Intraocular lens formula constant optimization and partial coherence interferometry biometry: Refractive outcomes in 8108 eyes after cataract surgery.

PURPOSE: To assess the benefits of intraocular lens (IOL)-constant optimization for IOLMaster biometry on refractive outcomes after cataract surgery for all surgeons and individual surgeons, define acceptable levels of error in IOL-constant optimization, and calculate the minimum number of eyes required for IOL-constant optimization. SETTING: Department of Ophthalmology, Gloucestershire Hospitals NHS Foundation Trust, Cheltenham, United Kingdom. DESIGN: Database study. METHODS: Hypothetical prediction errors were retrospectively calculated on prospectively collected data extracted from electronic medical records using manufacturers' and optimized IOL constants with Hoffer Q, Holladay 1, and SRK/T formulas for 2 IOLs. The acceptable IOL-constant optimization error margins, personalized IOL constants for individual surgeons, and minimum sample sizes for IOL-constant optimization were evaluated. RESULTS: Optimization of IOL constants reduced the mean absolute errors from 0.66 diopters (D) and 0.52 D to 0.40 D and 0.42 D for the Sofport AO IOL and Akreos Fit IOL, respectively. The percentage of eyes within ±0.25 D, ±0.50 D, and ±1.00 D of target refraction improved from for both IOL models. The IOL-constant errors exceeding 0.09 for the Hoffer Q, 0.09 for the Holladay 1, and 0.15 for the SRK/T produced inferior outcomes. Differences in personalized IOL constants between most surgeons were clinically insignificant. Calculating IOL constants to within 0.06, 0.06, and 0.10 for the Hoffer Q, Holladay 1, and SRK/T, respectively, required 148 to 257 eyes. CONCLUSIONS: Optimizing IOL constants for IOLMaster biometry substantially improved refractive outcomes, far exceeding any additional benefit of personalizing IOL constants for individual surgeons. FINANCIAL DISCLOSURE: No author has a financial or proprietary interest in any material or method mentioned. Additional disclosure is found in the footnotes.