You are here

Article Text

Article menu

Download PDFPDF

XML
Neurology
Research
Association of obstructive sleep apnoea with the risk of vascular outcomes and all-cause mortality: a meta-analysis
  1. Chengjuan Xie1,
  2. Ruolin Zhu1,
  3. Yanghua Tian1,2,
  4. Kai Wang1,2,3
  1. 1 Department of Neurology, The First Affiliated Hospital of Anhui Medical University, Hefei, China
  2. 2 Collaborative Innovation Centre of Neuropsychiatric Disorders and Mental Health, Anhui Province, Hefei, China
  3. 3 Department of Medical Psychology, Anhui Medical University, Hefei, China
  1. Correspondence to Dr Kai Wang; wangkai1964{at}126.com

Abstract

Objective This study aimed to conduct a meta-analysis to explore and summarise the evidence regarding the association between obstructive sleep apnoea (OSA) and the subsequent risk of vascular outcomes and all-cause mortality.

Methods Electronic databases PubMed, Embase and the Cochrane Library were searched to identify studies conducted through May 2016. Prospective cohort studies that reported effect estimates with 95% CIs of major adverse cardiac events (MACEs), coronary heart disease (CHD), stroke, cardiac death, all-cause mortality and heart failure for different levels versus the lowest level of OSA were included.

Results A total of 16 cohort studies reporting data on 24 308 individuals were included. Of these, 11 studies reported healthy participants, and the remaining five studies reported participants with different diseases. Severe OSA was associated with an increased risk of MACEs (relative risk (RR): 2.04; 95% CI 1.56 to 2.66; P<0.001), CHD (RR: 1.63; 95% CI 1.18 to 2.26; P=0.003), stroke (RR: 2.15; 95% CI 1.42 to 3.24; P<0.001), cardiac death (RR: 2.96; 95% CI 1.45 to 6.01; P=0.003) and all-cause mortality (RR: 1.54; 95% CI 1.21 to 1.97; P<0.001). Moderate OSA was also significantly associated with increased risk of MACEs (RR: 1.16; 95% CI 1.01 to 1.33; P=0.034) and CHD (RR: 1.38; 95% CI 1.04 to 1.83; P=0.026). No significant association was found between mild OSA and the risk of vascular outcomes or all-cause mortality (P>0.05). Finally, no evidence of a factor-specific difference in the risk ratio for MACEs among participants with different levels of OSA compared with those with the lowest level of OSA was found.

Conclusions Severe and moderate OSAs were associated with an increased risk of vascular outcomes and all-cause mortality. This relationship might differ between genders. Therefore, further large-scale prospective studies are needed to verify this difference.

  • meta-analysis
  • mortality
  • obstructive sleep apnea
  • vascular outcome

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/

https://doi.org/10.1136/bmjopen-2016-013983

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study:

    • This was a meta-analysis to elucidate the association of obstructive sleep apnoea (OSA) with fatal and non-fatal cardiovascular diseases, using a broad search strategy and predefined selection criteria and with no restriction of language or publication status.

    • The methodological quality of each study was assessed using the Newcastle-Ottawa Scale for prospective observational studies, and a meta-analysis, sensitivity analysis, subgroup analysis and bias assessment were also conducted.

    • Only prospective studies were included, eliminating selection and recall bias that could be of concern in retrospective case–control studies.

    • Summary relative risks were calculated to evaluate any potential difference between subsets according to the characteristics of the participants.

    • Different cut-off values for the apnoea–hypopnoea index might affect the relationship between OSA and vascular outcomes.

    Introduction

    Obstructive sleep apnoea (OSA) affects 24% of middle-aged men and 9% of women in the USA, but daytime sleepiness was reported in 17% and 22% of these subjects, respectively.1 OSA is an increasingly prevalent condition characterised by repetitive obstruction of the upper airway during sleep accompanied by episodic hypoxia, arousal and sleep fragmentation.2 Previous studies suggested that OSA was associated with an increased risk of glaucoma, diabetic kidney disease and metabolic syndrome.3–5 However, data on the association between OSA and the risk of subsequent vascular outcomes and mortality are both limited and inconclusive. Furthermore, whether these relationships differ according to the characteristics of patients with OSA also needs to be verified.

