PLoS ONE (Jan 2014)

Diabetes mellitus and risk of age-related macular degeneration: a systematic review and meta-analysis.

  • Xue Chen,
  • Shi Song Rong,
  • Qihua Xu,
  • Fang Yao Tang,
  • Yuan Liu,
  • Hong Gu,
  • Pancy O S Tam,
  • Li Jia Chen,
  • Mårten E Brelén,
  • Chi Pui Pang,
  • Chen Zhao

DOI
https://doi.org/10.1371/journal.pone.0108196
Journal volume & issue
Vol. 9, no. 9
p. e108196

Abstract

Read online

Age-related macular degeneration (AMD) is a major cause of severe vision loss in elderly people. Diabetes mellitus is a common endocrine disorder with serious consequences, and diabetic retinopathy (DR) is the main ophthalmic complication. DR and AMD are different diseases and we seek to explore the relationship between diabetes and AMD. MEDLINE, EMBASE, and the Cochrane Library were searched for potentially eligible studies. Studies based on longitudinal cohort, cross-sectional, and case-control associations, reporting evaluation data of diabetes as an independent factor for AMD were included. Reports of relative risks (RRs), hazard ratios (HRs), odds ratio (ORs), or evaluation data of diabetes as an independent factor for AMD were included. Review Manager and STATA were used for the meta-analysis. Twenty four articles involving 27 study populations were included for meta-analysis. In 7 cohort studies, diabetes was shown to be a risk factor for AMD (OR, 1.05; 95% CI, 1.00-1.14). Results of 9 cross-sectional studies revealed consistent association of diabetes with AMD (OR, 1.21; 95% CI, 1.00-1.45), especially for late AMD (OR, 1.48; 95% CI, 1.44-1.51). Similar association was also detected for AMD (OR, 1.29; 95% CI, 1.13-1.49) and late AMD (OR, 1.16; 95% CI, 1.11-1.21) in 11 case-control studies. The pooled ORs for risk of neovascular AMD (nAMD) were 1.10 (95% CI, 0.96-1.26), 1.48 (95% CI, 1.44-1.51), and 1.15 (95% CI, 1.11-1.21) from cohort, cross-sectional and case-control studies, respectively. No obvious divergence existed among different ethnic groups. Therefore, we find diabetes a risk factor for AMD, stronger for late AMD than earlier stages. However, most of the included studies only adjusted for age and sex; we thus cannot rule out confounding as a potential explanation for the association. More well-designed prospective cohort studies are still warranted to further examine the association.