advertisement

Editors Selection IGR 18-3

Comments

Comment by Harry Quigley on:

48060 Retinal Ganglion Cell Layer Thickness and Local Visual Field Sensitivity in Glaucoma, Raza AS; Cho J; De Moraes CGV et al., Archives of Ophthalmology, 2011; 129: 1529-1536

Find related abstracts

Hood and Kardon¹ found that a simple linear model (SLM) was successful in relating RNFL thicknesses for IT & ST optic disc sectors to corresponding 24-2 perimetric sensitivities, and this was independently corroborated by Leite et al.². A third independent study found that the SLM was successful in relating area of temporal neuroretinal rim with macular perimetric sensitivities.³ A new study from the Hood lab, Razi et al. 2011, took another important step by applying the SLM to SD-OCT data on thickness of the RGC+IPL segment. In a normal human retina, the visual field within 9° of fixation is served by one-third of the retinal ganglion cell population, but tested by less than 1/10 of the 24-2 test locations.

In a normal human retina, the visual field within 9° of fixation is served by one-third of the retinal ganglion cell population, but tested by less than 1/10 of the 24-2 test locations

The 10-2 tests 68 locations within 9° of fixation, and the authors show how to relate these visual field locations with retinal locations of ganglion cell bodies. This then allowed them to compare perimetric sensitivities at individual locations with local thickness of RGC+IPL. The only free variable for the SLM is the mean residual thickness for blind regions, which reflects non-neural components such as glia and vasculature. Raza et al. elegantly show how the utility of an anatomical measurement can be assessed in terms of the difference between mean normal and residual, along with confidence limits for normal between-subject variability and for variability of the residual. Raza et al. found that RGC+IPL inside 7°, unlike RNFL outside this region, had a trend towards greater loss in the neural component than in perimetric sensitivity. Shafi et al. 2011 did not find this with temporal rim area versus 10-2 sensitivities, perhaps due to differences in patient selection, residual thickness, maps of perimetric locations to anatomy, or extent of pooling across test locations. The SLM provides a powerful way to compare different imaging methods, and it would be useful to compare 10-2 data with both rim area and RGC+IPL thickness measured in a single group of patients.

• ← Previous
• Next →