Silicone oil (SO) still represents the main choice for long‐term intraocular tamponade in complicated vitreoretinal surgery. This review compared the complications associated with the use of SO and other vitreous substitutes after pars plana vitrectomy in patients with different underlying diseases. Meta‐analysis was conducted in accordance with PRISMA guidelines. We retrieved randomized clinical trials (RCTs), retrospective case–control and cohort studies evaluating the risk of using SO, published between 1994 and 2020, conducting a computer‐based search of the following databases: PubMed, Web of Science, Scopus and Embase. Primary outcome was the rate of complications such as intraocular hypertension, retinal re‐detachment, unexpected vision loss or hypotony. Secondary outcome was to compare the rate of adverse events of different SO viscosities, especially emulsification. Forty‐three articles were included. There were significant differences in intraocular hypertension (p = 0.0002, OR = 1.66; 95% CI = 1.27–2.18) and the rate of retinal re‐detachment (p < 0.0009, OR = 0.65; 95% CI = 0.50–0.64) between SO and other agents, including placebo. However, there were no differences in other complication rates. Silicone oil (SO)‐emulsification rate was non‐significantly higher in low than high SO viscosity, and results from other complications were comparable in both groups. The high quality of most of the studies included in this study is noteworthy, which provides some certainty to the conclusions. Among them is the high variability of the SO residence time. The fact that ocular hypertension and not hypotension is related to SO use. A clear relationship is not found for the so‐called unexplained vision loss, which affects a significant percentage of eyes. Re‐detachment cases are less if SO is used and that surprisingly there does not seem to be a relationship in the percentage of emulsification between the low‐ and high‐viscosity silicones. All these data warrant more standardized prospective studies.