Related Experiment Video
Updated: Jun 10, 2026
Comparison of Agreement and Accuracy using Binocular Wavefront Optometer with Autorefractor and Phoropter
Published on: September 16, 2025
IOL Calculators Variability in Toric Lens Planning: A Retrospective Paired Comparison of the Planning Discrepancies
Michal Klimek1,2, Maximilian Tamberger1,2, Stephanie Schauer1,2
1Department of Ophthalmology, Klinik Hietzing, Wiener Gesundheitsverbund, Vienna, Austria.
Purpose:
To determine the frequency, magnitude, and direction of toric intraocular lens (IOL) planning discrepancies between the Johnson & Johnson (J&J) official calculator and European Society of Cataract and Refractive Surgeons (ESCRS)-hosted toric engines, and to identify preoperative predictors of clinically relevant disagreements.
Methods:
This retrospective, single-center study included 182 eyes of 127 patients undergoing toric IOL implantation (TECNIS). For each eye, planning results were generated using identical biometric data (IOLMaster 700, CASIA2 keratometry) across all calculators. The primary comparison was J&J with posterior corneal astigmatism correction enabled (PCA-ON) versus ESCRS-Barrett - secondary comparisons included Kane, EVO 2.0, and Hoffer QST. Clinically relevant discrepancy (ClinRel) was defined as a toric cylinder step difference ≥ 1 and/or axis difference ≥ 10°. Logistic regression, including Firth-penalized estimation, identified predictors of ClinRel. The primary analysis set comprised one eye per patient (n=127).
Results:
ClinRel between J&J PCA-ON and Barrett occurred in 59.8% of eyes (76/127; 95% CI: 51.1-68.0%), driven almost only by step differences. J&J PCA-ON recommended a higher step in 56.7% of eyes, with only 2.4% showing the opposite. Discrepancies were concentrated in with-the-rule eyes (72.3%) and nearly absent in against-the-rule eyes (5.6%). Rates were higher for secondary engines (Kane 83.8%, EVO 74.4%, Hoffer QST 76.7%). An Alpins-style paired-planning analysis yielded a Correction Index of 1.20, with the Difference Vector concentrated at the vertical meridian (84°). Posterior corneal astigmatism grade was the strongest independent predictor (Firth-adjusted OR = 203.5; p=0.014). An empirical PCA threshold of 0.36 D (Youden index; sensitivity 91.6%, specificity 56.0%) identified eyes at elevated risk of calculator disagreement.
Conclusion:
Calculators' variability in toric IOL planning is substantial (~60% of eyes) and concentrated in with-the-rule eyes with elevated posterior corneal astigmatism. In this subgroup, consulting multiple calculators is recommended; in against-the-rule eyes, calculator choice rarely alters the recommendation.
