Complications After Removal of Silicone Oil

Droplets can migrate and cause damage to intraocular structures.


Indications for silicone oil (SO) use include tamponade for complex retinal detachments (RD) such as those with proliferative vitreoretinopathy (PVR), repair of giant retinal tears, repair of RDs in patients constrained to immediate air travel, or circumstances requiring more rapid postoperative visual rehabilitation (Table 1). Less commonly, silicone oil may be used during surgical treatment of endophthalmitis and management of postoperative hypotony. Silicone oil may itself be associated with complications such as glaucoma and keratopathy. However, even after oil removal, new issues may arise, often due to the pre-existing ocular disease, SO emulsification (Figure 1), or incomplete SO removal. This article aims to describe some potential issues and their treatment arising after silicone oil removal: recurrent RD, adhesion of oil droplets to the IOL, corneal decompensation, silicone oil-related glaucoma, and other issues leading to vision loss (Table 2).

Table 1: Uses of Silicone Oil at Time of Pars Plana Vitrectomy
  • Retinal detachments with proliferative vitreoretinopathy
  • Giant retinal tears
  • Severe proliferative diabetic retinopathy
  • Macular hole
  • Viral retinitis
  • Complicated pediatric retinal detachments
  • Trauma
  • Endophthalmitis

Figure 1. Emulsified silicone oil droplets seen in the posterior chamber and on the intraocular lens.

Table 2: Potential Complications After Silicone Oil Removal
  • Recurrent retinal detachment
  • Emulsification
  • Silicone oil in the anterior chamber
  • Keratopathy
  • Glaucoma
  • Chronic hypotony
  • Cataract
  • Adherence of oil droplets to an intraocular lens
  • Silicone oil invasion of the retina and optic nerve
  • Unexplained visual loss following removal


The Silicone Study established the efficacy of SO in a randomized, controlled clinical trial that compared 1000-centistoke SO to 14% C3F8 and 20% SF6 in patients with retinal detachment and grade C3 or greater PVR. The study reported that macular attachment was more frequent in eyes with SO (80%) than in eyes with SF6 (60%).1 Furthermore, eyes with SO had a higher percentage of best-corrected visual acuity (BCVA) of 5/200 or better than eyes with SF6, and eyes with SO had a lower prevalence of hypotony and keratopathy than eyes with SF6.1

The Silicone Study report number 2 included 265 eyes with rhegmatogenous RDs with advanced PVR enrolled from September 1987 to October 1990 and reported that silicone oil was equivalent to C3F8 in achieving complete posterior retinal attachment and BCVA of 5/200 or better.2 Eyes with SO had a lower prevalence of hypotony than those with C3F8 at less than 18 months, but there was an equal prevalence of hypotony among these 2 groups thereafter.2 There was an equal prevalence of keratopathy in eyes with SO and eyes with C3F8.2


The rate of recurrent RD in the Silicone Study following SO removal was high (Figure 2). In report 6 of the Silicone Study, after SO removal, 17 (20%) of the 84 eyes with attached retinas at the time of SO removal developed a recurrent RD.3 Similarly, a retrospective review from Germany of 225 patients who underwent PPV with SO tamponade and subsequent removal by 1 of 2 surgeons reported that 25.3% of patients developed recurrent RDs.4 Risk factors for RD in that review included number of prior unsuccessful RD surgeries, visual acuity before SO removal, incomplete removal of the vitreous base during the prior vitrectomies, absence of scleral buckle in eyes with PVR in which an inferior retinotomy had not been performed, and individual surgeon variability. The rate of recurrent RD was independent of the technique of SO removal and duration of SO endotamponade.4

Figure 2. Recurrent retinal detachment after silicone oil removal in a patient with retinal detachment with proliferative vitreoretinopathy.

A nonrandomized, consecutive case series study evaluated whether prophylactic laser photocoagulation at the time of SO removal decreased the rate of recurrent RD.5 Lower rates of recurrent postoperative RD occurred in patients who underwent prophylactic laser (4% of 26 eyes) than in those who did not (19% of 16 eyes).5 These differences were not statistically significant. Thus, prophylactic laser may be beneficial, but a larger randomized study would be needed to definitively answer this question.

In the era of small-gauge vitrectomy and widefield viewing with more complete vitreous base shaving, the rates of recurrent RD after SO removal may be lower than those in the Silicone Study. However, there are no large prospective comparative studies published on this issue.


