Peaks ofcircumpapillary retinal nerve fibre layer and associations in healthy eyes: the Beijing Eye Study 2011.

AIMS To investigate the distribution and associations of the circumpapillary retinal nerve fibre layer thickness (RNFLT) profile, characterised as peak height and peak position, in healthy eyes. METHODS 667 healthy participants (294 male) were randomly selected from the Beijing Eye Study 2011. RNFLT was measured at 768 points at 3.4 mm circumpapillary position using spectral-domain optical coherence tomography (OCT). The location and height of the superior temporally peak (PeakST), superior nasal peak (PeakSN) and inferior temporal peak (PeakIT) were assessed. RESULTS The RNFLT was thickest at PeakIT (194±25 µm; location: 288±12°), followed by PeakST (182±25 µm; 73±10°) and PeakSN (125±23 µm; 135°±13°). In multivariable analysis, peak RNFLT decreased with longer axial length (p<0.001; correlation coefficient beta: -0.18 to -0.15; all peaks), older age (all p<0.01, beta: -0.10; PeakST and PeakIT), female gender (p=0.026, beta: -0.09; PeakST), and larger parapapillary gamma zone and beta zone width (p≤0.004, beta: -0.16 to-0.11; PeakSN). The temporal peaks were located more closely to the horizontal line in women (p≤0.020, beta: 0.09-0.11) and with longer axial length (p<0.001, beta: 0.27-0.31), while they were located more inferiorly in eyes with larger Bruch's membrane openings (BMOs) (p≤0.01, beta:0.10~0.11). CONCLUSIONS Peak RNFLT decreased by 0.34 µm for each increase of year in age (PeakST and PeakIT), by 3.2-3.5 µm for each 1 mm increase in axial length (all three peaks), and was 4.5 µm thinner in women than in men. The position of temporal peaks depended on gender, axial length and BMO diameter. These associations should be taken into count in OCT-based RNFLT assessment for disease finding, especially in glaucoma evaluation.