ABSTRACT
PURPOSE: To compare the scanning laser polarimetry device with a fixed corneal compensator (GDx FCC) with the new version with a variable corneal compensator (GDx VCC) in healthy subjects. METHODS: A prospective, nonrandomized, comparative trial was carried out with both the FCC and the VCC systems on 130 healthy volunteers. The FCC measurements were divided into 2 subgroups, in accordance with the appearance of the macula. Subjects with a uniform aspect of the macula were included in the macula-'negative' group (macula-), whereas subjects with an irregular macular appearance with FCC were included in the macula-'positive' group (macula+). Data relative to all the GDx parameters were examined. For each parameter, we compared the data obtained by the FCC to the VCC models. Subsequently we compared each FCC subgroup (macula+ and macula-) to VCC. RESULTS: A total of 67 subjects entered the macula- group and 63 the macula+ group. FCC significantly overestimated the average parameters and underestimated the number, the modulation and also the ratio parameters. The only parameter showing no difference was symmetry. Similar trends were also evident when the subjects were divided into the macula+ and - subgroups. CONCLUSION: Individual variations of corneal birefringence, which are not completely corrected by FCC, result in over- or underestimation of retinal nerve fiber layer parameters. This measurement error cannot be eliminated, not even by only selecting patients with a uniform aspect of macular scan. For this reason GDx FCC does not seem to be an acceptable tool in the early diagnosis of glaucoma, and modern versions of this device should be preferred.
