Increased susceptibility to infection in experimental xerophthalmia.

Vitamin A-deficient rabbits were used to evaluate the role of secondary bacterial infection in the development of keratomalacia and to describe the resultant clinical and morphologic alterations. The conjunctival sacs of vitamin A-deficient rabbits at different stages of corneal involvement were inoculated with Pseudomonas aeruginosa topically. Approximately two weeks after inoculation, corneal ulceration with stromal melting developed in one of three eyes with severe punctate keratitis and in four of seven eyes with xerosis. Ulceration did not develop in any of the eight eyes with early epithelial graying or mild punctate keratitis. Inflammatory cells (primarily polymorphonuclear leukocytes) infiltrated the anterior corneal stroma of infected corneas. Liquefaction of collagen was observed in association with bacteria alone, as well as in association with polymorphonuclear leukocytes. No signs of infection were observed after conjunctival inoculation of Pseudomonas in the eyes of nine control rabbits.

Osmotic tolerance of rabbit and human corneal endothelium.

Rabbit and human corneas were mounted in a specular microscope and perfused with a balanced salt solution of varying osmolality (200 to 500 mOsm). Measurements of corneal thickness were made throughout the perfusion period, and at selected times the corneas were fixed and prepared for scanning and transmission electron microscopy. A hypo-osmotic perfusion medium caused an increase in corneal thickness; by comparison, a hyperosmotic perfusion medium decreased corneal thickness in both rabbit and human corneas. Despite the marked changes in corneal thickness and the water movement that occurred across the endothelium, the cellular ultrastructure remained intact. In reversal studies (return to 300-mOsm perfusion medium), corneal thickness returned to control values with no marked changes in endothelial cell structure. These data indicate that the corneal endotheium can tolerate a wide range of solution osmolalities (200 to 400 mOsm) without marked endotheial cell breakdown if the essential ions are present.

Effects of timolol on bovine corneal endothelial cultures.

The chronic use of timolol (Timoptic) maleate to control glaucoma may produce cytotoxic complications in the cornea. We have therefore compared the relative toxic effects of the commercial ophthalmic preparation with that of the pure compound. Commercial vehicle, either with or without 16 mM timolol maleate, killed cultures within the first five minutes of exposure. Pure timolol maleate, however, caused rapid but reversible cellular contractions, and cells remained viable in it for over 24 hours. Dilution with culture medium reduced both the cytotoxicity and the speed of the contractions. Incubation in 1:100 dilutions of vehicle or commercial drug preparations or in 0.16 mM pure timolol maleate did not alter cellular morphology. The results indicate that while undiluted vehicle is toxic, timolol maleate is not.

Experimental Pseudomonas keratitis in the rabbit: bacteriologic, clinical, and microscopic observations.

Uniformly severe corneal infections were produced in rabbits by intracorneal injection of a few viable Pseudomonas aeruginosa. The bacteria multiplied rapidly, and within 24 hr, about 10 million organisms were present. The numbers remained stable thereafter. Polymorphonuclear leukocytes (PMNs) began to infiltrate peripheral stroma 24 hr after inoculation. By 32 hr, ring-shaped dense accumulations of PMNs were apparent in the anterior stroma with moderate stromal edema. By 48 hr, the anterior one third of central stroma was severely involved with abscess formation and loss of epithelium, and PMNs had invaded full corneal thickness. The area of liquefactive necrosis eventually involved the entire cornea from limbus to limbus, and collagen staining was lost. Transmission electron microscopy revealed the accumulation of small electron-dense particles in association with collagen fibrils and degranulating PMNs.

Topical retinoic acid in the treatment of experimental xerophthalmia in the rabbit.

The purpose of this study was to evaluate the efficacy of topical retinoic acid (tretinoin) in the treatment of various stages of xerophthalmia caused by vitamin A deficiency. At the beginning of treatment, rabbits with similar disease in both eyes were given a normal diet and therapeutic doses (25,000 IU) of systemic vitamin A. One eye of each was treated with 0.1% retinoic acid in sesame oil one time per day. The other eye received sesame oil alone. Eyes with moderate disease (advanced keratinization and plaque formation) responded more quickly to topical retinoic acid (clear within three days) than did eyes treated with sesame oil (clear within seven days). No difference in time of response could be observed in corneas with milder or more severe disease (stromal infiltration and corneal vascularization).

Corneal endothelial cell shrinkage after critical point drying.

