Aqueous humor outflow system overview

Aqueous humor circulation through the anterior segment of the eye represents one of the many cardiac circulatory loops that also include the various arteriovenous, lymphatic, and cerebrospinal fluid circulations. Each of these circulatory loops is driven down a continuous pressure gradient initially set up by the heart. Aqueous humor is formed by the ciliary processes, passes from the posterior chamber to the anterior chamber through the pupil, and exits the eye at the anterior chamber angle. Aqueous returns to the venous system primarily by means of the conventional or canalicular pathway (83–96% of flow). The pathway is through the trabecular meshwork into Schlemm’s canal (hence the canalicular pathway). The Schlemm’s canal lumen communicates directly with the episcleral veins, completing the circulatory pathway for aqueous return to the heart.

Aqueous humor also returns to the heart by a secondary pathway known as the uveoscleral or unconventional route. The uveoscleral route accounts for from 5 to 15% of flow, an amount that decreases with age. Extracanalicular aqueous flow is through the anterior ciliary muscle and iris stroma to reach the supraciliary and suprachoroidal spaces. From these spaces the fluid passes through the sclera and the loose connective tissue around the penetrating nerves and vessels.

Considerable controversy at one time revolved around the issue of aqueous humor as a circulating or a stagnant fluid. However, Ascher, Goldmann and others document clearly through in-vivo observations that aqueous humor circulates and returns directly to the venous system. Abnormalities of this flow through the aqueous circulatory system provide the basis for our present concepts of both open- and closed-angle glaucoma.

Another prominent finding of these same investigators is pulsatile aqueous flow from the episcleral to aqueous veins; pulsatile flow that originates in Schlemm’s canal. Pulsatile flow is in synchrony with small pressure transients such as those induced by the ocular pulse, blinking, and eye movement. Findings of pulsatile aqueous flow are not integrated into the traditional framework of passive outflow across an unyielding syncytium in the juxtacanalicular space. However, a new conceptual model of aqueous outflow mechanisms integrates the pulsatile flow findings into the model.

It is generally accepted that the major portion of the normal resistance to conventional outflow resides in the region between the anterior chamber and external wall of Schlemm’s canal. Furthermore, it is thought that this region is the site of the abnormal resistance to outflow found in most cases of open-angle glaucoma. The nature of the resistance in this region in both the normal and glaucomatous eye is the subject of a continuing controversy.


Schlemm’s canal is a modified wall of a vessel. In other vessels, pressure gradients are higher in the vessel lumen. In contrast, pressures are higher external to the lumen of Schlemm’s canal. Fluid moves from the higher pressure in the lumen of vessels across vessel walls to the lower pressure in adjacent tissues as a response to the hydrostatic pressure gradient. Again, in contrast, aqueous flows from the anterior chamber, across the modified vascular wall represented by the trabecular meshwork, into the lower pressure vascular lumen of Schlemm’s canal. A series of adaptations is required as a result of the pressure gradient and fluid flow reversals. These adaptations are reflected in the unique tissue anatomy, geometry and responses to pressure in the wall of Schlemm’s canal that differ from those of the walls of other vessels.


The most obvious function of the outflow pathway is a circulatory path for aqueous humor return to the vascular system. A second function permits bulk aqueous flow of aqueous out of the anterior chamber but prevents blood reflux into the anterior chamber. The trabecular meshwork is thus a crucial part of the normal blood–aqueous barrier. The barrier is important for the optical properties of the eye and limits the entrance of potentially noxious substances.

A third important function is maintenance of a relatively stable intraocular pressure (IOP). A stable IOP range is crucial for the structural integrity and optical functioning of the eye. Stable IOP must be maintained despite different rates of aqueous humor formation, different levels of IOP, and different amounts of ciliary muscle tone.

A fourth function is filtration of foreign material and debris. Trabecular endothelial cells actively phagocytize foreign material and debris. Finally, as an adaptation to pressure gradient reversal, the meshwork functions as a suspensory system for the inner wall endothelium of Schlemm’s canal. The suspensory system consists of the ciliary muscle, scleral spur, trabecular lamellae, and juxtacanalicular cells all linked to Schlemm’s canal endothelium through a complex network of cytoplasmic processes.


The limbus is the transitional region between the cornea and sclera. At the inner surface, the limbus contains a scleral indentation called the scleral sulcus. At the anterior margin of the scleral sulcus is Schwalbe’s line. The scleral sulcus is defined posteriorly at its internal margin by the scleral spur, an anterior extension of sclera that partially encloses the posterior portion of the sulcus. The trabecular meshwork is interposed between the anterior chamber and Schlemm’s canal by means of its attachment anteriorly to Schwalbe’s line and posteriorly to the scleral spur and ciliary muscle. The trabecular meshwork thus composes the inner wall of Schlemm’s canal.


Schwalbe’s line (composed of collagen and elastic tissue) is an irregular elevation 50–150 μm wide that runs circumferentially around the globe ( Fig. 3-1 ). This line or zone marks the transition from trabecular to corneal endothelium, the termination of Descemet’s membrane, and the insertion of the trabecular meshwork into the corneal stroma. Secretory cells, called Schwalbe’s line cells, are present in this area that produce a phospholipid material thought to facilitate aqueous flow.

Fig. 3-1

Semi-diagrammatic representation of structures of the angle of the anterior chamber. Note the superimposed trabecular sheets with intratrabecular and intertrabecular spaces, through which aqueous humor percolates to reach Schlemm’s canal. SL, Schwalbe’s line; SS, scleral spur; IP, iris process; TM, trabecular meshwork; C, cornea; I, iris; SC, Schlemm’s canal; CB, ciliary body.

From Tripathi RC, Tripathi BJ: Functional anatomy of the anterior chamber angle. In: Duane TD, Jaeger EA, editors: Biomedical foundations of ophthalmology, vol 1, New York, Harper & Row, 1982.


The scleral spur is a fibrous ring that, on meridional section, appears as a wedge projecting from the inner aspect of the anterior sclera ( Figs 3-1 and 3-2 ). The spur is attached anteriorly to the trabecular meshwork and posteriorly to the sclera and the longitudinal portion of the ciliary muscle. The spur consists of collagen types I and III and about 5% elastic tissue oriented in a circumferential arrangement. When the ciliary muscle contracts, it pulls the scleral spur posteriorly ( Fig. 3-3 ). The largest trabecular lamellae near the anterior chamber are attached to the scleral spur and accordingly are rotated inward and posteriorly by ciliary muscle contraction. Rotation alters the position not only of the large lamellae, but also moves the entire attached meshwork further inward and posteriorly. Inward movement of the trabecular meshwork results in an enlargement of intertrabecular spaces and an increase in the size of Schlemm’s canal, reducing the tendency of the canal lumen to narrow or collapse. Varicose axons characteristic of mechanoreceptor nerve endings are present in the spur region and may measure stresses at the scleral spur induced by IOP changes or ciliary muscle contraction.

Fig. 3-2

Schematic view of different layers of the outflow system.

Modified from Shields MB: Textbook of glaucoma, Baltimore, Williams & Wilkins, 1987.

Fig. 3-3

(A) Schematic view of the system before pilocarpine treatment. (B) Administration of pilocarpine contracts the ciliary muscle, which pulls the scleral spur posteriorly and internally, opening the intertrabecular spaces and Schlemm’s canal.


In meridional section, the trabecular meshwork has a triangular shape, with its apex at Schwalbe’s line and its base at the scleral spur (see Fig. 3-1 ). The inner layers of the trabecular meshwork border the anterior chamber and are referred to as the uveal meshwork. The next more superficial layer is the corneoscleral meshwork. The juxtacanalicular space is the next layer, which is between the corneoscleral meshwork and Schlemm’s canal inner wall endothelium (see Fig. 3-2 ).

