Three-Dimensional Bruch’s Membrane Shape Change Over Time with Acetazolamide Treatment in the Idiopathic Intracranial Hypertension Treatment Trial (IIHTT)

Abstract

Purpose : We demonstrated that the Bruch’s membrane (BM) shape change from toward the vitreous to away from vitreous over time is more significant with treatment of diet+acetazolamide (ACZ) than diet+placebo using a 2D BM shape model in IIHTT; next, we extended the 2D BM shape model into a 3D version (Wang et al. ARVO 2015, 2016). In this work, after applying the 3D shape model to the IIHTT data, we further analyzed the 3D shape change over time and the treatment effect.

Methods : The 3D BM and BM opening (BMO) contour was first segmented in each spectral-domain optical coherence tomography (SD-OCT) optic-nerve-head (ONH) volumetric scan using our prior approach. Next, on the pre-segmented BM surface, 288 landmarks were placed along the obtained BMO contour; for every 10^o, eight equidistant landmarks were chosen covering 1.5 mm in the radial direction. The flowchart is shown in Fig. 1. By applying principal component analysis (PCA), the first component coefficient was computed to quantify the BM shape change towards/away from the vitreous in 3D (Fig. 2a).

Results : In the IIHTT OCT sub-study, of the 116 right eyes used to generate the baseline 3D BM shape model, 56 subjects had successfully registered ONH and HD raster scans at all study visits. In the ACZ (n=31) group, the 3D BM shape showed significant change after 3 months (p<0.05); the mean change from baseline to 3 (6) months was 0.32±0.94 (0.61±1.02). In the placebo group (n=25), the mean shape change from baseline to 3 (6) months was 0.08±0.51 (0.11±0.64); no significant changes over time (Fig. 2b). The mean 3D BM shape change from baseline to 6 months in the ACZ group was significantly greater than in the placebo group (p<0.01).

Conclusions : BM shape measures in 2D and 3D are highly correlated (ρ=0.81, p<0.01) and have been shown as potential biomarkers of successful treatment of papilledema. The 3D shape measure, compared to the 2D version, has an additional advantage of consolidating 3D information. Developing a fully-automated method and incorporating more principal components from the 3D BM shape models constitute future efforts.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

 

Fig. 1 Flowchart of generating 3D BM shape models

 

Fig. 2 (a) 3D BM shape model variations (b) the 3D shape measures in ACZ/placebo groups at baseline, three and six months