Retinal detachment (RD) with proliferative vitreoretinopathy (PVR) often requires surgery to restore normal anatomy and to stabilize or improve vision. PVR usually occurs in association with recurrent RD (that is, after initial retinal re-attachment surgery), but occasionally may be associated with primary RD. Either way, for both circumstances a tamponade agent (gas or silicone oil) is needed during surgery to reduce the rate of postoperative recurrent RD. The objective of this review was to assess the relative safety and effectiveness of various tamponade agents used with surgery for RD complicated by PVR. We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register) (the Cochrane Library 2019, Issue 1), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE (January 1946 to January 2019), Embase (January 1980 to January 2019), Latin American and Caribbean Literature on Health Sciences (LILACS) (January 1982 to January 2019), the metaRegister of Controlled Trials (mRCT) (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the WHO International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We did not use any date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 2 January 2019. We included randomized controlled trials (RCTs) on participants undergoing surgery for RD associated with PVR that compared various tamponade agents. Two review authors screened the search results independently. We used the standard methodological procedures expected by Cochrane. We identified four RCTs (601 participants) that provided data for the primary and secondary outcomes. Three RCTs provided data on visual acuity, two reported on macular attachment, one on retinal reattachment and another two on adverse events such as RD, worsening visual acuity and intraocular pressure. Study Characteristics Participants' characteristics varied across studies and across intervention groups, with an age range between 21 to 89 years, and were predominantly men. The Silicone Study was conducted in the USA and consisted of two RCTs: (silicone oil versus sulfur hexafluoride (SF6) gas tamponades; 151 participants) and (silicone oil versus perfluropropane (C3F8) gas tamponades; 271 participants). The third RCT compared heavy silicone oil (a mixture of perfluorohexyloctane (F6H8) and silicone oil) with standard silicone oil (either 1000 centistokes or 5000 centistokes; 94 participants). The fourth RCT compared 1000 centistokes with 5000 centistokes silicone oil in 85 participants. We assessed most RCTs at low or unclear risk of bias for most 'Risk of bias' domains. Findings Although SF6 gas was reported to be associated with worse anatomic and visual outcomes than was silicone oil at one year (quantitative data not reported), at two years, silicone oil compared to SF6 gas showed no evidence of a difference in visual acuity (33% versus 51%; risk ratio (RR) 1.57; 95% confidence interval (CI) 0.93 to 2.66; 1 RCT, 87 participants; low-certainty evidence). At one year, another RCT comparing silicone oil and C3F8 gas found no evidence of a difference in visual acuity between the two groups (41% versus 39%; RR 0.97; 95% CI 0.73 to 1.31; 1 RCT, 264 participants; low-certainty evidence). In a third RCT, participants treated with standard silicone oil compared to those receiving heavy silicone oil also showed no evidence of a difference in the change in visual acuity at one year, measured on logMAR scale ( mean difference -0.03 logMAR; 95% CI -0.35 to 0.29; 1 RCT; 93 participants; low-certainty evidence). The fourth RCT with 5000-centistoke and 1000-centistoke comparisons did not report data on visual acuity. For macular attachment, participants treated with silicone oil may probably experience more favorable outcomes than did participants who received SF6 at both one year (quantitative data not reported) and two years (58% versus 79%; RR 1.37; 95% CI 1.01 to 1.86; 1 RCT; 87 participants; low-certainty evidence). In another RCT, silicone oil compared to C3F8 at one year found no evidence of difference in macular attachment (RR 1.00; 95% CI 0.86 to 1.15; 1 RCT, 264 participants; low-certainty evidence). One RCT that compared 5000 centistokes to 1000 centistoke reported that retinal reattachment was successful in 67 participants (78.8%) with first surgery and 79 participants (92.9%) with the second surgery, and no evidence of between-group difference (1 RCT; 85 participants; low-certainty evidence). The fourth RCT that compared standard silicone oil with heavy silicone oil did not report on macular attachment. Adverse events In one RCT (86 participants), those receiving standard 1000 centistoke silicone oil compared with those of the 5000 centistoke silicone oil showed no evidence of a difference in intraocular pressure elevation at 18 months (24% versus 22%; RR 0.90; 95% CI 0.41 to 1.94; low-certainty evidence), visually significant cataract (49% versus 64%; RR 1.30; 95% CI 0.89 to 1.89; low-certainty evidence), and incidence of retina detachment after the removal of silicone oil (RR 0.36 95% CI 0.08 to 1.67; low-certainty evidence). Another RCT that compared standard silicone oil with heavy silicone oil suggests no difference in retinal detachment at one year (25% versus 22%; RR 0.89; 95% CI 0.54 to 1.48; 1 RCT; 186 participants; low-certainty evidence). Retinal detachment was not reported in the RCTs that compared silicone oil versus SF6 and silicone oil versus to C3F8. There do not appear to be any major differences in outcomes between C3F8 and silicone oil. Silicone oil may be better than SF6 for macular attachment and other short-term outcomes. The choice of a tamponade agent should be individualized for each patient. The use of either C3F8 or standard silicone oil appears reasonable for most patients with RD associated with PVR. Heavy silicone oil, which is not available for routine clinical use in the USA, may not demonstrate evidence of superiority over standard silicone oil.