ADVERTISEMENT

Issue Navigator

Volume 11 No. 04
Earn CME
Accepted Papers





Review Articles

Meta-analyses of the Association of Sleep Apnea with Insulin Resistance, and the Effects of CPAP on HOMA-IR, Adiponectin, and Visceral Adipose Fat

Imran H. Iftikhar, MD1; Camilla M. Hoyos, PhD2; Craig L. Phillips, PhD2,4; Ulysses J. Magalang, MD3
1Division of Pulmonary, Critical Care, and Sleep Medicine, University of South Carolina School of Medicine, Columbia, SC; 2Woolcock Institute of Medical Research, Central Clinical School, University of Sydney, Australia; 3Divison of Pulmonary, Allergy, Critical Care, and Sleep Medicine, Ohio State University Wexner Medical Center, Columbus, OH; 4Department of Respiratory and Sleep Medicine, Royal North Shore Hospital, Sydney, Australia

ABSTRACT

Objective:

We sought to conduct an updated meta-analysis of randomized controlled trials (RCTs) on the effect of continuous positive airway pressure (CPAP) on insulin resistance, as measured by homeostasis model assessment of insulin resistance (HOMA-IR), visceral abdominal fat (VAF), and adiponectin. Additionally, we performed a separate meta-analysis and meta-regression of studies on the association of insulin resistance and obstructive sleep apnea (OSA).

Methods:

All included studies were searched from PubMed (from conception to March 15, 2014). Data were pooled across all included RCTs as the mean difference in HOMA-IR and VAF, and as the standardized mean difference in the case of adiponectin analysis. From the included case-control studies, data on the difference of HOMA-IR between cases and controls were pooled across all studies, as the standardized mean difference (SMD).

Results:

There was a significant difference in HOMA-IR (−0.43 [95% CIs: −0.75 to −0.11], p = 0.008) between CPAP treated and non CPAP treated participants. However, there was no significant difference in VAF or adiponectin; (−47.93 [95% CI: −112.58 to 16.72], p = 0.14) and (−0.06 [95% CI: −0.28 to 0.15], p = 0.56), respectively. Meta-analysis of 16 case-control studies showed a pooled SMD in HOMA-IR of 0.51 (95% CI: 0.28 to 0.75), p ≤ 0.001, between cases and controls.

Conclusions:

The results of our meta-analyses show that CPAP has a favorable effect on insulin resistance. This effect is not associated with any significant changes in total adiponectin levels or amount of VAF. Our findings also confirm a significant association between OSA and insulin resistance.

Citation:

Iftikhar IH, Hoyos CM, Phillips CL, Magalang UJ. Meta-analyses of the association of sleep apnea with insulin resistance, and the effects of CPAP on HOMA-IR, adiponectin, and visceral adipose fat. J Clin Sleep Med 2015;11(4):475–485.


Obstructive sleep apnea (OSA), obesity, and insulin resistance are common disorders that often coexist. Although obesity is common to both the former and the latter, converging evidence indicates an independent association between the two.13 However, in the earlier studies, where large datasets were analyzed, this relationship was inferred based on statistical adjustments in the body mass index (BMI) and waist circumference.4,5 As a matter of fact, central obesity or visceral adipose fat (VAF) accumulation, much more than the degree of overall obesity, has also been shown to be independently associated with insulin resistance.68 Be that as it may, accounting for the effects of VAF in the association between OSA and insulin resistance has been relatively unexplored.1 Quantitative methods that provide an assessment of percentage body fat (e.g., DEXA scan) and VAF (such as MRI or CT scan) have been used in only a few randomized controlled trials (RCT).912 In addition to VAF, adipokines (such as leptin, and adiponectin) too, lie on the mechanistic pathway to insulin resistance, which in turn lies on the pathway to diabetes and ultimately to cardiovascular disease.2 Adiponectin, a well-studied adipokine, is synthesized by the adipocyte, and is thought to have insulin-sensitizing properties and considered protective against diabetes.1315 Therefore, hypoadiponectinemia is a risk factor for incident diabetes and coronary heart disease.14,16 Hypoadiponectinemia has also been more commonly observed in patients with OSA than in healthy individuals.2 However, the role of adiponectin, and more specifically the effects of CPAP on adiponectin, in OSA-related insulin resistance has not been clearly defined.3,17

In a previous meta-analysis, we showed a statistically significant effect of CPAP on improving insulin resistance in the non-diabetics.18 However, in that meta-analysis, a major study19 with a large number of study participants accounted for the overall results. Since the study was retracted by the authors,20 we sought to update the results of our meta-analysis by including two newer studies. Expanding on our prior results, we also attempted to study (a) the effects of CPAP on adiponectin and VAF as such, in general and in patients with OSA in specific, and (b) the magnitude of the difference in insulin resistance between those with OSA and those without in the cross-sectional case-control studies. We performed separate systematic searches and meta-analyses of RCTs on the effect of CPAP on (a) insulin resistance, as measured by homeostatic model assessment index (HOMA-IR), (b) VAF, and (c) adiponectin. We aimed to include interventional studies on Leptin and other pro-inflammatory cytokines. However, since there has been a relative paucity of published RCTs on either, they were not the focus of this meta-analysis. We also performed a separate meta-analysis with meta-regression of case-control, cross-sectional studies on the association between insulin resistance and OSA.

