Anatomic Relationship Among Descemet's Membrane, Trabecular Meshwork, Scleral Spur, and Ciliary Muscle

Abstract

Purpose: The purpose of this study was to examine anatomic relationships between Descemet's membrane (DM) and neighboring tissues.

Methods: Human eyes enucleated due to malignant uveal melanoma were histomorphometrically examined.

Results: In all 50 eyes included in the study (age = 59.2 ± 13.3 years), a cellular tissue (mean thickness at the DM end = 14.2 ± 19.5 µm), originating in the transitional zone, extended into the space between the peripheral DM and corneal stroma for a length of 142 ± 71 µm. This length decreased with the longer part of the DM covered with Hassall-Henle-bodies (beta = –0.30, B = –0.28, 95% confidence interval [CI] = –0.54 to –0.03, P = 0.03) and was not related with age (P = 0.84), axial length (P = 0.94), transitional zone length (P = 0.51), and the DM-scleral spur distance (P = 0.72). The DM thinned toward its end and the DM-like components appeared getting together and merging with transitional zone tissue fibers. The transitional zone length (mean = 267 ± 115 µm) varied from 82 µm to 586 µm, increased with longer DM-scleral spur distance (beta = 0.80, B = 0.67, 95% CI = 0.52 to 0.82, P < 0.001), and was not related with axial length (P = 0.74) or age (P = 0.18). The DM-scleral spur distance (mean = 660 ± 136 µm, range = 302 µm–979 µm) was not related with axial length (P = 0.60) or age (P = 0.06). The scleral spur dimensions (basis = 193 ± 103 µm, height = 151 ± 46 µm, length = 231 ± 75 µm) were not related with the DM-scleral spur distance (all P > 0.35), axial length (all P > 0.07), and age (all P > 0.25). In all eyes, the scleral spur tip was orientated anteriorly, with the longitudinal ciliary muscle inserting at its posterior side.

Conclusions: DM is firmly connected with the transitional zone tissue, and indirectly through the corneoscleral TM, with the scleral spur. With the latter connected through the longitudinal ciliary muscle and Bruch's membrane with the optic disc, the DM is part of an anatomic spherical unit of the globe.

Together with the corneal endothelium, Descemet's membrane (DM) forms the inner most layer of the cornea and is, together with Bowman's layer, the biomechanical most important structure of the cornea.^1–3 DM is produced by the corneal endothelium as its basal membrane and is composed of a posterior non-banded layer and an anterior banded layer.¹ DM has been described to end at Schwalbe's line anterior to the trabecular meshwork (TM), the corneoscleral portion of which continues to the scleral spur posteriorly and the uveal portion of which, bypassing the scleral spur, extends more posteriorly, and merges into the inner portion of the longitudinal ciliary muscle.^4–9 The scleral spur serves as the origin of the longitudinal part of the ciliary muscle, which connects to Bruch's membrane (BM) in the posterior region of the pars plana area.^4–11 In the ocular posterior segment, the peripapillary end of BM is attached to the peripapillary choroidal border tissue which continues into the peripapillary border tissue of the peripapillary scleral flange.^12,13 The collagen fibers of the peripapillary border tissue of the peripapillary scleral flange crisscross with the fibers of the peripapillary scleral flange and lamina cribrosa and firmly connects the posterior part of BM and its attached structures to the sclera.^12,13 The anterior part of BM is, through its adhesion to the posterior end of the longitudinal ciliary muscle, connected to the annular ring of the scleral spur.^10,11 The latter thus serves as an anterior fixation point of the inner eye ball (consisting of the uvea and all structures interior to it) to the sclera. With the biomechanical importance of DM long been known, it has remained obscure whether in addition to the scleral spur, DM may also play a role in the fixation of the inner eye ball to the outer coat.¹⁴ In particular, it has remained partially unclear so far, how DM is connected through the TM to the scleral spur. The region between the end of DM and the scleral spur may be thus of importance for the structural support for the eye. We therefore conducted this study to quantitatively examine details of the DM-TM-scleral spur transition, the morphometric data of which may be helpful for future investigations on the biomechanics of the anterior ocular segment. 

Methods

Enucleated human eyes were examined in the histological study. They had been removed due to malignant uveal melanomas. The Medical Ethics Committee II of the Medical Faculty Mannheim of the Heidelberg University confirmed that the study design was in agreement with the guidelines laid down in the World Medical Association Declaration of Helsinki. The Medical Ethics Committee II of the Medical Faculty Mannheim of the Heidelberg University approved the study and waived the necessity of an informed written consent by the patients, because the globes had been enucleated sometime in the period from 1980 to 1993, more than 30 years before the study was planned. Inclusion criteria were the absence of glaucomatous optic nerve atrophy, the absence of malignant melanoma cells in the iris, anterior chamber angle or anterior parts of the ciliary body, and a macroscopically and microscopically normal appearance of the anterior chamber angle and cornea. 

