Stromal Hydration Research Articles

The intraocular changes of anterior segment necrosis

Anterior segment necrosis is an acute or chronic process occasioned by embarrassment of the blood supply of the anterior segment of the eye. In the acute form this vascular obstruction leads to severe corneal oedema, necrosis of anterior uvea, hypotony and cataract formation. Depression of aqueous humour formation accounts for severe reduction of glucose levels in corneal stroma and aqueous humour lasting for two days after cautery of the long posterior ciliary arteries (LPCA) in rabbits. Lactate levels are initially significantly elevated but return to normal after one week. Stromal hydration was elevated for one week but then returned to normal. Corneal epithelial glycogen was diminished at one and two days after surgery but then returned to normal. Although unproven, oxygen deprivation probably plays a major role in endothelial ischaemia and therefore corneal oedema. It is concluded that the abnormalities seen in anterior segment necrosis stem from changes in aqueous metabolic components resulting from severely reduced aqueous turnover. Hyperbaric oxygen and intracameral metabolite substitution are unproven treatments but merit further experimental study.

Binding of peanut agglutinin on fresh and organ-cultured porcine corneal tissue.

To investigate the binding of peanut-agglutinin (PNA) on porcine corneal tissue, several methods of tissue processing were used. In cryostat sections of postmortem tissue, PNA labeling of epithelial cell membranes was detected. In the stroma, a fibrillar structure parallel to the corneal surface was found. The anterior part of Descemet's membrane was strongly positive, whereas the posterior Descemet membrane was negative. The endothelium was faintly positive. After 3 days of organ culture, the epithelial and endothelial staining patterns were not significantly changed. The stromal fibrils detected by PNA staining showed a considerable spatial disarrangement due to the stromal hydration.

A role for glutathione and glutathione reductase in control of corneal hydration

We have assessed the importance of the glutathione redox system of the corneal endothelial cells in the control of stromal hydration. The ability of freshly isolated corneas to maintain normal hydration during perfusion, while the activity of glutathione reductase was inhibited with 1,3- bis(2-chloroethyl)-1-nitrosourea (BCNU), was tested under a number of conditions. BCNU at 125 μ m led to mild swelling and at 500 μ m to more severe swelling. Swelling was also severe when 50 μ m H 2O 2 was added together with 125 μ m BCNU. The concentration of total glutathione (GSH + GSSG) in the endothelia of these corneas was decreased and the fraction found in the oxidized form was higher than in controls without BCNU. Glutathione reductase activity was inhibited in the perfused endothelia by 90%. With the addition of 0·5 m m GSSG during perfusion, swelling due to 125 μ m BCNU was at a slightly lower rate and reductase activity was inhibited only 79%. Complete proteetion was afforded against 125 μ m BCNU induced swelling by 1·0 m m GSH, corneas maintaining normal thickness for over 4 hr. The endothelial surface of the perfused corneas was shown by seanning electron microscopy to be little disturbed by the low concentration of BCNU, but at high concentration, or with H 2O 2 added, the cells were collapsed and had a heavily pitted appearance. We conclude that when the cornea is under oxidative stress (e.g. in the presence of H 2O 2) a rapid turnover of endothelial GSH via glutathione reductase and the hexose monophosphate shunt is required. Under conditions of less stress, the reduced need for GSH to counteract oxidative threats can be supplied either exogenously or by the partially inhibited reductase.

Rabbit cornea stromal hydration measured with proton NMR spectroscopy.

The suitability of water proton spin-lattice relaxation time T1 as a measure of corneal dysfunction is investigated in this paper. Proton nuclear magnetic resonance measurements were performed on excised rabbit corneas using the saturation recovery method to determine the proton spin-lattice relaxation time T1. Both freshly excised and progressively swollen rabbit corneas were studied. The experimental results are consistent with a two compartment model of "bound" water and "free" water. This model results in a linear correlation between 1 / T1 (measured) and 1/corneal hydration. This correlation suggests that water proton T1 values can be used as an index of corneal dysfunction.

Numerical solution of coupled transport equations applied to corneal hydration dynamics.

