Introduction ( Box 24.1 )
Exfoliation syndrome (XFS) is an age-related disease characterized by the production and progressive accumulation of a fibrillar extracellular material in many ocular tissues. First described in 1917 by Lindberg, it is the most common identifiable cause of open-angle glaucoma worldwide, comprising the majority of glaucoma in some countries. Its incidence increases progressively with age, while its widespread distribution, frequency, and association with other diseases are still poorly understood. The discovery in 2007 of nonsynonymous single-nucleotide polymorphisms (SNPs) in the lysyl oxidase-like 1 ( LOXL1 ) gene is expected to make a major impact not only in understanding XFS, but also in leading to new avenues of therapy.
Exfoliation syndrome is an age-related disease characterized by the production and progressive accumulation of a fibrillar extracellular material in many ocular tissues and is the most common identifiable cause of open-angle glaucoma worldwide, comprising the majority of glaucoma in some countries
Key symptoms and signs
All anterior-segment structures are involved in XFS. The diagnosis is made by finding typical white deposits of exfoliation material (XFM) on the anterior lens surface and/or pupillary border. The diagnosis should be suspected in the absence of XFM when ancillary pigment-related signs are present, which define patients as “exfoliation suspects.”
The classic pattern consists of three distinct zones that become visible when the pupil is fully dilated: a central disc, an intermediate clear zone created by the iris rubbing XFM from the lens surface during its normal excursions, and a granular peripheral zone ( Figures 24.1 and 24.2 ). XFM is often found at the pupillary border ( Figure 24.3 ).
Pigment loss from the pupillary ruff and iris sphincter region and its deposition on anterior-chamber structures are hallmarks of XFS ( Box 24.2 ). As the iris scrapes XFM from the lens surface, the material on the lens causes rupture of iris pigment epithelial cells, with concomitant pigment dispersion into the anterior chamber. This leads to iris sphincter transillumination, loss of the ruff, increased trabecular pigmentation ( Figure 24.4 ), and pigment deposition on the iris surface. Pigment dispersion in the anterior chamber is common after pupillary dilation and may be profuse. Marked intraocular pressure (IOP) rises can occur after pharmacologic dilation, and IOP should be measured routinely in all patients after dilation. XFM may be detected earliest on the ciliary processes and zonules, which are often frayed and broken ( Figure 24.5 ). Spontaneous subluxation or dislocation of the lens can occur.
Pigment loss from the pupillary ruff and iris sphincter region and its deposition on anterior-chamber structures lead to iris sphincter transillumination, loss of the pupillary ruff, pigment dispersion in the anterior chamber after pupillary dilation, and increased trabecular pigmentation, the degree of which has been reported to correlate with the presence and severity of glaucoma
An increasing number of reported systemic associations includes transient ischemic attacks, hypertension, angina, myocardial infarction, stroke, asymptomatic myocardial dysfunction, Alzheimer’s disease, and hearing loss ( Box 24.3 ). Some of these associations have been disputed and there is as yet no clear evidence of increased mortality in patients with XFS, which one might expect with these associations, nor has there been shown a clear-cut association of XFS with a systemic disease with conclusive evidence of a functional deficit caused by the presence of XFS.
An emerging clinical spectrum of associations with cardiovascular and cerebrovascular diseases elevates exfoliation syndrome to a condition of general medical importance. Systemic associations include transient ischemic attacks, circulatory abnormalities, myocardial dysfunction, Alzheimer’s disease, and hearing loss. The discovery of nonsynonymous single-nucleotide polymorphisms (SNPs) in the lysyl oxidase-like 1 (LOXL1) gene should lead to understanding of these associations and to new avenues of therapy
The prevalence of XFS increases steadily with age in all populations. The reported prevalence both with and without glaucoma has varied widely, reflecting true differences due to racial, ethnic, or other as yet unknown factors; the age and sex distribution of the patient cohort or population group examined; the clinical criteria used to diagnose XFS; the ability of the examiner to detect early stages and/or more subtle signs; the method and thoroughness of the examination; and the awareness of the observer. It comprises as much as 50% or more of the open-angle glaucoma in some countries, including Norway, Ireland, Greece, and Saudi Arabia. Previously thought rare in Africa, recent reports indicate that it comprises 25% of open-angle glaucoma in Ethiopia and in South African Zulus. In the USA, it is much more common in Caucasians than in persons of African ancestry, comprising about 12% of glaucoma populations. There are ethnic variations within countries and geographic variations even within adjacent towns in the same area. In central Norway, the prevalence in two adjacent towns (20%) was twice that in a third, adjacent town. In Nepal, XFS was found in 12% of members of one ethnic group, the Gurung, and only 0.24% of non-Gurung of similar ages. Although common in Japan and Mongolia, it is rare in southern China.
