This report summarizes the authors' 3-year experience with excimer laser photorefractive keratectomy (PRK) on 240 eyes of 161 patients. With constant laser emission parameters, nitrogen (N2) flow across the cornea was used on 79 eyes while 161 eyes had no N2 flow. 74 eyes were operated on without fixation with a suction ring. Postoperative pain management included patching and oral analgesics in 77 eyes and the use of topical Diclofenac or Ketorolac, and a therapeutic soft contact lens in 163 eyes. Follow up ranged from 1 month (206 eyes) to 36 months (10 eyes).

Laser ablation of human and rabbit corneas has been studied at 211 and 263 nm with 25 ps pulses. The pulses are obtained from the 4th and 5th harmonics of the output of a Nd:YLF laser system. This system provides laser pulses with maximum pulse energy of 25 (mu) J at 211 nm and 31 (mu) J at 263 nm simultaneously on a cornea at a repetition rate of 1 kHz. The linear incisions and morphology of the collateral tissue damage have been examined by electron microscopy. The dependence of the etch rate on laser fluence has been measured at both of wavelengths. The ablation threshold for human cornea is determined to be 3.0+/- 0.6mJ/cm² at 211 nm while the thresholds for rabbit cornea are 2.3+/- 0.5mJ/cm² at 211 nm and 8.0+/- 2mJ/cm² at 263 nm. We compare these results with the existing data on the corneal ablation by nanosecond UV pulses and discuss the deficiency of the photochemical model. A model of plasma ablation assisted by chromophore absorption has been proposed to interpret the experimental results.

In order to investigate the effects of 193-nm excimer laser radiation on corneal wound healing five bovine corneas were treated with subablative and ablative 193-nm excimer laser radiation, by simple light exposure and mechanical ablation of the corneal tissue or not treated. The corneas were metabolically labelled with a [³⁵S]-labeling reagent. The proteoglycans were then extracted and radioactivity was measured with a liquid scintillation spectrometer. Corneas after ablative 193-nm excimer laser radiation and after mechanical ablation had less radioactivity incorporated but had also less corneal tissue.

We update the continued progress of laser thermokeratoplasty (LTK) clinical trials being conducted in the U.S. for the treatment of hyperopia and hyperopic astigmatism. Data from the Phase II hyperopia investigations on 25 patients and from Phase I astigmatism trials on 30 patients is reviewed. From the hyperopia Phase IIa study, the near uncorrected visual acuity of 13 patients for whom complete follow-up results are available at 1 year shows that all 13 patients gained 2 or more lines of visual acuity (Ave. gain 3.5 lines), which indicates a significant improvement in near vision. A survey given to these hyperopia patients finds 8% could read without glasses preoperatively versus 58.3% at 1 year post-operatively. The preoperative uncorrected visual acuity of those patients treated for astigmatism in the Phase I trial showed 2 out of 30 patients or 6.7% seeing better than 20/40 versus 10 out of 27 patients or 37% at one year post-op. The one year data in both studies indicates that after an initial period of partial regression of effect, the residual correction remains relatively stable between 6 months and 1 year. Continued follow-up will be carried out to see if the corrections remain stable beyond 1 year.

We investigate the laser refractive scleroplasty (LRS) as a potential minimal-invasive method for correcting post-operative astigmatism. The scleral shrinkage near limbus was induced on 6 cadaver eyes using a 200 micrometers fiber optic probe coupled to a pulsed Ho:YAG laser. The diameter of the treatment spot was 0.8 mm. The output energy measured at tip was 60.2+/- 0.6 mJ. The treatments consisted of multiple sector patterns placed along the major axis of astigmatism parallel to the limbus, and round patterns placed along the limbus. Three treatment spots were applied on each side of the sector. The separation among sectors and limbus is 1 mm. Keratometry and topography of the cornea were measured after each sector or round pattern treatment. Effect of 5 and 10 pulses at each treatment spot were compared. Histology was performed to evaluate laser tissue damage. The major axis of astigmatism was shifted 90 degrees after the sector pattern treatment and amount of dioptric change increased when adding a new treatment or using more treatment pulses. However, the spherical equivalent of the eyes was essentially unchanged. The keratometry of the corneas remained the same after the round pattern treatment. Laser refractive scleroplasty may be applied for the correction of post-operative astigmatism.

We used the ArF excimer laser to determine the feasibility of performing prismatic photoablations in model eyes (plastic spheres simulating the eye), and in rabbit corneas. This would correct diplopia and small angles of deviation, and result in minimal refractive alterations. We modified excimer laser delivery system that achieved the desired corneal contour of prismatic ablations. 193-nm argon fluoride laser was used at fluence of 160 mJ/cm² and ablation rate 5 Hz. 5.0-mm diameter, 40 um corneal epithelial ablation were followed by 5.0- mm diameter, prismatic photokeratectomy (PPK). We were able to achieve prismatic photoablation of PMMA blocks and lenses. No other refractive changes accompanied the prismatic photoablation of PMMA blocks and lenses. No other refractive changes accompanied the prismatic effect. In rabbits re-epithelialization of the 5-mm ablations was complete by day 3, and corneal haze was not observed by gross examination. Epithelial hyperplasia and subepithelial scarring were noted at the deep edges. PPK holds important therapeutic potential for fine-tuning results of conventional strabismus surgery, and for patients with stable diplopia following nerve palsy and ocular surgery.

