Ultrastructure of the rabbit corneal stroma after experimental keratomileusis.

Experimental keratomileusis operations were performed on 13 rabbit eyes. The corneas of the rabbits, in particular the stroma, were investigated up to 24 months after the operation by light and electron microscopy. The most striking features of the corneal ultrastructure following experimental keratomileusis were the presence of abnormal collagen fibrils and electron-translucent vacuoles in the interface.

Ultrastructural characteristics of preretinal membranes from human eyes filled with silicone oil.

We removed preretinal membranes previously in contact with an intraocular silicone oil bubble from 22 human eyes for ultrastructural examination. All had membrane-bound intracytoplasmic vacuoles, which most likely had been filled with silicone oil before tissue processing. The vacuolation was more pronounced at the surface of the membrane. In the course of ingestion of silicone oil by preretinal cells, giant vacuoles may form by fusion of smaller vacuoles as these cells lack the ability to digest silicone oil.

Lectin binding pattern in human retinal pigment epithelium.

A comparative lectin histochemical study of human retinal pigment epithelium (RPE) was performed to investigate the lectin binding pattern of normal, reactive and proliferating RPE. Normal RPE with attached sensory retina was found to bind the lectins Con A, WGA, PNA and RCA I. Reactive and proliferating RPE in retinal detachment and in photocoagulation scars revealed the same lectin binding pattern although its cellular topography changed. RPE-macrophages showed an additional reaction with SBA. In periretinal membranes of human PVR the typical lectin binding pattern of Con A, WGA, PNA and RCA I was found in pigmented and in a subpopulation of non-pigmented cells, suggesting that these lectin-positive elements were of RPE-origin. Additionally, single pigmented cells positive for SBA were found indicating macrophage differentiation. Thus lectin histochemistry provides a tool for cytochemical identification of RPE and its morphologic variants by revealing a specific combination of sugar-binding sites.

[The effect of cyclooxygenase inhibitors on the course of hereditary retinal dystrophy in RCS rats]. / Die Wirkung von Cyclooxygenasehemmern auf den Verlauf der hereditären Netzhautdystrophie bei RCS-Ratten.

Daily administration of acetyl salicylic acid (ASA) and ibuprofen leads to an appreciable retardation in the process of retinal degeneration in the RCS rat which is dependent on the dosage given. The photoreceptor cell nuclei and inner segments are relatively well preserved in all regions of the retina. While the outer nuclear layer of 32 day old RCS rats is usually composed of only 3-4 rows, we found 8-9 nuclear rows exhibiting minimal pyknotic change in animals which had been treated with higher doses of ASA. These differences in layer thickness could be confirmed using morphometric analysis. The outer segments show evidence of degenerative change although they are in a clearly better condition than those found in untreated animals and in those animals treated at lower dose. Phagolysosomal structures which are not otherwise apparent in this strain of rat are detected only in the RPE cells of animals treated with higher dose. The animals treated with ibuprofen show essentially the same morphological changes although a corresponding effect in regard to the thickness of the outer nuclear layer was only achieved after a high dose. The determined dose for the optimal preservation (thickness) of the outer nuclear layer lies around 160 mg/kg body weight for acetyl salicylic acid and at 400 mg/kg body weight for ibuprofen.

[Scanning electron microscopy studies of the human lens capsule after capsulorhexis]. / Rasterelektronenmikroskopische Untersuchungen der menschlichen Linsenkapsel nach Kapsulorhexis.

BACKGROUND: In order to better understand the mechanism of tearing of the human lens capsule during circular capsulorhexis, scanning electron microscopic (SEM) examinations were made particularly of the rhexis edge. MATERIALS AND METHODS: Anterior segments from cornea donor eyes, as well as capsular pieces extracted during cataract surgery, were studied after fixation in glutaraldehyde, critical point drying, and sputtering with gold. RESULTS: The edges of the capsulorhexis were found to be very regular even in the area of zonular attachment. Neither the surface of the lens capsule nor the edge of the rhexis itself indicated any morphological influence on the direction of tearing. CONCLUSION: From the results we conclude that the rhexis of the lens capsule is only directed by the forces applied and not by particular morphological structures. To avoid radial tears, a deep anterior chamber, resulting in a relief of the anterior zonular portion seems most important. This minimizes radial forces on the anterior lens capsule, which provides the best condition for a safe rhexis.

