Obstructive sleep apnea (OSA) is a disease with metabolic and cardiovascular consequences and is associated with decreased serum concentrations of insulin-like growth factor-1 (IGF-1). The aim of this study was to investigate whether continuous positive airway pressure (CPAP) will increase serum IGF-1 concentration in patients with OSA.
Patients with moderate to severe OSA were recruited from a sleep clinic and serum IGF-1 was measured before initiation of CPAP and at follow-up after 4.8 ± 2.5 months. Patients adherent to CPAP treatment (usage ≥ 4 h/night) were compared with those considered to be nonadherent (usage < 4 h/night).
Complete data were obtained from 69 patients (86% male, age 56 ± 12 years, respiratory event index 43 ± 21 events/h, Epworth Sleepiness Scale score 12 ± 5). In those adherent to CPAP (n = 42), there was an increase in serum IGF-1 concentration with 21.1 (95% confidence interval [CI]: 13.1 to 29.2) μg/L compared to 4.7 (95% CI: −4.1 to 13.5) μg/L in the nonadherent group (n = 27) (P = .0083). In a linear multivariate model adjusting for sex, age, body mass index, respiratory event index, and mean oxygen saturation during the night recording, the change in serum IGF-1 concentration was significantly associated with adherence to CPAP treatment (adjusted β coefficient: 21.8, 95% CI: 10.2 to 33.4) and inversely associated with change in body mass index (adjusted β coefficient: −7.1, 95% CI: −11.3 to −3.0) and change in hemoglobin A1c (adjusted β coefficient: −1.8, 95% CI: −33 to −0.3).
CPAP usage ≥ 4 h/night is associated with increased serum IGF-1 concentration in male patients with OSA.
Palm A, Berne C, Igelström H, Åsenlöf P, Janson C, Lindberg E. The impact of continuous positive airway pressure on circulating IGF-1 in patients with obstructive sleep apnea. J Clin Sleep Med. 2018;14(3):385–391.
Current Knowledge/Study Rationale: Obstructive sleep apnea (OSA) is associated with decreased serum concentrations of insulin-like growth factor-1 (IGF-1) and some studies have reported an increase in concentration after treatment with continuous positive airway pressure (CPAP). The adherence to CPAP has only rarely been taken into account, and the aim of our study was to compare changes in serum IGF-1 concentration after initiation of CPAP treatment among those patients adherent to CPAP treatment and those who were not.
Study Impact: CPAP usage ≥ 4 h/night is associated with increased serum IGF-1 concentration in patients with OSA. Low serum IGF-1 concentration might not only be a surrogate marker for OSA but also a mediator between OSA and metabolic and cardiovascular disease.
Obstructive sleep apnea (OSA) is characterized by recurrent episodes of partial or complete airway obstructions during sleep, resulting in repetitive apneas and hypopneas and episodes of hypoxia as a result.1 OSA is associated with increased cardiovascular morbidity and mortality.2–4 These consequences are thought to be mediated via several mechanisms: direct effects of hypoxia, increased sympathetic activation, endothelial dysfunction, impaired glucose and triglyceride metabolism, and increased systemic inflammation.5 Treatment with continuous positive airway pressure (CPAP) reduces airway obstructions and prevents episodes of hypoxia and as a consequence, metabolic effects and cardiovascular diseases are reduced.6–9
Insulin-like growth factor-1 (IGF-1) is a polypeptide mainly of hepatic origin and the secretion is dependent on growth hormone production from the pituitary gland and portal insulin exposure.10 IGF-1 is an important mediator of the effects of growth hormones and has also insulin-like effects. Serum IGF-1 concentration is low in obese subjects11 and in patients with OSA12 and is furthermore associated with metabolic syndrome, insulin resistance,13,14 and cardiovascular disease.15,16
An increase in IGF-1 after CPAP treatment has been reported in some studies,17–23 whereas others have failed to identify such a relationship.24 However, patients' adherence to CPAP has only been taken into account in a minority of the previous studies.20 The aim of our study was to compare changes in serum IGF-1 concentration after initiation of CPAP treatment between those who were adherent (usage ≥ 4 h/night) and those who were nonadherent (usage < 4 h/night) to CPAP treatment.
