Prediction Power More Accurate for IOLs With Labeled Manufacturing Tolerance

Compiled by Callan Navitsky, Associate Editor; and Steve Daily, News Editor
IOLs available in 0.25 D increments with a labeled manufacturing tolerance increased the percentage of
patients within ±0.25 D of the targeted refraction to a statistically significant and clinically meaningful
degree compared with unlabeled IOLs available in 0.50 D increments, according to a study in the
Journal of Cataract & Refractive Surgery.1

In total, 118 eyes underwent cataract surgery and implantation of an IOL available in 0.25 D increments
and labeled with a manufacturing tolerance of ±0.11 D (n=67) or an IOL available in 0.50 D increments
without a labeled manufacturing tolerance (n=51).

Based on the SRK-T formula, the mean error of prediction after optimization was -0.03 ±0.35 D
(standard deviation) in the labeled group and -0.05 ±0.46 D in the unlabeled group (P=.64). The mean
absolute error of prediction was statistically significantly smaller in the labeled group (0.26 ±0.23 D)
than in the unlabeled group (0.37 ±0.28 D; P=.04). The mean and absolute errors were not statistically
significantly different with the Holladay 1 or Hoffer Q formula. Sixty-three percent of patients in the
labeled group and 43% in the unlabeled group (P=.03) were within ±0.25 D of the prediction error;
84% and 69%, respectively, were within ±0.50 D (P=.06).

1. Zudans JV, Desai NR, Trattler WB. Comparison of prediction error: labeled versus unlabeled intraocular
lens manufacturing tolerance. J Cataract Refract Surg. 2012;38(3):394-402.