ObjectiveTo observe the clinical effect of microincision vitreoretinal surgery (VRS) assisted with intravitreal injection of ranibizumab (IVR) in severe proliferative diabetic retinopathy (PDR) treatment. MethodsThis is a prospective non-randomized controlled clinical study. A total of 60 patients (70 eyes) with severe PDR diagnosed were enrolled and divided into IVR group (31 patients, 35 eyes) and control group (29 patients, 35 eyes). IVR group patients received an intravitreal injection of 0.05 ml ranibizumab solution (10 mg/ml) first, and 3 or 4 days later they received 23G microincision VRS. Control group patients only received 23G microincision VRS. The follow-up time was 3 to 12 months with an average of (4.5±1.8) months. The logarithm of the minimal angle of resolution (logMAR) best corrected visual acuity (BCVA), intraocular pressure, the central retinal thickness (CRT) and retinal reattachment, and the incidence of postoperative complications were comparatively analyzed. ResultsThere was no topical and systemic adverse reactions associated with the drug after injection in IVR group. The incidence of post-operative vitreous hemorrhage (VH) in IVR group and control group was 8.6% and 28.6% at 1 week after surgery, 0.0% and 17.1% at 1 month after surgery, 0.0% and 8.6% at 3 month after surgery respectively. The differences were statistically significant for 1 week (χ2=4.63, P < 0.05) and 1 month (χ2=4.56, P < 0.05), but was not statistically significant for 3 months (χ2=0.24, P > 0.05). The mean post-operative logMAR BCVA of IVR group (0.81±0.40) and control group (1.05±0.42) have all improved than their pre-operative BCVA, the difference was statistically significant (t=12.78, 4.39; P < 0.05). The mean logMAR BCVA of IVR group is higher than BCVA of control group, the difference was statistically significant (t=-2.36, P < 0.05). The average post-operative CRT in IVR group was thinner than that of control group, the difference was statistically significant (t=-2.53, P < 0.05). The incidence of a transient high intraocular pressure in IVR group (14.3%) was lower than that in control group (34.3%), the difference was statistically significant (t=4.79, P < 0.05). The incidence of retinal reattachment (t=0.35), epiretinal membrane (χ2=0.97), neovascular glaucoma (χ2=0.51) was no difference between these two groups (P > 0.05). ConclusionThe minimally invasive VRS assisted by IVR treatment for severe PDR can effectively prevent postoperative VH, reduce CRT and improve visual acuity.