• Retina/Vitreous

    Investigators evaluated the optimal dose of preoperative intravitreal bevacizumab in patients undergoing vitrectomy for proliferative diabetic retinopathy (PDR).

    Study design

    This randomized clinical trial involved 167 patients who underwent pars plana vitrectomy (PPV) for complications of PDR. They were randomized into 1 of 3 treatment groups: group A received 0.625 mg IVB (0.025 ml), group B received 1.25 mg IVB (0.05 ml) and group C received 2.5 mg IVB (0.1 ml). All bevacizumab treatments were administered 1 to 10 days prior to PPV.

    Outcomes

    At the 6-month follow-up, there were no significant differences in BCVA, intraoperative complications, postoperative complications or unplanned PPV rates between groups.

    There were no patients in group A (0.0%), 3 patients in group B (7.0%) and 5 patients in group C (8.5%) who demonstrated perioperative tractional retinal detachment during the period between bevacizumab administration and PPV (P=0.0283). This difference was significant between groups A and B (P=0.0494) and between groups A and C (P=0.0080).

    Limitations

    Investigators should have included a control group receiving no bevacizumab before PPV. A study with larger numbers would have been more conclusive regarding the outcome. Careful preoperative documentation of active neo-vascularization of the fibrous membranes would have been necessary, as old inactive fibrous proliferation will usually not contract with anti-VEGF medications.

    Clinical significance

    It is a common scenario in a retina practice to see patients with tractional retinal detachments from PDR. There have been reports of “crunch” phenomenon after intravitreal anti-VEGF medications in these cases. These findings show that giving a half the dose is a good option to reduce the risk of bleeding during vitrectomy. The half dose could be also used in patients with traction that does not involve the macula, where the injection could be administered a few days before initiation of pan retinal photocoagulation.