The prevalence of XFS in glaucoma patients is significantly higher than in age-matched nonglaucomatous populations. Approximately 25% of XFS patients have elevated IOP and one-third of these have glaucoma. This is approximately six times the chance of finding elevated IOP in eyes without XFS. It has been estimated (Lindberg Society) that clinically detectable XFS affects approximately 60–70 million people, so that there should be some 5–6 million persons with exfoliative glaucoma, or nearly 10% of the world’s glaucoma.
The prognosis of exfoliative glaucoma (XFG) is more severe than that of primary open-angle glaucoma (POAG) ( Box 24.4 ). Patients with XFS are twice as likely to convert from ocular hypertension to glaucoma and, when glaucoma is present, to progress. The mean IOP is greater in normotensive patients with XFS than in the general population and greater in XFG patients at presentation than in POAG patients. At any specific IOP level, eyes with XFS are more likely to have glaucomatous damage than are eyes without XFS. There is greater 24-hour IOP fluctuation, greater visual field loss and optic disc damage at the time of detection, poorer response to medications, more rapid progression, greater need for surgical intervention, and greater proportion of blindness.
The prognosis of exfoliative glaucoma is more severe than that of primary open-angle glaucoma at all levels of presentation and treatment
Treatment of exfoliative glaucoma
The sole focus of therapy in XFG should not be the reduction of IOP. Understanding the mechanisms leading to elevated IOP in XFS could allow us to develop new and more logical approaches to therapy.
Most ophthalmologists approach medical treatment with topical prostaglandin analogs and aqueous suppressants. In addition to lowering IOP, prostaglandin analogs may interfere with the disease process. Latanoprost treatment had a marked effect on the aqueous concentration of transforming growth factor (TGF)-ß 1 , matrix metalloproteinase (MMP)-2, and tissue inhibitor of matrix metalloproteinase-2 (TIMP-2) in XFG patients. Aqueous suppressants do not interfere with the mechanism of pigment liberation and trabecular blockage. Cholinergic agents, on the other hand, not only lower IOP, but by increasing aqueous outflow, should enable the trabecular meshwork to clear more rapidly, and by limiting pupillary movement, should slow the progression of the disease. Aqueous suppressants result in decreased flow through the trabecular meshwork. Treatment with aqueous suppressants may lead to worsening of trabecular function. Pilocarpine 2% qhs can provide sufficient limitation of pupillary mobility without causing visual side-effects. A prospective trial (International Collaborative Exfoliation Syndrome Treatment Study) comparing latanoprost and 2% pilocarpine qhs versus timolol/cosopt for patients with XFS and ocular hypertension or glaucoma has been completed and the data are currently being analyzed ( Box 24.5 ).
Blockage of aqeous outflow by a combination of pigment and exfoliation material in the intertrabecular spaces and juxtacanalicular meshwork is believed to be the major cause of elevated intraocular pressure. Therapy designed to prevent this buildup (directed therapy) is a major goal for the future. At the present time, 2% pilocarpine given at bedtime suffices to produce a 3-mm nonreactive pupil throughout the day and limits release of pigment and exfoliation material
Argon laser trabeculoplasty (ALT) is particularly effective, at least early on, in eyes with XFS. Approximately 20% of patients develop sudden, late rises of IOP within 2 years of treatment. Continued pigment liberation may overwhelm the restored functional capacity of the trabecular meshwork, and maintenance of miotic therapy (again 2% pilocarpine qhs) to minimize pupillary movement after ALT might counteract this. Selective laser trabeculoplasty needs further evaluation as an effective and safe alternative to ALT in the treatment of XFG.