Uveal meshwork

The uveal meshwork is adjacent to the anterior chamber. Iris processes are fine strands of the innermost layer of the uveal meshwork present in many eyes. The processes arise from the anterior surface of the iris, bridge the angle recess, and insert into the deeper uveal trabeculae or Schwalbe’s line (see Fig. 3-1 ). The rest of the uveal meshwork has a rope- or cord-like character, with randomly oriented interconnecting bands, and is only a few layers thick. The uveal meshwork inserts anteriorly into the region of Schwalbe’s line and posteriorly into the ciliary body and iris root. The inner layers are generally oriented radially although they branch and interconnect in multiple planes.

Corneoscleral meshwork

The corneoscleral meshwork consists of a series of 8–14 flattened, perforated parallel sheets or lamellae, each 5–12 microns thick. The sheets closer to the anterior chamber are anchored anteriorly to Schwalbe’s line. The sheets pass in a meridional fashion posteriorly to attach to the scleral spur. The anterior tendons of the longitudinal ciliary muscle fibers insert on the posterior portion of the corneoscleral meshwork as well as on the scleral spur. The inner trabecular lamellae closer to the anterior chamber are considerably thicker than the outer ones closest to Schlemm’s canal ( Fig. 3-4 ). Trabecular lamellae are attached to one another via cytoplasmic processes ( Fig. 3-4 ). The cytoplasmic processes originate from the surface of the endothelial cells covering the lamellae and meet in the intertrabecular space with a complex zone of apposition involving desmosomes and gap junctions. Intertrabecular collagen beams are difficult to find.

Fig. 3-4

Cell processes project from Schlemm’s canal endothelial cells (SCP) and attach to juxtacanalicular cell processes (JCP). Juxtacanalicular cell processes also attach to trabecular lamellae endothelial cell processes (TLP). Trabecular lamellae endothelial cells in addition have processes that also project to adjacent trabecular lamellae cell processes. Schlemm’s canal endothelial lining thus benefits from a distribution of attachments that extend to the entire system of trabecular lamellae. Tissue loading forces induced by IOP provide a means of determining resistance characteristics because outflow structures responsible for the resistance change shape. Schlemm’s canal endothelium responds to IOP-induced distending forces. Cell bodies, nuclei, and cytoplasmic processes of both Schlemm’s canal endothelium and juxtacanalicular cells undergo progressive deformation as a result of their role in maintaining resistance to progressive IOP-induced distention of Schlemm’s canal endothelium. The system of cell processes enables the trabecular lamellae to limit distention, thus countering IOP-induced forces acting on Schlemm’s endothelium. As a result of the countering tension, spaces between the resisting trabecular tissues progressively increase as IOP increases. At physiologic pressures (basal IOP), tensional integration is present because resistance forces are distributed throughout the trabecular tissues. Tensional integration provides an information processing network allowing finely graded responses to transient increases in IOP as well as longer term homeostasis through force-dependent mechanotransduction mechanisms.

The trabecular sheets have a generally circumferential orientation parallel to the limbal circumference. The sheets are fused in such a manner that only two or three layers are seen anteriorly. The sheets separate in an anterior–posterior plane so that 12–20 layers are detectable posteriorly.

Sheets of the trabeculae are perforated by elliptical (transtrabecular) openings with an equatorial orientation, with an average dimension of 12–30 microns. Perforations become progressively smaller from the superficial layers of the uveal meshwork to the deep layers of the corneoscleral meshwork (see Fig. 3-2 ). The perforations are not aligned, so aqueous humor must follow a circuitous route to reach Schlemm’s canal ( Fig. 3-5 ).

Fig. 3-5

Light micrograph of trabecular meshwork in 70-year-old human eye. Meridional section shows morphology of uveal trabeculae (UT) and corneoscleral trabeculae (CT). Note the rounded profile of inner uveal trabecular sheets (asterisks) compared with flattened profile of outer uveal and corneoscleral sheets, and a progressive narrowing of intertrabecular spaces (IT) in the latter region. The arrow denotes branching of a trabecular sheet. (Original magnification ×760.)

From Tripathi RC, Tripathi BJ: Functional anatomy of the anterior chamber angle. In: Duane TD, Jaeger EA, editors: Biomedical foundations of ophthalmology, vol 1, New York, Harper & Row, 1982.

Uveal and corneoscleral meshwork ultrastructure

The composition of the trabecular meshwork tissues has been compared to that of other highly compliant and resilient tissues, such as lung and other blood vessel walls. Ultrastructurally, the uveal and corneoscleral meshworks are the same, being composed of four concentric layers. A description of these layers follows.

First, trabecular sheets or lamellae have a central core of types I and III collagen and elastin with a typical 64 nm periodicity ( Fig. 3-6 ). Second, elastic fibers surround the core region with a spiraling pattern and a 100 nm periodicity. The fibers may be wound loosely or tightly thus conferring elastic properties to the meshwork. Third, the cortical zone (also referred to as the glassy membrane ) is a broad zone that contains collagen types III, IV, and V; laminin; fibronectin; and heparin sulfate proteoglycan. Types VI and VIII collagen are also present. Fourth is a continuous layer of endothelial cells that covers the trabecular lamellae; cells are joined by desmosomes as well as gap junctions. Intercellular clefts allow aqueous to pass freely.

Fig. 3-6

Electron micrograph of a meridional section of human corneoscleral trabecular meshwork (×11 000). TS, trabecular space; EN, endothelial cell; N, nucleus of endothelial cell; BM, basement membrane; LS, long-spacing collagen; C, collagen.

Courtesy of L Feeney, San Francisco.

Numerous cytoplasmic processes arise from the trabecular lamellae endothelial cells. These cytoplasmic processes are attached to cytoplasmic processes of adjacent lamellae and to juxtacanalicular cell cytoplasmic processes (see Fig. 3-4 ) by robust desmosomes. Endothelial cells lining the trabecular lamellae are anchored to a well-defined basement membrane by means of integrin attachments; this is in contrast to Schlemm’s canal endothelium, where a basement membrane is sparse or absent. Cytoskeletal elements include microfilaments (F-actin), intermediate filaments (vimentin) and microtubules (alpha-tubulin). Endothelial cells lining the trabecular lamellae are responsible for maintaining the structural topography and extracellular matrix composition of the lamellae in the face of constant oscillatory stresses. Tissue composition predicts anticipated tissue responses. Type I collagen provides tensile strength and type III collagen imparts resilience. Together these collagenous elements provide structural support in tension while elastin provides a recoverable response over wide excursions. The organization and distribution of collagen and elastin in the trabecular lamellae is like that of tendon, which provides a mechanism for reversible deformation in response to hydrodynamic tissue loading.

Juxtacanalicular space and cells

This space plays an important role in the debate about resistance sites. The juxtacanalicular space is 2–20 μm thick in non-pressurized eyes. The space separates the outer layers of the corneoscleral meshwork from the inner wall of Schlemm’s canal and has been called by a variety of names, including cribriform space, pericanalicular space, and endothelial meshwork. Star-shaped cells in the space are referred to as juxtacanalicular, subendothelial, and cribriform cells.

Extracellular matrix material, juxtacanalicular cells and elastic-like fibers are the prominent features of the juxtacanalicular space when viewing two-dimensional histologic sections in non-pressurized eyes. The juxtacanalicular space contains a ground substance of glycosaminoglycans, which include hyaluronic acid, chondroitin sulfate, dermatan sulfate, keratan sulfate and heparin sulfate as well as complex glycoproteins, types III, IV, and V collagen; curly collagen and fibronectin. Fibronectin content is increased in elderly patients and in patients with glaucoma. A network of elastic-like fibers in the juxtacanalicular space attaches both to the inner wall of Schlemm’s canal and to some of the tendons of the ciliary muscle.