METHODS

Data Sources

We searched in the database of PubMed from inception to March 15, 2014, for all eligible studies.

Search Strategy and Selection Criteria

For the meta-analysis of the effect of CPAP on HOMA-IR, VAF and adiponectin, we used the search terms; “continuous positive airway pressure,” “sleep apnea,” in combination with “insulin resistance,” “HOMA-IR,” “visceral abdominal fat,” and “adiponectin.” For the meta-regression and meta-analysis of the case-control studies on the association of insulin resistance and sleep apnea, we used the search terms; “sleep apnea,” “insulin resistance,” and “HOMA-IR.” One author (IHI) searched the database and short-listed the studies. Full manuscripts of the shortlisted articles were obtained and studied. Two investigators (IHI and UJM) performed the final review using pre-specified inclusion criteria. Figure 1 outlines the search strategy and study selection process. For the meta-analysis on the effect of CPAP on HOMA-IR, we considered only RCTs, involving adult participants with OSA but without diabetes, using CPAP as the intervention, to determine the mean difference in HOMA-IR. For the meta-analysis on the effect of CPAP on VAF, and for the meta-analysis on the effect of CPAP on adiponectin we considered only RCTs that measured VAF or adiponectin in adult participants (diabetic or non-diabetic), with OSA, using CPAP as the intervention. Finally, for the meta-analysis and meta-regression of the studies on association of sleep apnea and insulin resistance, we considered only cross-sectional case-control studies that enrolled adult study participants with OSA (diabetic or non-diabetic). Prospective case-control studies were excluded.

Flow diagram of articles identified and evaluated during the study selection process.

CPAP, continuous positive airway pressure; HOMA-IR, homeostatic model assessment index for insulin resistance; VAF, visceral abdominal fat; RCT, randomized controlled trials; PCOS, polycystic ovarian syndrome.

 

jcsm.11.4.475a.jpg

jcsm.11.4.475a.jpg
Figure 1

Flow diagram of articles identified and evaluated during the study selection process. CPAP, continuous positive airway pressure; HOMA-IR, homeostatic model assessment index for insulin resistance; VAF, visceral abdominal fat; RCT, randomized controlled trials; PCOS, polycystic ovarian syndrome.

(more ...)

A total of 6 RCTs finally met the inclusion criteria for the meta-analysis on the effect of CPAP on HOMA-IR. A total of 4 RCTs were shortlisted for the meta-analysis on the effect of CPAP on VAF. A total of 3 RCTs were shortlisted for the meta-analysis on the effect of CPAP on adiponectin. Finally, we considered only 16 case-control studies for performing a meta-analysis and meta-regression on the association of sleep apnea and insulin resistance. There were no disagreements between the authors on the inclusion or exclusion of a study. Quality assessment of included studies is provided in the supplemental material.

Data Abstraction

For all meta-analyses, we extracted from each paper, the first author's name, year of publication, number of participants, country of origin, study design, demographics (including mean age, BMI, gender distribution, presence of comorbidities and the apnea-hypopnea index [AHI]), and the methods of HOMA-IR, adiponectin, and VAF assessment. One author (IHI) independently extracted data from all included studies, which were independently verified by all authors (IHI, CMH, CLP, and UJM).

Quantitative Data Synthesis

The effect of CPAP on HOMA-IR was quantified by estimating and pooling the mean difference across all RCTs. The method of Follman et al., which assumes a correlation of 0.5 between pre- and post-intervention was adopted, when necessary.21 Data was pooled across all included studies as the mean difference in HOMA-IR (HOMA-IR analysis) and VAF (VAF analysis). From the studies included in the HOMA-IR analysis, where such data was provided, we performed 2 additional corresponding analyses: (a) an analysis of the pooled mean difference in BMI between CPAP and sham CPAP, and (b) a corresponding analysis of the pooled mean difference in ESS between CPAP and sham CPAP. For the VAF analysis, we performed 2 separate analyses: (a) a meta-analysis of all available RCTs that have examined the effect of CPAP on VAF, (b) a meta-analysis of VAF data from the RCTs included in HOMA-IR analysis. The adiponectin analysis was conducted differently, as different studies reported adiponectin in different units of measurement. Therefore, we pooled this data as the standardized mean difference in adiponectin. Finally, from the included case-control studies, data on the difference of HOMA-IR between cases and controls were pooled across all studies as the standardized mean difference. Effect sizes and 95% confidence intervals (CIs) were estimated by pooling available data using the Comprehensive Meta-Analysis V2 software (v 2.2.064, Biostat, Englewood, NJ, USA).

We also conducted meta-regression for the association of pre-stated variables of interest; age, AHI and BMI, CPAP compliance, and percentage of male participants at baseline. We used the mixed effects (unrestricted maximum likelihood) method. A priori, we decided to use the random effects methods, as most studies were different from each other in terms of the units of measurement, participants and geographic locations. Heterogeneity was further assessed with I2 index. Sensitivity analyses were carried out to evaluate the effects of each selected study on the overall results of the meta-analysis.

