ABSTRACT: Proliferative diabetic retinopathy (PDR) is the most severe case of diabetic retinopathy that can cause visual impairment. This study aimed to reveal the factors correlated with better postoperative visual acuity after a long follow-up in patients who underwent vitrectomy for PDR. We retrospectively analyzed the data set including systemic findings, ocular findings, and surgical factors from registered patients who could be completely followed up for 2 or 4 years after vitrectomy. We ultimately enrolled 128 eyes from 100 patients who underwent vitrectomy for PDR between January 2008 and September 2012 and were followed up for >2 years. Among them, 91 eyes from 70 patients could be followed up for 4 years. Factors related to the postoperative visual acuity of ?20/40 and ?20/30 after 2 and 4 years were investigated by logistic regression analysis. Better postoperative visual acuity correlated with the following factors: no rubeosis iridis ([?20/40 at 2 years; odds ratio {OR}, 0.068; 95% confidence interval {CI}, 0.012-0.39; P = 0.003], [?20/30 at 2 years; OR, 0.07; 95% CI, 0.01-0.40; P = 0.03], [?20/30 at 4 years; OR, 0.078; 95% CI, 0.006-0.96; P = 0.04]), no fibrovascular membrane [(?20/40 at 2 years; OR, 0.22; 95% CI, 0.061-0.81; P = 0.02), (?20/40 at 4 years; OR, 0.26; 95% CI, 0.07-0.94; P = 0.04), (?20/30 at 4 years; OR, 0.14; 95% CI, 0.04-0.52; P = 0.004)], existing vitreous hemorrhage (?20/30 at 2 years; OR, 9.55; 95% CI, 1.03-95.27; P = 0.04), and no reoperation ([?20/40 at 4 years; OR, 0.15; 95% CI, 0.03-0.78; P = 0.02], [?20/30 at 4 years; OR, 0.06; 95% CI, 0.07-0.54; P = 0.01]). Treatment provision before disease severity and treatment without complications were associated with good postoperative visual acuity.