Corneal surface ablation by 193 nm excimer laser and wound healing in rabbits.

The 193 nm argon fluoride excimer laser was used to ablate a 6 mm diameter area of the central rabbit cornea under various conditions of power, beam configuration and exposure time. High repetition rates or prolonged exposures produced charring and prevented rapid epithelial wound closure. Endothelial vacuolization, reduction in density, and displacement of cell material into Descemet's layer resulted in these experiments. A beam of low and uniform power intensity (40 pulses per second, 100 seconds at 23 mJ/cm2) reduced stromal damage, cellular infiltration, and epithelial irregularities including punctate staining and cell exfoliation. Epithelial rehealing occurred within two days. Basal lamina and hemidesmosomes were reformed by one week. Endothelial damage was not detected. Excimer laser ablation may allow removal of superficial dystrophies or scars, followed by rapid healing from normal corneal reparative processes.

Intraocular lens material evaluation by iris abrasion in vitro: a scanning electron microscope study.

Prosthetic posterior chamber intraocular lenses can abrade the iris pigmented epithelium, resulting in pigment dispersion, callus formation, and iris adhesions. To evaluate iris abrasion from intraocular lens materials, we developed an in vitro test and studied three materials: polymethylmethacrylate (PMMA), silicone, and PMMA surface-modified by covalent binding of a hydrophilic polymer (surface-modified PMMA). Each lens was rubbed 1,000 times on the posterior iris of pigmented rabbits using a total force of 0.5 gram. Scanning electron micrographs of the abraded irides were scored using a subjective, nonparametric numerical scale. Polymethyl-methacrylate lenses removed iris epithelium completely, exposing underlying connective tissue. Silicone lenses damaged an area longer and narrower than the PMMA lens and also exposed connective tissue. Surface-modified PMMA lenses caused only slight surface damage, including flattening or breaking of protruding suspensory ligaments. The test conditions used cannot precisely mimic prolonged lens implantation in living subjects. Nevertheless, the results indicate that iris-lens interaction may be reduced in vivo by using a surface-modified PMMA lens. This hypothesis should be further verified by animal implantation studies.

Polyvinylpyrrolidone iodine: corneal toxicology and epithelial healing in a rabbit model.

Polyvinylpyrrolidone-iodine (PVP-I) is a biocompatible broad spectrum antimicrobial that is widely used as an ophthalmic pre-operative prepping agent. In order to determine if PVP-I is safe for treating corneal ulcers and conjunctivitis, we evaluated the ocular toxicity of frequent dosing in a rabbit model. We evaluated ocular irritation in 18 rabbits according to the McDonald-Shadduck method. Concentrations of 0.5% PVP-I or less were practically non-irritating when administered six times per day. We evaluated corneal epithelial wound healing in 16 rabbits with standardized abrasions. Concentrations of 0.33% PVP-I showed epithelial healing comparable to gentamicin treated eyes and untreated controls. Concentrations of 0.5% PVP-I delayed epithelial healing by one day. Because of its broad spectrum and apparently low toxicity, PVP-I may be a useful treatment for corneal ulcers, keratitis and conjunctivitis.

Fibronectin in corneal wound healing.

Fibronectin is an extracellular structural protein with the unique ability to bind to both cells and collagen. It plays a major role in the development of the cornea. The universal appearance of fibronectin within 8 hours of corneal wounding has promoted major interest in its wound healing properties. The early clinical evidence for fibronectin treatment of recurrent corneal erosion and certain forms of keratitis sicca is strong. The current major problem preventing commercial use is the antigenic nature of fibronectin derived from separate species. Human pooled fibronectin has been suggested as a commercial source of this ubiquitously occurring protein.

Growth factor effects on corneal wound healing.

Corneal wound healing and the many factors which may promote or prevent it are of concern to both clinicians and researchers. Ophthalmologists are often confronted with corneal dystrophies and problematic injuries resulting in persistent and recurrent epithelial erosion and with permanent endothelial loss. Scientists have long noted that corneal epithelial and endothelial cell layers are among the most simple and accessible for experimental manipulation and non-invasive study in both animal and human models. Healing epithelium provides a classical model of cell locomotion over a surface, while endothelial cell viability and function is of fundamental importance to maintenance of corneal clarity and vision during the aging process as well as after disease, surgery, or laser treatment. Growth factors can now be produced homogeneously by genetic engineering, increasing the potential of commercial development for purposes such as corneal wound healing. This article reviews some known properties of growth factors whose effects have been at least partially characterized in corneal tissues.

Autofluorescence of the crystalline lens in diabetes.

Lenticular autofluorescence in a diabetic population was quantitated by scanning ocular fluorophotometry. Measurements were made through the lens along the optical axis. The profile of the scan from front to rear was characterized by anterior and posterior juxtacortical peaks and a central plateau. Maximum fluorescence values were compared with those of nondiabetic control subjects. Fluorescence increased linearly with age in both populations, but was significantly higher in the diabetic population of all ages. Posterior to anterior fluorescence peak ratios decreased as a parabolic function of age in both populations, but peak ratios were lower in the diabetic population. We conclude that in vivo measurements of lens autofluorescence correlate with premature lenticular senescence in diabetes.

