Intraocular Lens Power Calculation after Small Incision Lenticule Extraction

Abstract

With more than 1.5 million Small Incision Lenticule Extraction (SMILE) procedures having already been performed worldwide in an ageing population, intraocular lens (IOL) power calculation in post-SMILE eyes will inevitably become a common challenge for ophthalmologists. Since no refractive outcomes of cataract surgery following SMILE have been published, there is a lack of empirical data for optimizing IOL power calculation. Using the ray tracing as the standard of reference-a purely physical method that obviates the need for any empirical optimization-we analyzed the agreement of various IOL power calculation formulas derived from the American Society of Cataract and Refractive Surgeons (ASCRS) post-keratorefractive surgery online calculator. In our study of 88 post-SMILE eyes, the Masket formula showed the smallest mean prediction error [−0.36 ± 0.32 diopters (D)] and median absolute error (0.33D) and yielded the largest percentage of eyes within ±0.50D (70%) in reference to ray tracing. Non-inferior refractive prediction errors and ±0.50D accuracies were achieved by the Barrett True K, Barrett True K No History and the Potvin-Hill formula. Use of these formulas in conjunction with ray tracing is recommended until sufficient data for empirical optimization of IOL power calculation after SMILE is available.

Corneal power measurements by ray tracing in eyes after small incision lenticule extraction for myopia with a combined Scheimpflug Camera–Placido disk topographer

“…They evaluated the predictability of IOL power calculation after SMILE using ray tracing (Oculix based on Pentacam data) and found that ray tracing was the most accurate approach for IOL power calculation after myopic SMILE, with 81.9% and 98% of cases with an absolute PE of 0.5 D and 1 D or less, respectively. Another recent study has used ray tracing calculation as a benchmark in a theoretical study of IOL power calculation after SMILE [12]. These results, as well as those in our study, support the advantages of ray tracing for corneal power measurements and IOL calculation after SMILE.…”

“…The same result was obtained using Haigis formula in SMILE eyes by Pan et al [19] and Wei et al [31], using Pentacam to measure corneal power. In two recent theoretical studies, involving virtual implantation of an IOL, Haigis L and Shammas performed worse than Masket formula, ray tracing (Okulix), Barrett True K and Potvin-Hill formula [12,13]. As stated before, Gylkerne et al [10] found that LAISK and SMILE induce distinct changes in the anterior corneal shape, and thus, strategies developed to calculate IOL power calculation after LASIK may not be valid after SMILE.…”

“…In this setting, two approaches have been proposed. A theoretical study involving the virtual implantation of an IOL was employed in three studies [12][13][14] that evaluated the performance of ray tracing using Oculix [12,13], Barrett True K [12,14], conventional third-generation formulas (SRK-T, Holladay and Hoffer Q) [13,14],…”

“…Haigis and Haigis L [12][13][14], as well as the series of formulas developed for IOL calculation after LASIK and PRK of the American Society of Cataract and Refractive Surgery (ASCRS) IOL calculator [14] showing satisfactory results using ray tracing [13] and Barrett true K formulas [12,14], as well as Masket and Potvin-Hill formulas [12]. However, the Masket and Barrett True K formulas require previous data, which are not always available, or may not be reliable.…”

Advancements in intraocular lens power calculation formulas

“…One study using ray tracing as the standard of reference analyzed the agreement of various IOL power calculation formulas derived from the American Society of Cataract and Refractive Surgeons (ASCRS) postkeratorefractive surgery online calculator and found that the Masket formula showed the smallest mean prediction error (-0.36 AE 0.32 D) and MedAE (0.33 D), and yielded the largest percentage of eyes within AE0.50 D (70%). Noninferior refractive prediction error and AE0.50 D accuracies were achieved by the Barrett True-K, Barrett True K No History and the Potvin-Hill formula [34]. Use of these formulas in conjunction with ray tracing is recommended until sufficient data for empiric optimization of IOL power calculation after SMILE is available [34].…”