Article Text

Download PDFPDF
Factors predictive of cystoid macular oedema following endothelial keratoplasty: a single-centre review of 2233 cases
  1. James Myerscough1,2,
  2. Harry William Roberts1,
  3. Angeli Christy Yu3,4,
  4. Michael Mimouni5,
  5. Luca Furiosi3,4,
  6. Matteo Mandrioli4,
  7. Giuseppe Giannaccare6,
  8. Massimo Busin3,4
  1. 1 Department of Ophthalmology, Southend University Hospital, Southend, UK
  2. 2 Vision and Eye Research Institute, School of Medicine, Anglia Ruskin University, Cambridge, UK
  3. 3 Department of Translational Medicine, University of Ferrara, Ferrara, Italy
  4. 4 Department of Ophthalmology, Ospedali Privati Forlì "Villa Igea", Forlì, Italy
  5. 5 Ophthalmology, University of Toronto, Toronto, Ontario, Canada
  6. 6 Department of Ophthalmology, Magna Graecia University of Catanzaro, Catanzaro, Italy
  1. Correspondence to Dr Massimo Busin, Department of Morphology Surgery and Experimental Medicine, University of Ferrara, 44121 Ferrara, FE, Italy; mbusin{at}


Aims To describe the incidence of postoperative cystoid macular oedema (CMO) after endothelial keratoplasty (EK) and to identify its contributory risk factors.

Methods 2233 patients undergoing EK at Ospedali Privati Forlì ‘Villa Igea’, between January 2005 to October 2018 for Descemet stripping automated endothelial keratoplasty (DSAEK) and June 2014 to August 2018 for Descemet membrane endothelial keratoplasty (DMEK) with a minimum follow-up of 18 months were evaluated. Univariate and multivariate analyses were conducted to identify and quantify contributory risk factors. Receiver operating characteristic (ROC) curve analysis were performed to determine ideal cut-off points of continuous variables.

Results CMO was identified in 2.82% (n=63) of the cases. CMO occurred in 2.36% of DSAEK eyes and in 5.56% of DMEK eyes (p=0.001). Average onset of CMO was 4.27±6.63 months (range: 1–34 months) postoperatively. Compared with those who did not develop CMO, a higher proportion of patients in the CMO group had diabetes (24.2% vs 9.8%, p<0.001) (OR=3.16, 95% CI: 1.72 to 5.81, p<0.001), a higher proportion of patients who underwent DMEK rather than DSAEK (28.6% vs 14.1%, p=0.001) (OR=2.42, 95% CI: 1.35 to 4.33, p=0.003) and were older (70.5±10.0 vs 67.1±14.3 years, p=0.01). Using the cut-off of 67 years as identified by ROC curve analysis, subjects aged >67 years (OR=2.35, 95% CI: 1.30 to 4.26, p=0.005) were more likely to develop CMO. There were no other significant differences between the groups.

Conclusions Older age (>67 years), diabetes mellitus and DMEK have been identified as independent risk factors for postoperative CMO following EK. Close observation is necessary during the first postoperative year after EK, particularly in patients with risk factors.

  • cornea
  • treatment surgery

Data availability statement

Data are available upon reasonable request.

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.

Data availability statement

Data are available upon reasonable request.

View Full Text


  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests No support, financial or otherwise was received for this study. Massimo Busin has received (2006–2016) reimbursement of travel expenses and royalties from Moria (Antony, France). Harry Roberts has undertaken paid consultancy work for Alcon Inc (Fort Worth, TX, USA) in the past 12 months and has received honoraria from Thea Pharmaceuticals Ltd (Keele, UK). The other authors have no financial interest to disclose. No other acknowledgements are necessary.

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

Linked Articles

  • Highlights from this issue
    Frank Larkin