Sample size for estimating the quantiles of endothelial cell-area distribution

The estimation of corneal endothelium mean cell area (and, hence, mean cell density) is an important problem in clinical ophthalmology. Mitotic division of these cells is not known to occur, and cell deaths are followed by the enlargement of adjacent cells. As a consequence, cell-area distributions change drastically as functions of age and disease. Changes in cell-area distributions, in particular multimodality and skewness due to aging, are observed, and give rise to some difficult sampling problems. In this paper, sample quantiles are investigated as an alternative to the use of the sample mean. Asymptotic approximations are provided for the sample sizes required to estimate population quantiles with a desired precision. Asymptotic sample sizes are then compared with those obtained from tolerance limits. Empirical sample quantiles that can be used as benchmarks to compare corneas of normal individuals against corneas with unknown cell-area distributions are also presented. Aspects that merit further investigation are noted.

Oval lipid corneal opacities in beagles: VI. Quantitation of excess stromal cholesterol and phospholipid

We identified and quantitated the following lipids in corneas of normolipemic beagles that had oval corneal opacities: total lipids, cholesterol esters, free cholesterol, total phospholipids, phosphatidyl choline, phosphatidyl ethanolamine, sphingomyelin and total triacylglycerols. All except the total triacylglycerols were significantly higher in the 20 affected corneas than in ten normal age and sex-matched controls (P less than 0.005 for each). Few long-chain unsaturated fatty acids were detected. This lipid analysis and the previously reported natural history, ultrastructure, and histochemistry indicate that these opacities result from a primary disorder of corneal lipid metabolism that closely resembles central crystalline corneal dystrophy in man.

Clinical specular microscopy of posterior polymorphous endothelial dystrophy

By means of clinical specular microscopy in 27 patients with posterior polymorphous corneal endothelial dystrophy diagnosed by slit lamp examination and 26 relatives with normal findings by slit-lamp and specular microscopic examinations, we identified two clinical and specular microscopic patterns of posterior polymorphous dystrophy. (1) The localized vesicular or band pattern consisted of clear corneas with localized lesions resembling craters that were lined by abnormal cells and surrounded by enlarged, pleomorphic cells. (2) The geographic pattern was associated with Descemet's membrane haze and deep stromal haze, occasional peripheral iridocorneal adhesions, and glaucoma. By specular microscopy, this type had large, well circumscribed areas of abnormal pleomorphic cells with indistinct borders and increased reflective highlights. These areas were surrounded by normal-sized and smaller polygonal cells. The former cells appear to represent epithelial cells and the latter endothelial cells.

Rationale for and design of the National Eye Institute Prospective Evaluation of Radial Keratotomy (PERK) Study

A five-year, multicenter, collaborative clinical trial of radial keratotomy for myopia is being carried out at nine clinical centers. The study, funded by the National Eye Institute, is recruiting approximately 500 patients aged 21 years or older with -2.00 to -8.00 diopters of physiologic myopia who also meet other clearly defined eligibility criteria. Surgeons use a diamond-bladed micrometer knife to make eight radial incisions in the anterior cornea. The diameter of the central clear zone is determined by the amount of myopia and the depth of the incisions by intraoperative ultrasonic pachymetry. Investigators other than the surgeon gather all pre- and post-operative data. Examinations include measurement of visual acuity with standardized charts, verification of refractions by a second observer, measurement of corneal curvature by keratometry and photokeratoscopy, quantitation of glare with a glare tester, and measurement of the size of individual endothelial cells from specular photomicrographs with a computerized digitizer. A formal psychometric questionnaire is used to evaluate the patients' subjective response. Physician monitors insure adherence to the protocol during site visits and members of the Data and Safety Monitoring Board oversee the progress of the study. Biostatisticians at the Coordinating Center insure complete data collection, process the data, and assist in the interpretation of results.

Results of penetrating keratoplasty in 123 eyes with pseudophakic or aphakic corneal edema

The clinical records of 123 eyes that received a penetrating keratoplasty for aphakic or pseudophakic corneal edema with a mean follow-up time of 18 months were studied retrospectively. Thirty-five eyes from which an intraocular lens (IOL--85% iris fixated) was removed at the time of keratoplasty were compared with 88 eyes that never had an IOL. The mean interval between initial cataract surgery and penetrating keratoplasty was 28 months for the pseudophakic group from which the IOL was removed and 67 months for the aphakic group without IOL. In both groups, 97% had an anterior vitrectomy. Corneal grafts remained clear in 91% of the previously pseudophakic eyes and in 81% of the aphakic eyes (P less than or equal to 0.05). The final corrected visual acuity was worse in the pseudophakic eyes with IOL removed (16% saw 6/6 to 6/12 while 22% saw worse than 6/60) than in the aphakic eyes (34% saw 6/6 to 6/12 while 7% saw worse than 6/60). Postoperative maculopathy occurred in 43% of eyes from which IOLs were removed and in 22% of aphakic eyes.