Corneal endothelial cell densities of 30 rabbit eyes were determined in vivo by specular microscopy and after critical point drying by scanning electron microscopy. The range of shrinkage in area of the cells after critical point drying was 3.6% to 49.1% with a mean of 29.7%. This respresents a linear shrinkage of 1.8% to 28.7% with a mean of 16.4%. Comparisons of endothelial cell size or density between different critical point dried speciments therefore cannot be made without knowing how much shrinkage occurred in each individual specimen.

Xerophthalmia in vitamin A-deficient rabbits. Clinical and ultrastructural alterations in the cornea.

Xerophthalmia developed in the eyes of rabbits maintained on a vitamin A-deficient diet for 4 to 6 months. The earliest clinical change, a lusterless graying of the central corneal epithelium, was noted after 16 to 18 weeks on the diet. Multiple small punctate epithelial erosions appeared in the interpalpebral fissure zone within 7 to 10 days after the lusterless graying became evident. The erosions gradually became confluent, and a striking dry, glazed, peau d'orange appearance was noted. Polycystic microbullae appeared in the epithelium in some eyes. Thick keratinized epithelial plaques developed in all eyes 1 to 2 weeks after the appearance of severe peau d'orange. Electron microscopy of corneas with lusterless graying of the epithelium revealed swelling of the most superficial epithelial cells with flattened and shorter microvillous projections. In corneas with punctate epithelial erosions and keratinized plaques, microvilli were absent or decreased in number on superficial cells, and multilayered, keratinized epithelial cells were present on the surface of the cornea. The stroma appeared essentially normal with minimal edema at all stages when examined by electron microscopy. Intercellular edema was present in the endothelium in early- and late-stage xerotic corneas but could not be detected clinically. No significant clinical or microscopic alterations were seen in the corneas of control rabbits on normal diet or in rabbits on the vitamin A-deficient diet supplemented with vitamin A. The alterations seen in the corneas of vitamin A-deficient rabbits are similar to those which have been described in vitamin A-deficient humans. Rabbit therefore appears to be a good model for further studies of xerophthalmia.

The effects of aging and inflammation on corneal endothelial wound healing in rabbits.

After transcorneal freezing, the rates and patterns of corneal endothelial wound healing were compared in mature and young rabbits by autoradiographic analysis of 3H-thymidine incorporation and scanning electron microscopy. Healing was slower and less extensive in mature corneas than in young ones. Regardless of animal age, however, healing occurred by cell division and the migration of newly divided cells onto the wound surface. Spontaneously occurring severe inflammation appeared to reduce the ability of corneal endothelial cells to replicate their DNA.

Cytotoxicity of pivalylphenylephrine and pivalic acid to corneal endothelium.

We examined the effects of topically applied pivalylphenylephrine (PPE) and pivalic acid (PA) on the corneal endothelium of rabbits and the direct effects of PPE and PA on monolayer cultures of bovine corneal endothelium. The PPE-treated corneas without epithelium significantly increased in thickness, whereas no change in thickness was observed in corneas with epithelium intact. The PA did not alter the thickness of corneas with or without epithelium. Although 0.001% PE had no noticeable effect in two hours, 0.01% PPE caused breakdown of intercellular junctions in cultured cells in five minutes. Higher concentrations of PPE caused the cells to detach from the culture dishes within 30 minutes of treatment. Only 1% PA caused cell elongation and loss of intercellular contact after 60 to 90 minutes of exposure; lower concentrations did not effect cultured cells.

The effects of increased hydrostatic pressure upon normal and regenerated rabbit corneal endothelium.

The effect of increased hydrostatic pressure upon the ability of normal and regenerated endothelium to deturgesce preswollen, de-epithelialized rabbit corneas was studied. Stromal deturgescence occurred as a biphasic response when hydrostatic pressure at the endothelial surface was increased above baseline values. Initially there was a rapid phase of stromal thinning which was dependent upon hydrostatic pressure nad endothelial function. This was followed by a slower phase of corneal thinning which was independent of hydrostatic pressure at the endothelial surface for pressures between 15 and 50 mm Hg. The slow phase of thinning represents the steady-state ability of the endothelium to deturgesce the stroma. Regenerated rabbit endothelium functioned similarly to normal endothelium in deturgescing the stroma. In addition, short-term increases in hydrostatic pressure at the endothelial surface did not produce ultrastructural changes in normal or regenerated corneal endothelial cells.

Corneal endothelial autoradiography with the scanning electron microscope.

A new technique for performing corneal endothelial cell autoradiography with the scanning electron microscope is presented. The scanning electron microscope is able to detect silver grains deposited over tritiated thymidine--labeled nuclei of regenerating corneal endothelium.

