CONTROVERSY CONCERNING THE GEOGRAPHIC VARIATION OF EXFOLIATION SYNDROME (XFS)
A detailed account of the history, controversy, and wealth of available published information concerning the epidemiology of XFS is beyond the scope of this brief review. The reader is referred to a number of pertinent reviews on the subject.1–16 The present review focuses on a number of key publications17–25 and emphasizes up-to-date published reports on the subject.26–30 The following description outlines the various stages of development in the understanding of the topic. From a historical perspective the epidemiology of XFS and that of the ensuing glaucoma named exfoliative glaucoma (XFG) is one of the most controversial subjects in ophthalmic literature. Controversy has arisen concerning both the epidemiological and the clinical features of the condition1,3,6,8,10 and as we know even today the nomenclature of the condition remains debatable.4,7,8,14 Another historical problem is that that anecdotal findings on the prevalence of the condition reported by eminent ophthalmologists have arguably been accorded undue weight.
In the early literature the reported prevalence of XFS and XFG varied greatly in different ethnic groups.1,3,6 The condition was considered common in some countries (Finland, Norway, Russia, Greece) and rare in others (Sweden, Denmark, Britain, Germany, and the United States). In 1949, Thomassen31 investigated 2 glaucoma cohorts after dilatation of the pupil and gonioscopy. In the University Department of Ophthalmology in Oslo, Norway 45 of 57 glaucoma cases (79%) exhibited XFG. In Moorfields Hospital, London, England only 1 of 50 glaucoma cases (2%) was reported as showing evidence of exfoliation.31 In 1960, the German ophthalmologist Leydhecker32 claimed that he could not find 1 single case of XFG among 500 German glaucoma patients in Bonn.
In the literature of the late sixties and early seventies, the notion that XFS prevalence varied greatly between different countries came under closer scrutiny especially by the Norwegian ophthalmologist Aasved. Aasved33–35 challenged the prevailing concept of his time that XFS exhibited considerable geographic variation. First, he refuted the evidence presented to that date on the prevalence of XFS as unrepresentative, since it was not based on random population samples. Writing in 1971, Aasved34 suggested that he could find only 7 population studies in the literature; of these only 2 had screened >600 individuals. Motivated by disbelief in the reported differences in the prevalences of XFS in neighboring European countries, Aasved carried out a personal, prospective population survey using identical diagnostic criteria on the prevalence of XFS in 3 population cohorts over the age of 60 in Bergen, Norway; Bonn, Germany and Birmingham, England. He established that the prevalence of XFS was similar in the 3 ethnic groups—Bergen, 6.3%; Bonn, 4.7%; Birmingham, 4.0%.33,34 This is remarkable considering that one of these ethnic cohorts (in Bonn) was from the same location where Leydhecker in his retrospective study32 did not identify a single case with XFG in 500 chronic glaucoma patients (in whom presumably the prevalence of the condition would have been expected to be higher). Therefore, Aasved provided convincing epidemiological evidence to suggest that the prevalence of the condition in population groups is first, much higher than previously thought and second, similar in all geographic areas.33–35 Subsequent studies have either supported or contradicted Aasved’s view.
Ever since the publication of Aasved’s studies, a key question concerning the epidemiology of XFS is why are reported prevalence rates so variable? Addressing the subject, Ritch9,12 concluded that this reflects a combination of true differences caused by racial, ethnic, or other as-yet-unknown reasons; 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 of the condition; the thoroughness of the examination; and the awareness of the observer. Overall, cumulative published evidence seems to suggest that the prevalence of the condition varies significantly even within the same country and remarkable differences exist between and within various ethnic groups.3,24,36–42 However, none of the published studies to date is strictly comparable, or of similar scale to those of Aasved. Therefore, Tarkkanen’s assertion,43 inspired by Aasved’s work, that the prevalence of XFS is similar in those countries that population-based surveys have been conducted by the same investigators requires further confirmation. Although today there is compelling evidence6,7,8,13 that supports the view of significant geographic variation of XFS within and between ethnic groups, only a future, large prospective study conducted by the same investigators, using the same diagnostic methodology in similarly selected diverse ethnic populations will prove this hypothesis beyond doubt.
DO WE KNOW THE TRUE PREVALENCE OF XFS?
Current literature abounds with studies on the prevalence of XFS in different ethnic populations.5,8,13,18,40 Clearly XFS is a common condition, but its true prevalence is particularly difficult to assess.43,44 Considering the underlying reasons for this clinical problem, XFS is a chronic, asymptomatic disorder with a slow and insidious onset and subtle signs, which are difficult to appreciate clinically. Moreover, on routine examination of the anterior segment of the eye, without dilatation of the pupil, the diagnosis of XFS can be elusive. Examination with the slit lamp after pupillary dilatation is mandatory for the accurate diagnosis/identification of the condition, because the central disc on the anterior lens surface is difficult to appreciate and the peripheral granular zone remains invisible without adequate mydriasis.7,8 Similarly, the diagnostic sensitivity increases when XFS is actively sought by an experienced observer who is fully aware of the complete clinical picture. Although the clinical picture of advanced XFS is well established, remarkably little published evidence exists as to the early changes of the disorder.2,4,8,44 These early changes may be particularly subtle and confined to the mid periphery of the lens. Owing to the subtlety of the diagnostic signs and the poorly defined early stages of development of the disorder many epidemiological studies may have recorded an artificially low prevalence of XFS.3 Erroneously, investigators have tended to tacitly assume that the classic appearance of exfoliation material upon the pupillary margin and the appearance of “sugar frosting” on the anterior lens capsule appear rapidly, fully developed and thereafter undergo little change. This well established description, however, delineates only an advanced stage of XFS following years of gradual evolution. Even today the early stages of XFS remain ill-defined and no longitudinal epidemiological study has dealt with this subject. However, it is clear that a close relationship exists between the development of the classic clinical signs (ie, exfoliation material deposition) and a number of preceding signs comprising pigment release and deposition in ocular tissues.44 Krause et al45 were the first to provide controlled data on the subject of pigment liberation from the iris. They established that patients with XFS exhibited significantly more dense pigmentation gonioscopically compared with normal subjects. They interpreted this finding to be consistent with increased pigmentation being the first stage of the disorder.45 Since then it has become well recognized that this process gives rise to a number of characteristic pigmentary diagnostic features.46–48 There is now convincing evidence to suggest that pigmentary signs not only accompany XFS, but precede the fully established condition and can predict its development.