To assess the risk of publication bias, funnel plots of standard error and difference in means were constructed and bias was assessed with the Begg and Mazumdar rank correlation test. A p value of < 0.05 on these tests was considered statistically significant for evidence of publication bias.

RESULTS

For the meta-analysis of the effect of CPAP on HOMA-IR, a total of 6 studies qualified for inclusion, enrolling 340 study participants (172 in CPAP intervention group and 168 in control).9,12,2225 For the meta-analysis of the effect of CPAP on VAF, a total of 4 studies with 206 study participants were analyzed in this meta-analysis.912 Demographic characteristics of the study population in each trial for the above mentioned meta-analyses are outlined in Table 1. A total of 3 RCTs with 200 participants (101 in CPAP and 99 in sham CPAP groups) qualified for inclusion in the meta-analysis of the effect of CPAP on adiponectin.9,26,27 Baseline demographic characteristics of the study population in each trial are outlined in Table 2. A total of 16 studies (2,456 participants in “Cases” and 3,094 participants in “Controls”) were included in the meta-analysis and meta-regression for determining the association of sleep apnea with insulin resistance.2843 Baseline characteristics of the study participants are detailed in Table 3.

Characteristics of studies for HOMA-IR and VAF analysis.

 

jcsm.11.4.475.t01.jpg

table icon
Table 1

Characteristics of studies for HOMA-IR and VAF analysis.

(more ...)

Baseline characteristics of RCTs on adiponectin.

 

jcsm.11.4.475.t02.jpg

table icon
Table 2

Baseline characteristics of RCTs on adiponectin.

(more ...)

Baseline characteristics of participants with sleep apnea in studies on the association of sleep apnea and insulin resistance

 

jcsm.11.4.475.t03.jpg

table icon
Table 3

Baseline characteristics of participants with sleep apnea in studies on the association of sleep apnea and insulin resistance.

(more ...)

Effect of CPAP on HOMA-IR

Pooled estimate of the 6 included studies showed a mean difference of −0.43 (95% CI: −0.75 to −0.11), p = 0.008, (I2 = 0%, p = 0.75), favoring the CPAP group (Figure 2).9,12,2225 On meta-regression, the mean difference did not show any relationship with baseline age, BMI, AHI, CPAP compliance, or number of male participants (Table 4). For 3 of the studies included in this HOMA-IR analysis, a corresponding analysis of the pooled mean difference in BMI between CPAP and sham CPAP showed a decrease of −0.06 (95% CI: −0.39 to 0.27), p = 0.72, (I2 = 0%, p = 0.99), as shown in Figure 3.9,12,25 Similarly, for 3 of the included studies in the HOMA-IR analysis, a corresponding analysis of the pooled mean difference in ESS between CPAP and sham CPAP showed a decrease of −1.53 (95% CI: −4.13 to 1.06), p = 0.24, (I2 = 82%, p < 0.01), as shown in Figure 3.9,23,25

Pooled mean difference in HOMA-IR between CPAP and sham CPAP.

The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

 

jcsm.11.4.475b.jpg

jcsm.11.4.475b.jpg
Figure 2

Pooled mean difference in HOMA-IR between CPAP and sham CPAP. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

(more ...)

Meta-regression of pooled mean difference in HOMA-IR.

 

jcsm.11.4.475.t04.jpg

table icon
Table 4

Meta-regression of pooled mean difference in HOMA-IR.

(more ...)

Pooled mean difference in BMI and ESS for the studies analyzed in HOMA-IR meta-analysis.

BMI, body mass index. ESS, Epworth Sleepiness Scale. HOMA-IR, homeostatic model assessment index for insulin resistance. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

 

jcsm.11.4.475c.jpg

jcsm.11.4.475c.jpg
Figure 3

Pooled mean difference in BMI and ESS for the studies analyzed in HOMA-IR meta-analysis. BMI, body mass index. ESS, Epworth Sleepiness Scale. HOMA-IR, homeostatic model assessment index for insulin resistance. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

(more ...)

Effect of CPAP on VAF

The pooled mean difference in VAF from 4 RCTs was −47.93 (95% CI: −112.58 to 16.72), p = 0.14, (I2 = 95.85%, p < 0.01) as shown in Figure 4.912 A corresponding analysis of the pooled mean difference of VAF for 2 of the studies included in the HOMA-IR analysis, showed a decrease of −10.59 cm3 (95% CIs: −37.44 to 16.25), p = 0.43, (I2 = 49%, p = 0.15), as shown in Figure 5.9,12

Pooled mean difference in VAF between CPAP and sham CPAP.

VAF, visceral abdominal fat. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

 

jcsm.11.4.475d.jpg

jcsm.11.4.475d.jpg
Figure 4

Pooled mean difference in VAF between CPAP and sham CPAP. VAF, visceral abdominal fat. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

(more ...)

Subgroup analysis of the pooled mean difference in VAF between CPAP and sham CPAP based on the studies included in HOMA-IR analysis.

