BACKGROUND Intravitreal silicone oil droplets have been found in the vitreous. The aim of this study is to compare the rates of silicone oil released by different brands of commonly used syringes for intravitreal injection after agitation by flicking. METHODS Three models of two brands of syringes were analyzed for their rates of silicone oil release: Saldanha Rodrigues (SR) 1 mL insulin syringe (SR, Brazil, syringe 1), Becton-Dickinson (BD) Plastipak 1 mL insulin syringe (Brazil, syringe 2), and BD Safety-Glide 1 mL insulin syringe (USA, syringe 3). All syringes were tested under four different conditions: positive control (fluid with addition of silicone oil) without agitation (group 1, n = 5); positive control with agitation (group 2, n = 3); fluid only without agitation (group 3, n = 5); and fluid only with agitation (group 4, n = 5). Masked graders performed all analyses using light microscopy. RESULTS All syringes (1, 2, and 3) released silicone oil droplets in the positive control group regardless of the agitation status (groups 1 and 2). When no oil was added and the syringes were not agitated, only syringe 1 released silicone oil droplets (40% of samples). After agitation, syringes 1 and 3 released silicone oil droplets in all samples. Quantitative analysis showed a significantly (P = 0.011; 11.2 ± 2.9 vs. 0.6 ± 0.9, respectively) higher mean number of silicone oil droplets released by syringe 1 after agitation compared to no agitation. Syringe 1 also had significantly (P = 0.002, 11.2 ± 2.9 vs. 0.0 ± 0.0 vs. 2.2 ± 0.8, respectively) more droplets than syringes 2 and 3 after agitation. CONCLUSIONS Syringes commonly used for intravitreal injections frequently release silicone oil droplets when agitated by flicking, especially the SR insulin ones. We recommend that they not be agitated at the time of intravitreal injection and that the manufacturers consider producing syringes adapted for intraocular use.