Purpose: To compare the refractive results of 4 different options for the Barrett True-K formula in eyes with previous myopic excimer laser surgery.
Setting: IRCCS-Fondazione Bietti, Rome, Italy.
Design: Retrospective case series.
Methods: Biometric measurements obtained with a rotating Scheimpflug camera (Pentacam) were entered into the Barrett True-K formula. Clinical history (laser-induced refractive change) and the measured posterior corneal curvature were entered as optional. Four variants of the Barrett True-K formula were investigated: (1) with history and measured posterior corneal power, (2) with history and predicted posterior corneal power, (3) no history with measured posterior corneal power, and (4) no history with predicted posterior corneal power. The prediction error (PE) was calculated as the difference between the measured and predicted postoperative refraction values.
Results: In 50 eyes (50 patients), the Barrett True-K formula with history and measured posterior corneal power resulted in the lowest standard deviation of the PE (0.52 diopters [D]), lowest median (0.245 D) and mean (0.413 D) absolute errors, and highest percentage of eyes with a PE within ±0.25 D (54%), ±0.50 D (70%), and ±0.75 D (84%). The Barrett True-K no-history formula with predicted posterior corneal power yielded the worst refractive outcomes. When the 4 options were compared, statistically significant differences were detected among the median absolute errors (P = .0017) and the percentage of eyes with a PE within ±0.25 D (P < .0001).
Conclusions: Using historical data and measured posterior corneal power improved the refractive accuracy of the Barrett True-K formula in eyes with previous myopic excimer laser surgery.