SECTION 1  GLAUCOMA IN THE WORLD

1  Prevalence and Geographical

Variations
WINNIE NOLAN and JENNIFER LY YIP

Summary Epidemiological Methods

Glaucoma is the commonest cause of irreversible visual PREVALENCE AND INCIDENCE
morbidity worldwide. The covert nature of the disease requires
representative surveys to determine the true burden of Epidemiological studies quantify and interpret frequency of
glaucoma. Good-quality surveys with standardized definitions disease, and factors that affect this. Two important mea-
and methods are the starting point with which to tackle this sures that form the basis of all epidemiological studies are
global public health problem and in recent decades the number incidence and prevalence. Incidence is the number of new
of well-conducted prevalence surveys has increased
considerably. However, data from areas such as Latin America
cases in a given population over a specified period of time.
and Africa are needed to further quantify this problem. Global Prevalence is the number of all cases in a given population
and regional strategies can then be developed to address the at one point in time, and this is commonly determined from
challenge of glaucoma blindness in the Vision 2020 agenda. cross-sectional studies. Both measures alone are descriptive
in nature and, when used to compare frequencies in differ-
ent populations or subgroups, the analysis and search for
risk factors and causal factors can begin. This will lead to
new understanding and practical application in clinical
care and public health.
Introduction
STUDY DESIGNS: POPULATION-BASED SURVEYS
Glaucoma is the commonest cause of irreversible blindness
worldwide.1 The World Health Organization (WHO) esti- Prevalence is a useful measure of the burden of disease,
mates for the number of people blind from glaucoma in especially in conditions with long duration such as glau-
2002 were 4.4 million (12.3% of people blind worldwide). coma. Early studies used convenience samples from hospital
The majority of glaucoma in the world remains undiag- clinics; however, as glaucoma is primarily asymptomatic,
nosed and so we rely on data collected from epidemiological these studies would not provide an accurate assessment of
surveys to estimate numbers with the disease. In recent glaucoma prevalence. Therefore, more recent estimates are
decades there have been a number of population-based usually ascertained from cross-sectional surveys. The scien-
surveys investigating the prevalence of eye disease. One of tific value of a survey is dependent on its internal and exter-
the limitations of using prevalence data has been the lack nal validity. Glaucoma prevalence should be determined
of a standardized definition of glaucoma across the differ- by population-based studies, which select a representative
ent surveys. The increasing use of the International Society sample. As glaucoma is generally a disease of the older
of Geographical and Epidemiological Ophthalmology population, most glaucoma surveys focus on people who
(ISGEO) definition of glaucoma2 (Table 1-1) means it is now are 40 years or older, as this decreases the required sample
possible to obtain a global picture of the numbers of indi- size and saves resources.
viduals affected by glaucoma. It also allows comparison of Internal validity is dependent on factors that can distort
glaucoma prevalence and types in different regions, so the results, leading to false estimates, whether this is due to
highlighting populations and subgroups at increased risk of chance, bias, or confounding. Chance errors can be mini-
the disease. mized with adequate sample size. A low participation rate
With the accumulation of epidemiological data it is clear is an important source of bias in surveys, as nonpartici-
that glaucoma affects all populations, but that some regions pants may have a different experience of disease. Bias in
and racial subgroups are more affected either due to having ascertainment of disease or risk factors can result from
a higher disease prevalence or because the large population imprecise methods and protocols. Clear criteria should be
of those regions means the absolute numbers of individuals used to examine the participants and define the outcome.
with glaucoma is very large. The ISGEO guidelines2 described below have been adopted
In this chapter the methods of acquiring epidemiological internationally in many surveys, and are a useful standard
data will be explained and the geographical variations in to allow comparisons between different studies.
the prevalence and types of glaucoma will be illustrated Humans are prone to error, and objective measures are
together with a discussion of contributing risk factors. valued in scientific studies. The set of instruments used to

1
2 SECTION 1  •  Glaucoma in the World

assess the different parameters required in a glaucoma Table 1-1  International Society of Geographical and
survey will vary. Gold standard instruments such as Epidemiological Ophthalmology Classification of
Goldmann applanation tonometer and Humphrey visual Glaucoma for Use in Population-Based Surveys
fields may not be feasible in some community-based proj- Glaucoma
ects. If more than one set of equipment or examiners are
Category 1 Diagnosis (Structural and Functional Evidence)
employed in the study, then inter-observer assessments are
Cup:disc ratio (CDR) or CDR symmetry ± 97.5th percentile for the
required to demonstrate that both teams are comparable in normal population
performance. Or
Variation in disease frequency between different popula- Neuroretinal rim width reduced to ≤0.1 CDR (between 11 to 1 o’clock
tions is a major source of epidemiological hypotheses in the or 5 to 7 o’clock)
investigation of causal mechanisms. These differences can +
A definite visual field defect consistent with glaucoma
be real or artefact. The first step is to establish that a real
variation exists and then look for an association between Category 2 Diagnosis (Advanced Structural Damage  
with Unproved Field Loss)
the disease and a risk factor in these groups. Differences in
age structures can also lead to apparent differences in prev- CDR or CDR asymmetry ≥99.5th percentile for the normal population
alence. Populations with higher proportions of older people Category 3 Diagnosis (Optic Disc Not Seen)
can lead to higher estimates of glaucoma compared to Visual acuity <3/60 and IOP >99.5th percentile
younger populations. One common method to account for Or
differences in age structure is age standardization. Visual acuity <3/60 and evidence of glaucoma filtering surgery