Trabeculectomy results are comparable to those in POAG. Trabeculotomy is also successful. Jacobi et al described a procedure termed trabecular aspiration, designed to improve outflow facility by eliminating the trabecular blockage by pigment and XFM. Deep sclerectomy and similar procedures including a deroofing of Schlemm’s canal are becoming popular choices in some centers owing to the reduced risk profile of nonpenetrating surgery. In one series, XFG patients had significantly better success than POAG patients following deep sclerectomy with an implant. Moreover, phacoemulsification combined with penetrating and nonpenetrating procedures does not seem to influence success rate adversely.
Several lines of evidence, including regional clustering, familial aggregation, and genetic linkage analyses, had previously supported a genetic predisposition to XFS. Recently, a genome-wide association study detected two common SNPs in the coding region of the LOXL1 gene on chromosome 15q24 that were specifically associated with XFS and XFG in two Scandinavian populations from Iceland and Sweden, accounting for virtually all XFS cases. These disease-associated polymorphisms appeared to confer risk of glaucoma mainly through XFS. The combination of alleles formed by the two coding polymorphisms determined the risk of developing XFG, which is increased by a factor of 27 if the high-risk haplotype is present. Individuals carrying two copies of this high-risk haplotype would have a 700-times increased risk of developing XFG. Moreover, these genetic alterations also lead to decreased tissue expression of LOXL1 dependent on the individual haplotype. These genetic findings have been confirmed in populations of European descent in Iowa, New York, Utah, Boston, and Australia. One different SNP and one common SNP have been reported in a Japanese population. The LOXL1 gene variations are not associated with POAG or primary angle closure.
LOXL1 is a member of the lysyl oxidase family of enzymes, which are essential for the formation, stabilization, maintenance, and remodeling of elastic fibers and prevent age-related loss of tissue elasticity. It is involved in cross-linking tropoelastin to mature elastin using elastic microfibrils as a scaffold, thus serving both as a cross-linking enzyme and as a scaffolding element which ensures spatially defined elastin deposition. The functional consequences of the LOXL1 gene variants in XFS are not yet known; however, inadequate tissue levels of LOXL1 could predispose to impaired elastin homeostasis and to increased elastosis. Genetic variation in LOXL1 may be a factor in spontaneous cervical artery dissection, a cause of stroke in younger patients. Reduced LOXL1 levels are also found in patients with varicose veins and venous insufficiency. Overactivity of lysyl oxidase, with localization of the enzyme in blood vessel walls and in plaque-like structures, has been found in Alzheimer’s disease. Mice deficient in LOXL1 develop pelvic organ prolapse secondary to a generalized connective tissue defect, and women with prolapse have reduced mRNA for LOXL1. Marked elastosis with elastic fiber degeneration has been observed in the skin and connective tissue of the lamina cribrosa in XFS eyes. Although further studies correlating the genetic variants and tissue alterations associated with XFS are needed, these new findings already provide a basis for both genetic testing and novel treatment approaches.
Various nongenetic factors, including dietary factors, autoimmunity, infectious agents, and trauma, have also been hypothesized to be involved in the pathogenesis of XFS. Reports dealing with sunlight exposure (ultraviolet radiation) are conflicting. Eskimos are the only people reported to have no XFS, but it is common in Lapps living at the same latitude. Persons living at lower latitudes develop XFS at younger ages, whereas those living at higher altitudes had a greater prevalence in two series but not in a third. In one series, XFS was detected more frequently in eyes with blue irides versus brown irides. Herpes simplex virus type 1 was detected by polymerase chain reaction in 13.8% of iris and anterior capsule specimens of patients with XFS compared to 1.8% of controls. Younger patients have developed XFS after penetrating keratoplasty using buttons from elderly donors. Altogether, it appears that XFS represents a complex, multifactorial, late-onset disease, involving both genetic and nongenetic factors in its pathogenesis.
Pathogenesis of exfoliation syndrome and exfoliative glaucoma
A precise understanding of the pathogenesis of XFS remains elusive. However, the pathologic process in intra- and extraocular tissues is characterized by the progressive accumulation of an abnormal fibrillar matrix, which is the result of either an excessive production or an insufficient breakdown or both, and which is regarded as pathognomonic for the disease based on its unique light microscopic and ultrastructural criteria.
