A novel Hartmann Shack-based topography system: repeatability and agreement for corneal power with Scheimpflug + Placido topographer and rotating prism auto-keratorefractor

Session Details

Session Title: Photoablation, Cross-Linking and Intracorneal Ring Segment

Session Date/Time: Monday 07/09/2015 | 08:00-10:30

Paper Time: 10:12

Venue: Room 16

First Author: : G.Prakash UNITED ARAB EMIRATES

Co Author(s): :    D. Srivastava   S. Choudhary                 

Abstract Details

Purpose:

To analyze the repeatability and agreement of corneal power using a new Hartman type topographer in comparison to Scheimpflug+Placido and autorefractor devices.

Setting:

Dept of Cornea and Refractive Surgery Services, NMC Eye Care, NMC Specialty Hospital, Abu Dhabi, UAE

Methods:

In this cross sectional, observational study , 100 normal eyes (100 consecutive candidates) without any previous ocular surgery or morbidity except refractive error were evaluated. All candidates underwent three measurements each on full gradient, Hartman type topographer (FG) (iDesign, AMO), Scheimpflug+Placido topographer (SP) (Sirius, CSO) or rotating-prism auto-keratorefractor (AR) (KR1, Nidek). The parameters assessed were flat keratometry (K1), steep keratometry (K2), steep axis (K2 axis), mean K, J0 and J45. Intra-device repeatability and inter-device agreement was evaluated.

Results:

The intra-device means were not significantly different (ANOVA, p <0.05). Intraclass correlations (ICC) were >0.98 except for J0 and J45. In terms of intra-measurement standard deviation (Sw) the SP and FG groups fared better than AR group (p<0.001, ANOVA). On Sw vs Average plots, no significantly predictive fit was seen (p<0.05, R2<0.1 for all the values). On inter-device agreement analysis, there was no difference in means (ANOVA,p<0.05). ICC ranged from 0.92 to 0.99 (p<0.001). Regression fits on Bland Altman plots suggested no clinically significant effect of average values over difference in means.

Conclusions:

The repeatability of Hartman type topographer in normal eyes is comparable to SP combination device and better than AR. The agreement between the three devices is good. However, we recommend against interchanging these devices between followups or pooling their data.

Financial Interest:

NONE