effects of nasal continuous positive airway pressure in male patients with obstructive sleep apnoea. Journal of Internal Medicine 2003 254 447–454. (https://doi.org/10.1046/j.1365-2796.2003.01212.x)
SchmollerAEberhardtFJauch-CharaKSchweigerUZabelPPetersASchultesBOltmannsKM. Continuous positive airway pressure therapy decreases evening cortisol concentrations in patients with severe obstructive sleep apnea. Metabolism-Clinical and Experimental 2009 58 848–853. (https://doi.org/10.1016/j.metabol.2009.02.014)
HenleyDERussellGMDouthwaiteJAWoodSABuchananFGibsonRWoltersdorfWWCatterallJRLightmanSL. Hypothalamic-pituitary-adrenal axis activation in obstructive sleep apnea: the effect of continuous positive airway pressure therapy. Journal of Clinical Endocrinology and Metabolism 2009 94 4234–4242. (https://doi.org/10.1210/jc.2009-1174)
MokhlesiBGrimaldiDBeccutiGVan CauterE. Effect of one week of CPAP treatment of obstructive sleep apnoea on 24‐hour profiles of glucose, insulin and counter‐regulatory hormones in type 2 diabetes. Diabetes Obesity and Metabolism 2017 19 452–456. (https://doi.org/10.1111/dom.12823)
VgontzasANZoumakisMBixlerEOLinH-MProloPVela-BuenoAKalesAChrousosGP. Impaired nighttime sleep in healthy old versus young adults is associated with elevated plasma interleukin-6 and cortisol levels: physiologic and therapeutic implications. Journal of Clinical Endocrinology and Metabolism 2003 88 2087–2095. (https://doi.org/10.1210/jc.2002-021176)
PruessnerJCWolfOTHellhammerDHBuske-KirschbaumAVon AuerKJobstSKaspersFKirschbaumC. Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity. Life Sciences 1997 61 2539–2549. (https://doi.org/10.1016/S0024-3205(97)01008-4)
CaufriezAMoreno-ReyesRLeproultRVertongenFVan CauterECopinschiG. Immediate effects of an 8-h advance shift of the rest-activity cycle on 24-h profiles of cortisol. American Journal of Physiology: Endocrinology and Metabolism 2002 282 E1147–E1153. (https://doi.org/10.1152/ajpendo.00525.2001)
GokosmanoğluFGüzelAKanEKAtmacaH. Increased prevalence of obstructive sleep apnea in patients with Cushing’s syndrome compared with weight and age-matched controls. European Journal of Endocrinology 2017 176 267–272. (https://doi.org/10.1530/EJE-16-0815)
WangL-UWangT-YBaiY-MHsuJ-WHuangK-LSuT-PLiC-TLinW-CChenT-JChenM-H. Risk of obstructive sleep apnea among patients with Cushing's syndrome: a nationwide longitudinal study. Sleep Medicine 2017 36 44–47. (https://doi.org/10.1016/j.sleep.2017.04.016)
BergerGHardakEShahamBAvitanEYiglaM. Preliminary prospective explanatory observation on the impact of 3-month steroid therapy on the objective measures of sleep-disordered breathing. Sleep and Breathing 2012 16 549–553. (https://doi.org/10.1007/s11325-011-0541-x)
SainiJKriegerJBrandenbergerGWittersheimGSimonCFolleniusM. Continuous positive airway pressure treatment. Effects on growth hormone, insulin and glucose profiles in obstructive sleep apnea patients. Hormone and Metabolic Research 1993 25 375–381. (https://doi.org/10.1055/s-2007-1002123)
HoyosCMKillickRKeenanDMBaxterRCVeldhuisJDLiuPY. Continuous positive airway pressure increases pulsatile growth hormone secretion and circulating insulin-like growth factor-1 in a time-dependent manner in men with obstructive sleep apnea: a randomized sham-controlled study. Sleep 2014 37 733–741. (https://doi.org/10.5665/sleep.3580)
CastellaniCFranciaGDalle CarbonareLFerrariMVivaECeriniRZaccarellaATrevisiolLDaviMV. Morphological study of upper airways and long-term follow-up of obstructive sleep apnea syndrome in acromegalic patients. Endocrine 2016 51 308–316. (https://doi.org/10.1007/s12020-015-0659-x)
De MenisEGiustinaAColaoADegli UbertiEGhigoEMinutoFBogazziFDrigoRCattaneoAAimarettiG. Assessment of the awareness and management of sleep apnea syndrome in acromegaly. The COM.E.TA (Comorbidities Evaluation and Treatment in Acromegaly) Italian Study Group. Journal of Endocrinological Investigation 2011 34 60–64. (https://doi.org/10.1007/BF03346696)
DaviMVDalle CarbonareLGiustinaAFerrariMFrigoALo CascioVFranciaG. Sleep apnoea syndrome is highly prevalent in acromegaly and only partially reversible after biochemical control of the disease. European Journal of Endocrinology 2008 159 533–540. (https://doi.org/10.1530/EJE-08-0442)
