Accuracy Of Iol Power Calculation In Eyes With Lens Exchange

In eyes with IOL exchange, to compare the accuracy of IOL power calculations using the Barrett Rx and “MR-biometry” method that uses the existing IOL parameters (model and power) and the biometry measurements and manifest refraction obtained prior to the IOL exchange.

Department of Ophthalmology, Baylor College of Medicine

We reviewed the refractive outcomes of consecutive cases who underwent IOL exchange between 2015 to 2024 (approximately 300 eyes). Exclusion criteria were: 1) IOL exchange for dislocated lens, 2) unknown lens parameters in the eye, and 3) no manifest refraction or distance-corrected vision 20/40 or worse at least 3 weeks after the IOL exchange. IOL power was calculated using the Barrett Rx and MR-biometry method. Refractive prediction errors were calculated by subtracting the predicted refraction using these two methods from the actual postoperative refraction. We compared the accuracy of these two methods using the WHWK (Wilcox-Holladay-Wang-Koch) Statistical Software with R.

Compared to MR-biometry method, significant differences (all P<0.05) were found for groups with IOL location: 1) fully in bag (n=155): Barrett Rx had more myopic MeanPE (-0.18 vs 0.08 D); 2) in bag with anterior optic capture (OC) (n=15): Barrett Rx had more myopic MeanPE (-1.01 vs -0.29 D), higher RMSAE (1.16 vs. 0.50 D), higher MeanAE (1.03 vs 0.41 D), and lower % of eyes within ±0.5 D PE (20% vs 67%); 3) in sulcus with posterior OC (n= 48): Barrett Rx had more myopic MeanPE (-0.71 vs -0.08 D), higher RMSAE (1.10 vs. 0.80 D), higher MeanAE (0.92 vs 0.62 D), and lower % of eyes within ±0.5 D PE (29% vs 50%); and 4) in sulcus (n=23): Barrett Rx had more myopic MeanPE (-0.96 vs -0.63 D).

Compared to the Barrett Rx, MR-biometry method produced less myopic MeanPEs in all groups, lower RMSAE and MeanAE, and higher % of eyes with PEs within ±0.5 D in eyes with IOLs in the bag with anterior optic capture and in the sulcus with posterior optic capture.