Timing of Preoperative Anti-VEGF for Vitrectomy in Proliferative Diabetic Retinopathy: A Competing-Risks Analysis of a Large Cohort

Summary

Preop anti-VEGF for PDR PPV: Fewer post-op VH, but no reduced TRD risk. Best results with injection 4-7 days pre-op. Retrospective, 608 eyes.

Abstract

Objective: To evaluate whether timing of preoperative anti-VEGF therapy given before vitrectomy for proliferative diabetic retinopathy (PDR) influences risks of postoperative vitreous hemorrhage (VH) and tractional retinal detachment (TRD).

Design: Single-center, retrospective cohort study.

Participants: A total of 608 eyes of 608 patients undergoing PDR vitrectomy were classified into a No-Pretreatment group (n = 100) or an Anti-VEGF group (n = 508). The Anti-VEGF group was stratified by injection-to-surgery interval: 1 to 3, 4 to 7, 8 to 14, and >14 days.

Methods: Multivariable Cox proportional hazards models and Fine and Gray competing-risks models were used to analyze the time to first postoperative recurrence (defined as VH or TRD requiring intervention).

Main outcome measures: The incidence and independent predictors of postoperative VH and TRD.

Results: In this large clinical cohort, preoperative anti-VEGF preceding vitrectomy for PDR was associated with fewer postoperative VH events but did not reduce TRD risk. An injection-to-surgery interval of 4 to 7 days had the lowest overall postoperative recurrence (VH + TRD), whereas delays beyond 14 days were associated with highest risk of TRD. Visual acuity outcomes did not differ significantly with injection timing. These findings highlight the importance of considering injection timing in surgical planning.

Conclusions: In this large clincal cohort, the association between preoperative anti-VEGF and postoperative outcomes in PDR vitrectomy was complication-specific: the therapy was associated with fewer postoperative VH events but was not associated with reduced TRD risk. The 4 to 7-day interval after injection was associated with the lowest rate of postoperative recurrence, whereas delays beyond 14 days were associated with an increased risk of TRD. These findings highlight the importance of considering injection timing in surgical planning, while acknowledging that visual acuity outcomes did not differ significantly among groups.