Article Text

This article has a correction. Please see:

Markers of inflammation: data from the MOSAIC randomised trial of CPAP for minimally symptomatic OSA
  1. J R Stradling1,
  2. S E Craig1,
  3. M Kohler1,2,
  4. D Nicoll1,
  5. L Ayers3,
  6. A J Nunn4,
  7. D J Bratton4
  1. 1Oxford Centre for Respiratory Medicine and Oxford Biomedical Research Centre, Churchill Hospital Campus, Oxford University Hospitals NHS Trust, Oxford, UK
  2. 2Sleep Disorders Centre and Pulmonary Division, University Hospital of Zurich, Zurich, Switzerland
  3. 3Department of Clinical Immunology, Churchill Hospital Campus, Oxford University Hospitals NHS Trust, Oxford, UK
  4. 4MRC Clinical Trials Unit at UCL, Aviation House, London, UK
  1. Correspondence to Professor John R Stradling, Oxford Centre for Respiratory Medicine and Oxford Biomedical Research Centre, Churchill Hospital Campus, Oxford University, Oxford OX3 7LJ, UK; john.stradling{at}


The Multi-centre Obstructive Sleep Apnoea Interventional Cardiovascular (MOSAIC) trial compared 6 months of CPAP therapy, versus no CPAP, in 391 patients with minimally symptomatic obstructive sleep apnoea (OSA). We now report some exploratory outcomes, markers of systemic inflammation (interleukin 6 (IL-6), IL-10, C reactive protein, tumour necrosis factor). We found no consistent changes (all p values >0.13).

Trial registration number: ISRCTN 34164388.

  • Sleep apnoea
  • Cytokine Biology

Statistics from

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.


Intermittent hypoxia may generate free oxygen radicals and oxidative stress, activate expression of adhesion molecules and proinflammatory cytokines, and in turn damage the vascular endothelium with chronic low grade inflammation, ultimately leading to atheroma formation and vascular events.1 There are very few randomised controlled treatment trials in patients with OSA, and a recent meta-analysis relied exclusively on uncontrolled and non-randomised trials for its conclusion that there were small reductions in some markers of inflammation following CPAP.1 The MOSAIC randomised controlled trial looked at the effects of 6 months CPAP in 391 patients with minimally symptomatic OSA. We have now analysed stored plasma for markers of systemic inflammation, from MOSAIC patients with sufficient samples remaining, recruited at the Oxford, Reading and Taunton centres (303 patients).


Details of the MOSAIC study have been published previously.2 ,3 The online repository for this current research letter contains details of the main entry criteria, statistical analysis and analyses performed on the blood samples: interleukin 6 (IL-6), IL-10 (anti-inflammatory), highly sensitive C reactive protein (CRP) and tumour necrosis factor. Patients with CRP values over 8 mg/L, at baseline or 6 months, were discarded from each analysis as they were assumed to represent significant intercurrent infection. As in the primary outcomes paper, the analysis was on an intention-to-treat basis, but with secondary analyses to explore the effect of CPAP compliance, as this was low overall (median 2:39, IQR 0:36 to 4:59, hour:min/night).


Patient characteristics, pre and post levels of oxygen desaturation index (ODI), and compliance data are in the original paper.2 Table 1 shows the results of the blood assays. There were no consistent changes in any of the outcomes, with all p values >0.13, and no differences were seen when comparing high and low CPAP compliers (online data supplement). Sensitivity analyses using all subjects, regardless of whether the CRP was >8 mg/l or not, were no different (data not shown).

Table 1

Baseline, 6 month, mean change and treatment effect data for all outcomes (excluding patients with CRP values >8 at baseline or follow-up)


We have shown no changes in a limited set of inflammatory markers following 6 months of CPAP therapy versus untreated controls, in patients with mild-to-moderate OSA from the MOSAIC trial. This is despite there being a clear improvement in sleepiness,2 and an improvement in endothelial function (albeit in a subset of patients).3 Low CPAP compliance is unlikely to be the explanation for these negative findings, given that there was no suggestion of an effect even in the higher CPAP compliers (≥4 h/night). Our results are in agreement with our earlier 1 month randomised controlled trial in a more severe group of patients with OSA.4 It is also in agreement with a recently published randomised controlled trial which showed no effect of 24 weeks CPAP on highly sensitive CRP in obese patients with moderate-to-severe OSA, despite a small fall in systolic blood pressure.5 The small effect on CRP, IL-6 and tumour necrosis factor seen in some uncontrolled trials is likely to be the result of regression to the mean.1 Reductions in catecholamine levels seem a more likely explanation for the improved endothelial function.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors JRS takes responsibility for the overall content as guarantor. JRS, SEC and MK were involved in the planning, conduct and reporting of the work described in this article. DN was involved in the planning and conduct of the work described in this article. LA and DJB were involved in the conduct and reporting of the work described in this article. AJN was involved in the planning and reporting of the work described in this article.

  • Funding Original MOSAIC trial: British Heart Foundation (UK)—PG/05/068, Oxford Health Services Research Committee. Current analyses—Oxford Radcliffe Charitable Funds (Registered charity no. 1057295).

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval COREC No: 05/Q1604/159.

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

  • Data sharing statement All data in this article are freely available on request.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd and British Thoracic Society