Abstract
Purpose:
To examine the spatial relationships between the retinal inner nuclear layer (INL), outer nuclear layer (ONL), retinal pigment epithelium (RPE) layer, Bruch's membrane (BM), and choriocapillaris in the parapapillary region.
Methods:
Human eyes enucleated due to uveal melanomas or secondary angle-closure glaucoma were histomorphometrically examined. We compared the tissue dimensions between four groups of eyes categorized based on the presence/absence of high myopia and glaucoma.
Results:
The investigation consisted of 100 globes (axial length: 25.6 ± 3.1 mm; range: 20.0–35.0 mm). In non-highly myopic nonglaucomatous eyes, the INL, ONL, RPE, BM, and choriocapillaris ended approximately at the end of the RPE layer, with no significant (all P ≥ 0.10) difference between the layers in their distance to the RPE-layer end. From non-highly myopic nonglaucomatous eyes to non–highly myopic glaucomatous eyes, highly myopic nonglaucomatous eyes, and eventually highly myopic glaucomatous eyes, the choriocapillaris, INL, and ONL increasingly extended into the beta zone, most marked for the choriocapillaris and least marked for the ONL. A larger extension of the choriocapillaris into the parapapillary beta zone correlated with longer axial length (standardized regression coefficient β, 0.24; B, 23.0; 95% confidence interval [CI], 1.6–44.5; P = 0.04) and wider parapapillary beta zone (β, 0.59; B, 0.32; 95% CI, 0.22–0.41; P < 0.001); a larger extension of the INL correlated with longer axial length (β, 0.34; B, 43.7; 95% CI, 11.6–75.7; P = 0.009), longer gamma zone (β, 0.52; B, 0.28; 95% CI, 0.15–0.41; P < 0.001), and diagnosis of non–highly myopic glaucoma (β, 0.28; B, 267; 95% CI, 80.8–454; P = 0.006); and a larger extension of the ONL into parapapillary beta zone correlated with longer axial length (β, 0.50; B, 32.2; 95% CI, 21.6–42.8; P < 0.001) and wider parapapillary beta zone (β, 0.28; B, 0.10; 95% CI, 0.04–0.16; P < 0.001).
Conclusions:
Nonglaucomatous non–highly myopic eyes differ from highly myopic eyes and glaucomatous eyes in the spatial relationship of the parapapillary tissue layers.
The parapapillary region of the optic nerve head has been differentiated into four zones.
1–13 Alpha is characterized by irregular pigmentation upon ophthalmoscopy and shows histologically an irregularly structured retinal pigment epithelium (RPE) on top of Bruch's membrane (BM). The beta zone, located between the alpha zone and the optic disc border, is ophthalmoscopically a bright region with visible sclera and large choroidal vessels. It is characterized by the presence of BM and the absence of the RPE. The gamma zone, if present, is located between the beta zone and the optic disc border, can ophthalmoscopically be hardly differentiated from the beta zone, and is histologically characterized by the absence of BM and absence of RPE. The delta zone has been defined as a region within the gamma zone and corresponds to an elongated and thinned peripapillary scleral flange.
14 The peripheral border of the delta zone (and of the peripapillary scleral flange) is the merging line of the optic nerve dura mater with the posterior sclera. The central border of the delta zone (and of the scleral flange) is the peripapillary border tissue of the scleral flange.
15 While the four parapapillary zones have histomorphometrically and clinically been defined, the spatial relationship between them and the other tissue layers, namely, the choriocapillaris and the retinal layers, has remained partially unclear yet. We therefore conducted this histomorphometric study to examine the spatial relationships between the retinal inner nuclear layer, the photoreceptor layer, and the choriocapillaris within the parapapillary zones in four groups of eyes categorized based on the presence or absence of high myopia and of glaucoma.
The investigation consisted of 100 globes of 100 patients (61 [61%] men) aged 62.1 ± 15.1 years (range, 30-89 years) with a mean axial length of 25.6 ± 3.1 mm (range, 20.0–35.0 mm). Out of the 100 eyes, 49 (49%) eyes, 39 (39%) eyes, and 15 (15%) eyes had an axial length of ≥25 mm, ≥26 mm, and ≥29 mm, respectively. The study cohort could be subdivided into a non–highly myopic group without glaucoma (41 eyes), a non–highly myopic group with glaucomatous optic nerve atrophy (20 globes), a highly myopic group without glaucoma (19 eyes), and a highly myopic eyes with glaucomatous optic neuropathy (20 globes).