1. A quantitative basis for the currently accepted theory on the regulation of corneal hydration was derived using the technique of finite element analysis to integrate a set of coupled flow equations. The model was based on non-equilibrium thermodynamics and incorporated the transport and permeability properties of the corneal epithelium and endothelium as well as the gel properties of the central connective tissue layer. 2. Considerable errors were introduced in the prediction of corneal hydration dynamics (unsteady-state behaviour) unless allowance was made for the development of trans-stromal gradients in pressure and solute concentration. 3. Thickness of in vitro rabbit corneal epithelium and stroma were measured with an automatic specular microscope during responses to changes in the osmolarity of the tear-side bathing medium. The time course of these experiments was fitted with the mathematical model to obtain a set of membrane phenomenological coefficients and transport rates. 4. The model with the redetermined membrane parameters was tested by predicting the influence of other variations in boundary conditions with excellent match to several well-documented experimental observations, including an explanation for the slight stromal swelling observed in hibernating mammals. 5. The regulation of corneal stromal hydration can be explained accurately by balance between the dissipative flows across the serial array of corneal layers and the active HCO3 transport by the endothelium, supporting the earlier 'pump-leak' hypothesis. 6. It was found that stromal retardation of fluid flow, as well as gradients in solute concentration, significantly influences the dynamics of corneal stroma hydration. Tissue gel properties may be a more important factor in coupled transport across cell layers than generally appreciated.

Unsteady aspects of corneal thickness control

Unsteady-state calculations of rabbit corneal thickness dynamics show that the stromal hydration changes slowly in response to the normal sleep/wake variations in tear osmotic pressure. Consequently, the corneal thickness never varies very far from thatthickness which it would attain were the tear tonicity held constant at a value between those characteristic of the open eye and the closed eye. This intermediate “time-average” tonicity exceeds that of the stromal fluid, and it is shown that, in the normal eye, the consequent osmotic driving force across the epithelium, averaged over time, can generate sufficient flow out across this cell layer to balance completely stromal imbibition across the endothelium. The implications of this result as regards the role of active deturgescent transport are discussed, and some directions for synthesizing current hydration control concepts are indicated.

Mass Transfer in the Cornea: II. Ion Transport and Electrical Properties of a Series Membrane Tissue

The electrical and active transport properties of isolated rabbit cornea are investigated by computer experimentation. The tissue is modeled as a series membrane system and the passive ion fluxes through it are described by the frictional formulation of irreversible thermodynamics. From short-circuit current (SCC) data, it is found that the epithelial sodium pump rate (P) is not appreciably changed when much of the sodium in the solution bathing the anterior corneal surface (concentration = c(11)) is replaced by choline, with choline-free medium posteriorly. Simulations of open-circuited corneas, using the mean P computed from the SCC data, yield corneal and stromal potentials in agreement with experiment. The stromal fluid is calculated to become more hypotonic as c(11) is diminished, a result consistent with posttest measurements of the sodium content of experimental stromata. The apparent decrease in "bound sodium" which accompanies the reduction of c(11) is a result of the associated changes in steady stromal hydration; the epithelial sodium pump does not contribute to corneal deturgescence. The inclusion of a simple epithelial structure in the computations changes the value of P but affects neither its constancy nor the calculated behavior of the cornea under open-circuit conditions. A general algebraic relation among pump rates and ion fluxes in short-circuited series membrane systems bathed in complex media is derived and used to construct a relation between P and SCC for the cornea. This equation yields pump rates in good agreement with the computer results and is used to show that (a) P is independent of c(11) if d(SCC)/dc(11) is a constant related to the over-all corneal permeability to sodium, and (b) a Lineweaver-Burke plot of 1/SCC vs. 1/c(11) can appear to be linear at constant P.

Solute permeability of the corneal endothelium and Descemet's membrane

The permeability of in vitro rabbit corneal endothelium to seven radioactive substances of various molecular weights, ranging from chloride (atomic weight 35·5) to dextran B (molecular weight 60,000–90,000), was measured. The permeability varied proportionally with the molecular weight, and linearly with respect to molecular radius. The permeability of the endothelium alone was calculated from data on the permeability of stroma and Descemet's membrane, a preparation identical to that used for endothelial determinations except for the absence of endothelium, and it was found that the endothelium comprises only 40–80% of the total resistance of the stroma, Descemet's membrane and endothelial preparation. A newly designed lucite chamber was used and a simple corneal layer removal technique devised. Electron microscopy of completely denuded stroma, and of the separation of endothelium, with or without Descemet's membrane from the stroma showed that such layer removal was complete. The permeability to mannitol of stroma alone and stroma plus Descemet's membrane at high (90%) and near-normal (76%) hydration was identical. The permeability of stroma alone and stroma plus Descemet's membrane to inulin was also equal at any specific hydration, indicating that Descemet's membrane offers no resistance to the passage of these two nonelectrolytes. A significant difference was found between the inulin permeability at high and near-normal stromal hydration (with or without Descemet's membrane) with increased permeability occurring at high hydration. This increased permeability is presumably a reflection of the widening of the collagen interfibrillar distance. At near-normal hydration, the permeability of the denuded stroma alone and stroma plus Descemet's membrane to inulin or mannitol is at least 54% (inulin) to 85% (mannitol) greater than that of the endothelium.

Anterior segment necrosis in rabbits. Metabolic and histologic changes.