As also described in greater detail in previous reports, the eyes had been fixed at once after surgical enucleation.^11,12 They had been kept in a fixation solution consisting of 4% formaldehyde and 1% glutaraldehyde for 1 week at room temperature. After that week, the eye diameters were measured, and a central part, measuring 8 mm in thickness and running from the corneal center through the pupil and through the optic nerve head, was removed from of the globes. It was dehydrated in alcohol and imbedded in paraffin. Histological slides approximately 5 to 8 µm thick were stained by hematoxylin and eosin or using the Periodic Acid Schiff (PAS) method. The meridional orientation of the histological sections depended on the location of the tumors, with the sections running through the pupil, the optic nerve head, and through the center of the tumor. 

Using a millimeter scale built into the objective of a light microscope, we histomorphometrically assessed the cornea and structures of the anterior chamber angle (Figs. 1–5). We measured: 

• - Thickness of DM at its end, in a distance of one millimeter centrally to it, in the corneal midperiphery, and in the corneal center (see Fig. 1);
• - Presence of Schwalbe's line defined as localized thickening of the peripheral end of DM (not to be confounded with a Hassall-Henle body [HHB]);
• - Thickness of a collagenous layer just beneath (i.e. anterior to) DM (“DMCL”) close to the end of DM, in the corneal midperiphery, and in the corneal center (see Fig. 1);
• - Number of peripheral PAS-positive nodular thickenings of DM located close to the end of BM (i.e. HHBs; see Fig. 2);
• - Length (and thickness in the region of the end of DM) of a cellular collagenous tissue, extending from the transitional zone centrally beyond the end of DM into the space between the anterior DM surface and the corneal stroma, and connecting to the stromal surface of DM (see Figs. 2–4). We defined the transitional zone as the region extending from the end of DM posteriorly and consisting of a relatively compact, non-spongious or meshwork-like tissue. It ended latest at the level of the anterior end of Schlemm's canal (see Fig. 5a);
• - Distance between DM end and the most anterior point of the scleral spur (see Fig. 5a);
• - Length of the transitional zone (see Fig. 5a);
• - Thickness of the corneoscleral part of the TM (defined as the continuation of the transitional zone tissue and extending to the tip of the scleral spur) at the level of Schlemm's canal (see Fig. 5a);
• - Thickness of the juxtacanalicular part of the TM (defined as the TM tissue located between the corneoscleral TM and Schlemm's canal) at the level of Schlemm's canal (see Fig. 5);
• - Thickness of the uveal TM tissue at the scleral spur tip (see Figs. 5a, 5b);
• - Length of the scleral spur at its base, its height and its length, defined as the distance between the midpoint of its base and the scleral spur tip (see Fig. 5a);
• - Length and width of Schlemm's canal.

Figure 1.

 

Histo-photograph of the cornea showing Descemet's membrane (black arrows), a pre-Descemet-collagenous layer (green arrows) and corneal endothelial cells (blue arrow; case # 13919; sex = male; age = 52 years).

Figure 2.

 

