Anti-Vascular Endothelial Growth Factor Injection-to-Surgery Interval in Diabetic Vitrectomy: A Systematic Review and Meta-Analysis.

Abstract

Topic: To evaluate how outcomes of pars plana vitrectomy (PPV) for proliferative diabetic retinopathy (PDR) vary with the preoperative anti-vascular endothelial growth factor (anti-VEGF) injection-to-surgery interval.

Clinical relevance: Preoperative anti-VEGF facilitates vitrectomy in PDR, but the influence of the interval is unclear; longer intervals may increase fibrovascular contraction and tractional retinal detachment (TRD) risk.

Methods: This prospectively registered review (CRD420250651229) searched MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials to March 17, 2026 for comparative studies of PPV for PDR with preoperative anti-VEGF. The prespecified confirmatory comparison was anti-VEGF versus no anti-VEGF, with early postoperative vitreous hemorrhage (VH) as the principal outcome; interval analyses (no anti-VEGF, 1 to 3, 4 to 7, 8 to 14, and more than 14 days) were exploratory. Odds ratios (OR) and mean differences (MD) with 95% confidence intervals (CI) were pooled using random-effects models (restricted maximum likelihood, Hartung-Knapp, prediction intervals) and sensitivity models for sparse events and clustered arms. Certainty of evidence was rated separately for the anti-VEGF effect and interval effects.

Results: Thirty-three studies (3,419 eyes; 12 randomized) were included. Anti-VEGF was associated with lower early postoperative VH (OR, 0.40; 95% CI, 0.23 to 0.69), lower intraoperative VH (OR, 0.20; 95% CI, 0.10 to 0.40), fewer reoperations (OR, 0.58; 95% CI, 0.41 to 0.82), and less endodiathermy (OR, 0.21; 95% CI, 0.10 to 0.43). Final visual acuity was slightly better but heterogeneous (MD, -0.15 logarithm of the minimum angle of resolution; 95% CI, -0.29 to 0.00). Late VH, retinal detachment, silicone oil use, and TRD did not differ. Single-group proportions were lowest at 1 to 3 and 4 to 7 days, but the interval modeled continuously was not associated with either hemorrhage outcome. Certainty was moderate for early postoperative VH and low to very low otherwise.

Conclusion: Preoperative anti-VEGF is associated with less intraoperative and early postoperative hemorrhage and fewer reoperations at diabetic vitrectomy, with the lowest hemorrhage rates when injection precedes surgery by 1 to 7 days. The evidence does not identify a precise optimal interval; higher TRD and reoperation rates beyond 14 days derive from a single cohort. Interval associations are hypothesis-generating given observational, low-certainty evidence.