Repeatability and reproducibility of Orbscan II
AbstractThis study aimed to determine the repeatability and reproducibility of Orbscan for anterior and posterior best fit sphere (BFS), simulated keratometry (Sim-K), and central (CCT) and mid-peripheral (PCT) corneal thickness measurements in healthy eyes. Orbscan was performed in 40 healthy eyes (20 subjects) three consecutive times on each cornea during three visits scheduled over one week. Repeatability and reproducibility coefficients [Bland and Altman’s coefficient (BAC), coefficient of variation (CV) and intraclass correlation (ICC)] were calculated for Orbscan anterior and posterior BFS, Sim-K and corneal pachymetry (central, superior, inferior, nasal and temporal locations). Repeatability was calculated using three consecutive measurements during each visit, while reproducibility was calculated using the average of the measurements obtained at each visit. High repeatability was found for all Orbscan measurements (r2<0.01; P>0.05, two-way ANOVA) with BAC and CV <1% (except in PCT coefficients; from 0.97% to 1.67%) and ICC close to 0.98- 0.99 for all visits. High reproducibility was also found for all Orbscan measurements (r2<0.01; P>0.05 two-way ANOVA). BAC values were less than 1% for both BFS and Sim- K, and between 1.21 and 2.20% for corneal pachymetry. CV values were less than 1% (except in superior, nasal and temporal PCT, where they ranged from 1.06 to 1.30%). ICC was close to 0.98-0.99 for all measurements. The BAC of reproducibility was higher than the CV of reproducibility. PCT showed less repeatability and reproducibility than CCT. The Orbscan provides non-invasive, repeatable and reproducible measurements of anterior and posterior BFS, simulated keratometry. and central and mid-peripheral pachymetry in healthy eyes.
- Abstract views: 2461
- PDF: 1953
Copyright (c) 2012 Raúl Martín, Maria Izquierdo, Victoria de Juan, Guadalupe Rodriguez, Itziar Fernandez
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.