(a) Histo-photograph of the periphery of a cornea, showing Descemet's membrane (DM; black arrows), and a cellular tissue (between red arrows) which extends in a wedge-shaped manner from the periphery (in the image on the left upper side) into the space between the DM and the corneal stroma. This tissue contains fibers (between short green arrows) anchoring to DM and extending posteriorly in direction to the trabecular meshwork. Blue arrow = corneal endothelial cell; perforated red arrow = central end of the sub-DM tissue; perforated black arrow = the end of the DM; yellow arrow = Hassall-Henle body (case # 13912; sex = male; age = 59 years). (b) Histo-photograph of the periphery of a cornea, showing DM (black arrows), and a cellular tissue (between red arrows) which extends in a wedge-shaped manner from the periphery (in the image on the left side) into the space between the DM and the corneal stroma. This tissue contains fibers (between short green arrows) anchoring to DM and extending posteriorly in direction to the trabecular meshwork. Blue arrow = corneal endothelial cell; perforated black arrow = the end of the DM; yellow arrow = Hassall-Henle body (case # 13919; sex = male; age = 52 years). (c) Histo-photograph of the periphery of a cornea, showing DM (black arrows), and a cellular tissue (between red arrows) which extends in a wedge-shaped manner from the periphery (in the image on the upper left side) into the space between DM and the corneal stroma. This tissue contains fibers (between short green arrows) anchoring to DM and extending posteriorly in direction to the trabecular meshwork. Blue arrow = corneal endothelial cell; perforated red arrow = central end of the sub-DM tissue; perforated black arrow = end of DM; yellow arrow = Hassall-Henle body (case # 14002; sex = female; age = 63 years). (d) Histo-photograph of the periphery of the cornea (the same eye as shown in Fig. 2c, here, it was slightly in a more peripheral location), showing DM (black arrows), and a cellular tissue (between red arrows) which extends in a wedge-shaped manner from the periphery (in the image on the upper left side) into the space between DM and the corneal stroma. This tissue contains fibers (between short green arrows) anchoring to DM and extending posteriorly in direction to the trabecular meshwork. Blue arrow = corneal endothelial cell; perforated black arrow = the end of the DM; yellow arrow = Hassall-Henle body (case # 14002; sex = female; age = 63 years). (e) Histo-photograph of the periphery of the cornea, showing DM (black arrows), and a cellular tissue (between red arrows) which extends in a wedge-shaped manner from the periphery (in the image on the upper left side) into the space between the DM and the corneal stroma. This tissue contains fibers (between short green arrows) anchoring to DM and extending posteriorly in direction to the trabecular meshwork. Blue arrow = corneal endothelial cell; perforated red arrow = central end of the sub-DM tissue. Note that the corneal endothelial cells extend into the peripheral DM region, in which on the inner DM side the sub-DM tissue (red arrows) is present. Green perforated arrows = artifact due to debris beneath the cover of the histological slide (case # 13919; sex = male; age = 52 years).

Figure 3.

 

Histo-photograph of the anterior region of the anterior chamber angle of the same eye shown in Figure 2e. DM (black arrows) ends at the perforated black arrow, whereas a cellular tissue (between the red arrows) extends from the periphery (in the image on the upper left side) in a wedge-shaped manner into the space between the DM and the corneal stroma. This tissue contains fibers (between the short green arrows) anchoring to DM and extending posteriorly in the direction to the trabecular meshwork. Blue arrow = corneal endothelial cell; yellow arrow = Hassall-Henle body. Note that the DM gradually reduces in its thickness toward its end (perforated black arrow). Note that the sub-DM tissue consists of a PAS-positive part and a more collagenous-fibrous part (case # 13919; sex = male; age = 52 years).

Figure 4.

 

Histo-photograph of the anterior region of the anterior chamber angle of the same eye shown in Figure 2e. In the transitional zone, the cellular tissue (between the red arrows) continues in the direction to the anterior chamber angle (in the image on the left upper side), where it gets more spongious and trabecular meshwork-like (black arrows) in the region of the trabecular meshwork. Note that the long filament-like structure running through the transitional zone in its entire length (between the short green arrows; case # 13919; sex = male; age = 52 years).

Figure 5.

 

(a) Histo-photograph of the anterior chamber angle, showing the end of Descemet's membrane (red arrow), the peripheral end of the transitional zone (green arrow), and the scleral spur (yellow asterisks). Black bar = thickness of the uveal trabecular meshwork tissue on top of the scleral spur tip; in direct continuation (yellow arrow) from the longitudinal ciliary muscle, the uveal trabecular meshwork extends toward the transitional zone; yellow bars = combined uveal part (inner part) and corneoscleral part (outer part) of the trabecular meshwork, connecting the longitudinal ciliary muscle (through the uveal part) and the scleral spur (through the corneoscleral part) with the transitional zone and Descemet's membrane; between red asterisks and Schlemm's canal (green asterisks) = juxtacanalicular trabecular meshwork; blue line = basis of the scleral spur; red line = length of the scleral spur (case # 14002; sex = female; age = 63 years). (b) Histo-photograph of the anterior chamber angle, showing the scleral spur (demarcated by yellow asterisks), with the scleral spur top covered by a tissue (black arrows) in continuation of the trabecular meshwork (“uveal trabecular meshwork”); red asterisks = separation line between the corneoscleral portion of the trabecular meshwork and the juxtacanalicular portion of the trabecular meshwork (case # 13919; sex = male; age = 52 years).