VAF, visceral abdominal fat. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

 

jcsm.11.4.475e.jpg

jcsm.11.4.475e.jpg
Figure 5

Subgroup analysis of the pooled mean difference in VAF between CPAP and sham CPAP based on the studies included in HOMA-IR analysis. VAF, visceral abdominal fat. The diamond reflects the pooled estimate of mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

(more ...)

Effect of CPAP on Adiponectin

The pooled standardized mean difference in adiponectin from 3 RCTs was −0.06 (95% CI: −0.28 to 0.15), p = 0.56 (I2 = 0%, p = 0.75), as shown in Figure 6.9,26,27

Pooled standardized mean difference in adiponectin between CPAP and Sham CPAP.

The diamond reflects the pooled estimate of standardized mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

 

jcsm.11.4.475f.jpg

jcsm.11.4.475f.jpg
Figure 6

Pooled standardized mean difference in adiponectin between CPAP and Sham CPAP. The diamond reflects the pooled estimate of standardized mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

(more ...)

Meta-Analysis and Meta-Regression of Studies for the Association of Sleep Apnea and Insulin Resistance

The pooled standardized mean difference of HOMA-IR between the cases and controls from 16 case-control studies was 0.51 (95% CI: 0.28 to 0.75), p ≤ 0.001, (I2 = 88%, p < 0.01), as shown in Figure 7.2843 Meta-regression on the basis of baseline characteristics showed an association of increasing HOMA-IR in sleep apneics with baseline BMI but not with age, AHI, or gender composition in the studies (Table 5).

Pooled standardized mean difference of HOMA-IR between cases and controls.

HOMA-IR, homeostatic model assessment index for insulin resistance. The diamond reflects the pooled estimate of standardized mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

 

jcsm.11.4.475g.jpg

jcsm.11.4.475g.jpg
Figure 7

Pooled standardized mean difference of HOMA-IR between cases and controls. HOMA-IR, homeostatic model assessment index for insulin resistance. The diamond reflects the pooled estimate of standardized mean difference with 95% confidence intervals (CIs). The squares represent the individual studies

(more ...)

Meta-regression of the effect of baseline characteristics on the difference of HOMA-IR between cases and controls.

 

jcsm.11.4.475.t05.jpg

table icon
Table 5

Meta-regression of the effect of baseline characteristics on the difference of HOMA-IR between cases and controls.

(more ...)

Sensitivity Analysis

For the meta-analyses of the effect of CPAP on HOMA-IR, VAF, and adiponectin, the systematic exclusion of any single study in all of the analyses did not change the statistical significance of the overall results.

Publication Bias Assessment

The funnel plots for the analyses of the effect of CPAP on HOMA-IR and adiponectin showed no bias on visual inspection and with Begg and Mazumdar rank correlation tests or Egger's test of the intercept (Figure 8A and 8C). The funnel plot for the effect of CPAP on VAF showed some bias on visual inspection but none was detected with Begg and Mazumdar rank correlation tests or Egger's test of the intercept (Figure 8B). For the meta-analysis of the association studies, the plot showed some bias (Figure 8D).

Assessment of publication bias.

(A) Funnel plot for detecting publication bias for the analysis of the effect of CPAP on HOMA-IR. Kendall's tau b = 0.00, p = 0.50. Intercept (B0) as −0.36 (95% CI: −3.03 to 2.30) p = 0.36. (B) Funnel plot for detecting publication bias for the analysis of the effect of CPAP on VAF. Kendall's tau b = −0.30, p = 0.23. Intercept (B0) as −4.15 (95% CI: −20.47 to 12.17) p = 0.23. (C) Funnel plot for detecting publication bias for the analysis of the effect of CPAP on Adiponectin. Kendall's tau b = 0.00, p = 0.50. Intercept (B0) as 1.04 (95% CI: 12.43 to 14.53) p = 0.25. (D) Funnel plot for detecting publication bias for the analysis of studies on the association between sleep apnea and insulin resistance. Kendall's tau b = 0.37, p = 0.02. Intercept (B0) as 2.48 (95% CI: −0.15 to 5.12) p = 0.03. HOMA-IR, homeostatic model assessment index for insulin resistance. VAF, visceral abdominal fat. CI, confidence intervals; CPAP, continuous positive airway pressure.

 

jcsm.11.4.475h.jpg

jcsm.11.4.475h.jpg
Figure 8

Assessment of publication bias. (A) Funnel plot for detecting publication bias for the analysis of the effect of CPAP on HOMA-IR. Kendall's tau b = 0.00, p = 0.50. Intercept (B0) as −0.36 (95% CI: −3.03 to 2.30) p = 0.36. (B) Funnel plot for detecting publication bias for the analysis...

(more ...)