Dispersion, in contrast to emulsification, is the splitting of a liquid into small bubbles. Emulsification occurs when the small bubbles are unable to coalesce with the larger bubble.6 A lipid coating forms around the smaller bubbles, preventing their re-entry into the larger bubble (Figure 3). There are multiple factors that potentially influence SO emulsification. Surfactants decrease a liquid’s surface tension within a medium, which subsequently increase the risk of emulsification. Intrinsic surfactants include serum, fibrin, and fibrinogen.6 Extrinsic surfactants include sterilization detergents present on vitrectomy tubing or the cutter or contaminants that may be on the sclerotomy cannula.6 Prior studies suggest that longer duration of SO fill,7 less complete fill,8 and use of PFO9 may also increase rates of emulsification.

Figure 3. Electron micrograph of emulsified oil droplets. Each small droplet is coated in lipid, preventing the droplets from coalescing to form a larger bubble.

The rates of emulsification among various viscosities of SO has been reported.10 Theoretically, higher viscosity SO with higher molecular weights are more resistant to deformation and therefore less likely to emulsify. One in vitro study examining the relative resistance to emulsification in SO of differing viscosity and polymer composition found emulsification was more likely in both the lower viscosity and the lower molecular weight silicone oils.11 Similar observations were seen in a randomized control trial of 86 eyes receiving either 1000-centistoke or 5000-centistoke SO, in which authors found the 1000-centistoke group had a higher frequency of oil emulsification necessitating early removal of SO.12 However, in a large retrospective study of outcomes and complications after RD repair with 1000-centistoke or 5000-centistoke SO, there was no difference in silicone oil emulsification between the 2 groups.13

Emulsified oil can adhere to an intraocular lens (IOL) or migrate into the anterior chamber (AC), causing damage to the corneal endothelium or the trabecular meshwork. While emulsification may prompt removal of SO, it is important to recognize that not all oil droplets can be removed, even with repeated fluid-air exchanges (Figure 4).

Figure 4. B-scan ultrasound of emulsified oil after silicone oil removal. Small oil droplets remained after multiple fluid-air exchanges.


Adhesion of silicone oil droplets to an IOL may occur (or become apparent) after removal of the majority of SO (Figure 5). Prior studies indicate that although SO does adhere to multiple types of IOL styles, it adheres most significantly to silicone IOLs.14 This adhesion may result in various visual disturbances in the form of decreased visual acuity, visual distortion, glare, or micropsia.

Figure 5. Silicone oil droplets adherent to a silicone intraocular lens after silicone oil removal.

Removal of these adherent droplets can be challenging and is generally reserved for patients who are very symptomatic. Simple aspiration and irrigation of the IOL may be partially effective.15 Various solvents, such as semifluorinated alkanes, have been trialed in the removal of the oil droplets. Although they may be more effective compared to simple irrigation, they are potentially toxic and require complete removal from the vitreous cavity.15 In some cases, it appears that the SO interacts with the IOL (especially silicone IOLs) to make a permanent alteration in its surface, thus resulting in a permanent optical aberration.16

In cases where patients experience significant visual disturbance resulting from oil adherence to the IOL, patients may elect to undergo IOL explantation. In a retrospective case series of patients undergoing IOL removal, results disclosed that 17 of 63 IOLs were removed as a result of oil opacification.17 Of these 17 IOLs, 15 were silicone lenses.17


The Silicone Study report 7 disclosed an incidence of keratopathy of 27% at 2 years, equal to that of C3F8.18 Risk factors for keratopathy include oil in the AC, hypotony, pseudophakia, aphakia, a postoperative aqueous flare, and reoperations. Even after the SO removal, small oil droplets may remain and may migrate into the AC when patients are in the supine position. Prolonged contact of oil droplets with the corneal endothelium may ultimately result in worsening of band or bullous keratopathy.


The overall incidence of SO-induced glaucoma is reported to be 6% to 56%.19 The variability may be due to different intraocular pressure (IOP) criteria and variable duration of follow-up information in published series. One retrospective study of secondary outcomes following vitreoretinal surgery reported an average of 6 mmHg IOP rise in the first 2-3 weeks following the first vitreoretinal procedure.20 The authors concluded that visual outcomes were significantly worse in eyes with hypotony (<6 mmHg) or elevated IOP (>21 mmHg) but that there was mild progression of the mean vertical cup-to-disc ratio even despite effective IOP control in the majority of patients.20

Risk factors for secondary glaucoma include aphakia, diabetic retinopathy, pre-existing glaucoma, and possibly lower viscosity oil.19 There are multiple proposed mechanisms for SO-induced glaucoma, including pupillary block, inflammation, synechial angle closure, rubeosis iridis, and oil migration into the AC causing mechanical blockage of the trabecular meshwork and/or trabeculitis.19

Many eyes have hypotony following repair of complex RD. Surgeons are reluctant to remove silicone oil in eyes with hypotony because the IOP may drop further after oil removal. These eyes usually have very poor functional vision.


