摘要： To determine the rates and risk factors of recurrent retinopathy of prematurity (ROP) treated by laser photocoagulation, intravitreal bevacizumab (IVB) monotherapy, or intravitreal ranibizumab (IVR) monotherapy. In this retrospective cohort study, consecutive infants with Type 1 ROP who received laser, IVB, or IVR treatments were followed for at least 75 weeks of postmenstrual age. Data analysis was performed between March 2010 and February 2017 in Chang Gung Memorial Hospital, Linkou, Taiwan. A total of 176 infants (340 eyes) were included in this study. The mean follow-up was 197.3 ± 110 weeks. All of the baseline demographic and ROP characteristics among the laser, IVB, and IVR groups were similar. The overall recurrence rate after treatment was 44 of 340 eyes (12.9%). The IVB group had a recurrence rate of 10.0%, followed by the laser group (18.0%) and the IVR group (20.8%); however, these rates were not signi?cantly different (P = 0.0528). Compared with the laser group, the IVB and IVR groups exhibited recurrence at later ages (43.4 ± 3.5 weeks for the IVB group, 42.3 ± 2.0 weeks for the IVR group, and 39.5 ± 2.8 weeks for the laser group; P = 0.0058). The mean interval of recurrence from initial treatment in the laser group was 3.6 ± 1.4 weeks compared with 8.8 ± 3.9 weeks and 8.3 ± 1.6 weeks in the IVB and IVR groups, respectively (P = 0.0001). Overall, the independent risk factors of recurrence included an early postmenstrual age at initial treatment (P = 0.0160), Zone I (P = 0.0007), low Apgar score (P = 0.0297), and multiple births (P = 0.0285). There was no signi?cant difference in progression to retinal detachment among the three groups (laser: 3/61, 4.9%; IVB: 2/231, 0.9%;and IVR: 1/48, 2.1%; P = 0.2701). Laser, IVR, and IVB are effective for Type 1 ROP. Retinopathy of prematurity recurrence requiring re-treatment was encountered as late as 50 weeks of postmenstrual age after IVB or IVR but earlier after laser. Longer follow-up for infants treated with anti–vascular endothelial growth factor is needed, especially in patients with signi?cant risk factors such as an early postmenstrual age at initial treatment, Zone I ROP, low Apgar score, and multiple births.