In 1971, Bartholomew49 first provided epidemiological clues of an early stage in the development of XFS. He demonstrated an early “pregranular” stage in the evolution of the disorder in the Pondo tribe in South Africa (observed in 5.1% of persons between 30 and 39 y of age). A hypothesis for the early development and clinical appearance of the condition has also been suggested by Jerndal.2 According to his theory the posterior layer of the iris epithelium begins to lose pigment and tiny dots of pigment appear in radial clusters upon the lens capsule. Subsequently, there is progressive loss of the pupillary ruff and an increasing number of pigmentary signs.2 These observations have raised the possibility that a group of apparently “normal” eyes without manifest exfoliation deposits may be affected by the disorder and this group has since been named “exfoliation suspects.” Importantly, a population-based study in 3468 Northern Chinese subjects with an age ≥50 years24 identified a greater rate of exfoliation suspects (3.44%) than those with definite XFS (2.38%). Including exfoliation suspects the prevalence of XFS in the Chinese population (5.82%) according to this study24 was substantial; between 14 and 29 times greater than the 2 previously reported prevalence studies in the Chinese (0.4% and 0.2%),36,37 an ethnic cohort in whom XFS had hitherto been considered rare.
A number of researchers have shown that further insight into the development and prevalence of the condition might be gained by ultrastructural studies on tissue samples from exfoliation suspects who do not demonstrate clinical evidence of exfoliation material on biomicroscopy. The “unaffected” eye of individuals with clinically unilateral XFS has been the principal focus of interest. First, Speakman and Ghosh50 found exfoliation material in conjunctival biopsy specimens from both eyes in 7 patients with clinical evidence of unilateral XFS. Similar evidence was obtained from ultrustructural studies in conjunctival specimens from exfoliation suspect patients who only exhibited pigmentary signs.47 Importantly, these patients exhibited only pigmentary signs, without biomiscroscopical evidence of exfoliation material in either eye. Therefore, the concept that there is a preclinical histologic stage in many, if not all XFS patients is now firmly established. Moreover, the terms “unilateral’’ and “monocular’’ in XFS are probably misleading.9 Consistently, when only 1 eye is clinically involved, the fellow eye also demonstrates findings such as abnormal aqueous humor dynamics, elevated intraocular pressure or glaucomatous damage resembling the clinical status of the fellow eye.4,8,10 Further, early pigment-related signs of XFS are found in most apparently unaffected fellow eyes, and exfoliation fibers can be detected on conjunctival biopsy in virtually all. As a result, there is a group of individuals who demonstrate “unilateral” disease but actually manifest asymmetric XFS.44,47 More importantly for the prevalence of the disorder, a group of seemingly “normal” exfoliation suspects suffers from histologic disease only.
It remains to be elucidated to what extent in each ethnic group the presence of these patients would influence the reported prevalence of the disorder. It is reasonable to assume that these ultrustructural and pigmentary findings are encountered early in the disease process and increase during the course of the disease. It is uncertain whether the ratio of clinically apparent XFS compared with only morphologic disease is different in the various ethnic groups, or whether this is uniformly encountered in every cohort.44 Moreover, although the various pigmentary signs are clinically visible early in the course of the disorder, their diagnostic sensitivity and specificity are yet to be evaluated in longitudinal studies.
Another limitation of many epidemiological studies is the considerable difficulty in the detection of lenticular signs of XFS in patients with either lens opacities, or cataract. This reduction in diagnostic sensitivity was highlighted in a published histologic study of 100 consecutive cataractous lenses.51 Importantly, XFS was detected morphologically in 33% of these cataractous lenses, whereas, clinically only 16% of these cases had been previously diagnosed before cataract surgery. In another investigation Larsen52 examined 100 postmortem eyes of 50 patients over age 70 and found that while 12% had light microscopic evidence of exfoliation material, only 3% had visible biomicroscopical evidence of the disease. Moreover, in all population-based studies a probable weakness is missing the diagnosis of XFS in cataract-operated patients. Although sometimes XFS deposits can be seen upon the intraocular lens after cataract surgery, clearly the diagnosis of XFS may easily be missed in those subjects who have undergone cataract surgery.
It is therefore beyond doubt that there is a significant rate of underdiagnosis with XFS if we include those with only histologic XFS, exfoliation suspects with pigmentary signs only, those with early disease and those in whom the diagnosis is missed. Published prevalence figures unquestionably do not reflect the true prevalence of XFS. Indeed, in their review Ritch and Schloetzer-Schrehardt8 have put forward the concept that in a given population of patients with chronic glaucoma, the actual prevalence of XFG is probably about twice that which is visible on clinical examination. The same ratio between detected and undetected disease may apply in eyes with XFS.
With early diagnosis elusive in a significant proportion of subjects it is worthwhile questioning how should we pursue in the future an earlier and more accurate diagnosis of XFS. The investigation of XFS and its precursor the exfoliation suspect, is complicated by the extensive, variable range of subtle abnormalities witnessed in the long course of this condition. These findings highlight the enigmatic nature and puzzling geographic distribution of the disorder. Future epidemiological studies that will address these issues are essential if the complex epidemiology of this condition is to be delineated.
