Retinal nerve fiber layer measurements using laser scanning polarimetry in different stages of glaucomatous optic nerve damage

PURPOSE: To evaluate the diagnostic value of polarimetric measurements of the retinal nerve fiber layer (RNFL) thickness in different stages of glaucomatous optic nerve damage. METHODS: The study included 92 eyes of 46 controls (age 41.0+/-13.7 years) and a heterogeneous group of 232 eyes of 135 patients with different stages of glaucomatous optic nerve damage (age 54.0+/-10.2 years; 68 patients with primary open-angle glaucoma, 56 with normal-pressure glaucoma and 11 patients with secondary glaucoma due to primary dispersion syndrome or pseudoexfoliation syndrome). All control subjects and patients underwent complete ophthalmological examinations including scanning laser polarimetry of the RNFL using the GDx (Laser Diagnostic Technologies, San Diego, Calif.) and 15 degrees color stereo optic disc photographs. Only subjects and patients with disc area less than 3.4 mm(2) were included in the study. The total glaucoma group were divided into four subgroups according to the morphological criteria of the neuroretinal rim. RESULTS: The stage of morphological glaucomatous optic nerve damage was classified as follows: stage 0: n=92, stage 1: n=103, stage 2: n=65, stage 3: n=40, and stage 4: n=19. Differences in mean polarimetric retardation between controls and eyes with glaucoma were significant for all parameters except the variable symmetry. The most significant differences between controls and eyes with glaucomatous optic nerve damage were found with the "number" variable assigned by the neural network analysis ( P<0.001). With increasing stage of glaucomatous optic nerve damage, separation of the variable "the number" increased significantly. At a predetermined specificity of 90% the sensitivity of the groups with different stages of morphological glaucomatous optic nerve damage increased from 32% for stage 1 to 90% for stage 4. CONCLUSION: Polarimetric measurement of the RNFL thickness is significantly associated with morphological glaucomatous optic nerve damage. The fast performance, easy handling, and low cost of RNFL polarimetry mean that it can be included in the routine examination of glaucoma patients. Further study and refinement of this technique are indicated to improve its usefulness in both clinical diagnosis and in population-based case identification.