Dye-enhanced diode laser photocoagulation, using the dye indocyanine green (ICG), has shown some potential in the treatment of choroidal neovascularizations (CNV). A diode laser system was developed and optimized to emit at the absorption maximum of ICG. In a clinical study at two retinal centers, more than 70 patients, the majority of which had age-related macular degeneration, were treated. Eighteen cases with ill-defined subfoveal CNV were followed an average of 11 months after laser treatment. The results show success in resolving the CNV with an average long-term preservation of visual function equal to or superior to data provided by the Macular Photocoagulation Study for confluent burns of low intensity applied to the CNV. Details of the technique and discussion of the controversies inherent in such a treatment strategy will be presented.

Infrared fundus angiography, using the fluorescent dye indocyanine green (ICG), has shown great potential in delineating choroidal neovascularization (CNV) otherwise not detectable. A digital retinal imaging system containing a diode laser for illumination has been developed and optimized to perform high sensitivity ICG angiography. The system requires less power and generates less pseudo-fluorescence background than nonlaser devices. During clinical evaluation at three retinal centers more than 200 patients, the majority of which had age-related macular degeneration, were analyzed. Laser based ICG angiography was successful in outlining many of the ill-defined or obscure CNV as defined by fluorescein angiography. The procedure was not as successful with classic CNV. ICG angiograms were used to prepare and guide laser treatment.

Over the last two years, ophthalmologists have begun to use indocyanine green angiography as a supplement to fluorescein angiography. Unlike fluorescein, indocyanine green absorbs near infrared and emits slightly longer infrared light. Therefore, indocyanine green angiography images structures deeper in the retina and through blood, pigment, and turbid serous fluid that accumulates in a number of diseases. In addition, indocyanine green shows very different properties of dye leakage than does fluorescein and this can be used to identify certain abnormal blood vessels that grow beneath the retinas of patients with macular degeneration. Finally, indocyanine green fluoresces only 4% as efficiently as fluorescein which has presented a major technical problem until highly amplified video systems have become available to ophthalmologists. We have used a scanning laser ophthalmoscope to perform video indocyanine green angiography at the Doheny Eye Institute since November, 1991. In this paper, I will present several clinical cases that demonstrate the clinical usefulness of indocyanine green angiography.

Low-energy laser irradiation at subthreshold energies has various effects on the eye and other body systems. The known bioeffects in the visual system are manifested mostly in retinal and other ocular tissues morphology. The only ophthalmic low energy laser irradiation effect which has been adequately and thoroughly studied and analyzed is its degeneration postponing activity on the crushed rodent optic nerve. This bioeffect, which demonstrates typical dose-response relationship is of temporary nature. There are many descriptions of low energy laser irradiation curing various eye diseases. None of these claims is scientifically and medically substantiated and all low energy laser therapies must be considered experimental at best.

Evaluation of the retinal nerve fiber layer (RNFL) is important in the diagnosis of glaucoma and other optic nerve diseases. To aid understanding of the mechanism of RNFL reflectance, anatomical data were obtained from electron micrographs of cross- sections of amphibian RNFL and possible and possible scattering structures (axonal membranes, microtubules, and neurofilaments) were modeled as arrays of parallel thin fibers. Microtubules and neurofilaments are already thin fibers, and axonal membranes were decomposed into arrays of adjacent parallel fibers. Internal fields were assigned to each fiber by means of Born approximation or according to rules that depended on membrane orientation. Far field scattering was then calculated by field summation. Calculated scattering was much larger from axonal membranes than from microtubules or neurofilaments. Calculated spectra from 400- 700 nm showed increased scattering at shorter wavelengths for both axonal membranes and microtubules. Calculated backscattering from axonal membranes was approximately equal for incident light polarized parallel (TM mode) and perpendicular (TE mode) to the cylinder axis, but from microtubules TE mode scattering was less than TM mode.

The 193-nm excimer laser is known for its ability to precisely ablate soft biological tissues in the air environment with sub- micron depth control and sub-micron damage zones in the surrounding. The lack of a convenient delivery system and strong absorption of this radiation by biological liquids prevented, until recently, microsurgical applications of this laser. We have constructed special tips that are capable of delivering enough energy for effective removal of soft tissues in a strongly absorbing liquid environment. These tips attach to an articulated arm-based delivery system. This instrument was applied to vitreoretinal membranes removal. The accepted technique for these membranes removal is mechanical peeling and cutting which is associated with strong traction of the retina and this occasionally results in retinal damage. It was demonstrated in this study that the 193-nm excimer laser is capable of safely and precisely cutting and ablating these membranes which enable their removal without exerting any tractional forces on the retina. The effective cutting regime of retina and vitreoretinal membranes occurred at energy fluence of about 250- to 350-mJ/cm²/pulse with a corresponding cutting depth of 50 to 150 micrometers /pulse. The results obtained in this study suggest that this technology could be applicable to a wide variety of intraocular procedures.

A new real-time, scanning slit, confocal microscope designed for the clinical in-vivo examination of the human cornea and anterior segment is demonstrated. The key feature of the in-vivo confocal microscope is the use of two conjugate adjustable slits, and a two-sided oscillating mirror which is used both for scanning and for descanning. The light detection efficiency of this microscope is such that is possible to image both the wing cells, and the basal epithelial cells in the normal, living human eye. This paper also demonstrates the advantage of the slit-scanning confocal microscope over alternative designs based on Nipkow disk pinhole systems. The confocal microscopes based on the Nipkow disk with the pinholes are not capable of imaging either wing cells or basal epithelial cells in the normal in vivo human eye. The images shown in this paper are all from single video frames. No analog or digital image processing was used. The high resolution and high contrast shown in the images that are photographs of single video frames are due to the microscope design.

