| Home Features Departments News & Trends CME May Article Instructions Exam September Article Instructions Exam 
|
 Continuing Medical Education New Approaches to Age-Related Macular DegenerationAs the leading cause of severe, irreversible vision loss and legal blindness in the United States, age-related macular degeneration (AMD) is a major public health concern. As our elderly population increases, the interest in developing successful treatments will only intensify.Much of the research effort in AMD is directed toward the exudative or wet form of AMD, which is characterized by choroidal neovascularization, subretinal fibrosis and disciform scarring. The hallmarks of the non-exudative or dry form are abnormalities of the retinal pigment epithelium and drusen.  This five-part CME article reviews several new approaches to laser treatment of both types of AMD. Some of these treatments have yet to receive U.S. Food and Drug Administration approval, and, where appropriate, their status is indicated in their individual sections. The initial segment looks at efforts to treat AMD before it reaches the wet or exudative stage. Drusen in the eye of a patient over 50 years old strongly suggests the presence of AMD. In addition, patients with drusen are at significantly greater risk of developing choroidal neovascularization (CNV). Study parameters on the prophylactic use of laser photocoagulation in eyes with drusen have varied to such a degree that the issue remains unresolved. In addition, no study to date has shown definitively that prophylactic therapy prevents the development of CNV. A new multi-centered trial, assessing whether subthreshold laser treatment has a beneficial effect on eyes with drusen, is described.  The next two segments review new approaches to laser photocoagulation in patients who have advanced beyond the dry stage. Laser photocoagulation has been validated as a proven treatment for certain types of CNV. Ophthalmologists performing macular photocoagulation with the goal of achieving white retinal burn have historically preferred green and/or yellow lasers. Several studies have explored the pros and cons of using various laser wavelengths for macular photocoagulation, though no significant difference in treatment outcome was found in most cases. However, several new approaches to laser photocoagulation suggest that treatments without visible retinal burn may give equally effective results. Red or infrared laser wavelengths may offer a better choice for safer, less-invasive option than the green or yellow wavelengths.  Transpupillary thermotherapy (TTT) is another subthreshold laser
photocoagulation technique that can result in closure of subfoveal CNV with
less damage to the neurosensory retina and less effect on visual acuity than
conventional laser photocoagulation techniques. Preliminary results suggest
that TTT is safe and may be efficacious in treating occult CNV.
Next, another new technology may enhance laser treatment. High-speed indocyanine green (ICG) choroidal angiography is a new approach to treating the afferent, or feeder, vessels that supply blood to retinal and choroidal vascular lesions. The application of high-speed digitized scanning and data processing of ICG generated images permits the analysis of human choroidal blood flow in real time. This segment describes how high-resolution digital infrared cameras, along with desktop computers capable of managing digital information, have made high-speed digital ICG choroidal angiography possible, and details the use of the technology in clinical practice. Finally, photodynamic therapy (PDT) continues to draw high interest among surgeons and patients alike as offering the possibility of an alternative to laser photocoagulation that may halt or delay vision loss in certain cases. The concluding segment describes research into another of the new agents that are seeking regulatory approval for use in treating patients with PDT. Part 1. Treating Drusen with Prophylactic Laser Therapy Thomas R. Friberg, MD, Pittsburgh Preventive laser treatment of patients with high-risk drusen may reduce the incidence of AMD. The presence of drusen in the eye of a patient more than 50 years old strongly suggests that AMD has occurred. This is because drusen themselves can be considered by-products of impaired recycling of rod and cone outer segment debris: Lipoproteins within the outer segment membranes are continually discarded and must be recycled by the retinal pigment epithelium (RPE) cells located adjacent to and beneath the photoreceptors. As a patient ages, this process diminishes, and lipofuscin-laden debris may accumulate at the base of the RPE cells and along Bruch’s membrane. Clinically, drusen vary in size from a few microns in diameter to much larger confluent complexes ranging from hundreds to thousands of microns in size (See Figure 1A and 1B). Occasionally, confluent drusen may appear as a solid retinal pigment epithelial detachment (RPED). Drusen might signal an altered pathophysiology of the RPE cells, not only at the site where they appear but also in areas distant from drusen. While exact statistics vary, patients age 65 and older who have drusen in both eyes have approximately an 18 percent chance of vision loss from AMD over a three-year period.1 Patients with an exudative lesion in one eye are already at a higher risk-about 10 percent per year-of losing vision in their fellow eye.2
While several novel therapies show substantial promise for treating eyes with exudative macular degeneration, few prophylactic strategies have proved clearly effective. Vitamins, minerals, micronutrient supplements and diets replete with carotenoids, lutein and other nutrients have been promoted as having long-term benefits for patients with dry AMD.3,4 The advantages are subtle and difficult to measure, though, and clinicians seek a more definitive therapy. One therapy currently under investigation is prophylactic laser treatment for patients with drusen.5 Although numerous studies have shown that using photocoagulation to treat eyes harboring macular drusen promotes drusen resorption, most of these studies have been limited and uncontrolled, and laser treatment and follow-up parameters vary widely among them. For example, some of the studies used visible argon or krypton wavelengths for treatment; some required direct treatment of the drusen, while others promote favorable outcomes by placing grids of varying extent. In some instances, prophylactic drusen therapy has even proved harmful, particularly to patients with an existing exudative lesion in their fellow eye who have had prophylactic treatment for drusen in the remaining eye. In one study using argon laser photocoagulation, patients treated prophylactically in their remaining eye had an increased incidence of CNV.5 Promising results, however, have emerged from a previous pilot study employing a grid of 48 810-nm laser spots with a 125-µm diameter. Laser intensity was randomized. Published two-year follow-up results indicated that heavier laser treatment is not necessary to eliminate drusen and that a small but significant number of treated patients enjoyed a visual benefit when the drusen disappeared.6 Unfortunately, no study to date has shown definitively that prophylactic therapy prevents the development of CNV. In fact, one estimate asserts that to achieve a 20-percent reduction in CNV event rates in patients with drusen, a prospective study would need to randomize between 1,200 and 2,500 bilaterally eligible patients.