KEY FINDINGS FROM EPIDEMIOLOGICAL STUDIES
When XFS was discovered, it gained little attention anywhere in the world, and the impression was formed that it was a condition mainly encountered in Scandinavia, Greece, and a few other countries.5,6 This is history, and today we know that XFS is a common pathology that occurs worldwide. Nevertheless, even today little is known about the prevalence of the disorder in many countries in Asia, Africa, and Latin America. Moreover, many published studies are unsatisfactory. When we critically evaluate the epidemiology literature on XFS, one has to bear in mind the different types of patient groups examined. Until recently, in many ethnic cohorts random sampling has largely been neglected, and therefore results on XFS prevalence are often incomparable. Available data may be grouped within the following categories:
- Information was collected from patients voluntarily attending ophthalmologists for some reason. These are studies documenting the occurrence of XFS and indicating its frequency in the particular ophthalmic group under examination. Most of these papers present figures on the frequency of XFS in selected glaucoma and/or cataract cohorts.
- Screening studies of larger ophthalmic population groups selected on a nonrandom basis. Observations show the frequency of XFS in the examined population, but do not reflect the real distribution in the area because of selection bias.
- Prevalence studies (ie, publications presenting the number of XFS cases at any given time related to the number of individuals in that particular area) are, strictly speaking, the only reports meeting the criteria for proper epidemiological studies. In other words, the information has been compiled from a randomly selected population group, or from the total population in a given area.
- Incidence studies. So far, few XFS incidence studies exist26,28,53–55 (ie, studies recording the number of XFS cases arising in a given time interval in a certain area).
The large number of category 1 and 2 reports (ie, frequency studies) has been previously summarized3,6,8,44 and demonstrate that XFS is a common condition of worldwide significance. Moreover, as XFS is the precursor of XFG, it is a condition of interest to anyone dealing with glaucoma-related issues, (etiology, pathogenesis, or therapy). Future epidemiological projects should be planned as category 3 and/or 4 studies (ie, prevalence and/or incidence studies), because this evidence will enhance our understanding on the pathogenesis and clinical importance of the disorder. Any calculation of XFS distribution in a population, as well as interregional comparisons, makes sense only when based on reliable prevalence and incidence studies. Otherwise fictional differences may be observed. The following aspects should therefore be considered when studying XFS:
- The pupil should always be maximally dilated, because at least 10% to 25% of exfoliation cases are missed without pharmacological dilatation of the pupil.1,7,8
- XFS prevalence increases rapidly above 60 years of age, but early, atypical occurrence of the condition in some young patients has been observed in some ethnic cohorts. Depending on the cohort under investigation this parameter should be included in the design.
- Exfoliation suspects, histologic XFS and early XFS cases with subtle signs can readily be overlooked. The protocol of future epidemiological studies should incorporate evaluation and careful monitoring of these cases.
- The diagnostic sensitivity increases when XFS is actively sought by experienced investigators who are fully aware of the complete clinical picture of XFS. Accurate diagnosis of XFS requires comprehensive knowledge of all pigmentary and exfoliation clinical signs and active interest in the condition. All investigators should therefore have appropriate clinical experience commensurate with identification of the full spectrum of the condition.
It should be noted, however, that even results from population-based studies may be incomparable because of significant differences in design (eg, substantial age differences between examined groups, etc.). Studies performed by the same investigators, using the same methodology and examining population cohorts in different regions of the same country may help. In the Middle Norway survey, population cohorts from 3 separate areas were examined contemporaneously using identical diagnostic criteria.42 In this population-based study, 10.2%, 21.0%, and 19.9% of the 3 neighboring Norwegian population cohorts exhibited typical XFS. Interestingly, marked differences in XFS prevalence was observed between regions lying only some 100 km apart. In Crete, Greece the XFS prevalence for 4 neighboring regions (Heraklion, Chania, Lasithi, and Rethymnon) was also diverse: 11.5%, 13.4%, 16.9%, and 27%, respectively.41 In contrast, in Hlabisa and Temba, 2 different areas in South Africa, the recorded XFS prevalences were fairly similar (7.7% and 6.0%, respectively).17
An intriguing observation made for the first time in the Middle Norway survey was of a significantly higher number of married couples, with XFS affecting both spouses with a higher frequency than could be expected from random occurrence.42 Along with the observation of 2 homozygote pairs of twins, 1 concordant and 1 discordant for the presence of XFS, this was thought to indicate an environmental influence for the geographic distribution of XFS. However, it is more likely that a combination of genetic and nongenetic factors are involved in the pathogenesis of the disorder (ie, a multifactorial disorder).
It is currently known that exfoliation material can be found in both intraocular and extraocular tissues, and XFS is now considered a generalized disorder of the extracellular matrix.8,9,14 To date, however, it remains unknown whether exfoliation fibers in various systemic organs are similar in their composition and are associated with an identical matrix to that seen in intraocular exfoliation material. Further, it remains to be elucidated if exfoliation fibers detected in systemic organs can form typical exfoliation deposits adversely influencing function, and what (if any) other contributing factors are required.8,10 It is noteworthy that the morphologic presence of the generally sparse systemic exfoliation aggregates have yet to be directly linked with a specific functional deficit. Therefore, the systemic ramifications of detecting XFS in patients remain to be elucidated. So far, no definitive associations between XFS and cardiovascular disease, cerebrovascular disease, systemic hypertension, or diabetes mellitus have been established.12,13,56 Importantly, unlike other more serious systemic conditions, which influence mortality, XFS patients do not seem to have a reduced life expectancy compared with those without exfoliation.57–59
OTHER EPIDEMIOLOGICAL FEATURES
It is well established that XFS is age dependent. The youngest reported patient with XFS was 17 years of age and had undergone intraocular surgery involving iris excision in infancy.60 Overall, the few, younger, atypical cases appear to have a different pathogenesis from the classic, age-related pathogenesis documented to date. Indeed, in at least 50 published reports in the Western and Japanese literature the youngest patients were over the age of 50 years.44,60 In contrast, it is worth noting that a distinct younger pattern may truly exist in some racial groups. In the Pondo tribe of South African Bantu, Bartholomew61 documented an overall prevalence of 6.4% in the 30 to 39 age group (10/157 persons), principally due to the inclusion of subjects demonstrating a pregranular stage of the disorder (ie, exfoliation suspects). These comprised 8 of 157 cases examined (5%). Along with the Australian Aboriginal cohort reported by Taylor et al,62 a unique “earlier involvement pattern” possibly influenced by genetic, or environmental factors has been documented for XFS in these ethnic cohorts.
