Objective: To compare the silicone oil removal time and rate of three different methods. Methods: This clinical study included one eye from each of 52 consecutively enrolled patients who underwent silicone oil removal at the Eye Hospital, Wenzhou Medical University from October 2016 to July 2017. The patients were categorized into three groups for intraocular silicon oil removal: manual (Group A=19), vitrectomy (Group B, n=17), and modified vitrectomy (Group C, n=16). For Group A, a 10 ml syringe was connected to the 23-gauge cannula through a short section of blood transfusion tube. The oil was removed by pulling the plunger to the end where it was held to create sufficient negative pressure. For Group B, the small side of a blood transfusion tube was connected to the cannula on one side of the eye while the other side was connected to a vitrectomy apparatus. For Group C, the expanded side of a blood transfusion tube was connected to the cannula on one side of the eye while the other side was connected to a vitrectomy apparatus. The negative pressure of Group B and Group C was provided by the vitrectomy apparatus.Silicone oil removal time and rate (volume divided by the time taken for removal) were recorded. All statistical analyses were performed using analysis of variance and Fisher exact test. Results: The mean times of silicone oil removal in Groups A, B, and C were 4.2 ± 1.4 min, 6.1 ± 1.5 min, and 5.1 ± 1.2 min
respectively (F=8.9, P < 0.001). Furthermore, the mean time for Group A was shorter than for Groups B and C in pairwise comparisons (P < 0.05 ). The silicone oil removal rate in Groups A, B, and C were 1.42 ± 0.31 ml/min, 0.94 ± 0.21 ml/min, and 1.14 ± 0.20 ml/min respectively (F=18.0, P < 0.001). Moreover,the removal rate for Group A was higher than for Groups B and C, and Group C was higher than Group B in pairwise comparisons (P < 0.05). The incidence of low intraocular pressure (≤6 mmHg) at day 1 after surgery in Groups A, B, and C were 1/19, 0/17, and 1/16 respectively (P=0.76). Recovery of all intraocular pressures was achieved by 1 week after surgery. No other severe intraoperative or postoperative complications were noted for three approaches. Conclusions: All three methods for removal of intraocular silicon oil were safe and efficient. The removal time of the manual approach was the shortest and the rate was the highest of the three approaches.