From Foster PJ, Burhmann R, Quigley HA, et al. The definition and
DEFINITIONS AND DIAGNOSTIC CRITERIA classification of glaucoma in prevalence surveys. Br J Ophthalmol
2002;86:238–242.
Differences in diagnostic criteria can result in difficulties
when comparing differences in glaucoma prevalence. An PRIMARY OPEN-ANGLE GLAUCOMA
example of this is in comparing prevalence figures for
primary angle-closure glaucoma (PACG) from different According to the ISGEO classification, primary open-
surveys. The ISGEO guidelines have helped to standardize angle glaucoma (POAG) is defined as glaucomatous
definitions used in PACG studies. Previous criteria used optic neuropathy in the presence of an open-angle and no
symptoms rather than structural or functional evidence for other ocular abnormality to account for a secondary
diagnosing glaucoma. As a result, in the first publication of mechanism.
the Mongolia glaucoma survey, the prevalence of PACG was
1.4%,3 but using the revised grading system, the prevalence PRIMARY ANGLE-CLOSURE GLAUCOMA
is 0.8%. Similar problems arise when different criteria are
used to evaluate glaucoma progression. Differences can The ISGEO classification of primary angle-closure glau-
occur if only disc or visual field evidence is used compared coma (PACG) is a revised classification which places the
to a combination of both factors. The definitions used for emphasis on evidence of glaucomatous optic neuropathy
perimetric progression should also be carefully assessed in together with gonioscopic evidence.
each study, as this may be the true cause of an apparent The ISGEO classification of primary angle closure (PAC)
variation between different areas. Care should also be taken has three stages:
with studies that use self-reporting as the method for case ■ Iridotrabecular contact (ITC): Appositional contact
ascertainment. At least 50% of glaucoma is undiagnosed between the peripheral iris and posterior trabecular
in the population, and using this method to determine asso- meshwork (previously termed primary angle-closure
ciation with risk factors would be biased. Current diagnos- suspect).
tic guidelines suggest that glaucoma should be diagnosed ■ Primary angle closure (PAC): Presence of ITC (as
independently of symptoms or IOP. above) together with signs of trabecular meshwork
obstruction, e.g. peripheral anterior synechiae, elevated
Definition of Glaucoma for Use in IOP, excessive pigment deposition on surface of trabecu-
lar meshwork or ischemic sequelae such as glaucom-
Epidemiological Surveys flecken or iris whirling. The optic disc is healthy.
■ Primary angle-closure glaucoma (PACG): PAC with
In an attempt to overcome the problem of varying evidence of glaucomatous optic neuropathy.
diagnostic criteria for glaucoma The Working Group for
Defining Glaucoma of ISGEO developed a new scheme for SECONDARY GLAUCOMAS
the diagnostic classification of glaucoma.2 This classifica-
tion emphasizes the importance of visually significant end- Secondary glaucoma was proposed by the ISGEO authors to
organ (optic nerve head) damage as a requirement for the be the presence of glaucomatous optic neuropathy together
diagnosis of glaucoma. Since the publication of the ISGEO with signs of other pathological processes.
definitions a consensus meeting of experts in the field
resulted in a modification of the diagnostic criteria for APPLICATIONS OF EPIDEMIOLOGICAL DATA
primary angle-closure glaucoma.4
Table 1-1 shows an abbreviated version of the ISGEO Obtaining glaucoma prevalence data is the starting point
classification of glaucoma. for prevention of blindness programs targeted at reducing
 1  •  Prevalence and Geographical Variations 3

the number of cases of blindness. The epidemiological data compared with POAG in Blacks and Caucasians living in
have several uses: these geographic regions. However, Asian Americans have
a higher frequency of angle-closure glaucoma compared to
■ Application of data to population figures to estimate the
their Caucasian counterparts, in line with the risk of their
absolute numbers affected with glaucoma in different
source population. The differences in glaucoma types in
geographic regions.
■ Highlighting regions or subgroups with a higher preva-
migrants indicate the importance of genetic and possibly
cultural influences on these diseases. While it is indeed
lence of glaucoma-related visual impairment.
■ Determination of
likely that PACG is less frequent than POAG in Caucasians,
association between demographic
unless gonioscopic assessment is included as part of the
factors, e.g. age, sex, ethnicity, and glaucoma, and
comprehensive examination of glaucoma cases detected in
ocular factors such as biometric measurements or genetic
surveys, then the true prevalence of PACG may be
factors, which may be causal mechanisms contributing
underestimated.
to the pathogenesis of glaucoma.
■ Comparing prevalence between different regions so that

prioritization in resources and research programs can be LATIN AMERICA

given to populations most at risk.
Two surveys have focused specifically on the Hispanic or
Latino population who live in the United States. Proyecto
VER (conducted in Arizona) found a POAG prevalence
Regional Variation in Glaucoma somewhere between that shown in White and Black people
Prevalence and Type living in North America, but increasing at a significantly
more rapid rate with age compared to other ethnic groups.11
The Los Angeles Latino Eye Study (LALES) demonstrated a
EUROPE, NORTH AMERICA
high prevalence of POAG (4.74%). This is likely to be partly
Primary open-angle glaucoma (POAG) is the predominant due to the less stringent definitions used, which required
glaucoma in Europe, the United States of America (USA), either optic disc damage or visual field evidence to make the
and in the European-derived population of Australia (Table diagnosis of glaucoma.12 The population of the LALES
1-2).5–8 Within these regions the highest prevalence of study had mainly Mexican ancestry whereas that of
POAG is in the African and Caribbean-derived populations Proyecto VER had a greater number of people with Native
that live in the USA and in the Caribbean.5,9,10 Primary American roots. Either way, the data may or may not be
angle-closure glaucoma (PACG) is relatively uncommon applicable to Latin America as a whole, bearing in mind
that the region is populated by people with Indigenous, His-
panic, and African ethnicity. A more recent study from
Brazil where 72% of the local population is self-reported as
Table 1-2  Prevalence Estimates for Primary Open- White, supports a POAG prevalence between that of White
Angle Glaucoma from Population Surveys in Persons and Black populations. Furthermore, there was also a
over 40 Years Old by Region higher prevalence of PACG in this population.13
Region Author Prevalence (%)

