[Variability and reproducibility of 3 methods for measuring the thickness of the nerve fiber layer].

OBJECTIVE To estimate the variability and reproducibility of confocal tomography (HRT), scanning laser polarimetry (GDx) and optical coherence tomography (OCT-Cirrus) to determine the thickness of the layer of ganglion fibers. METHOD A total of 75 normal eyes were examined twice. Inter-individual variability was analyzed after standardizing the results. The coefficient of variation was used to measure the variability between tests, and the Pearson coefficient was used to analyze the correlation between variables. RESULTS The inter-individual variability was similar in GDx (8.9%) and OCT (11.1%), but very high in HRT (30.0%). No instrument detected significant changes with age. The coefficient of variation of the total thickness between the examinations of the same subject was significantly lower (P<.05) in GDx (1.4) than in OCT (2.0), but very high in HRT (6.4). The same was true when analyzing the upper fibers (GDx=1.8, OCT=2.9, HRT = 6.6), but not with the lower ones, where the only significant differences were observed with HRT (GDx = 2.2, OCT = 2.7, HRT = 7.0). Among the results of OCT and GDx, there was a significant correlation when comparing the first (r=0.46, P<.0001) and second examinations (r=0.52, P<.0001). However there was no significant relationship between the data provided by HRT for the two remaining instruments (P>.05). CONCLUSIONS There is a wide variation in the inter-individual and inter-test measurement of the thickness of the of nerve fibers layers using HRT. GDx has, in this respect, slight advantages over OCT.