The PTAMD Study The PTAMD (Prophylactic Treatment of Age-Related Macular Degeneration) Study, a prospective, multi-centered, randomized, controlled trial, is designed to determine whether very minimal laser treatment placed in a 48-spot grid surrounding the foveola has a beneficial effect on eyes with drusen. Indeed, the subthreshold treatment is so light that the lesions are subclinical and invisible to the laser surgeon directly after treatment. Presently, the protocol specifies a single treatment for eligible patients, with follow-ups using fundus photography, fluorescein angiography and clinical examination, as well as rigorous measurements using ETDRS charts. Although the PTAMD Study was originally designed to enroll bilaterally eligible patients (those with drusen in both eyes) and unilaterally eligible patients (those with an exudative lesion in one eye and drusen in the remaining eye), enrollment in the unilateral group was suspended recently. Data has suggested that this treatment is unlikely to benefit such patients. Enrollment in the bilateral arm of the study continues, however, and will require more than 1,200 patients, assuming the original sample size estimates are valid. (See sidebar, left, for complete eligibility requirements.) Because patients with drusen are at significantly greater risk of developing CNV even without prophylactic laser treatment, careful investigation continues into the efficacy of this type of treatment. Some clinicians believe that previously published results justify continuing prophylactic laser therapy, but investigators for the PTAMD Study and other researchers of macular degeneration therapy do not recommend it for AMD. Over the next several years, we anticipate having solid data to support or refute the benefits of preventive laser photocoagulation. Positive results may include improvement of visual acuity and contrast sensitivity, as well as a decreased risk of choroidal neovascular events.
Part 2. Infrared Laser Treatment Giorgio Dorin, Mountain View, Calif. A New Approach to Minimally Invasive Photocoagulation Therapy for CNVM and Macular Disorders Laser photocoagulation has been validated by the Macular Photocoagulation Study (MPS) as the first proven treatment for well-defined extrafoveal, juxtafoveal and some types of subfoveal choroidal neovascular membranes (CNVMs). Several studies have explored the pros and cons of using various laser wavelengths for macular photocoagulation, such as 488, 514, 521, 532, 568, 577, 647 and 810 nm. No specific benefits nor significant difference in treatment outcome was found for one wavelength over another, with the exception of the argon blue 488 nm wavelength, which has been determined theoretically undesirable for its absorption by macular xanthophyll. This result is not surprising, since the MPS protocol required that all wavelength treatments achieve the same common endpoint: a “white” retinal burn. The absence of wavelength specificity can be easily understood when one considers that the white retinal burn does not result from the direct interaction between the laser beam and the retina, which is basically transparent to all laser lines between 514 nm and 810 nm. Rather, retinal burn results from indirect thermal damage from intense heat generated elsewhere, mainly at the retinal pigment epithelium (RPE), which is the main absorber for all wavelengths in the above range. The pursuit of this common endpoint in the MPS defeated and masked any wavelength-specific energy-heat conversion in the three major endogenous absorbing chromophores of the macula: the melanin, xanthophyll and hemoglobin. Melanin in the RPE and in choroidal melanocytes, for example, is the strongest laser-absorbing ocular pigment and the major energy-heat conversion site for all wavelengths. The spectral response curve of melanin’s absorption of low to high wavelengths is a gradually declining line. That is, shorter wavelengths have higher absorption with lower penetration; as a result, green and yellow (514, 521, 532, 568 and 577 nm) lasers will more easily produce retinal burn mediated by the conduction of heat generated at the RPE than will red and infrared (647 and 810 nm) lasers. For this reason, ophthalmologists performing macular photocoagulation with the goal of achieving white retinal burn have historically preferred green and/or yellow lasers. As we will see, however, if treatments without visible retinal burn can give equally effective results, then red or infrared laser wavelengths may offer a better choice for safer, less-invasive photocoagulation protocols.