All studies across the globe have uniformly shown increasing age to be a significant risk factor for the prevalence of XFS. For example, in the Chennai Glaucoma Study23 the odds for XFS increased 6 times from 8.4 for the 50- to 59-year group to 51.2 for the 70 years and above age group. Here it is noteworthy that a relevant key factor for the reported wide variation in XFS prevalence is the divergence (between 40 and 60 y) of minimum age in the various populations investigated. As increasing age is a constant risk factor for XFS prevalence this can explain some of the differences encountered. In addition, a consistent feature emerging from published evidence is that environment and geography are important associations in XFS.
The reported series on the relationship between XFS and sex are often irreconcilable. Some authors have shown a higher prevalence for female individuals.13,63 However the preponderance of female individuals has not been confirmed in other populations3,22,40 and large ophthalmic cohorts. For example in an Australian Aboriginal population only 10 of 37 patients with XFS were female.62 In some ophthalmic cohorts there is a marked preponderance of male individuals with the condition. However, cultural and social factors influencing the proportions of male individuals and female individuals presenting to ophthalmologists were not discussed by these authors. There are however population-based studies reporting a significant relationship between XFS and male sex.23,25,40,41
The Chennai Glaucoma Study23 examined 7.774 subjects aged 40 years and above and reported an overall prevalence of 3.73% and an age-adjusted and sex-adjusted prevalence of 3.41%. For the first time this study convincingly demonstrated that the prevalence of XFS was significantly higher in the rural, as opposed to the urban population cohort (4,74% vs. 2,05%; P<0.001). By showing a significant difference between rural and urban residence this Indian study suggests again a potential causative role for sunlight exposure in XFS development.23 In contrast, glaucoma was seen in only 8.3% of those with XFS. On multivariate logistic regression analysis, after adjusting for age, sex, and residence, there was no significant association of XFS with male sex. To date 4 population-based studies have reported the prevalence of XFS in India for subjects of 40 years and above. The prevalence of XFS in the Chennai Glaucoma Study was 3.8%23 in the Aravind Comprehensive Eye Survey 6%,23 in the Andhra Pradesh Eye Disease Study 3.01%,64 whereas in the Central India Eye and Medical Study it was only 1.96%.25 From population-based cross-sectional studies the relationship between XFS and cataract is also inconsistent.
INCIDENCE OF XFS
The information available on the incidence of XFS is limited. In a study by Karger et al55 conducted over a period of 15 years, the estimated overall age and sex-adjusted annual incidence of XFS was 25.9 per 100,000 population. The age-adjusted incidence was higher in female individuals than in male individuals (32.7 vs. 16.9 per 100,000; P<0.001). The prevalence of XFS increased with age, from 2.8 per 100,000 in persons 40 to 49 years to 205.7 per 100,000 in persons above 79 years (P<0.001).55 In the Reykjavik Eye Study26 the 12-year incidence of XFS was higher in women (9.2%) than in men (6.6%) and higher in older age groups than in younger ones. A total of 8.0% of participants developed XFS in at least 1 eye during the follow-up period. Most individuals deemed on biomicroscopy to be clinically unilaterally affected converted to bilateral status over 12 years. In the Chennai Eye Disease Incidence Study28 in 6 years XFS developed in 2.03% of the population with the rural incidence being 3 times greater than the urban incidence. The incidence of XFS was not associated with female sex, but as in all other studies it increased significantly with age. Eyes with XFS were at a greater risk of nuclear cataract developing, or needing to undergo cataract surgery.
ASSOCIATION OF GLAUCOMA WITH XFS
The association of glaucoma with exfoliation has been reviewed by Vesti and Kivelä7 and Ritch et al.8,10 Briefly, the presence of exfoliation material observed by slit-lamp examination is a prerequisite for the clinical diagnosis of XFG. It has long been known that XFS and XFG often coexist in the eye. By repeated examinations, the sequence of events has been established: XFS occurs first, followed by the development of ocular hypertension and XFG in a significant proportion of patients, which means that the presence of XFS may serve as a marker for the future development of XFG. New insights have increased the importance of accurate diagnosis. Ritch9 has calculated that cumulative epidemiological evidence implicates XFS as the cause of about 20% to 25% of cases of open-angle glaucoma, making it the most common identifiable cause worldwide. As highlighted above, many XFS cases exist before the clinically visible appearance of classic exfoliation aggregates on the lens surface. A better understanding of the geographic distribution, clinical evolution, and the triggering mechanism between XFS and XFG should lead to new approaches to XFG management and, eventually, prevention, or elimination of the disorder as a cause of glaucoma.4,9,12
Most studies pinpoint XFS as a significant risk factor for glaucoma, for example in the Blue Mountain study,65 glaucoma was ∼8-fold more common in eyes with exfoliation than in eyes without it. In contrast, in the Reykjavik Eye Study13,21 XFS was not documented as a significant risk factor for open-angle glaucoma. These and other discrepancies demonstrate that the precise interaction between XFS and glaucoma is still unclear, and that future surveys will have to adopt a standardized setup if methodological confusion is to be prevented.
