Purpose
To test the glaucoma diagnostic accuracy of a method of estimating retinal ganglion cell (RGC) counts, using a combination of standard automated perimetry (SAP) and retinal nerve fiber layer (RNFL) thickness assessment by optical coherence tomography (OCT), in a Brazilian cohort.
Methods
Observational cross-sectional study including 73 eyes of 44 glaucomatous patients and 95 eyes of 58 healthy subjects. Eyes were classified as glaucomatous if they had evidence of glaucomatous optic neuropathy and repeatable visual field defects. Controls were recruited from the general population. All eyes were tested with 24-2 SAP and spectral domain OCT. Estimates of RGC counts were obtained according to a previously described algorithm (Medeiros et al. Arch Ophthalmol. 2012 Sep;130(9):1107-16). Receiver operating characteristic (ROC) curves were used to evaluate diagnostic accuracy. Data were corrected for age (ROC regression model) and correlation between eyes.
Results
SAP Mean Deviation (mean ± SD) was -5.23 ± 6.27 and -1.13 ± 1.17 dB (P<0.001), and mean RNFL thickness was 82.20 ± 18.02 and 100.7 ± 10.24 μm (P<0.001) in glaucomatous and healthy eyes, respectively. Mean estimated RGC counts were 683,192 ± 256,824 and 988,961 ± 146,360 in glaucomatous and healthy eyes, respectively (P<0.001). Estimated RGC counts performed better than OCT and SAP isolated for discriminating glaucomatous from healthy eyes, with ROC curve areas of 0.85, 0.81 and 0.67, respectively. There was a strong correlation between RGC estimates obtained from SAP and OCT data for all exams from the 168 eyes included in the study group (r = 0.73; P<0.001).
Conclusions
Estimates of RGC counts based on a combination of structural and functional tests had excellent accuracy for discriminating glaucomatous from healthy eyes, and constitute a better method for staging the disease.