    Several meta-analyses have illustrated that continuous positive airway pressure (CPAP) interventions aimed at OSA may reduce the risk of cardiovascular outcomes. Kim et al 6 showed that CPAP treatment for OSA was associated with a lower incidence of stroke and cardiac events. Furthermore, Bratton et al 7 indicated that use of both CPAP and mandibular advancement devices was associated with a reduction in the blood pressure among patients with OSA. Nadeem et al 8 suggested that CPAP treatment for OSA seemed to improve dyslipidaemia (decrease in total cholesterol and low-density lipoprotein and increase in high-density lipoprotein), whereas it did not appear to affect the triglyceride levels. These studies demonstrated that patients with OSA who received interventions had a reduced risk of cardiovascular diseases. Therefore, clarifying the relationship between OSA and vascular outcomes is particularly important as it has not been definitively determined. This study attempted to perform a large-scale examination of the available prospective studies to determine the association of OSA with the potential risk of vascular outcomes and all-cause mortality.

    Methods

    Data sources, search strategy and selection criteria

    This study was conducted and reported according to the Meta-analysis of Observational Studies in Epidemiology protocol (Checklist S1).9

    Any prospective cohort study that examined the relationship between OSA and vascular outcomes or all-cause mortality was eligible for inclusion into this study, and no restrictions were placed on language or publication status (eg, published, in press or in progress). Electronic databases PubMed, Embase and the Cochrane Library were searched for articles published through May 2016, using the terms ‘sleep apnea’ OR ‘obstructive sleep apneas’ AND (‘cardiovascular disease’ OR ‘stroke’ OR ‘cardiac death’ OR ‘mortality’ OR ‘death’ OR ‘CVD’ OR ‘myocardial infarction’ OR ‘coronary events’) AND ‘clinical trials’ AND ‘human’ as the search terms (online supplementary 1). Manual searches of reference lists were also conducted from all the relevant original and reviewed articles to identify additional eligible studies. The medical subject heading, methods, patient population, design, exposure and outcome variables of these articles were used to identify the relevant studies.

    Supplementary file 1

    The literature search was independently undertaken by two authors using a standardised approach. Any inconsistencies between these two authors were settled by the primary author until a consensus was reached. The study was eligible for inclusion if the following criteria were met: (1) the study had a prospective cohort design; (2) the study investigated the association between OSA and the risk of major adverse cardiac events (MACEs), coronary heart disease (CHD), stroke, cardiac death, all-cause mortality and heart failure; and (3) the authors reported effect estimates (relative risk (RR), HR or OR) and 95% CIs for comparisons of different levels of OSA versus lowest OSA level. All case–control studies were excluded because various confounding factors could bias the results.

    Data collection and quality assessment

    The data collected included the first author’s name, publication year, country, sample size, mean age at baseline, percentage of male patients, body mass index (BMI), disease status, assessment of OSA, follow-up duration, effect estimate and its 95% CI, reported endpoints and covariates in the fully adjusted model. For studies that reported several multivariable adjusted RRs, the effect estimate that was maximally adjusted for potential confounders was selected.

    The Newcastle-Ottawa Scale (NOS), which is quite comprehensive and has been partially validated for evaluating the quality of observational studies in the meta-analysis, was used to evaluate the methodological quality.10 The NOS is based on the following three subscales: selection (four items), comparability (one item) and outcome (three items). A ‘star system’ (range: 0–9) was developed for assessment (table 1). The data extraction and quality assessment were conducted independently by two authors. Information was examined and adjudicated independently by an additional author referring to the original studies.

    View this table:
    Table 1

    Baseline characteristic of studies included in the systematic review and meta-analysis

    Statistical analysis

    The relationship between OSA and the risk of vascular outcomes or all-cause mortality based on the effect estimate (OR, RR or HR) and its 95% CI was examined in each study. HR was considered to be equivalent to RR in cohort studies. Given the low incidence of vascular outcomes and all-cause mortality, ORs could be considered as accurate estimates of RRs.11 A semiparametric method was first used to evaluate the association of mild OSA (apnoea–hypopnea index (AHI): 5–15), moderate OSA (AHI: 15–30) and severe OSA (AHI >30) with the risk of vascular outcomes or all-cause mortality in order to analyse the trend between OSA levels and vascular outcomes or all-cause mortality risk.12 For each individual study, each category of AHI was reclassified based on its calculated midpoint (for closed categories) or median (for open categories, assuming a normal distribution for AHI). The control category was composed of participants with the lowest AHI or normal participants in that study. Furthermore, when an individual study provided more than one median AHI level for classification among the three categories (ie, mild, moderate or severe OSA), a fixed-effects model was used to calculate their summary RRs and 95% CIs to obtain effect estimates for each category.13 If the study data were not broken down by AHI but rather by oxygen desaturation index, classification into the OSA categories was carried out based on the judgement of the clinicians. A random-effects model was then used to calculate summary RRs and 95% CIs for mild, moderate and severe OSA versus normal.14 Finally, the ratio of RRs and the corresponding 95% CIs between subgroups were estimated using specific RRs and 95% CIs in each group based on the country, mean age, gender, BMI, disease status and duration of the follow-up period.15