KamenickyPViengchareunSBlanchardAMeduriGZizzariPImbert-TeboulMDoucetAChansonPLombesM. Epithelial sodium channel is a key mediator of growth hormone-induced sodium retention in acromegaly. Endocrinology 2008 149 3294–3305. (https://doi.org/10.1210/en.2008-0143)
KamenickyPBlanchardAFrankMSalenaveSLetierceAAziziMLombèsMChansonP. Body fluid expansion in acromegaly is related to enhanced epithelial sodium channel (ENaC) activity. Journal of Clinical Endocrinology and Metabolism 2011 96 2127–2135. (https://doi.org/10.1210/jc.2011-0078)
HerrmannBWessendorfTAjajWKahlkeSTeschlerHMannK. Effects of octreotide on sleep apnoea and tongue volume (magnetic resonance imaging) in patients with acromegaly. European Journal of Endocrinology 2004 151 309–315. (https://doi.org/10.1530/eje.0.1510309)
van HauteFRTaboadaGFCorrêaLLLimaGAFontesRRielloAPDominiciMGadelhaMR. Prevalence of sleep apnea and metabolic abnormalities in patients with acromegaly and analysis of cephalometric parameters by magnetic resonance imaging. European Journal of Endocrinology 2008 158 459–465. (https://doi.org/10.1530/EJE-07-0753)
GasperiMMartinoEManettiLArosioMPorrettiSFagliaGMariottiSColaoALombardiGBaldelliR. Prevalence of thyroid diseases in patients with acromegaly: results of an Italian multi-center study. Journal of Endocrinological Investigation 2002 25 240–245. (https://doi.org/10.1007/BF03343997)
TolisGAngelopoulosNGKatoundaERombopoulosGKaltzidouVKaltsasDProtonotariouALytrasA. Medical treatment of acromegaly: comorbidities and their reversibility by somatostatin analogs. Neuroendocrinology 2006 83 249–257. (https://doi.org/10.1159/000095535)
ChansonPTimsitJBenoitOAugendreBMoulonguetMGuillausseauP-JWarnetALubetzkiJ. Rapid improvement in sleep apnoea of acromegaly after short-term treatment with somatostatin analogue SMS 201–995. Lancet 1986 327 1270–1271. (https://doi.org/10.1016/S0140-6736(86)91405-4)
BergCWessendorfTMortschFForstingMTeschlerHWeischerTMannKSallerBHerrmannB. Influence of disease control with pegvisomant on sleep apnoea and tongue volume in patients with active acromegaly. European Journal of Endocrinology 2009 161 829–835. (https://doi.org/10.1530/EJE-09-0694)
KuhnEMaioneLBouchachiARozièreMSalenaveSBrailly-TabardSYoungJKamenickyPAssayagPChansonP. Long-term effects of pegvisomant on comorbidities in patients with acromegaly: a retrospective single-center study. European Journal of Endocrinology 2015 173 693–702. (https://doi.org/10.1530/EJE-15-0500)
ChemlaDAttalPMaioneLVeyerA-SMroueGBaudDSalenaveSKamenickyPBobinSChansonP. Impact of successful treatment of acromegaly on overnight heart rate variability and sleep apnea. Journal of Clinical Endocrinology and Metabolism 2014 99 2925–2931. (https://doi.org/10.1210/jc.2013-4288)
PekerYSvenssonJHednerJGroteLJohannssonG. Sleep apnoea and quality of life in growth hormone (GH)‐deficient adults before and after 6 months of GH replacement therapy. Clinical Endocrinology 2006 65 98–105. (https://doi.org/10.1111/j.1365-2265.2006.02555.x)
Bergan-RollerHESheridanMA. The growth hormone signaling system: insights into coordinating the anabolic and catabolic actions of growth hormone. General and Comparative Endocrinology 2018 258 119–133. (https://doi.org/10.1016/j.ygcen.2017.07.028)
NolteWRädischCRodenbeckAWiltfangJHüfnerM. Polysomnographic findings in five adult patients with pituitary insufficiency before and after cessation of human growth hormone replacement therapy. Clinical Endocrinology 2002 56 805–810. (https://doi.org/10.1046/j.1365-2265.2002.01531.x)
HartmanMLXuRCroweBJRobisonLLErfurthEMKleinbergDLZimmermannAGWoodmanseeWWCutlerGBJrChipmanJJ. Prospective safety surveillance of GH-deficient adults: comparison of GH-treated vs untreated patients. Journal of Clinical Endocrinology and Metabolism 2013 98 980–988. (https://doi.org/10.1210/jc.2012-2684)
KarimiMKoranyiJFrancoCPekerYEderDNAngelhedJ-ELönnLGroteLBengtssonB-ÅSvenssonJ. Increased neck soft tissue mass and worsening of obstructive sleep apnea after growth hormone treatment in men with abdominal obesity. Journal of Clinical Sleep Medicine 2010 6 256.
SalvatoniAVeronelliENosettiLBeriniJde SimoneSIughettiLBosioLChiumelloGGrugniGDelùG et al. Short-term effects of growth hormone treatment on the upper airways of non severely obese children with Prader-Willi syndrome. Journal of Endocrinological Investigation 2009 32 601–605. (https://doi.org/10.1007/BF03346516)