Importantly, in some ethnic cohorts the high prevalence of XFS may translate into a higher overall prevalence of open-angle glaucoma. This has been highlighted in the Thessaloniki Eye Study11,19,20,22 where the prevalence of open-angle glaucoma in a Greek population 60 years of age or older was relatively high (5.5%) if both definitions used by the authors are used. In this population-based study XFS was present in 11.9% of participants of whom 15.2% already demonstrated XFG. It should be stated here however, that the opposite may not equally apply: low XFS prevalence figures may not be consistent with commensurate lower open-angle glaucoma rates in various regions.
Nevertheless, it is almost universally suggested in the world literature that male individuals with XFS may suffer more commonly from XFG.5,6 To date, this susceptibility of male individuals to XFG has not been adequately explained. Furthermore, the intraocular pressure has been reported by some investigators to be significantly higher in male patients with XFG. Ringvold et al66 conducted a population-based survey in Norway. In their material (1871 persons) 14.8% of the male individuals and 18.2% of the female individuals had XFS, whereas XFG was present in 35.9% of the affected male individuals in contrast to 25.7% for the affected female individuals.66 Considering patient cohorts coming to surgery, the male to female ratio is even higher. For example, a Greek study67 established a male to female ratio of 2.8:1 in patients undergoing filtration surgery. Similarly in an ophthalmic cohort from Hungary the male to female ratio was 1.7:1.68
The majority of published reports on XFS have suggested that the disorder occurs unilaterally in a significant number of cases.1,3,6,8 In the Framingham study the condition occurred unilaterally in 76.5% of affected persons aged 52 to 85 years.63 In contrast, in Australia, only 56.7% of the Aboriginal population studied by Taylor et al62 showed unilateral exfoliation. Tarkkanen43 has claimed that “in many patients the condition remained unilateral over a number of years or forever.” This concept was supported by other investigators,44 however, it is now firmly established that bilaterality in XFS increases with age thus many cases that are unilateral at the time of initial diagnosis become clinically bilateral at a later date.8–10 Overall, the average age for bilateral XFS is higher than that that for unilateral XFS.10,14,16 To date, however, it is not clear how clinically evident XFS would compare with morphologic XFS only.
The relationship between race and prevalence of the disease is intriguing, complex, and poorly understood. It is recognized that there are many difficulties in attempting to assess the impact of race in the prevalence of a common, subtle condition such as XFS by different investigators using different protocols. As stated above XFS is common in Greece.19,20,44 In the United States Gradle and Sugar69 reported that XFG was common in individuals of Greek origin. In their series from Chicago they detected 12 patients of Greek birth comprising 15.6% of their cases. Similarly, in Australia, Gillies70 found XFS to be more common in individuals of Greek, Latvian, and Russian origin. Khazanda71 studied 696 ophthalmic patients with XFS/XFG in Pakistan and found the disease to be more common in the mountainous areas of North West Frontier Province in the Pathan tribe. He was mystified by this phenomenon and speculated that the high prevalence of the disorder may be due to the fact that historically, this region has been the gateway to Indo-Pakistan for the invasions of Greeks and other ethnic groups. According to the author, the common ancestry between the Pathans and the Greeks (Aryans) may be the answer to this question.71
There is cumulative evidence suggesting that XFS occurs at a high prevalence among black South Africans. In a random cross-sectional sample of 1840 black subjects aged 40 years, or older the prevalence of XFS was 7.7% in the Hlabisa and 6% in the Temba tribes.17 The prevalence increased with age, with 18.9% (Hlabisa) and 16.5% (Temba) of those 70 or older, being affected. Significantly, these prevalence rates are not far below the highest published figures from Scandinavia, or Greece.13,14 Moreover, in this South African study there was no apparent ethnic or regional variation in the prevalence of XFS. In contrast, the prevalence of XFS appears to be low in black Americans. Olawoye et al18 established that the burden of XFS is high in sub-Saharan Africa, but well-designed population studies are needed to determine why clinic-based studies report virtually no XFS in some countries (Ghana and Tanzania), whereas nearby countries report greater proportions (Nigeria and Ethiopia).18
XFS has been considered rare in Chinese people. In the first study to document the prevalence of XFS in Chinese people, Young et al36 investigated 500 subjects at 2 hospitals in Hong Kong. They reported that only 2 of 500 consecutive ophthalmic patients (0.4%) exhibited XFS. As the authors pointed out themselves, precise figures on prevalence can be obtained only by investigating population groups. Further, since all patients enrolled in this study had a diagnosis of presumptive cataract it would have been difficult to detect all cases with XFS. Nevertheless, in the population-based Singaporean Tanjong Pagar Study37 the prevalence of XFS was also reported to be very low (0.2%). In contrast, in the population-based Beijing Eye Study24 which examined 3.022 Chinese subjects from Northern China, the overall prevalence of XFS (definite and suspected) was much higher (5.82%). The prevalence of XFS increased from 1.1% among those who were 50 to 54 years old to 3.5%, 5.7% and 11.8% among those who were 60 to 64 years, 70 to 74 years, and aged 80 or older respectively. This population-based study appears to refute the previously held concept that XFS is a rare condition in the Chinese people. It should be noted however, that the substantially higher prevalence figure in this Chinese cohort was in part due to the inclusion of exfoliation suspects.24 The higher prevalence figure for XFS in Chinese subjects from the Beijing Eye Study appears to be similar to those reported in a number of large population-based studies from Asia. Specifically, from 3 populations studies in South India (3% to 6%),23,25,27,64 a Japanese study (Hisayama study, 3.4%)38 and a study in Burma (the Meiktila Eye Study, with 3.4% in a population over the age of 40 y being affected).39
In the United States and Israeli case-control groups investigated by Pasquale et al72 a positive association was observed between residential latitude away from the equator and XFS. More time spent outdoors in the summer over a lifetime was associated with subsequent XFS development. In contrast, at the US site, sunglass wear was associated with reduced odds of XFS. Taken with the evidence that work over water, or snow in the US was associated with an increased odds of XFS, these data suggest that ocular exposure to light from reflective surfaces is important for XFS development.72 Here a discrepancy arises with cohorts like Greenland Inuits and Peruvians with a relatively low prevalence of XFS despite significant exposure to reflected sunlight from snow that could contribute to XFS.40 Pasquale et al72 have proposed that these individuals have relatively thick irides that ameliorate damage caused by the high degree of reflected ultraviolet radiation. Another intriguing possibility and a unifying concept in XFS development in some, but not all exposed individuals is the possibility that beyond ultraviolet irradiation exposure, what determines the manifestation of the disorder is the individual level of protection to ultraviolet irradiation. Those subjects with significantly reduced levels of ascorbic acid in the aqueous humor may ultimately develop XFS.73 A biochemical study documented a significantly lower level of ascorbic acid in the aqueous humor of patients with XFS.73 In view of the fact that ascorbic acid is a major protective factor against free-radical action, a role for free-radical action is logical in XFS pathobiology. Ultraviolet radiation, a main source of free radicals, has been previously implicated in various ocular disorders such as cataract and age-related macular degeneration, and has been suggested as one of the causes for the variable prevalence of XFS around the world.74 Importantly, the ascorbic acid concentration in the aqueous humor in humans is ∼40 times greater than that of plasma.73,75 Therefore, it appears that the most likely etiology for the reduced ascorbic levels in the aqueous in XFS is either impaired ascorbic acid transport through the ciliary epithelium, or depletion of ascorbic acid induced by increased levels of free radicals in XFS.76 Another possibility for the low ascorbic acid level is due to the impaired ciliary body ascorbate pump in XFS. It is therefore conceivable that it is the reduced protection against ultraviolet radiation induced from a lower concentration of ascorbic in those susceptible individuals that develop XFS may play a role in the variable prevalence of XFS in certain ethnic groups and the higher prevalence of cataract seen in this condition.