    Heterogeneity between studies was investigated using the Q statistic, and P values <0.10 were considered as indicative of significant heterogeneity.16 17 Subgroup analyses were conducted for mild, moderate and severe OSA and the risk of MACEs based on the country, mean age, gender, BMI, disease status and duration of the follow-up period. A sensitivity analysis was also performed by removing each individual study from the meta-analysis.18 Several methods were used to check for potential publication bias. Visual inspections of funnel plots for MACEs were conducted. The Egger19 and Begg20 tests were also used to statistically assess publication bias for MACEs. All reported P values were two sided, and P values <0.05 were regarded as statistically significant for all included studies. Statistical analyses were performed using the STATA software (V.12.0).

    Results

    Literature search

    The results of the study selection process are shown in figure 1. An initial electronic search yielded 3282 articles, of which 3236 duplicates and irrelevant studies were excluded, and 46 potentially eligible studies were selected. After detailed evaluations, 16 prospective studies were selected for the final meta-analysis.21–36 No new studies qualified for inclusion after a manual search of the reference lists of these studies. The general characteristics of the included studies are presented in table 1.

    Figure 1
    Figure 1

    Study selection process.

    Study characteristics

    A total of 16 studies with 24 308 individuals qualified for this study. The follow-up period for participants was 2.9–18.0 years, while 73–10 149 individuals were included in each study. Eight studies were conducted in the USA, four in Spain, one in Sweden, one in Portugal, one in Hungary and one in Canada. Furthermore, 11 studies reported healthy participants, and the remaining five studies reported participants with different diseases. The mean BMI ranged from 26.8 to 34.0 kg/m2. Fourteen studies used polysomnography (PSG), and the remaining one study used limited PSG to assess the levels of OSA. The study quality was assessed using the NOS (table 1). Overall, one study had a score of 9, six studies had a score of 8, seven studies had a score of 7 and the remaining two studies had a score of 6.

    OSA and MACE risk

    The summary RRs showed that mild OSA was not associated with MACEs (RR: 0.98; 95% CI 0.87 to 1.11; P=0.741; figure 2 and table 2). Furthermore, the pooled analysis results for moderate and severe OSA indicated that they had a harmful effect on the risk of MACEs (moderate: RR: 1.16; 95% CI 1.01 to 1.33; P=0.034; figure 3 and table 2; severe: RR: 2.04; 95% CI 1.56 to 2.66; P<0.001; figure 4 and table 2). A subgroup analysis for MACEs was conducted to minimise heterogeneity among the included studies and evaluate the relationship between OSA and MACEs in specific subpopulations (table 3). Overall, participants with moderate OSA were associated with an increased risk of MACEs if individuals did not have other diseases (RR: 1.16; 95% CI 1.01 to 1.33; P=0.034). Furthermore, no significant association was found between severe OSA and MACEs if the study included only women (RR: 1.98; 95% CI 0.64 to 6.06; P=0.234); in other subsets, severe OSA was associated with an increased risk of MACEs (table 3). Finally, no evidence of a factor-specific difference was found in the RR for MACEs among participants with OSA compared with controls (table 3).

    View this table:
    Table 2

    Summary of the relative risks of all outcomes evaluated

    View this table:
    Table 3

    Subgroup analyses for MACEs

    Figure 2
    Figure 2

    Association between mild OSA and MACEs. MACES, major adverse cardiac events; OSA, obstructive sleep apnoea; RR, relative risk.

    Figure 3
    Figure 3

    Association between moderate OSA and MACEs. MACES, major adverse cardiac events; OSA, obstructive sleep apnoea.

    Figure 4
    Figure 4

    Association between severe OSA and MACEs. MACES, major adverse cardiac events; OSA, obstructive sleep apnoea; RR, relative risk.

