Reproducibility of high-resolution optical coherence tomography measurements of the nerve fibre layer with the new Heidelberg Spectralis optical coherence tomography

Aim Conventional time-domain OCT technology for detection of retinal nerve fibre layer (RNFL) neurodegeneration suffers from technical inaccuracy owing to a lack of exact scan centring around the optic disc as well as a true follow-up possibility. In this study, the authors evaluated a novel high-resolution spectral-domain OCT device (SD-OCT) with an incorporated eye-tracking feature in its ability to objectively measure the RNFL thickness (RNFLT) by testing intraobserver reproducibility in a series of healthy volunteers. Methods Triplicate circumferential RNFL scans of six peripapillary sectors were obtained from both eyes of all 31 participants. The authors compared the measurements of RNFLT during three separate examination days under miotic (Mi) and mydriatic (My) pupil conditions using a high-speed (HS) and high-resolution (HR) scan-acquisition mode. To examine the intersession reproducibility of the SD-OCT measurements, the mean, SD and coefficient of variation (COV) were calculated. Results No significant differences were found in all groups, independent of the mode of image acquisition and examination day (p always >0,05). Under all conditions, low COVs between 0.545% and 3.97% (intrasession COV on baseline) were found. The intersession COV with activated follow-up mode ranged between 0.29% and 1.07%. In both settings, the temporal sector showed the highest COV values. Conclusions True follow-up measurement of identical peripapillary regions may enable clinicians to detect discrete levels of retinal thickness change over time. This constitutes a crucial prerequisite for a reliable monitoring of subtle RNFL changes in neurodegenerative disorders.