CONCLUDING REMARKS ON XFS EPIDEMIOLOGY
As we currently have sufficient information that XFS occurs worldwide, it is hard to see how our epidemiological knowledge can be augmented by more frequency studies (ie, categories 1 and 2).
What we need are well-designed prevalence and/or incidence surveys on XFS (ie, categories 3 and 4), which will facilitate a better understanding on the impact of the condition worldwide and may help our search for etiological factors.
XFS is now established as a significant risk factor for XFG. The etiology of XFG is still unclear. Moreover, it is not known why only about 30% of XFS eyes develop XFG within their lifetime. Country-to-country comparisons pose difficulties because of significant methodological discrepancies. Therefore, in order to facilitate future prevalence and incidence comparisons worldwide, a consensus-based agreement on an optimized study design should be used. This approach can be coordinated and facilitated by a supranational scientific body (eg, Lindberg Society). An essential criterion for such a study would be inclusion of clearly defined XFS suspects.
In the light of the inadequacies of previous epidemiological studies and the considerations discussed above there is much scope for further research on the epidemiology of the condition. Issues that remain ambiguous include the mode and rate of conversion of exfoliation suspects to XFS and subsequently to XFG. At the present time our knowledge concerning the incidence rate of XFS/XFG in exfoliation suspects is incomplete. Long-term–controlled studies are needed here as well. As yet there is no adequate information on the prevalence of histologic change only, or early clinical XFS and this will certainly be a promising line of future research. Long-term, prospective combined clinical and morphologic studies involving conjunctival biopsies to document the presence of exfoliation material in exfoliation suspects are required to establish the true prevalence of the condition. Ultrastructural assessment of conjunctival biopsies will permit evaluation of histologic only XFS, the diagnostic value of a host of preceding pigmentary signs and will allow proper clinicomorphologic correlations. Finally, it will be interesting to test whether these early changes and the prevalence of exfoliation suspects are similar in nature and frequency in different ethnic cohorts.
1. Tarkkanen AH. Pseudoexfoliation of the lens capsule. Acta Ophthalmol. 1962;71(suppl):9–98.
2. Jerndal TCairns JE. Open angle glaucoma and the pseudo-exfoliation syndrome
. Glaucoma. London: Grune & Stratton; 1986:661–677.
3. Forsius H. Exfoliation syndrome
in various ethnic populations. Acta Ophthalmol Suppl. 1988;184:71–85.
4. Ritch R. Exfoliation syndrome
and occludable angles. Trans Am Ophthalmol Soc. 1994;92:845–944.
5. Ringvold A. Epidemiology
of glaucoma in Northern Europe. Eur J Ophthalmol. 1996;6:26–29.
6. Ringvold A. Epidemiology
of the pseudoexfoliation syndrome
. A review. Acta Ophthalmol Scand. 1999;77:371–375.
7. Vesti E, Kivelä T. Exfoliation syndrome
and exfoliation glaucoma. Prog Ret Eye Res. 2000;19:345–368.
8. Ritch R, Schlötzer-Schrehardt U. Exfoliation syndrome
. Surv Ophthalmol. 2001;45:265–315.
9. Ritch R. Exfoliation syndrome
. Curr Opin Ophthalmol. 2001;12:124–130.
10. Ritch R, Schlötzer-Schrehardt U, Konstas AG. Why is glaucoma associated with exfoliation syndrome
? Prog Retin Eye Res. 2003;22:253–275.
11. Anastasopoulos E, Founti P, Topouzis F. Update on pseudoexfoliation syndrome
pathogenesis and associations with intraocular pressure, glaucoma and systemic diseases. Curr Opin Ophthalmol. 2015;26:82–89.
12. Ritch R. Exfoliation syndrome
: beyond glaucoma. Arch Ophthalmol. 2008;126:859–861.
13. Arnarsson AM. Epidemiology
of exfoliation syndrome
in the Reykjavik Eye Study. Acta Ophthalmol. 2009;87(thesis 3):1–17.
14. Aboobakar IF, Johnson WM, Stamer WD, et al. Major review: exfoliation syndrome
; advances in disease genetics, molecular biology, and epidemiology
. Exp Eye Res. 2016;154:88–103.