Properties of Red and Infrared Wavelength Treatment In cases where ocular media with pathologic or age-related opacities and vitreous hemorrhage prohibit the use of green and yellow lasers, some ophthalmologists have used red and infrared wavelengths, available from krypton (647 nm) and diode (810 nm) lasers, respectively. The invisible 810-nm wavelength in particular has several desirable properties. These include:
Despite these qualities, the 810-nm diode laser did not become popular for macular photocoagulation mainly because of its deeper penetration, which requires the use of higher power over a longer exposure to produce the endpoint of a conventional “argon-like” retinal burn and this causes patient discomfort. Recently, however, the notion has developed that the “white” visible burn endpoint used in conventional photocoagulation protocols represents a supra-threshold, full-thickness retinal burn, which is progressively enlarging and is probably therapeutically redundant. This thinking has raised interest in developing less invasive laser therapies.1 Basic science works2,3 and reported outcomes of clinical trials employing minimal impact, neuroretina-sparing photocoagulation protocols4,5,6,7,8 suggest that their therapeutic effectiveness is at least equal to that of conventional, more destructive treatments. The mechanism of action of laser retinal photocoagulation is still not completely understood, but it is becoming increasingly apparent that minimal-impact photothermal stimulation of cellular apoptosis (rather than supra-threshold acute necrosis) may suffice to trigger the beneficial effects of photocoagulation. Minimal impact thermally modulated laser treatments have been successfully applied to the closure of extrafoveal CNV feeder or modulating vessels with no apparent damage to the neurosensory retina or to the RPE’s optical properties at the time of treatment.10 Sub-visible threshold treatments also have been applied to macular grids for macular edema4,5,6,7,8 and for high-risk drusen,11,12 as well as for panretinal photocoagulation (PRP). Transpupillary Thermotherapy (TTT)13,14 for subfoveal occult CNVM is a typical example of a sub-visible threshold photocoagulation protocol, in which treatment involves a “large spot-low irradiance-long duration” laser exposure to create and maintain moderate photothermal elevation (hyperthermia). Advantages of Diode Laser Treatment The 810-nm diode laser presents a number of theoretical and practical advantages for neuroretina-sparing photocoagulation for the following biophysical and technical characteristics:
Using the appropriate laser energy delivery mode, you can select different  combinations of exposure duration, pulse lengths, duty cycles,
repetition rates and number of pulses, to precisely regulate the thermal
gradient created in the RPE/inner choroid. Thus, you can obtain RPE threshold
stimulation or neovascular closure with little or no damage to the neurosensory
retina and with little change in the RPE’s optical properties.
MicroPulse Mode (10% Duty Cycle). An example of a 200 msec exposure enveloping 100 MicroPulses with a 2.0 msec period (0.2 msec “ON” + 1.8 msec “OFF” time), 500 Hz repetition rate and 0.2/2.0 msec = 10% duty cycle. Giorgio Dorin, an electronic engineer with a specialty in nuclear medicine, was named an honorary member of Societá Italiana Laser in Oftalmologia for his contribution to the development of ophthalmic laser application. He is the director of clinical applications development at Iridex Corp.
