Posters

Repeatability of measurements taken with two optical biometers

Poster Details

First Author: A.Belkin ISRAEL

Co Author(s):    A. Abulafia   Y. Ton   S. Ofir   A. Ohayon   G. Kleinmann   E. Assia

Abstract Details

Purpose:

To compare the repeatability of biometric measurements taken with two optical biometers, and to analyze the clinical significance of the findings

Setting:

Ein-Tal Eye Center, Tel Aviv, Israel

Methods:

Full biometric measurements were performed using the Lenstar LS 900 and IOLMaster 500 on 70 eyes of 35 healthy volunteers (age > 18). Three identical exams were performed with both biometers at different time points: two were performed 3-5 hours apart, and a third 5-10 days later. All exams were carried out by a single experienced optometrist, and all measurements adhered to strict Hill Criteria. Exclusion criteria were known ocular pathology, contact lens wear, and previous ocular surgery. Repeatability of the following parameters was examined: axial length (AL), anterior chamber depth (ACD), corneal power at the steep and flat meridians, average corneal power, astigmatic power and axis, and astigmatic power vectors.

Results:

A total of 70 eyes of 35 patients were eligible for analysis. The average age was 41.5 ± 12.8 years, and 85% of participants were female. Intra-class correlation (ICC) showed excellent repeatability in all measured parameters for both biometers. Bland & Altman plots demonstrated a smaller 95% confidence interval (CI) for the mean differences between measurements using the Lenstar as compared to the IOLMaster in average corneal power, anterior chamber depth, and corneal astigmatism. A vector analysis of the corneal astigmatism revealed a lower centroid and SD between the repeated measurements of the Lenstar (0.01D ± 0.16) as compared to the IOLMaster (0.04D ± 0.27). When inserting biometric data into IOL calculation formulas, a potential prediction difference between measurements of more than 0.5D in post-operative refraction was seen in significantly less cases when using the Lenstar as compared to the IOLMaster (0% Vs 10% for Holladay 1; p= 0.016, 1.4% Vs 11.4% for Hoffer Q; p= 0.016, 1.4% Vs 12.9 % for Haigis ; p= 0.008).

Conclusions:

The Lenstar is more repeatable than the IOLMaster in the measurement of average corneal power, astigmatic power and axis, and IOL power calculations. The superior repeatability in power vectors could prove very important in toric IOL calculations. FINANCIAL INTEREST: NONE