We carried out the statistical analysis applying a statistical software package (SPSS for Windows, version 27.0; IBM-SPSS, Chicago, IL, USA). The data were presented as mean ± standard deviation. Differences between eyes were assessed using the Mann-Whitney test for unpaired samples. Differences between various parameters, measured in the same eyes, were examined using the Wilcoxon test for paired samples. We calculated the standardized regression coefficient beta, the non-standardized regression coefficient B and its 95% confidence intervals (CIs). All P values were two-sided and considered statistically significant, if the values were less than 0.05. 

Results

The study included 50 eyes of 50 patients (age = 59.2 ± 13.3 years, range = 30–85 years) with a mean axial length of 23.9 ± 1.9 mm (range = 20.0–30.0 mm). The horizontal globe diameter was 23.2 ± 1.0 mm, and the vertical eye diameter measured 23.2 ± 1.2 mm. All eyes had been enucleated due to a malignant uveal melanoma. 

In all eyes, a cellular tissue in continuation of the tissue of the transitional zone extended into the space just anterior to the DM surface in the corneal periphery (“TZ-DM-tissue”; see Figs. 2–4). The length of the TZ-DM-tissue, measured from the end of DM to the end of the tissue in the corneal centripetal direction was 142 ± 71 µm, and its thickness at the end of DM was 14.2 ± 19.5 µm (see the Table). Its length decreased significantly with longer length of DM covered with HHBs (beta = –0.30, B = –0.28, 95% CI = −0.54 to −0.03, P = 0.03; Fig. 6). If the outlier with a value of more than 400 µm for the length of DM with an extended transitional zone tissue was dropped from the analysis, the association between the TZ-DM-tissue length and the length of DM covered with HHBs was marginally significant (beta = –0.28, B = –0.07, 95% CI = –0.14 to 0.000, P = 0.05). The length of the TZ-DM-tissue was not significantly associated with age (P = 0.84), axial length (P = 0.94), length of the transitional zone (P = 0.51), and distance between the DM end and the scleral spur (P = 0.72). The thickness of the TZ-DM-tissue (measured at the DM end) was not significantly related with age (P = 0.69), axial length (P = 0.89), length of the transitional zone (P = 0.55), and distance between the DM end and the scleral spur (P = 0.23). 

Table.

 

Histomorphometric Measurements (Mean ± Standard Deviation) of Histologic Structures, Measured in Enucleated Human Eyes

 

Figure 6.

 

Scattergram showing the association between the length transitional zone tissue extending into the space between Descemet's membrane and the corneal stroma, and the length of Descemet's membrane covered by Hassall-Henle bodies on its inner surface.

None of the eyes showed a localized marked thickening of the end of DM (except for the HHBs), so that a histological correlate of Schwalbe's line, as described gonioscopically, was not detected. Instead, in all eyes, DM-like components related to TM fibers, which stained the same as DM with PAS, looked like getting together and merging with the DM. The filaments of the DM end had tight contact with, or merged with, the tissue of the transitional zone, which showed in some eyes PAS-positive linear structures (see Figs. 2a–e, 3, 4). 

The mean number of HHBs was 8.2 ± 9.2, which were distributed along the posterior surface of DM for a mean length of 269 ± 230 µm. The length of DM covered with HHBs increased significantly with older age (beta = 0.35, B = 6.15, 95% CI = 0.94–11.4, P = 0.02; Fig. 7). The HHB number increased, but not significantly, with age (P = 0.07). The HHB number and the length of DM covered by HHBs were not associated with axial length (P = 0.41 and P = 0.49, respectively). 

Figure 7.

 

Scattergram showing the association between the length of Descemet's membrane covered by Hassall-Henle bodies on its inner surface and age.

The thickness of the DMCL was 3.3 ± 0.8 µm at the end of the region with HHBs, 3.5 ± 0.9 µm in the corneal midperiphery, and 3.7 ± 2.7 µm in the corneal center. All thickness measurements of the DMCL were not related with age or axial length (all P > 0.10). In none of the eyes, was any cell body located between the DM and the DMCL. The DMCL stopped close to the DM end, at the beginning of the TZ-DM-tissue. The DMCL did not continue into the TZ-DM-tissue. In contrast, the TZ-DM-tissue had a wedge-shaped form, with its tip orientated in direction to the center of the DM and its base located at the end of the DM. 

The transitional zone measured in length 267 ± 115 µm (range = 82 µm to 586 µm). The length of the transitional zone significantly increased with longer distance between the DM end and the scleral spur (beta = 0.80, B = 0.67, 95% CI = 0.52–0.82, P < 0.001; Fig. 8). It was not significantly related with axial length (P = 0.74) or age (P = 0.18). 