 Preoperative fundus photograph and OCT. Greater than 500 µm of thickening is observed in the topographic display of the OCT. Cross section of the retina reveals subretinal fluid. Visual acuity is 20/50. Part 3. Treatment of Subfoveal Choroidal Neovascularization with Transpupillary Thermotherapy Elias Reichel, MD, Boston Studies show that transpupillary thermotherapy appears to be a viable technique for treatment of CNV. Transpupillary thermotherapy (TTT) is a subthreshold laser photocoagulation technique that can result in closure of subfoveal CNV with relative sparing of the neurosensory retina compared to conventional laser photocoagulation techniques. Visual acuity test results suggest that with TTT retinal function is spared; the observation of retinal status after TTT (for example, retinal color change) shows that the therapy causes little or no damage to the neurosensory retina. These preliminary results suggest that TTT is safe and may be efficacious in treating occult CNV. Although TTT also has been observed to cause closure of classic or well-defined CNV as well, this effect requires further study.  Postoperative fundus photograph showing mild pigmentary changes within the macula. The thickness of the macula is diminished and vision has improved to 20/32. Theoretical Models TTT creates a prolonged but moderate temperature increase in the choroid, retinal pigment epithelium (RPE), retina and abnormal CNV. The temperature increase occurs within one second after treatment begins. Empiric data suggests that early retinal coagulation necrosis, in which retinal whitening or color change may be first observed, occurs at 51°C. According to theoretical models developed by Martin Mainster, PhD, MD, the temperature rise during TTT approaches 49°C. By contrast, conventional laser photocoagulation raises the temperature to near 75°C. The principles of TTT are similar to those of conventional laser photocoagulation in that melanin is the chromophore that absorbs the infrared light. Thus, darkly pigmented lesions or pigmented fundi must be taken into account and the laser power setting must be adjusted accordingly. Because Infrared wavelength (810 nm) maximally penetrates the choroid and RPE, but relatively spares the clear structures of the eye, including the neurosensory retina, it is favored over conventional photocoagulation techniques. Unless the retina contains pigment or blood, infrared laser also results in minimal retinal absorption. TTT is performed with the IRIS laser using the large spot delivery device (the laser software allows treatment exposure times of up to several minutes, making it effective for treating retinoblastomas and choroidal melanomas, as well as CNV). Additionally, large spot diameters of 0.8 mm, 1.2 mm, 2.0 mm, and 3.0 mm can be increased with magnifying lenses. For example, a 2.0x magnifying lens used with a 3.0-mm large spot aperture will create a 6.0-mm diameter spot on the retina. Laser power should be doubled when using such lenses. The method of determining appropriate power settings for TTT is significantly different from that of traditional laser photocoagulation, in which the power setting is proportional to the radius squared (the area) of the spot size. With TTT, heat dissipation from a large spot requires that the power setting be proportional to the diameter of the spot instead. If the appropriate power setting for a 3.0-mm spot is 1000 mW, for example, then the power setting for a 2.0-mm spot would be 670 mW; for a 1.2-mm spot it would be 400 mW. Likewise, the appropriate setting for a 3.0-mm spot using a 2x-magnification lens would be 2000 mW. The status of choroidal circulation is especially important with TTT. If it has been compromised, it may be necessary to lower the laser power setting. Signs of choroidal compromise may include geographic atrophy of the RPE and loss of the choriocapillaris (caused by prior laser treatment). In the same way, if intraocular pressure is significantly increased during TTT, choroidal compromise may occur, and loss of the normal cooling mechanism of the choroid may result in overtreatment. Treatment Techniques Preliminary results suggest that TTT is safe for treating subfoveal occult CNV. Data from several centers shows a less than 2 percent chance of severe loss of central vision. I have already mentioned some of the reasons for vision loss in the previous section; in these special cases, making treatment alterations (such as reducing the power setting) may lower the incidence of vision loss. TTT is typically conducted without retrobulbar anesthesia—a topical anesthetic is preferred. The Goldmann lens (diode coated) is commonly used for lesions less than or equal to 3 mm. Wide-field diode coated magnifying lenses can be used also, as long as the power level is adjusted for the size of the spot on the retina. Larger lesions may be treated either with overlapping spots or by using magnifying lenses. The HeNe (helium neon) aiming beam should be at low to moderate intensity. The circle is then bisected by a moderate-to-high slit beam to visualize the retina and the RPE, allowing for concomitant observation of the treatment area and retina. The therapeutic temperature increase is reached in approximately one second.
If retinal color change occurs, treatment should continue, but with power reduced by 20 percent to finish the minute. For a 3-mm spot size, the settings for treating occult CNV (in lightly pigmented fundi) is 800 mW for one minute. With classic CNV, where lesions are smaller and where increased pigmentation and less fluid may be present, you may require a lower power setting and smaller spot sizes. In all cases, intraocular pressure (IOP) must remain stable, since moderate or even mild IOP elevation can significantly reduce choroidal blood flow and inhibit heat dissipation. If indirect lenses are used for treatment, they must not be torqued or tilted; otherwise the spot may become distorted and induce astigmatism, making the spot smaller and oval-shaped and increasing the power density. Retinal burn may result. Several groups have performed studies that suggest that TTT is an effective treatment for CNV. It is hoped that the TTT4CNV study will prove its efficacy for treating occult CNV; other studies are evaluating TTT’s viability for treating classic CNV. Study Parameters A nationwide study involving 22 centers was begun in March 2000 with patients who have symptomatic occult CNV with signs of exudation. Patient recruits have fibrovascular pigment epithelial detachments (PEDs) and late leakages of undetermined source that are smaller than 3 mm. In pilot studies, these early exudative AMD lesions have shown the best response to TTT. In all, 336 patients will be recruited. Accepted patients will have vision between 20/50 and 20/400. Occurrence of severe loss of vision immediately after TTT is unusual and occurs approximately 1 percent of the time. Two-thirds of the eyes in the study will be treated and one-third will receive sham treatment. Patients with serous PEDs and extensive geographic atrophy will be excluded. TTT safety results should be available six months after treatment; the study endpoint is one year after treatment. TTT appears to be a promising technique for treating CNV. For the most part, the procedure spares the neurosensory retina, with associated resorption of intraretinal and subretinal fluid. Approximately two-thirds of eyes treated with TTT have shown stabilized or improved vision over a one-year period. Although severe post-treatment vision loss can occur, it is an unusual complication, and its occurrence rate compares favorably to that of photodynamic therapy (PDT). Dr. Reichel is an associate of the New England Eye Center, Tufts University School of Medicine.
