Purpose: To assess the effect of atypical pattern of retardation (APR) on retinal nerve fibre layer (RNFL) measurements made by scanning laser polarimetry (SLP) with variable corneal compensation (GDx-VCC) in glaucomatous eyes.
Methods: One eye each of 30 glaucomatous patients (average mean deviation (MD): -6.4+/-4.8) with APR on GDx-VCC retardation map were selected. In total, 34 glaucomatous, age- and severity-matched eyes (average MD: -7.0+/-5.3) and 36 age-matched healthy subjects, both with a normal pattern of retardation (NPR) represented control groups. APR on retardation maps was characterized by alternating peripapillary circumferential bands of low and high retardation, or high retardation areas arranged in a spokelike pattern, or high retardation nasal and temporal splotchy areas. Typical scan score (TSS) was extracted for each included eye. GDx-VCC parameters (mean+/-SD) in the two groups of glaucomatous eyes were compared with healthy eyes' corresponding values (Mann-Whitney U-test). Areas under receiver operating characteristic (AUROC) curves were generated to assess the APR effect on the parameters' diagnostic ability.
Results: All parameters discriminated adequately between healthy and glaucomatous eyes with NPR (AUROCs > or =0.9 for nine parameters). On the contrary, considering healthy and glaucomatous eyes with APR, four thickness parameters could not separate the two groups and AUROCs > or =0.85 appeared only for Inferior and Superior Ratio, NFI, Max Modulation.
Conclusion: APR may void the effect of custom compensation and provide spurious RNFL thickness measurements. When a printout of glaucomatous eyes with APR is evaluated, it is proper to rely on ratios, modulation parameters, and NFI, since the diagnostic ability of thickness parameters is significantly reduced.