[Retinal photocoagulation with a pulsed, frequency-doubled Nd:YAG laser (532 nm)]. / Retinale Photokoagulation mit einem gepulsten, frequenzverdoppelten Nd:YAG-laser (532 nm).

The small difference in wavelength between an argon laser (514 nm) and a frequency-doubled Nd:YAG laser (532 nm), together with the advantage of the solid-state technology, makes the Nd:YAG laser likely to play a major role in retinal photocoagulation in the near future. For technical reasons all frequency-doubled Nd:YAG lasers work in a quasi-continuous mode, emitting a burst of highly repetitive short laser pulses during the exposure time desired. We investigated the side effects due to high peak irradiances of those short laser pulse trains (Crystal Focus Nd:YAG laser, Emerald; pulse duration 1-10 microseconds, repetition rate 13 KHz) in rabbits in comparison with a standard argon laser system (Zeiss, Visulas, Argon II). The energy necessary for blanching the retina was similar in both cases. As opposed to the argon laser system, subretinal bubbles were regularly visible ophthalmoscopically with the Nd:YAG system, when average powers as high as 200 mW were used. The ED50 power for bubble formation is about 2-3 times above the ED50 power for blanching. Thermal calculations show that this bubble formation effect is likely to be related to the peak power of the short pulses. The hemorrhage threshold is similar in both systems. However, light microscopically there is no difference between the two laser systems. Panretinal photocoagulation (300-500 microns, 100-200 ms) in patients with proliferative diabetic retinopathy produced such bubbles about once per 1000 lesions.

The effect of long-term irradiation with fluorescent light on the rat fundus.

In the rat, continuous exposure to so-called "cold light" of 800 lux leads to a nearly complete loss of retinal photoreceptors, which occurs within 4 weeks. This process is followed by vascularization of the retinal pigment epithelium (RPE). The origin, extent, and expansion of such proliferations within 6 months after the onset of irradiation were evaluated by means of light microscopy, transmission electron microscopy, scanning electron microscopy, corrosion cast preparation, india ink injection, and trypsin digest preparation. The vessels within the RPE were found to be fenestrated and to form a network with the vessels of the neurosensory retina. Invasion of Bruch's membrane and formation of chorioretinal anastomoses were not observed. The occurrence of vessels within the RPE coincided with a massive regression of capillaries in the neurosensory retina.

Bubble formation as primary interaction mechanism in retinal laser exposure with 200-ns laser pulses.

BACKGROUND AND OBJECTIVE: Retinal laser photocoagulation is generally performed by laser pulses of a few hundred milliseconds. The tissue interaction mechanism is a pure thermal interaction mechanism. As pulse duration gets shorter, different, non-thermal interaction mechanisms start to appear. The time domain for a change of tissue interaction mechanism seems to be in the ns and micros range. The goal of this study was to characterize the tissue interaction mechanism with 200-ns laser pulses, which approximate the thermal relaxation time of single melanin granules. MATERIALS AND METHODS: The retinas of 19 eyes of 10 rabbits were irradiated by 10 and 500 repetitive laser pulses (wavelength, 532 nm; repetition rate, 500 Hz; pulse duration, 200 ns; per pulse energy, 0-120 microJ; retinal spot size, 100 microm). The effects were evaluated by fluorescein angiography, ophthalmoscopy and by theoretical thermal calculations. Light microscopy and transmission electron microscopy were additionally performed on lesions irradiated by 500 pulses. RESULTS: Single pulse threshold energies for angiographic visibility were 3.5 microJ (10 pulses) and 2.1 microJ (500 pulses), for ophthalmoscopic visibility 9.0 microJ (10 pulses) vs. 8.6 microJ (500 pulses). At energy levels above ophthalmoscopic visibility macroscopically visible bubble formation inside the retina could be observed. This occurred at energy levels of 35 microJ (10 pulses) vs. 17 microJ (500 pulses). Microscopic evaluation of lesions irradiated with 500 pulses and energies at the angiographic threshold showed a damage primarily to the RPE. Additional outer segment damage of the photoreceptors could be found. A gap between damaged RPE cells and the outer segments could be repeatedly found as well as damaged RPE cells, which were detached from intact Bruch's membrane. Temperature calculation shows that temperatures above 100 degrees C may exist around single melanin granules. CONCLUSION: The studies suggest that RPE damage may occur by bubble formation around single melanin granules.