Part 4. Evolution of a Classic Approach for the Treatment of Neovascularization Robert P. Murphy, MD Bert M. Glaser, MD Imaging the choroidal circulation with High-Speed ICG and treating choroidal vessels that modulate blood flow to the lesion. Treating the afferent or feeder vessels supplying retinal and choroidal vascular lesions in an effort to close them was one of the first applications of photocoagulation in humans. In the 1960s, the earliest treatments for proliferative diabetic retinopathy with the ruby laser incorporated focal treatment of the afferent or feeder arterioles of neovascularization of the retina, especially for neovascularization of the disc.1,2 By the 1970s, xenon arc coagulation replaced ruby laser coagulation; however, treatment of the afferent blood supply of vascular lesions remained a mainstay of therapy.3 Also in the 1970s, argon laser treatment began to replace the earlier coagulation techniques, but treatment of the feeder vessel supplying retinal and choroidal vascular lesions remained an important part of the treatment technique.4 Some experts in the field still recommend treating the arterioles feeding retinal neovascularization in sickle cell retinopathy,5 and treatment of the feeder vessel supplying choroidal neovascularization secondary to choroidal rupture.6 In its Manual of Operations and in its published treatment recommendations for treatment of recurrent choroidal neovascularization, the Macular Photocoagulation Study recommended treatment of feeder vessels supplying the neovascularization.7,8Treatment of feeder vessels for recurrent choroidal neovascularization after subfoveal surgery has also be advocated.9 Other examples of treatment of feeder vessels supplying vascular lesions include the treatment of Von Hipple lesions and the neovascularization of Eales' disease. High-Speed ICG Choroidal Angiography Fluorescein angiography is extremely limited in its ability to image the afferent or feeder blood vessels of neovascular lesions of the choroid. Indocyanine Green (ICG) angiography has permitted remarkable improvements in our ability to image the choroidal circulation. The longer excitation and emission wavelengths of ICG reduce scatter and allow better transmission through blood and pigment in the choroid. Improved technology in digital angiography and the development of high-resolution infrared cameras in the last several years have facilitated the spatial and temporal resolution required to image human choroidal vessels. Most significantly, since the feeder vessels fill and empty rapidly (often within two seconds), the ability of high-speed ICG angiography to capture six to 40 frames per second is a great enhancement over older systems with functional capture rates of only one frame every few seconds. The development of computerized subtraction techniques with ICG allowed visualization of feeder vessels in choroidal neovascularization.10 Because of its complexity and the lengthy analysis times required, though, this technique is not well-suited to routine clinical use. The application of high-speed digitized scanning and data processing of ICG-generated images has permitted the analysis of human choroidal blood flow in real time. The combination of high-resolution digital infrared cameras, along with desktop computers capable of managing digital information, has made possible high-speed digital ICG choroidal angiography (HSICG). With this technique, a high-resolution digital video loop of the arteriolar and venous components of the choroid is made available within minutes. The ophthalmologist can evaluate the video loop, playing it forward and backward on a desktop computer at varying speeds until all desired information of the choroidal circulation is obtained. Investigators in Italy and in Israel have published reports of successful treatment of choroidal neovascularization utilizing HSICG.11,12 Our group has also reported on success in treating occult CNV, neovascular pigment epithelial detachments, and posterior polypoidal choroidopathy.13-15 Angiographic Technique Sequential intravenous injection and imaging of both HSICG and fluorescein is essential. The fluorescein angiogram adds important data to the high-speed ICG angiogram by delineating the area of leakage of the neovascularization. ICG is injected in a small volume (approximately 0.3 cc) with a rapid balanced salt solution flush to optimize the resolution of the HSICG. Images are captured at six to 40 frames per second with the ability to play back the sequential imaging as a dynamic (phi-motion) movie. Resolution should be 512 x 512 pixels or higher. Recording of the HSICG study must begin just prior to the entry of ICG into the choroidal circulation. Transit of ICG through the afferent vessel supplying the choroidal neovascularization is rapid. The feeder vessel may be visible for only a few seconds or less. Performing the HSICG demands significantly more skill and knowledge of the angiographer than routine fluorescein angiography. Image Analysis Analyzing dynamic HSICG images is considerably more complex and time-consuming than evaluating the relatively few images in a fluorescein angiogram. The latter requires simple pattern recognition of static images. HSICG requires evaluating dynamic images and making determinations of the timing of sequences of events. In this aspect, it is somewhat similar to the skills required of a cardiologist or neurosurgeon evaluating a dynamic study of the heart or brain circulation. First, the arteriolar filling is distinguished from the venous filling of the normal choroidal circulation. Vascular filling of choroidal occult neovascularization is usually slightly delayed with respect to arteriolar filling of the normal choroid. This permits temporal separation of the normal from the abnormal circulations. The afferent vessel or vessels supplying flow to a neovascular complex necessarily fill before there is filling of the neovascularization. These relationships can be determined by repeated evaluations of the filling cycles. The venous drainage of the neovascularization is then determined and distinguished from