Chemical burns of the cornea can cause irreversible scarring, leading to visual impairment. The tandem scanning confocal microscope (TSM) was used to evaluate stromal changes in vivo following exposure to alkali, the most devastating form of chemical ocular injury. The corneas of anesthetized rabbits were exposed to filter papers impregnated with NaOH. A 25x water immersion objective lens was used with the TSM, and images captured with a CCD camera. Normal keratocytes appeared as ovoid nuclei. Collagen lamellae were not visible. Alkali led to immediate opacification of the extracellular matrix and loss of keratocytes. The former was quantified by en-face serial optical sectioning and subsequent off-line densitometry of the captured image. Wound healing was monitored as spindle-shaped cells appeared at the wound edge. This was accompanied by the production of fibrillary extracellular matrix. After 3 days, branched cellular processes 100 to 200 micrometers long were extended. By 1 week, dense aggregates of ovoid fibroblasts and whorls of collagen fibers had formed. Despite the overlying scarring, the deeper stromal layers and endothelium were still visible with this technique. We suggest that the TSM would be a useful clinical instrument for the evaluation and treatment evaluation of patients with chemical injuries of the cornea.

The ability to visualize the innervation in the living human cornea is demonstrated. A new, real-time scanning slit confocal for the examination of the in vivo eye is described, and its unique imaging capabilities are demonstrated. A new flying slit confocal microscope, illuminated with a halogen lamp, which has unique imaging characteristics for in vivo human confocal microscopy has been developed. This real-time confocal microscope can image the anterior segment and image en face views of the cornea which includes wing cells, basal epithelial cells, nerves, nuclei and cell bodies of the stromal keratocytes, and endothelial cells. The high contrast images are acquired in a single video frame and do not require any image processing, frame averaging or other techniques improve the single frame images. This paper demonstrates the imaging of the nerve fibers in the anterior cornea. Both stromal nerves, and fibers in the corneal epithelium are imaged. The contrast and depth of focus of this new instrument is not seen with other in vivo real-time confocal microscopes designed for ocular use.

A scanning laser ophthalmoscope (SLO) equipped with an acousto- optical modulator (ACM) was used to make focal acuity and contrast sensitivity measurements in individuals with macular damage. The depth of modulation achieved by the ACM was determined by imaging the SLO raster pattern onto a Pulnix TM 745 video camera and evaluating the intensity distribution with a Big Sky BVA10 beam view analyzer. Contrast levels remained approximately constant over the entire range of SLO input raster power settings. A delta Technologies image processing system produced Landolt ring test stimuli at the center of the raster pattern. Contrast thresholds were determined at various retinal locations by having subjects fixate a specific location on a fixed grid imaged on the raster pattern. This procedure insured that the test stimuli were always imaged in the center of the raster pattern thereby avoiding peripheral variations in the raster pattern intensity distribution. Measurements of contrast sensitivity where focal test targets fell within the macular damage area demonstrated elevated contrast thresholds relative to retinal locations where focal test targets evaluated the border regions between normal and pathological retina.

A scanning laser ophthalmoscope (SLO) employs an extremely low power laser that is focused to a 10 micron spot on the human retina and scanned across the retina at the same rate and pattern that an electron beam travels across a television screen. An image is acquired by measuring the retina's reflectance along each raster. An acousto-optic modulator, in the path of the laser beam, can be used to write targets directly onto the retina. since the SLO can project targets onto the retina while simultaneously viewing the retina, retinal function at specific locations can be unambiguously assessed. We developed software so that sine wave contrast sensitivity gratings were projected onto subjects' retinas with contrasts ranging from 0 to 63.5%. Subjects determined whether the gratings tilted 45 degree(s) to the right or 45 degree(s) to the left. A staircase protocol using 6 reversals was employed to determine contrast sensitivity at each spatial frequency. In early work with the system, 16 eyes of 11 normal subjects were tested. For 8 eyes the intraclass correlation coefficient of reliability was 0.85. This system yields reproducible measures of contrast sensitivity. It may have advantages over other methods to measure contrast sensitivity because the location of tested retina can be unambiguously assessed and illumination levels from the instrument directly measured and controlled. In addition, the small entrance pupil of the SLO can reduce the effects of cataracts measurements of contrast sensitivity.

We report on the use of a recently developed scatterometer to make objective measurements of haze that develops following excimer laser ablation of cornea. Phototherapeutic keratectomies (PTK), 6.0 mm dia, 0.25 mm transition zone, 150 micrometers stromal depth, were done on 4 pigmented rabbits using the VISX Model 20/20 laser system. Scatterometer measurements were made on the normal cornea prior to ablation and at various times up to 125 days postablation and are compared to slit-lamp photographs. Scattering from 3 of the eyes exhibited nearly identical behavior, peaking 20-25 days after treatment at 10 to 12 times the preablation level and then diminishing slowly. Scattering from the fourth eye peaked 14 days after treatment at 28 times the preablation level and remained well above the level of the other three eyes throughout the experimental period. These results suggest that identical PTK treatments can lead to different degrees of subepithelial haze.

There is little time for acquiring coherent data in corneal topography. Especially in height contouring, only a single flash exposure excludes movement artifacts. Starting with a single flash image, from a projected low frequency line grating on a fluorescein stained cornea, there are basically two ways of analyzing the acquired data. First by locating the local phase of the grating which is not very accurate and second, a straight forward frequency demodulation on the complete recorded image. The frequency demodulation had one great advantage, all points were both horizontally and vertically correlated. One major problem was the sensitivity to noise in the amplitude of the recorded image. This problem was solved by limiting the maximum allowable slope in the measured object to +/- 60 degree(s) and using steep digital filters in both spatial and frequency domains. With these improvements we could reconstruct the corneal height image with a lateral resolution of approximately 30 $umm and an axial resolution of +/- 3 micrometers . The accurate information of the entire corneal surface that became available, gave us reasons to doubt the reliability of some commonly used parameters; e.g., the radius of curvature assumes one optical axis but the calculated vectors from the corneal surface do not point to one axis. Because of the nonspherical shape, there probably is a focal volume instead of a focal point.