AFRICA
ASIA
Tanzania (East Africa) Buhrmann 3.1 Asia is thought to harbor almost half of the glaucoma cases
Kwazulu-Natal (South Africa) Rotchford 2.7 worldwide and the numbers affected are projected to
Temba (South Africa) Rotchford 2.9 increase considerably over the next 20 years.14 This is due
Tema (West Africa) Budenz 8.0 to the presence of a number of heavily populated countries
WEST INDIES
including the two most populous, namely China (estimated
population 1.3 billion) and India (estimated population 1.1
Barbados (Blacks) Leske 7.0
billion), which make up almost a third of the world’s popu-
UNITED STATES AND EUROPE lation.15 Indonesia (245 million) is another densely and
Baltimore Tielsch 1.1 heavily populated country in the region.
Beaver Dam Klein 2.1 There is considerable variation in the distribution of
Egna-Neumarket, Italy Bonomi 2.0 glaucoma subtypes in Asia. East Asians of China, Mongolia,
AUSTRALIA and South-East Asian people of Chinese descent all experi-
Melbourne Wensor 1.7 ence a higher prevalence of PACG compared to people of
other ethnic origins. The reported prevalences range from
ASIA (SOUTH)
0.5% in Handan, a rural area of northern China to 2.5% in
Aravind, India Dandona 1.7 Myanmar.16,17 The highest reported prevalence within
Bangladesh Rahman 2.5 China itself is from Inner Mongolia with 1.4% of those >40
Chennai, India Vijaya 1.62 years affected.18
ASIA (EAST) Conversely, the Japanese and Koreans have reported
Mongolia Foster 0.5 some of the highest prevalences of POAG in the world of
Singapore Foster 2.1 >3% in the >40 population, with a majority of OAG with
Tajimi, Japan Iwase 3.9 low IOP at diagnosis.19,20 People living in South Asia also
experience a high frequency of POAG, with a markedly
4 SECTION 1  •  Glaucoma in the World

higher prevalence in urban compared to rural areas. Preva- Europe, and Australia.28 There was also statistical evidence
lence of PACG is higher than Western countries, but com- in all racial groups of heterogeneity, that is, a true variation
paratively lower than East Asia.21,22 between studies rather than variation by chance. This was
attributed to differences in age groups for the different
studies, survey methods, and year of publication which
AFRICA
reflects the change in diagnostic criteria used. In this study,
The most comprehensive data available from Africa were the prevalence is presented by different racial groups,
provided by surveys conducted in East Africa (Tanzania) and regardless of location; therefore, glaucoma prevalence of
South Africa. These reported combined primary and sec- Blacks includes surveys from Africa, the West Indies, and
ondary glaucoma prevalence figures of just over 5% (see Black populations within Europe and the USA. This would
Table 1-2).23–25 The predominant glaucoma was POAG but overlook potential differences in prevalence estimates
pseudoexfoliation (in Black South Africans), aphakic, and caused by environmental factors, which would be another
angle-closure mechanisms composed the remaining cases. source of heterogeneity. To address these issues, Table 1-2
Glaucoma prevalence among Black people living in the USA shows a summary of different prevalence estimates by
is measured as being four or more times that in Caucasians.5 region from more recent surveys with comparable methods
As the ancestors of African-Americans and the Caribbean in participants over 40 years old.
population came from West Africa, it is suspected that glau- These and other prevalence figures have been applied to
coma prevalence in this region may be equally high. This is the projected global populations for 2010 to estimate the
now confirmed by findings from a population-based survey absolute numbers of individuals with glaucoma worldwide.
in Ghana, which found glaucoma to be present in 6.5% of These calculations estimate that there will be almost 45
the adult population.26 It is very likely that the prevalence million individuals with primary open-angle glaucoma
and mechanism of glaucoma varies between heterogeneous (POAG) by 2010.14 A breakdown for these figures by region
populations of the African continent. However, what has is seen in Table 1-3.
been repeatedly demonstrated is that glaucoma affects a
higher proportion of African-derived people, has a younger PRIMARY ANGLE-CLOSURE GLAUCOMA:
age of onset, and that it may result in a greater visual mor-
PREVALENCE AND NUMBERS AFFECTED
bidity than in other populations.24,26
Table 1-4 shows the global variations in prevalence of
PRIMARY OPEN-ANGLE GLAUCOMA: PREVALENCE PACG. Definitions of PACG vary in published studies and the
majority of the surveys in Western countries have focused
AND NUMBERS AFFECTED
on detecting POAG. A systematic review of PACG in Euro-
A meta-analysis by Rudnicka et al.27 reviewed all POAG pean-derived populations gives a prevalence estimate of
surveys available in the literature, and estimated pooled 0.4%.29 The Proyecto VER survey of Hispanic patients in
prevalence by race. The plotted summary of all studies the USA reported a PACG prevalence of 0.1%,11 but there is
reviewed is shown in Figure 1-1, which provides a useful little population-based data from Central or South America
outline of differences in prevalence estimates. The variation other than that from the Projeto Glaucoma survey in Brazil,
in prevalence estimates between different races grouped in which found a PACG prevalence of 0.7% in a mixed ethnic-
this manner is evident, with prevalence for Whites ranging ity population.13 In Africa the data available confirm that
from <0.5% to >10%. The overall pooled prevalence esti- angle closure does exist in this population, with a preva-
mate was 2% (95% CI: 1.61–2.70%), which is higher than lence of 0.5% in South Africa and Tanzania.23–25
1.69% (1.53–1.85%) presented in another meta-analysis Recently, there has been an expansion in the number of
using individual data from recent studies from the USA, population-based glaucoma surveys conducted in Asia.

