Prevalence

Journal of Current of Glaucoma

Glaucoma in India:
Practice, A Review
Sept.-Dec. 2007;1(2):7-11

Prevalence of Glaucoma in India: A Review

Ronnie George, Lingam Vijaya
Medical Research Foundation, Sankara Nethralaya, Chennai, India

Glaucoma is estimated to affect 60.5 million persons worldwide field ) damage to diagnose glaucoma. The Chennai Glaucoma
by the year 2010.1 The estimated prevalence of glaucoma for Study (CGS) and the West Bengal Glaucoma Study (WBGS)
India is 11.9 million.2 These estimates have been derived from used the ISGEO guidelines for diagnosing glaucoma in
population based studies conducted worldwide. The availability population based studies.7
of population based data from India is relatively recent as
compared to Western countries. We present a summary of the Ocular hypertension: The rates of OHT have been reported
findings from different population based studies in the country. from the VES,2 APEDS8 and ACES.4 There are wide variations
These have been five populations based studies, three from in the reported rates. The APEDS reported the lowest rate of
the state of Tamil Nadu, one from Andhra Pradesh and one 0.42% (95% CI: 0.11, 1.12%). There were 1.1% (95% CI: 0.84,
from Bengal (Table 1).2-6 Data from North India is unfortunately 1.41%) persons with OHT among those aged 40 years and older
lacking. These studies have been carried out from 1993 to 2003. in ACES. The VES reported that 3.08% (95% CI: 1.98, 4.19%)
To our best knowledge at least one more population based were ocular hypertensive. Pseudoexfoliation was commonly
study (from Central India) is underway. All these studies have associated with ocular hypertension 4.22% of those with PXF
used differing methodology and diagnostic criteria for glaucoma. in the ACES9 and 9.33% of those with PXF in the CGS (Rural)
To improve comparability the prevalence reported by different were ocular hypertensive.10
studies has been age standardized to the population of India Primary open angle glaucoma: There are again wide variations
(2001 census) where possible. A summary of the study design in the prevalence of POAG between the five studies (Table 2).
and diagnostic criteria is provided in Table 1. The VES reported the lowest rates of 0.41%.2 There are a number
of reasons for this, the prime cause being that the study was
Diagnostic Criteria
limited to the age group of 30-60 years. The prevalence of POAG
The diagnostic criteria for glaucoma used by the different increased with age in all the studies discussed here. On
studies have varied . All have required optic disc changes to comparing the prevalence of POAG in those is the 30-60 age
diagnose glaucoma. Most have not used intraocular pressure group from APEDS (1.01%)8 the VES reported prevalence is not
for glaucoma diagnosis except when visual field or optic disc dissimilar (0.41%, 95% CI: 0.08, 0.81%). The other possible
data is not available. This is in accordance with the International reason is the low rate of visual field performance in the VES.
Society for Geographical and Epidemiological Ophthalmology Since a visual field defect was essential for diagnosis, those
(ISGEO) emphasis on structural (optic disc) or functional (visual who did not perform visual fields would not be classified as

Table 1: Summary of the different population based studies from India

Study Study period Setting Age group Number examined IOP Optic Disc Visual Fields
(response rate %)

VES 1994 Urban 30-60 972 (50.3) +/- + +

APEDS 1996-2000 Rural, Urban All ages 10,273(87.3) - + +/–
ACES 1995-97 Rural 40 + 5150 (93.0) - + +/–
CGS 2001-03 Rural, Urban 40 + 7774 (81.0) -* + +/–
WBGS 1998-99 Rural 50 + 1324(83.1) -* + +/–

*The CGS and the WBGS used the ISGEO criteria (with minor modifications) to diagnose disease. An IOP level that exceeds the 99.5Th percentile for
a normal population is used to diagnose disease only when the optic disc cannot be visualized and visual fields are not possible.