Because of the inherent limitation of techniques using the cornea as a mirror, we developed a corneal topographer based on diffusely emitted images obtained from two orthoscopically projected grids on the cornea. The additive moire pattern on the cornea was the starting point of the accurate reconstruction of the corneal topography including limbal and apical zone. In general, moire contouring instruments have inherent limitations on their accuracy because of limitations on maximum line frequencies versus depth of field. We eliminated this problem by using discrete Fourier analysis directly on the separately measured grids. For the reconstruction of the corneal height image, we removed all disturbance with a combination of fluorescence-, polarizing- and redundancy-techniques. With these improvements in hardware and extra digital filter techniques in software, we reconstructed the corneal height image with an axial resolution of 3 micrometers and a lateral resolution of approximately 40 X 30 micrometers in a 21.5 X 14.5 X 4 mm³ measuring area. Starting from this 3D topographic cast, other corneal parameters like optical axis, radius of curvature and corneal astigmatism could be calculated.

Our current study has looked at the use of a CW hydrogen fluoride laser to weld human cadaver corneas. The laser used was a Helios CL-II laser operating from 2.3 to 2.6 micrometers . A 6-mm full thickness linear incision was made in the center of the cornea. Two 10-0 nylon sutures were placed to hold the corneal stromal edges apposed during welding. Each specimen was mounted on a motorized micrometer stage. The beam was passed over the incision at a rate of between 0.5-2.0 mm per minute. After welding, the specimens were tested for wound strength by increasing intraocular pressure until the wound leaked. Two eyes had corneal welds performed and pressures of up to twice baseline intraocular pressure could be sustained after the stay sutures were removed. Light and electron microscopy demonstrated full thickness welds to be present. Electron micrographic sections demonstrated apparent interdigitation of collagen fibers between adjacent lamellae. Tissue welding may become an alternate means of wound closure in eye surgery.

The THC:YAG laser was used to create 93 thermal sclerostomies ab externo in 81 glaucomatous eyes of 76 patients. This holmium laser is a long-pulsed, compact, self-contained, solid-state laser operating in the near infrared. A 1-mm conjunctival stab incision was made 5-15 mm from the sclerostomy site to allow entry of a specially designed 22-gauge optic probe that delivers energy at a right angle to the long axis of the fiber. Pulse energies of 80 mJ to 120 mJ were used with a repetition rate of 5 pulses per second. Total energy levels to produce full-thickness sclerostomies ranged from 1.4 to 7.2 J. Seventy-eight eyes received the antimetabolite 5-fluorouracil. Success was defined as postoperative intraocular pressure <22 mmHg, or a >30% reduction in intraocular pressure if preoperative pressure was <22 mmHg. Estimated probability of success was 0.66 at 12 months and 0.57 at both 24 months and 30 months. Mean intraocular pressure of successful cases was 12.7 mmHg, 12.9 mmHg, and 13.0 mmHg at 12, 24, and 30 months, respectively. This is the 30-month report of an on-going clinical trial.

Ablation of corneal tissue with excimer laser light is an effective way to correct refractive errors of the eye. For this purpose a beam-stop (iris diaphragm or interchangeable masks) is illuminated by the laser radiation. The beam-stop is imaged onto the cornea, and circular or elliptic ablations are produced. The computer-controlled process varies the diameter of the ablation area in a way that the inner portions of the treatment zone receive more laser energy than the outer portions, thus flattening the curvature of the refractive surface. For the treatment of hyperopia, the outer portions of the ablation area receive more laser energy to steepen the surface profile of the cornea. The beam delivery system employs several sets of circular, elliptic and ring shaped masks which are etched into a stainless-steel tape.

We describe a new multipurpose maskless method of corneal surgery based on the point-by-point scanning of a focused laser beam which allows the application of a low-energy excimer laser. The crucial scanning parameters to achieve a smooth corneal surface have been investigated. A computer program for the simulation and optimization of the point-by-point scanning process has been developed and tested on contact lenses consisting of PMMA. In addition, a method of measuring the eye-movement by means of the computer-assisted interpretation of photographs was proved for its application in an eye-tracking-system.

The purpose of this study is to demonstrate that topographic irregularities in the form of central islands of higher refractive power can be seen following excimer laser refractive surgery. We reviewed the computerized corneal topographic maps of 35 patients undergoing excimer laser PRK for compound myopic astigmatism or anisometropia from 8/91 to 8/93 at the USC/Doheny Eye Institute. The topographic maps were generated by the Computed Anatomy Corneal Modeling System, and central islands were defined as topographic areas of steepening of at least 3 diopters and 3 mm in diameter. A grading system was developed based on the presence of central islands during the postoperative period. Visually significant topographic steep central islands may be seen in over 50% of patients at 1 month following excimer laser PRK, and persist at 3 months in up to 24% of patients without nitrogen gas blowing. Loss of best corrected visual acuity or ghosting is associated with island formation, and may prolong visual rehabilitation after excimer laser PRK.