Table 1-3  Estimated Numbers with Open-Angle Glaucoma (OAG) and Angle-Closure Glaucoma (ACG) Worldwide, 2010
World Region Number with OAG World OAG % Number with World ACG % OAG and ACG
ACG Combined
China 8309001 18.6 7473195 47.5 15782196
Europe (including USA, Australia) 10693335 23.9 1371405 8.7 12064740
India 8211276 18.4 3733620 23.7 11944896
Africa 6212179 13.9 245844 1.6 6458023
Latin America 5354354 12 322804 2.1 5677158
Japan 2383802 5.3 278643 1.8 2662466
South East Asia 2116036 4.7 20141584 13.6 4257620
Middle East 1440849 3.2 177869 1.1 1618718
World 44720832 15744965 60465796

From Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006; 90:262–267.
 1  •  Prevalence and Geographical Variations 5

Citation Study (age range)

Asian
19 Hu (40+)
18 Arkell (15-70+)
22 Jacob (30-60)
21 Foster (40-89)
24 Dandona (0-102)
29 Rahman (35-85)
27 Ramakrishnan (40-90)
17 Aisbirk (40+)
23 Foster (40-81)
26 Bourne (50-70+)
11 Shiose (30-70+)
20 Rauf (30-80+)
25 Metheetrairut (60-104)
28 Iwase (40-80+)
Subtotal 1.41, 95% CI (1.00, 2.00); Q13 = 160.65, p <0.0001

Black
30 Wallace (35-74)
35 Ekwerekwu (30-80+)
34 Rotchford (40-80+)
33 Buhrmann(40-80+)
32 Wormald (35-60+)
36 Rotchford (40-97)
2 Tielsch (40-80+)
12 Leske (40-86)
37 Ntim-Amponsah (30-100)
31 Mason (30-70+)
Subtotal 4.23, 95% CI (3.07, 5.83); Q9 = 173.44, p<0.0001

White
4 Hollows (40-74)
12 Leske (40-86)
41 Bankes (20-80+)
6 Bengtsson (58.5-68.5)
2 Tieisch (40-80+)
10 Dielemans (55-75+)
44 Giuffre (40-99)
40 Salmon (40-70+)
49 Wensor (40-90+)
5 Coffey (50-80+)
1 Leibowitz (<65-75+)
47 Bonomi (40-80+)
39 Anton (40-79)
38 Quigley (41-90+)
3 Kiein (43-75+)
9 Mitchell (49-80+)
46 Cedrone (40-80+)
50 Kozobolis (40-80+)
48 Reidy (65-100)
7 Ringvold (65-89+)
8 Ekstrom (65-74)
42 Martinez (65-90+)
43 Gibson (76-85+)
45 Hirvela (70-95)
Subtotal 209, 95% CI (1.61, 2.70); Q23 = 361.74, p <0.0001

Overall 2.14, 95% CI (1.72, 2.65); Q47 = 1230.81, p <0.0001

0.003 0.01 0.03 0.05 0.1 0.25 0.5 1 2 4 8 10 15 20

Prevalence (%)
Figure 1-1  Meta-analysis of prevalence of open-angle glaucoma stratified by racial group. (Reproduced from: Rudnicka AR, et al. Varia-
tions in primary open-angle glaucoma prevalence by age, gender and race: a Bayesian meta-analysis. Invest Ophthalmol Vis Sci 2006;
47:4254–4261. With permission from the Association for Research in Vision and Ophthalmology.)
6 SECTION 1  •  Glaucoma in the World

Table 1-4  Prevalence Estimates for Primary Angle- These estimates of numbers of people with PACG predict
Closure Glaucoma by Region that China will account for almost 50% of angle-closure
Region Author Prevalence
glaucoma worldwide and that 85% of all individuals with
(%) PACG will be from Asia.14 These figures are estimates but
have the benefit of the reliable epidemiological data that
UNITED STATES, EUROPE, AND AUSTRALIA have been collected in Asia. The increase in predicted
Baltimore Whites Personal 0.4 numbers of glaucoma-affected people in the next ten years
communication
is partly due to an increase in life expectancy. The higher
Beaver Dam Klein 0.04
rate of visual loss in PACG compared with POAG heightens
Melbourne Wensor 0.06 the concern over the potential numbers with glaucoma
Wales Hollows 0.09 blindness in Asia and is the stimulus for current research
Egna-Neumarkt, Italy Bonomi 0.6 strategies aimed at effective methods of treatment and early
Proyecto VER (Hispanic) Quigley 0.1 detection of the disease.
LATIN AMERICA
Brazil Sakata 0.7 SECONDARY GLAUCOMAS
ASIA (SOUTH)
Secondary glaucomas are defined by the presence of a
Andra Pradesh (>30 yr) Dandona 0.7
primary ophthalmic or systemic disease causing raised
Aravind Ramakrishnan 0.5 intraocular pressure, which if sustained then leads to glau-
Chennai Vijaya 0.87 comatous optic neuropathy.30 Although it is relatively
ASIA (EAST) uncommon when compared with POAG, these cases have a
Beijing, China Hu 1.37 greater risk of blindness. This is partly as a result of the
Guangzhou, China (>50 yr) He 1.5 underlying disease, but also due to the higher levels of IOP
Nepal Thapa 0.39 seen in these cases and the fact that they are often refrac-
Tajimi, Japan Yamamoto 0.6 tory to conventional glaucoma treatment. The geographi-
Mongolia Foster 0.8
cal distribution of these types of glaucoma will follow that
Singapore Foster 0.8
of the causal ocular pathology. For example, in countries
where intracapsular cataract surgery is still practiced there
Thailand (>50 yr) Bourne 0.9
is a higher prevalence of aphakic glaucoma.24,33 However,
Myanmar Casson 2.9
with the increasing use of extracapsular cataract surgery
Sri Lanka Casson 0.57 and intraocular lens (IOL) implantation in most parts of the
AFRICA world, the incidence of this type of glaucoma should be on
Tanzania Buhrmann 0.5 the decrease.
Temba, South Africa Rotchford 0.5
Pseudoexfoliative Glaucoma
Cape Coloreds, South Africa Salmon 2.3
The finding of pseudoexfoliative (PXF) material on the lens
surface is strongly associated with glaucoma and raised
intraocular pressure.34 Pseudoexfoliative glaucoma is some-
times classified as a secondary glaucoma and otherwise as
a subtype of primary open-angle glaucoma. Glaucoma
Some of the Asia-based surveys carried out in the past patients with PXF have higher IOPs than POAG patients
lacked validity due to methodological weaknesses. The without PXF35 and it is a risk factor for blindness due to
strength of many of the recent surveys lies in their adher- cataract or glaucoma.35,37 One of the problems in establish-
ence to the ISGEO glaucoma definitions, and the inclusion ing the epidemiological distribution of PXF is that to be sure
of gonioscopy as part of the definitive clinical examination of its presence or absence on the lens surface requires pupil
on all subjects. dilation, otherwise it may be missed. PXF is a frequent
Surveys from Asia confirm that angle-closure glaucoma finding in the Scandinavian countries of Northern Europe
occurs more frequently in this region although POAG is such as Iceland (31% glaucoma cases)38 but other regions
equally if not more common. PACG prevalence figures where it is prevalent include southern India,35 Mongolia,3
range from 0.8% and 0.87% in Mongolia3 and Southern and in the Mediterranean countries such as Turkey and
India21 to 1% in Singapore,30 1.5% in Guangzhou, China,31 Greece. It is rare in the Chinese and is so far unreported in
and a high prevalence rate of 2.9% in Myanmar.17 African-American/Caribbean people, West Africans,39 and
The higher figure for the Guangzhou study can be in Tanzania,23 but has high prevalence in Black South Afri-
accounted for by the older age of the population being cans (16% all glaucoma cases)24,25 and Ethiopians.40 The
examined (>50 years) compared to the other study popula- clustering of PXF in families and populations suggests there
tions. An association between visually significant cataract is a genetic basis for the condition and this is now supported
and presence of angle closure in Myanmar suggests that the by the finding of LOX 1 gene polymorphisms in people with
higher prevalence of PACG in this population could be PXF41 although nongenetic factors have also been put
reduced by improving diagnosis and surgery for cataract.32 forward as potential contributors.42 The epidemiology and
When PACG prevalence figures are applied to the popula- etiology of PXF is an area that merits further research as
tion numbers for Asia the real potential magnitude of the it is the commonest identifiable cause of open-angle glau-
burden of PACG is immediately apparent (see Table 1-3). coma worldwide.
 1  •  Prevalence and Geographical Variations 7

