Purchase this article with an account.
vahid mohammadzadeh, Alessandro Rabiolo, Navid Amini, Esteban Morales, Sharon Henry, Anne L Coleman, Simon K. Law, Joseph Caprioli, Kouros Nouri-Mahdavi; Longitudinal Structure-Function (SF) Relationships in The Macula. Invest. Ophthalmol. Vis. Sci. 2019;60(9):3920.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate the relationship between longitudinal changes in macular OCT thickness and changes in central visual field (VF) sensitivity in a cohort of glaucoma eyes with central damage at baseline.
123 eyes from UCLA's Advanced Glaucoma Progression Study with ≥3 years of follow-up and 5 macular OCT (Spectralis SD-OCT) and 10-2 VF tests were selected. OCT superpixels and VF locations were matched correcting for retinal ganglion cell displacement. We grouped the central 24 superpixels and matched locations into 3 concentric circles (3.4°, 5.6°, and 6.8° eccentricities, Figure 1) and defined superior and inferior hemiretina/hemifield pairs. We estimated structural and functional rates of change with linear regression and calculated change from baseline for ganglion cell/inner plexiform layer (GCIPL) and total deviation values at matching superpixels/locations, circles, and hemiretina/hemifield pairs. We investigated the correlation between SF rates of change and corresponding changes from baseline with parametric and nonparametric analyses.
The median (IQR) follow-up and number of exams were 3.7 (3.2-4.3) years and 7 (6-9). The highest correlation was observed between the rates of change in the inferior hemiretina and superior hemifield (Pearson’s r=0.571, p<0.001). Among the 3 circles explored, only structural and functional rates of change at 6.8° eccentricity demonstrated a significant association (r=0.353; p<0.001). In contrast, changes in structure and function from baseline demonstrated significant correlations at all circles (r = 0.298, 0.486, and 0.294 at 3.4°, 5.6°, and 6.8°, respectively). Inferior hemiretina/superior hemifield SF changes from baseline were significantly correlated (r=0.352; p<0.001, Figure 2). GCIPL rates of change decreased as baseline GCIPL diminished (spearman’s r=–0.309; p<0.001). 29% and 26% of superpixels and locations demonstrated a clinically significant change from baseline (i.e., >3 microns and 3 dB).
Longitudinal SF changes show fair to moderate correlations and potentially provide complementary information. The correlation of SF changes from baseline was stronger and was detected more frequently compared to those for rates of change, likely a reflection of relatively short follow-up. Event analysis may be more helpful during early follow-up of glaucoma patients for detection or prediction of change.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only