We postulated that ablation of trabecular meshwork can be accomplished without damage to Schlemm's canal and adjacent tissue. We performed Erbium:YAG laser trabecular ablation on human autopsy eyes and confirmed the results by histopathologic examination. We used an Erbium:YAG laser (2.94 micrometers ) equipped with a sapphire probe and a quartz tip (350 micrometers diameter). Energy levels of 5, 10, 15, 20, and 40 mJ per pulse were used. The probe was placed directly onto the trabecular meshwork viewed through an operating microscope and focal ablation was performed with single pulses. The tissue was then processed and examined under light and scanning microscopy. The histopathologic study of the trabecular meshwork revealed that a single pulse at an energy level of 15 mJ was optimal for ablating through the trabecular meshwork to Schlemm's canal with a minimal amount of coagulative damage to the surrounding tissue. The mean width of ablated tissue was 150 micrometers +/- 29 micrometers (n equals 18) and the mean depth of ablation was 133 +/- 48 micrometers (n equals 18) at this energy level. The mean width of coagulative damage to adjacent tissue was 16 +/- 8 micrometers (n equals 18). The ablation and coagulative damage increased with increasing energy levels. The Erbium:YAG laser equipped with a contact probe effectively ablates trabecular meshwork with little surrounding coagulative damage. This feature may minimize scarring and result in a high long-term patency rate of outflow channels created by laser trabecular ablation.

A number of recent studies have demonstrated the efficacy of transscleral Nd:YAG and diode cyclophotocoagulation in the reduction of intraocular pressure. The choice of an appropriate wavelength for this treatment modality has been based on a tradeoff of maximal transmission through the superficial layers of conjunctiva and sclera, and maximal absorption in the ciliary body. A UV-Vis-NIR spectrophotometer was used to determine the optical properties of the conjunctiva, sclera, and the ciliary body. These optical properties were used in a Monte Carlo Model to examine the fluence and the resulting rate of heat generation in each layer for the wavelengths of Nd:YAG (1064 nm), semiconductor diode (850 nm), Ruby (693 nm), Krypton yellow (568 nm), and Argon (514 nm) lasers. The results provided a basis for a theoretical prediction of the extent of thermal damage in the ciliary body as a result of transsclerral cyclophotocoagulation. Key Words: transscleral cyclophotocoagulation, Monte Carlo Model, rate of heat generation, Nd:YAG laser, Argon laser, Krypton laser, Ruby laser, semiconductor diode laser.

Goniotomy is an effective treatment for primary infantile glaucoma. Because a cloudy cornea may prevent a clear view of the anterior chamber angle through the operating microscope, we investigated whether an endoscope can be combined with a cutting laser to perform laser goniotomy in a surgical model of primary infantile glaucoma. The anterior chambers of cadaver procine eyes were deepened with a viscoelastic material. A 300-micron-diameter silica fiber coupled to an Olympus 0.8-mm-diameter flexible fiber optic endoscope entered the anterior chambers through 4-mm corneal incisions. The anterior chamber angles were clearly observed on a videoscreen as the endoscopic fiber optic laser scalpel approached the pectinate ligaments. With the guidance of a He-Ne aiming beam, the anterior chamber angle pectinate ligaments were cut over a 160 degree arc with a pulsed Ho:YAG laser (2.1 micrometers wavelength, 50 mJ, 5 Hz repetition). The specimens were fixed in glutaraldehyde and processed for scanning electron microscopy, or fixed in formalin and processed for light microscopy. The treated area demonstrated incision of the pectinate ligaments with opening of the underlying trabecular meshwork. The edoscopic fiber optic laser scalpel is capable of cutting the pectinate ligaments in a surgical model of primary infantile glaucoma. Therefore, it may be a useful instrument for performing goniotomy when a cloudy cornea in primary infantile glaucoma prevents visualization of the anterior chamber angle with a goniotomy lens.

Several years ago when our research group at Premier Laser Systems began to explore the possibility of developing a new multipurpose ophthalmic laser, we started by doing two things: we listed the specific clinical applications which we felt were most needed, and we delineated the various desirable characteristics which we believed an ideal intraocular laser system should have.

The use of the pattern-electroretinogram (PERG) as a noninvasive diagnostic tool for the early detection of glaucoma has been supported by a number of recent studies. We have developed a unique device which uses a laser interferometer to generate a sinusoidal fringe pattern that is presented to the eye in Maxwellian view for the purpose of producing a PERG response. The projection system stimulates a large visual field and is designed to bypass the optics of the eye in order to measure the true retinal response to a temporally alternating fringe pattern. The contrast, spatial frequency, total power output, orientation, alternating temporal frequency, and field location of the fringe pattern presented to the eye can all be varied by the device. It is critical for these parameters to be variable so that optimal settings may be determined for the normal state and any deviation from it, i.e. early or preclinical glaucoma. Several interferometer designs and optical projection systems were studied in order to design a compact system which provided the desired variable pattern stimulus to the eye. This paper will present a description of the clinical research instrument and its performance with the primary emphasis on the optical system design as it relates to the fringe pattern generation and other optical parameters. Examples of its use in the study of glaucoma diagnosis will also be presented.

We report the development of a new stroboscopic light source system and technique for performing intraoperative fluorescein angiography during penetrating keratoplasty for aphakic or pseudophakic bullous keratopathy. A controllable pulse xenon light source system with a fiber optic endoilluminator probe is used to perform high-quality intraoperative fluorescein angiography during penetrating keratoplasty in pigmented rabbits and human subjects. Following corneal trephination and extraction of the intraocular lens, a temporary Cobo keratoprosthesis is secured while a 20-gauge endoilluminator is inserted into the vitreous cavity through a limbal incision. The endoilluminator is advanced to a retinal illumination area of approximately 3 DD and 10% fluorescein is injected intravenously. A microscope camera coupled to a 50:50 beamsplitter photographs the passage of fluorescein dye while the surgeon maintains an unaltered view through the operating microscope. Angiograms through a keratoprosthesis show excellent contrast and resolution, comparable to standard fluorescein angiography. Fine peripapillary vessels are seen reproducibly and with great detail in the rabbits. All the phases of retinal angiography can be seen, including arteriolar constriction and capillary nonperfusion in one of four human subjects examined. High quality intraoperative fluorescein angiography can be performed in patients undergoing penetrating keratoplasty for aphakic/ pseudophakic bullous keratopathy. With this technology, preexisting retinal disorders such as cystoid macular edema might be identified in the perioperative setting allowing for important management decisions to be made intraoperatively.