Juxtacanalicular cells and their cytoplasmic processes are the principal feature of the space when a three-dimensional view is achieved with scanning electron microscopy in pressurized eyes. Juxtacanalicular cell cytoplasmic processes attach to processes arising from Schlemm’s canal inner wall endothelium (see Fig. 3-4 ). Other juxtacanalicular cell processes attach to cytoplasmic processes arising from the endothelium of the trabecular lamellae. Well-characterized robust desmosomes, capable of sustaining cellular stresses, are the mechanism of cell process attachment. Desmosomes attach to intracellular intermediate and actin filament support systems that enable the filaments to distribute stresses throughout the cytoplasm of involved cells to form integrated tissue relationships.

The juxtacanalicular cells thus serve the function of anchoring Schlemm’s canal inner wall endothelium to the trabecular lamellae. By distributing IOP-induced stresses across the entire trabecular lamellae system that supports the inner wall endothelium of Schlemm’s canal, the stresses are tensionally integrated, which is essential in cellular response mechanisms (see Fig. 3-4 ). Tensional integration at physiologic pressures provides constant cellular pre-stress. Pre-stress allows immediate graded responses to oscillatory pressures caused by continuously varying IOP transients that force Schlemm’s canal endothelium toward the canal lumen.



Schlemm’s canal is a vascular sinus with a lumen that communicates around the entire globe. The lumen has a flattened elliptical cross-section with a total circumference of ≈36 mm. As the lining wall of a vessel, Schlemm’s canal endothelium has properties of a vascular endothelium. The lumen structure has also been likened to that of a lymphatic channel. The canal is surrounded by sclera, trabecular meshwork, and the scleral spur. Generally Schlemm’s canal has a lumen that is 190–370 μm in length in the radial plane. In hypotony, the shape varies, but when Schlemm’s canal shape is triangular, the lumen typically measures ≈50 microns at its posterior base and narrows to about 5–10 microns at its apex. However, the diameter of the canal lumen is IOP dependent and the space can be absent at high pressures or very large at low pressures.

Schlemm’s canal inner wall endothelium

The canal inner wall endothelium is of special significance because the wall represents the barrier aqueous must cross to get from the juxtacanalicular space to Schlemm’s canal. This inner wall endothelium of Schlemm’s canal also plays a very significant role in debate about resistance sites. The inner wall endothelium forms a continuous monolayer of long, slender endothelial cells with their long axes parallel to the canal lumen. The cells have an average diameter of 20–50 microns and a thickness of 0.2 microns. Tight junctions (zonule occludentes) and desmosomes join the cells to one another and form a continuous belt-like region of contact encircling the apex of the cells. Such contacts normally represent physiologic barriers to perfusion of fluid and particles.

The tight junctions are traversed by slit pores, which are meandering channels in the cell junctions. The frequency and diameter of slit pores in the tight junctions indicate that they can account for only a small fraction of aqueous humor flow. Gap junctions provide communication pathways. As in the cells lining trabecular lamellae, cytoskeletal elements include microfilaments (F-actin), intermediate filaments (vimentin) and microtubules (alpha-tubulin). A vascular endothelial origin of these cells is emphasized by the presence of von Willebrand factor (factor VIII-related antigen), and specialized cellular inclusions including Weibel-Palade bodies. The above studies, as well as others, indicate that Schlemm’s canal endothelial cells have a different origin than juxtacanalicular cells or endothelial cells lining the trabecular lamellae.

The basement membrane beneath Schlemm’s canal external or corneoscleral wall is continuous. By contrast, Schlemm’s canal inner wall endothelium is not continuous, a feature it shares with the lymphatic vessels. The basement membrane of Schlemm’s canal inner wall is poorly defined, inconstant, frequently interrupted and of variable thickness. The rudimentary basement membrane may be explained by the fact that pressure gradients force Schlemm’s canal endothelium toward the vascular lumen of Schlemm’s canal. By contrast, the higher luminal pressure gradients of other vasculature force the endothelium against the basement membrane. Schlemm’s canal inner wall endothelium has a special adaptation for this gradient reversal. Instead of a basement membrane, numerous cytoplasmic processes arise from its adluminal surface and ultimately connect with processes attached to cytoplasmic processes of endothelial cells lining the trabecular lamellae (see Fig. 3-4 ). Tension exerted on Schlemm’s canal endothelium by pressure is transmitted through the cell processes to endothelial cells lining the trabecular lamellae. The tension is further transferred to the integrin attachments of the meshwork that anchor the endothelial cells to the basement membrane of the lamellae (see Fig. 3-4 ). The endothelium–integrin–basement membrane complex of the cells lining the trabecular lamellae serves as a surrogate for the absent basement membrane at Schlemm’s canal inner wall endothelium. The arrangement provides an integrin-dependent, force-sensing architecture like that in the vasculature.

Schlemm’s canal inner wall endothelial cells undergo pressure-dependent configuration changes as they progressively separate from the underlying juxtacanalicular space in response to IOP increases (see Fig. 3-4 ). The outer wall of Schlemm’s canal is apposed to the scleral wall and is organized as a single layer of endothelium continuous with the endothelium of the inner wall. The outer wall does not generally undergo pressure-dependent configuration changes.


The glycocalyx is a thin layer of negatively charged (anionic) material 60–90 nm thick that coats the luminal surface of vascular endothelia and the entrance to intracellular clefts. The glycocalyx modulates adhesion between cells and also surface characteristics that determine adhesion and flow properties. The layer consists of a network of fibrous proteins with sugar-based side chains. The layer acts as a size-selective molecular sieve that impedes the passage of plasma proteins. Vascular endothelia have an experimentally determined effective pore size of 4–5 nm.

A small pore system is present at intracellular clefts, but the openings in the clefts alone are too large (20 nm) in the open pathway around junctional strands to account for the effective pore size. Instead, the matrix of fibrous protein molecules in the clefts is organized so that spaces between these molecular chains have an effective size of 4–5 nm, small enough to restrict diffusion, create hydraulic resistance and reflect macromolecules such as albumin. Partial digestion of the glycocalyx by enzyme perfusion raises the hydraulic permeability of the wall.

In vascular endothelia, cationic ferritin binds to and reveals the presence of the glycocalyx, while neutral or anionic ferritin is not bound. Consistent with the presence of a glycocalyx found in other vascular endothelia, cationized ferritin, but not anionic ferritin, exhibits a striking binding to the luminal surface of Schlemm’s canal endothelium and the associated intercellular clefts. As in vascular endothelia elsewhere, the glycocalyx may play an important role in maintaining hydraulic resistance characteristics as well as maintaining a barrier to passage of proteins such as albumin.

Distending cells that form invaginations or pseudovacuoles, ‘giant vacuoles’

Schlemm’s canal inner wall endothelium stretches to form progressively larger hemispherical outpouchings into the canal lumen as IOP increases ( Figs 3-7 and 3-8 ). A frontal plane through such a hollow hemisphere results in a central ‘vacuolated’ area surrounded by the cell body, including the nucleus. In the enucleated eye, the number and size of the pseudovacuoles increases with progressive increases in IOP but the effect is reversible when IOP is lowered. The structures regularly maintain a communication with the juxtacanalicular space and are generally described as balloon-shaped invaginations from the basal aspect of the cell surface. The formation of the pseudovacuoles continues in enucleated eyes and is not inhibited by hypothermia, consistent with a mechanical rather than an active metabolic rearrangement of the cell.