Geographical Variation   ophthalmologists working with the Greenland Inuit

people.48–50 The values for age-specific mean central ante-
in Risk Factors rior chamber depth (ACD) measured in Mongolians were
shown by Foster et al. to lie between those found in Cauca-
RACE AND ETHNICITY sian populations (which have deeper chambers) and those
of the Greenland Inuit (with shallower chambers) demon-
There is much debate about the role and use of race in strating the inverse relationship between mean population
epidemiological research. The biological concept of race is central ACD and the prevalence of PACG.51
outdated. There is less genetic variation between races or Recent research using anterior segment imaging mea-
groups separated geographically compared to individuals surements provides further evidence of the relationship
from the same ethnic group.43 Ethnicity refers to a group between the lens and risk of angle closure. In population-
with shared cultural and geographical origins, and is being based studies there is a strong association between a higher
increasingly used as a euphemism for race, although they lens vault (perpendicular distance between the anterior
are distinct but overlapping entities. Ethnicity is difficult to pole of the lens and a horizontal line joining two scleral
measure accurately, and at present, is dependent on self- spurs on anterior segment OCT scans) and angle closure.52,53
assessment. However, in studies that use ethnicity or race With more research into risk factors in populations with a
as a source of variation, the implication is that it encom- high prevalence of PACG it may be possible to use some of
passes genetic, environmental, and cultural differences. these anterior segment parameters as tools for early detec-
The variation in glaucoma prevalence found in different tion of the disease.
races, ethnicities, and geographical locations will be due to
multiple causes.
PREVALENCE OF UNDETECTED GLAUCOMA
The proportion of glaucoma cases detected in population-
INTRAOCULAR PRESSURE
based surveys that were previously undiagnosed is high in
The evidence for a causal association between intraocular all regions of the world. The figure for the high-income
pressure (IOP) and glaucoma is robust. Randomized, con- areas such as Europe and Australia is fairly consistent at
trolled trials have shown that removal of the risk factor about 50% (Table 1-5). In the middle- and low-income
(higher IOP) can reduce disease.44,45 A ‘high’ IOP is arbi- regions of Asia and Africa the percentage is much higher,
trary based on the distribution of IOP in the population. In approaching 90% and above, and can be linked with late
Western populations, 21 mmHg is traditionally used as the presentation of glaucoma cases and the higher risk of glau-
cut-off point, as this represents the mean (16 mmHg) plus coma blindness in these populations. Contributing factors
2 standard deviations. The distribution of IOP in Asian to the late presentation of glaucoma patients in developing
populations is shifted to the left, that is, they have lower countries include poor public knowledge about glaucoma
IOPs with a lower mean, and the cutoff point is approxi- and limited access to healthcare services. In developed
mately 1–2 mmHg lower. In Mongolia and Singapore, the countries with better access to optometry and eye care ser-
mean and cutoff were 13 mmHg and 19 mmHg, respec- vices the principal reason for the poor glaucoma detection
tively.3,46 A higher prevalence of normal tension glaucoma rate is the lack of a good screening test for a largely asymp-
(NTG) has been observed in the Japanese population.47 The tomatic disease. However, within populations such as the
distribution of IOP in this population presents a cutoff point UK, demographic factors including ethnicity (African-
of 20 mmHg with a mean of 14.5 mmHg. Prevalence esti- Caribbean and Asian), older age, and lower socioeconomic
mates for NTG used 21 mmHg as the boundary; however, status have been associated with a higher risk of late pre-
authors from the Tajimi study suggested that there were sentation with glaucoma.54,55
very few patients that would have been recategorized as a Late presentation is more likely to lead to visual morbid-
result of this. There is a suggestion that African-Caribbean ity. These issues indicate that global and national health
people have higher IOP. The Barbados Eye Study (BES) inequalities could contribute to glaucoma blindness.
showed that the distribution of IOP in Blacks has a mean of
18.7 mmHg and standard deviation of 5.2 mmHg, which
was higher than the Mixed and White participants in the
study.9
Table 1-5  Proportion of People with Glaucoma Cases
Detected in Population-Based Surveys that were
PRIMARY ANGLE-CLOSURE GLAUCOMA: RISK Previously Undiagnosed
FACTORS AND MECHANISMS Population Studied Percentage of Cases
Previously Undiagnosed
A more detailed description of the factors predisposing indi-
viduals to angle closure is set out in another chapter but it South Africa (Temba)24 87% (POAG)
is worth touching on some of them in this section. It was Chennai – Southern India56 98.5% (POAG)
established by Ron Lowe that there is an association between Tema (West Africa)26 97% (POAG)
certain biometric characteristics and an increased risk of Los Angeles Latinos (LALES)12 75% (POAG)
developing angle closure, these being a smaller eye, shal- Australia – Blue Mountains Eye Study57 51% (POAG)
lower central anterior chamber, and steep corneal curva- Melbourne (Visual Impairment Project)7 50% (POAG)
ture and relatively anterior positioned lens. These risk Rotterdam58 53% (POAG)
factors for angle closure were also reported by Danish
8 SECTION 1  •  Glaucoma in the World

