Abstract
Purpose:
To evaluate the accuracy of intraocular lens (IOL) power calculations in eyes with previous radial keratotomy (RK).
Methods:
In 62 consecutive cases with best-corrected visual acuity of 20/32 or better and follow-up more than one month after cataract surgery, we evaluated the accuracy of IOL calculations performed using corneal power values from six devices: 1) IOLMaster Ks, 2) Lenstar Ks, 3) Atlas 4-mm zone and 1-4-mm rings, 4) EyeSys effective refractive power (EffRP), 5) Galilei SimK and 4-mm total corneal power (TCP), and 6) RTVue net corneal power (NCP). Using both the original axial length (AL) and an optimized AL in long eyes (>25 mm), we calculated IOL power with two formulas: 1) double-K Holladay 1 as used by the ASCRS calculator, and 2) modified Shammas formula, in which unmodified corneal powers were used. Refractive prediction errors (RPE) were calculated, with positive values indicating an outcome that is more hyperopic than the target refraction and smaller standard deviations (SD) implying better consistency of performance of the formula.
Results:
RPE values (median ± SD in diopters) with the double-K Holladay 1 and modified Shammas formulas were: 1) -0.17 ± 1.03 and +0.20 ± 0.96 for IOL Master Ks; 2) -0.33 ± 0.92 and +0.15 ± 0.90 for Lenstar Ks; 3) -0.22 ± 0.88 and +0.28 ± 0.79, and -0.19 ± 0.96 and +0.22 ± 0.88 for Atlas 4-mm zone and 1-4-mm rings, respectively; 4) -0.54 ± 0.89 and +0.09 ± 0.76 for EyeSys EffRP; 5) -0.44 ± 1.04 and +0.10 ± 0.96, and -1.99 ± 1.35 and -1.43 ± 1.24 for Galilei SimK and 4-mm TCP, respectively; and 6) -1.13 ± 1.16 and -0.68 ± 1.05 for RTVue NCP. Optimizing the AL in long eyes created more myopic RPEs with the double-K Holladay 1 but reduced median RPE by 0.1 to 0.28 D with modified Shammas formula. Using the average IOL power with devices included in the ASCRS calculator (IOLMaster, Lenstar, Atlas and EyeSys,), RPE was -0.28 ± 0.89 D; % of eyes with RPE of >+1.0 D, +1.0 to +0.01 D, 0.00 to -1.0 D and <-1.0 D were 9.7%, 27.4%, 48.4% and 14.5%, respectively. The percentage of eyes that had RPE > 1 D were 16.7%, 22.5% and 40.0% in eyes with <4, 5-8 and >8 RK incisions, respectively. Flatter corneas were not associated with higher RPEs.
Conclusions:
IOL power calculations in RK eyes are challenging, especially in eyes with 8 or more RK incisions. More accurate corneal power measurements in RK eyes are desirable.
Keywords: 567 intraocular lens •
733 topography