Ultrastructure and composition of exfoliation material
Exfoliation fibers are clearly distinguishable from any other known form of extracellular matrix. Light microscopy reveals XFM to be periodic acid–Schiff (PAS)-positive, eosinophilic, bush-like, nodular, or feathery aggregates on the surfaces of anterior-segment tissues. On transmission electron microscopy, the aggregates consist of randomly arranged, fuzzy fibrils, 25–50 nm in diameter, frequently with 20–25 or 45–50 nm cross-banding. These composite fibers are generally associated with 8–10 nm microfibrils, which resemble elastic microfibrils and which aggregate laterally into mature fibers. However, the microfibrillar core of the complex fibers is usually hidden by a coating of electron-dense amorphous material.
The exact chemical composition of XFM remains unknown. Indirect histochemical and immunohistochemical evidence suggests a complex glycoprotein/proteoglycan structure composed of a protein core surrounded by abundant glycoconjugates, including various glycosaminoglycans (heparin sulfate, chondroitin sulfate, hyaluronan) indicating excessive glycosylation. The protein components contain epitopes of the elastic fiber system, such as elastin, tropoelastin, amyloid-P, and vitronectin. Components of elastic microfibrils, such as fibrillin-1, microfibril-associated glycoprotein (MAGP-1), and the latent TGF-ß-binding proteins (LTBP-1 and LTBP-2), are associated with XFM deposits in intra- and extraocular locations and co-localize with latent TGF-ß 1 on exfoliation fibers.
A recent direct analytical approach using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) showed XFM to consist of the elastic microfibril components fibrillin-1, fibulin-2, and vitronectin, the proteoglycans syndecan and versican, the extracellular chaperone clusterin, the cross-linking enzyme lysyl oxidase, and some other proteins, confirming many of the previously reported immunohistochemical data. Together, these findings support the notion that XFM represents an elastotic material arising from abnormal aggregation of elastic microfibril components interacting with multiple ligands.
Origin of exfoliation material
Ocular XFM is closely associated with the nonpigmented ciliary epithelium, pre-equatorial lens epithelium, iris pigment epithelium, trabecular and corneal endothelia, and virtually all cell types in the iris stroma and vasculature, all showing signs of active fibrillogenesis. Passive distribution of XFM by the aqueous humor may be responsible for abnormal deposits on the central anterior lens capsule, the zonules, the anterior hyaloid surface, vitreous, and intraocular lenses. In extraocular locations, fibers are found in close proximity to connective tissue fibroblasts, vascular wall cells, smooth and striated muscle cells, and cardiomyocytes.
Differential gene expression
With gene expression analyses, XFS tissues contained a number of differentially expressed genes, which were mainly involved in extracellular matrix metabolism and in cellular stress. One set of genes consistently upregulated in anterior-segment tissues comprised the elastic microfibril components fibrillin-1, LTBP-1 and LTBP-2, the cross-linking enzyme transglutaminase (TGase)-2, TIMP-2, TGF-ß 1 , several heat shock proteins (Hsp 27, Hsp 40, Hsp 60), proinflammatory cytokines, apolipoprotein D, and the adenosine receptor (AdoR)-A3. Genes reproducibly downregulated in XFS tissues included TIMP-1, the extracellular chaperone clusterin, the antioxidant defense enzymes glutathione-S-transferases (mGST-1, GST-T1), components of the ubiquitin-proteasome pathway (ubiquitin conjugating enzymes E2A and E2B), several DNA repair proteins (ERCC1, hMLH1, GADD 153), the transcription factor Id-3, and serum amyloid A1.
Together, these findings provide evidence that the underlying pathophysiology of XFS is associated with an excessive production of elastic microfibril components, enzymatic cross-linking processes, overexpression of TGF-ß 1 , a proteolytic imbalance between MMPs and TIMPs, low-grade inflammatory processes, increased cellular and oxidative stress, and an impaired cellular stress response, as reflected by the downregulation of antioxidative enzymes, ubiquitin-conjugating enzymes, clusterin, and DNA repair proteins ( Box 24.6 ).