Choroidal neovascularization (CNV) involves the growth of new blood vessels that originate from the choroid through a break in the Bruch membrane into the sub–retinal pigment epithelium (sub-RPE) or subretinal space. CNV is a major cause of visual loss.
In the history, patients with CNV describe the following:
Physical findings in patients with CNV include the following:
See Clinical Presentation for more detail.
Laboratory studies may be indicated if certain underlying medical conditions, such as pseudoxanthoma elasticum (PXE), are present. Imaging studies include the following:
Fluorescein angiography
FA was once an essential tool in diagnosing and managing CNV. Spectral domain OCT has largely replaced FA as the imaging modality of choice in the management of CNV. Angiographic patterns that have been described for CNV include the following:
According to its location relative to the center of the fovea, CNV has been classified as follows:
Indocyanine green angiography
ICG has a peak absorption and fluorescence in the near infrared range, which allows visualization of choroidal pathology through overlying serosanguineous fluid, pigment, or a thin layer of hemorrhage that usually blocks visualization during FA.
Because ICG is bound tightly to the plasma proteins, less dye escapes from the choroidal circulation, allowing better definition of choroidal vasculature.
Three types of ICG patterns that are assumed to represent CNV may be imaged, as follows:
High-speed or dynamic ICG angiography uses a scanning laser ophthalmoscope that takes up to 32 frames per second. These images are recorded like a movie, and the flow in and out of the vessels can actually be seen. The main use of dynamic ICG angiography is in the identification of CNV feeder vessels that are located in the Sattler layer of the choroid.
Optical coherence tomography
With the advent of anti-VEGF therapy, OCT plays a major role in the management of CNV; despite its many advantages, however, OCT cannot fully replace FA in the management of CNV.
Well-defined CNV is seen as a fusiform thickening of the RPE-choriocapillaris band.
Poorly-defined CNV is seen as a diffuse area of choroidal hyperreflectivity that blends into the normal contour of the normal RPE band.
In CNV, a normal boundary between the choriocapillaris and the RPE cannot be defined.
A subretinal hemorrhage is seen as a layer of moderate reflectivity that elevates the neurosensory retina and causes optical shadowing, resulting in a lower reflectivity of the underlying RPE and choroid.
A serous RPE detachment is characterized by complete shadowing of the underlying structures.
A hemorrhagic RPE detachment shows a moderately reflective layer beneath the detached RPE.
A fibrovascular RPE detachment demonstrates moderate reflectivity throughout the entire sub-RPE space under the elevation.
Detachments of the neurosensory retina appear as elevations of a moderately reflective band above the RPE band.
RPE tears can be seen as thick elevated areas of high reflectivity; the underlying choroid is completely shadowed, whereas the adjacent choroid reveals a hyperreflective image because of the absence of RPE.
Retinal edema or thickness can be measured objectively by defining the anterior and posterior borders of the retina.
A proposed classification scheme of CNV following photodynamic therapy (PDT) is as follows[1] :
See Workup for more detail.
Anti-VEGF treatment counters angiogenesis and increased vascular permeability; accumulation of subretinal fluid secondary to increased permeability is an important component of decreased vision in CNV.[2]
The major limitation of anti-VEGF treatment is the injection burden. Most patients require multiple injections. Therefore, a number of different protocols are looking at combining photodynamic therapy, corticosteroids, and anti-VEGF drugs.[3, 4, 5, 6, 7]
Currently, the treatment of choice for CNV secondary to exudative age-related macular degeneration (ARMD) is intravitreal anti-VEGF therapy.
Intravitreal ant-VEGF agents used for the treatment of CNV include the following:
Other treatment approaches
See Treatment and Medication for more detail.
Choroidal neovascularization describes the growth of new blood vessels that originate from the choroid through a break in the Bruch membrane into the sub–retinal pigment epithelium (sub-RPE) or subretinal space. Choroidal neovascularization (CNV) is a major cause of visual loss.
