Quantitative measurement of retinal nerve fiber layer thickness by scanning laser polarimetry in normal eyes

Objective To assess the reproducibility of scanning laser polarimetry GDxVCC in retinal nerve fiber layer (RNFL) thickness measurement in normal eyes, and state the normative database of normal people's RNFL thickness parameters measured by GDxVCC. Methods One hundred and twenty-three eyes of 67 normal subjects were underwent RNFL thickness measurement by GDxVCC. The measurements were repeated for 3 times, and the average thicknesses were calculated. The differences of RNFL thickness measurements between sex, right and left eyes were compared. The differences of RNFL thickness measurements between superior part and inferior part of the same eyes were also compared. The intraclass correlation coefficients (ICC) of 3 measurements in every eye were calculated. Results The average RNFL thickness of 123 eyes of 67 normal subjects were: (70.30±6.76) μm in superior part, (67.35±6.77) μm in inferior part and (56.87±4.53) μm in mean; the average TNSIT standard deviation was 23.68±4.61; the average symmetric value of right and left eyes was 0.86±0.11. There was significant difference between superior part and inferior part of the same eyes (t=4.952, P=0.000). And the inferior RNFL thickness and TNSIT standard deviation between right and left eyes showed significant difference (P=0.005, 0.002), while the superior RNFL thickness and mean RNFL thickness between right and left eye showed no statistical difference (P=0.086, 0.529). And there is no significant difference in RNFL thickness parameters between different sexes. The ICC values of all TSNIT parameters were 0.5, and the ICC values of superior, inferior and mean RNFL thickness were 0.8. Conclusion The reproducibility of GDxVCC in retinal nerve fiber layer (RNFL) thickness measurement in normal eyes was very good. But the effects of age on RNFL thickness measurement by GDxVCC need further research.