A comparison of intraocular lens power calculation formulas in eyes with keratoconus
Free Paper Details
First Author: J.Chaves PORTUGAL
Co Author(s): J. Bernardes M. Raimundo J. Gil C. Lobo J. Murta
To comparatively evaluate the accuracy of newer intraocular lens (IOL) calculation formulas (Barrett Universal II, Kane, Ladas Super Formula AI, Hill-RBF 2.0, EVO, PhacoOptics®) and common second and third-generation formulas (SRKII, SRK/T, Haigis, Holladay 1) in the prediction of postoperative refraction in eyes with keratoconus (KC)
Tertiary university hospital
Retrospective study including eyes with stage 1 and 2 keratoconus (Amsler-Krumeich classification) submitted to uncomplicated cataract surgery. All patients underwent optical biometry (Carl Zeiss IOLMaster 500 and IOLMaster 700) and corneal tomography (Oculus Pentacam®) evaluation. The post-operative spherical equivalent for the same implanted IOL was estimated using SRKII, SRK/T, Holladay, Barrett Universal II, Kane, Ladas, Hill-RBF 2.0, EVO and PhacoOptics® formulas. Both IOL Master’s SimK and Pentacam’s 4.5-mm equivalent keratometry readings (EKR) values were used.Outcomes included the median absolute prediction error (MedAE) and the proportion of eyes within 0.50, 1.00 and 2.00 diopters (D) of the pre-operative prediction
Twenty-one eyes, 12 eyes with stage 1 KC (61.2%) and 9 eyes with stage 2 KC (42.8%), were included. When SimKs were used, MedAE was Hill-RBF 2.0 (0.64D), Kane (0.78D), EVO (0.96D), Barret (1.01D), SRK/T (1.32D), Ladas (1.34D), PhacoOptics (1.44D), Holladay 1 (1.63D), SRKII (1.84D). With 4.5-mm EKR, MedAE was Barret (0.93D), Haigis (0.95D), Kane (0.95D), Ladas (0.99D), Hill-RBF 2.0 (1.01D), EVO (1.06D), Holladay 1 (1.11D), SRK/T (1.14), PhacoOptics® (1.15D), SRKII (1.90D). The formula with the lowest MedAE, Hill-RBF 2.0, yielded a prediction error within 0.5D, 1D and 2D in 33.3%, 50% and 83.3% of cases, respectively.
To the best of our knowledge, this is the first study to evaluate the accuracy of recent “super-formulas” with eyes with keratoconus. As expected, eyes with keratoconus still represent a challenge regarding IOL power calculation due to unstable keratometry. Nonetheless, recent formulas like Barrett Universal II, Kane, Hill-RBF v2.0, Ladas Super Formula and EVO showed reasonably good accuracy, particularly when 4.5-mm EKR values were used. The Hill-RBF v2.0 formula had the lowest median absolute prediction error and highest proportion of eyes within 0.25D, 0.50D and 1.00D of the predicted target.