# Evaluation of the Accuracy of Intraocular Lens Power Calculation Formulas for Cataract Surgery in Patients with Keratoconus

## TL;DR

This study compares the accuracy of different formulas used to calculate intraocular lens power for cataract surgery in patients with keratoconus.

## Contribution

The study evaluates and compares the performance of various IOL power calculation formulas specifically in patients with keratoconus.

## Key findings

- The SRK/T formula had the lowest mean absolute error (MAE) and highest accuracy in predicting refractive outcomes.
- The Holladay 1 and Hoffer Q formulas resulted in a hyperopic refractive shift, while SRK II tended toward myopia.
- Among modern formulas, Barrett True-K showed the highest accuracy in eyes with keratoconus.

## Abstract

To compare the refractive accuracy of different intraocular lens (IOL) power calculation formulas in eyes with keratoconus (KCN) undergoing cataract surgery.

This retrospective case series included the medical records of patients with KCN who underwent optical biometry and cataract surgery with IOL implantation. The predicted spherical equivalent (SE) values were calculated using the Holladay 1, Hoffer Q, SRK/T, and SRK II formulas. Additionally, a subgroup analysis was performed for eyes with available data on anterior chamber depth to compare the accuracy of Haigis, Barrett Universal II, Barrett True-K, EVO 2.0, Kane, and Kane KCN formulas. The mean prediction error (PE), mean absolute error (MAE), median absolute error, and the percentage of eyes within a PE of
±
0.25 diopters (D),
±
0.50 D,
±
0.75 D, and
±
1.00 D were calculated.

Forty-seven eyes of 30 patients were included. The MAE was significantly different among the Holladay 1, Hoffer Q, SRK/T, and SRK II formulas. The Holladay 1 and Hoffer Q formulas led to a hyperopic refractive shift. The SRK/T and SRK II formulas tended toward a myopic refractive outcome. The MAE was lowest for the SRK/T formula (0.39 D), followed by the Holladay 1 (0.48 D), Hoffer Q (0.59 D), and SRK II (0.87 D). Statistical analysis revealed a significantly lower MAE for the SRK/T formula compared to the Hoffer Q and SRK II formulas (P

<
0.05). The percentage of eyes within a PE of
±
0.50 D was 70.2% for SRK/T, 57.44% for Holladay 1, 48.93% for Hoffer Q, and 29.78% for SRK II. The subgroup analysis comprising 11 eyes showed no significant difference among six other formulas (Haigis, Barrett Universal II, Barrett True K, EVO 2.0, Kane, and Kane KCN), with Barrett True-K having the least MAE.

The SRK/T was the most accurate IOL power calculation formula in this study, and Holladay 1 could be an alternative choice. SRK II had the lowest accuracy in predicting refractive outcomes. Among modern formulas, Barrett True-K demonstrated the highest accuracy in eyes with KCN.