The corneal endothelium. Normal and pathologic structure and function

A summary of normal and abnormal endothelial structure and function is presented. Endothelium originates from neural crest and it elaborates a banded basement membrane in utero. It is involved in mesenchymal dysgenesis of the anterior segment, like the central defect of Peters' anomaly. Cytoplasmic organelles include mitochondria that provide energy for the metabolic pump, rough endoplasmic reticulum that participate in secretion of extracellular matrix, and a terminal web that may participate in cell migration. The endothelium's main function is to control corneal hydration and nutrition with a leaky barrier formed by the apical gap and macula occludens junctions that keep some water out of the stroma but allow nutrients to pass, and with an ATPase-dependent metabolic pump that is located in the lateral plasma membranes. Endothelial wound healing involves flattening and enlargement of cells to maintain an intact monolayer as well as production of abnormal collagenous material posterior to Descemet's membrane. HLA antigens located in the plasma membrane may participate in corneal endothelial graft rejection. Clinical assessment of the endothelium involves three modalities: specular microscopy to study endothelial morphology, fluorophotometry to measure barrier function, and pachymetry to measure corneal thickness.

Use of Healon in anterior segment trauma

Healon can assist in the repair of penetrating segment injuries by maintaining a deep anterior chamber that allows easier suturing of the wound and easier repositioning of the iris, by protecting the corneal endothelium during surgery, and by separating peripheral iridocorneal adhesions. We describe this technique in repair of an eye with a limbus-to-limbus corneal laceration.

Oval lipid corneal opacities in beagles

Naturally occurring, oval opacities at the junction of the middle and inferior thirds of the cornea were found in an experimental beagle dog colony and were examined by light and electron microscopy. Abnormalities were limited to the anterior stroma and consisted of extracellular rhomboid crystalline spaces. Spherical or elliptical vacuolar spaces, sometimes containing electron-dense amorphous debris, distorted stromal lamellae. Stromal keratocytes in affected areas revealed a range of changes from hyperplasia of organelles to degeneration and necrosis.

Oval lipid corneal opacities in beagles: ultrastructure of normal beagle cornea

The ultrastructure of normal Beagle cornea was similar to the human cornea, except the Beagle lacked an anterior limiting lamina (Bowman's layer) and banding in the anterior portion of the posterior limiting lamina (Descemet's membrane). The epithelium was an average of 19 cells thick with 3 basic types: squamous, polygonal, and basal. The subepithelial stroma with a mean thickness of 9 microns was hypocellular, and the bundles of collagen were randomly oriented. The majority of the stroma consisted of closely packed, parallel collagen lamellae interspersed with keratocytes and occasional nerve fibers. The posterior limiting lamina was homogeneous and thick with increasing age. The endothelial cells formed a monolayer and had apical tight junctions.

Posterior collagenous layer of the cornea. Ultrastructural classification of abnormal collagenous tissue posterior to Descemet's membrane in 30 cases

We studied the transmission electron microscopy of the abnormal collagenous tissue posterior to Descemet's membrane in 30 human corneas that represented 19 different diseases. We named this abnormal tissue the "posterior collagenous layer" of the cornea and classified it into three morphologic types: (1) banded posterior collagenous layer, characterized by 110-nm wide-spacing material, (2) fibrillar posterior collagenous layer, characterized by loose meshwork of collagenous fibrils 20 to 30 nm in diameter, and (3) fibrocellular posterior collagenous layer, characterized by fibroblasts in connective tissue. The published ultrastructural studies of 88 corneas that represented 13 different clinical disorders fit this classification well. The posterior collagenous layers often formed multiple lamellae, but no pattern was invariably associated with a specific clinical disorder. We think these multiple layers represent successive periods of deposition of collagen by a stressed corneal endothelium and provide a lifelong record of senescent and pathologic events.

Isolation and characterization of fibronectin from bovine aqueous humor

Aqueous humor collected from freshly enucleated bovine eyes was passed over a gelatin affinity column, and bound material was eluted with 4M urea. When fractionated by sodium dodecyl sulfate electrophoresis, the eluted material co-migrated with human plasma fibronectin and protein standards of approximately 220,000 daltons. Amino acid analyses demonstrated a composition similar to that of human plasma fibronectin and fibronectin synthesized by cultured cells. The concentration of fibronectin in aqueous humor is approximately 100 times less than in human plasma.

