Abstract
Purpose :
To predict the post-operative refractive error in cataract surgery eyes with previous corneal refractive surgery.
Methods :
A total of 75 patients with previous myopic and hyperopic LASIK (40 female / 35 male; age: 66 ± 6 y.o.; axial length range: 20.85 - 29.84 mm; RMS HOAs range: 0.11 - 0.81 microns for a 4-mm pupil) were included in the study. Anterior and posterior corneal Zernike coefficients (from Pentacam, Sirius and MS39 corneal topographers), axial distances (corneal thickness, post-operative effective lens position and vitreous chamber depth) and IOL geometry (SN60WF, ALCON Laboratories; Precizon Toric, OPHTEC BV) were exported to ZEMAX (Radiant ZEMAX; Focus software) where the patient-specific eye models were built. Optical quality was described in terms of the metric Visual Strehl (VSOTF). The defocus and astigmatism producing the maximum VSOTF was taken as the optimal target and compared with the post-operative subjective refraction.
Results :
The Bland-Altman analyses of the data show the agreement between the predicted refraction of the proposed optimized methodology for IOL power calculation and the post-operative subjective refraction. The bias and limits of agreement (LoA) were -0.02±0.52 [LoA: -1.05, 1.00], -0.03±0.20 [LoA: -0.43, 0.35] and -0.03±0.19 [LoA: -0.39, 0.38] D for M, J0 and J45, respectively. Average spherical equivalent was -0.36±0.80 D (subjective) and -0.35±0.73 D (predicted), average power vector J0 was 0.10±0.33 D (subjective) and 0.10±0.37 D (predicted), and average power vector J45 was -0.03±0.12 D (subjective) and -0.02±0.15 D (predicted), with ICC of 0.79 for M, 0.71 for J0 and 0.55 for J45, respectively; showing high reliability between the predicted refraction and the post-operative subjective refraction in all patients. The percentage of eyes within ±0.5D was of 82.6% (M), 84.1% (J0) and 82.6% (J45), while the agreement within ±1.0D was of 93.3% (M), 98.6% (J0) and 97.3% (J45).
Conclusions :
We develop a patient-specific method for IOL selection based on virtual ray tracing. Strong agreement between the predicted refraction and the post-operative subjective refraction in a large case series of cataract surgery eyes with previous corneal refractive surgery suggest that the proposed optimized methodology provide an accurate tool for IOL selection. The predicted refraction was influenced by the presence of corneal aberrations.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.