    OSA and CHD risk

    The pooled data of meta-analysis showed that mild OSA was not associated with the risk of CHD (RR: 1.25; 95% CI 0.95 to 1.66; P=0.117; table 2 and online supplementary 2), whereas moderate OSA (RR: 1.38; 95% CI 1.04 to 1.83; P=0.026; table 2 and online supplementary 2) and severe OSA (RR: 1.63; 95% CI 1.18 to 2.26; P=0.003; table 2 and online supplementary 2) were associated with a significantly increased risk of CHD. Stratified analyses according to gender were conducted for different levels of OSA versus normal group, and it was found that patients with severe OSA had significantly increased the risk of CHD in men (RR: 1.65; 95% CI 1.06 to 2.57; P=0.027). No other significant differences were detected (table 4).

    Supplementary file 2

    View this table:
    Table 4

    Gender difference for other outcomes

    OSA and stroke risk

    Pooled analysis results indicated no association between mild OSA (RR: 1.29; 95% CI 0.69 to 2.41; P=0.424; table 2 and online supplementary 2) and moderate OSA (RR: 1.35; 95% CI 0.82 to 2.23; P=0.245; table 2 andonline supplementary 2) and stroke, whereas severe OSA was associated with an increased risk of stroke (RR: 2.15; 95% CI 1.42 to 3.24; P<0.001; table 2 and online supplementary 2). Subgroup analysis on the basis of gender indicated that severe OSA had a harmful effect on the risk of stroke in men (RR: 2.86; 95% CI 1.10 to 7.41; P=0.031; table 4).

    OSA and cardiac death risk

    The summary RRs showed that mild OSA (RR: 1.80; 95% CI 0.68 to 4.76; P=0.236; table 2 and  online supplementary 2) and moderate OSA (RR: 1.11; 95% CI 0.53 to 2.35; P=0.781; table 2 and online supplementary 2) were not associated with cardiac death risk, whereas severe OSA significantly increased the risk of cardiac death (RR: 2.96; 95% CI 1.45 to 6.01; P=0.003; table 2 and online supplementary 2). Subgroup analysis showed that severe OSA was associated with an increased risk of cardiac death in men (RR: 2.87; 95% CI 1.13 to 7.27; P=0.026; table 4).

    OSA and all-cause mortality risk

    No significant association was found between mild OSA (RR: 1.26; 95% CI 0.77 to 2.07; P=0.354; table 2 and online supplementary 2), moderate OSA (RR: 1.04; 95% CI 0.60 to 1.79; P=0.895; table 2 and online supplementary 2) and all-cause mortality risk. However, severe OSA had a harmful impact on the all-cause mortality (RR: 1.54; 95% CI 1.21 to 1.97; P<0.001; table 2 and online supplementary 2). Stratified analysis suggested that severe OSA increased the risk of all-cause mortality in men (RR: 1.72; 95% CI 1.22 to 2.43; P=0.002) and women (RR: 3.50; 95% CI 1.23 to 9.97; P=0.019; table 4).

    OSA and heart failure risk

    The summary results indicated no significant differences between mild OSA (RR: 1.02; 95% CI 0.78 to 1.34; P=0.868), moderate OSA (RR: 1.07; 95% CI 0.74 to 1.54; P=0.719) and severe OSA (RR: 1.44; 95% CI 0.94 to 2.21; P=0.097) and the risk of heart failure (table 2 and online supplementary 2). Subgroup analysis reported similar results compared with the overall analysis (table 4).

    Publication bias

    Review of the funnel plots could not rule out the potential publication bias for MACEs (figure 5). The Egger and Begg test results showed no evidence of publication bias for MACEs of mild OSA (P value for Egger: 0.132; P value for Begg: 0.221) and moderate OSA (P value for Egger: 0.052; P value for Begg: 0.452). Although the Begg test showed no evidence of publication bias for MACEs of severe OSA (P=0.118), the Egger test showed potential evidence of publication bias for MACEs of severe OSA (P<0.001). The conclusion did not change after adjustment for publication bias using the trim-and-fill method.37

    Figure 5
    Figure 5

    Funnel plots. MACES, major adverse cardiac events; OSA, obstructive sleep apnoea; RR, relative risk.

    Discussion

    The present study was based on prospective cohort studies and explored all possible correlations between OSA and the outcomes of MACEs, CHD, stroke, cardiac death, all-cause mortality and heart failure. This large quantitative study included 24 308 individuals from 16 prospective cohort studies with a broad range of populations. The findings from the present meta-analysis suggested that mild OSA had no significant impact on the risk of vascular outcomes and all-cause mortality, moderate OSA was associated with an increased risk of MACEs and CHD and severe OSA had a harmful effect on the risk of MACEs, CHD, stroke, cardiac death and all-cause mortality.