Fig. 3-7

The walls of Schlemm’s canal (SC) and adjacent trabecular meshwork in a composite sectional and three-dimensional view. The endothelial lining of the trabecular wall of Schlemm’s canal is very irregular, and the cells show luminal bulges corresponding to cell nuclei (N) and macrovacuolar configurations (v). The latter may represent cellular invaginations occurring from the basal aspect and eventually opening on the apical aspect of the cell to form transcellular channels (arrows), through which aqueous humor flows down a pressure gradient. The endothelial lining of the trabecular wall is supported by an interrupted, irregular basement membrane and a zone of pericanalicular connective tissue (PT) of variable thickness. The cellular element predominates in this zone, and the fibrous elements, especially elastic fibers, are irregularly arranged in a net-like fashion. Here the open spaces are narrower than those of the trabecular meshwork (TM). The corneoscleral trabecular sheets show frequent branching, and the endothelial covering may be shared between adjacent sheets. The corneoscleral wall (CW) of Schlemm’s canal is more compact than the trabecular wall, with a predominance of lamellar arrangement of collagen and elastic tissue.

From Tripathi RC, Tripathi BJ: Functional anatomy of the anterior chamber angle. In: Duane TD, Jaeger EA, editors: Biomedical foundations of ophthalmology, vol 1, New York, Harper & Row, 1982.

Fig. 3-8

(A) Electron micrograph of the endothelial lining of Schlemm’s canal (SC), showing the majority of the vacuolar configurations (V) at this level of section having direct communication (arrows) with the subendothelial extracellular spaces, which contain aqueous humor in life. (Original magnification ×23 970.) (B) Scanning electron micrographs of Schlemm’s canal endothelium fixed while at 22 mmHg IOP. Micrographs illustrate ‘serious sources of error’ in quantitation of pore frequency and size. The upper micrographs provide detail of two bulges in which small ‘natural’ openings were identified. The lower micrographs demonstrate artifactual tears in the surface of the thin-walled bulges, and in the lower right micrograph, an opening in the base of the cavity can be seen. (Upper left, ×2500; upper right, ×4500; lower left, ×2500; lower right, ×2000.)

(A) (From Tripathi RC, Tripathi BJ: Functional anatomy of the anterior chamber angle. In: Duane TD, Jaeger EA, editors: Biomedical foundations of ophthalmology, vol 1, New York, Harper & Row, 1982.) (B) (From Grierson I, Lee WR: Pressure effects on the endothelium of the trabecular meshwork and Schlemm’s canal: a study by scanning electron microscopy, Albrecht Von Graefes Arch Klin Exp Ophthalmol 196 : 255–265, 1975. Published with permission from Albrecht Von Graefes Arch Klin Exp Ophthalmol. Copyright by Springer-Verlag.)

The term giant vacuole was used in the 1950s and 1960s to describe these distending cells. Authorities for many years have recognized that ‘giant vacuoles’ represent a passive cellular response to pressure gradient changes. Cellular distention or invagination more accurately describes the structures. The term ‘giant vacuole’ captures the imagination and is still used at times. The term is somewhat unfortunate because it does not reflect the anatomic or physiologic characteristics of the structures. It suggests a metabolically driven mechanism of intracellular fluid transport to Schlemm’s canal rather than focusing attention on the actual physiologic behavior that involves remarkable pressure-induced cellular distention and recoil.

Some authorities suggest that the pressure-induced cellular distention causes these invaginations or pseudovacuoles to form and recede in a cyclic fashion causing transient transcellular pores to form in the distending wall (see Fig. 3-7 ). These authorities propose the pores as a primary pathway for fluid movement.

Schlemm’s canal endothelium pores

Direct aqueous passage through the non-fenestrated endothelium of the inner wall of Schlemm’s canal is controversial, but such a mechanism of passage is supported by a number of studies. If aqueous passes directly through the inner wall endothelium, it must do so by a mechanism very different from that of other non-fenestrated endothelia. Hydraulic conductivity is calculated from knowing the surface area of an endothelium and the total fluid volume crossing that area. Such calculations indicate that hydraulic conductivity required to explain aqueous flow into Schlemm’s canal is 100 times greater than any known non-fenestrated endothelium. Transient transcellular pores offer one possible mechanism to explain the required high hydraulic conductivity. The proposed pores are of two different types: the first type is transcellular pores and the second type is pores at intercellular junctions.

Transcellular or cytoplasmic channels in Schlemm’s canal endothelial cells are proposed as one mechanism for aqueous passage to Schlemm’s canal. Tripathi has suggested that these channels form and recede in cyclic fashion (see Fig. 3-7 ). The cycle begins with an invagination on the trabecular side of the endothelial cell and progresses to a transcellular channel with a small pore opening into Schlemm’s canal. The model envisions only a small fraction of the invaginations open into the canal at any one time. It also proposes infrequent pore opening as a mechanism to allow the endothelium to provide the majority of the normal resistance to outflow. Transcellular pores are reported in the inner wall endothelium of Schlemm’s canal in a number of studies and some studies have also shown evidence of red blood cell and tracer passage through pores.

Paracellular pores or pathways are another proposed mechanism that may serve as an alternative or added explanation for aqueous flow to Schlemm’s canal. Investigators postulate that aqueous passes between the endothelial cells, with pressure changes causing an opening and closing of cell junctions. As evidence, investigators observe that intercellular pores or interendothelial junctions respond to changes in IOP and pharmacologic agents. The number and size of the pores increase with increasing levels of IOP, while at low levels of IOP, the invaginations and the pores disappear.

It is important to point out that some authorities dispute this theory of aqueous egress and question whether the invaginations are really part of a fluid transport system or are merely artifacts. Bill has estimated the total number and area of the pores and believes they are too numerous to account for the normal resistance to outflow. Some pores in the inner wall endothelium of Schlemm’s canal as demonstrated by scanning electron microscopy are accepted as artifactual, especially those with angular shapes. Assessment of transcellular pore frequency and size has serious sources of error because it requires arbitrary acceptance or rejection of pores across a broad spectrum of sizes and shapes (see Fig. 3-8 ). The pore spectrum ranges from very large round openings with complete absence of the anterior surface of the distending cellular invaginations (collapse) to highly angular, slightly angular or small structures with a completely round appearance. Subtle gradations in edge appearance make the grading decision difficult (see Fig. 3-8 ). Limiting assessment to round structures does not resolve the issue because round transcellular pores are also known to be a reproducible artifact of preparation conditions for scanning electron microscopy.

A recent study concluded that transcellular ‘pores are artifacts of tissue fixation or processing conditions’ but that intercellular pores may be non-artifactual. In a subsequent study, the presence of transcellular and intracellular pores was correlated, consistent with formation by a common mechanism. The latter study concluded that ‘non-linear regression of pore density versus fixative volume produced a pore density at zero fixative volume that was not statistically different from zero. If true, this implies that all (or nearly all) inner wall pores observed by scanning electron microscopy are fixation artifacts.’ Schlemm’s canal endothelium becomes extremely stretched and attenuated as IOP increases. Progressively thinner cells that develop with increasing IOP may be prone to the progressively increased artifact associated with scanning electron microscopy.

Several studies have not found pores, but rather have found Schlemm’s canal inner wall endothelium to be a continuous lining. Studies also demonstrate that the inner wall endothelium acts as a barrier to the passage of particles such as ferritin or red blood cells.

Sonderman’s canals invaginate into the trabecular meshwork

Meandering invaginations of the inner wall endothelium in some instances appear to form circular or oval endothelial-lined diverticulae within the meshwork. By light microscopy, these diverticulae have been reported to provide a communication between Schlemm’s canal and the anterior chamber. Transmission electron microscopy studies, however, have since indicated there is no direct communication.


