Intraocular lens power calculation for eyes with high and low average keratometry readings: Comparison between various formulas

Purpose To compare the accuracy of intraocular lens power prediction for eyes with average keratometry (K) readings greater than 46.00 diopters (D) and lower than 42.00 D. Setting Ein-Tal Eye Center, Tel-Aviv, Israel. Design Retrospective case series. Methods Eyes having cataract extraction surgery with steep and flat preoperative corneal curvatures determined with the Lenstar-LS900 device were enrolled. Refractive prediction errors for the Barrett Universal II, Haigis, Hill-RBF, Hoffer-Q, Holladay 1, Holladay 2, Olsen, and SRK/T formulas were compared. Optimized K values for the SRK/T formula were back-calculated for each group. Validation was performed using an additional dataset. Results The study comprised 171 eyes (79, K reading >46.00 D; 92, K reading <42.00 D). For K readings greater than 46.00 D, myopic errors were noted using the SRK/T and Hill-RBF formulas and hyperopic errors using the Olsen C-constant and Haigis (−0.31 D, −0.17 D, 0.18 D, and 0.17 D, respectively). The percentage of eyes with an absolute error within ±0.50 D from target refraction ranged from 60.8% (SRK/T) to 83.0% (Hill-RBF). For K readings lower than 42.00 D, myopic errors were seen using the Haigis, Hill-RBF, Hoffer-Q, and Olsen-C formulas (−0.31 D, −0.14 D, −0.22 D, and −0.17 D, respectively) and a hyperopic error using the SRK/T formula (0.16 D). The refractive prediction within ±0.50 D ranged between 75.0% (Haigis) and 96.7% (Barrett Universal II). Conclusions Power calculation for eyes with flat corneas and steep corneas requires the use of specific formulas for accurate postoperative results. An adjustment method of the SRK/T formula is proposed.