Selecting patients for penetrating keratoplasty

Proper selection of patients for successful penetrating keratoplasty depends on four important factors: (1) psychosocial factors, including visual need, the conditions under which the patient lives, and the patient's ability to nurture this tissue over many years; (2) the general condition of the eye, including the health of the ocular surface, intraocular pressure, motility, anterior segment anatomy, and vitreoretinal disease; (3) the structure of the host cornea--in terms of its thickness, vascularization, epithelial and stromal integrity, and sensation--as well as the corneal condition for which the operation is being done; and (4) the skills of the ophthalmologist managing the graft, particularly those involved in preoperative patient selection and postoperative graft management, as well as his technical surgical skills.

Oval corneal opacities in beagles. III. Histochemical demonstration of stromal lipids without hyperlipidemia

We found oral stromal avascular corneal opacities in 128 eyes of 75 beagles from 497 studied. There were three morphologic types that progressed in severity with time; nebular, racetrack, and white arc. Histochemical study of the earliest morphologic type (nebular) revealed neutral fats, cholesterol, phospholipids, and sometimes fatty acids both intracellularly and extracellularly. We found no elevation of serum cholesterol or triglycerides except in dogs with the most advanced morphologic type (white arc) and no alteration in thyrometabolic function. We think that oval stromal opacities in beagle corneas are a primary disorder of corneal lipid metabolism closely resembling the central crystalline dystrophy of Schnyder and may be an animal model for this human disease.

Ultrastructure and successful keratoplasty of sclerocornea in Mietens' syndrome

A 5-year-old boy with Mietens' syndrome had bilateral microsclerocornea, hypoplastic nose, bilaterally absent radii, elbow flexion contractures, absent left fibula, growth retardation, and normal intelligence. Both corneas measured 9.25 mm in diameter and showed diffuse anterior stromal opacification with focal nebular densities and extensive superficial vascularization. Penetrating keratoplasty in one eye remained clear and compact two years postoperatively and visual acuity improved. Histopathologically, vascularized collagenous tissue occupied the anterior one fourth of the corneal stroma and contained bundles of collagen fibrils 75 to 90 nm in diameter. Descemet's membrane showed abnormal anterior lamination.

Four methods of measuring human corneal endothelial cells from specular photomicrographs

We measured central corneal endothelial cell density and area from contact specular photomicrographs of ten normal and ten abnormal corneas, comparing the precision, cost, and speed of four methods: a rectangle, planimeter, digitizer, and cell sizer. The rectangle, planimeter, and digitizer gave results that differed less than 10% from each other; therefore, the three methods can be used interchangeably for clinical purposes. There are statistically significant differences among the three techniques that may be important in basic research. The cell sizer gave a rapid, less precise estimate of mean cell area and cell density. The planimeter and digitizer measured individual endothelial cell size, and the latter entered data directly into a computer that printed both a copy of the endothelial mosaic and a histogram of cell size frequency, and computed cell density and mean cell area. We make the following recommendations: Count cells in a rectangle used for routine clinical measurement. use a cell sizer for rough estimation, as in an eyebank setting. Use a computerized digitizer to study individual endothelial cell size.

Crystalline corneal opacities in the Siberian Husky

Bilaterally symmetric opacities were detected in the corneal stroma of 78 (14%) of 560 Siberian Huskies, aged 7 months to 12 years, examined in ophthalmology screening clinics. The opacities were round or horizontally oval and consisted of a diffuse gray homogeneous haze in the anterior stroma or an array of fine polychromatic crystals in the posterior stroma, or both. The corneas were not inflamed. The frequency of occurrence and density of the opacities increased with age. Several affected dogs were closely related, but a specific inheritance pattern could not be established. Light and electron microscopy disclosed clusters of extracellular, thin, needle-shaped, crystalline clefts. Histochemical stains on frozen sections identified neutral fats, phospholipids, and cholesterol as components of the crystals.

Lepromatous iridocyclitis diagnosed by anterior chamber paracentesis

A 34-year-old man with a three-month history of intraocular inflammation after ocular trauma with a fir branch, had an acute unilateral fulminant iridocyclitis. The iris had a thick, gray, cheesy membrane composed of nodular lepromata. The patient denied a history of Hansen's disease, despite the dermatologic and facial features that suggested the diagnosis. Anterior chamber paracentesis and scleral nodule biopsy demonstrated Mycobacterium lepra. The iridocyclitis resolved after treatment with dapsone, corticosteroids, and rifampin.

Primary radiation therapy for meibomian gland carcinoma

Three patients with Meibomian gland carcinoma were treated initially with irradiation. This therapy was palliative in one case that eventually required orbital exenteration because of recurrent disease. A second patient was clinically tumor-free ten months after treatment when he died of a myocardial infarction. The third patient has shown no evidence of recurrent disease in the 25 months since treatment.