Mechanisms of CNV are not well understood. Virtually any pathologic process that involves the RPE and damages the Bruch membrane can be complicated by CNV. CNV may be considered as a wound healing response to an insult of the RPE. A protein derived from the RPE, pigment epithelium derived factor (PEDF), was found to have an inhibitory effect on ocular neovascularization. Another peptide, vascular endothelium growth factor (VEGF), is a well-known ocular angiogenic factor.
The balance between antiangiogenic factors (eg, PEDF) and angiogenic factors (eg, VEGF) is speculated to determine the growth of CNV. The cause of VEGF upregulation in CNV remains unclear. VEGF upregulation is known to occur secondary to hypoxia, high glucose and protein kinase c activation, advanced glycation end products, reactive oxygen species, activated oncogenes, and a variety of cytokines.
VEGF has been temporally and spatially correlated with the development of CNV. Histopathologic specimens obtained from submacular surgery reveal the presence of VEGF in CNV. In addition, several researchers have induced CNV formation in animal models by overexpressing VEGF. Once secreted, VEGF binds to its tyrosine kinase receptors in endothelial cells activating several signal transduction pathways. Activation of VEGF induces vascular permeability, endothelial cell proliferation, and cell migration. The end product is the formation of a network of new vessels. These new vessels were previously thought to occur secondary to angiogenesis.
Angiogenesis can be defined as the growth of new vessels from preexisting vessels. Recent evidence suggests that CNV forms from both angiogenesis and vasculogenesis.[9] Vasculogenesis may be defined as the de novo growth of new blood vessels.
In an experimental model of CNV, it has been estimated that up to 20% of endothelial cells are bone marrow–derived progenitor cells that have been mobilized from the bone marrow.[10] These endothelial progenitor cells join the activated endothelial resident cells and incorporate into the nascent vascular tubular structure. The inhibition of endothelial progenitor cells mobilization from the bone marrow significantly reduced the size of the CNV lesion.[11, 12] Migration of endothelial cells requires remodelling of the extracellular matrix. Integrins and metalloproteinases play an important role at this stage. With time, vascular maturation and stability is achieved. Vascular maturation is intimately associated with platelet-derived growth factor (PDGF)-BB, which recruits pericytes to the new vessels.
As new choroidal blood vessels grow, they may extend into the sub-RPE space (Gass type 1) or into the subretinal space (Gass type 2). The location, growth pattern, and type (1 or 2) of CNV depend on the patient's age and the underlying disease. Bleeding and exudation occur with further growth, accounting for the visual symptoms. Alternatively, abnormal blood vessels may originate from an intraretinal location and grow into the subretinal space. This pattern of growth has been named retinal angiomatous proliferation (RAP), or type 3 neovascularization.[13]
In the Wisconsin Beaver Dam Study, prevalence of CNV associated with age-related macular degeneration (ARMD) was 1.2% in adults aged 43-86 years.[14] Myopia is the second most common cause of CNV in the United States and Europe. CNV is estimated to occur in 5-10% of myopes; 60-75% of these are subfoveal.
Disciform scars secondary to CNV from presumed ocular histoplasmosis syndrome (POHS) were present in 0.1% of people living in endemic areas. In multiple evanescent white dot syndrome (MEWDS), development of CNV is rare. In multifocal choroiditis, estimates of CNV range from 25-40% of patients. In punctate inner choroidopathy (PIC), 33% of patients develop CNV. Of these, 50% are subfoveal and result in visual acuities between 20/80 and 20/200.
CNV occurs in 5% of patients with birdshot chorioretinopathy. CNV occurs in virtually all choroidal ruptures during the healing phase; most involute spontaneously. In 15-30% of patients, CNV may recur and lead to a hemorrhagic or serous macular detachment with concomitant visual loss.
ARMD is the most common cause of visual loss in people older than 50 years in the developed world. Up to 90% of visual loss in ARMD is secondary to CNV.