Visual loss may be associated with SO tamponade and its removal. In a study of 9 patients over a 10-year period with 5000-centistoke SO, the median BCVA with the oil in place was 20/40. The median BCVA after SO removal was 20/120.21 The mechanism of this visual loss remains unknown, although there are multiple hypotheses. One is the concept of a potassium sink, in which the formed vitreous is a normal reservoir for potassium released by Muller cells. This is blocked by SO and may lead to a potassium buildup.21 Elevated intraretinal potassium may cause apoptosis and macular dysfunction.22 Ultraviolet light damage has also been postulated as a possible mechanism, because eyes with SO may be more susceptible to phototoxicity due to higher transmission of certain frequencies of UV light through SO than vitreous gel.23 Additionally, SO may dissolve lutein and zeaxanthin, which are thought to have photoprotective roles.24 Lastly, ocular factors such as cystoid macular edema, redetachment, and epiretinal membrane may contribute to vision loss associated with the use and removal of SO. Prior studies indicate that there are higher levels of basic fibroblast growth factor, interleukin 6, and protein in retro-silicone oil fluid compared to other vitreous cavity fluid samples, which may contribute to perisilicone proliferation and epiretinal membrane formation.25


There are few reports of intraretinal SO seen on OCT following SO removal. The mechanism is not entirely well understood, but it may be related to elevated IOP. One study found SO droplets in the retina of a patient after macular hole repair with ILM peeling. The authors postulated that defects in the ILM provided an entryway for the emulsified oil.26

Another report of a patient who underwent multiple RD repairs with otherwise unexplained visual loss and patchy central visual field deficits described numerous small echogenic particles in the vitreous cavity and a hyper-reflective structure within the optic nerve head.27 Swept-source OCT revealed multiple hyper-reflective spaces within the prelaminar optic nerve head, and adaptive optics imaging confirmed that these were similar size and shape to the oil droplets in the vitreous cavity.27 The authors concluded that oil droplets migrating into the retina and optic nerve head may be associated with the progressive visual loss. Similarly, Shields and Eagle described a case of SO infiltration in the entire length of the optic nerve in a patient who underwent RD repair following trauma.28 Their report contains histopathologic images of numerous silicone vacuoles within the optic nerve.


Although SO tamponade is very important in vitreoretinal surgery, particularly in complex cases with PVR, it can be associated with adverse events and visual loss for a variety for reasons. Many of the adverse sequelae of SO tamponade and its removal are a result of SO microemulsification. Although the majority of the SO is removed surgically, it can be impossible to remove all small oil droplets, despite multiple fluid-air exchanges. Although these events may have limited negative impact, emulsified droplets can migrate throughout the eye and cause damage to other intraocular structures. Knowledge of these potential complications is important in management decisions for patients undergoing surgery with SO. RP


