Postoperative hyperopic shift after cataract surgery in silicone oil‐filled eyes

Abstract

Vitrectomy with silicone oil (SO) instillation has been widely used in complicated vitreoretinal diseases to achieve retina stability. Because progressive lens opacity is common after vitrectomy and SO infusion, the removal of SO and cataract extraction with intra-ocular lens (IOL) implantation may be performed in the same setting. The refractive outcomes of cataract surgery with simultaneous SO removal have shown acceptable results. However, the refractive outcome after cataract surgery is less predictable in some complicated cases in which intravitreal SO is retained to achieve a long-term tamponade effect. We investigate the refractive outcome of cataract surgery in 21 SO-filled eyes with complicated vitreoretinal diseases and compare the results with those in 42 vitrectomized eyes without SO instillation. The SO used in our study was 5000 centistokes. Axial length (AL) measurements were performed using contact A-scan echography (Sonomed A-2500, New York, NY, USA). To adjust the falsely elongated AL due to the low sound velocity in SO, the following formula was used: AL = anterior chamber depth + lens thickness + vitreous length *0.68. IOL power was calculated using the SRK-T formula. In all SO-filled eyes, we generally chose a more myopic calculated target refraction when deciding the IOL power to compensate for the possible hyperopic shift due to a higher refractive index of SO (Stefansson et al. 1988; Smith et al. 1990; Hotta & Sugitani 2005). Foldable acrylic IOLs (AcrySof SA60AT for power ≧6 D or AcrySof MA60MA for power<6 D; Alcon Surgery) were implanted into the capsular bag in all eyes. Intergroup comparison was shown in Table 1. The refractive outcomes in the SO-filled group were more variable (1.07 ± 2.26 D versus −0.57 ± 1.15 D) and showed a significant tendency of hyperopic shift (+2.89 ± 2.15 D versus + 0.04 ± 1.0 D, p < 0.001). Eyes in the SO-filled group had a higher proportion of cases undergoing scleral buckling procedure, a shorter anterior chamber depth (ACD) and a longer adjusted AL in average. In one eye in the SO-filled group, the refraction changed from +3.25 D after cataract surgery to 0 D after SO removal 3 years later. The uncertainty of postoperative refraction after cataract surgery in SO-filled eyes is due to several perioperative factors. In our study, the SO-filled group had significantly longer average AL and shorter average ACD, which might be related to a higher rate of scleral buckling procedure performed in those eyes. A longer AL could augment the SO-induced acoustic signal attenuating effect and make the measurement less reliable (Ghoraba et al. 2002). Previous studies found that a hyperopic shift in refraction would be induced when vitreous was replaced with SO through vitrectomy in phakic or pseudophakic eyes (Stefansson et al. 1988; Smith et al. 1990; Hotta & Sugitani 2005). This tendency of SO-induced hyperopic shift was also noted in our study. However, the amount of hyperopic shift showed great variation among different studies. The posterior convex curvature of the IOL is thought to be related to the amount of hyperopic shift (Hotta & Sugitani 2005). Incomplete SO filling in the vitreous cavity causing changes in curvature and contact area of SO-fluid interface in different head positions may also result in variability in refractions and AL measurement (Dick et al. 1997). In conclusion, the refractive results of cataract surgery in SO-filled eyes were generally more variable than in simply vitrectomized eyes. Because a mean hyperopic shift is about 3.0 D in our study, we suggest choosing −2.5 to − 3.0 D target refraction, which may partially compensate for the SO-induced hyperopic shift, if SO is intended to be left for a long-term tamponade effect.