Myopia is the seventh greatest cause of registered blindness in the United States and Europe. CNV is responsible for most of this visual loss.[15, 16]
POHS is an uncommon cause of visual loss. Incidence and prevalence in the blind of Tennessee, an area endemic for histoplasmosis, were reported to be 2.8% and 0.5%, respectively.[17]
No gender predilection exists.
Certain diseases (ie, choroidal ruptures, angioid streaks, myopic macular degeneration, multifocal choroiditis, PIC, MEWDS) that may be complicated by CNV have gender proclivity.
CNV is associated with multiple ocular conditions, so the age distribution of CNV reflects the underlying condition.
For instance, younger patients are affected with POHS, multifocal choroiditis, MEWDS, and PIC.
Older patients will be affected by CNV secondary to ARMD.
Location, growth pattern, type (1 or 2) of CNV, and underlying condition determine the prognosis.
The 5-year follow-up of the MPS ARMD study showed that 46% of eyes with extrafoveal CNV that were photocoagulated had severe visual loss (loss of 6 lines or more from baseline) compared to 42% of eyes that were followed. In the juxtafoveal study, 49% of treated eyes compared to 58% of observed eyes had severe visual loss. In the subfoveal study, 22% of treated eyes and 47% of observed eyes had severe visual loss.[18, 19, 20, 21]
The 5-year follow-up of the MPS POHS study has shown that 12% of eyes with extrafoveal CNV that were photocoagulated had severe visual loss compared to 42% of eyes that were followed.[18] In the juxtafoveal study, 12% of treated eyes also had severe visual loss compared to 28% of eyes that were followed. The natural history of untreated subfoveal CNV secondary to POHS shows that 14-23% of patients retain 20/40 or better visual acuity. Pilot studies of photocoagulation of subfoveal CNV secondary to POHS were inconclusive. Photocoagulation of peripapillary CNV secondary to POHS reduced severe visual loss from 26% of control eyes to 14% of treated eyes.
The 5-year follow-up of the MPS idiopathic CNV study showed that 27% of eyes with extrafoveal CNV that were photocoagulated had severe visual loss compared to 44% of eyes that were followed.[18] Few eyes were entered in the juxtafoveal study; at the 3-year follow-up, 10% of treated eyes versus 27% of observed eyes had severe visual loss. The natural history for idiopathic subfoveal CNV is not necessarily associated with profound loss of vision. Size of CNV seemed to be an important prognostic factor. If the CNV was smaller than 1 disc area at initial examination, prognosis was better. In the MPS, the median visual acuity for untreated extrafoveal and juxtafoveal idiopathic CNV was 20/80.
CNV in myopia does not necessarily imply bad prognosis. Up to 50% of patients can expect spontaneous improvement or stabilization of vision. About 25% of extrafoveal CNV becomes subfoveal, but up to 25% of subfoveal CNV involutes spontaneously.
Once diagnosed with maculopathy secondary to CNV, the patient is asked to self-monitor each eye with a near card and an Amsler grid.
If a disturbance is detected, prompt examination is encouraged.
History may include the following:
See the list below:
Virtually any pathologic process that involves the RPE and damages the Bruch membrane can be complicated by CNV.
In a study examining the relationship of smoking to CNV secondary to presumed ocular histoplasmosis syndrome (POHS), Chheda et al found that the risk of smokers having CNV secondary to POHS is 3 times higher than that of nonsmokers. The risk increased with age and decreased with increasing level of educational information.[22]
Laboratory studies may be indicated if certain underlying medical conditions, such as pseudoxanthoma elasticum (PXE), are present.
Fluorescein angiography (FA) is an essential tool in diagnosing and managing CNV. Several angiographic patterns have been described for CNV.
A lesion that hyperfluoresces in the early phases of the angiogram, maintains well-demarcated borders, and leaks late (obscuring its borders) is a classic CNV.