  1. Silicone Study Group, Lean JS, Boone DC, et al. Vitrectomy with silicone oil or sulfur hexafluoride gas in eyes with severe proliferative vitreoretinopathy: results of a randomized clinical trial. Silicone Study report 1. Arch Ophthalmol. 1992;110(6):770-779.
  2. Silicone Study Group, McCuen BW, Azen SP, et al. Vitrectomy with silicone oil or perfluoropropane gas in eyes with severe proliferative vitreoretinopathy: results of a randomized clinical trial. Silicone Study Report 2. Arch Ophthalmol. 1992;110(6):780-792.
  3. Hutton WL, Azen SP, Blumenkranz MS, et al. The effects of silicone oil removal. Silicone Study report 6. Arch Ophthalmol. 1994;112(6):778-785.
  4. Jonas JB, Knorr HL, Rank RM, Budde WM. Retinal redetachment after removal of intraocular silicone oil tamponade. Br J Ophthalmol. 2001;85(10):1203-1207.
  5. Dessouki A, Smiddy WE. Prophylactic endolaser photocoagulation with silicone oil removal. Middle East J Ophthalmol. 2002;10:26-29.
  6. Miller JB, Papakostas TD, Vavvas DG. Complications of emulsified silicone oil after retinal detachment repair. Semin Ophthalmol. 2014;29(5-6):312-318.
  7. Toklu Y, Cakmak HB, Ergun SB, Yorgun MA, Simsek S. Time course of silicone oil emulsification. Retina. 2012;32(10):2039-2044.
  8. de Silva DJ, Lim KS, Schulenburg WE. An experimental study on the effect of encircling band procedure on silicone oil emulsification. Br J Ophthalmol. 2005;89(10):1348-1350.
  9. Dresp JH, Menz DH. Interaction of different ocular endotamponades as a risk factor for silicone oil emulsification. Retina. 2005;25(7):902-910.
  10. Parel JM, Milne P, Gautier S, Jallet V, Villain F. Silicone oils: physicochemical properties. In: Ryan S, Schachat A, Wilkinson C, et al, eds. Retina. 4th ed. Amsterdam, The Netherlands: Elsevier; 2005: vol. 3.
  11. Crisp A, de Juan E Jr., Tiedeman J. Effect of silicone oil viscosity on emulsification. Arch Ophthalmol. 1987;105(4):546-550.
  12. Zafar S, Shakir M, Mahmood SA, Amin S, Iqbal Z. Comparison of 1000-centistoke versus 5000-centistoke silicone oil in complex retinal detachment surgery. J Coll Physicians Surg Pak. 2016;26(1):36-40.
  13. Scott IU, Flynn HW Jr., Murray TG, et al. Outcomes of complex retinal detachment repair using 1000- vs 5000-centistoke silicone oil. Arch Ophthalmol. 2005;123(4):473-478.
  14. Apple DJ, Isaacs RT, Kent DG, et al. Silicone oil adhesion to intraocular lenses: an experimental study comparing various biomaterials. J Cataract Refract Surg. 1997;23(4):536-544.
  15. Wong SC, Ramkissoon YD, Lopez M, et al. Use of hydroxypropylmethylcellulose 2% for removing adherent silicone oil from silicone intraocular lenses. Br J Ophthalmol. 2009;93(8):1085-1088.
  16. Apple DJ, Federman JL, Krolicki TJ, et al. Irreversible silicone oil adhesion to silicone intraocular lenses. A clinicopathologic analysis. Ophthalmology. 1996;103(10):1555-1561.
  17. Patel NA, Gangasani N, Yannuzzi NA, Melo G, Flynn HW, Smiddy WE. Indications and outcomes for the removal of intraocular lens implants in a retinal surgery practice. Ophthalmic Surg Lasers Imaging Retina. 2019;50(8):504-508.
  18. Abrams GW, Azen SP, Barr CC, et al. The incidence of corneal abnormalities in the Silicone Study. Silicone Study report 7. Arch Ophthalmol. 1995;113(6):764-769.
  19. Gedde SJ. Management of glaucoma after retinal detachment surgery. Curr Opin Ophthalmol. 2002;13(2):103-109.
  20. Tranos P, Asaria R, Aylward W, Sullivan P, Franks W. Long-term outcome of secondary glaucoma following vitreoretinal surgery. Br J Ophthalmol. 2004;88(3):341-343.
  21. Williams PD, Fuller CG, Scott IU, Fuller DG, Flynn HW Jr. Vision loss associated with the use and removal of intraocular silicone oil. Clin Ophthalmol. 2008;2(4):955-959.
  22. Cazabon S, Groenewald C, Pearce IA, Wong D. Visual loss following removal of intraocular silicone oil. Br J Ophthalmol. 2005;89(7):799-802.
  23. Azzolini C, Docchio F, Brancato R, Trabucchi G. Interactions between light and vitreous fluid substitutes. Arch Ophthalmol. 1992;110(10):1468-1471.
  24. Refojo MF, Leong FL, Chung H, Ueno N, Nemiroff B, Tolentino FI. Extraction of retinol and cholesterol by intraocular silicone oils. Ophthalmology. 1988;95(5):614-618.
  25. Asaria RH, Kon CH, Bunce C, et al. Silicone oil concentrates fibrogenic growth factors in the retro-oil fluid. Br J Ophthalmol. 2004;88(11):1439-1442.
  26. Chung J, Spaide R. Intraretinal silicone oil vacuoles after macular hole surgery with internal limiting membrane peeling. Am J Ophthalmol. 2003;136(4):766-767.
  27. Mrejen S, Sato T, Fisher Y, Spaide RF. Intraretinal and intra-optic nerve head silicone oil vacuoles using adaptive optics. Ophthalmic Surg Lasers Imaging Retina. 2014;45(1):71-73.
  28. Shields CL, Eagle RC Jr. Pseudo-Schnabel’s cavernous degeneration of the optic nerve secondary to intraocular silicone oil. Arch Ophthalmol. 1989;107(5):714-717.