In the non–highly myopic nonglaucomatous group, the photoreceptor layer, choriocapillaris, and inner nuclear layer ended in the majority of eyes at the end of the RPE layer. The photoreceptor layer ended before the RPE end in 3 (3/41 or 7%) eyes, at the RPE end in 31 (76%) eyes, and beyond the RPE end in 7 (17%) eyes. The choriocapillaris ended before the RPE end in 2 (5%) eyes, at the RPE end in 32 (78%) eyes, and beyond the RPE end in 7 (17%) eyes. The inner nuclear layer ended before the RPE end in 7 (17%) eyes, at the RPE end in 29 (71%) eyes, and beyond the RPE end in 5 (12%) eyes. The various layers did not differ significantly (all P ≥ 0.10) in the distance between their end to the RPE end.
In the non–highly myopic glaucomatous eyes, the photoreceptor layer, choriocapillaris, and inner nuclear layer ended in the majority of eyes beyond the RPE end. The photoreceptor layer ended before the RPE end in 2 (2/20 or 10%) eyes, ended at the RPE end in 6 (30%) eyes, and extended beyond the RPE end in 12 (60%) eyes. The choriocapillaris ended before the RPE end in none of the eyes, at the RPE end in 8 (40%) eyes, and beyond the RPE end in 12 (60%) eyes. The inner nuclear layer ended before the RPE end in none of the eyes, at the RPE end in 7 (35%) eyes, and beyond the RPE end in 13 (65%) eyes. The distance between the inner nuclear layer end to the RPE-layer end was significantly larger than the distance between the photoreceptor layer end to the RPE-layer end (133 ± 183 µm vs. 66.5 ± 96.8 µm;
P = 0.001) (
Table).
In the highly myopic nonglaucomatous eyes, the photoreceptor layer, choriocapillaris, and inner nuclear layer ended in the majority of eyes beyond the RPE end. The photoreceptor layer ended before the RPE end in 1 (1/19 or 5%) eye, at the RPE end in 7 (35%) eyes, and beyond the RPE end in 11 (58%) eyes. The choriocapillaris ended before the RPE end in 1 (6%) eye, at the RPE end in 9 (50%) eyes, and beyond the RPE end in 8 (45%) eyes. The inner nuclear layer ended before the RPE end in 2 (12%) eyes, at the RPE end in 5 (31%) eyes, and beyond the RPE end in 9 (57%) eyes. The distances between the end of the various tissues to the RPE-layer end did not vary significantly (P > 0.05) between the inner nuclear layer, photoreceptor layer, and choriocapillaris.
Also in the highly myopic glaucomatous eyes, the photoreceptor layer, choriocapillaris, and inner nuclear layer ended in the majority of eyes beyond the RPE end. The photoreceptor layer ended before the RPE end in 1 (1/20 or 5%) eye, at the RPE end in 7 (35%) eyes, and beyond the RPE end in 12 (60%) eyes. The choriocapillaris ended before the RPE end in none of the eyes, at the RPE end in 9 (45%) eyes, and beyond the RPE end in 11 (55%) eyes. The inner nuclear layer ended before the RPE end in 1 (5%) eye, at the RPE end in 6 (30%) eyes, and beyond the RPE end in 13 (65%) eyes. The distance between the inner nuclear layer end and the RPE-layer end was significantly larger than the distance between the photoreceptor layer end and the RPE-layer end (266 ± 395 µm vs. 211 ± 286 µm;
P = 0.03). The differences in the distances between the other tissue layer ends to the RPE-layer end were not significant (
Table).
Comparing the groups with each other revealed that the photoreceptor layer extended significantly further (
P = 0.001) beyond the RPE end in the highly myopic glaucomatous group (
P < 0.001), the highly myopic nonglaucomatous group (
P = 0.001), and the non–highly myopic glaucomatous group (
P = 0.01) than in the non–highly myopic nonglaucomatous group (
Table). In a similar manner, the choriocapillaris and the inner nuclear layer extended wider beyond the RPE end in both the highly myopic group and in the non–highly myopic glaucomatous group than in the non–highly myopic nonglaucomatous group (
Table) (
Figs. 4–
7).
The alpha zone was significantly the largest in the non–highly myopic glaucomatous group compared with the non–highly myopic nonglaucomatous group (
P = 0.02), in which it was larger than in the highly myopic nonglaucomatous group (
P = 0.006) and in the highly myopic nonglaucomatous group (
P = 0.006), while both latter groups did not vary significantly in alpha zone size (
P = 0.95). The beta zone was significantly (
P < 0.001) the smallest in the non–highly myopic nonglaucomatous group compared to any of the three other groups, which did not differ significantly in beta zone width (all
P > 0.50). The gamma zone and delta zone were significantly the smallest in the non–highly myopic nonglaucomatous group and the non–highly myopic glaucomatous group as compared to both highly myopic groups (
Table).