A lesion whose borders cannot be determined by FA is an occult CNV. Fibrovascular pigment epithelial detachment (PED) and late leakage of undetermined source (LLUS) represent patterns of occult CNV. A fibrovascular PED is a lesion that is elevated solidly and hyperfluoresces irregularly to different degrees. The lesion may be well demarcated or poorly demarcated. LLUS is seen during FA as an irregular, indistinct, late, sub-RPE leakage.
According to its location relative to the center of the fovea, CNV has been classified as extrafoveal (200-1500 µm), juxtafoveal (1-199 µm), and subfoveal.
Indocyanine green (ICG) is a water-soluble tricarbocyanine dye that contains 5% sodium iodide; it rapidly binds almost completely to globulins after intravenous injection. ICG has a peak absorption and fluorescence in the near infrared range. This allows visualization of choroidal pathology through overlying serosanguineous fluid, pigment, or a thin layer of hemorrhage that usually blocks visualization during FA. Because ICG is bound tightly to the plasma proteins, less dye escapes from the choroidal circulation, allowing better definition of choroidal vasculature.
Three types of ICG patterns that are assumed to represent CNV may be imaged. A hot spot is a well-defined focal hyperfluorescent area that is less than one disc area in size. Hot spots usually fluoresce early. A plaque refers to a hyperfluorescent lesion that is larger than one disc area in size. A plaque usually does not fluoresce early, and its intensity diminishes late. Finally, some eyes harbor a combination of plaques and hot spots. In these eyes, the hot spots may be at the edge of the plaque, may overlie the plaque, or may be far from the plaque.
High-speed or dynamic ICG angiography uses a scanning laser ophthalmoscope that takes up to 32 frames per second. These images are recorded like a movie, and the flow in and out of the vessels can actually be seen. The main use of dynamic ICG angiography is in the identification of CNV feeder vessels that are located in the Sattler layer of the choroid.
CNV causes thickening and fragmentation of the highly reflective RPE-choriocapillaris band. If the CNV is well defined, it is seen as a fusiform thickening of the RPE-choriocapillaris band. In contrast, poorly defined CNV is seen as a diffuse area of choroidal hyperreflectivity that blends into the normal contour of the normal RPE band. A normal boundary between the choriocapillaris and the RPE cannot be defined.
A subretinal hemorrhage is seen as a layer of moderate reflectivity that elevates the neurosensory retina and causes optical shadowing, resulting in a lower reflectivity of the underlying RPE and choroid. Serous, hemorrhagic, or fibrovascular RPE detachments reveal focal RPE elevations with shadowing of the structures beneath the elevated areas. Serous detachments are characterized by complete shadowing of the underlying structures. A hemorrhagic RPE detachment shows a moderately reflective layer beneath the detached RPE. Fibrovascular RPE detachments demonstrate moderate reflectivity throughout the entire sub-RPE space under the elevation.
Detachments of the neurosensory retina appear as elevations of a moderately reflective band above the RPE band. RPE tears can be seen as thick elevated areas of high reflectivity. The underlying choroid is completely shadowed, whereas the adjacent choroid reveals a hyperreflective image because of the absence of RPE. Retinal edema or thickness can be measured objectively by defining the anterior and posterior borders of the retina.
Rogers and coworkers have proposed an optical coherence tomography (OCT) classification scheme of CNV following photodynamic therapy (PDT).[1]
Stage I occurs shortly after PDT and lasts for about a week. It is characterized by an inflammatory reaction that causes an increase in intraretinal fluid in a circular fashion that corresponds with the treatment spot.
Stage II represents the restoration of a near-normal foveal contour with diminished subretinal fluid occurring 1-4 weeks after treatment.
Stage III represents reperfusion and involution of CNV. It typically occurs 4-12 weeks following treatment and is subdivided into 2 categories based on the ratio of subretinal fibrosis to fluid present. Stage IIIa contains a greater subretinal fluid to fibrosis ratio, indicating active CNV. Lesions in Stage IIIb have more prominent fibrosis with minimal intraretinal fluid, indicating inactive CNV.
Further involution of CNV may lead to cystoid macular edema, signifying Stage IV.