Autofluorescence and light transmission in the aging crystalline lens.

We quantitated blue-green light transmission and autofluorescence of the human crystalline lens in vivo, using an automated scanning fluorophotometer (Fluorotron) coupled with a lens system designed for high resolution of the ocular anterior segment. Lenses were scanned through the dilated pupil along the optical axis, generating a fluorescence profile consisting of anterior and posterior juxtacortical peaks and a central plateau. Fluorescence increased linearly with increasing age. We calculated percent transmission of excitation (410-500 nm) and emission (510-670 nm) as the ratio of posterior to anterior juxtacortical peaks. Light transmission decreased as a parabolic function of age and was correlated with both fluorescence increase and observed lens brunescence.

Trans-scleral iontophoresis of gentamicin.

Bactericidal levels of gentamicin were obtained in vitreous humor by iontophoresis of antibiotic directly through the sclera. A silicone rubber tube of small diameter filled with gentamicin sulfate formed the electrode. A two milliampere current applied for three minutes to each of four perilimbal sites introduced gentamicin sufficient to maintain therapeutic levels for more than 24 hours. The proportion of drug reaching vitreous versus aqueous humor was dependent on electrode position relative to retina and pars plana. An endogenous antibacterial agent was apparently released during iontophoretic stimulation and interfered with the bacterial growth inhibition assay. Ketoconazole, a water-insoluble antifungal agent, has also been introduced successfully into vitreous humor by anodal iontophoresis after protonation in weak acid.

The effects of topical drugs and preservatives on the tears and corneal epithelium in dry eye.

Medications used in 'Dry Eye' patients are reviewed for their effects on the corneal surface including the overlying tear film. Preservatives are discussed, since they affect the properties of commercial preparations which may be instilled frequently as a substitute for normal tears. The major beneficial effect which a topically applied agent can have on epithelium is to supplement or stabilise the tear film. Thimerosal sometimes triggers a sensitivity reaction, and other mercurial compounds are unstable. Benzalkonium chloride compromises both corneal epithelium and tear film. Some cationic detergents, including chlorhexidine digluconate and polyquat, cause less disruption at prophylactic concentrations. The use of a small drop size is helpful in preventing toxic effects of preservatives. All preservatives should be avoided when unit doses of a sterile tear replacement, such as saline, can be made available. Topical antibiotics should be used only to control known bacterial infections, avoiding high concentrations of bacitracin, gentamicin, and neomycin. Steroids and antibiotic/steroid combinations must be used with great caution, and only when uncontrolled ocular inflammation justifies the risk of possible corneal ulceration.

Corneal penetration and ocular bioavailability of drugs.

This article reviews the objectives which must be considered in producing optimal formulations for topical ophthalmic use. The effects of preservatives, vehicles, and adjunct agents are described. Anatomical considerations which impact bioavailability of the drug at the desired target site are discussed in detail. Model systems that can aid in determining the best formulations for preclinical and clinical testing are also reviewed. The long term objective of this review is the development of formulations for optimal efficacy and safety, considering all the requirements of the patient.

Aqueous flow in human eyes is reduced by forskolin, a potent adenylate cyclase activator.

Forskolin, which lowers intraocular pressure in rabbits, monkeys, and human subjects, was tested for its effect on the rate of aqueous humor flow and on outflow facility. Topical sodium fluorescein was used to measure the rate of aqueous humor flow in forskolin or placebo treated eyes. Tonography was used to determine outflow facility before forskolin administration and at 3 hr after administration of the drug. In eight human subjects who showed reduction of intraocular pressure in response to a single drop of forskolin, flow was reduced by an average of 34%, compared to the contralateral eye during the same period of time. No significant change in outflow facility occurred. The action of forskolin in reducing intraocular pressure was the direct result of a reduction in net aqueous flow.

Preservative alteration of corneal permeability in humans and rabbits.

Isotonic, neutral buffered solutions of benzalkonium chloride or chlorhexidine digluconate were applied topically to one eye of rabbits or human subjects. Contralateral control eyes received phosphate buffered saline as placebo. One-half hour later, the tear film of both eyes was loaded with nonpreserved sodium fluorescein. Anterior chamber fluorescence levels were measured at 1 hr intervals to determine corneal permeability changes attributable to preservative action. In rabbits, corneal permeability increased with rising preservative concentration. Benzalkonium chloride 0.01% increased anterior chamber fluorescence level 1.8 (+/- 0.2 SEM) times over control eyes, while chlorhexidine digluconate 0.01% caused 1.5 (+/- 0.2 SEM) to one ratio of fluorescence in treated/untreated eyes. In human subjects, neither preservative produced significant permeability change at 0.01% concentration. However, benzalkonium chloride 0.02% caused 1.23 (+/- 0.08 SEM) permeability increase. The results support the hypothesis that rabbits are more sensitive to single-dose applications of preservatives than humans.