Figure 8.

 

Scattergram showing the association between the distance between the end of Descemet's membrane and the scleral spur and the length of the transitional zone.

The distance between the end of the DM and the scleral spur tip was 660 ± 136 µm (range = 302 µm to 979 µm). It increased with longer length of the transitional zone (beta = 0.76, B = 0.66, 95% CI = 0.50–0.83, P < 0.001), and was not significantly related with axial length (P = 0.60) and age (P = 0.06). 

The dimensions of the scleral spur were 193 ± 103 µm as the length of the basis, 151 ± 46 µm as the height, and 231 ± 75 µm for the length of the scleral spur, measured from the mid of its basis to the tip. These dimensions were not related with the DM-scleral spur distance (all P > 0.35), axial length (all P > 0.07), and age (all P > 0.25). 

The thickness of the combined uveal and corneoscleral part of the TM at the level of Schlemm's canal was 33 ± 39 µm, and of the juxtacanalicular part (in direction to Schlemm's canal) 54 ± 43 µm. Both parameters were not related with age, axial length, dimensions of the scleral spur, tissue thickness on top of the scleral spur tip, and the DM-scleral spur distance and transitional zone length (all P > 0.25 and P > 0.30, respectively). The thickness of the uveal TM tissue at the tip of the scleral spur was 42 ± 56 µm. It was not significantly related with age, axial length, scleral spur dimensions, DM-scleral spur distance, and length of the transitional zone (all P > 0.05). 

The length and maximal width of Schlemm's canal was 343 ± 97 µm and 61 ± 24 µm, respectively, with a mean cell count on the inner and outer wall of Schlemm's canal of 38.3 ± 25.0 cells and 29.6 ± 27.1 cells, respectively. All Schlemm's canal related parameters were not significantly associated with the axial length or age (all P > 0.10). 

Discussion

This histomorphometric study suggests that DM is connected with the scleral spur through the transitional zone tissue which originates on the outer surface of the peripheral DM and continues into the corneoscleral TM which finally attaches to the scleral spur. In association with the longitudinal ciliary muscle originating at the posterior side of the scleral spur, the finding suggests that DM forms an anatomic unit with the transitional zone tissue, corneoscleral TM, scleral spur, the outer part of the longitudinal ciliary muscle, BM in the pars plana region and in the retinal region, the peripapillary border tissue of the choroid, and finally with the peripapillary border tissue of the peripapillary scleral flange and thus the posterior sclera. It also suggests that the inner eye ball is connected not only through the peripapillary choroidal border tissue and through the scleral spur, but also through the transitional zone tissue with the ocular outer shell, formed by the sclera and DM. DM did not abruptly end in the corneal periphery but thinned toward its end and DM-like components appeared getting together and merging with transitional zone tissue fibers. DM did not show a prominent region at its end (as a potential correlate of Schwalbe's line). The corneoscleral part of the TM originated at the anterior surface and at the tip of the scleral spur. Both, the uveal and the corneoscleral part of the TM continued into the transitional zone tissue and indirectly to the peripheral DM. 

The observations made in our study agree with findings reported in previous investigations. Applying inverted two-photon excited fluorescence microscopy, Marando and colleagues examined the collagen and elastin architecture in the transition zone between the DM periphery and TM in unfixed human corneal buttons.¹⁵ They found that collagen-rich extensions of the ciliary body (“ciliary body tendon”) together with elastin fibers originating from the TM inserted into the cornea between DM and the posterior corneal stroma. The collagen fibers from the ciliary body had direct contact with the anterior surface of DM, with their insertion narrowing in direction to the optical axis. It resulted in a wedge-shaped appearance, as also found in our study (see Figs. 2–4). Marando and associates also described that about 260 µm centrally to the DM end, the ciliary body collagen fibers fanned out and merged with collagen fibers located in the pre-DM region, forming a dense collagenous comb-like structure orthogonal to the edge of DM.¹⁵ The fibers were associated with a delicate elastin network of interwoven fibers originating from the TM. Using inverted two photon excitation microscopic imaging, Park et al. examined the anatomic relationship between the TM and the ciliary muscle tendons.¹⁶ They described that the elastin fibers of ciliary muscle tendons were connected to the elastin network within the TM lamellae, forming intricate networks within the TM. Using electron microscopy, Dua and colleagues reported that collagen beams emerging from the pre-DM layer (also called “Dua's layer”) in the corneal periphery divided, subdivided and merged with, or continued as, TM beams, with both, the pre-DM collagenous layer and the TM, containing long-spacing collagen.¹⁷ In an immunohistochemical study, Kepez and colleagues found collagen types 2, 3, and 4, laminin and myocilin predominantly in the transitional zone between the cornea and the TM, whereas Wnt-1, integrin alpha 3, and tenascin C were highly concentrated in the TM.¹⁸ The corneal stroma predominantly showed collagen type 1. The collagen and elastin fibers of the TM were continuous with the pre-DM layer, with the structures being firmly attached to each other. In contrast to Dua and colleagues who described that the collagen beams emerging from the pre-DM layer in the corneal periphery merged with, or continued as, TM beams, we did not find a major direct continuation of the pre-DM layer fibers into the TZ-DM-tissue. In contrast, the TZ-DM-tissue ended in a wedge-shaped form on the anterior surface of the peripheral DM. It suggests that it is DM, and not the pre-DM layer, which forms a biomechanical unit with the TZ-DM-tissue, and in continuation posterior, to the TM and scleral spur. 