In Stage V, CNV and the subretinal fluid resolve, leading to fibrosis and retinal thinning.
Despite the many advantages of OCT, FA remains the imaging modality of choice in the management of CNV. Currently, OCT cannot replace FA in the management of CNV.
With the advent of anti-VEGF therapy, OCT plays a major role in the management of CNV. Most clinicians use the presence of fluid on the OCT scan as an indication of CNV activity and the need for further treatment.
Optical coherence tomography angiography (OCT-A) is a noninvasive imaging modality that visualizes the vascular anatomy of the fundus by detecting motion contrast between repeated OCT B-scans at the same site. Since the only moving objects in the fundus are the erythrocytes, the images produced represent a map of blood flow. OCT-A is depth resolved and permits differentiation of the different capillary plexuses.
OCT-A has several limitations, including a limited field of view, image artifacts, and inability to demonstrate leakage. Current en face OCT-A suffers from 2 major problems: flattening of the depth information within a given layer, causing loss of the vascular interrelationships within the layer, and the need for segmentation. In many eyes with pathologic changes, accurate segmentation is not possible because the layers cannot be readily identified.[23]
Since OCT-A does not require dye injection, it is much safer than fluorescein or indocyanine green angiography. Intravenous access is unnecessary, and there is no risk of anaphylaxis. With the current technology, the sensitivity of OCT-A for CNV detection varies from 50%-100%.[24, 25]
New capillaries and fibroblasts originate from the choroid and grow through a defect in the Bruch membrane into the subretinal space (type 2 CNV) or the sub-RPE space (type 1 CNV). Reactive hyperplastic RPE is present at the advancing edge of CNV.
Specimens obtained from surgical excision of CNV reveal that the most common cellular components are vascular endothelium and RPE. These were present in more than 85% of samples. Fibrocytes and macrophages also have been identified in more than 50% of specimens. Extracellular components include collagen and fibrin. VEGF has been identified in the specimens obtained during submacular surgery.
Current knowledge of molecular events in the pathogenesis of choroidal neovascularization (CNV) has allowed CNV to be targeted with very specific antiangiogenic factors. Targeting VEGF allows a two-hit strategy: antiangiogenesis and antipermeability. VEGF is 50,000 times more potent than histamine in inducing vascular permeability. An important component of decreased vision is the accumulation of subretinal fluid secondary to increased vascular permeability.[2]
The major limitation of anti-VEGF treatment is the injection burden. Most patients require multiple injections. Therefore, a number of different protocols are looking at combining photodynamic therapy, corticosteroids, and anti-VEGF drugs.[3, 4, 5, 6, 7]
Currently, the treatment of choice for CNV secondary to exudative ARMD is intravitreal anti-VEGF therapy. A reduced biological response to both intravitreal ranibizumab and bevacizumab has been reported by several authors.[26, 27, 28, 29, 30, 31] A distinction between tachyphylaxis and drug tolerance should be made.[32] Tachyphylaxis refers to the loss of drug effectiveness following repetitive use during a short period of time. In general, drug effectiveness is restored after a short drug holiday. In contrast, drug tolerance develops slowly over time. Increasing the drug dosage or shortening the dosing interval improves its effectiveness. A drug holiday does not restore its effectiveness.[32]
Several mechanisms have been proposed to explain these phenomena. VEGF blockade may lead to an increase in other angiogenic signaling pathways as a compensatory mechanism.[33] Up-regulation of VEGF production by macrophages within CNV has also been proposed.[28] Anti-bevacizumab and anti-ranibizumab auto-antibodies have been documented in the systemic circulation of patients undergoing chronic anti-VEGF therapy for exudative AMD. These auto-antibodies may neutralize the effect of anti-VEGF agents.[28] } CNV lesion composition might change with time with more mature and therefore less VEGF sensitive vessels.[28, 33]
A retrospective case series reported that tachyphylaxis occurred in 5 of the 59 patients treated with intravitreal bevacizumab.[28] In this study, the median time to develop tachyphylaxis with intravitreal bevacizumab was 100 weeks, with a median number of 8 intravitreal injections. Another retrospective case series identified tachyphylaxis in 2% of patients being treated with ranibizumab.[26]
Several strategies, including drug holidays, increasing the drug dosage, combination therapy, and switching from one anti-VEGF drug to another anti-VEGF agent, have been advocated to counteract these phenomena.[28, 29, 31, 32] Gasperini and colleagues showed that in 81% of cases, the switch from ranibizumab to bevacizumab and vice versa was at least somewhat effective in further reducing subretinal fluid.[29]
Several other antiangiogenic compounds are currently in different stages of development.[34] These agents include genetic therapy with vectors carrying anti-angiogenics,[35] si (small interference) RNA-VEGF, various PDGF inhibitors, angiopoietin inhibitors, and combretastatin A4.
