Prediction of postoperative intraocular lens position in cataract surgery

Poster Details

First Author: A.Tamaoki JAPAN

Co Author(s):    T. Kojima   Y. Nagase   H. Nakamura   K. Ichikawa     

Abstract Details

Purpose:

The predictive anterior chamber depth (ACD) is one of the main sources of refractive error in the intraocular lens (IOL) power calculation. Our purpose of this study is to investigate methods to predict the effective postoperative ACD.

Setting:

Social Insurance Chukyo Hospital, Nagoya, Japan

Methods:

Total 83 eyes of 68 patients were enrolled in the study. All patients underwent cataract surgery and anterior chamber depth was successfully measured with Lenstar LS900?Haag-Streit AG, Koeniz, Switzerland?laser interferometry before and after cataract surgery. Stepwise multiple regression analysis was performed for 62 eyes of 51 consecutive cataract patients in order to predict the postoperative ACD, and the regression equation was calculated. The axial length, mean keratometric value, preoperative ACD, crystalline lens thickness and corneal diameter were applied as candidates of explanatory variables. Then regression equation was evaluated for a new group of 21 eyes of 17 patients.

Results:

Postoperative predictive ACD was significantly correlated with 3 explanatory variables of preoperative ACD, the corneal diameter, and the lens thickness ( p<0.0001 , r2 =0.684).?The mean absolute error (MAE) between predictive ACD and the postoperative ACD was 0.23 ± 0.24mm. ACD calculated from the regression equation was correctly predicted with 13 eyes (61.9%) within ± 0.2mm, 19 eyes (90.5%) within ± 0.5mm. In 83 eyes, the MAE between predicted ACD and postoperative ACD by the SRK/T and Haigis formula was 1.10 ± 0.52mm and 0.53 ± 0.45mm, respectively. ACD calculated from SRK/T and Haigis formula was predicted with none (0%) and 17 eyes (20.5%) within ± 0.2mm, respectively.

Conclusions:

In recent years, the prediction of postoperative intraocular lens position has become possible due to the advancement of modern biometric techniques. However, prediction should require further accuracy in order to apply it to the intraocular lens calculation. FINANCIAL DISCLOSURE?: No