Reproducibility of Circumpapillary Retinal Nerve Fiber Layer Thickness Measurements Evaluated by Circle and Annulus Area With and Without Correction for Ocular Rotation.

PURPOSE

To compare the reproducibility of annulus area measurements of the circumpapillary retinal nerve fiber layer thickness (cpRNFLTannulus) using spectral-domain optical coherence tomography (SD-OCT) with that of conventional cpRNFLT (cpRNFLTring) measurements and evaluate the effect of correction of ocular rotation on the reproducibility.

MATERIALS AND METHODS

Peripapillary SD-OCT raster scans were performed on 2 different days in 48 normal subjects and 59 patients with glaucoma using the Topcon OCT-1000. In the raster scan dataset, the cpRNFLTring and cpRNFLTannulus were averaged along a circle (3.4-mm diameter) and in an annulus area (diameters from 2.8 to 4.0 mm) centered on the barycenter of the disc. The measurement reproducibility, effects of ocular rotational correction, and factors affecting the reproducibility were studied.

RESULTS

The intervisit reproducibility values, expressed as the coefficients of variation, for the cpRNFLTannulus, were significantly smaller than those for the cpRNFLTring measurements in sectors 30 degrees wide in glaucomatous eyes (P=0.006) and in sectors 10 degrees wide in normal and glaucomatous eyes (P<0.001). The ocular rotational angles averaged 7.1 and 7.2 degrees with intervisit differences of 1.4 and 1.8 degrees, respectively, in normal and glaucomatous eyes. Correction for the ocular rotation had little effect on the reproducibility. The mean deviation and cpRNFLT were correlated negatively with the coefficient of variation (β=-0.93 to -0.83; P=0.027 to 0.028).

CONCLUSIONS

The cpRNFLTannulus measurement yielded better reproducibility than the cpRNFLTring measurement, and multiple linear regression analysis showed that the more advanced the glaucomatous damage, the worse the intervisit reproducibility.