Pegaptanib sodium [8]
Pegaptanib sodium is an aptamer against VEGF165, the isoform identified with pathological angiogenesis. An aptamer is an oligonucleotide that acts like a high affinity antibody to VEGF, neutralizing it before it can contact its receptor.
Pegaptanib sodium is given as an intravitreal injection every 6 weeks.
Overall, pegaptanib sodium was able to decrease visual loss when compared to placebo in a similar fashion to that of PDT therapy with verteporfin. Only 6% of eyes were reported to have an improvement in visual acuity of 3 or more lines after 12 months of follow-up. Unlike therapy with verteporfin, all eyes with exudative ARMD benefited from treatment regardless of lesion composition. In addition, the trials using pegaptanib sodium included eyes with larger lesions than those eyes in the trials using verteporfin.[36]
Complications associated with the intravitreal injection of pegaptanib sodium are few but include retinal detachment and endophthalmitis.
Ranibizumab is a recombinant monoclonal antibody Fab fragment that neutralizes all active forms of VEGF-A.
Ranibizumab is delivered as a monthly intravitreal injection.
The US Food and Drug Administration approved the use of ranibizumab for the treatment of all angiographic subtypes of subfoveal neovascular ARMD.
Intravitreal ranibizumab is the first treatment that significantly improves visual acuity in up to 40% of eyes.[37] An extension study of patients who completed 1 of 3 different randomized clinical trials of ranibizumab for exudative age-related macular degeneration showed that intravitreal injections of ranibizumab were well tolerated for more than 4 years. However, less frequent follow-up led to fewer injections, which in turn led to a loss of the initial gains in visual acuity.[38]
Although infrequent, complications associated with this treatment include endophthalmitis and severe uveitis.
Bevacizumab is a humanized, recombinant monoclonal immunoglobulin G (IgG) antibody that binds and inhibits all VEGF isoforms and is currently approved for systemic use in metastatic colorectal cancer and non–small cell lung cancer.
Off-label use of intravitreal bevacizumab for CNV secondary to ARMD was first reported in 2005. Most of the reports of bevacizumab are uncontrolled, open-label case series that have suggested functional and anatomical efficacy, short-term safety, and low cost.
Results from several studies suggest that bevacizumab may be useful in the treatment of CNV secondary to multiple etiologies including myopia,[39] angioid streaks,[40] inflammatory conditions,[41, 42] and ARMD.[43, 44]
A retrospective study reported findings in 180 patients with choroidal neovascularization secondary to age-related macular degeneration who were injected with either 1.5 mg or 2.5 mg and were followed for a minimum of 24 months.[45] An average of 5 injections using a PRN protocol demonstrated improvement or stability in vision. No statistically significant differences between doses were noted.
The VIEW 1 and VIEW 2, two similarly designed double-masked, randomized multicenter clinical trials, demonstrated that intravitreal aflibercept dosed monthly or every 2 months after a loading dose of 3 monthly doses was noninferior to monthly ranibizumab.[46] The US Food and Drug Administration has approved aflibercept for the treatment of all angiographic subtypes of subfoveal neovascular ARMD.