A model laser surgical probe was designed and built to employ laser induced breakdown (LIB) in cutting fibrovascular membranes within the vitreous cavity of the eye. The probe is a simple design of a gradient index (GRIN) lens attached to the tip of a multimode fiber. It is designed to fit through a sclerotomy incision and enter the vitreaous cavity for work anterior to the retina. The laser light is focused close to the tip of the probe without causing GRIN lens damage. Thus a widely divergent beam behind the focus will diminish potential laser damage posterior to the target tissue. A Nd:YAG 1064-nm 10-ns pulsed laser was used with the probe inserted into a cell of tap water. Threshold data for laser induced breakdown was taken ad fit to a probit curve. The data was compared with past LIB threshold data. The radiant exposure versus distance from the probe was plotted via spot size measurements. This measurement gave a rough indication of the distance the probe must remain from the retina to be well below retinal maximum permissible exposure (MPE) levels. In-vitro threshold measurements of bovine vitreous were taken and compared to the water threshold. Finally, collagen membranes were lased with the probe to demonstrate its functional application.

A fiber optic point-source was designed for in vivo measurement of light transmission and scattering. The point-source consists of an optical fiber, with a core diameter of about 8 micrometers , mounted in a stainless steel cannula. At the tip of the cannula the entrance face of a 1-mm micro prism is fixed. The hypotenuse face of the prism is coated with aluminum and acts as a mirror reflecting the light out through the exit face. The arrangement allows light fed into the optical fiber to be emitted in a direction perpendicular to the axis of the cannula. The cannula is bent to allow its in vivo introduction into the eye at the posterior pole. The prism is positioned close to the retina along the optical axis of the eye. The light is transmitted through the eye in a reversed beam path (from vitreaous towards cornea). In the vitreous, the light source emits within a numerical aperture of about 0.1 allowing the exiting light to be well confined through the dilated pupil. After careful calibration of the light source, accurate measurements can easily be performed by positioning detector equipment outside the eye. The small size of the point-source (in the range of the eye's resolution) allows measurement of light scatter at both wide and narrow angles. With this intraocular fiber optic technique, it is possible to perform several types of experiments, including measurements of light transmission and scattering, and also studies of fluorescence phenomena in different parts of the ocular media.

In course of instrumentation technology developments for spaceborne experiments under microgravity in areas of nucleation control during crystal growth and critical point phenomena, miniaturized optrodes were achieved. Described optrodes can be easily integrated into ophthalmoscopes as well. By combination of monomode-fibres and focusing lenses, light source powers were minimized and photon correlation spectroscopy (PCS) measurements were conducted with lowest intensity levels on human eye lenses. In vivo measurements enabled resolution of various radii fractions. Detailed analysis and applied statistical categorization of results are described. Measurements are evaluated for the first time with a 288-channel real-time correlator. PCS is studied concerning baseline determination addressing multi-exponential fit procedures. Potential extensions of PCS to developments of simultaneous multi-angle profilometry are indicated. Approaches to aggregation mapping are presented, addressing demands of depth related resolution within scattering volumes. Velocity studies reflecting local blood propagation, achieved by described opto-electronics on possibly different capillary diameters are included. Interpretations are given concerning distortions observed at sections of measured autocorrelation functions.

One of the major problems of keratoprosthesis (KPro) is postoperative expulsion resulting from inadequate haptic fixation. To overcome this obstacle, a new design was conceived using a retrocorneal fixation. An earlier version of our KPro was made of two pieces: an optical element threaded in a surrounding haptic. Both pieces were made of medical grade PMMA. Although encouraging results were obtained, some anatomic failures occurred (20% at 5 years). To improve fixation and sealing, a biocolonizable skirt was recently added to the keratoprosthesis. This skirt should prevent migration of biocontaminants inside the eye. Made of porous PTFE, the skirt was ultrasonically held inside a one-piece PMMA keratoprosthesis by ultrasonic microwelding. Thus far, implantations carried out in 8 patients show a good tolerance.

