Abstract
Purpose :
The fovea to disc (FtoD) angle has been associated with a greater circumpapillary retinal nerve fiber layer (cpRNFL) thickness in the inferior (I), as compared to the superior (S), quadrant in myopic eyes.[1] Given that the angle between the raphe and the fovea-to-disc line (RtoD angle, in Fig. 1) is, on average, less than 180°,[2,3] this angle may be a contributing factor to differences in the cpRNFL in the I vs. S quadrants. Using an en-face technique based upon a single swept source optical coherent tomography (ssOCT) scan, we studied healthy eyes to determine the association between cpRNFL thickness and FtoD and RtoD angles, the latter of which is not affected by head/eye torsion.
Methods :
28 healthy eyes (54.7 ±11.5 yrs) were included if they had a spherical refractive errors between +3 and -6D and a wide-field ssOCT cube scan (9x12 mm, DRI-OCT, Topcon, Inc). Using special purpose software (ATL 3D-Suite),[4] and the inner limiting membrane (ILM) as a reference, 52 um thick en-face slabs of maximum reflective intensity were generated (Fig. 1).[5] The upper and lower limits of the raphe (dashed lines in Fig. 1) were marked based upon where clear bundles appeared to end. The raphe was estimated from the line of best fit through the midpoints between the upper and lower limits. Based upon the centers of the fovea and disc, the FtoD and RtoD angles were determined (Fig. 1). The thickness of the S and I quadrants was determined using commercial segmentation.[6]
Results :
The ratio of I to S quadrant thickness increased with decreased RtoD angle (r=0.46; Fig. 2, top). It also increased with increased FtoD angle (r=0.34; Fig. 2, bottom), although FtoD is affected by head/eye torsion.
Conclusions :
The en-face technique combined with the high resolution, wide-field scans allows reasonably good visualization of the raphe. Using this technique, as predicted, we found that the relative thickness of the I quadrant compared to the S quadrant increased as the RtoD angle decreased. Taking the RtoD angle into consideration may improve the usefulness of quadrant cpRNFL analysis, including the use of the ISNT rule, for diagnostic purposes.
1. Choi et al. IOVS 2014; 2. Chauhan et al. Ophthal. 2014; 3. Huang et al. IOVS, 2014; 4. Fortune et al. IOVS 2014; 5. Hood et al. IOVS 2015. 6. Hood et al. TVST 2016.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.