15. Dewundara S, Pasquale LR. Exfoliation syndrome
: a disease with an environmental component. Curr Opin Ophthalmol. 2015;26:78–81.
16. Sein J, Galor A, Sheth A, et al. Exfoliation syndrome
: new genetic and pathophysiologic insights. Curr Opin Ophthalmol. 2013;24:167–174.
17. Rotchford AP, Kirwan JF, Johnson GJ, et al. Exfoliation syndrome
in black South Africans. Arch Ophthalmol. 2003;121:863–870.
18. Olawoye OO, Pasquale LR, Ritch R. Exfoliation syndrome
in sub-Saharan Africa. Int Ophthalmol. 2014;34:1165–1173.
19. Topouzis F, Wilson MR, Harris A, et al. Prevalence
of open-angle glaucoma in Greece: the Thessaloniki Eye Study. Am J Ophthalmol. 2007;144:511–519.
20. Topouzis F, Wilson MR, Harris A, et al. Risk factors for primary open-angle glaucoma and pseudoexfoliative glaucoma in the Thessaloniki Eye Study. Am J Ophthalmol. 2011;152:219–228.
21. Arnarsson A, Damji KF, Sverrisson T, et al. Pseudoexfoliation in the Reykjavik Eye Study: prevalence
and related ophthalmological variables. Acta Ophthalmol Scand. 2007;85:822–827.
22. Anastasopoulos E, Topouzis F, Wilson MR, et al. Characteristics of pseudoexfoliation in the Thessaloniki Eye Study. J Glaucoma. 2011;20:160–166.
23. Krishnadas R, Nirmalan PK, Ramakrishnanan R, et al. Pseudoexfoliation in a rural population of southern india: the araavind comprehensive eye survey. Am J Ophthalmol. 2003;135:830–837.
24. You QS, Xu L, Wang YX, et al. Pseudoexfoliation: normative data and associations. The Beijing Eye Study 2011. Ophthalmology. 2013;120:1551–1558.
25. Jonas JB, Nangia V, Matin A, et al. Pseudoexfoliation: normative data and associations. The Central India Eye and Medical Study. PLoS ONE. 2013;8:e76770.
26. Arnarsson A, Sasaki H, Jonasson F. Twelve-year incidence of exfoliation syndrome
in the Reykjavik Eye Study. Acta Ophthalmol. 2013;91:157–162.
27. Vijaya L, Asokan R, Panday M, et al. The prevalence
of pseudoexfoliation and the long-term changes in eyes with pseudoexfoliation in a South Indian population. J Glaucoma. 2016;25:e596–e602.
28. Vijaya L, Asokan R, Panday M, et al. Six-year incidence and baseline risk factors for pseudoexfoliation in a south Indian population: the Chennai Eye Disease Incidence Study. Ophthalmology. 2015;122:1158–1164.
29. Kilic R, Karagoz N, Cetin AB, et al. The prevalence
of exfoliation syndrome
in Turkey. Acta Ophthalmol. 2016;94:e105–e108.
30. Kang JH, Wu J, Cho E, et al. Contribution of the Nurses Health Study to the epidemiology
of cataract, age-related macular degeneration, and glaucoma. Am J Public Health. 2016;106:1684–1689.
31. Tomassen TL. On the so-called capsular glaucoma. Acta Ophthalmologica. 1949;27:423–427.
32. Leydhecker W. Glaucom bei der sogenannten Abschilferung der Linsenkapsel. Glaucom ein Handbuch. Berlin: Springer-Verlag; 1960:163–68.
33. Aasved H. The geographic distribution of fibrillopathia epitheliocapsularis, so-called senile exfoliation, or pseudoexfoliation of the anterior lens capsule. Acta Ophthalmol. 1969;47:792–810.
34. Aasved H. Mass screening for fibrillopathia epitheliocapsularis, so-called senile exfoliation or pseudoexfoliation of the anterior lens capsule. Acta Ophthalmol (Copenh). 1971;49:334–343.
35. Aasved H. Prevalence
of fibrillopathia epitheliocapsularis (pseudoexfoliation) and capsular glaucoma. Trans Ophthalmol Soc UK. 1975;99:293–295.
36. Young AL, Tang WWT, Lam DSC. The prevalence
of pseudoexfoliation syndrome
in Chinese people. Br J Ophthalmol. 2004;88:193–195.
37. Foster PJ, Seah SKL. The prevalence
of pseudoexfoliation syndrome
in Chinese people: the Tanjong Pagar Survey. Br J Ophthalmol. 2005;89:239–240.
38. Miyazaki M, Kubota T, Kubo M, et al. The prevalence
of pseudoexfoliation syndrome
in a Japanese population: the Hisayama study. J Glaucoma. 2005;14:482–484.
39. Abdul-Rahman AM, Casson RJ, Newland HS, et al. Pseudoexfoliation in a rural Burmese population: the Meiktila Eye Study. Br J Ophthalmol. 2008;92:1325–1328.
40. Forsius H, Forsman E, Fellman J, et al. Exfoliation syndrome
: frequency, gender distribution and association with climatically induced alterations of the cornea and conjunctiva. Acta Ophthalmol. 2002;80:478–484.
41. Kozobolis VP, Papatzanaki M, Vlachonikolis IG, et al. Epidemiology
of pseudoexfoliation in the island of Crete (Greece). Acta Ophthalmol Scand. 1997;75:726–729.
42. Ringvold A, Blika S, Elsås T, et al. The Middle-Norway eye screening study. I. Epidemiology
of the pseudo-exfoliation syndrome
. Acta Ophthalmol. 1988;66:652–658.
43. Tarkkanen AH. Exfoliation syndrome
. Trans Ophthalmol Soc UK. 1986;105:233–236.
44. Konstas AGP. Morphological and clinical studies on the exfoliation syndrome
and open angle glaucoma. PhD Thesis
, Glasgow, 1993;1–367.