the afferent arteriole supply. In this manner, the feeder vessel can be identified. Treatment Technique Small-diameter laser treatment is applied to the extrafoveal portion of the feeder vessel. Laser spot sizes range from 75-200µm. We prefer using 810-nm diode laser treatment, because there is decreased spread of the laser beam, and visible color change of the retina and retinal pigment epithelium can be minimized. However, all currently available wavelengths have been successfully used to close feeder vessels. Millipulsing the 810 diode laser at cycles of approximately 0.1 seconds on, followed by 0.1 seconds off, may aid in the goal of maintaining a thermal threshold deep enough in the choroid to close the feeder vessel without causing visible color change of the neurosensory retina. A total treatment time for each lesion is two minutes or longer. Since the choroidal vessel being treated is not usually visible clinically, and because there is usually no visible clinical endpoint in terms of color change of tissue pigments, it is necessary to assess the response of treatment by obtaining a HSICG at various time points following treatment. Evaluating Results of Treatment The primary goal of treatment is to cause closure or significant attenuation of filling of the feeder vessel previously determined to be one supplying blood flow to the neovascularization. Significant delay in filling of the feeder vessel is often sufficient to cause clinical and angiographic improvement. To determine whether closing the feeder vessel has had a favorable effect on closing the neovascularization, the combined use of fluorescein angiography and HSICG angiography is essential. Decreased amounts or rates of fluorescein leakage and delayed or absent filling of the feeder vessel are favorable responses. Effectively monitoring results of treatment permits early detection of recurrent or new feeder vessels. This mandates the frequent use of HSICG and fluorescein angiography postoperatively in eyes with choroidal neovascularization. Because the feeder vessel requiring treatment is usually not visible, and because its treatment requires great precision with a very small laser spot size, this technique requires the highest degree of photocoagulation skills on the part of the treating ophthalmologist. Careful mapping of retinal and choroidal landmarks and thoughtful analysis of end-point parameters are essential. Treatable Types of Choroidal Neovascularization Predominantly occult CNV with a smaller classic component accounts for the most common form of subfoveal CNV in age-related macular degeneration. Often, separate branches of the afferent or feeder vessel supply the classic and occult portions. Both must be treated to close the CNV. Initial CNV closure can be followed by re-opening of either the treated feeder vessel or of other, less obvious afferent vessels, progressing to clinically significant levels of blood flow. Both must be closed to control the CNV. Repeat examinations with HSICG and fluorescein angiography are essential. CNV composed entirely of the occult (fluorescein) pattern will often have a long, linear feeder vessel that may have a beaded appearance. Resolution of subretinal fluid and visual improvement can occur within days to weeks following closure of the feeder vessels. However, these vessels can re-open within one to two months. Follow-up examination and angiography are essential. Predominantly classic CNV may have one or more feeder vessels providing blood flow. All must be closed to successfully control the CNV. The lesions must be carefully followed with clinical exam and angiography. Other types of choroidal neovascularization may also benefit from this diagnostic and treatment approach. Recurrent CNV following previous confluent ablative laser treatment will often have a prominent feeder vessel entering the neovascularization from a vessel emanating from the previous laser scar. Treatment can usually be applied within the previous scar to close the feeder vessel. Neovascular pigment epithelial detachments will sometimes have a prominent external feeder vessel best seen with 30 degree ICG angiography. These vessels can be quite large and very prominent. Reduction in flow in these can have a favorable effect on the PED and on vision, even if flow is not completely stopped. Resolution of subretinal fluid with improvement of visual acuity can improve within weeks in some cases. Results Feeder vessels can be imaged in more than 75 percent of eyes with choroidal neovascularization. Selective closure of feeder vessels can result in resolution of subretinal fluid and improvement of vision in a significant portion of eyes. Decreasing the blood flow to the CNV usually prevents further growth of the CNV. Complications Since only a tiny area of tissue is being subjected to the thermal effects of the infrared laser, the complications of treatment are minimal. Treatment of the venous drainage vessel of CNV can cause hemorrhage beneath the RPE and retina. This risk is greatly reduced by proper evaluation of the HSICG, and this complication is rare. All forms of neovascularization can progress when the afferent vessels are not adequately closed and even when the visible afferent vessels are closed. New choroidal vessels can bring a new blood supply to the CNV after the initial feeder vessels are closed, permitting progression of the CNV. In rare cases, the fibrous component of the CNV progresses even when the obvious vascular component is controlled. In some cases, the feeder vessels cannot be localized well enough to permit treatment. Feeder vessel treatment of choroidal neovascularization has been a standard part of our treatment arsenal for over 30 years. It has now become more effective because of advances in both diagnosis and treatment. Choroidal vascular imaging, facilitated by HSICG angiography coupled with more effective feeder vessel closure made possible by improved laser technology are responsible for dramatic improvements of a classic approach. Drs. Glaser and Murphy practice in the Glaser Murphy Retinal Treatment Centers in Baltimore and Washington.