7
 Ronnie George, Lingam Vijaya

Table 2: Age wise prevalence of POAG in different studies

APEDS POAG ACES POAG CGS (Rur) POAG CGS (Urban) POAG WBGS POAG
(n = 934) (n = 5150) (n = 3924) (n = 3850) (n = 1269)

40-49 395 1.27 2066 0.34 1585 0.63 1419 2.26 -

50-59 260 2.31 1466 1.57 985 1.62 1120 3.57 628 2.55
60-69 184 4.89 1201 1.83 892 2.58 906 4.08 409 2.69
70+ 95 6.32 417 2.88 462 3.25 405 6.42 231 4.76
Reported
prevalence % 2.56 1.7 1.62 3.51
(95% CI) (1.22,3.91) (1.3,2.1) (1.42,1.82) (3.04,4.0) 2.99

VES reported prevalence ( 30-60 yrs): 0.41%( 95%CI: 0.08,0.81)

glaucoma, the high (51.5 %) non response rates could affect Table 3: Age standardized rates for POAG, PACG, angle closure
case detection. With the exception of the VES the reported disease and PXF for the population aged above 40 years
(Standardized to the population of India 2001 census)
prevalence rates for POAG appeared to be higher in urban
population (CGS (Urban)11 and APEDS8). While the urban Study PXF POAG PACG Angle closure
populations were older than the rural population this is not the disease
reason for the increased prevalence as the age standardized APEDS 1.94 2.85 1.21 3.71
rates (to the Indian Population 2001) (Table 3) show the same ACES 6.15 1.26 NR NR
trend. Urban India is known to have higher reported rates of CGS (Rural) 3.62 1.58 0.85 7.75
diabetes and cardio-vascular diseases. We speculate that
CGS (Urban) NR 3.46 0.80 10.43
lifestyle changes and their related cardio-vascular influences
WBGS NR 3.06 NR NR
may account in part of this increase in prevalence.
The proportion of persons with POAG who presented with NR : Not reported/ age wise prevalence not reported. VES data not
a “normal” IOP(defined as two standard deviations above the available. WBGS data reported only for the 50 + age group (Age
Standardized POAG prevalence: 3.06%. Comparable adjusted rates for
population mean) was significant in all the prevalence studies.
POAG in the 50 + age group: APEDS =4.00%, ACES = 1.93%, CGS(R)
Sixty five percent of those with POAG in APEDS,8 45% in ACES,4 =2.28%, CGS (U) = 4.34%.
67% in CGS (rural)12 and 82% in CGS (urban)11 had a normal
presenting intraocular pressure. This is again one of the
contributory factors for the large proportion of undiagnosed Myopia was also a risk factor for POAG in the ACES. No
disease – a single normal IOP does not rule out disease. Optic association with refractive error was reported by any of the
disc evaluation is mandatory in order to identify those with other studies. 4
glaucoma.
Angle closure disease: The disparities in reported prevalence
Risk factors: Increasing age was the consistent risk factor for of angle closure disease are perhaps most representative of the
all studies. Males were at greater risk of POAG in the ACES.4 influence of the different methodology and diagnostic criteria
No gender difference was reported by any of the other studies. used by these studies (Table 4). Detailed information regarding

Table 4: Age wise prevalence of angle closure disease in different studies

APEDS APEDS APEDS CGS (RUR) CGS (RUR) CGS (RUR) CGS (RUR) CGS (Urban)CGS (URB)CGS (URB)
(n = 934) (ACG) PACS (n = 3924) PACG PAC PACS (n=3850) PACG PAC

40-49 395 0.00 0.76 1585 0.44 0.38 4.29 1419 0.07 1.27 4.86
50-59 260 1.54 3.08 985 1.02 0.81 7.31 1120 0.63 3.30 7.77
60-69 184 2.17 3.80 892 1.01 1.01 7.96 906 2.21 4.19 9.27
70+ 95 3.16 5.26 462 1.73 1.08 7.58 405 1.48 3.21 9.38

Reported
prevalence 1.08 2.21 0.87 0.71 6.27 0.88 2.75 7.24
% (95% CI) (0.36,1.80) (1.15,3.27) (0.58,1.16) (0.45,0.98) (5.51,7.03) (0.6,1.16) (2.01,3.49) (6.58,8.02)

