OBJECTIVE

To investigate vitreous levels of reactive oxygen species (ROS) in patients with proliferative diabetic retinopathy (PDR) and analyze ROS levels among different groups of PDR patients.

DESIGN

Retrospective case-control study.

PARTICIPANTS

Thirty-nine eyes of 39 patients with PDR and 16 eyes of 16 non-PDR patients (control group) that underwent primary vitrectomy for complications of PDR and other conditions (control group), with a follow-up time > or = 12 months.

METHODS

Proliferative diabetic retinopathy patients were classified into 3 groups according to the extent of fibrovascular proliferation: (1) no or focal adhesions at < or =3 sites (n = 17); (2) > or =1 broad adhesions or vitreous-retinal adhesions around disc, macula, and arcade (n = 12); and (3) vitreous-retinal attachment extending to the periphery or no posterior vitreous detachment with or without retinal detachment (RD) (n = 10). The control group (n = 16) contained non-PDR patients. Vitreous samples were obtained during measurement of vitrectomy and vitreous levels of ROS by luminol-enhanced chemiluminescence assay.

MAIN OUTCOME MEASURES

Reactive oxygen species levels were recorded as mean (+/- standard deviation) chemiluminescence counts per 10 seconds. Correlations of vitreous levels of ROS among the 3 PDR groups and anatomic prognosis were evaluated. Multiple linear regression analysis of selective potential risk factors was performed to investigate the main determinants of ROS levels.

RESULTS

Vitreous ROS levels were significantly higher in patients with PDR (125.76+/-351.72 chemiluminescence counts per 10 seconds) than in control patients (0.37+/-0.72 chemiluminescence counts per 10 seconds; P<0.0001). Reactive oxygen species levels were 1.86+/-1.63 (group 1), 24.47+/-22.68 (group 2), and 457.94+/-597.01 (group 3); the difference among groups was significant (P = 0.001). Regression analysis indicated that only patient grouping (according to the severity of fibrovascular proliferation) had a strong dependent association with ROS levels (P = 0.001). Final anatomic results revealed that recurrent untreatable RD occurred in 3 patients of group 3, who also had the highest ROS levels.

CONCLUSIONS

Reactive oxygen species levels were significantly elevated in the vitreous fluid of PDR patients, and patients with a more advanced clinical PDR appearance had higher ROS levels. These findings suggest an association between ROS and the pathogenesis of PDR. Reactive oxygen species might be correlated with PDR severity.