Part 5. Photodynamic Therapy: New Hope in the Fight Against Neovascular AMD Edgar L. Thomas, M.D.,Beverly Hills, Calif. This new therapy will help slow vision loss for the many patients who face losing their vision to this debilitating condition. The Treatment Process Photodynamic therapy (PDT) began in the early 1900s when topical dyes were applied to skin tumors, which were then exposed to light, producing necrosis of the lesions. Sophistication increased as newer compounds, such as hematoporphyrin, were found to have a selective affinity for tumor cells, especially the tumor's vascular endothelium. With the advent of lasers, a high level of selectivity could be achieved. This prevented phototherapeutic damage to normal tissue surrounding the lesion. Because of this selective sensitization of the abnormal tumor vascular endothelium, we made the postulation in the early 1980s that the choroidal neovascular endothelium present in the AMD lesions might have a similar response with selective closure preserving overlying retinal photoreceptor cells and pigment epithelium. In a model of CNV in the cynomologous monkey, we demonstrated efficacy of photodynamic enhancement of closure of the experimentally induced CNV. The mechanism might be selective to the neovascular endothelium by an effect on the mitochondrial enzyme systems and possible induction of apoptosis-a programmed pattern of cell death within the CNV. Improvement in safety and efficacy of newer photodynamic drugs has spawned clinical trials. This year, the U.S. Food and Drug Administration approved photodynamic therapy for subfoveal CNV with a predominantly "classic" pattern of CNV for verteporfin (Visudyne-CIBA Vision/ QLT). In classic CNV, the neovascular structure fills early, quickly becomes more hyperfluorescent and progressively leaks until the end. Occasionally, a radial pattern of CNV vessel anatomy can be seen.  SnET2 PDT for CNV has completed enrollment and was afforded "fast track" status by the FDA to allow rapid approval when the data demonstrates a significant benefit over the placebo controls. Evaluation of the early data will be performed this year. Lu-Tex (Alcon/Pharmacyclics) is currently in early Phase III trials and may benefit the patients by offering both an angiographic capability as well as a photosensitizing one. Figure 1. Arrow delineates the superior margin of a predominantly classic subfoveal choroidal neovascularization (CNV) at onset of metamorphopsia. The outer rim of atrophic RPE is the basic underlying disease; the dark area around the CNV is elevated blocked fluorescence indicative a component of sub-RPE CNV. He was offered the potential of photodynamic therapy for the left eye and the potential benefits of non-thermal PDT. Two weeks post treatment, his visual acuity had returned to 20/40. (See case study, page 116.) Patient Selection Criteria The SnET2 Photodynamic Therapy for Age-Related Maculopathy trial completed enrollment of 934 eligible patient eyes in December 1999. The on-going study will continue to follow up and treat patients for two years from the time of treatment/enrollment. The selection of patients eligible for the Phase III clinical trial was based on the Phase I/II open-labeled data, which included AMD, high myopia or an idiopathic etiology. The majority of the patients were AMD in this dose escalation study. The double-masked, placebo-controlled Phase III entry criteria were:
 Because retreatment of a recurrence of CNV is allowed in the Phase
III trial, the treatment zone was enlarged to include recurrent lesions <4.5
mm or 3 DD. The Phase I/II study had demonstrated efficacy and safety for the
two intermediate doses of SnET2 at 0.5 mg/kg and 0.75 mg/kg of body weight.
