In the first group, a total of 110 eyes of 57 healthy subjects were included. Four eyes from four subjects were excluded because of the suboptimal quality of the OCT scans. In the second group, 10 fellow eyes of patients with unilateral retinal disease (comprising six cases of branch retinal vein occlusion, three cases of central serous chorioretinopathy, and one case of ischemic optic neuropathy) were separately included and analyzed.
In the first group, there were 34 women and 23 men, with a mean (±SD) age of 31.7 ± 6.8 years (range, 18–45). Among them, 48 were of European and 9 were of Afro-Caribbean descent. The mean axial length was 24.29 ± 1.39 mm (range, 20.7–26.9). The distribution of all parameters was in accordance with a Gaussian distribution (not shown).
Within eyes, superior, nasal, and inferior MRT were on average not significantly different (332 ± 16, 329 ± 17, and 325 ± 15 μm, respectively) and correlated strongly (ICC, 0.875–0.915), whereas MRT was significantly lower temporally (306 ± 16 μm; P < 0.001). The FIRA values in the temporal, superior, and inferior areas correlated highly (ICC > 0.860); the nasal FIRA showed smaller ICCs than other areas (0.775–0.806). The pit diameter was higher horizontally than vertically (2205 vs. 2450 μm; P < 0.001). Between fellow eyes, there was a high degree of symmetry, as demonstrated by the high ICCs of all morphometric parameters (CFT, 0.850; pit surface, 0.932; pit depth, 0.952; pit diameter, 0.810; nasal MRT, 0.805; temporal MRT, 0.725; and FIRA between 0.865 and 0.873).
Examination of OCT scans showed a clear trend toward an increasing pit depth and increased inner layer separation with decreasing CFT (
Video 1 and
Fig. 2). In seven eyes (6.3% of the whole population) of four subjects, a shallow, triangle-shaped foveal depression with continuity of the inner nuclear layer (INL) over the fovea was observed (
Fig. 2, top). By contrast, eight eyes (7.3% of the whole population) of six subjects had a flat central foveal reflex with a large separation of the inner layers (
Fig. 2, bottom). The difference in CFT between these two types of foveae was significant (161 vs. 232 μm;
P < 0.001).
Morphometric variables of the whole population of this group are shown in
Table 1. Among the measured parameters, those related to the morphology of the pit (depth, surface, and FIRA) showed the highest variability, whereas MRT showed the lowest.
The correlation matrix of the study parameters is given in
Table 2. This showed that parameters related to the pit depth and separation of inner layers correlated inversely with central thicknesses. The only parameter significantly associated with axial length was the pit diameter. Age did not correlate with foveal shape or structure (data not shown).
Figure 3 illustrates the variations in the morphometric parameters between different subgroups of patients. The Afro-Caribbeans had significantly thinner central thicknesses and a higher degree of separation of inner retinal layers than did the Europeans. CFT was indeed on average 9% thinner and the pit surface 21% wider in Afro-Caribbeans than in Europeans (
P < 0.001 for both). The women also showed a trend toward a thinner CFT (average 10.5 μm;
P = 0.059).
In the group of participants who underwent fluorescein angiography, we observed that the perifoveal capillary ring projected on the inner edge of the ganglion cell layer (
Fig. 4, top). Overall, the surface of the FAZ correlated with the degree of inner layer separation (
Fig. 4, bottom).