Along the external wall of Schlemm’s canal a series of obliquely oriented septa are present at the entrance to collector channels. Because of their oblique orientation, in single sections septa sometimes appear to be attached to Schlemm’s canal posterior wall, but suspended unattached in Schlemm’s canal anteriorly. Septa are composed of dense parallel collagen bundles that are continuous with and have staining characteristics identical to collagen bundles of the sclera. The septa typically join the collagenous walls of the canal rather than attaching to the trabecular meshwork. Although the canal is occasionally separated into channels, it is usually a single lumen. Collector channels are at times quite large (50–70 micron) and course circumferentially adjacent to the canal for a considerable distance. In individual histologic sections, collagenous partitions between large collector channels and Schlemm’s canal may be confused with septa within the canal and thus appear to divide the canal into more than one channel.

Schlemm’s canal valves spanning across Schlemm’s canal

Schlemm’s canal valves arise from the inner wall endothelium of Schlemm’s canal, develop a cylindrical configuration and course across the canal to attach to the external wall ( Fig. 3-9 ). The valve walls are continuous with Schlemm’s canal inner wall endothelium as documented by light, scanning, and transmission electron microscopy. Laboratory evidence of a lumen continuous with the juxtacanalicular space consists of studies with the dissecting, light, scanning, and transmission electron microscope. Tracer studies demonstrate a communication between the anterior chamber and Schlemm’s canal through the lumen of the structures using both antegrade and retrograde techniques. Pigment granules and amorphous extracellular matrix material are present in the lumen of the aqueous valves, a finding like that in the juxtacanalicular space. The valve walls distend and recoil in response to changes in IOP (see Fig. 3-9 ). The presence of Schlemm’s canal valves is documented at the operating microscope in living human eyes during Schlemm’s canal surgery. In vivo , aqueous enters Schlemm’s canal in a pulsatile fashion through structures with a size and configuration like those characterized in the laboratory (see Fig. 3-9 ). Functionally, the aqueous valves have been proposed as part of a mechanical pump discharging aqueous to Schlemm’s canal. The lumen of the valves is compressed between the walls of Schlemm’s canal at a relatively low IOP of 25 mmHg.

Fig. 3-9

Trabecular meshwork (TM) is collapsed and Schlemm’s canal (SC) is widely dilated. Schlemm’s canal valves (SCV) arise from inner wall of Schlemm’s canal, then course across the canal to attach to the external or corneoscleral wall (CSW). (A) Intraocular pressure below episcleral venous pressure. Schlemm’s canal valves are stretched across the dilated canal, the valve walls recoil and the valve lumen is small. Pressure surrounding the valves is higher than pressure in the lumen so blood cannot easily reflux through the valve lumen. (B) Aqueous outflow system appearance with IOP at physiologic levels. Schlemm’s canal valves are oriented circumferentially in Schlemm’s canal, the valve walls distend and the valve lumen is large. (C, D, E) Aqueous discharge to Schlemm’s canal during one systolic pulse wave in the human eye. The transparent trabecular meshwork (TM) permits visualization of blood intentionally refluxed into Schlemm’s canal. (C) Clear aqueous in funnel-shaped area at base of trabecular meshwork (arrow). (D) Regularly recurring sequential aqueous flow into more distal cylindrical region with site of initial swirling eddies of aqueous and blood indicating initial aqueous entry to Schlemm’s canal (arrow). (E) Clear aqueous column ejected into Schlemm’s canal as indicated by swirling eddies of mixing aqueous and blood cause circumferential flow along Schlemm’s canal for a considerable distance (arrow) beyond initial mixing area identified in D .

( A and B from Johnstone MA: Pressure-dependent changes in the configuration of the endothelial tubules of Schlemm’s canal, Am J Ophthalmol 78 : 630–8, 1974. Copyright by American Journal of Ophthalmology.) ( C–E , video courtesy of Robert Stegmann.)

Herniations or protrusions of Schlemm’s canal inner wall

As the inner wall endothelium of Schlemm’s canal distends outward in response to pressure, the distending endothelial wall forms projections, herniations or protrusions into Schlemm’s canal. These protrusions do not attach between the walls of the canal, and the distention is completely reversible with movement away from the external wall when pressure is low. Evidence indicates that there is no direct opening of the herniations into the canal. The original study of the aqueous valves illustrates, but does not emphasize, that the aqueous valves are always attached to the external wall of Schlemm’s canal and have a valve-like arrangement at the level of the external wall.

One study interpreted the herniations as being the same structures as the aqueous valves and concluded that they could not carry aqueous to Schlemm’s canal. However, a salient feature of the aqueous valves is their attachment to the external wall, thus suspending them within the canal. The study of the protrusions or herniations completely separated the walls of Schlemm’s canal, in the process disrupting the ‘tissue strands’ or aqueous valves spanning the canal, and excluded the regions of disruption from observation. Although the study was valuable in further characterizing the herniations or protrusions, the study could not address the appearance or function of the valves spanning Schlemm’s canal.

Collector channels, aqueous veins and episcleral veins

Schlemm’s canal is drained by a series of collector channels that in turn drain into a complex system of intrascleral, episcleral, and subconjunctival venous plexus. The collector channels arise from the outer wall of Schlemm’s canal at irregular intervals (0.3–2.8 mm) that average 1.2 per mm creating a total of 20–30 collector channels. At the origin of some collector channels, torus or lip-like openings are observed that are associated with septa. Septa at collector channel ostia limit or prevent trabecular tissue from completely occluding the opening. A few (4–6) direct collector channels (≈70 micron diameter) proceed directly from Schlemm’s canal through the sclera thus communicating directly with aqueous veins on the surface of the eye. Indirect collector channels are smaller (≈50 micron diameter), more numerous (15–20) and enter into the intrascleral drainage network. A few (4–6) intermediate types are present. Aqueous veins empty into episcleral and conjunctival veins. Where aqueous and episcleral veins join, characteristic laminar flow of aqueous humor and blood is seen on slit-lamp examination at the limbus. A number of manifestations of pulsatile discharge of aqueous into the episcleral veins is also seen.


Glaucoma results from an abnormality of the resistance characteristics of the outflow system, but the actual nature of that resistance remains controversial. The region of the trabecular beams is an unlikely source of significant resistance because of the large openings in the area and the lack of significant extracellular matrix material in the region. Investigators propose two very different models of the resistance location and mechanism. The first model envisions the main resistance localized to the juxtacanalicular space. The juxtacanalicular space acts as a syncytium of extracellular matrix material and elastic-like fiber network that attaches to Schlemm’s canal endothelium. The syncytium must provide a sufficiently stable geometry so that the extracellular matrix material can act as a passive filter regulating resistance. After passing through the juxtacanalicular resistance, aqueous passes through low-resistance pores in Schlemm’s canal endothelium.

The second model places the initial resistance to IOP-generated forces at Schlemm’s canal endothelium. The model necessitates redistribution of IOP-induced resistive forces at Schlemm’s canal endothelium to structural elements throughout a tensionally integrated trabecular meshwork. The force redistribution takes place via cytoplasmic process attachments to Schlemm’s canal endothelium. A second component of the Schlemm’s canal endothelium/trabecular meshwork resistance model is pressure-induced distention of Schlemm’s canal inner wall: such a distention leads to apposition between Schlemm’s canal walls. Schlemm’s canal wall apposition thus becomes a resistance element integral to the model.


The juxtacanalicular region is posited as a reasonable candidate for much of outflow resistance. Especially in non-pressurized eyes, the space is narrower than the region of the trabecular lamella and contains a greater concentration of extracellular and cellular elements than the rest of the meshwork. The juxtacanalicular space contains hyaluronic acid, other glycosaminoglycans (GAGs), other glycoproteins and fibronectin. An elastic-like fiber network along with cellular elements, fibrils, and structural proteins is described as creating a three-dimensional cellular sponge or syncytium. One may deduce that such a stable syncytium will be able to provide a resistance unit restricting flow.