    A previous meta-analysis suggested that OSA was associated with stroke, but its relationship with ischaemic heart disease and cardiovascular mortality needs further research.38 However, this study could not illustrate the impact of different levels of OSA on the risk of serious cardiovascular outcomes. Furthermore, Dong et al 39 suggested that moderate-to-severe OSA significantly increased the risk of cardiovascular diseases, in particular, the risk of stroke. Similarly, Ge et al 40 indicated that severe OSA is a strong independent predictor of cardiovascular and all-cause mortality. CPAP treatment was associated with decreased cardiovascular mortality. However, these two studies could not evaluate the association of OSA with the risk of vascular outcomes and all-cause mortality in specific subpopulations. In addition, Wang et al 41 suggested that severe OSA significantly increased the risk of CHD and stroke, and all-cause mortality. A positive association with CHD was observed for moderate OSA but not for mild OSA. However, whether this relationship differs according to the characteristics of participants remains unclear. Finally, Xie et al 42 conducted a meta-analysis to evaluate the relationship between OSA and recurrent vascular events and all-cause mortality. However, they just compared the highest AHI versus lowest AHI, whereas the degree of OSA and subsequent adverse outcomes were not available. Therefore, a comprehensive meta-analysis of these prospective cohort studies was performed to evaluate any possible correlates between OSA and vascular outcomes.

    No significant difference was observed between mild OSA and the risk of vascular outcomes. However, several studies included in this study reported inconsistent results. Young et al 25 suggested that mild OSA significantly increased the risk of CHD by 92%, whereas Punjabi et al 28 indicated that mild OSA might have a harmful effect on the risk of CHD. This might be because these two studies used healthy individuals as controls, which may make them more susceptible to acquired significant conclusion. Furthermore, most of these studies did not take into account potential confounders for the risk of cardiovascular disease. Moderate-to-severe OSA might play an important role in the risk of vascular outcomes. Shah et al 29 concluded that OSA increased the risk of coronary events or death from cardiovascular causes. Nearly all included studies reported adverse outcomes for severe OSA. Finally, Previous studies indicated that OSA was a cause of diabetes, which was an independent risk factor for MACEs.43

    Subgroup analyses reported similar conclusions. Gender might have an impact on the relationship between OSA and CHD, stroke or cardiac death, although the sex difference was not statistically significant. The possible reasons could be the lower prevalence of severe OSA in women and the later age of onset of OSA in women than in men. Furthermore, OSA in women always occurred after menopause. Physiological response to OSA is another reason for this non-significant difference. Finally, these conclusions might be unreliable because smaller cohorts were included in each subset. Therefore, further large-scale studies were needed to verify this difference. Therefore, a relative result was given, and a synthetic and comprehensive review was provided.

    No significant difference was found between mild or moderate OSA and all-cause mortality, while severe OSA was associated with an increased risk of all-cause mortality. Furthermore, these significant associations were also observed in men and women separately. Although the effect estimate in women was larger than that in men, no gender difference was found in the relationship between OSA and all-cause mortality. This might be because the number of studies that reported the relationship between severe OSA and all-cause mortality was smaller than expected, and a broad 95% CI was acquired. Therefore, the association of severe OSA with all-cause mortality in women was variable and should be verified in future large-scale prospective studies.

    Three strengths of this study should be highlighted. First, only prospective studies were included, which eliminated selection and recall bias, and could be of concern in retrospective case−control studies. Second, the large sample size allowed us to quantitatively assess the association of OSA with the risk of vascular outcomes and mortality, and thus the findings were potentially more robust than those of any individual study. Third, the summary RRs were calculated to evaluate any potential difference between subsets according to the characteristics of participants.

    The limitations of this study were as follows: (1) the adjusted models were different across the included studies, and these factors might have played an important role in developing vascular outcomes; (2) in a meta-analysis of published studies, publication bias was an inevitable problem; and (3) the analysis used pooled data (individual data were not available), which restricted performing a more detailed relevant analysis and obtaining more comprehensive results.

    The results of this study suggested that moderate-to-severe OSA might play an important role in the risk of vascular outcomes, especially for men. Future studies should focus on specific populations to analyse the gender difference to study the association between OSA and vascular outcomes.