45. Krause U, Helve J, Forsius H. Pseudoexfoliation of the lens capsule and liberation of iris pigment. Acta Ophthalmologica. 1973;51:39–46.
46. Madden JG, Crowley MJ. Factors in the exfoliation syndrome
. Br J Ophthalmol. 1982;66:432–437.
47. Prince AM, Streeten BW, Ritch R, et al. Preclinical diagnosis of pseudoexfoliation syndrome
. Arch Ophthalmol. 1987;105:1076–1082.
48. Wishart PK, Spaeth GL, Poryzees EM. Anterior chamber angle in the exfoliation syndrome
. Br J Ophthalmol. 1985;69:103–107.
49. Bartholomew RS. Pseudo-capsular exfoliation in the Bantu of South Africa. I. Early or pre-granular clinical stage. Br J Ophthalmol. 1971;55:693–699.
50. Speakman JS, Ghosh M. The conjunctiva in senile lens exfoliation. Arch Ophthalmol. 1976;94:1757–1759.
51. Krause U, Tarkkanen A. Cataract and pseudoexfoliation. A clinicopathological study. Acta Ophthalmol (Copenh). 1978;56:329–334.
52. Larsen JS. Senile exfoliation (pseudo-exfoliation, fibrillopathia epitheliocapsularis) of the lens capsule in a postmortem material. Microscopic investigation of 100 eyes. Acta Ophthalmol (Copenh). 1969;47:676–684.
53. Topouzis F, Anastasopoulos E. Incidence of pseudoexfoliation syndrome
. Am J Ophthalmol. 2011;152:219–228.
54. Ekström C. Incidence of open-angle glaucoma in central Sweden. Acta Ophthalmol. 2008;86:747–754.
55. Karger RA, Jeng SM, Johnson DH, et al. Estimated incidence of pseudoexfoliation syndrome
and pseudoexfoliation glaucoma in Olmsted County, Minnesota. J Glaucoma. 2003;12:193–197.
56. Hietanen J, Soisalon-Soininen S, Kivelä T, et al. Evaluation of the clinical association between exfoliation syndrome
and abdominal aortic aneurysm. Acta Ophthalmol Scand. 2002;80:617–619.
57. Shrum KR, Hattenhauer MG, Hodge AD. Cardiovascular and cerebrovascular mortality associated with ocular pseudoexfoliation. Am J Ophthalmol. 2000;129:83–86.
58. Slettedal JK, Sandvik L, Ringvold A. Ocular pseudoexfoliation syndrome
and life span. EBioMed. 2015;2:765–769.
59. Svensson R, Ekstrom C. Pseudoexfoliation and mortality: a population-based 30-year follow-up study. Acta Ophthalmol. 2015;93:162–164.
60. Konstas AGP, Ritch R, Bufidis T, et al. Exfoliation syndrome
in a 17-year-old girl. Arch Ophthalmol. 1997;115:1063–1067.
61. Bartholomew RS. Pseudocapsular exfoliation in the Bantu of South Africa. II. Occurrence and prevalence
. Br J Ophthalmol. 1973;57:41–45.
62. Taylor HR, Hollows FC, Moran D. Pseudoexfoliation of the lens in Australian Aborigines. Br J Ophthalmol. 1977;61:473–475.
63. Hiller R, Sperduto RD, Kruegger DE. Pseudoexfoliation, intraocular pressure and senile lens changes in a population-based survey. Arch Ophthalmol. 1982;100:1080–1082.
64. Thomas R, Nirmalan PK, Krishnaiah S. Pseudoexfoliation in southern India: the Andhra Pradesh Eye Disease Study. Invest Ophthalmol Vis Sci. 2005;46:1170–1176.
65. Kanthan GL, Mitchell P, Burlutsky G, et al. Pseudoexfoliation syndrome
and the long-term incidence of cataract and cataract surgery: The Blue Mountains Eye Study. Am J Ophthalmol. 2013;155:83–88.
66. Ringvold A, Blika S, Elsas T, et al. The Middle-Norway eye-screening study. II. Prevalence
of simple and capsular glaucoma. Acta Ophthalmol. 1991;69:273–280.
67. Konstas AG, Allan D. Pseudoexfoliation glaucoma in Greece. Eye. 1989;3:747–753.
68. Sziklai P, Suveges I. Glaucoma capsulare in patients with open-angle glaucoma in Hungary. Acta Ophthalmol Suppl. 1988;184:90–92.
69. Gradle HS, Sugar HS. Glaucoma capsulare. Am J Ophthalmol. 1947;30:12–19.
70. Gillies WE. Racial incidence of pseudoexfoliation of the lens capsule. Br J Ophthalmol. 1972;56:474–477.
71. Khanzada AM. Exfoliation syndrome
in Pakistan. Pak J Ophthalmol. 1985;2:7–9.
72. Pasquale LR, Jiwani AZ, Zehavi-Dorin T, et al. Solar exposure and residential geographic history in relation to exfoliation syndrome
in the United States and Israel. JAMA Ophthalmol. 2014;132:1439–1445.
73. Koliakos GG, Konstas AGP, Schrehardt-Schloetzer U, et al. Ascorbic acid concentration is reduced in the aqueous humor of patients with exfoliation syndrome
. Am J Ophthalmol. 2002;134:879–883.
74. Taylor HR. Pseudoexfoliation, an environmental disease? Trans Ophthalmol Soc UK. 1979;99:302–307.
75. Watson DG, Iqbal Z, Midgley JM, et al. Measurement of ascorbic acid in human aqueous humour and plasma and bovine aqueous humour by high-performance liquid chromatography with electrochemical detection. J Pharm Biomed Anal. 1993;11:389–392.
76. Hollo G, Katsanos A, Konstas AGP. Management of exfoliative glaucoma
: challenges and solutions. Clin Ophthalmol. 2015;9:907–919.