Two centuries ago, Pellier de Quengsy suggested the implantation of a glass plate in the cornea, and the Chevalier de Tadini the replacement of the crystalline with a glass lens. This was the beginning of ophthalmic prostheses. Since, material sciences have made tremendous progress with respect to ocular implantation. As postoperative implant rejection and complications have to be minimized, several requirements on the materials to be selected can be given: In order to minimize stress locations and local tissue deformation, the physical properties of the implants should be similar to those of the replaced tissue. The material should remain stable in time and not degrade and, in some instance, they should be colonizable. The implants should not induce reactions to surrounding tissues, and therefore should be non-carcinogenic, nonallergenic and non-immunogenic. When the material is implanted within the optical path, its optical properties should be similar to those of the replaced tissue. Especially, a high transparency is required in the visible part of the spectrum (400-700 nm) with a low scattering. As the retina and the lens need to be protected from UV radiations and, to a lesser extend, from radiations in the deep-blue, the materials should also present a high degree of absorption at those wavelengths. To avoid potential complications at long term, an absorption as large as that of the eye tissues should also exist in the infrared part of the spectrum. Because light has to be focused sharply on the retina, the implant geometry and index of refraction have to accurately selected. In addition, the surface quality and the homogeneity of the implants have to be as high as possible. Ideally, to respect the natural physiology of the eye, the materials' index of refraction should match that of the tissue it is design to replace. Consequently, whenever possible the implant material should mimic the properties of ocular tissues. Certain glasses and crystals (sapphire) mounted on metal alloys or ceramics fixation devices (haptic), have been used for the fabrication of intraocular lenses and keratoprostheses. However, the most well-known material for ophthalmic implantation is the thermoplastic polymer PMMA (Polymethylmethacrylate) obtained by polymerization of its monomer, methylmethacrylate. Its ocular stability was accidentally demonstrated after the second world war [1]. Ridley took first advantages of this material for intraocular lenses [2] and, a few years later, MacPherson and Anderson for intracomeal implants [3]. Synthetic polymers are made of very large molecules consisting of either three-dimension (bulk) or two-dimension (linear chains) crosslinked repeating monomers. Thermoplastics are organic polymers having in common a long repetitive carbon chain [-C-]. The mechanical properties of these polymers depend principally on the van der Waals intermolecular forces. In some other polymers the various chains are chemically linked to each other by covalent bonds. Carbon, however, is not the only atom possessing the ability to form very large molecule with other atoms. Silicon is another element that exhibits similar properties. The chain is made in this case of repetitive siloxane units [-Si-O-], whereby the molecular weight of the polymer is determined by the chain's length and type of cross-linking. At low molecular weight, a few thousands centistokes (ctsk), and end-crosslinking, polymers are fluid. Because of their viscosity and their ability to repel water, they are often referred to as silicone oils [4]. Gels are another form of matter, an intermediate between solid and liquid, that can also be used for the fabrication of implants. Gels consist of long chain molecules (generally polymers) that are bulk cross-linked to create a topological complex network immersed in a liquid medium. The latter prevents the polymer network for collapsing while the former traps the liquid. Such gels have been found to be very sensitive to external parameters such as temperature, pH, pressure and salt concentration. The rigidity of a gel is function of the number of crosslinks per linear chain composing the bulk. Most ocular tissues, such as the vitreous, a loose low rigidity network made of proteins (<99 % H20), and the corneal stroma (or the sciera), a dense network composed of collagen (<75%H20),could be considered as gels.

Most complications of keratoprosthesis (KPro) occur at the tissue-to-implant interface. The ideal prosthesis would eliminate this interface by having the tissue actually grown into the support material forming the haptic. We present a 2-year clinical human study of a novel biocolonizable KPro on 24 eyes of 24 patients. To promote tissue stability, a 9 mm (Phi) skirt made of a new microporous fluorocarbon was used. The optical core or the KPro optic was made of a medical grade polymethylmetacrylate (PMMA) commonly used world-wide to fabricate intraocular lenses. The optic was 5 mm in diameter and 2.67 mm long. The average follow-up was 14 months. Mean corrected visual acuity was 20/100. Anatomic failures occurred in 5 cases in the first year (1 lens dislocation, 1 endophthalmitis, 3 extrusions). These preliminary results appear encouraging. However, we did not eliminate all the complications with this biocompatible inert microporous polymer. Intensive research in mechanical, chemical, and surface biocompatibility is required to develop a true artificial cornea.

Over the past century, several refractive surgical techniques have been devised to correct ametropia. These can be categorized in three difterent ways: (I) The first is the "subtraction" method, consisting in the removal of tissue from the patient's cornea1 . Removal is almost always performed on the central optical zone. Mechanical instruments or photoablative lasers are used for these procedures. Today, the ArF I 93nm excimer and, potentially, the 21 3nm frequency quintupled Nd:YAG solid state lasers have replaced the keratomileusis mechanical techniques originally developed by José I. Barraquer; (II) The second is the Nadditivew method, involving the implantation of donor tissue or a polymeric implant in the host corneal stroma1 or under the host comeal epithelium2. One subclassification of the second method consists of implanting a lenticule over the central optical zone while the others rely on bending the host peripheral cornea either using a ring implant; (Ill) The third method consists of thermal shrinking portions of the stromal collagen at difterent locations on the cornea to correct myopia, hyperopia, or astigmatism. Several kinds of heat sources have been investigated, including hot wires cautery and RF diathermy, but the mid-infrared lasers have given the best clinical results thus fare.

In the overview optical models of cornea, sclera, and crystalline lens humor will be presented. On the basis of these models eye tissue transmittance spectra and scattering indicatrices for the main informative elements of the mueller matrix will be analyzed. This paper will discuss some problems of eye tissue optical characteristics control, and possibilities and perspectives of elastic scattering spectroscopy in cataract diagnostics.

Optical models of crystalline lens were analyzed, to increase understanding of high transparency and loss of it with aging and for pathology tissue. Experimental and calculated human crystalline lens spectra in a wide range of wavelengths are presented. Calculations were made for the model of short-term- order scatterers, taking into account interference effects of scatterers and specific features of absorbing species, such as protein-bound tryptophan, kynurenine, and age-related chromophores leading to generalized yellowing of the lens proteins. Satisfactory qualitative coincidence of calculated spectra with spectrophotometric data for isolated human lens is shown. Measured angular dependencies of scattering matrix elements are sensitive to modification related to aging and cataract appearance.