VES reported prevalence (30-60 yrs): 4.32 (95% CI: 3.01,5.63), ACES reported prevalence (40 yrs or more): 0.5% (95% CI:0.3,0.7)

8
 Prevalence of Glaucoma in India: A Review

the prevalence of angle closure disease (including PACG) is axial length and a shallower anterior chamber and thicker lenses
not available from the ACES and WBGS. than normals by the CGS. 16
The VES was the first to report that PACG comprised a
Pseudoexfoliation is one of the risk factors for glaucoma and its
significant proportion of those with glaucoma.2 A number of
prevalence in clinic based studies from India has been variously
persons were also reported to have occludable angles
reported to be between 1.87 and 13.5 %.9,10,17 Population based
(equivalent to PACS). While the definition of “Occludability”
data on the prevalence of Pseudoexfoliation are available from
was relatively lenient – this probably reflects clinical practice at
three of the studies, it is reported to affect between 3% and 6%
many institutions in the country and is more stringent than the
of those aged of 40 years or older. The prevalence of glaucoma
recent recommendations of the World Glaucoma Association.
among those with Pseudoexfoliation was reported to be 3.02%
The VES definition of PACG encompassed those now classified
in the APEDS,17 7.5% in ACES9 and 13% in the rural cohort of
as PAC in addition to those with PACG. Five persons (0.51%)
the CGS10. The risk of Pseudoexfoliation increased with
were reported to have angle closure disease with disc and field
increasing age.9,10,17 No gender predilection was reported for
damage. These would correspond to a diagnosis of PACG; this
the disease. Those predominantly involved in outdoor activities
prevalence is similar to other reported rates of PACG by studies
were at greater risk of PXF than those whose occupation was
in South India.4,13,14,15
classified as indoor.17 A trend towards increased PXF among
Subsequent reports from the APEDS13 did report a lower
those from the lower socio-economic classes was also noted in
prevalence of angle closure disease. The difference was most
the APEDS, this was however not significant. 17
marked in the lower prevalence of PACS. This can partly be
The wide variations in the prevalence of Pseudoexfoliation
explained by a stricter criterion for occludability ( non visibility
may represent inter-regional differences or may arise out of
of the pigmented trabecular meshwork in 270° or more of the
differing examination techniques.
filtering angle).
The prevalence of angle closure disease in the CGS14,15 is Secondary glaucoma: The rates of secondary glaucoma have
again similar to that reported by the VES. The urban area of the been reported by some studies. The WBGS reported a rate of
CGS had higher rates of PAC, PACS and PACG than the rural 0.08%,6 APEDS : 0.21% in those aged of 30 and above 8,13 and
cohort.14,15 The VES and the CGS used similar gonioscopic 0.3% in the ACES.4 These reports are probably an underestimate
criteria to diagnose a narrow angle (needing 180° of the angle to as many of these eyes may not have had useful vision or have
be closed as defined) unlike other studies that needed three- a disorganized anterior segment and would nor meet the
fourths of the angle to be narrow before grading as occludable. diagnostic criteria of the studies.
This difference in grading using 180 or 270° is unlikely to result The report from the CGS highlighting the high rates of
in such large discrepancies in the prevalence of angle closure glaucoma (11.2% of the aphakes/pseudophakes examined)
disease. Glaucoma specialists are more likely to identify among those who had undergone cataract surgery directs
gonioscopic abnormality, gonioscopy in both the CGS and the attention towards a large at risk population in India.18 Those
VES was performed by persons with specialized glaucoma who have undergone cataract surgery are in the same age cohort
training, and could be one of the reasons contributing to the at risk for glaucoma. An inadequate pre-operative assessment
similarity in results. or poor quality of surgery may precipitate glaucoma is
In all studies the vast majority of person with PACG were susceptible individuals. The higher rates of glaucoma among
asymptomatic, the disease was silent and chronic. Failure to those with aphakia are indications of the proportion of iatrogenic
perform gonioscopy could result in misdiagnosis as angle glaucoma.
closure glaucoma. The CGS did report that of those with PACG
Blindness and glaucoma: Glaucoma is the second leading cause
who were previously diagnosed to have glaucoma 40% were
of blindness in the adult population in India.19,20 The proportion
misdiagnosed as POAG.
of those diagnosed to be bilaterally blind because of POAG in
Risk factors: Increasing age was a common risk factor for PACG The APEDS, ACES, CGS (rural), CGS (Urban) and WBGS was
in all studies. Female gender (CGS) was a risk factor for PACG 11.1%,1.6%,3.2%,1.5% and 5.2% respectively. The
and PAC,14,15 more women in the APEDS had PACG but this corresponding figures for PACG for APEDS, CGS (rural), CGS
difference was not statistically significant.13 Hyperopia was also (Urban) and WBGS were 16.6%, 2.9% and 5.9%. The WBGS
a risk factor in the Chennai glaucoma a study (Urban),15 a trend reported only 3 cases of PACG none of whom were blind. In
of increased risk with hyperopia was noted both in APEDS and those studies that reported significant rates of angle closure
CGS (Rural).13,14 Among systemic conditions diabetes mellitus disease PACG caused between one and four times the proportion
was reported to be a risk factor.15 Biometric data at the baseline of blindness that POAG did.
visit was available only from the Chennai Glaucoma Study. Eyes The high rates of blindness in India can be explained, at
with angle closure disease had been reported to have shorter least partially, by the large proportion of undiagnosed disease

