Purpose To compare the effectiveness of intravitreal ziv-aflibercept (IVZ) to intravitreal bevacizumab (IVB) administered preoperatively to patients undergoing pars plana vitrectomy (PPV) for severe manifestations of proliferative diabetic retinopathy (PDR).

Design Randomised clinical trial (RCT).

Methods Two hundred and six patients with PDR-related complications requiring PPV were randomised into one of two treatment groups: Group A received IVZ (1.25 mg/0.05 mL) 1–10 days before PPV, while Group B received IVB (1.25 mg/0.05 mL) 1–10 days before PPV. The primary outcome was best-corrected visual acuity (BCVA) at 6 months follow-up. Secondary outcome measures were perioperative tractional retinal detachment (TRD) rates, surgical times, intraoperative and postoperative complications and incidence of unplanned PPV during the 6 month study interval.

Results One hundred and seventy three subjects underwent PPV and completed the 6-month follow-up interval. Group A subjects had better BCVA at 6 months (p=0.0035), shorter surgical times (p=0.0013) and were less likely to have a recurrence of vitreous haemorrhaging in the postoperative period (p=0.0101) when compared with subjects in Group B. There were no significant differences among the treatment groups with regards to baseline characteristics, perioperative TRD development, intraoperative complications and incidence of unplanned PPV during the 6 month study interval.

Conclusions This RCT demonstrated better final visual outcomes, shorter operating times and less vitreous haemorrhage recurrences in the postoperative period when subjects received IVZ compared to IVB prior to PPV for the treatment of PDR-related complications.