Abstract
Purpose: :
To determine the reproducibility of retinal nerve fiber layer (RNFL) measurements with an eye tracking Spectral Domain Optical Coherence Tomography (SD-OCT) around the optic nerve in normal and glaucomatous eyes.
Methods: :
The peripapillary RNFL of 40 normal and glaucomatous eyes was imaged on a commercially available high resolution SD-OCT (Spectralis, Heidelberg Engineering). RNFL thickness was measured with undilated pupils, around the disc with 16 automatically averaged consecutive circular B-scans (3.4 mm diameter, 768 A-scans). Three examinations were used to have one baseline measurement and two follow up units within 24 hours. The two follow ups were performed with and without an follow up online tracking system (FUTS). The FUTS automatically aligns the B-scan position of the follow up measurement to exactly the same position of the baseline measurement.
Results: :
Reliability, as measured by intraclass correlation coefficient (ICC) and the coefficient of variation (CV), was calculated for the average RNFL thickness overall and in 4 quadrants (Fig.1). The ICC and CV for the overall average RNFL thickness showed no difference between with and without online tracking (both ICC 0.987, CV 1.62%). Looking into 4 quadrants the follow up tracking system showed better reproducibility for the temporal, nasal and inferior quadrant, being less in the superior quadrant. The CV for the nasal quadrant was significant better using the tracking system. Both techniques provided very good results for the test-retest variability (Table 1).
Conclusions: :
Reproducibility of RNFL measurements using the eye tracking SD-OCT is better as reported with Stratus OCT. The use of an online tracking system can even improve the good results without any tracking system.
Keywords: imaging/image analysis: clinical • optic nerve • anatomy