Article Text

Download PDFPDF

Obstructive sleep apnoea is associated with the development and progression of diabetic retinopathy, independent of conventional risk factors and novel biomarkers for diabetic retinopathy
  1. Supreeth Rudrappa1,
  2. Graham Warren2,
  3. Iskandar Idris1,3
  1. 1Department of Diabetes and Endocrinology, Sherwood Forest Hospitals Foundation Trust, Nottinghamshire, UK
  2. 2Trent Research Design Services, University of Nottingham, Nottinghamshire, UK
  3. 3School of Graduate Entry Medicine, University of Nottingham, Derby, UK
  1. Correspondence to
    Dr Iskandar Idris, School of Graduate Entry Medicine, University of Nottingham, Royal Derby Hospital, Uttoxeter Road, Derby DE12, UK; iidris{at}

Statistics from

Accumulating evidence suggests an association between obstructive sleep apnoea (OSA) and diabetic retinopathy.1 The mechanism for this is unclear but a previous study suggested that hypertension and high body mass index are strongly implicated.2 Increases in serum vascular endothelial growth factor and other biomarkers3–5 have also recently been shown to be independently predictive for retinopathy. Their role in mediating OSA induced retinopathy, independent of conventional risk factors, is not known. We therefore performed an exploratory study to corroborate the link between OSA with diabetic retinopathy and the association of novel serum biomarkers on OSA induced diabetic retinopathy.

In all, 31 male obese patients with Type 2 diabetes (body mass index 28–40; HbA1c 7.0%–12.0%) attending a hospital diabetes obesity clinic on stable medication regimen with no significant comorbidities were recruited. All subjects underwent an inhospital sleep study via the Visilab system. OSA was diagnosed based on the presence of apnea–hypopnea index (AHI) >5 events/h (mild), 15–30 events/h (moderate) and AHI ≥30 (severe). Indication for continuous positive airway pressure therapy at our unit is AHI of 5–14 events/h with excessive daytime sleepiness or any AHI>15. Assessments included clinical parameters, lifestyle, food questionnaire, Epworth Score, HbA1c levels, fasting glucose, plasma vascular endothelial growth factor, tumour necrotic factor α, monocyte chemoattractant-protein 1, matrix metalloproteinases, highly sensitive-CRP and interleukin 6 levels. Retinopathy and maculopathy gradings were obtained from the National diabetic eye screening programme record which involves digital photography of the retina and systematic image grading by optometrists and/or ophthalmologist. The study was approved by the Nottingham Research Ethics committee and all participants provided written consent.

Mean age was 54.9±11.3 years, body mass index 35.3±4.8 kg/m2, HbA1c level 9.2±2.4% and diabetes duration 11.0±8.3 years. All clinical and biochemical parameters were equally matched between OSA (n=17) and non-OSA patients (n=16) except for Epworth and the AHI scores (table 1). Retinopathy score (p=0.04) but not maculopathy score (p=0.15) was significantly worse in the OSA group. The proportion of patients with proliferative diabetic retinopathy was significantly higher in the OSA group (N=4 vs 0; χ2 4.8; p=0.01). Following stepwise multiple regression analysis to adjust for potential confounders to include novel serum biomarkers for diabetic retinopathy and inflammation, only OSA is an independent significant predictor of the total retinopathy scores (p=0.008) and remained in the regression model after stepwise forced exclusion of covariates which did not influence the prediction for diabetic retinopathy progression.

Table 1

Baseline parameters of patients recruited into the study

The present study supports recent observation between OSA and diabetic retinopathy. The new observation here is that this association persists even after adjustment for novel biomarkers for diabetic retinopathy. In addition, there appears to be an association between OSA and severity of retinopathy. Severity of OSA however did not predict total retinopathy score. Novel serum biomarkers previously linked with retinopathy progression did not appear to independently drive the association between OSA and diabetic retinopathy. OSA induced neovascularisation of diabetic retinopathy at least in this study appears to be driven by local angiogenesis (ie, cellular expression and vitreous level of inflammatory marker) rather than systemic vasculogenesis.6 The lack of association between OSA and maculopathy however is surprising, given the putative mechanism thought to explain the association between OSA and diabetic retinopathy. Some limitations of the study are the relatively small sample size, maximum and minimum oxygen saturation data on sleep study not being formally reported, neck size and exclusive Caucasian population, which limit the generalisibility of this study. Nevertheless, this study showed that OSA may be the most important independent predictor of retinopathy, but not maculopathy progression independent of conventional and novel biomarkers for retinopathy progression. Further adequately powered study is required to clarify the mechanism of OSA induced diabetic retinopathy progression.


View Abstract


  • Contributors GW assisted in the study analysis. SR participated in data interpretation and wrote the first draft of the paper. II conceived the study, coordinated the study and obtained ethical approval. All authors read, contributed towards and approved the final manuscript.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Nottingham Research Ethics Committee.

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

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.