We identified 86 patients in whom moderate to severe OSA (respiratory event index [REI] ≥ 15 events/h) was diagnosed at the sleep clinic of Uppsala University Hospital. We attempted to measure patients' serum IGF-1 concentrations at the initiation of CPAP treatment and at follow-up after 4.8 ± 2.5 months. Data were collected from March 2010 to March 2012. Patients without information about IGF-1 at baseline and/or at follow-up (n = 15) and patients with OSA treated with mandibular device because of inability to use CPAP (n = 2) were excluded from subsequent analysis. The patients had been included consecutively into a study with the main purpose to evaluate the effects on OSA of a tailored behavioral medicine intervention program targeting physical activity and eating behavior.25 Inclusion criteria were body mass index (BMI) ≥ 25 kg/m2 and a sedentary lifestyle with a self-reported leisure time physical activity less than 30 min/d, 5 d/wk. Exclusion criteria were symptomatic heart disease despite medication and participation in a current weight reduction program. Approximately 1 month after initiation of CPAP treatment, a clinical checkup was performed. If the patients had problems with their CPAP treatment, CPAP settings or CPAP interface were adjusted in order to optimize treatment comfort and adherence.
Adherence to CPAP treatment at follow-up was assessed based on self-reported information on mean number of nights/ week and mean number of hours/night of CPAP usage. CPAP usage was categorized into the following groups: nonadherent (defined as an average CPAP usage < 4 h/night) and adherent (defined as an average CPAP usage ≥ 4 h/night).26
Information about CPAP usage was missing in 14 patients at the 6-month follow-up visit. Patients were considered adherent to CPAP if they were adherent at the first clinical visit to the CPAP nurse after 1 month and also at the long-term follow-up after the study completion at 28 ± 7 months according to the medical records.
Diagnosis of OSA
OSA was diagnosed by a single night sleep recording using the Embletta type 3 portable monitor (Embla, Reykjavik, Iceland) or Breas SC 20 (Breas Medical AB, Mölnlycke, Sweden). Total sleep time was estimated by use of the subject's sleep diary in conjunction with visual assessment of the overnight tracing. Apneas were defined as cessation of airflow in nasal pressure for at least 10 seconds with continuing abdominal and thoracic movements whereas hypopneas were defined as ≥ 50% reduction in baseline airflow for at least 10 seconds in combination with an oxygen desaturation of ≥ 3%. REI was calculated as the number of apneas and hypopneas per hour of sleep. Oxygen desaturation index (ODI3) was calculated as the number of desaturations of ≥ 3% per hour of sleep.1 Moderate to severe OSA was defined as an REI ≥ 15 events/h. Daytime sleepiness was assessed using the Epworth Sleepiness Scale (ESS).27 Scores range from zero to 24, and higher scores indicate greater severity of sleepiness.
Weight and height were measured and BMI was calculated as weight in kilograms divided by height in meters squared (kg/m2). The patients were asked about smoking habits. A structured interview was performed by the research nurse and the presence of former thromboembolic disease, diabetes type I and II, atrial fibrillation, hypertension, coronary disease, heart failure, lung disease, or stroke was recorded. The study was approved by the Ethics Committee of the Medical Faculty at Uppsala University (dnr 2009/004).
A fasting venous blood level was obtained for determination of total serum IGF-1 concentrations and hemoglobin A1c (HbA1c). IGF-1 was determined by using a commercial immunochemiluminescent assay, Liasion (DiaSorin S.p.A, Saluggia, Italy), which uses two monoclonal antibodies prepared against two different antigenic sites of IGF-1 molecule.28 HbA1c was analyzed immunologically on a Cobas 6000 instrument with Tina-quant hemoglobin A1c reagents (Roche Diagnostics, Mannheim, Germany).
Statistical analyses were performed using Stata version 12.1 (StataCorp LLC, College Station, Texas, United States). Differences between groups were compared with a χ2 test for categorical variables and with a t test for continuous variables. Univariate linear regression was used to examine association between baseline concentration and change in IGF-1 concentration with explanatory variables. A multivariate linear regression model adjusting for CPAP adherence and change in BMI and with independent variables added separately was used to further examine the association between change in IGF-1 concentration and explanatory variables. To further analyze the association between changes in IGF-1 and CPAP adherence, the calculations were also performed after stratifying the study population by sex, severity of sleepiness, age, and participation in the tailored behavioral medicine intervention program.
Complete data were obtained from 69 patients (86% male, mean age 56 ± 12 years [range 29–79], mean REI 43 ± 21 events/h [range 17–92], mean BMI 34.2 ± 5.0 kg/m2 [range 26– 48]) (Figure 1 and Table 1). There were no significant differences among patients who were adherent and nonadherent to CPAP in sex distribution, age, REI, BMI, IGF-1 concentration, HbA1c concentration, ESS, smoking habits, or comorbidities.
Flow chart of study population.
BMI = body mass index, CPAP = continuous positive airway pressure, IGF-1 = insulin-like growth factor-1, REI = respiratory event index.
