ORIGINAL ARTICLES

Sources of longitudinal variability in optical coherence tomography nerve-fibre layer measurements

L Kagemann^1,2, T Mumcuoglu¹, G Wollstein¹, R Bilonick¹, H Ishikawa^1,2, K A Townsend¹, M Gabriele^1,2, J G Fujimoto³ and J S Schuman^1,2

¹ UPMC Eye Center, Ophthalmology and Visual Science Research Center, Eye and Ear Institute, Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
² Department of Bioengineering, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, PA, USA
³ Department of Electrical Engineering and Computer Science and Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA, USA

Correspondence to:
G Wollstein, UPMC Eye Center, Department of Ophthalmology, University of Pittsburgh School of Medicine, 203 Lothrop Street, Eye and Ear Institute, Suite 834, Pittsburgh, PA 15213, USA; wollsteing{at}upmc.edu

Aims: The purpose of this study was to compare the day-to-day reproducibility of optical coherence tomography (OCT; StratusOCT, Carl Zeiss Meditec, Dublin, CA) measurements of retinal nerve-fibre layer (RNFL) measurements at time points 1 year apart.

Methods: One eye in each of 11 healthy subjects was examined using the StratusOCT fast RNFL scan protocol. Three fast RNFL scans with signal strength 7 were obtained on each of 3 days within a month. This protocol was repeated after 12 months. A linear mixed effects model fitted to the nested data was used to compute the variance components.

Results: The square root of the variance component that was attributed to the differences between subjects was 7.17 µm in 2005 and 7.28 µm in 2006. The square roots of the variance component due to differences between days within a single subject were 1.95 µm and 1.50 µm, respectively, and for within day within a single subject were 2.51 µm and 2.55 µm, respectively. There were no statistically significant differences for any variance component between the two testing occasions.

Conclusions: Measurement error variance remains similar from year to year. Day and scan variance component values obtained in a cohort study may be safely applied for prediction of long-term reproducibility.

Funding: National Institutes of Health R01-EY13178-07, R01-EY11289-21, P30-EY008098, The Eye and Ear Foundation (Pittsburgh, PA) and unrestricted grant from Research to Prevent Blindness.

Competing interests: JGF and JSS receive royalties from intellectual property licensed by Massachusetts Institute of Technology to Carl Zeiss Meditec. GW receives grant support from Carl Zeiss Meditec and Optovue.

Ethics approval: Institutional Review Board and Ethics Committee approval were obtained for the study. This study followed the tenets of the Declaration of Helsinki and was conducted in compliance with the Health Insurance Portability and Accountability Act (HIPPA).

Patient consent: Informed consent was obtained from all subjects.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?