The vascular component of proliferative vitreoretinopathy membranes: an immunohistochemical and ultrastructural study.

BACKGROUND: The basic cellular components of proliferative vitreoretinopathy (PVR) membranes are well studied. Endothelial cells have also been documented. The importance of the vascular element in PVR has received little attention, as vascular components are clinically inapparent. The aim of this study was to obtain a better characterization and quantification of occurrence of the vascular component. METHODS: Serial sections of surgically excised PVR membranes were examined with Lectin histochemistry (25 membranes with ulex europaeus agglutinin I [UEA I] and rhicinus communis agglutinin I [RCA I]), immunohistochemistry for von Willebrand factor (31 membranes), and electron microscopy. RESULTS: Vascular endothelial cells were identified by visualization of UEA I and RCA I binding sites or by marking for von Willebrand factor. A total of 28.6% of the PVR membranes showed a vascular component. Vascular components consisted mostly of capillary-sized vessels; larger vessels were rarely found. Ultrastructurally, most vascular elements were found to be capillaries of the nonfenestrated type. Membranes from eyes that underwent PVR surgery with silicone oil tamponade showed vascular components less frequently (18.5%) than did membranes from eyes without silicone oil (43.8%).

[Retinal lesions after transvitreal use of ultrasound]. / Netzhautläsionen nach transvitrealer Ultraschallapplikation.

Removal of a dislocated natural lens into the vitreous cavity is now performed using vitreous surgery techniques combined with intravitreal phacoemulsification via the pars plana. In contrast to the earlier external surgical approach to luxated lenses, postoperative complications, particularly retinal detachment, are rare. However, retinal damage may occur when ultrasound is used at therapeutic intensities. We therefore examined ultrasound-induced retinal lesions produced in rabbit eyes by treating the retina directly with ultrasound using the type of tip conventionally used for pars plana lensectomy (Fragmatom). Histological evaluation showed that acoustic energy at low intensities led primarily to damage of photoreceptor cell outer and inner segments, which correlated with a discrete pigment reaction visible on ophthalmoscopy. More severe lesions were seen in destruction of the inner retina and resulted in retinal blanching or caused a small retinal break. High energy led to a full-thickness retinal defect with rupture of choroidal vessels and heavy bleeding into the vitreous cavity. Within these lesions the retinal pigment epithelium and Bruch's membrane were also disturbed. Our ophthalmoscopical and histological findings indicate that the mechanism of ultrasound-induced chorio-retinal lesions is not exclusively thermal in nature and differs from other coagulation modalities.

Retinal lesions experimentally produced by intravitreal ultrasound.

Currently, vitreous surgical techniques combined with phakoemulsification are an established procedure for the removal of luxated natural lenses or dislocated lenticular material. This has proven to be a safe procedure, but occasionally retinal lesions have been inadvertently created. We describe the ophthalmoscopic and morphologic features of ultrasonically induced retinal lesions in rabbit eyes using a similar technique. The retina was treated directly using a conventional ultrasound tip for pars plana lensectomy (Fragmatome). Ophthalmoscopically slight lesions corresponded to an area of retinal whitening. More severe lesions showed a destruction of the inner retina and created a retinal break. Extensive effects also involved the choroid and led to a retinal defect with rupture of choroidal vessels and heavy bleeding into the vitreous cavity. Histologic evaluation showed that the acoustic energy primarily led to damage of the outer retina, later involving the inner retina and the choroid as well.