The problems of biometry in combined silicone oil removal and cataract extraction: a clinical trial

To evaluate the problems and accuracy of biometry in eyes that have been operated on by silicone oil removal combined with cataract extraction with or without intraocular lens implantation.

Twenty-nine consecutive cases that have been operated on for silicone oil removal and cataract, with or without intraocular lens implantation, were included in a prospective study. Axial length measurement was done in silicone oil-filled eyes after changing the sound speed in vitreous cavity to 987 m/sec. The SRKT formula was used for intraocular lens calculation. The cataract was removed by phacoemulsification through corneal section or phacofragmentation through the standard vitrectomy sclerotomy. Intraocular lens implantation was done either on the peripheral anterior capsule (single piece, PMMA) or in the bag (acrylic foldable three pieces). The final refraction after at least 3 months' follow-up was measured. The difference between the predicted and the actual refraction was evaluated. A control group of 30 consecutive cases (20 nonmyopic and 10 highly myopic) of cataract extraction and intraocular lens implantation in eyes not filled with silicone oil was included using the same evaluation methods.

The mean deviation of the final from the predicted refraction was 1.87 +/- 2.11 diopters (range, +3.25 to -9 diopters). A total of 72.4% had a deviation of +/-2 diopters. The mean deviation in high myopia was 3.04 +/- 2.68 diopters (12 cases) (range, +/-1.75 to -9 diopters), while in nonmyopic patients, the mean deviation was 1.04 +/- 1.04 diopters (17 cases) (range, +3.25 to -3.25 diopters). The difference between highly myopic and nonmyopic eyes was statistically significant (P < 0.05). Most high errors occurred in highly myopic eyes. There was no statistical difference between the types of silicone oil (1000 versus 5000) or the mode of cataract extraction (phacoemulsification versus pars plana phacofragmentation). In the control group, the mean deviation in high myopia was 1.48 +/- 1.02 diopters (10 cases) (range, +/-0.5 to -3.5 diopters), while in nonmyopic patients, the mean deviation was 0.65 +/- 0.75 diopters (20 cases) (range, +2.75 to -1.75 diopters). The difference was statistically significant (P < 0.05).

Intraocular lens calculation in silicone oil-filled eyes is accurate in eyes that are not highly myopic. Some highly myopic eyes with posterior staphyloma have great deviation. Unsuitable formula, artifacts, or large eyes beyond the machine range may be the cause of errors of deviation of refraction. Comparable results are obtained in eyes regardless of the type of silicone oil (1000 or 5000) and techniques of intraocular lens implantation (in the sulcus or in the bag).