Purpose:
To compare the loss in sensitivity measured with standard automated perimetry (SAP) to local retinal ganglion cell (RGC) layer thickness measured with frequency domain optical coherence tomography (fdOCT).
Methods:
The central ±10° region of 12 patients with glaucoma (7 NTG; 3 XFG; 2 POAG) was tested with SAP (10-2 SITA standard) and scanned with a fdOCT (3D-OCT 1000, Topcon) volume macular scan (6x6 mm region with 128 B-scans). 20 individuals with normal vision served as controls. To compare RGC thickness to SAP (TD: total deviation), two experienced readers segmented the RGC+ layer (i.e. RGC plus inner plexiform layer), as previously described,[1,2] for horizontal B-scans at 11 locations (0, ±1, ±3, ±5 ±7, and ±8°). To account for displacement of the RGC around the fovea, the field points were displaced to correspond to the location of the corresponding RGCs.[3] RGC+ thickness vs. SAP (TD) data were analyzed for 9 eccentricities from 1.4 to 9°.
Results:
RGC+ thickness correlated well (Spearman) with SAP loss within 6° of the fovea (r= 0.72 to 0.67). The correlation was worse (0.54 to 0.20) from 7° outward. A linear model [4] relating RGC+ thickness to linear SAP loss provided a reasonable fit (R²=0.56 to 0.45) for eccentricities less than 6°, but a poorer fit beyond (0.26 to -0.36). However, the fdOCT thickness associated with a given SAP loss tended to be lower than predicted by the model for smaller field losses.
Conclusions:
Within the central 6°, RGC+ layer thickness correlates with local sensitivity loss when the displacement of the RGCs is taken into consideration. The agreement becomes worse beyond 6°.1. Hood et al (2009) IOVS. 2. Wang et al (2009) Arch Ophthal. 3. Drasdo et al (2007) Vis Res; 4. Hood & Kardon (2007)
Keywords: ganglion cells • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • visual fields