Refractive error after intravitreal bevazicumab for threshold disease in retinopathy of prematurity: Two‐years follow‐up

Abstract

AbstractPurpose To examine the association between myopic refractive error at two years of life in infants who postnatally received an intravitreal bevacizumab injection or standard laser therapy for threshold retinopathy of prematurity (ROP) in fundus zone I or zone II.Methods In the retrospective non‐randomized interventional comparative study, infants who consecutively received a single intravitreal bevacizumab (0.375 mg) injection (study group) were compared with infants who had previously undergone standardized retinal argon laser therapy (control group). The follow‐up examination included cylcoplegic refractometry.Results The study group consisted of 7 children (13 eyes) and the control group included 14 children (27 eyes). Both groups did not differ significantly in gestational age and follow‐up. At the end of follow‐up of 24 months, refractive error (median:+1.00 diopters (D) (+4.0D to ‐10.00D) versus median:‐7.50D (+2.50D to ‐15.0D);P=0.001)), astigmatism (median:0.25D (0D to 2D) versus median:1.25 D (0D to 5.75D);P=0.001)) and prevalence of high myopia (≤‐6D) (15±10% versus 52±10%;P=0.04) were significantly less in the study group than control group. After adjusting for gestational age, follow‐up time, gender and birth weight, refractive error (P=0.007), astigmatism (P=0.01) and prevalence of high myopia (P=0.03; Odds Ratio:0.13) were significantly associated only with laser therapy versus bevacizumab therapy.Conclusion A single intravitreal bevacizumab injection as compared to conventional retinal laser coagulation was associated with a lower degree of myopization and less astigmatism at two year follow‐up.