Electrical potential, resistance, and fluid secretion across isolated ciliary body.

Rabbit ciliary epithelium was mounted with the sclera and lens still attached. The potential measured in physiological bathing medium was 1.29 +/- 0.18 mV initially, but dropped to 0.66 +/- 0.34 mV, then became stable for 2 hr or more. The resistance of the epithelium was measured by comparing total trans-tissue resistance before and after treatment with Triton X-100, ethanol, or distilled water. The calculated resistance with correction for epithelial surface area was found to be 1504 +/- 452 omega, in six experiments. Fluid secretion was measureable against a hydrostatic gradient when identical solutions bathed both sides of the preparation. Ouabain caused opposite responses when applied to either side of the preparation, indicating that it does not readily cross the epithelial barrier. The evidence presented suggests that the ciliary epithelium is a 'tight' rather than a 'leaky' epithelium, and that fluid secretion in the normal ciliary epithelium occurs by active transport rather than by ultrafiltration.

Corneal cytotoxicity of topically applied drugs, vehicles and preservatives.

The toxicity and side effects of drugs and their adjuncts are reviewed according to effect on corneal epithelium, stroma, and endothelium, when topically applied to healthy or wounded surfaces, or introduced into the anterior chamber of the eye. Corneal research techniques are reviewed, including scanning and transmission electron microscopy, electrophysiological techniques, and specular microscope observation. Clinical interpretation of these results is discussed. Reviews of specific toxicologic techniques and of medications are cited. Present research makes it increasingly apparent that not only medications but also their preservatives cannot be considered innocuous. Preservatives may be effective yet less damaging to the ocular integument when used at lower concentrations than those presently available commercially. Complete avoidance of preservatives by the use of special dispensing systems is found increasingly desirable.

Preservative cytotoxic threshold for benzalkonium chloride and chlorhexidine digluconate in cat and rabbit corneas.

Benzalkonium chloride (BAC) and chlorhexidine digluconate (CDG) were applied to rabbit and cat corneal epithelium in clinically used concentrations. Corneas were fixed 1/2 hr later and examined by scanning electron microscopy (SEM). BAC was found to produce a progressive increase in damage at concentrations between 0.001% and 0.01% as determined by SEM. CGD produced less damage than BAC at any concentration. Cats lacrimated copiously and blinked frequently after instillation of drops; rabbits did not. No significant difference was found between the two species, however, in their response to the preservative agents tested. It is presumed that binding of these surface active agents occurs almost immediately and is unaffected by tear film dilution. Studies measuring permeability increase in the human eye after preservative use are required to allow clinical interpretation of the data presented here.

Electrophysiologic and morphologic effects of ophthalmic preparations on rabbit cornea epithelium.

The effects of several components of ophthalmic preparations on isolated rabbit cornea were studied by continuous electrophysiologic monitoring followed by fixation for scanning electron microscopy (SEM). Benzalkonium chloride (0.001 percent), thimerosal (0.0004 percent), and amphotericin B (0.0025 percent) all briefly increased ion transport, then greatly decreased epithelial resistance. Severe disruption of surface cell layers occurred simultaneously with resistance decrease. Silver nitrate (0.00017 percent) stimulated transport with less accompanying morphologic damage. Tetracaine (0.05 percent) disrupted epithelial function and caused exfoliation of several cell layers. Chlorobutanol (0.1 percent) produced a nearly complete loss of the squamous cell layer. Chloramphenicol, epinephrine, and pilocarpine produced minor changes in structure and electrophysiology at full clinical concentration. It was concluded that low concentrations of preservatives in ophthalmic preparations disrupt the barrier and transport properties of the corneal epithelium.

The normal and abnormal human corneal epithelial surface: a scanning electron microscope study.

The surface morphology of 108 corneal buttons obtained at keratoplasty showed specific patterns for each disease process. The surface over a traumatically scarred cornea was identical to that of undamaged sites, showing microvilli and microplicae in various numbers and combinations. Keratoconus specimens showed many dark cells, frequently noted to have surface blebs 0.25 to 3 micrometer in size over the entire cone in the nipple type, and in a broad basal band inside the cone or over the entire button in the sagging-cone type. Some blebs contained cytoplasm and 250 A glycogen-like granules. In larger, dark cells, holes were found in the blebs and the plasma membrane was degenerated. Corneal epithelial edema was manifested by a large irregular surface caused by the anterior bulge of edematous cells, many attached by only a small area, and variable-sized depressions, often the size of epithelial cells. More than a year after stromal scarring from herpetic keratitis, many epithelial cells lay loosely on the surface whereas other epithelial cells were edematous and partially detached from the surface cell sheet. Localized heaping of rounded epithelial and inflammatory cells persisted in some areas.