Brolucizumab (RTH258, ESBA 1008) is the newest anti-VEGF agent developed. It is the smallest, with a molecular weight of 26 kDa. It is a humanized single-chain antibody fragment against all VEGF isoforms. Recent clinical trials demonstrated that brolucizumab compared favorably to aflibercept in the management of CNV due to exudative age-related macular degeneration.[47] Brolucizumab is expected to gain FDA approval in the near future.
The Macular Photocoagulation Study (MPS) proved the efficacy of laser photocoagulation in the treatment of CNV secondary to ARMD, POHS, and idiopathic causes.
The goal is to completely obliterate CNV.
Partial treatment of CNV is not beneficial when compared to observation. Extrapolate these results to other conditions that are complicated by CNV on a case-by-case basis. Many patients and their physicians choose not to elect immediate loss or several lines of vision in an attempt to have a very modest visual improvement in 18 months. Extrapolation of MPS results to CNV secondary to myopia probably is not indicated in juxtafoveal CNV. Cases of enlargement of laser scars through the fovea with subsequent visual loss have been reported.
PDT uses light-activated drugs and nonthermal light to achieve selective destruction of CNV with minimal effects on the surrounding normal tissues. Randomized clinical trials have shown that PDT with verteporfin is effective in reducing visual loss in certain eyes with CNV secondary to ARMD. In eyes with at least some classic CNV, treatment with verteporfin reduced visual loss. Subgroup analysis revealed that eyes with a classic component of greater than 50% fared much better than those eyes with a classic component of less than 50%. In eyes with a classic component of less than 50%, no difference existed in visual loss between the eyes treated with placebo and the eyes treated with verteporfin.[48] Another study reported that therapy with verteporfin for occult CNV secondary to ARMD was effective in slowing the progression of visual loss. However, such benefit was only seen after the second year of follow-up. Subgroup analysis revealed that eyes with a visual acuity of 20/50 or worse or eyes with lesions smaller than 4 disc areas in size had a better outcome. Further analysis of the data revealed that lesion size rather than lesion composition is a strong predictor of visual benefit following PDT with verteporfin.[49] Despite all the encouraging initial results, PDT provides marginal benefit. Most eyes will continue losing vision, though at a slower rate, and only 15% of eyes will manifest some visual improvement. PDT in combination with intravitreal triamcinolone, bevacizumab, or ranibizumab may have better visual outcomes than PDT alone in patients with ARMD.
High-speed ICG confocal angiography guided laser photocoagulation of feeder vessels is reportedly beneficial in selected patients with exudative ARMD but remains unproven.
Uncontrolled studies have recommended surgical excision of subfoveal CNV via pars plana vitrectomy. The goal is to remove CNV but to leave the underlying RPE and choriocapillaris intact. Surgical excision of type 2 CNV would be more beneficial than type 1 CNV. Pilot studies resulted in substantial numbers of patients with worse vision, many with unchanged vision, and a small number with apparent improved vision. The current rhetoric is that stabilization may occur with surgery.[50] The Submacular Surgery Trial (SST), a randomized multicenter prospective trial sponsored by the National Eye Institute (NEI), confirmed that submacular surgery in eyes with CNV secondary to ARMD generally does not have a good visual outcome.[51] In addition, with CNV secondary to idiopathic causes and POHS, submacular surgery offers a modest benefit in eyes with a baseline visual acuity of 20/100 or worse.
Two surgical methods to translocate the fovea have been developed to treat subfoveal CNV. The previously subfoveal CNV is now juxtafoveal or extrafoveal; then, standard laser photocoagulation or PDT can be performed without damaging the fovea. Caution is warranted because high rates of retinal detachment, proliferative vitreoretinopathy (PVR), macular holes, recurrent CNV, cystoid macular edema (CME), and hemorrhage have been reported.