Anterior segment necrosis was induced in rabbits by extensive cautery of both long posterior ciliary arteries. Metabolic and histologic changes were studied at intervals after operation. Severe depression of glucose levels in corneal stroma and aqueous lasted for two days after cautery. Lactate levels, initially very high, fell below normal at one week but had returned to normal at six weeks. Stromal hydration was elevated for one week after cautery, then became normal. Corneal epithelial glycogen was severely depleted one and two days after operation, but epithelium otherwise appeared normal. Ischemia of anterior segment was reversed after three to six weeks but irreversible changes of iris, ciliary body, and lens persisted. It is concluded that corneal complications seen in anterior segment necrosis stem from changes in aqueous metabolic components resulting from severely reduced aqueous turnover.

The anterior segments of rabbits after alkali burns. Metabolic and histologic alterations.

Alkali-burned rabbit eyes were analyzed for stromal and aqueous glucose and lactate, corneal hydration, and histologic changes. Severe alkali burns depressed aqueous glucose levels to an average of 35% of normal in 24 hours and to 5% of normal in 7 and 14 days. The stromal glucose level of 4.2μmol/gm H 2 O was lowered to an average of 60% of normal in 24 hours and to 26% of normal in 7 and 14 days. Stromal and aqueous lactates did not change. Stromal hydration was increased 24 hours after burning but was normal 7 and 14 days after burning. Milder alkali burns of the anterior segment produced less alteration of aqueous and stromal glucose concentrations. Severe hypotony and reduced open vascular channels in the iris and ciliary body suggest inadequate aqueous production through insufficient circulation.

General theory of tissue swelling with application to the corneal stroma

Equations describing the time course of swelling in a very general system are derived and are used to study the free swelling of corneal stroma. An integral solution is obtained, which demonstrates how the swelling rate depends on swelling pressure and tissue permeability at the space-average stromal hydration, and which can be integrated by quadrature to give tissue hydration as a function of time. Only limited comparisons of predicted stromal swelling curves with experimental results can be made, and these are inconclusive. Experiments to test the theory more severely are suggested.

Physiological response of the cornea to an artificial epithelium

The effect of epithelial removal and absence upon stromal hydration and thickness was studied in rabbits, cats, and humans. The epithelium was replaced with a methyl-methacrylate contact lens, affixed to the denuded stroma by cyanoacrylate adhesive. In rabbits, stromal thickness and hydration showed an increase from the day of lens application. After 20 days, the stromal sodium concentration remained unchanged, but potassium levels had decreased, supporting the histological observation of some degradation of anterior keratocytes. Cats tolerated the lens quite well, with normal stromal thickness being maintained as long as the seal around the lens was complete. Humans exhibited excellent tolerance to the procedure, with a stable stromal thickness being maintained for up to 6 months, to date. In view of the success in replacing the epithelial layer of the cornea with an intert impermeable barrier, such as a contact lens, it appears likely that the epithelium is not required, beyond its fluid barrier characteristics, to maintain normal thickness and hydration of the corneal stroma in cats and humans.

A comparative study of stromal swelling in sea lamprey spectacle and brook trout cornea

Brook trout corneal stroma immersed in water increased to 350 % of its original wet weight; whereas, the stroma of sea lamprey spectacle which contains sutural fibers increased to only 150 %. The normal arrangement and spacing of collagen fibrils was preserved in the stroma of the sea lamprey spectacle. In contrast, in the osmotically swollen brook trout corneas, large interlamellar spaces were present which contained fragments of lysed keratocytes. Control stromal tissues from both sea lamprey spectacle and trout cornea placed in 0.5 % saline showed small changes in wet weight and maintenance of normal collagen alignment. The sutural fibers present in the sea lamprey spectacle appear to prevent stromal hydration by opposing the osmotic swelling of the stromal cells, thereby limiting the development of interlamellar spaces and preventing the loss of transparency.

Corneal edema and intraocular pressure. II. Clinical results.

The literature pertaining to the relationship between intraocular pressure and edema of the cornea was reviewed in an earlier communication (Ytteborg and Dohlman).1Results of animal experiments were reported in the same paper. The immediate effect of varying the intraocular pressure on epithelial and stromal hydration, as well as on transparency, was observed. Raising or lowering of the intraocular pressure for a few hours had little effect onstromalhydration.Epithelialedema seemed to appear sooner and progress more rapidly as (a) the level of intraocular pressure was increased, and/or (b) stromal swelling increased after endothelial damage. The experiments were short so that the cornea, in most cases, did not have time to reach a new steady state of hydration. In the present paper, the long term effects were studied in human subjects having low, normal, or raised intraocular pressure, assumed to have remained steady for some time. Observations