Glycosaminoglycans have been proposed as a key physiologic component of this resistance. The GAGs are found as components of larger proteoglycans and generally function as a part of these larger molecules. The GAGs are heavily hydrated and able to trap a large amount of water. Therefore GAGs are able to fill a very large hydrodynamic volume. Through hydration and fluid trapping mechanisms, the GAGs are proposed to reduce the functional diameter of flow channels through the juxtacanalicular tissues. A funneling mechanism dependent on the presence of GAGs and Schlemm’s canal inner wall pores has been proposed to alter effective resistance to flow, although two studies failed to find a correlation between outflow and pore density.

The previously discussed evidence is indirect, but direct evidence is cited to indicate that the region accounts for about 75% of resistance. Because the evidence is direct, it assumes special importance and warrants careful scrutiny. The investigators carefully pointed out that the micropipette used to cannulate Schlemm’s canal was 25 times the thickness of the highly compliant inner wall endothelium of Schlemm’s canal and that the actual location of the tip was not known at the time of measurements. They also referenced the previously identified compliance characteristics of the tissues as a possible cause of inaccurate interpretation of their results. Two other studies, involving Schlemm’s canal microcannulation, did not find a high proportion of the resistance within the juxtacanalicular space.

Alterations of extracellular matrix materials occur after laser trabeculoplasty. Upregulation of metalloproteinase synthesis also accompanies improvement of aqueous outflow following laser trabeculoplasty. This relationship between metalloproteinase upregulation and outflow improvement is offered as evidence that alterations in the extracellular matrix distribution in the juxtacanalicular space contribute to outflow resistance. However, the response to injury is complex as is illustrated by evidence of repopulation of trabecular meshwork cells following injury. Of interest, most of the extracellular matrix subject to remodeling responses is in the trabecular beams.

Some evidence does not favor the juxtacanalicular space resistance model. The model requires relatively unchanging juxtacanalicular space geometry so that the space may maintain its resistance characteristics. Studies of tissue biomechanics involving tissue loading described later in this chapter demonstrate that the juxtacanalicular space undergoes a two- to three-fold enlargement in response to physiologic increases in pressure. The same pressure increases induce marked increases in resistance, making the juxtacanalicular space a less likely site of significant resistance in normal eyes.

Although spaces are present in the juxtacanalicular region, in hypotonous eyes that may be construed as being filled and held open by a syncytium of extracellular matrix material, studies of biomechanics that examine boundary conditions indicate that extracellular matrix material does not act as a space-occupying syncytium. When IOP is reduced to zero in living eyes, blood refluxes into Schlemm’s canal in response to the episcleral venous pressure gradient of about 8–9 mmHg. Under these conditions, the juxta-canalicular space does not act as a space-occupying syncytium. Rather, the juxtacanalicular space is almost completely obliterated. A pressure gradient reversal of as little as 5 mmHg causes obliteration of the space. The elimination of the juxtacanalicular space in response to a modest pressure reversal is thought to indicate that the spaces within the juxtacanalicular region are dependent on hydrostatic pressure rather than on the rigidity induced by a syncytium of extracellular matrix material.

Histochemical studies provide additional insights. For many years it was assumed that the apparently open spaces of the trabecular meshwork were filled completely with a GAG gel that was washed out in conventional histologic processing. In recent years it has been possible to localize histochemically hyaluronan, and it is now clear that most of the open spaces are not filled with gel. Some hyaluronan is found in the juxtacanalicular region, but the amount decreases with age and no morphological study has demonstrated extracellular matrix that could generate the measured outflow resistance.


Different lines of evidence support the idea of Schlemm’s canal endothelium as the major site of resistance to aqueous outflow. Tracer studies demonstrate accumulation of material at the inner wall of Schlemm’s canal. Other evidence offered in support of Schlemm’s canal endothelium as the main resistance site comes from the improvement in outflow facility that follows experimental infusion of certain substances, such as iodoacetic acid, N -ethylmaleimide, cytochalasin B or D, EDTA, and colchicine. Histologic studies suggest that these agents alter the inner wall of Schlemm’s canal. However, disruption of Schlemm’s canal causes washout of extracellular material in the juxtacanalicular space. Because of the contemporaneous loss of extracellular matrix material with the above studies, disruption of Schlemm’s canal endothelium is of unclear value in discriminating between the juxtacanalicular space and Schlemm’s canal endothelium as the primary resistance site.


Studies of cellular biomechanics point to Schlemm’s canal inner wall endothelium and the trabecular meshwork acting as a unified resistance unit.


The principles of biomechanics require, in turn, study of tissue geometry, tissue composition, laboratory effects of tissue loading, boundary conditions, and, finally, in-vivo effects of tissue loading. Tissue composition and geometry, issues discussed previously in this chapter, determine constraints and possible responses of the tissues to external forces.

Tissue loading studies subject tissues to normally encountered forces to determine force-induced responses. Load-bearing structural elements respond by characteristic changes in configuration. In other words, tissues causing the resistance are the ones that undergo configuration changes appropriate to the loading forces they experience. The applicable loading force in the aqueous outflow system is IOP. Boundary conditions define the maximum limits of tissue responses to induced forces. In-vivo tissue loading responses are discussed in a later section and are the most crucial test of the validity of conclusions from the laboratory.


Tissue loading induced by IOP, both in vitro and in vivo , consistently demonstrates progressive distention of Schlemm’s canal inner wall endothelium that correlates with IOP increases. Evidence from these same studies follows, demonstrating that the IOP-induced load on the endothelium is then distributed to the entire trabecular meshwork (see Fig. 3-4 ).

To induce tissue loading, the pressure gradient is systematically raised above zero (in the enucleated eye, pressure above hypotony, or in living eyes, pressure above episcleral venous pressure ≈8–9 mm). Schlemm’s inner wall begins its outward distention when the pressure gradient is as low as 5 mmHg. Distention of Schlemm’s canal inner wall continues progressively, both within the physiologic pressure range and beyond. Concomitently, inner wall distention causes the juxtacanalicular space to enlarge, as much as two- to three-fold. Because of their anchoring attachments to the distending wall of Schlemm’s canal endothelium, trabecular lamellae move progressively outward toward Schlemm’s canal lumen, thus developing progressively increased spacing between lamellae. Cytoplasmic processes throughout the meshwork undergo progressive changes from a parallel to a perpendicular orientation. The processes, initially short and stubby, undergo elongation and thinning both in the juxtacanalicular and intertrabecular spaces. A more pronounced longitudinal orientation of the cytoskeletal filaments of the processes develops as IOP increases.

At the cellular level, Schlemm’s canal endothelial cell membrane and cytoplasmic contents, as well as the nuclear envelope and its contents, change shape in a progressive fashion from a spherical configuration in hypotony to an elongated plate-like configuration. At cell process origins, the cytoplasm and nucleus reorganize from a neutral to an elongated cone-shaped configuration in response to tension. Juxtacanalicular cells undergo a change in configuration involving the cell membrane, the cytoplasm, the nuclear envelope, and the nuclear contents, all of which develop a progressively more cone-shaped appearance directed toward cell process origins. Such cellular changes are reflective of stresses induced by progressive tension that develops between Schlemm’s canal endothelium and the restraining trabecular lamellae.

