Materials and Methods: We retrospectively reviewed fi les of patients who underwent pars plana vitrectomy for isolated vitreous hemorrhage. Only patients without tractional detachment who had non-clearing vitreous hemorrhage were included in the study. All patients underwent standard 3-port 23 gauge pars plana vitrectomy. The patients who underwent pars plana vitrectomy were divided into 4 groups according to the intraocular tamponade used: fl uid (group 1), air (group 2), perfl uoropropane gas (group 3) and silicone oil (group 4). We recorded complications (such as re-hemorrhage, fi brovascular proliferation, neovascular glaucoma, retinal detachment) observed with intraocular tamponade in place and within the fi rst 1 month after tamponade effectiveness disappeared.

Results: There were 17 patients in group 1, 13 in group 2, 14 in group 3, and 13 in group 4. The age (p = 0.251), gender (p = 0.262) and diabetes duration (p = 0.112) of the patients were similar in all groups. The number of patients experienced complications was 3 in group 1 (17.6%), 3 in group 2 (23.1%), 3 in group 3 (21.4%) and 2 in group 4 (15.4% ). Complication rates were comparable in all groups (p = 0.957).

Conclusion: No tamponade is required to reduce postoperative complications when no tear is detected with successful hemostasis during the surgery in vitrectomy performed for isolated vitreous hemorrhage in patients with diabetic retinopathy. Keywords : Air, diabetic retinopathy, pars plana vitrectomy, perfl uoropropane, silicone oil