The measurements of the distance between DM end to the tip of scleral spur (660 ± 136 µm) as obtained in our study are similar to those reported by Kasuga and colleagues with a mean length of 694.9 ± 109.0 µm in men and 713.2 ± 106.9 µm in women, without a significant difference between men and women.¹⁹ Kasuga defined TM length as the length between the tip of the scleral spur to the end of Descemet's membrane.¹⁹ 

The findings made in our study add to the existing knowledge about the anterior chamber angle anatomy by providing histomorphometric measurements and by further discussing the potential functional sequels of these anatomic relationships. Previous studies have shown that BM is a biomechanically important structure, with a biomechanical strength, related to the tissue thickness, higher than that of the sclera.¹⁴ BM could withhold an intraocular pressure of up to 100 mg Hg if the sclera has been opened for an area of about 0.5 mm × 1.0 mm.¹⁴ In non-highly myopic eyes, acquired defects in BM are associated with localized staphylomas of the sclera.²⁰ The same holds true for congenital BM defects. As a corollary, scleral staphylomas in highly myopic eyes are often associated with BM defects at the margin of the staphyloma.^21–23 It has also been speculated, that BM may be the structure which primarily makes the optical axis longer in myopic axially elongating eyes.^24,25 These observations led to the model of an inner eye ball which contains the uvea, including choroid, ciliary body and iris, and all tissues interior to it. This inner eye ball is covered by the sclera, which can easily be separated from the inner eye ball by just peeling the sclera off the uvea. The inner eye ball is firmly connected to the sclera only at two positions: posteriorly through the peripapillary border tissue of the choroid, connecting the peripapillary end of BM with the peripapillary scleral flange in its transition into the lamina cribrosa; and anteriorly at the scleral spur. BM in the region of the retina continues into a BM-like structure in the pars plana region of the ciliary body, where, in the posterior region of the pars plana region, it connects to the longitudinal part of the ciliary muscle.^10,11 The latter connects to the scleral spur. The observations presented in the present study extend that anatomic, and potentially functional, model by showing the anatomic connection between the scleral spur and the peripheral DM region through the transitional zone tissue and corneoscleral TM. This arrangement is complemented by the direct connection between the inner part of the longitudinal ciliary muscle with the uveal TM, which extends into the transition zone tissue and finally to the peripheral DM. Allover, it results in a continuous anatomical, and potentially functional, unit extending from DM through the transitional zone tissue and TM, scleral spur (in the case of the uveal TM without direct involvement of the scleral spur), longitudinal ciliary muscle, BM of the pars plana and retina, eventually to the peripapillary border tissues of the choroid and peripapillary scleral flange, and thus the sclera. 

A potential counter-argument against a functional importance of the transitional tissue and its extension to DM may be that the corneoscleral coat has a thickness ranging from 0.5 mm to 1.0 mm, whereas the thickness of the wedge-shaped tissue embedded under the peripheral DM was 14 µm or just 2 to 3% of the thickness of the corneoscleral coat.²⁶ An analogy, however, may be the ratio of the thickness of the posterior sclera (ranging between 0.4 mm and 0.9 mm in dependence of the ocular location) and the thickness of BM with about 5 µm or about 1% of the scleral thickness.^26,27 BM as part of the inner eye ball has been discussed to be the major structure biomechanically supporting the inner eye ball, whereas the sclera is just the outer coat.²⁵ As such, the sclera has a firm contact with the inner eye ball only at the scleral spur anteriorly and through the peripapillary choroidal border tissue posteriorly. According to the observations described in the present study, an additional firm contact between the inner eye ball and its outer coat is achieved by the connection between the scleral spur and the peripheral region of DM through the TM and the transitional tissue. Another argument in favor of a potential biomechanical role of the transitional zone tissue is that DM is the biomechanically most important structure of the cornea, and that it is the transitional zone tissue together with its extension on DM surface, which directly connects DM to the scleral spur. 