DISCUSSION

The results of our current meta-analysis confirm and expand our previous findings.18 We previously reported that CPAP improves insulin resistance in non-diabetics. This finding remains unchanged despite the removal of data from a retracted large randomized controlled trial and the addition of two new studies. In this regard, our results are in line with some of the earlier studies,44,45 including a limited meta-analysis that was based upon observational studies using a pre- to post-intervention model.46

An important additional finding based on our analysis of four RCTs14,1618 was that short-term CPAP does not appear to reduce visceral fat. Furthermore, by separately analyzing data from two of these studies, we showed that the improvements in insulin resistance with CPAP are independent of any effects on visceral adiposity. This was despite our meta-regression clearly showing that baseline BMI is the single most significant determinant of insulin resistance in these patients. However, BMI is by itself only a crude measure for abdominal visceral fat and in most of these association studies, visceral adiposity was not taken into consideration. These findings from our meta-analysis suggest that insulin resistance may improve through mechanisms that are independent of changes in abdominal fat. Indeed, more recent studies show an inability of CPAP to reduce weight, with some data even suggesting that weight increases with CPAP, especially in those who are CPAP adherent.47,48 It is important to note, however, that an increase in weight with CPAP may also reflect an increase in lean mass, which has been shown in two recent studies.49,50 This increase in lean mass may by itself explain the overall improvement in insulin sensitivity following CPAP treatment. However, longer-term randomized trials that quantify both fat and lean mass changes with CPAP together with changes in insulin resistance are required to clarify mechanisms.

Our results show that CPAP does not have an effect on total circulating levels of adiponectin, the levels of which account for up to 73% of the variance in insulin sensitivity.51 The negative finding of CPAP on adiponectin is consistent with most but not all studies.3,52 These conflicting results could be due to the uncontrolled observational study designs. However, a potential explanation could also be that adiponectin levels were studied in the study participants who had varying levels of obesity. For example, positive findings on the effect of CPAP on adipokines were reported in non-obese apneics.5357 Obesity is strongly associated with lower levels of adiponectin which is thought to be due to adipose tissue dysfunction.5860 Most of the study participants in our meta-analysis were obese. It is possible that in the continuing presence of obesity, the effects of CPAP on adiponectin may be dampened by the expanded fat mass. In addition, only the total levels of adiponectin have been examined. Adiponectin exists in high-molecular-weight (HMW) and low-molecular-weight (LMW) complexes in the circulation, with the HMW complex being the active form of this adipokine.61 It has been shown that the complex distribution, rather than the absolute amount of adiponectin, correlates with improvement in insulin sensitivity.61 Therefore, it is entirely possible that the improvement in insulin resistance with CPAP is mediated through effects on HMW adiponectin, but this has never been examined.

Our study does not address whether sleep apnea is causally linked to the development of diabetes or that it worsens glucose control in those who already have diabetes. Although it was not the scope of our meta-analysis, factors other than severity of sleep apnea and obesity, could also confound the association of sleep apnea and insulin resistance and response to CPAP. For example, extreme durations of sleep have also been linked to insulin resistance.62,63 Indeed, CPAP compliance remains an important determinant in the response of insulin resistance to CPAP.12 It can also be argued that short term (< 3 months) treatment of CPAP may not have much of an impact on improving insulin resistance.64

Our meta-analysis has several strengths. First, in our meta-analysis of the effect of CPAP on HOMA-IR, we analyzed only RCTs, as opposed to the previously performed meta-analyses by other authors where data were incorporated from observational studies only.46,65 Second, we believe our meta-analysis has sufficient power (99.9) to detect a difference in insulin resistance of −0.3 assuming an SD of 1. Finally, in all of the analyses, we were able to use meta-regression to study the effect of baseline characteristics on outcomes.

Our study also has several limitations. First, most of the studies used HOMA-IR as an index of insulin resistance, rather than using the more sensitive hyperinsulinemic euglycemic clamp procedure. However, this method is more complex and costly. Additionally, HOMA-IR assessment has been shown to have a reasonable correlation with the indices derived from the hyperinsulinemic euglycemic clamp procedure. Second, the CPAP intervention in most studies was short. In reality, it is possible that longer durations of CPAP treatment could elicit slightly different outcomes. Third, most of the study participants were males.

In conclusion, this current meta-analysis shows that CPAP treatment of OSA improves insulin resistance in OSA patients without diabetes. This effect is not associated with any significant changes in total adiponectin levels or amount of VAF. Whether or not CPAP has any effect on the complex distribution of adiponectin or body fat in other depots remains to be examined in future studies.

DISCLOSURE STATEMENT

This was not an industry supported study. Dr. Magalang reported support in part by HL093463 (UJM) and UL1TR000090 of the Ohio State University Center for Clinical and Translational Science. The other authors have indicated no financial conflicts of interest.

REFERENCES

1 

Aurora RN, Punjabi NM, authors. Sleep apnea and metabolic dysfunction: cause or co-relation? Sleep Med Clin. 2007;2:237–50. [PubMed Central][PubMed]

2 

Aurora RN, Punjabi NM, authors. Obstructive sleep apnoea and type 2 diabetes mellitus: a bidirectional association. Lancet Respir Med. 2013;1:329–38. [PubMed]

3 

Punjabi NM, Polotsky VY, authors. Disorders of glucose metabolism in sleep apnea. J Appl Physiol. 2005;99:1998–2007. [PubMed]

4 

Al-Delaimy WK, Manson JE, Willett WC, Stampfer MJ, Hu FB, authors. Snoring as a risk factor for type II diabetes mellitus: a prospective study. Am J Epidemiol. 2002;155:387–93. [PubMed]