Spotlight 1  China Study Spotlight 2  What Do Prevalence and

Geographic Variations Tell Us?
Jost B Jonas
In recent population-based studies on Chinese people, the Kazuhisa Sugiyama and Farrah Ja’afar
prevalence of open-angle glaucoma varied between 1.6% The prevalence and geographic variations in glaucoma
in a Singaporean urban population,1 3.8% in South China have been studied extensively. The prevalence of primary
(Liwan District, Guangzhou),2 and 3.1% in Beijing in open-angle glaucoma (POAG) varies greatly among racial
North China.3 These prevalence data found in mainland and ethnic groups, and it is the highest in Black
China were similar to figures reported from Japan and populations. However, the prevalence of normal-tension
India, and were lower than glaucoma prevalence data glaucoma (NTG) is very high in Japan (Tajimi Study1), and
reported in some studies on Caucasian and African also in Korea (Namil Study).
population groups.3,4 One of the reasons for the Risk factors are clinically useful to assess the risk for
discrepancies between the studies may be ethnic glaucoma for the individual patient. Among them, age has a
differences with a presumably relatively high rate of more powerful influence on prevalence of POAG than racial
glaucoma in the African population groups. Other reasons and ethnic grouping does. The age-specific prevalence is an
may be differences in the examination technique and important assessment tool for ophthalmologists.
different definitions of glaucoma. An age-related increase A significant relationship between intraocular pressure
(IOP) and glaucoma has been demonstrated in
of the glaucoma prevalence as well as associations
epidemiological studies not only in hypertensive POAG but
between intraocular pressure and glaucoma and between also in NTG (Tajimi Study). These studies suggest that IOP
myopia and glaucoma were reported in all these studies. reduction helps to stop or at least suppress progression of
The ratio of open-angle glaucoma to primary angle- glaucoma. However, it seems to be difficult to halt
closure glaucoma was 2.6 to 1 in the Beijing Eye Study,3 progression in some NTG cases. Besides IOP, other
and it was 1.4 to 1 (2.1% to 1.5%) in the Liwan Eye mechanisms could be involved. The lower cerebrospinal
Study.2 In other Asian studies, the prevalence ratio of fluid (CSF) pressure may increase the risk for NTG.2 Studies
open-angle glaucoma to primary angle-closure glaucoma show that patients with NTG have lower CSF pressures,
was 3.9% to 0.6% in the Japanese Tajimi study,4 2.5% to which increases translaminar pressure differences.
0.1% in the Singapore Malay Eye Study,5 and 2.4% to An association between myopia and POAG/NTG is
0.8% in the Singaporean Tanjong Pagar study.1 Reasons supported by large population-based surveys. The
for discrepancies between the studies on the prevalence of geographic variations of myopia may affect the prevalence
angle-closure glaucoma may be, among others, differences of POAG/NTG.
in the prevalence of pseudophakia/aphakia since cataract With regard to the type of glaucoma, prevalence surveys
surgery may prevent the development (and progression) suggest blindness is more common with PACG and
of primary angle-closure glaucoma. The prevalence of secondary glaucoma than with POAG in Asian populations.
glaucoma-related bilateral blindness or unilateral We need screening programs to detect individuals with
blindness was significantly higher in the primary angle- occludable angles for preventing PACG, especially in
closure glaucoma groups than in the open-angle developing countries. On the other hand, more than 50%
of glaucoma patients remain undiagnosed in the United
glaucoma groups. It suggests that primary angle-closure
States and Europe. The Tajimi Study revealed that 90% of
glaucoma has a worse visual outcome and prognosis as POAG patients were previously undiagnosed. Therefore, a
compared with open-angle glaucoma. mass screening method for glaucoma is urgently needed.
References Three-dimensional evaluation of the optic disc by
1. Foster PJ, Oen FT, Machin D, et al. The prevalence glaucoma specialists is thought to be the ideal method, but
of glaucoma in Chinese residents of Singapore: imaging devices, such as OCT or confocal scanning laser
a cross-sectional population survey of the Tanjong ophthalmoscopy, may be useful to detect glaucoma in
Pagar district. Arch Ophthalmol 2000; community health screening.3
118(8):1105–1111. The primary goal of glaucoma therapy is to minimize
the lifetime risk for loss of vision and vision-related quality
2. He M, Foster PJ, Ge J, et al. Prevalence and
of life. The epidemiological studies in various regions give
clinical characteristics of glaucoma in adult us useful data regarding the proportion of individuals
Chinese: a population-based study in Liwan District, with glaucoma who were bilaterally blind and it appears
Guangzhou. Invest Ophthalmol Vis Sci 2006; to be higher in Blacks. Advanced stage and young age at
47(7):2782–2788. the time of diagnosis, inadequate IOP control, high rate of
3. Wang YX, Xu L, Yang H, Jonas JB. Prevalence of progression despite treatment, and undiagnosed glaucoma
glaucoma in North China. The Beijing Eye Study. Am J are all considered to be high risk for blindness in POAG/
Ophthalmol 2010; 150:917–924. NTG. It is vital to note that risks of blindness or vision loss
4. Iwase A, Suzuki Y, Araie M, et al. The prevalence of cannot be calculated assuming that all patients progress
primary open-angle glaucoma in Japanese: the Tajimi at the average rate.
Study. Ophthalmology 2004;111(9):1641–1648. References
5. Shen SY, Wong TY, Foster PJ, et al. The prevalence and 1. Iwase A, Suzuki Y, Araie M, et al. The prevalence of
types of glaucoma in Malay people: the Singapore primary open-angle glaucoma in Japanese: the Tajimi
Malay eye study. Invest Ophthalmol Vis Sci Study. Ophthalmology 2004;111:1641–1648.
2008;49(9):3846–3851. 2. Ren R, Jonas JB, Tian G, et al. Cerebrospinal fluid
pressure in glaucoma: a prospective study.
Ophthalmology 2010;117:259–266.
3. Ohkubo S, Takeda H, Higashide T, et al. A pilot study to
detect glaucoma with confocal scanning laser
ophthalmoscopy compared with nonmydriatic
stereoscopic photography in a community health
screening. J Glaucoma 2007;16:531–538.
 1  •  Prevalence and Geographical Variations 9

