Unexplained Vision Loss Associated With Intraocular Silicone Oil Tamponade in Rhegmatogenous Retinal Detachment Repair

November 2023

Sushant Wagley MD
Iowa Retina Consultants
Des Moines, IA

 Pakravan P, Shaheen A, Patel V, Villalba MF, Dib B, Lai J, Rohowetz L, Chau V, Patel NA, Tzu JH, Wang AL, Alhoyek S, Scott N, Samara WA, Goduni L, Jung JJ, Russell JF, Mantopoulos D, Hajrasouliha AR, Savoie BT, Haddock LJ, Berrocal AM, Sridhar J, West MR, Yannuzzi NA. Unexplained Vision Loss Associated With Intraocular Silicone Oil Tamponade in Rhegmatogenous Retinal Detachment Repair. J Vitreoretin Dis. 2023;7(4):299-304.


Silicone oil (SO) is a commonly used intraocular tamponade agent in complex vitreoretinal surgeries or when altitude restrictions limit gas use. Although rare, there are reports of unexplained vision loss following the use of SO. This study set out to report the clinical features and visual outcomes of patients who experienced at least 2 lines of unexplained vision loss after SO tamponade.

This was a retrospective, multicenter, case series examining 29 eyes over an eight-year period. Eyes with other identifiable cause of vision loss such as prolonged rise in intraocular pressure, cataracts, epi-retinal membranes, cystoid macular edema were excluded. The most common indication for SO tamponade was rhegmatogenous retinal detachment (RRD). Mean age at the time of SO placement was 50.5 years. One thousand centistoke viscosity was the most commonly used (83%) SO agent followed by 5000-centistoke (17%), and both viscosities were found to be culpable in unexplained vision loss. The mean duration of SO tamponade was 148 days. In this study, 62% developed unexplained vision loss while under SO and 38% developed vision loss after SO removal.

In the group that lost vision while still under SO, BCVA was 20/85 pre-SO placement which decreased to 20/340 at the last visit before SO was removed. While BCVA improved after SO was removed, it was still lower than baseline with final BCVA at 20/235. In these eyes, the mean duration of SO tamponade at the time of vision loss was 116 days.

In the group that lost vision after SO was removed, BCVA was 20/104 at baseline with improvement to 20/72 at the last visit before SO removal. However, at last follow up after SO removal the mean BCVA had decreased to 20/219. In this group, the mean duration of tamponade at time of vision loss was 145 days.

Both groups showed ganglion cell layer thinning on optical coherence tomography.

Limitations included the retrospective, descriptive nature of the study. There was also heterogenicity of surgeons and techniques over multiple institutions. The authors could not determine the rate at which vision loss occurs and authors could not identify risk factors leading to vision loss with SO. However, this is the largest review to date and the authors present a real-world analysis into understanding how SO can lead to vision loss.

In conclusion, this study highlights that unexplained vision loss with SO tamponade can occur while the SO is in the eye as well as after the SO has been removed. It can be seen with both 1000-centistoke and 5000-centistoke oil. Vision loss can present in both macula-involved and macula-sparing RRD. The most common imaging finding was ganglion cell layer loss on OCT pointing toward a mechanism of neurodegenerative damage from the tamponade agent. This study serves as the first step in further analysis and understanding this rare but important topic in vitreoretinal surgery.