Purpose: As advanced glycation endproducts (AGEs) and the vitreous body are both considered to be involved in the development of diabetic retinopathy and hydroimidazolone is one of the most prominent AGEs, we compared vitreous and serum hydroimidazolone in diabetes patients with proliferative diabetic retinopathy (PDR) to levels in non-diabetes controls, co-measuring vitreous albumin and vascular endothelial growth factor (VEGF).
Methods: In a cross-sectional case-control study design, we used immunoassays to compare vitreous and serum hydroimidazolone and VEGF levels in 23 consecutive type 2 diabetes mellitus patients with a median known diabetes duration of 12 years (range 1-38 years) with those in 32 non-diabetes age-matched controls undergoing vitrectomy.
Results: Vitreous hydroimidazolone was increased in the PDR group (median 1.3 U/ml, range 0.5-3.3 U/ml) compared with controls (median 0.8 U/ml, 0.5-2.5 U/ml) (p = 0.026). Hydroimidazolone levels in serum and vitreous correlated (r = 0.49, p = 0.019). In PDR, vitreous VEGF levels were increased (median 1600 pg/ml, range 20-14 700 pg/ml) compared with controls (median 7 pg/ml, range 2-500 pg/ml) (p < 0.001). Similarly, vitreous albumin concentration was increased in PDR (median 1.6 g/l, range 0.7-3.0 g/l) compared with controls (median 0.3 g/l, range 0.08-1.9 g/l) (p < 0.001). Albumin could not explain the differences in vitreous VEGF levels in a logistic regression analysis. No correlation was found between vitreous levels of VEGF and hydroimidazolone (r = 0.12, p = 0.59).
Conclusions: Increased vitreous hydroimidazolone is associated with diabetic retinopathy, possibly due to a blood-retinal barrier breakdown. Irrespective of origin, it may add to the ocular damage and needs further causal investigation. Increased VEGF in diabetic vitreous is probably of intraocular origin. It is not associated with vitreous hydroimidazolone.