Tissue loading by IOP thus provides evidence at both the tissue and cellular levels, placing trabecular meshwork resistance to IOP at Schlemm’s canal endothelium. Attachments of Schlemm’s canal endothelium to the underlying trabecular meshwork provide a dynamic tensional integration between the endothelium and the underlying load-bearing trabecular tissues. In contrast, tissue loading studies provide no evidence of hydraulic resistance in the juxtacanalicular space. Juxtacanalicular space enlargement and reduced compaction of both extracellular and cellular elements occurs. This juxtacanalicular space enlargement progressively reduces the ability of the juxtacanalicular space to act as a determinate of resistance, yet as IOP increases, measured resistance to aqueous outflow increases, making this region an unlikely source of resistance.


High IOP induces trabecular meshwork distention and Schlemm’s canal lumen collapse ( Fig. 3-10 ). As IOP progressively increases, Schlemm’s canal endothelium progressively distends into the lumen of the canal. At higher IOPs, Schlemm’s canal endothelium becomes appositional to the external or corneoscleral wall of Schlemm’s canal, effectively occluding much of the canal lumen. No further excursions can occur. Aqueous cannot easily pass across Schlemm’s canal endothelium in these regions and circumferential flow to collector channel ostia is progressively compromised. For example, in one report, one-third of sections had over 75% of the angle closed at 20 mmHg. This finding in enucleated eyes may result from an absence of ciliary body tone and normal episceral back pressure. However, in living primates, the walls of the canal also become appositional, with fairly extensive apposition present at a relatively low IOP of 20–25 mmHg.

Fig. 3-10

Boundary conditions explore the limits of tissue excursions in response to physical forces. In the outflow system, boundary conditions are determined by maximal trabecular meshwork (TM) excursions induced by IOP. Upper figures : micrographs of boundary conditions. Lower figures : illustrations of boundary conditions. Upper figures A and C : micrographs of two eyes of same primate with eyes fixed simultaneously in vivo . B : human eye fixed in hypotony. (A) (Low IOP) IOP = 0 mmHg, episcleral venous pressure ≈8 mmHg. Schlemm’s canal (SC) is large and trabecular tissues are completely collapsed with obliteration of juxtacanalicular space. No further excursion is possible. (B) (Neutral IOP) Eyes fixed in hypotony, trabecular tissues in neutral position. (C) (High IOP) IOP = 25 mmHg during fixation. Schlemm’s canal lumen is reduced to a potential space. Schlemm’s canal endothelium distends to reach Schlemm’s canal external or corneoscleral wall (CSW). No further excursion can take place when the external wall is reached. The juxtacanalicular space is large. Large spaces are present between the trabecular lamellae. Red blood cells (RBC) are present in SC. (N, nucleus of Schlemm’s canal endothelial cell.)

Modified from Johnstone MA, Grant WM: Pressure-dependent changes in structure of the aqueous outflow system in human and monkey eyes, Am J Ophthalmol 75:380, 1973. Published with permission from the American Journal of Ophthalmology.

Low IOP induces trabecular meshwork collapse and Schlemm’s canal lumen dilation (see Fig. 3-10 ). When IOP is reduced below episcleral venous pressure (≈4–8 mmHg), Schlemm’s canal endothelium moves inward toward the anterior chamber. The trabecular lamellae nearest Schlemm’s canal are compressed together to form a uniform, solid-appearing sheet of tissue. No further excursions of Schlemm’s canal endothelium can occur. Additionally, no blood crosses this tissue, leading in the initial report of the behavior to the proposal that the configuration provides a means of assuring maintenance of the blood aqueous barrier.



Grant’s microsurgical and perfusion studies stand as the foundation for characterizing the location of aqueous outflow resistance. Thus, it is important to understand the evidence and conclusions from his work. His earliest studies are often quoted, which indicated that 75% of aqueous outflow resistance is at the level of the trabecular meshwork. Grant’s subsequent studies provide a very different picture, pointing to apposition between the walls of Schlemm’s canal as the explanation for much of normal outflow resistance as well as abnormal resistance in glaucoma.

Grant’s studies in normal enucleated eyes demonstrates that removal of the trabecular meshwork eliminates 75% of the resistance to aqueous outflow. Trabecular meshwork removal also eliminates abnormal resistance found in glaucoma eyes. The assumption was made that tissue removal with a cystitome was limited to the trabecular meshwork because gross examination during the procedure was consistent with a relatively precise removal of the meshwork tissue.

Later, Grant and co-workers duplicated the original studies, duplicating the previous 75% improvement in aqueous outflow, this time including histologic evaluation. Histologic studies revealed that the cystitome damaged the external wall, collector channel ostia and structures within Schlemm’s canal. Grant and co-workers concluded that the outflow improvement seen in his earlier studies could not be entirely attributable to resistance generated by the trabecular meshwork and Schlemm’s canal inner wall, because structures within the canal and along the external wall were also damaged.

Additional studies point to a relationship between the walls of Schlemm’s canal as the probable cause of much of outflow resistance. Perfusion studies by Grant and colleagues demonstrate increasing resistance with increasing IOP, a finding that is more prominent on glaucoma eyes. Perfusion of the anterior chamber without peripheral iridectomy causes reverse pupillary block and chamber deepening, by pulling the scleral spur backward thus reducing Schlemm’s canal wall apposition. When the walls of Schlemm’s canal are held more widely apart by this method, the increasing resistance with increasing pressure is eliminated. These findings led Grant and co-workers to propose a new model involving apposition of Schlemm’s canal inner wall and outer wall as the cause of increasing resistance. Furthermore, the abnormally steep increase in resistance they found in glaucoma eyes led them to propose that the abnormality in glaucoma was a result of this same variable resistance mechanism associated with compression of the trabecular tissues against Schlemm’s canal outer wall.

Further studies showed that removal of the external wall of Schlemm’s canal causes greater than 50% reduction in outflow resistance, just as does removal of the internal wall. The excess sum of resistances related to internalization or externalization of Schlemm’s canal led them to conclude that an intact and unyielding outer wall of Schlemm’s canal limits stretching of the inner wall and is required for maintenance of normal resistance. The result is attributed to reduced ability of fluid to enter Schlemm’s canal in regions of apposition and also reduced ability of aqueous to move circumferentially in Schlemm’s canal to collector channel entrances.

Systematic tissue loading by IOP provided direct evidence that the trabecular meshwork progressively distends to come into apposition with the external wall of Schlemm’s canal in enucleated, but more importantly in living, eyes. Extensive Schlemm’s canal wall apposition begins to develop at relatively low pressures (20–25 mmHg) in the living eye with normal ciliary body tone and episceral venous pressure. The septa around collector channel ostia tend to keep the two walls separated, but these regions only represent a small per cent of Schlemm’s canal circumference. Van Buskirk and Grant found that aqueous humor did not flow more than 10° around the canal of enucleated human eyes. When a 30° trabeculotomy was done, facility calculations indicated that aqueous only flowed circumferentially a total of 10° to each side of the Schlemm’s canal opening. When lens depression then further separated the walls of the canal, facility calculations indicated that aqueous flowed circumferentially a total of 35° on each side of the trabeculotomy site.

In support of the conclusion that facility improvement with circumferential flow is improved by reduction in Schlemm’s canal apposition, the authors noted the effect of limited trabeculotomy reported by Barany et al. Dilation of Schlemm’s canal by pilocarpine still has a large effect in improving facility, indicating that circumferential flow is minimal after a limited trabeculotomy, and is consistent with a persistent facility improving benefit of pilocarpine related to its ability to reduce apposition of Schlemm’s canal walls.

Van Buskirk subsequently correlated lens depression experiments that dilate Schlemm’s with perfusion and histologic data. Resistance changes correlated with the degree of apposition between Schlemm’s canal walls. Outflow facility was affected by lens depression 1% at 2.5 mmHg when there was no canal wall apposition, but 89% at 25 mmHg where significant apposition of Schlemm’s canal was present. As IOP increases causing increasing apposition of Schlemm’s canal walls, lens depression to separate the walls is progressively more effective in improving flow with a remarkably high correlation coefficient of 0.97.

