advertisement
Editors Selection IGR 17-3
Structure and function measurements: GDxVCC in myopic eyes
Comment by David Greenfield on:
20382 Performance of GDx VCC in eyes with peripapillary atrophy: comparison of three circle sizes, Kunimatsu S; Tomidokoro A; Saito H et al., Eye, 2008; 22: 173-178
Glaucoma diagnosis and monitoring in myopic eyes is challenging. Standard automated perimetry often demonstrates enlargement of the physiologic blind spot, and eyes with considerable axial length elongation or posterior staphyloma often have a refraction-associated scotoma. Clinical examination of the optic disc is difficult in eyes with significant myopic tilt or parapapillary atrophy. Computerized imaging of the optic disc and retinal nerve fiber layer (RNFL) is limited by several factors including absence of highly myopic subjects from inclusion in the normative database, artifact introduced by scanning within the zone of parapapillary atrophy (PPA), and a high prevalence of atypical birefringence patterns using the GDxVCC.
Scanning within the zone of PPA should be avoidedThe study by Kunimatsu et al. (189) compares the performance of GDxVCC in myopic eyes with a mean refractive error -3.9D and PPA using three scanning circles of varying diameters. Several findings in this study are noteworthy. Not surprisingly, scanning in the zone of PPA was highest using the smallest diameter circle (43%) compared to the medium (14%) and large (4%) diameter circles. In both normal and glaucomatous subjects, RNFL thickness measurements significantly decline with increasing diameter of the measurement circle due to the physiologic attenuation in the RNFL with increasing distance from the optic disc margin. This may adversely impact measurement reproducibility for some parameters (e.g., coefficient of reproducibility for TSNIT average in glaucomatous eyes was 4.5 vs 6.2 using the small and medium circle, respectively). Furthermore, this may adversely impact the discriminating power for some parameters (e.g., area under the ROC curve for TSNIT average was 0.9 vs 0.8 using the small and medium circle, respectively). This interesting study emphasizes the challenges associated with image interpretation in myopic eyes with PPA. As demonstrated by Kunimatsu and colleagues, scanning within the zone of PPA should be avoided. This applies not only to GDxVCC imaging, but also to time-domain OCT and HRT in which eyes with significant PPA often have incorrect contour line placement. Future advances in RNFL measurement algorithms will likely demonstrate that three-dimension RNFL maps are superior to cross-sectional slices.
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One