Anatomically, the scleral spur is orientated anteriorly. At a first glance, this would appear to be contradictory to the scleral spur insertion of the posteriorly directed longitudinal ciliary muscle. If the ciliary muscle contracts, the scleral spur would bend posteriorly, unless the scleral spur is fixed anteriorly. A mobile and flexible insertion of the muscle at the scleral spur would markedly decrease the muscle's contractile strength during accommodation. The findings of the present study, in agreement with the observations made in the previous investigations, suggests that a biomechanically firm connection between the scleral spur and the end of the DM exists.^3,6,7 In the peripheral region of DM, the tissue of the transition zone appeared to be firmly attached to the anterior surface of DM. The tissue of the transition zone continued posteriorly into the corneoscleral part of the TM which connected to the tip of the scleral spur. One may infer that the traction of the transition zone tissue and of the corneoscleral TM fixed the tip of the scleral spur in an anterior direction, and simultaneously provided a stable and firm basis for the insertion of the longitudinal ciliary muscle posteriorly. It completes the biomechanical sphere consisting of DM anteriorly, continuing into the tissue of the transitional zone and the corneoscleral TM and inserting at the scleral spur, from where the longitudinal ciliary muscle through its adhesion to BM in the pars plana region and through BM of the retinal pigment epithelium eventually inserted at the optic disc border through the peripapillary choroidal border tissue in the zone between the peripapillary scleral flange and the lamina cribrosa.²⁸ 

Future studies may examine and discuss the biomechanical sequels of these anatomic relationships. It may include the observations, that a contraction of the longitudinal ciliary muscle moves the ora serrata and the lens zonula fibers anteriorly as a major part of the process of accommodation.²⁹ Such an action would be possible only in the case of a firm and stable fixation of the scleral spur anteriorly. It may also include an indirect effect of a longitudinal ciliary muscle contraction on the BM at the posterior pole, perhaps leading to a widening of the subfoveal choroidal space and a change in the morphology of the optic nerve head. Simultaneously, a contraction of the longitudinal ciliary muscle will have an effect anteriorly by a posteriorly directed traction on the scleral spur. The latter may not be able to markedly move backward due to its firm connection through the corneoscleral TM and through the transition zone tissue to the DM end. Previous studies have shown that the action of the ciliary muscle has a marked influence on the function of the TM with respect of a regulation of the aqueous humor outflow resistance.^4,5,7,30 In that context, it may also be of interest, whether the biomechanics and the biodynamics of the inter-action between the longitudinal ciliary muscle and the TM is influenced by a central defect in DM, as produced by DM endothelial keratoplasty (DMEK).³¹ The biomechanical role of the scleral spur in its anatomic and physiological association with the TM on one side and the ciliary muscle on the other side has also previously been assessed and discussed, for example, by Moses and Grodzki and by Moses and Arnzen.^32,33 

Another, potential novel finding in our study was the presence of PAS-positive fibers different from DM and anchoring to DM (see Figs. 2a–e, 3, 4). These variations in anchoring fibers might reflect differences in the proportion of corneoscleral and uveal meshwork components attaching to the DM, and may be addressed in future investigations. In particular, their potential importance for the biomechanics of the DM-transitional zone-TM-scleral spur complex may be examined in greater detail. 

Another finding of our study may be of clinical interest. The length of the transition zone showed a marked interindividual variability, ranging between 82 µm and 586 µm with a mean of 267 ± 115 µm. It agrees with results of previous studies.^34,35 Wahlig and colleagues used optical coherence tomography (OCT) and measured the dimensions of the anterior chamber angle in healthy Chinese.³⁴ Based on transmission electron images, they assumed that the transitional zone made out 20% of the DM-scleral spur distance. The authors estimated the length of the transitional zone to be 156 ± 20 µm. The clinical importance of the interindividual variability in the length of the transitional zone may be the application of argon laser trabeculoplasty and selective laser trabeculoplasty. These are anti-glaucomatous interventions during which a laser beam is directed into the anterior third of the TM, as assessed gonioscopically, or directed at the anterior border of a more or less pigmented line of the TM.³⁶ Histologically, this zone predominantly represents the transition zone between DM and the TM. The interindividual variability in the length of the transition zone may perhaps be one of the reasons for the variation in the therapeutic effect of the laser interventions to reduce the intraocular pressure.³⁶ The interindividual variability in the length of the transitional zone may also be of interest for the discussion of the presence of adult stem cells in that region.^37–40 

