Vitreal interleukin (IL)-1beta (IL-1beta), IL-6, IL-8, tumor necrosis factor-alpha (TNF-alpha) and nitric oxide (NO) levels have previously been determined in patients with proliferative diabetic retinopathy (PDR). However, at present there is no cohort study linking serum levels of NO and many inflammatory cytokines such as TNF-alpha, IL-1beta, soluble IL-2 receptor (sIL-2R), IL-6 and IL-8 to the grade of the microvascular complications. To determine the relation between the stages of DR and the levels of serum NO, TNF-alpha, IL-1beta, sIL-2R, IL-6 and chemokine IL-8 in patients with diabetes compared with healthy controls. Fifty-three consecutive patients with diabetes (25 men, 28 women) with or without DR and 15 non-diabetic healthy subjects (seven men, eight women) as controls were included in this prospective study. As an indicator for NO, serum total nitrite (NO2- + NO3-) levels (end-product of NO) were measured by the Griess reaction. Serum TNF-alpha, IL-1beta, sIL-2R, IL-6 and IL-8 levels were determined by a spectrophotometric technique using an Immulite chemiluminescent immunometric assay. The patients with diabetes were classified into three groups according to the stage of DR: no DR (NDR; n = 16), non-proliferative DR (NPDR; n = 18) and PDR (n = 19). The data were analysed using a Mann-Whitney U-test and the results were expressed as mean +/- SE (range). The levels of IL-1beta and IL-6 were below the detection limits of the assay (for each, <5.0 pg/ml) in all patients with diabetes and controls. Soluble IL-2R levels ranged from 260 to 958 U/ml, with the highest values observed in the patients with PDR. In 47 of the 53 samples (89%) tested for diabetic patients, IL-8 levels were above the detection limits of the assay (5.0 pg/ml). IL-8 levels ranged from <5.0 to 25.0 pg/ml, with the highest mean values observed in PDR patients. TNF-alpha was detectable in 46 of 53 patients with diabetes (87%), ranging from <4.0 to 26.4 pg/ml, with again the highest values obtained in the patients with PDR. Serum NO levels ranged from 80 to 188 micromol/l, with the highest values obtained in patients with PDR. Taken together, the mean serum NO, sIL-2R, IL-8 and TNF-alpha levels increased with the stage of DR and the highest levels were found in patients with PDR. The PDR patients had significantly (for each, P < 0.001) higher serum NO (166.8 +/- 3.2 micromol/l), sIL-2R (807.9 +/- 33.3 U/ml), IL-8 (17.9 +/- 0.4 pg/ml) and TNF-alpha (15.0 +/- 0.8 pg/ml) levels compared with NPDR patients (149.5 +/- 2.1, 659.4 +/- 23.4, 12.9 +/- 1.1, 11.5 +/- 0.6, respectively), NDR patients (115.9 +/- 5.8, 373.8 +/- 15.0, 8.3 +/- 1.0, 6.6 +/- 0.9, respectively) and controls (116.6 +/- 2.3, 392.4 +/- 16.6, 7.2 +/- 0.3, 7.3 +/- 0.5, respectively). Serum levels of these parameters for NPDR patients were also significantly (for each, P < 0.01) higher compared with those of NDR patients and controls. On the other hand, serum NO, sIL-2R, IL-8 and TNF-alpha levels of patients with NDR were comparable with those of controls (for each, P > 0.05). The results of the present study suggest that NO, sIL-2R, IL-8 and TNF-alpha may play important roles in the pathophysiology and progression of DR. We think that these potentially inflammatory cytokines and NO with their endothelial implications may act together during the course and progression of DR. These molecules may serve as therapeutic targets for the treatment and/or prevention of diabetes with its systemic and ocular microvascular complications.