    References

    1. 1.↵
      2. Young T ,
      3. Palta M ,
      4. Dempsey J , et al
      . The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230–5.doi:10.1056/NEJM199304293281704
      OpenUrlCrossRefPubMedWeb of Science
    2. 2.↵
      2. Young T ,
      3. Peppard PE ,
      4. Taheri S
      . Excess weight and sleep-disordered breathing. J Appl Physiol 2005;99:1592–9.doi:10.1152/japplphysiol.00587.2005
      OpenUrlCrossRefPubMedWeb of Science
    3. 3.↵
      2. Liu S ,
      3. Lin Y ,
      4. Liu X
      . Meta-analysis of association of obstructive sleep apnea with glaucoma. J Glaucoma 2016;25:1–7.doi:10.1097/IJG.0000000000000357
      OpenUrl
    4. 4.↵
      2. Leong WB ,
      3. Jadhakhan F ,
      4. Taheri S , et al
      . The association between obstructive sleep apnea on diabetic kidney disease: a systematic review and meta-analysis. Sleep 2016;39:301–8.doi:10.5665/sleep.5432
      OpenUrl
    5. 5.↵
      2. Xu S ,
      3. Wan Y ,
      4. Xu M , et al
      . The association between obstructive sleep apnea and metabolic syndrome: a systematic review and meta-analysis. BMC Pulm Med 2015;15:105.doi:10.1186/s12890-015-0102-3
      OpenUrl
    6. 6.↵
      2. Kim Y ,
      3. Koo YS ,
      4. Lee HY , et al
      . Can continuous positive airway pressure reduce the risk of stroke in obstructive sleep apnea patients? a systematic review and meta-analysis. PLoS One 2016;11:e0146317.doi:10.1371/journal.pone.0146317
    7. 7.↵
      2. Bratton DJ ,
      3. Gaisl T ,
      4. Wons AM , et al
      . CPAP vs mandibular advancement devices and blood pressure in patients with obstructive sleep apnea: a systematic review and meta-analysis. JAMA 2015;314:2280–93.doi:10.1001/jama.2015.16303
      OpenUrlCrossRefPubMed
    8. 8.↵
      2. Nadeem R ,
      3. Singh M ,
      4. Nida M , et al
      . Effect of CPAP treatment for obstructive sleep apnea hypopnea syndrome on lipid profile: a meta-regression analysis. J Clin Sleep Med 2014;10:1295–302.doi:10.5664/jcsm.4282
      OpenUrl
    9. 9.↵
      2. Stroup DF ,
      3. Berlin JA ,
      4. Morton SC , et al
      . Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12.
      OpenUrlCrossRefPubMedWeb of Science
    10. 10.↵
      2. Wells G ,
      3. Shea B ,
      4. O’Connell D
      . The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa Hospital Research Institute 2009.
    11. 11.↵
      2. Zheng YL ,
      3. Lian F ,
      4. Shi Q , et al
      . Alcohol intake and associated risk of major cardiovascular outcomes in women compared with men: a systematic review and meta-analysis of prospective observational studies. BMC Public Health 2015;15:773.doi:10.1186/s12889-015-2081-y
    12. 12.↵
      2. Bradley TD ,
      3. Floras JS
      . Obstructive sleep apnoea and its cardiovascular consequences. Lancet 2009;373:82–93.doi:10.1016/S0140-6736(08)61622-0
      OpenUrlCrossRefPubMedWeb of Science
    13. 13.↵
      2. DerSimonian R ,
      3. Laird N
      . Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–88.doi:10.1016/0197-2456(86)90046-2
      OpenUrlCrossRefPubMedWeb of Science
    14. 14.↵
      2. Ades AE ,
      3. Lu G ,
      4. Higgins JP
      . The interpretation of random-effects meta-analysis in decision models. Med Decis Making 2005;25:646–54.doi:10.1177/0272989X05282643
      OpenUrlCrossRefPubMedWeb of Science
    15. 15.↵
      2. Huxley RR ,
      3. Woodward M
      . Cigarette smoking as a risk factor for coronary heart disease in women compared with men: a systematic review and meta-analysis of prospective cohort studies. Lancet 2011;378:1297–305.doi:10.1016/S0140-6736(11)60781-2
      OpenUrlCrossRefPubMedWeb of Science
    16. 16.↵
      2. Deeks JJ ,
      3. Higgins JP ,
      4. Altman DG
      . Analysing data and undertaking meta‐analyses: John Wiley & Sons, Ltd, 2008.
    17. 17.↵
      2. Higgins JP ,
      3. Thompson SG ,
      4. Deeks JJ , et al
      . Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60.doi:10.1136/bmj.327.7414.557
      OpenUrlFREE Full Text
    18. 18.↵
      2. Altman DG ,
      3. Bland JM
      . Interaction revisited: the difference between two estimates. BMJ 2003;326:219.doi:10.1136/bmj.326.7382.219
      OpenUrlFREE Full Text
    19. 19.↵
      2. Egger M ,
      3. Davey Smith G ,
      4. Schneider M , et al
      . Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34.doi:10.1136/bmj.315.7109.629
      OpenUrlAbstract/FREE Full Text
    20. 20.↵
      2. Begg CB ,
      3. Mazumdar M
      . Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088–101.doi:10.2307/2533446
      OpenUrlCrossRefPubMedWeb of Science
    21. 21.↵
      2. Mooe T ,
      3. Franklin KA ,
      4. Wiklund U , et al
      . Sleep-disordered breathing and myocardial ischemia in patients with coronary artery disease. Chest 2000;117:1597–602.doi:10.1378/chest.117.6.1597
      OpenUrlCrossRefPubMedWeb of Science
    22. 22.↵
      2. Gottlieb DJ ,
      3. Yenokyan G ,
      4. Newman AB , et al
      . Prospective study of obstructive sleep apnea and incident coronary heart disease and heart failure: the sleep heart health study. Circulation 2010;122:352–60.doi:10.1161/CIRCULATIONAHA.109.901801
      OpenUrlAbstract/FREE Full Text
    23. 23.↵
      2. Campos-Rodriguez F ,
      3. Martinez-Garcia MA ,
      4. de la Cruz-Moron I , et al
      . Cardiovascular mortality in women with obstructive sleep apnea with or without continuous positive airway pressure treatment: a cohort study. Ann Intern Med 2012;156:115–22.doi:10.7326/0003-4819-156-2-201201170-00006
      OpenUrlCrossRefPubMedWeb of Science
    24. 24.↵
      2. Marin JM ,
      3. Carrizo SJ ,
      4. Vicente E , et al
      . Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005;365:1046–53.doi:10.1016/S0140-6736(05)74229-X
      OpenUrlCrossRefPubMedWeb of Science
    25. 25.↵
      2. Young T ,
      3. Finn L ,
      4. Peppard PE , et al
      . Sleep disordered breathing and mortality: eighteen-year follow-up of the Wisconsin sleep cohort. Sleep 2008;31:1071–8.
      OpenUrlPubMedWeb of Science
    26. 26.↵
      2. Redline S ,
      3. Yenokyan G ,
      4. Gottlieb DJ , et al
      . Obstructive sleep apnea-hypopnea and incident stroke: the sleep heart health study. Am J Respir Crit Care Med 2010;182:269–77.doi:10.1164/rccm.200911-1746OC
      OpenUrlCrossRefPubMedWeb of Science
    27. 27.↵
      2. Arzt M ,
      3. Young T ,
      4. Finn L , et al
      . Association of sleep-disordered breathing and the occurrence of stroke. Am J Respir Crit Care Med 2005;172:1447–51.doi:10.1164/rccm.200505-702OC
      OpenUrlCrossRefPubMedWeb of Science
    28. 28.↵
      2. Punjabi NM ,
      3. Newman AB ,
      4. Young TB , et al
      . Sleep-disordered breathing and cardiovascular disease: an outcome-based definition of hypopneas. Am J Respir Crit Care Med 2008;177:1150–5.doi:10.1164/rccm.200712-1884OC
      OpenUrlCrossRefPubMedWeb of Science
    29. 29.↵
      2. Shah NA ,
      3. Yaggi HK ,
      4. Concato J , et al
      . Obstructive sleep apnea as a risk factor for coronary events or cardiovascular death. Sleep Breath 2010;14:131–6.doi:10.1007/s11325-009-0298-7
      OpenUrlCrossRefPubMedWeb of Science
    30. 30.↵
      2. Yaggi HK ,
      3. Concato J ,
      4. Kernan WN , et al
      . Obstructive sleep apnea as a risk factor for stroke and death. N Engl J Med 2005;353:2034–41.doi:10.1056/NEJMoa043104
      OpenUrlCrossRefPubMedWeb of Science
    31. 31.↵
      2. Martínez-García MA ,
      3. Soler-Cataluña JJ ,
      4. Ejarque-Martínez L , et al
      . Continuous positive airway pressure treatment reduces mortality in patients with ischemic stroke and obstructive sleep apnea: a 5-year follow-up study. Am J Respir Crit Care Med 2009;180:36–41.doi:10.1164/rccm.200808-1341OC
      OpenUrlCrossRefPubMedWeb of Science
    32. 32.↵