5 

Punjabi NM, Shahar E, Redline S, Gottlieb DJ, Givelber R, Resnick HE, authors. Sleep-disordered breathing, glucose intolerance, and insulin resistance: the Sleep Heart Health Study. Am J Epidemiol. 2004;160:521–30. [PubMed]

6 

Evans DJ, Hoffmann RG, Kalkhoff RK, Kissebah AH, authors. Relationship of body fat topography to insulin sensitivity and metabolic profiles in premenopausal women. Metab Clin Exp. 1984;33:68–75. [PubMed]

7 

Kobayashi H, Nakamura T, Miyaoka K, et al., authors. Visceral fat accumulation contributes to insulin resistance, small-sized low-density lipoprotein, and progression of coronary artery disease in middle-aged non-obese Japanese men. Jpn Circ J. 2001;65:193–9. [PubMed]

8 

Cefalu WT, Wang ZQ, Werbel S, et al., authors. Contribution of visceral fat mass to the insulin resistance of aging. Metab Clin Exp. 1995;44:954–9. [PubMed]

9 

Hoyos CM, Killick R, Yee BJ, Phillips CL, Grunstein RR, Liu PY, authors. Cardiometabolic changes after continuous positive airway pressure for obstructive sleep apnoea: a randomised sham-controlled study. Thorax. 2012;67:1081–9. [PubMed]

10 

Kritikou I, Basta M, Tappouni R, et al., authors. Sleep apnoea and visceral adiposity in middle-aged male and female subjects. Eur Respir J. 2013;41:601–9. [PubMed]

11 

Sivam S, Phillips CL, Trenell MI, et al., authors. Effects of 8 weeks of continuous positive airway pressure on abdominal adiposity in obstructive sleep apnoea. Eur Respir J. 2012;40:913–8. [PubMed]

12 

Weinstock TG, Wang X, Rueschman M, et al., authors. A controlled trial of CPAP therapy on metabolic control in individuals with impaired glucose tolerance and sleep apnea. Sleep. 2012;35:617–25B. [PubMed Central][PubMed]

13 

Masserini B, Morpurgo PS, Donadio F, et al., authors. Reduced levels of adiponectin in sleep apnea syndrome. J Endocrin Invest. 2006;29:700–5.

14 

Shehzad A, Iqbal W, Shehzad O, Lee YS, authors. Adiponectin: regulation of its production and its role in human diseases. Hormones (Athens). 2012;11:8–20.

15 

Zhang XL, Yin KS, Wang H, Su S, authors. Serum adiponectin levels in adult male patients with obstructive sleep apnea hypopnea syndrome. Respiration. 2006;73:73–7. [PubMed]

16 

Hajer GR, van Haeften TW, Visseren FL, authors. Adipose tissue dysfunction in obesity, diabetes, and vascular diseases. Eur Heart J. 2008;29:2959–71. [PubMed]

17 

Polak J, BB, Punjabi NM, authors; Pack AI, editor. Obstructive sleep apnea and glucose metabolism. Sleep apnea: sleep apnea pathogenesis, diagnosis and treatment. 2nd ed. London: Informa Healthcare, 2012.

18 

Iftikhar IH, Khan MF, Das A, Magalang UJ, authors. Meta-analysis: continuous positive airway pressure improves insulin resistance in patients with sleep apnea without diabetes. Ann Am Thorac Soc. 2013;10:115–20. [PubMed Central][PubMed]

19 

Sharma SK, Agrawal S, Damodaran D, et al., authors. CPAP for the metabolic syndrome in patients with obstructive sleep apnea. N Engl J Med. 2011;365:2277–86. [PubMed]

20 

Sharma SK, Agrawal S, Damodaran D, et al., authors. Retraction: CPAP for the metabolic syndrome in patients with obstructive sleep apnea. N Engl J Med. 2011;365:2277–86. N Engl J Med 2013;369:1770. [PubMed]

21 

Follmann D, Elliott P, Suh I, Cutler J, authors. Variance imputation for overviews of clinical trials with continuous response. J Clin Epidemiol. 1992;45:769–73. [PubMed]

22 

Comondore VR, Cheema R, Fox J, et al., authors. The impact of CPAP on cardiovascular biomarkers in minimally symptomatic patients with obstructive sleep apnea: a pilot feasibility randomized crossover trial. Lung. 2009;187:17–22. [PubMed]

23 

Coughlin SR, Mawdsley L, Mugarza JA, Wilding JP, Calverley PM, authors. Cardiovascular and metabolic effects of CPAP in obese males with OSA. Eur Respir J. 2007;29:720–7. [PubMed]

24 

Duarte FH, Jallad RS, Amaro AC, Drager LF, Lorenzi-Filho G, Bronstein MD, authors. The impact of sleep apnea treatment on carbohydrate metabolism in patients with acromegaly. Pituitary. 2013;16:341–50. [PubMed]

25 

Lam JC, Lam B, Yao TJ, et al., authors. A randomised controlled trial of nasal continuous positive airway pressure on insulin sensitivity in obstructive sleep apnoea. Eur Respir J. 2010;35:138–45. [PubMed]

