BACKGROUND Advanced glycosylation end products (AGEs) are thought to play an important role in the pathophysiology of diabetes. Particularly, these products have been implicated in the pathogenesis of proliferative diabetic retinopathy. The majority of these products are formed from a vast range of precursor molecules, the variable chemical nature of which contributes to AGE heterogeneity. There is a growing population of structurally defined AGE adducts such as pyrraline, pentosidine, CML and crossline that have been found to be elevated in diabetic tissues. In the present study, the levels of the glycoxidation product pentosidine were determined in vitreous samples obtained during vitrectomy from eyes with proliferative diabetic retinopathy (PDR), proliferative vitreoretinopathy (PVR), and retinal detachment (RD). Samples from cadaveric control eyes were also included in the study. The levels of pentosidine were compared among the groups. METHODS Seventy-three vitreous samples were collected from eyes undergoing vitrectomy for PDR (n=33), PVR (n=28) and RD (n=12). Eighteen samples from cadaveric control eyes were also included in the study. A modified Bradford's method was used to assay protein content, and vitreous levels of pentosidine were determined by high-performance liquid chromatography after acid hydrolysis and pretreatment with SP-Sephadex. Statistical analyses were performed using a two-sided Mann-Whitney U test. RESULTS The levels of pentosidine [median (interquartile range)] were 0.92 (0.55-1.26) pmol/mg of protein in the PDR cases, 1.12 (0.46-1.80) pmol/mg of protein in PVR, and 1.02 (0.24-1.44) pmol/mg of protein in RD. In the cadaveric control eyes pentosidine levels were 0.97 (0.68-1.30) pmol/mg of protein. The pentosidine levels of the four groups did not differ significantly. CONCLUSIONS The levels of the glycoxidation product pentosidine (expressed as pmol/mg of protein) in the vitreous of eyes with PDR do not differ significantly from those in the vitreous of eyes with PVR, RD or cadaveric control eyes. Although these results do not refute the findings of previous studies that evaluated globally total AGE levels and the existence of diabetic vitreopathy, further investigation is needed to fully understand their relevance in this multifactorial disorder.