Purpose: To evaluate interleukin-8 (IL-8), nitric oxide (NO) and glutathione (GSH) profiles in vitreous humor and blood samples in patients with proliferative diabetic retinopathy (PDR) and in patients with proliferative vitreoretinopathy (PVR) and to compare the levels with those of controls. Patients and Methods: NO concentrations were determined by using the Greiss reaction in plasma and vitreous humor samples. GSH levels were determined in both blood and vitreous humor samples, using DTNB, a disulfide chromogen. Vitreous IL-8 were assayed by ELISA. Twenty-three patients with PDR, 18 patients with PVR and 21 cadavers as the control group were included in the study. Results: Plasma and vitreous NO levels were found to be 25.6 ± 2. 1 and 36.9 ± 3.0 μmol/l in patients with PDR, 27.0 ± 4.7 and 34.3 ± 2.9 μmol/l in patients with PVR and 17.4 ± 2.7 and 15.9 ± 1.4 μmol/l in controls, respectively. Vitreous humor and plasma NO levels did not show any statistically significant difference between PDR and PVR groups. However, the values for vitreous in both groups were significantly higher than those of controls (p < 0.0001). Although IL-8 levels in vitreous samples of patients with PDR were not significantly different (79.6 ± 9. 7 pg/ml) from those of patients with PVR (42.2 ± 7.3 pg/ml) (p = 0.06), the levels in both groups were significantly higher than those of controls (19. 0 ± 3.9 pg/ml) (p < 0.0001 and p < 0.05, respectively). Blood and vitreous GSH levels were found to be 5.3 ± 0.4 μmol/g·Hb and 0.58 ± 0.16 μmol/l in patients with PDR and 8.4 ± 0.5 μmol/g·Hb and 15.7 ± 2.2 μmol/l in patients with PVR and 12.0 ± 1.1 μmol/g·Hb and 0.26 ± 0.03 mmol/l in controls, respectively. Vitreous and blood GSH levels were significantly lower in patients with PDR compared to those with PVR (p < 0.0001 for both). Conclusion: Elevated levels of vitreous and plasma NO and vitreous IL-8 in PDR and PVR implicate a role for these parameters in the proliferation in these ocular disorders. GSH concentrations both in vitreous and blood samples of the PVR and PDR patients were much less than those observed in the control group. Lower GSH concentrations detected in PDR in comparison with those in PVR in vitreous humor and to a lesser degree in blood may play an important role in pathogenesis of new retinal vessel formation in patients with PDR. This also suggests that oxidative stress may be involved in the pathogenesis of PVR and particularly that of PDR. Copyright © 2003 S. Karger AG, Basel.