In total, 42 patients (61%) were categorized as adherent and 27 patients (39%) as nonadherent to CPAP treatment. Two of the patients who were nonadherent to CPAP started to use CPAP immediately after the follow-up control and achieved at that time an adherence of ≥ 4 h/night. The overall CPAP adherence ratio in the entire group was 44 of 69 (64%) (Figure 1).
At baseline, there was a negative association between IGF-1 and BMI and a positive association between IGF-1 concentration and mean oxygen saturation during the sleep recording (Table 2). No association was found between IGF-1 concentration at baseline and age, REI, ODI3, HbA1c, or ESS.
Univariate linear regression model with IGF-1 at baseline as dependent variable.
Univariate linear regression model with IGF-1 at baseline as dependent variable.
The mean time between initiation of CPAP treatment and the follow-up measurement of IGF-1 was 4.8 ± 2.5 months. In the group adherent to CPAP the mean rise in IGF-1 was 21.1 μg/L (95% confidence interval [CI]: 13.1 to 29.2) at the follow-up compared to 4.7 μg/L (95% CI: −4.1 to 13.5) in the nonadherent group (P = .008) (Figure 2 and Table 3). At follow-up, the ESS score decreased more in those who were adherent compared to those who were nonadherent to CPAP. In the individuals who were nonadherent, there was a greater reduction of BMI and also in HbA1c compared to those who were adherent. A linear multivariate model showed an association between change in IGF-1 versus CPAP adherence and inverse associations between change in IGF-1 versus change in BMI and change in HbA1c (Table 4). No association was found between change in IGF-1 and sex, age, REI, ODI3, mean saturation during night recording, BMI or ESS at baseline, change in ESS, or participation in the tailored behavioral medicine intervention program.
IGF-1 at baseline and follow-up.
Outliers (defined by Tukey′s criteria) are marked by circles. CPAP = continuous positive airway pressure, IGF-1 = insulin-like growth factor-1.
IGF-1 at baseline and follow-up.
Characteristics at follow-up.
Linear regression model with dependent variable “Change in IGF-1.”
Linear regression model with dependent variable “Change in IGF-1.”
In males (n = 59), those adherent to CPAP (n = 35) had an increase in IGF-1 of 23.3 ± 27.2 μg/L compared to 3.5 ± 22.7 μg/L (P = .005) in those who were not adherent to CPAP (n = 24). In females (n = 10), those adherent to CPAP (n = 7) had an increase in IGF-1 of 10.6 ± 14.4 μg/L compared to 14.3 ± 19.1 μg/L (P = .74) in those who were not adherent to CPAP (n = 3). Among those adherent to CPAP and had an ESS score < 10 (n = 10, mean ESS 6.5 ± 2.5), the mean increase in IGF-1 was 26.6 ± 22.7 μg/L compared to 19.4 ± 26.8 μg/L (P = .4504) in those with an ESS score ≥ 10 (n = 32, mean value 14.5 ± 2.8).
Those adherent to CPAP with age older than 50 years (n = 26) had an increase in IGF-1 of 21.5 ± 18.9 μg/L compared to 20.5 ± 35.0 μg/L in those age 50 years or younger (n = 16, P = .90). Those adherent to CPAP with fewer than 3 months between initiation of CPAP treatment and follow-up (n = 5) had an increase in IGF-1 of 23.6 ± 36.9 μg/L compared to 20.8 ± 24.6 μg/L in those with more than 3 months between initiation of CPAP and follow-up (n = 37) (P = .824).
Excluding patients with diabetes (n = 15) did not significantly change any of the results (data not shown).
The main finding of this study is that adherence to CPAP (usage ≥ 4 h/night) was associated with elevated serum IGF-1 concentration, at least in males. This supports the result from a previous study of 78 patients (86% male) in which those adherent to CPAP (usage ≥ 4 h/night) after 7.8 ± 1.3 months had an increase in IGF-1 in contrast to those who were nonadherent.20 In females in both studies, no change in serum IGF-1 concentration was observed. However, these two studies with 9 and 11 females, respectively, were not powered to come to conclusions about sex differences.
In the current study, there was no difference in increase of IGF-1 if the patients adherent to CPAP had been under treatment for a shorter or longer period than 3 months. It is not clear how soon after initiation of CPAP treatment a rise in IGF-1 occurs. Data from previous studies have diverged. In one study, there was a rise in IGF-1 after 12 but not after 6 weeks of CPAP.23 Another study showed elevated serum IGF-1 after 1 month of CPAP treatment but further analysis revealed an equal rise of IGF-1 in the sham CPAP group.
