AIM: To investigate the clinical efficacy of 23G /25G minimally invasive vitrectomy assisted by intravitreal injection of Ranibizumab(IVR)in the treatment of severe proliferative diabetic retinopathy(PDR).

METHODS: The respective analysis of 56 eyes of 56 patients with severe PDR treated by minimally invasive vitrectomy from June 2014 to July 2016 was performed. Thirty eyes treated by IVR-assisted minimally invasive vitrectomy as Group A and 26 eyes treated by minimally invasive vitrectomy only as Group B. IVR(0.5mg/0.05mL)was performed on the eyes 5-7d before PPV in Group A. Operative duration, hemorrhage during the operation, endodiathermy times, incidence of iatrogenic retinal holes and best corrected visual acuity(BCVA)between the two groups were observed. The follow-up period was 6mo.

RESULTS: The average operation time in Group A was 98.10 ± 14.23min, Group B was 113.89±17.66min, the difference was statistically significant(t=3.703, P=0.0005). Hemorrhage during the operation was observed in 6 eyes in Group A(20%)and 14 eyes in Group B(54%), the difference was statistically significant(χ²=6.95, P=0.0084). Endodiathermy time in Group A was significantly less than in Group B(1.30±1.15 vs 2.73±1.71; t=3.72, P=0.001). The incidence of iatrogenic retinal holes was observed in 2 eyes in Group A(7%)and 7 eyes in Group B(27%). The difference was statistically significant(χ²=4.24, P=0.0396)in the two groups. The BCVA at post-operation 1mo was significantly improved in comparison with pre-operation, but no significant difference in the post-operative BCVA between the two groups(87% vs 73%; χ²=1.630, P=0.202).

CONCLUSION: Minimally invasive vitrectomy assisted with the preoperative intravitreal injection of ranibizumab in the treatment of severe PDR can effectively shorten the operation time, lessen the incidence of hemorrhage, endodiathermy time and iatrogenic retinal holes. No significant difference in the post-operative BCVA between the two groups.