Earlier Silicone Oil Removal Linked to Less Visual Loss
Key Takeaways
- In eyes with macula-on RRD treated with primary ppV and SO, longer tamponade was associated with more UVL than SOR within 60 days, with UVL reported in 42% of eyes with tamponade lasting more than 60 days and in none with SOR within 60 days.
- Shorter SO tamponade was associated with significantly better final postoperative BCVA than longer tamponade.
- Primary retinal reattachment was achieved in all eyes, and no redetachment was reported after SOR during follow-up.
- CFT fell during SO tamponade and recovered after removal, with no reported between-group difference on OCT.
In the Barth et al. study of silicone oil duration and unexplained visual loss in macula-on rhegmatogenous retinal detachment, investigators retrospectively reviewed consecutive eyes with macula-on rhegmatogenous retinal detachment (RRD) treated with primary pars plana vitrectomy (ppV) and silicone oil (SO) from 2020 through 2025. The analyzed sample included 37 eyes from 36 patients, with 25 eyes in a short-term group undergoing silicone oil removal (SOR) within 60 days and 12 in a long-term group with removal after 60 days. Eligibility required at least 2 months of follow-up after SOR, and exclusions included trauma, glaucoma, proliferative vitreoretinopathy grade C or higher, and prior intraocular surgery other than uncomplicated cataract surgery. Assessed outcomes were best corrected visual acuity (BCVA), intraocular pressure (IOP), central foveal thickness (CFT), unexplained visual loss (UVL), reattachment, and postoperative complications after 23- or 25-gauge ppV with perfluorocarbon liquid, endolaser and/or cryopexy, and 5000-centistoke SO, with removal during a second vitrectomy. The comparison therefore tested whether earlier SOR within this primary SO cohort tracked with different functional and anatomic outcomes.
Tamponade exposure separated clearly, with mean SO duration of 36 days in the short-term group versus 92 days in the long-term group. In the functional outcome comparison by silicone oil removal timing, UVL, defined as loss of at least 2 Snellen lines without a correlate on postoperative central optical coherence tomography (OCT), occurred in 0 of 25 short-term eyes and 5 of 12 long-term eyes (p=0.002). In the short-term group, mean BCVA was 0.3 logMAR before surgery and 0.3 logMAR after SOR, while in the long-term group it declined from 0.3 to 0.5 logMAR; final BCVA after SOR was significantly better with shorter tamponade (p=0.033). All UVL events were documented during tamponade rather than after removal.
Baseline characteristics and intraoperative parameters did not differ between groups, and primary retinal reattachment was achieved in all eyes with no redetachment after SOR during a mean follow-up of 7 months. Across eyes with OCT data, mean CFT moved from 251 μm before surgery to 236 μm during tamponade and back to 253 μm after removal, with no between-group difference reported. Qualitative flattening of the foveal contour was reversible and was not linked to worse BCVA. IOP patterns and postoperative complication patterns were similar between groups, elevated pressure episodes were controlled topically, and complications included secondary glaucoma, macular edema, epiretinal membrane, and corneal decompensation.
The authors described the cohort as small, retrospective, non-randomized, and numerically imbalanced, with few UVL events. Longer tamponade often reflected organizational or patient-related delays rather than planned assignment, which leaves room for confounding, and structural comparisons were constrained because baseline OCT was available in only 19 of 37 eyes. In that context, the findings support an association between longer SO exposure and worse visual outcomes in this specific macula-on cohort rather than a causal threshold.
The authors concluded that earlier SOR within this cohort was associated with better postoperative vision and no observed UVL events while anatomic reattachment outcomes were maintained. They also noted that these data do not define a low-risk tamponade duration.
Clinician Questions
How was unexplained visual loss defined after silicone oil tamponade in macula-on retinal detachment?
In this macula-on RRD cohort, unexplained visual loss was defined as a loss of at least 2 Snellen lines with an unremarkable postoperative central OCT scan, and the OCT review was performed by two graders unmasked to tamponade duration. All UVL events were documented during SO tamponade rather than after SOR.
Which macula-on retinal detachment cases were included in this silicone oil duration comparison?
The comparison included consecutive macula-on RRD eyes treated with primary ppV and SO at a German single center between 2020 and 2025, with exclusions for trauma, glaucoma, proliferative vitreoretinopathy grade C or higher, and prior intraocular surgery other than uncomplicated cataract surgery. The findings therefore apply to a selected primary SO cohort rather than all retinal detachment repairs.
Did the study identify a low-risk duration threshold for silicone oil tamponade?
The authors did not identify a low-risk duration threshold for SO tamponade. They described the 60-day cutoff as an investigator-set comparison boundary, and they framed the findings as observational because the cohort was small, non-randomized, and affected by nonrandom reasons for delayed removal.