Modern intraocular lens power calculation formulas: A comparative review

ABSTRACT

Accurate intraocular lens (IOL) power calculation is central to achieving predictable refractive outcomes after modern cataract surgery. With the evolution of cataract surgery into a refractive procedure, increasing use of premium IOLs, and availability of advanced optical biometry, small residual refractive errors have become clinically relevant. Traditional formulas such as SRK/T, Hoffer Q, Holladay 1, Haigis, and Holladay 2 have performed reliably in eyes with average biometric parameters, but their accuracy may be less consistent in eyes with extreme axial length or altered corneal profiles. This structured narrative review evaluates the comparative performance of modern IOL power calculation formulas, including Barrett Universal II (BUII), Kane, Hill-RBF, EVO, PEARL-DGS, and Olsen, with traditional formulas included as comparators. A literature search was performed using PubMed/MEDLINE and Google Scholar. Studies were included if they compared two or more IOL formulas in cataract surgery and reported quantitative refractive outcomes such as mean absolute error (MAE), median absolute error (MedAE), or the percentage of eyes within ±0.5 D of target refraction. Eleven studies were included in the comparative analysis. Across the included studies, modern formulas generally demonstrated lower prediction errors and higher refractive accuracy than traditional formulas, although no single formula was consistently superior across all datasets. BUII showed favorable MAE in several studies, whereas Kane frequently demonstrated favorable MedAE. Hill-RBF, EVO, Olsen, and PEARL-DGS also performed strongly in selected datasets, particularly when modern biometry was used. Subgroup analysis suggested that formula performance varied with axial length, biometry platform, IOL type, and refractive target. Modern IOL formulas offer improved refractive predictability, but formula choice should be individualized according to ocular biometry, surgical context, IOL type, and availability of validated biometric inputs.