An additional study compared resistance characteristics at low pressure (7 mmHg) with those at higher pressure (25 mmHg) by doing sequential internal trabeculotomy. At a higher pressure, where the canal walls are more extensively appositional, trabeculotomy has a greater effect on reducing resistance, further supporting the concept of Schlemm’s canal apposition as the mechanism causing the variable resistance.



The traditional model of aqueous flow is that of a passive, non-energy-dependent bulk fluid movement down a pressure gradient with aqueous leaving the eye primarily by the canalicular route. The model is recognized as being somewhat oversimplified because of a component of uveoscleral flow. In the model, aqueous is forced through a syncytium of extracellular matrix material in the juxtacanalicular space that acts as a passive resistance unit controlling pressure and flow. Evidence favoring the model is the finding of similar aqueous flow rates in living and enucleated eyes.


Aqueous flow through the outflow system has long been regarded as a passive phenomenon as noted above, but a recent model proposes that the outflow system acts as a biomechanical pump. The outflow system is constantly subjected to oscillatory pressure transients caused by the ocular pulse, blinking and eye movement. In this model, elastic and contractile tissues of the trabecular meshwork and valves within Schlemm’s canal stretch in response to transient pressure increases. The energy stored during distention is released when the pressure transients decay, causing the tissues to recoil to their prior configuration. The pressure transients thus enable energy-dependent pulsatile fluid movement through the outflow system.

Laboratory evidence of a highly compliant trabecular meshwork and valves predict the presence of a pumping mechanism in the outflow system coupled to IOP transients. Aqueous outflow responses to pressure transients in humans represent a means of validating the predictions; in effect, a means of observing in-vivo tissue loading. The unique optical properties of the eye provide an opportunity to examine the effects of tissue loading at a scale usually associated with laboratory histologic examination. These observations at high magnification (×80) demonstrate pulsatile flow of aqueous from the anterior chamber into Schlemm’s canal (see Fig. 3-9 ), and from Schlemm’s canal into collector channels, a finding first observed by Stegmann. Pulsatile flow is synchronous with the ocular pulse. Pulsatile aqueous flow from Schlemm’s canal into the episcleral veins is observed in response to the ocular pulse as well as in response to blinking and eye movements.

In the aqueous pump model, short-term pressure homeostasis occurs via alterations in the stroke volume of aqueous discharged to the episcleral veins in response to pressure transients. Long-term homeostasis occurs by intrinsic mechanotransduction mechanisms that optimize the pump function, analogous to homeostatic mechanisms elsewhere in the vasculature.


A number of authorities have raised the question of whether the trabecular meshwork directly regulates IOP by extrinsic mechanisms. These authorities have postulated the existence of a receptor in the trabecular meshwork that responds to pressure or tissue distention and then provides feedback to modulate some process, such as aqueous humor formation, ciliary muscle tone, or glycosaminoglycan synthesis. Although this concept remains intriguing, there is no evidence to support that such a feedback system exists. In fact, in the studies done to date, elevations or depressions of IOP are not accompanied by substantial changes in the rate of aqueous humor formation, suggesting that pressure regulation is not aqueous secretion dependent (see Ch. 2 ).


A lesser amount of the aqueous humor exits the eye by an alternate route through the ciliary muscle, the iris, the sclera, and other structures of the anterior segment ( Fig. 3-11 ). This alternate pathway is known by a number of terms, including uveoscleral , unconventional , extracanalicular , and uveovortex flow . As Bill pointed out, flow through the trabecular meshwork and Schlemm’s canal seems to involve a well-designed system. In contrast, uveoscleral flow seems more primitive and resembles a leak more than a well-designed fluid transport system.

Fig. 3-11

Aqueous humor leaving the eye by trabeculocanalicular flow and uveoscleral flow.

Aqueous humor enters the ciliary muscle through the uveal trabecular meshwork, the ciliary body face, and the iris root. The fluid passes posteriorly between the bundles of the ciliary muscle until it reaches the supraciliary and suprachoroidal spaces. Aqueous humor leaves the eye through the spaces around the penetrating nerves and blood vessels and through the sclera. Even large molecules such as horseradish peroxidase and albumin can pass through intact sclera.

A few investigators have questioned whether aqueous humor can exit the eye by entering the uveal vascular system. Tracer studies indicate that there is some exchange of substances between the aqueous humor and the plasma in the uveal blood vessels. However, the net fluid flow into the uveal vascular system is quite low for a number of reasons. First, the iris capillaries have thick walls that restrict movement of water and ions. Furthermore, pressure in the uveal capillaries is higher than IOP. This pressure difference partially offsets the difference in oncotic pressure between the plasma and the tissue fluid of the uveal tract. Thus there is little driving force for fluid to cross the capillary walls.

Uveoscleral flow seems to be present in most species, but the portion of the aqueous humor transported by this system varies considerably. For example, unconventional flow constitutes about 3% of the total outflow in rabbit eyes but represents more than 50% of the total outflow in some species of monkeys. In human eyes, the unconventional pathway is estimated to carry 5–25% of the total aqueous outflow. It should be pointed out that direct measurements of uveoscleral flow in humans have been limited to a few eyes, many of which were scheduled for enucleation because of intraocular tumors. It is possible that such eyes are atypical and that the results are not representative of normal eyes. Calculations based on non-invasive measurements indicate that uveoscleral outflow may constitute as much as 35% of the outflow. Furthermore, in primate and human studies, uveoscleral outflow increases up to four-fold when the anterior segment is inflamed.

Uveoscleral flow has been studied with a variety of tracer substances, including fluorescein, radiolabeled molecules, and small plastic spheres. In human experiments, I-albumin can be traced by autoradiography from the anterior uveal tract to the posterior pole.

Uveoscleral flow increases when IOP is raised from atmospheric pressure to the level of episcleral venous pressure. However, above this pressure level uveoscleral flow is largely independent of IOP. An increased IOP provides a greater driving force for uveoscleral flow, but it also compacts the anterior ciliary muscle bundles. These two factors must nearly offset one another, because in the uninflamed eye, the facility of uveoscleral flow is quite low at 0.02–0.052 μl/min/mmHg.

The main resistance to uveoscleral flow is the tone of the ciliary muscle. Factors that contract the ciliary muscle (such as pilocarpine) lower uveoscleral flow, whereas factors that relax the ciliary muscle (such as atropine) raise uveoscleral flow. Uveoscleral outflow is increased significantly by prostaglandins. Prostaglandins in low dose are among the most potent IOP-lowering agents available.

As mentioned above, pilocarpine decreases and atropine increases uveoscleral flow. This is consistent with a large body of work indicating that the therapeutic effect of pilocarpine in most glaucoma patients reflects increased trabecular outflow (caused by contraction of the ciliary muscle). Some studies have shown that pilocarpine antagonizes therapeutic prostaglandin agents, but clinical experience is mixed in this area. A few studies indicate that epinephrine may lower IOP, in large part by increasing uveoscleral flow. Cyclodialysis is an operation designed to lower IOP by detaching a portion of the ciliary body from the scleral spur. There is evidence that cyclodialysis acts to increase uveoscleral flow.



The Goldmann equation can be rearranged to give a simplified view of the factors that determine the ease with which aqueous humor leaves the eye by conventional outflow:

<SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='C=FPO-Pv’>C=FPOPvC=FPO-Pv
C = F P O – P v

Only gold members can continue reading. Log In or Register to continue

Feb 12, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Aqueous humor outflow system overview

Full access? Get Clinical Tree

Get Clinical Tree app for offline access