Foveal avascular zone in diabetic retinopathy: quantitative vs qualitative assessment

Eye (Lond). 2005 Mar;19(3):322-6. doi: 10.1038/sj.eye.6701456.


Purpose: To describe the relations between foveal avascular zone (FAZ) size and outline in patients presenting diabetic retinopathy.

Methods: 110 high-quality fluorescein angiograms from 110 diabetics were chosen from our digital retinal image databank. Patients with significant media opacities, macular scars, macular hard exsudates, high ametropia, and associated macular pathology were excluded. Both FAZ perimeter and surface area were measured with image analysis software. FAZ outline was graded according to ETDRS report Number 11 (from 0=normal to 4=capillary outline completely destroyed). Data were compared to that of 31 healthy controls. FAZ surface in diabetics was compared to that of controls and FAZ surface was compared to FAZ grade, FAZ perimeter and retinopathy stage in diabetics. Quantitative variables were compared using the U-test of Mann-Whitney or Kruskal-Wallis test and correlations between quantitative variables were estimated with the Spearmann coefficient.

Results: All patients presented diabetic retinopathy (54 BDR, 30 PPDR, 26 PDR). FAZ size was larger in diabetics than controls (P<0.001). In diabetics, FAZ size increased with FAZ grade (P<or=0.006 except between grades 1 and 2=NS) and with retinopathy stage (P<or=0.024). As retinopathy advanced, there was a higher proportion of altered FAZ outlines (P=0.003).

Conclusions: This study confirms capillary alteration to be the cause of increase in FAZ size in diabetics and presents an alternative evaluation method of the FAZ to FAZ size measurement. No qualitative studies using the ETDRS FAZ grading scale have been performed to our knowledge.

MeSH terms

  • Adult
  • Aged
  • Capillaries / pathology
  • Diabetic Retinopathy / pathology*
  • Disease Progression
  • Female
  • Fluorescein Angiography
  • Fovea Centralis / blood supply
  • Fovea Centralis / pathology*
  • Humans
  • Image Processing, Computer-Assisted / methods
  • Male
  • Middle Aged
  • Severity of Illness Index