These two doses were chosen for the Phase III study. The placebo control was
the use of the vehicle intravenously, but not the drug. The randomization
schema for eyes eligible for the study was 2(.50mg/kg): 2(.75mg/kg): 1(vehicle
control). Figure 2. Two weeks post PDT treatment for CNV, the arrow points to the previous superior border of hyperfluorescence that is now absent, the CNV having been occluded by the PDT modality. Eighty percent of the eyes were in the treatment groups and 20 percent were controls. All eyes received light doses of 35 J/cm2 or 60 seconds of exposure to 664 nm irradiance from a diode laser delivery system. FDA and Institutional Review Board approval was obtained for the study. Individual informed consent was obtained from each patient. Clinical Results The Phase I/II open-label study showed very promising results with improvement of vision of ETDRS 1.9 lines in the 0.5mg/kg dose of SnET2 (Pharmacia/Miravant) and 3.5 lines of improvement of the eyes with the 0.75 mg/kg dose at the 90-day follow-up period. Patients studied at six months-not part of the original trial protocol, but looked at retrospectively-showed a persistence of vision remaining at or above baseline in more than 50 percent of cases regardless of the treatment dose used. All patients received the drug in the Phase I/II study. Current Clinical Trials The Phase III SnET2 Photodynamic Therapy for Subfoveal CNV in AMD is completely enrolled with 934 patients from 59 clinical centers in the United States. These patients represent two parallel studies. The demographics of the enrolled patients was presented at the May meeting of the Association for Research and Vision in Ophthalmology. The average age of the patients was 77 years and the duration of known AMD was 2.2 months. The average lesion diameter was 2.35 mm. Visual acuity distribution using ETDRS grading was 17 percent less than 31 letters, 25 percent between 31 and 45 letters, and 58 percent greater than 45 letters. Laser to the fellow eye for CNV from AMD had been performed in 19.5 percent of cases, and no laser had been performed in 80.5 percent of the patients. Current smokers comprised 15.7 percent; previous smokers not currently smoking comprised 47.5 percent; and patients who never smoked were 36.6 percent of the study group. Right eye and left eye distribution was 51.1 percent to 48.9 percent, respectively. Male/female distribution was 42.3 percent and 57.7 percent. All females were post-menopausal. Only 1.9 percent of the patients were non-Caucasians. Practice Impact Our early experience with increasing volumes of patients requiring PDT for CNV in classic AMD eyes has shown a significant increase in patient time within the office. An infusion monitor during the sensitizer injection and a need for immediate treatment at the termination of the infusion places an increased burden on the ophthalmologist to be more readily available to treat at the appropriate time. This means better coordination with the staff and being unable to start new exams or other therapies in close proximity to the end of the infusion. Planning for multiple consecutive PDT patients doesn't seem to solve the problem and may actually produce a significant amount of "down" time during the setup, infusion and delivery. Interspersing the PDT patients with routine follow-up patients seems to work the best with the provision one can extract himselves from the activity of the moment and arrive at the laser at the appropriate time to deliver the photoactivating light. Equipment, Staffing Data from the first approved PDT for CNV in AMD therapy determined a retreatment rate of 3.4 times in the first year and closer to two the second year. Preliminary Phase I/II data from SnET2 (Pharmacia/Miravant/Iridex) showed that following a single SnET2-PDT treatment, more than half of the patients reviewed maintained baseline or better vision compared to entry. The possibility exists for less frequent need of recurrent treatment, though the long-term data will be needed to determine if this is borne out. Control and accountability for the very expensive drugs mean the need for secure drug control and limited access to its storage area. Case Study A 67-year-old white male with AMD and a subfoveal CNV had been unsuccessfully treated with argon laser photocoagulation in the right eye nine months prior to the onset of the subfoveal CNV lesion in the left eye (See Figure 1, page 113.) He had been followed every other month and had noted the onset of metamorphopsia and decreased vision for two weeks prior to the angiogram. His visual acuity had dropped from 20/25 to 20/200 and a moderate serous detachment of the sensory retina was also present. The patient was aware of a significant reduction in metamorphopsia. With subfoveal argon laser photocoagulation, the left eye would not have had the potential vision at this level because of the destructive nature of full-thickness retinal ablation. Certainly, the risk of recurrence is significant, but with the safety margin of PDT for CNV in AMD, he has a potential for further significant responses to PDT should it be necessary in the future. He has had the opportunity to compare early results of PDT. Compared to his experience with the previous eye and argon laser, this eye has demonstrated the powerful effect of the advance in selective closure of subfoveal CNV with PDT in AMD. Dr. Thomas was a pioneer in photodynamic therapy, first suggesting its role for CNV in age-related macular degeneration in the mid-1980s. He is the co-lead investigator for the Pharmacia/Miravant/Iridex SnET2 Photodynamic Therapy Trial for Subfoveal AMD and for the Vitrase for Vitreous Hemorrhage study sponsored by Ista Pharmaceuticals. Click here to complete your participation in this activity. |
||||||||||