In univariate analysis, the distance of the photoreceptor layer end to the RPE-layer end correlated with the distance of the choriocapillaris end to the RPE-layer end (
P < 0.001), the distance of the inner nuclear layer end to the RPE-layer end (
P < 0.001), the width of the gamma zone (
P < 0.001) and the delta zone (
P < 0.001), and axial length (
P < 0.001) (
Fig. 4). It was not significantly associated with age (
P = 0.10) and sex (
P = 0.99). In multivariable linear regression analysis, a larger distance of the photoreceptor layer end to the RPE-layer end remained significantly (correlation coefficient
r2 = 0.42) associated with longer axial length (β, 0.50; B, 32.2; 95% CI, 21.6, 42.8;
P < 0.001) and wider beta zone (β, 0.28; B, 0.10; 95% CI, 0.04, 0.16;
P < 0.001).
A longer distance between the choriocapillaris end and the RPE-layer end was associated (multivariable analysis;
r2 = 0.40) with longer axial length (β, 0.24; B, 23.0; 95% CI, 1.6, 44.5;
P = 0.04) and longer beta zone (β, 0.59; B, 0.32; 95% CI, 0.22, 0.41;
P < 0.001), while gamma zone width was not significantly correlated (
P = 0.07). A larger distance between the inner nuclear layer end to the RPE-layer end correlated (multivariable analysis;
r2 = 0.60) with longer axial length (β, 0.34; B, 43.7; 95% CI, 11.6, 75.7;
P = 0.009) (
Fig. 5), longer gamma zone (β, 0.52; B, 0.28; 95% CI, 0.15, 0.41;
P < 0.001), and the diagnosis of non–highly myopic glaucoma (β, 0.28; B, 267; 95% CI, 80.8, 454;
P = 0.006). It was not significantly correlated with sex (
P = 0.94).
This histomorphometric study showed that in non–highly myopic nonglaucomatous eyes, the choriocapillaris, the photoreceptor layer, and the inner nuclear layer extended approximately to the RPE-layer end. In non–highly myopic glaucomatous eyes, all three layers, the choriocapillaris, the photoreceptor layer, and the inner nuclear layer, extended usually beyond the RPE-layer end into the RPE-free zone of the BM (i.e., beta zone), with the photoreceptor layer ending closest to the RPE-layer end and the inner nuclear layer ending most distant to the RPE-layer end (i.e., closest to the optic disc border). In highly myopic nonglaucomatous and glaucomatous eyes, the photoreceptor layer and the inner nuclear layer extended furthest into the RPE-free compartment (beta zone) and beyond into the BM-free compartment (i.e., gamma zone) (
Figs. 1–
7).
The observations made in our study agree with results of previous histomorphometric and clinical studies that the beta zone is larger in non–highly myopic glaucomatous eyes and in highly myopic eyes than in non–highly myopic nonglaucomatous eyes.
1–13,18,19 It complements the previous results by the findings that in both non–highly myopic glaucomatous eyes and highly myopic eyes, the choriocapillaris and the retinal layers extend beyond the RPE end into the beta zone and, in the case of some highly myopic eyes, additionally into the gamma zone. Psychophysically, the presence of a photoreceptor layer in the peripheral parts of the beta zone may not imply that this region of the beta zone does not belong to an enlarged blind spot in perimetry, since it may be unlikely that photoreceptors can function without the underlying RPE.
20
The observation of an overhanging of the retinal layers into the beta zone in the glaucomatous group and in the highly myopic group in contrast to nonglaucomatous non–highly myopic eyes may give the basis for different hypotheses. In the case of highly myopic eyes, the notion has been put forward that axial elongation occurs through a retina-induced growth of BM in the midperiphery of the fundus.
21 The notion is supported by the findings of the present and previous studies that the optic nerve exit in moderately myopic eyes runs in an oblique anterior direction. Such a morphology is explainable by a posterior shift of BM opening in relationship to the lamina cribrosa within the optic nerve head canal. The posterior BM opening shift may be induced by a growth of BM in the fundus midperiphery. The BM opening shift leads to an absence of BM in the temporal parapapillary region (i.e., gamma zone). If BM as the underlying layer of the retina is drawn into the posterior direction, the retina, connected to the optic disc by the inner limiting membrane and the retinal nerve fibers, may not fully follow the BM and may partially stay back. It would lead to an overhanging of the retinal layers into the gamma zone in moderately myopic eyes. Since the deep retinal layer (i.e., the outer nuclear layer) is closer to the shifting BM, and since the inner nuclear layer is closer to the retinal nerve fiber layer, the notion would also agree with the finding of the present study that the inner nuclear layer as compared to the outer nuclear layer extended wider into the gamma zone. While the findings of the present study support such a notion, the design of the study does not allow proving the hypothesis.