Mid-IR lasers that operate near the absorption peak of water have a short penetration depth in ocular tissues. Ablation of tissue can be accomplished with minimal coagulative damage to underlying structures. We used an erbium:YAG laser equipped with a contact probe to create retinotomy sites in the human retina of eye bank eyes. An erbium:YAG laser (2.94 micrometers ) equipped with an infrared transmitting glass fiber and a sapphire tip (400 micrometers ) was used to directly ablate the surface of the retina. We administered both single and multiple pulses to the macula and peripheral retina using energy levels from 4 to 16 mJ per pulse. The retinas were then examined histopathologically to evaluate the extent of ablation and coagulative damage. Single pulses at low energy levels were noted to cause ablative damage to the nerve fiber layer and ganglion cell layer without a notable coagulative effect. The mean ablation depth at lower energy levels was less than the mean ablation depth at higher energy levels. Extensive laser application produced disruption of the retinal pigment epithelium, choroid and sclera. the erbium:YAG laser equipped with a contact probe is an effective means of creating retinotomies in human autopsy eyes. When used in the single pulse mode at lower energy levels, the erbium:YAG laser appears capable of removing superficial retinal layers without damaging deeper structures.

An ophthalmic laser system is reported, when this laser is used to perform an operation, an interference effect can produce, then the laser power density acted upon local eye tissue is increased.

A pulsed erbium:YAG laser equipped with a 1-m zirconium fluoride fiber coupled with a sapphire tip was used to photovaporize human autopsy lens tissue. Measured diameter of photovaporization of human lens cortex with erbium:YAG laser ranged from 117 micrometers at a fluence of 3.98 J/cm² to 227 micrometers at fluence of 15.92 J/cm². Measured diameter of photovaporization of human lens nucleus with Erbium:YAG laser ranged from 265 micrometers at a fluence of 11 J/cm² and 295 micrometers at a fluence of 15 J/cm². Endolenticular phacovaporization with the Erbium:YAG laser appears to be a potentially efficient and precise means of removing the human lens.

Thirty-six cases of crystal hyaline cortex emulsification operations for immature cataract with an YAG laser are reported in this paper. According to the basic theory of interaction between laser and tissues, the mechanisms of laser induced hyaline cortex emulsification in the operation are discussed, and it is put forward that the mechanical effect of laser plasma shock wave and photochemical effect of plasma flash on eye tissues result in degeneration and coagulation of the crystal hyaline cortex tissue protein.

Twenty-three cases of anterior capsular aftercataracts after implantation of the intraocular lens in the posterior chamber have been treated by a Nd:YAG laser. Afteroperatively, visions of those patients who had serious visual obstacles have been improved very obviously, and the curative effect is good. No damage on the intraocular lens.

The EyeSys Corneal Analysis System and the TMS Topographic Modeling System were compared, both when properly aligned and under specified conditions of misalignment, in their ability to map a sharp transition in curvature in order to model the type of surface topography that might be produced after photorefractive keratectomy. Results indicate that neither system can accurately reconstruct the transition zone, and that the source of the error is data averaging across the meridians to produce a smooth display for the EyeSys system, and lesser averaging combined with a fundamental error in reconstructing the peripheral region for the TMS system. In addition, the TMS is shown to possess greater sensitivity to errors of decentration and defocus for the bicurve test surface, while the EyeSys system is shown to be relatively immune to similar conditions of misalignment.

This study was performed to obtain information about matrix metalloproteinase (MMP) expression in excimer-wounded corneas and to determine whether MMPs expression correlates with the depth of the ablation. 6-mm excimer keratectomy (60 or 180 micrometers ) was performed using the 193-mm ArF excimer laser on 12 NZW rabbits. Corneas treated with mechanical epithelial debridement and untreated corneas served as controls. Rabbits were killed at 20 and 30 hr after laser ablation. Zymography after SDS extraction was performed on regenerated central epithelium and the central stroma to determine MMPs expression. We observed enzymatic activity of a 92 KDa band in the epithelium of excimer-ablated corneas but not in that following debridement wounds and untreated controls. The expression of the 92 KDa MMP was most pronounced with the deeper excimer ablation. A 72 KDa band of enzymatic activity present in the stroma of all treated and control eyes was also seen in the epithelium of excimer-ablated corneas. These proteolytic enzymes may play an important role in wound healing and remodelling after excimer keratectomy.

The possibility for selective thermodenaturation (coagulation) of pigment epithelium layer by an appropriate choice of parameters of short laser pulses, while the functioning of neighboring layers is preserved, is shown on the basis of mathematical simulation.

Testing of the excimer lasers used in PRK requires special considerations in terms of ease of use, day-to-day reliability, and high resolution to see details of beam interference effects. SensorPhysics employs a patented photochromic material on a polyester substrate to record permanent, instant records of the laser and laser system output. Since each SensorCard is used only once concerns about detection device deterioration are not an issue. The SensorCards have a demonstrated resolving power on the order of 0.1 micrometers . A small, portable reading device is used to convert the SensorCard optical density to a mJ/cm² value. Special software also measures beam uniformity to +/- 1% to provide both qualitative and quantitative analysis. Results of use in clinic environments will be presented. In particular detection of exposure `islands' will be demonstrated. The techniques employed are similar to those we developed for UV laser micromachining and lithography four years ago.

Fluorescein angiography is currently the main method for evaluation of the retinal vascular patency. Ashton noted that capillary patency to the small fluorescein molecule may differ from that of the larger red blood cells. He concluded that fluorescein angiography is not able to demonstrate a developing stenosis, that might be the precipitating cause of a later capillary closure in various microvasculopathies. Sarelius et al have shown, in hamster cheek pouch and cremaster muscle, that fluorescently labeled erythrocytes in known concentrations can be used for the direct measurement of capillary flow parameters. The only assumption that this method relies on, is that the labeled cells are rheologically normal and therefore reflect the behavior of the total cell population. We have developed a new method for an in-vivo, real-time demonstration of the blood cell flow in the retinal capillary net. Based on the assumption presented by Sarelius et al, measurement and analysis of the retinal capillary blood cell flow is also possible from the results achieved by the new method.