Our results underscore that adequate adherence to treatment is crucial for a favorable outcome of CPAP treatment. CPAP reduces airway obstructions and prevents episodes of hypoxia and as a consequence, adverse metabolic and cardiovascular effects of OSA are reduced. Randomized controlled trials have shown the effect of CPAP on surrogate endpoints such as hypertension,6,7 endothelial function,8 and insulin resistance.9 Observational studies have shown the effect of CPAP on cardiovascular diseases and on cardiovascular death.4,29 Randomized controlled trials, however, have not shown the effect of CPAP on cardiovascular outcome on a group level,30–33 but when performing subgroup analysis, those who used CPAP ≥ 4 h/night displayed significantly lower incidence of cardiovascular outcomes.31,32
The adherence rate in this study (64%) is in line with the adherence rates in previous studies in which adherence rates varied between 46% to 83% when adherence is defined as ≥ 4 h/night usage.26
We found no association between change in IGF-1 and ESS at baseline or change in ESS. However, the patients were rather homogenous in regard to ESS score, with an overwhelming majority with ESS scores higher than 10 (72%) and ESS at baseline of 12.3 ± 4.5. The absence of association between IGF-1 and ESS could possibly be explained by the limited number of patients with low ESS scores. In a previous study of 35 patients with OSA with more pronounced diversity in daytime sleepiness, only those with sleepiness (mean ESS score 16) had a rise in IGF-1 after initiation of CPAP, whereas no effect was seen in controls with similar OSA severity but a mean ESS score 4.19
In accordance with previous studies there was an association between decreased IGF-1 and increasing BMI and sleep-disordered breathing.13,17 However, those who were nonadherent to CPAP had a decline in BMI as well as in HbA1c concentration, whereas this finding was not present in the adherent group. This result is in line with a recent meta-analysis where 3,181 subjects from 25 studies were pooled. Those with OSA treated with CPAP showed a significant increase in weight and BMI compared to controls without active treatment.34 An explanation can be that patients who fail to adhere to CPAP treatment might become more motivated to change their eating behavior or physical activity level. Those who succeeded to use their CPAP and experience clinical improvement may have been less motivated to change their lifestyle. Our results were unchanged even after exclusion of those with diabetes. No analysis was performed only on the patients with diabetes (n = 15) due to power limitations.
Secretion of IGF-1 is controlled by growth hormones predominantly during slow wave sleep.35 When OSA is treated successfully with CPAP, the sleep quality improves and the amount of slow wave sleep increases, which in turn might explain the increase in IGF-1. Like OSA, growth hormone deficiency is associated with an impaired glucose metabolism, endothelial and vasodilatory function, and systemic inflammation.36 Decreased IGF-1 is associated with impaired glucose metabolism13,14 and with cardiovascular diseases.15,16 This suggests that decreased IGF-1 in patients with OSA not only is a consequence of OSA but also forms an important pathway of the increased cardiovascular risk.
Strength and Weaknesses
The strengths of this study were the sample size and that the study groups were stratified by adherence to CPAP. When interpreting the results of this study, some limitations must be taken into account. This study was not randomized. It was not population based with a subsequent risk of selection bias. Information about CPAP adherence was assessed from self-reported information with a risk of overestimation of CPAP usage. Thus, patients who were nonadherent to CPAP may have been misclassified as adherent, and that could cause an underestimation of the difference in change in IGF-1 between the groups. The results should also be interpreted with caution because of the low number of female participants. Power analyses revealed that at least 10 females in each group would have been needed to discard the hypothesis of a similar difference between adherent and nonadherent CPAP users in females as in males.
CPAP usage ≥ 4 h/night is associated with increased serum IGF-1 concentration, at least in male patients with OSA. Those with the greatest improvement of IGF-1 after initiation of CPAP may be those who would most benefit from CPAP from a metabolic and a cardioprotective point of view. Low serum IGF-1 concentration might not only be a surrogate marker for OSA but also a mediator between OSA and metabolic and cardiovascular disease. It is suggested that future longitudinal studies be designed to evaluate the effect of CPAP on cardiovascular morbidity and mortality and should include analysis of IGF-1 to clarify this issue. It is also important that future studies within this field ensure a sufficient number of female patients as well as a wide distribution of daytime sleepiness severity.
All authors have seen and approved the manuscript. The authors report no conflicts of interest. This study was supported by grants from the Swedish Research Council, the Swedish Heart-Lung Foundation and the Centre for Research and Development, Uppsala University/Region Gävleborg, Sweden. Part of this work has been presented as a poster at the Annual Congress for Swedish Respiratory Society, Örebro, Sweden, April 2016 and as a poster at the Annual Congress for European Respiratory Society, London, UK, September 2016. Work for this study was performed at the Department of Medical Sciences, Respiratory, Allergy and Sleep Research, Uppsala University, Sweden.