Background  Intravitreal ranibizumab (IVR) has become the first-line drug for treatment of retinopathy of prematurity (ROP). At present, IVR is significantly better than retinal photocoagulation (RP) in zone I ROP, and there is no obvious difference in zone Ⅱ ROP between them. However, there are few reports on the long-term effect in zone Ⅱ, especially the refractive status.

  Objective  To compare the recurrence, anatomical and refractive outcomes of intravitreal ranibizumab (IVR) with laser photocoagulation for zone Ⅱ type 1 retinopathy of prematurity (ROP) at 1-year corrected age.

  Methods  Totally 48 eyes of 24 infants with type 1 ROP in Zone Ⅱ that received IVR and 44 eyes of 22 infants that received laser photocoagulation from January 2018 to December 2019 were followed-up at the corrected age of 1 year. Another 20 infants with spontaneously regressed ROP were enrolled in the study as the control group. Fundus examination was performed at baseline and each follow-up. ROP recurrence, anatomical and refractive outcomes were assessed at 1 year corrected age.

  Results  Totally 66 infants (132 eyes, median gestation age 29.0027.00-31.00 weeks, 40 males) were followed up. There were 24 cases (48 eyes, median gestation age 28.00 IQR: 26.00-30.00 weeks, 14 males) in the IVR group, 22 cases (44 eyes, median gestation age 28.50 IQR: 28.00-31.00 weeks, 13 males) in the RP group, and 20 cases (40 eyes, median gestation age 27.50 IQR: 27.00-30.75 weeks, 13 males) in the control group with spontaneously resolved ROP. ROP recurred in 9 eyes of 5 infants (18.75%) in the IVR group and 2 eyes of 1 infant (4.54%) in the laser therapy group (P= 0.029). And after 2 or 3 injection, ROP regressed. At 1 year corrected age, the differences in axial length were not significant between the two groups. The median spherical equivalent refraction (SE) was significantly lower in eyes treated with laser (MdIQR: -0.75 -2.50, -0.75D) than the IVR group (MdIQR:-0.06-1.47, 1.00D) and control group (MdIQR:0.50-1.00, 1.19D). The frequencies of high myopia (>1.00 D) in the laser group (59.09%) was significantly higher then that in the IVR group (31.25%) and the control group (15.0%)(P<0.001). There was no significant difference in the axial length and the incidence of high myopia (< -5.00 D) among the three groups (P >0.05).

  Conclusion  Both IVR and RP are effective in the treatment of type 1 ROP in zone II. Though the recurrence rate is high after IVR treatment, reinjection is still effective. At the same time, the incidence of myopia and astigmatism corrected by IVR at 52 weeks after birth is low, but the long-term safety needs to be further evaluated.