26 

Kohler M, Ayers L, Pepperell JC, et al., authors. Effects of continuous positive airway pressure on systemic inflammation in patients with moderate to severe obstructive sleep apnoea: a randomised controlled trial. Thorax. 2009;64:67–73. [PubMed]

27 

West SD, Nicoll DJ, Wallace TM, Matthews DR, Stradling JR, authors. Effect of CPAP on insulin resistance and HbA1c in men with obstructive sleep apnoea and type 2 diabetes. Thorax. 2007;62:969–74. [PubMed Central][PubMed]

28 

Barcelo A, Barbe F, de la Pena M, et al., authors. Insulin resistance and daytime sleepiness in patients with sleep apnoea. Thorax. 2008;63:946–50. [PubMed]

29 

Basoglu OK, Sarac F, Sarac S, Uluer H, Yilmaz C, authors. Metabolic syndrome, insulin resistance, fibrinogen, homocysteine, leptin, and C-reactive protein in obese patients with obstructive sleep apnea syndrome. Ann Thorac Med. 2011;6:120–5. [PubMed Central][PubMed]

30 

Bulcun E, Ekici M, Ekici A, authors. Disorders of glucose metabolism and insulin resistance in patients with obstructive sleep apnoea syndrome. Int J Clin Pract. 2012;66:91–7. [PubMed]

31 

Coughlin SR, Mawdsley L, Mugarza JA, Calverley PM, Wilding JP, authors. Obstructive sleep apnoea is independently associated with an increased prevalence of metabolic syndrome. Eur Heart J. 2004;25:735–41. [PubMed]

32 

Gruber A, Horwood F, Sithole J, Ali NJ, Idris I, authors. Obstructive sleep apnoea is independently associated with the metabolic syndrome but not insulin resistance state. Cardiovasc Diabetol. 2006;5:22. [PubMed Central][PubMed]

33 

Ip MS, Lam B, Ng MM, Lam WK, Tsang KW, Lam KS, authors. Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med. 2002;165:670–6. [PubMed]

34 

Kapsimalis F, Varouchakis G, Manousaki A, et al., authors. Association of sleep apnea severity and obesity with insulin resistance, C-reactive protein, and leptin levels in male patients with obstructive sleep apnea. Lung. 2008;186:209–17. [PubMed]

35 

Kim NH, Cho NH, Yun CH, et al., authors. Association of obstructive sleep apnea and glucose metabolism in subjects with or without obesity. Diabetes Care. 2013;36:3909–15. [PubMed Central][PubMed]

36 

Kritikou I, Basta M, Vgontzas AN, et al., authors. Sleep apnoea, sleepiness, inflammation and insulin resistance in middle-aged males and females. Eur Respir J. 2014;43:145–55. [PubMed]

37 

Mazzuca E, Battaglia S, Marrone O, et al., authors. Gender-specific anthropometric markers of adiposity, metabolic syndrome and visceral adiposity index (VAI) in patients with obstructive sleep apnea. J Sleep Res. 2014;23:13–21. [PubMed]

38 

McArdle N, Hillman D, Beilin L, Watts G, authors. Metabolic risk factors for vascular disease in obstructive sleep apnea: a matched controlled study. Am J Respir Crit Care Med. 2007;175:190–5. [PubMed]

39 

Onat A, Hergenc G, Uyarel H, et al., authors. Obstructive sleep apnea syndrome is associated with metabolic syndrome rather than insulin resistance. Sleep Breath. 2007;11:23–30. [PubMed]

40 

Otake K, Sasanabe R, Hasegawa R, et al., authors. Glucose intolerance in Japanese patients with obstructive sleep apnea. Intern Med. 2009;48:1863–8. [PubMed]

41 

Peled N, Kassirer M, Shitrit D, et al., authors. The association of OSA with insulin resistance, inflammation and metabolic syndrome. Respir Med. 2007;101:1696–701. [PubMed]

42 

Qian X, Yin T, Li T, et al., authors. High levels of inflammation and insulin resistance in obstructive sleep apnea patients with hypertension. Inflammation. 2012;35:1507–11. [PubMed]

43 

Tkacova R, Dorkova Z, Molcanyiova A, Radikova Z, Klimes I, Tkac I, authors. Cardiovascular risk and insulin resistance in patients with obstructive sleep apnea. Med Sci Monit. 2008;14:CR438–44. [PubMed]

44 

Punjabi NM, Sorkin JD, Katzel LI, Goldberg AP, Schwartz AR, Smith PL, authors. Sleep-disordered breathing and insulin resistance in middle-aged and overweight men. Am J Respir Crit Care Med. 2002;165:677–82. [PubMed]

45 

Tassone F, Lanfranco F, Gianotti L, et al., authors. Obstructive sleep apnoea syndrome impairs insulin sensitivity independently of anthropometric variables. Clin Endocrinol. 2003;59:374–9.

