Comparison of accuracy of a toric calculator with predicted vs measured posterior corneal astigmatism

Purpose: To compare the accuracy of postoperative residual astigmatism prediction using the Barrett toric calculator with predicted vs measured posterior corneal astigmatism (PCA). Setting: Cullen Eye Institute, Baylor College of Medicine, Houston, Texas. Design: Retrospective case series. Methods: We included 602 eyes with monofocal nontoric intraocular lens implantation. Biometry and PCA were obtained from the IOLMaster 700. Anticipated postoperative refractive astigmatism was calculated with the Barrett toric calculator for predicted and measured PCA, and the astigmatism prediction errors (PEs) for each were calculated using vector analysis. The vector PE magnitudes and percentage of eyes within certain amounts of vector PEs were compared between 2 methods. Results: Compared with the Barrett toric calculator with predicted PCA, the Barrett toric calculator with measured PCA produced significantly smaller mean vector PE (0.54 diopter [D] vs 0.57 D) and higher percentage of eyes with vector PE of <= 0.5 D (57.6% [347/602] vs 52.5% [316/602]) (both P <.05). In eyes with predicted residual astigmatism of <= 0.5 D, the Barrett toric calculator with measured PCA again yielded a significantly higher percentage of eyes with vector PE of <= 0.5 D (51.2% [226/441] vs 44.7% [197/441], P <.05). Conclusions: Accuracy of residual astigmatism prediction is improved using the Barrett toric calculator with measured PCA rather than predicted PCA.