      2. Munoz R ,
      3. Duran-Cantolla J ,
      4. Martínez-Vila E , et al
      . Severe sleep apnea and risk of ischemic stroke in the elderly. Stroke 2006;37:2317–21.doi:10.1161/01.STR.0000236560.15735.0f
      OpenUrlAbstract/FREE Full Text
    33. 33.↵
      2. Leão S ,
      3. Conde B ,
      4. Fontes P , et al
      . Effect of Obstructive Sleep Apnea in Acute Coronary Syndrome. Am J Cardiol 2016;117:1084–7.doi:10.1016/j.amjcard.2015.12.053
      OpenUrl
    34. 34.↵
      2. Fornadi K ,
      3. Ronai KZ ,
      4. Turanyi CZ , et al
      . Sleep apnea is not associated with worse outcomes in kidney transplant recipients. Sci Rep 2014;4:6987.doi:10.1038/srep06987
      OpenUrlCrossRefPubMed
    35. 35.↵
      2. Kendzerska T ,
      3. Gershon AS ,
      4. Hawker G , et al
      . Obstructive sleep apnea and risk of cardiovascular events and all-cause mortality: a decade-long historical cohort study. PLoS Med 2014;11:e1001599.doi:10.1371/journal.pmed.1001599
      OpenUrlCrossRefPubMed
    36. 36.↵
      2. Won CH ,
      3. Chun HJ ,
      4. Chandra SM , et al
      . Severe obstructive sleep apnea increases mortality in patients with ischemic heart disease and myocardial injury. Sleep Breath 2013;17:85–91.doi:10.1007/s11325-012-0653-y
      OpenUrlCrossRefPubMedWeb of Science
    37. 37.↵
      2. Duval S ,
      3. Tweedie R
      . A nonparametric “Trim and Fill” method of accounting for publication bias in meta-analysis. J Am Stat Assoc 2000;95:89–98.doi:10.1080/01621459.2000.10473905
      OpenUrlCrossRefWeb of Science
    38. 38.↵
      2. Loke YK ,
      3. Brown JW ,
      4. Kwok CS , et al
      . Association of obstructive sleep apnea with risk of serious cardiovascular events: a systematic review and meta-analysis. Circ Cardiovasc Qual Outcomes 2012;5:720–8.doi:10.1161/CIRCOUTCOMES.111.964783
      OpenUrlAbstract/FREE Full Text
    39. 39.↵
      2. Dong JY ,
      3. Zhang YH ,
      4. Qin LQ
      . Obstructive sleep apnea and cardiovascular risk: meta-analysis of prospective cohort studies. Atherosclerosis 2013;229:489–95.doi:10.1016/j.atherosclerosis.2013.04.026
      OpenUrlCrossRefPubMed
    40. 40.↵
      2. Ge X ,
      3. Han F ,
      4. Huang Y , et al
      . Is obstructive sleep apnea associated with cardiovascular and all-cause mortality? PLoS One 2013;8:e69432.doi:10.1371/journal.pone.0069432
    41. 41.↵
      2. Wang X ,
      3. Ouyang Y ,
      4. Wang Z , et al
      . Obstructive sleep apnea and risk of cardiovascular disease and all-cause mortality: a meta-analysis of prospective cohort studies. Int J Cardiol 2013;169:207–14.doi:10.1016/j.ijcard.2013.08.088
      OpenUrlCrossRefPubMed
    42. 42.↵
      2. Xie W ,
      3. Zheng F ,
      4. Song X
      . Obstructive sleep apnea and serious adverse outcomes in patients with cardiovascular or cerebrovascular disease: a PRISMA-compliant systematic review and meta-analysis. Medicine 2014;93:e336.doi:10.1097/MD.0000000000000336
    43. 43.↵
      2. Wang X ,
      3. Bi Y ,
      4. Zhang Q , et al
      . Obstructive sleep apnoea and the risk of type 2 diabetes: a meta-analysis of prospective cohort studies. Respirology 2013;18:140–6.doi:10.1111/j.1440-1843.2012.02267.x
      OpenUrlCrossRefPubMed

    Footnotes

    • Contributors CX carried out the studies, participated in collecting data and drafted the manuscript. RZ performed the statistical analysis and participated in its design. YT and KW helped to draft the manuscript. All authors read and approved the final manuscript.

    • Funding This research was supported by the National Basic Research Program of China (nos. 973 Program 2015CB856405 and 2012CB720704) and the National Natural Science Foundation of China (nos. 31571149, 91432301, 81301176 and 81171273).

    • Competing interests None declared.

    • Patient consent Obtained.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement No additional data available.