Purchase this article with an account.
W. D. Dilworth, T. Mumcuoglu, G. Wollstein, L. Kagemann, K. A. Townsend, R. A. Bilonick, H. Ishikawa, M. L. Gabriele, J. S. Schuman; The Long Term Stability of StratusOCT Reproducibility. Invest. Ophthalmol. Vis. Sci. 2007;48(13):520.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Statistical design of longitudinal studies requires an estimation of measurement reproducibility over time. The purpose of the present study was to determine the stability of optical coherence tomography (OCT; StratusOCT, Carl Zeiss Meditec, Dublin, CA) retinal nerve fiber layer (RNFL) measurement reproducibility over time.
One eye each in 11 subjects (5 male 6 female, 7 OD 4 OS, 39 ±;10 years of age) was examined using the StratusOCT fast RNFL scan protocol, consisting of an average of 3 circumpapillary scans, each with a diameter of 3.4 mm. Three fast RNFL scans with signal score>6 were obtained on each of 3 days within a month. This protocol was repeated after 12 months. Intervisit and intravisit interclass correlation (ICC) and standard deviation (SD) were calculated for mean RNFL, and each quadrant at year 1 and 2. The calibration of the imaging device was tested on a model eye on a monthly basis during both years, and the unit was routinely calibrated by the manufacturer between years 1 and 2.
The coefficients of variation (CVs) of the model eye measurements for years 1 and 2 were 0.29% and 0.32% respectively. The intervisit CVs of mean RNFL thickness were 1.29% and 1.20% for years 1 and 2 respectively.
Drift in the imaging device calibration may contribute to variability observed in patients. The reproducibility of mean RNFL measurements remains stable over one year; however, the reproducibility of RNFL measurements in individual quadrants varies. This should be taken into consideration when planning a longitudinal study.
This PDF is available to Subscribers Only