A recent histologic study reported on morphologic differences in the parapapillary beta zone between glaucomatous eyes and myopic eyes.
17 It was postulated that the myopic beta zone developed due to an axial elongation-associated enlargement of the RPE opening of the optic nerve head canal.
17 It might explain why myopic eyes with the beta zone as compared to glaucomatous eyes with the beta zone had a significantly smaller alpha zone with RPE irregularities and parapapillary RPE drusen. The notion of an axial elongation-related enlargement of the RPE opening fits with the observations made in the present study. If the RPE opening enlarges, the RPE would slip on the BM away from the optic disc border. Since, as also pointed out above, the inner retinal layers more than outer retinal layers are connected to the optic disc, the RPE movement away from the optic disc may not fully be followed by the photoreceptor layer and even less by the inner nuclear layer. It would lead to an overhanging of the photoreceptors and, more so, of the inner nuclear layer into the (myopic) beta zone, as observed in the present study. One may speculate that the discrepancy between the outer retinal layers and inner retinal layers in the amount they follow the shift of the BM and the RPE may lead to an intraretinal strain between the inner and outer retinal layers, perhaps eventually resulting in a peripapillary myopic retinoschisis. Again, while the findings of the present study support such a notion, the design of the study does not allow proving the hypothesis.
Interestingly, the choriocapillaris also extended beyond the RPE end into the beta zone. It may suggest that it was not primarily a choriocapillaris insufficiency that might have led to an RPE loss and the development of a myopic beta zone. The histomorphometric findings obtained in our study on enucleated human globes agree with clinical images of the optic nerve head obtained by optical coherence tomography (OCT) and show an extension of the photoreceptor layer and inner nuclear layer into the beta zone and gamma zone in myopic eyes.
The observation of an overhanging of the deep and middle retinal layers into the beta zone of the glaucomatous eyes examined in the present study fits with the notion that the development of the beta zone in glaucomatous eyes may be associated with a loss of parapapillary RPE cells, as suggested in a previous OCT-based clinical study.
22 Correspondingly, a recent histomorphometric study reported that non–highly myopic glaucomatous eyes as compared to highly myopic eyes showed RPE drusen and a thickening of the parapapillary BM or sublaminar deposits on top of the BM in their parapapillary region in association with a (glaucomatous) beta zone.
17 Fitting with the notion of an RPE cell loss as cause for the development of the beta zone in glaucomatous eyes is that the choriocapillaris extended beyond the RPE end into the beta zone. It suggests that an insufficiency of the choriocapillaris may not have been the reason for the absence of RPE in the beta zone in glaucomatous eyes.
Limitations of our investigations should be considered when its results are discussed. First, the globes examined showed severe intraocular diseases that had led to their enucleation, so the results of our study cannot directly be transferred to eyes without these disorders. Second, the eyes in the glaucomatous group in our study had a severe form of secondary angle-closure glaucoma, so it is primarily doubtful whether the findings are valid also for eyes with primary or secondary open-angle glaucoma. Third, the patients included in the study were Caucasians, so future studies may address whether the results can be extended to other ethnicities. Fourth, the recruitment of patients occurred in a hospital-based manner, so a referral bias cannot be ruled out. Fifth, as in any histomorphometric study, artifacts due to the histologic processing will have influenced the tissue dimensions. If, however, one assumes that the different tissue layers were affected in a similar manner by these artifacts, the spatial relationship between the tissue layers may have been mostly independent of such potential sources of error.
In conclusion, while in nonglaucomatous non–highly myopic eyes, the photoreceptor layer, the inner nuclear layer, and the choriocapillaris stop at or shortly before the RPE-layer ends in the parapapillary region, they extend into the RPE-free compartment in non–highly myopic glaucomatous eyes and in highly myopic eyes. In highly myopic eyes, the photoreceptor layer and inner nuclear layer extend furthest in direction to the optic disc border. Future studies may explore whether these spatial relationships between the tissue layers in the parapapillary region can be explained by an axial elongation-related shift of the RPE layer and BM away from the optic disc border in highly myopic eyes or by a loss of parapapillary RPE cells in glaucomatous eyes.
Disclosure: J.B. Jonas, EP 3 271 392 (P), JP 2021–119187 (P), and US 2021 0340237 A1 (P); R.A. Jonas, EP 3 271 392 (P), JP 2021–119187 (P), and US 2021 0340237 A1 (P); S.B. Jonas, EP 3 271 392 (P), JP 2021–119187 (P), and US 2021 0340237 A1 (P); S. Panda-Jonas, EP 3 271 392 (P), JP 2021–119187 (P), and US 2021 0340237 A1 (P)