Subjects receiving 16% C3F8 gas tamponade during vitrectomy will have better visual acuity and similar postoperative complications to subjects who receive silicone oil tamponade during vitrectomy.

Severe vision loss in patients with proliferative diabetic retinopathy (PDR) frequently results from complications related to neovascularization and fibrovascular proliferation. Patients with PDR are typically considered candidates for pars plana vitrectomy (PPV) when non-clearing vitreous hemorrhaging, tractional retinal detachment (TRD) development with or without rhegmatogenous retinal detachment (RRD) or extensive fibrovascular proliferation occur. Visual prognosis is often guarded in patients with PDR undergoing PPV with higher grades of vitreoretinal (VR) adhesion because of the high rate of both intra-operative and postoperative complications. Releasing all significant traction caused by VR adhesion during PPV is critical for achieving long-term anatomic success and obtaining better visual outcomes in these patients with more advanced disease. Failure to relieve VR adhesion often leads to persistent, recurrent or de novo TRD with or without RRD in the perioperative and postoperative period, which then results in one or more reoperations and ultimately a poor visual prognosis. Maneuvers to remove VR adhesion during PPV are challenging and can result in posterior and/or peripheral retinal holes, which then may lead to postoperative proliferative vitreoretinopathy (PVR) and recurrent or de novo RRD. Silicone oil and long-acting gas tamponade have been employed during PPV with reported success in cases with complex RRD and PVR. And even though silicone oil and long-acting gas tamponade are currently in wide use for patients with PDR and high-grade VR adhesion undergoing PPV (especially when intra-operative breaks occur), little clinical data is available regarding which tamponading agent is best-suited for this patient population. Previous studies have been retrospective in nature and are tarnished by selection bias when tamponading agents are compared (i.e. cases with more severe grades of VR adhesion and intra-operative breaks typically are selected to receive silicone oil tamonade over gas). Presently, there are not any randomized controlled trials reported on this topic. In this randomized controlled trial, we compare silicone oil tamponade to 16% C3F8 gas tamponade in patients with PDR undergoing PPV for the management of TRD (with or without RRD) and/or extensive fibrovascular proliferation with or without vitreous hemorrhaging.

Inclusion Criteria:

1. Subject age is 18-85 years.
2. Subject consents to study participation and is capable of 6 months of follow-up.
3. The subject has type I or II Diabetes Mellitus with active PDR in the study eye.
4. Best-corrected spectacle visual acuity (BCSVA) on the Snellen eye chart ranges from 20/40 to hand motions in the study eye.
5. The subject is determined to need a PPV because of reduced BCSVA from 1) a TRD with or without RRD and/or vitreous hemorrhaging, or 2) extensive fibrous proliferation with or without vitreous hemorrhaging. When TRD is the reason for PPV, the TRD must be threatening (within one disc diameter) or involving the fovea. When fibrovascular proliferation is the reason for PPV, it must be extensive (>3 clock hours) and threatening (within one disc diameter) or involving the fovea. The presence of subretinal fluid on B-scan ultrasonography, clinical exam, or OCT determines whether or not to classify the indication as TRD (subretinal fluid is present) or fibrous proliferation (no subretinal fluid present). Most subjects will have concurrent vitreous hemorrhaging, so it is critical that a B-scan ultrasound be performed to confirm a TRD or fibrous proliferation.
6. Only one eye per patient is eligible for the study.

Exclusion Criteria:

1. Subject is known to have a significant retinal/optic nerve disease otherwise unrelated to Diabetes Mellitus, which in the opinion of the examiner is responsible for two or more lines of reduced BCSVA (macular degeneration, optic neuritis, glaucoma, amblyopia, etc.) in the study eye.
2. Subject is known to have macular ischemia, which in the opinion of the examiner, is responsible for two or more lines of reduced BCSVA in the study eye.
3. Subject has a significant corneal opacity, which in the opinion of the examiner, is responsible for two or more lines of reduced BCSVA (corneal scar, ectasia, etc.) in the study eye.
4. Subject has been documented to have had a macula-involving retinal detachment for greater than 6 months in the study eye.
5. Subject has had a previous vitrectomy (anterior or PPV) in the study eye.
6. Subject has uncontrolled neovascular glaucoma (intraocular pressure > 30 mmHg despite medical/surgical treatment) in the study eye.
7. Subject has uncontrolled hypertension (systolic > 200 mmHg or diastolic > 120 mmHg) despite adherence to a multiple anti-hypertensive medication regimen.