27. Rudnicka AR, Mt-Isa S, Owen CG, et al. Variations in primary open-
References angle glaucoma prevalence by age, gender, and race: a Bayesian meta-
1. Resnikoff S, Pascolini D, Etyaále D, et al. Global data on visual impair- analysis. Invest Ophthalmol Vis Sci 2006;47(10):4254–61.
ment in the year 2002. Bull World Health Organization 2004; 28. Friedman DS, Wolfs RC, O’Colmain BJ, et al. Prevalence of open-angle
82:844–51. glaucoma among adults in the United States. Arch Ophthalmol
2. Foster PJ, Burhmann R, Quigley HA, et al. The definition and classifi- 2004;122(4):532–8.
cation of glaucoma in prevalence surveys. Br J Ophthalmol 29. Day AC, Baio G, Gazzard G, et al. The prevalence of primary angle
2002;86:238–42. closure glaucoma in European derived populations: a systematic
3. Foster PJ, Baasanhu J, Alsbirk PH, et al. Glaucoma in Mongolia. A review. Br J Ophthalmol 2012;96:1162–7.
population-based survey in Hovsgol province, northern Mongolia. 30. Foster PJ, Oen FT, Machin D, et al. The prevalence of glaucoma
Arch Ophthalmol 1996;114:1235–41. in Chinese residents of Singapore: a cross-sectional population survey
4. www.globalaigs.org/pages/Consensus/3/3rdconsensus.php. of the Tanjong Pagar district. Arch Ophthalmol 2000;118:
5. Tielsch J, Sommer A, Katz J, et al. Racial variations in the prevalence 1105–11.
of primary open-angle glaucoma: the Baltimore Eye Survey. JAMA 31. He M, Foster PJ, Ge J, et al. Prevalence of clinical characteristics of
1991;266:369–74. glaucoma in adult Chinese: a population-based study in Liwan dis-
6. Bonomi L, Marchini G, Marraffa M, et al. Epidemiology of angle- trict, Guangzhou. Invest Ophthalmol Vis Sci 2006;47:2782–8.
closure glaucoma. Prevalence, clinical types, and association with 32. Chan W, Garcia JA, Newland HS, et al. Killing tow birds with one
peripheral anterior chamber depth in the Egna-Neumarkt glaucoma stone: the potential effect of cataract surgery on the incidence of
study. Ophthalmology 2000;107:998–1003. primary angle-closure glaucoma in a high-risk population. Clin Exp
7. Wensor MD, McCarty CA, Stanislavsky YL, et al. The prevalence of Ophthalmol 2012;40:128–34.
glaucoma in the Melbourne Visual Impairment Project. Ophthalmol- 33. Arvind H, George R, Raju P, et al. Glaucoma in aphakia and
ogy 1998;105:733–9. pseudophakia in the Chennai glaucoma study. Br J Ophthalmol
8. Klein BE, Klein R, Sponsel WE, et al. Prevalence of glaucoma. The 2005;89:699–703.
Beaver Dam Eye Study. Ophthalmology 1992;99:1499–504. 34. Grodum K, Heijl A, Bengtsson B. Risk of glaucoma in ocular hyperten-
9. Leske C, Connell A, Schahat A, et al. The Barbados Eye Study: sion with and without pseudo exfoliation. Ophthalmology 2005;
prevalence of open-angle glaucoma. Arch Ophthalmol 1994;112: 112(3):386–90.
821–9. 35. Thomas R, Nirmalan PK, Skishnaiah S. Pseudo exfoliation in south-
10. Mason RE, Kosoko O, Wilson MR, et al. National survey of the preva- ern India. The Andra Pradesh Eye Disease Study. Invest Ophthalmol
lence and risk factors of glaucoma in St Lucia, West Indies I: preva- Vis Sci 2005;46:1170–6.
lence findings. Ophthalmology 1989;96:1363–8. 36. Arvind H, Raju P, Paul PG, et al. Pseudo exfoliation in south India. Br
11. Quigley HA, West SK, Rodriguez J, et al. The prevalence of glaucoma J Ophthalmol 2003;87(11):1321–3.
in a population-based study of Hispanic subjects: Proyecto VER. Arch 37. Kanthan GL, Mitchell P, Burlutsky G, et al. Pseudo exfoliation syn-
Ophthalmol 2001;119:1819–26. drome and long-term incidence of cataract and cataract surgery: The
12. Varma R, Ying-Lai M, Francis BA, et al. Prevalence of open-angle Blue Mountains Eye Study. Am J Ophthalmol 2013;155:83–8.
glaucoma and ocular hypertension in Latinos: the Los Angeles Latino 38. Jonasson F, Damji KF, Arnarsson A, et al. Prevalence of open-angle
Eye Study. Ophthalmology 2004;111:1434–9. glaucoma in Iceland: Reykjavik Eye Study. Eye 2003;17(6):747–53.
13. Sakata K, Sakata LM, Sakata VM, et al. Prevalence of glaucoma in a 39. Herndon LW, Chalia P, Ababio-Danso B, et al. Survey of glaucoma in
South Brazilian population: Projeto Glaucoma. Invest Ophthalmol Vis an eye clinic in Ghana, West Africa. J Glaucoma 2002;11:421–5.
Sci 2007;48:4974–9. 40. Teshome T, Regassa K. Prevalence of pseudo exfoliation syndrome in
14. Quigley HA, Broman AT. The number of people with glaucoma world- Ethiopian patients scheduled for cataract surgery. Acta Ophthalmol
wide in 2010 and 2020. Br J Ophthalmol 2006;90:262–7. Scand 2004;82(3 Pt 1):254–8.
