Prediction of the trabecular iris angle after posterior chamber phakic intraocular lens implantation

Abstract

Purpose: To create an equation for predicting the trabecular iris angle (TIA) and to verify its accuracy after implantable collamer lens (ICL) implantation.

Setting: Nagoya Eye Clinic, Nagoya, Japan.

Design: Retrospective evaluation of a screening approach.

Methods: 174 eyes (174 patients) that underwent ICL implantation were included. Patients were randomly assigned to the prediction equation group (116 eyes) or verification group (58 eyes). Anterior segment optical coherence tomography (AS-OCT) (CASIA2 TOMEY) was performed before and 3 months after ICL surgery. For the prediction group, a prediction equation was created with the preoperative AS-OCT parameters and ICL size as independent variables and the postoperative anterior chamber depth (ACD) as the dependent variable. Then, by applying the predicted post-ACD and preoperative AS-OCT parameters as independent variables and TIA after ICL surgery as the dependent variable, a prediction equation was created to predict the postoperative TIA (post-TIA) after ICL surgery. Each prediction equation was created using stepwise multiple regression analysis, and its accuracy was verified by a Bland-Altman plot in the verification group.

Results: The explanatory variables (standardized partial regression coefficient) selected in the post-TIA prediction equation were post-ACD (0.629), TIA750 (0.563), iris curvature (0.353), pupil diameter (-0.281), iris area (-0.249), and trabecular iris space area 250 (-0.171) (R2 = 0.646) (n = 116). There were no clinically significant systematic errors between measured and predictive post-TIA values in the verification group (n = 58). The mean absolute prediction error was 3.43 ± 2.22 degrees.

Conclusions: Post-TIA was accurately predicted from the predicted post-ACD and other preoperative AS-OCT parameters.