The effect of phenylephrine on the cornea.

Rabbit corneas were treated with three drops of phenylephrine hydrochloride with the epithelium intact or denuded. Corneal thickness was measured before and after drug treatment, and at various times after treatment the corneas were fixed for scanning and transmission electron microscopic observation. The results of this study show that phenylephrine caused a dramatic increase in corneal thickness (drug-induced edema) and cellular vacuolation within the keratocytes and endothelial cells in the corneas without the epithelium. Corneal thickness did not change and the ultrastructural changes were minimal following drug application in those corneas with the epithelium intact. Results of this study also suggest that phenylephrine has a cytotoxic effect on the corneal endothelium and keratocytes when used in corneas where the epithelium has been removed. In coreas with intact epithelium, the damage was less severe and limited to the epithelium.

The pH tolerance of rabbit and human corneal endothelium.

The endotheliums of rabbit corneas were perfused in an in vitro perfusion specular microscope up to 3 hr with solutions varying in pH from 3.5 to 10.0. Corneal thickness was monitored throughout the experiment, and at appropriate times the corneas were prepared for SEM and TEM. Analysis of the corneal thickness data and interpretation of the electron micrographs reveals that outside of the pH range of 6.5 to 8.5, structural and functional alterations occur. Direct cellular damage, as well as disruption of junctional complexes, lead to a breakdown in the barrier function of the corneal endothelium. The extent of this breakdown is dependent upon both the magnitude of the pH change and the exposure time. Further experiments on banked human eyes support this finding.

Binkhorst Lecture. (Part 2). Experimental cataract surgery--electron microscopy.

Scanning and transmission electron microscopic examinations were performed on cat corneas after the steps usually taken to perform cataract extraction were completed. The endothelium did not replicate, but showed nuclear division or segmentation occasionally. In each case, the anterior segment was free from any evidence of inflammation one year after surgical treatment--unless a pseudophakos had been inserted: in some of these cases, significant numbers of white cells were present, a condition that may contribute to anterior segment complications from implants.

Corneal endothelial cell function after storage in MK medium and hydrocortisone.

We stored rabbit corneas either in MK medium or MK medium plus hydrocortisone 10(-6)M for 7 days. The corneas stored in MK medium with hydrocortisone had better temperature reversal characteristics than those stored in MK medium alone.

Effect of phenylephrine on normal and regenerated endothelial cells in cat cornea.

Topical commercial phenylephrine HCl (Neo-Synephrine 10%) has been shown to cause an increase in corneal thickness and reversible vacuolization of corneal endothelial cells in rabbits. Using an in vivo model of regenerated corneal endothelial cells in the cat, we compared the cytotoxicity of phenylephrine-HCl 10% to regenerated and to normal, nonregenerated cells. Following removal of the epithelium, topical application of the drug causes the appearance of anterior and posterior bands of stromal edema and reversible vacuolization in both normal and regenerated endothelial cells. Phenylephrine was not more damaging to the regenerated cells. Polymorphonuclear leukocytes infiltrated between the regenerating cells 24 hr after treatment but did not appear to destroy them. Phenylephrine may therefore be implicated as a causative factor of corneal edema and postoperative inflammation.

Effects of transcorneal freezing on protein content of aqueous humor and intraocular temperature in rabbit and cat.

The effects of transcorneal freezing on protein content of aqueous humor and intraocular temperature at the posterior surface of the cornea, the angle, the iris, and the ciliary processes were determined in rabbits and cats. Normal aqueous protein concentration was 40 +/- 2 mg/dl in rabbits and 43 +/- 4 mg/dl in cats. In rabbits, total aqueous protein content reached its highest level (2790 +/- 302 mg/dl) for 3 hr after transcorneal freezing, decreased by 50% after 4 hr, and was not significantly different than normal at 7 days. In cats, total aqueous protein content also reached its highest level (1610 +/- 290 mg/dl) 3 hr after corneal freezing. Fluctuations occurred thereafter, but protein content was not significantly different from normal after 7 days. The temperature at the corneal endothelium always decreased to below 0 degrees C with a 10 to 25 sec application of the cryoprobe to the cornea in rabbit and cat. Intraocular temperature did not decrease below 24 degrees C at the angle or ciliary processes during application of the cryoprobe for up to 25 sec, whereas the temperature at the pupillary margin of the iris sometimes decreased to near 0 degrees C with a 15 to 25 sec application.