Low-dose radiation therapy has been effective in inhibiting neovascularization in different tissues. A randomized clinical trial reported better visual outcomes in eyes with exudative ARMD receiving radiation therapy of 24 Gy given in 6 fractions of 4 Gy each compared to observation.[52] However, other trials do not support radiation therapy as a treatment alternative in eyes with CNV secondary to ARMD. Long-term effects are unknown, and radiation retinopathy is definitely a concern.
Diagnosis and treatment is often difficult. Consider referring to a retinal specialist who is experienced with these conditions.
After 5 years of follow-up, the MPS reported that 55% of patients with exudative ARMD, 33% of patients with POHS, and 34% of patients with idiopathic CNV had a recurrent or persistent CNV after laser photocoagulation.[18] These recurrences, regardless of etiology, tended to be toward the foveal side and were associated with visual loss. In most cases, photocoagulation of these recurrent CNV is indicated. Laser treatment of peripapillary CNV may be complicated by thermal damage to the papillomacular bundle.
Surgical excision of CNV may be complicated by retinal detachment, postvitrectomy cataract, choroidal hemorrhage, epimacular membrane, and macular hole. CNV recurrence following excision occurred in up to 44% of cases. How to effectively manage these recurrences is unclear.
The Age-Related Eye Disease Study (AREDS) has shown that eyes with a high risk of developing CNV secondary to age-related macular degeneration benefit from antioxidants and zinc. The recommended daily dosages are 500 mg of vitamin C, 400 IU of vitamin E, 15 mg of beta carotene, 80 mg of zinc oxide, and 2 mg of cupric oxide.[53] Smokers should not take the AREDS formulation because beta carotene has been associated with an increased risk of developing pulmonary neoplasia.
Two weeks following laser photocoagulation, the patient should be observed and undergo fluorescein angiography. Pay special attention to the borders, especially the foveal border, of the laser treatment zone to detect any persistence. If no leakage is detected, the patient should have another fluorescein angiogram 4 weeks later or if there are new changes on the Amsler grid. If no leakage is detected again, another angiogram should possibly be obtained 4-6 weeks later.
Clinical examination cannot replace FA during the first 18 months after laser treatment, because most persistent and recurrent leakage occurs during this period.
Decreased vision with choroidal neovascularization is caused in part by the accumulation of subretinal fluid secondary to increased vascular permeability. Anti-VEGF agents arrest the neovascularization and reduce vision loss.
Clinical Context: A benzoporphyrin derivative monoacid (BPD-MA), consists of equally active isomers BPD-MAC and BPD-MAD, which can be activated by low-intensity, nonthermal light of 689-nm wavelength. After activation with light and in presence of oxygen, verteporfin forms cytotoxic oxygen free radicals and singlet oxygen. Singlet oxygen causes damage to biological structures within range of diffusion. This leads to local vascular occlusion, cell damage, and cell death. In plasma, verteporfin is transported primarily by low-density lipoproteins (LDL). Tumor and neovascular endothelial cells have increased specificity and uptake of verteporfin because of their high expression of LDL receptors. Effect can be enhanced by use of liposomal formulation.
Reduction of leakage from abnormal, neovascular vessels, resulting in reduced visual loss.
Clinical Context: A recombinant fusion protein consisting of portions of human VEGF receptors 1 and 2 extracellular domains fused to the Fc portion of human IgG1. Inhibits all VEFG isoforms and placental growth factor.
Clinical Context: Selective VEGF antagonist that promotes vision stability and reduces visual acuity loss and progression to legal blindness. VEGF causes angiogenesis and increases vascular permeability and inflammation, all of which contribute to neovascularization in age-related wet macular degeneration.
Clinical Context: Ranibizumab is a recombinant monoclonal antibody Fab designed to bind and inhibit VEGF-A, a protein that is believed to play a critical role in the formation of new blood vessels of exudative ARMD. First approved treatment with visual improvement for exudative ARMD.
Clinical Context: A nonspecific monoclonal anti-VEGF. Off-label drug with apparent similar efficacy of ranibizumab.