The dimensions of the scleral spur have also been measured previously by Nesterov et al. and by Swain and colleagues.^41,42 Swain and colleagues reported that the mean length of the scleral spur was significantly shorter in eyes with primary open-angle glaucoma than in normal globes (0.15 ± 0.01 mm versus 0.20 ± 0.01 mm; P < 0.0001).⁴² These values are close to measurements obtained in our study with a mean scleral spur length at is basis of 193 ± 103 µm (see the Table). 

In all eyes examined, we did not detect a PAS-positive prominence of the end of the DM which might have been the equivalent of Schwalbe's line as described gonioscopically. Interestingly, Alwadani and colleagues described that a posterior embryotoxon was not an anterior displaced Schwalbe’s line but a peripheral corneal stromal nub variable in location with abnormal extracellular matrix.⁴³ Fitting with the notion of an absence of a prominent end of DM (except for HHBs), the DM thinned at its end, in direct adhesion to the anterior portion of the transition zone tissue (see Figs. 2, 3). 

The DMCL could be detected in all eyes examined. It appeared not to have a direct firm connection with the anterior end of the transitional zone tissue, which was mainly adherent to the anterior surface of the peripheral DM. Due to the design of our study as a retrospective investigation, we could not definitely confirm that the DMCL was an anatomic entity of its own and to be differentiated from the remaining corneal stroma (as described by Dua as “Dua's layer”), or whether it is the first corneal stromal lamella in contact with the DM.^17,18,44 

The length of the transitional zone extension into the space anterior to DM decreased with longer length of DM covered with HHBs (see Fig. 6). While the reasons for such a relationship have remained unclear, it may be discussed whether a potential biomechanical force on the anterior surface of the peripheral DM may be associated with the formation of HHBs on the posterior DM surface. 

When the findings obtained in our study are discussed, its limitations should be considered. First, we did not perform immunohistochemical examinations to classify the various cells in the transitional zone and in the TM. Second, serial sections were not available. Third, all eyes had been enucleated due to a malignant choroidal melanoma which might have influenced the appearance of the anterior chamber angle, even if the anterior chamber had to be free of tumor cells and eyes with glaucomatous optic nerve damage had been excluded. Fourth, the tissue dimensions had been changed due to post mortem tissue swelling and fixation-related tissue shrinkage. The measurements can therefore not directly be transferred into clinical conditions. One may, however, take into account that collagenous structures such as DM may not undergo marked changes in their length due to short-term postmortem swelling or due to the fixation process. Fifth, the PAS staining method was used following aldehyde fixation of the enucleated eyes. PAS is a staining method detecting aldehydes, which are induced by the periodic acid acting on mucopolysaccharides. Because glutaraldehyde as a fixative agent may also introduce aldehydes, the question may arise whether the PAS-positive label was specific for mucopolysaccharides or whether it was an artifact of fixation. Because the globes stayed in the fixative agent for 1 week only after their enucleation, and because the PAS method has been a standard procedure in ophthalmic pathology, it may be unlikely that a major artifact was caused by applying the PAS staining method. Sixth, the assumption of the wedge-shaped tissue underlying the peripheral DM to be the anterior anchor for the longitudinal ciliary muscle, although the scleral spur and the corneoscleral TM could not definitely be concluded because the connection between the wedge-shaped tissue and the DM was not explicitly shown in the study. Future investigations may explore such a connection by electron microscopy, by examining flat mounts, or other techniques. 

In conclusion, DM appears to be connected with the scleral spur through the transitional zone tissue which originates in the space between the anterior DM surface and the corneal stroma in the corneal periphery and which continues through the corneoscleral portion of the TM to the scleral spur. In addition, the DM is connected through the transitional zone tissue and the uveal portion of the TM, bypassing the scleral spur, with the anterior end of the longitudinal ciliary muscle. The DM potentially forms a biomechanical and biodynamical unit with these structures, in association with the longitudinal ciliary muscle originating at the posterior side of the scleral spur and indirectly extending, through the BM of the pars plana region and of the retina, to the optic disc. 

Acknowledgments

Data Availability Statements: All data are available upon reasonable request from the corresponding author. 

Disclosure: J.B. Jonas, None; S. Panda-Jonas, None; J.S. Mehta, None; R.A. Jonas, None