Glaucoma Progression Detection with Spectral Domain Optical Coherence Tomography (SD-OCT) and Time Domain (TD)-OCT

Abstract

 

Purpose:

 

To evaluate the ability of longitudinal retinal nerve fiber layer (RNFL) measurements obtained with SD-OCT and TD-OCT to detect glaucoma progression.

 

Methods:

 

Circumpapillary TD-OCT (Stratus OCT, Fast RNFL scan), and SD-OCT scans with and without inter-visit scan alignment (RTVue-100, ONH and RNFL scans, respectively), visual fields (VF), and disc photos were obtained from 110 eyes (24 healthy, 48 glaucoma suspect, and 38 glaucoma) having ≥4 visits. Progression was defined by VF Guided Progression Analysis, VF index, and glaucoma experts’ evaluation of disc photos. Mean and quadrant RNFL thickness rate of change over time were analyzed for progressor (PR) and non-progressor (NPR) eyes using linear mixed effects models accounting for image quality.

 

Results:

 

The average length of follow-up was 3.2 years, with 10 eyes progressing. The average baseline VF mean deviation (MD) was -0.64 dB for NPR and -1.43 dB for PR. No statistically significant difference in the rate of RNFL thickness change between PR and NPR was observed for both TD- and SD-OCT (Table). TD- and non-aligned SD-OCT (but not in the aligned SD-OCT group) had a more negative slope (thinning) for PR eyes compared to NPR (except for SD-OCT nasal and inferior quadrants).

 

Conclusions:

 

Circumpapillary scanning with both TD-OCT and SD-OCT showed RNFL thinning in the PR group; however, the change in PR was not statistically significantly different from that in the NPR group. SD-OCT aligned ONH scan RNFL thickness measurements, did not show a more negative slope in the PR group. This might reflect the variability associated with interpolation in the alignment algorithm of the ONH scan pattern. (TD-Time Domain, SD-Spectral Domain, SD ALN-Aligned)  

 

Keywords: imaging/image analysis: clinical • nerve fiber layer