46 

Yang D, Liu Z, Yang H, Luo Q, authors. Effects of continuous positive airway pressure on glycemic control and insulin resistance in patients with obstructive sleep apnea: a meta-analysis. Sleep Breath. 2013;17:33–8. [PubMed]

47 

Quan SF, Budhiraja R, Clarke DP, et al., authors. Impact of treatment with continuous positive airway pressure (CPAP) on weight in obstructive sleep apnea. J Clin Sleep Med. 2013;9:989–93. [PubMed Central][PubMed]

48 

Redenius R, Murphy C, O'Neill E, Al-Hamwi M, Zallek SN, authors. Does CPAP lead to change in BMI? J Clin Sleep Med. 2008;4:205–9. [PubMed Central][PubMed]

49 

Hoyos CM, Yee BJ, Phillips CL, Machan EA, Grunstein RR, Liu PY, authors. Body compositional and cardiometabolic effects of testosterone therapy in obese men with severe obstructive sleep apnoea: a randomised placebo-controlled trial. Eur J Endocrinol. 2012;167:531–41. [PubMed]

50 

Munzer T, Hegglin A, Stannek T, et al., authors. Effects of long-term continuous positive airway pressure on body composition and IGF1. Eur J Endocrinol. 2010;162:695–704. [PubMed]

51 

Bacha F, Saad R, Gungor N, Arslanian SA, authors. Adiponectin in youth: relationship to visceral adiposity, insulin sensitivity, and beta-cell function. Diabetes Care. 2004;27:547–52. [PubMed]

52 

Harsch IA, Wallaschofski H, Koebnick C, et al., authors. Adiponectin in patients with obstructive sleep apnea syndrome: course and physiological relevance. Respiration. 2004;71:580–6. [PubMed]

53 

Harsch IA, Koebnick C, Wallaschofski H, et al., authors. Resistin levels in patients with obstructive sleep apnoea syndrome--the link to subclinical inflammation? Med Sci Monit. 2004;10:CR510–5. [PubMed]

54 

Minoguchi K, Tazaki T, Yokoe T, et al., authors. Elevated production of tumor necrosis factor-alpha by monocytes in patients with obstructive sleep apnea syndrome. Chest. 2004;126:1473–9. [PubMed]

55 

Ryan S, Taylor CT, McNicholas WT, authors. Selective activation of inflammatory pathways by intermittent hypoxia in obstructive sleep apnea syndrome. Circulation. 2005;112:2660–7. [PubMed]

56 

Ryan S, Taylor CT, McNicholas WT, authors. Predictors of elevated nuclear factor-kappaB-dependent genes in obstructive sleep apnea syndrome. Am J Respir Crit Care Med. 2006;174:824–30. [PubMed]

57 

Yokoe T, Minoguchi K, Matsuo H, et al., authors. Elevated levels of C-reactive protein and interleukin-6 in patients with obstructive sleep apnea syndrome are decreased by nasal continuous positive airway pressure. Circulation. 2003;107:1129–34. [PubMed]

58 

Sun K, Kusminski CM, Scherer PE, authors. Adipose tissue remodeling and obesity. J Clin Invest. 2011;121:2094–101. [PubMed Central][PubMed]

59 

Halberg N, Khan T, Trujillo ME, et al., authors. Hypoxia-inducible factor 1alpha induces fibrosis and insulin resistance in white adipose tissue. Mol Cell Biol. 2009;29:4467–83. [PubMed Central][PubMed]

60 

Sun K, Halberg N, Khan M, Magalang UJ, Scherer PE, authors. Selective inhibition of hypoxia-inducible factor 1alpha ameliorates adipose tissue dysfunction. Mol Cell Biol. 2013;33:904–17. [PubMed Central][PubMed]

61 

Pajvani UB, Hawkins M, Combs TP, et al., authors. Complex distribution, not absolute amount of adiponectin, correlates with thiazolidinedione-mediated improvement in insulin sensitivity. J Biol Chem. 2004;279:12152–62. [PubMed]

62 

Ayas NT, White DP, Al-Delaimy WK, et al., authors. A prospective study of self-reported sleep duration and incident diabetes in women. Diabetes Care. 2003;26:380–4. [PubMed]

63 

Gottlieb DJ, Punjabi NM, Newman AB, et al., authors. Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med. 2005;165:863–7. [PubMed]

64 

Harsch IA, Schahin SP, Bruckner K, et al., authors. The effect of continuous positive airway pressure treatment on insulin sensitivity in patients with obstructive sleep apnoea syndrome and type 2 diabetes. Respiration. 2004;71:252–9. [PubMed]

65 

Yang D, Liu Z, Yang H, authors. The impact of effective continuous positive airway pressure on homeostasis model assessment insulin resistance in non-diabetic patients with moderate to severe obstructive sleep apnea. Diabetes Metab Res Rev. 2012;28:499–504. [PubMed]

SUPPLEMENTAL MATERIAL

Quality assessment of randomized controlled trials (using jadad scale).

 

jcsm.11.4.475.t0S1.jpg

table icon
Table S1

Quality assessment of randomized controlled trials (using jadad scale).

(more ...)

Quality assessment tool for observational cohort and cross-sectional studies.

 

jcsm.11.4.475.t0S2.jpg

table icon
Table S2

Quality assessment tool for observational cohort and cross-sectional studies.

(more ...)

Checklist summarizing compliance with PRISMA guidelines.

 

jcsm.11.4.475.t0S3.jpg

table icon
Table S3

Checklist summarizing compliance with PRISMA guidelines.

(more ...)