15. www.globalhealthfacts.org. 41. Challa P. Genetics of pseudo exfoliation syndrome. Curr Opin Ophthal-
16. Liang Y, Friedman DS, Zhou Q, et al. Prevalence and characteristics mol 2009;20:88–91.
of primary angle closure disease in a rural adult Chinese population: 42. Damji KF, Bains HS, Stefansson E, et al. Is pseudo exfoliation syndrome
The Handan Eye Study. IOVS 2011;52:8672–9. inherited? A review of genetic and nongenetic factors and a new
17. Casson RJ, Newland HS, Mueke J, et al. Prevalence of glaucoma observation. Ophthalmic Genet 1998;19:175–85.
in rural Myanmar: The Meiktila Eye Study. Br J Ophthalmol 43. Rose S, Kamin L, Lewontin R. Not in our genes: biology, ideology and
2007;91:710–14. human nature. New York: Penguin; 1984.
18. Song W, Shan L, Cheng F, et al. Prevalence of glaucoma in a rural 44. The AGIS Investigators. The Advanced Glaucoma Intervention Study
northern China adult population: A population based survey in Kailu (AGIS): 7. The relationship between control of intraocular pressure
County, Inner Mongolia. Ophthalmology 2011;118:1982–8. and visual field deterioration. Am J Ophthalmol 2000;130(4):
19. Shiose Y, Kitazawa Y, Tsukahara S, et al. Epidemiology of glaucoma 429–40.
in Japan – a nationwide glaucoma survey. Jpn J Ophthalmol 45. Heijl A, Leske MC, Bengtsson B, et al. Reduction of intraocular pres-
1991;35:133–55. sure and glaucoma progression: results from the Early Manifest Glau-
20. Kim CS, Seong GJ, Lee NH, et al. Prevalence of primary open-angle coma Trial. Arch Ophthalmol 2002;120(10):1268–79.
glaucoma in central South Korea, the Namil study. Ophthalmology 46. Foster PJ, Machin D, Wong TY, et al. Determinants of intraocular
2011;118(6):1024–30. pressure and its association with glaucomatous optic neuropathy in
21. Vijaya L, George R, Arvind H, et al. Prevalence of angle-closure Chinese Singaporeans: the Tanjong Pagar Study. Invest Ophthalmol
disease in a rural southern Indian population. Arch Ophthalmol Vis Sci 2003;44(9):3885–91.
2006;124:403–9. 47. Iwase A, Suzuki Y, Araie M, et al. The prevalence of primary open-
22. Dandona L, Dandona R, Mandal P, et al. Angle closure glaucoma in angle glaucoma in Japanese: the Tajimi Study. Ophthalmology
an urban population in Southern India. The Andhra Pradesh Eye 2004;111(9):1641–8.
Disease Study. Ophthalmology 2000;107:1710–16. 48. Lowe RF. Primary angle-closure glaucoma: a review of ocular biom-
23. Buhrmann RR, Quigley HA, Barron Y, et al. Prevalence of glaucoma etry. Aust J Ophthalmol 1977;5:9–17.
in a rural East African population. Invest Ophthalmol Vis Sci 49. Alsbirk PH. Anterior chamber depth and primary angle-closure glau-
2000;41:40–8. coma. I. An epidemiologic study in Greenland Eskimos. Acta Ophthal-
24. Rotchford AP, Kirwan JF, Muller MA, et al. Temba glaucoma study: a mol 1975;53:89–104.
population-based cross-sectional survey in urban South Africa. Oph- 50. Clemmesen V, Alsbirk PH. Primary angle-closure glaucoma in Green-
thalmology 2003;110:376–82. land. Acta Ophthalmol 1971;49:47–58.
25. Rotchford AP, Johnson GJ. Glaucoma in Zulus: a population-based 51. Foster PJ, Alsbirk PH, Baasanhu J, et al. Anterior chamber depth in
cross-sectional survey in a rural district in South Africa. Arch Oph- Mongolians: variation with age, sex, and method of measurement.
thalmol 2002;120:471–8. Am J Ophthalmol 1997;124:53–60.
26. Budenz DL, Barton K, Whiteside-de-Vos J, et al. Prevalence of glau- 52. Tan GS, He M, Zhao W, et al. Determinants of lens vault and associa-
coma in an urban West African population: The Tema Eye Survey. tion with narrow angles in patients from Singapore. Am J Ophthalmol
JAMA Ophthalmol 2013;131:651–8. 2012;154:39–46.
10 SECTION 1  •  Glaucoma in the World

53. Lee RY, Huang G, Cui QN, et al. Association of lens vault with 57. Mitchell P, Smith W, Attebo K, et al. Prevalence of open-angle glau-
narrow angles among different ethnic groups. Curr Eye Res 2012; coma in Australia. The Blue Mountains Eye Study. Ophthalmology
37:486–91. 1996;103:1661–9.
54. Fraser S, Bunce C, Wormald R. Risk factors for late presentation in 58. Dielemans I, VIngerling JR, Wolfs RCW, et al. The prevalence of
chronic glaucoma. Invest Ophthalmol Vis Sci 1999;40:2251–7. primary open-angle glaucoma in a population-based study in The
55. Fraser S, Bunce C, Wormald R, et al. Deprivation and late presentation Netherlands. Ophthalmology 1994;11:1851–5.
of glaucoma: case-control study. Br Med J 2001;322:639–43.
56. Vijaya L, George R, Paul PG, et al. Prevalence of open-angle glaucoma
in a rural south Indian population. Invest Ophthalmol Vis Sci
2005;46:4461–7.