AIM: To evaluate the clinical efficacy of heads-up 3D vision system in surgery for proliferative diabetic retinopathy with traction retinal detachment.

METHODS: We retrospectively reviewed 32 patients(38 eyes)of PDR complicated with local TRD(no traction retinal hole)who underwent 25G minimally invasive vitrectomy in our hospital from August 2018 to March 2019. The patients were divided into two groups according to the observation system during the operation. 19 eyes of 16 patients in the experimental group were operated with heads-up 3D vision system, and 19 eyes of 16 patients in the control group were operated with traditional microscope. The operation time, iatrogenic retinal hole and silicone oil injection were recorded in the two groups. The patients were followed up for at least 6mo to observe the best corrected visual acuity and the occurrence of postoperative complications.

RESULTS: In the experimental group, iatrogenic retinal hole and silicone oil injection occurred in 1 eye during operation. The retina was completely reattached after operation. One day after operation, vitreous hemorrhage occurred in 4 eyes, which was self-absorbed after 2-4wk. Intraocular hypertension occurred in 6 eyes in postoperative 2wk, which could be controlled by drug treatment. Vitreous hemorrhage occurred in 2 eyes after 6wk and the best corrected visual acuity of 6mo was more than 0.3 in 15 eyes. In the control group, iatrogenic retinal hole occurred in 4 eyes, silicone oil was injected in 5 eyes during operation. The retina was completely reattached after operation. One day after operation, vitreous hemorrhage occurred in 6 eyes, which was self-absorbed after 2-4wk. Intraocular hypertension occurred in 5 eyes in postoperative 2wk, which could be controlled by drug treatment. Vitreous hemorrhage occurred in 3 eyes after 6wk, and the best corrected visual acuity of 6mo was more than 0.3 in 14 eyes. The operation of all patients was completed successfully, and no serious complications such as endophthalmitis occurred, but the operation time of the experimental group was significantly shorter than that of the control group(37.3±4.8min vs 41.2±5.1min, P=0.020).

CONCLUSION:Application of heads-up 3D vision system in PDR combined with TRD vitrectomy can shorten the operation time and improve the operation efficiency.