To evaluate the role of cryotherapy of sclerotomy sites in prevention of late (after 4 weeks) post-vitrectomy diabetic hemorrhage. In a prospective randomized double-masked clinical trial, a total of 124 eyes of 119 patients underwent pars plana vitrectomy for complications of proliferative diabetic retinopathy. At the end of vitrectomy, the eyes were randomly assigned to two groups. The case group received two cryotherapy spots at each sclerotomy site, whereas the eyes in the control group received no further intervention. Evaluations were repeated at months 1, 2, 4, and 6. The primary outcome measure was occurrence of vitreous hemorrhage after the first month and up to the sixth month. Ultrasound biomicroscopy (UBM) examination of the sclerotomy sites was performed concurrent with the occurrence of vitreous hemorrhage or at the last follow-up. Thirty-six eyes were excluded mostly because of early postoperative vitreous hemorrhage. Finally, the data of 42 eyes of the cases and 46 eyes of the controls were used for analysis. Late vitreous hemorrhage occurred in 17 out of 88 eyes (19.3%). This rate was significantly higher in the cases compared to the controls, 28.6% (12 eyes) versus 10.9% (five eyes) within 6 months (P = 0.036). There was no statistically significant difference in the UBM findings of the sclerotomy sites between the two groups. No relationship was noticed between morphologic patterns detected by UBM and occurrence of late vitreous hemorrhage. Cryotherapy of the sclerotomy sites at the end of the operation is not helpful for prevention of late post-vitrectomy diabetic hemorrhage, and it may even increase this risk.