9
 Ronnie George, Lingam Vijaya

in the population’s studies. The vast majority of disease was

undiagnosed with more than 90% being diagnosed during the
study examination. The ACES reported that 50% of those with
glaucoma had undergone an eye examination by an
ophthalmologist previously.4 Even in this group 80% of were
still undetected at the time of the survey. This highlights the
low case detection rates in the ophthalmic community in India.

Incidence Studies in Glaucoma

The only study from which incidence data is currently available
from India is the Vellore Eye Survey, the Chennai Eye Disease
Fig. 1: Prevalence of POAG and age
Incidence Survey (CEDIS) a follow up from the Chennai
Glaucoma Study cohort is currently underway.
The VES re-examined all those classified as PACS,21 PAC22 increases substantially with age. The majority of the Indian
and OHT23 five years after the initial examination in addition to population is less than 30 years of age. With increasing life
examining a random subset of previously normal persons. expectancy the number of those at risk of glaucoma will grow
Eleven (22%(95% CI: 9.8 to 34.2)) of 50 persons with PACS exponentially over time. Improving our case detection rates is
progressed to develop PAC over a five years period, none of utmost importance if we are to make an impact on the burden
progressed to PACG (seven synechial and four appositional).21 of disease in the country. The vast majority of glaucoma in
All those who progressed had bilateral PACS at baseline. India is undiagnosed; there is an urgent need to adopt compre-
Though persons with PACS had significantly shallower anterior hensive eye examinations, which include IOP measurement,
chamber depth than normal no biometric risk factors for gonioscopy and optic disc evaluation, as a routine.
progression could be identified. One previously normal person
progressed to PAC with time, the relative risk for progression REFERENCES
among those with PACS to PAC was 24.2 (95%CI: 3.2 to 182.4). 1. Quigley HA, Broman AT. The number of people with glaucoma
Eight (28.5%, 95% CI: 12-45%) of 28 persons diagnosed to have worldwide in 2010 and 2020. Br J Ophthalmol 2006;90(3):262-
PAC at baseline progressed to PACG over similar time frame.22 7.
Of these nine had undergone YAG PI, of whom one (11.1%) 2. Jacob A, Thomas R, Koshi SP, Braganza A, Muliyil J. Prevalence
progressed. Seven (36.8%) of the 19 who refused YAG PI at the of primary glaucoma in an urban south Indian population. Indian
baseline progressed. None were blind due to glaucoma. J Ophthalmol 1998;46(2):81-6.
In both groups no patients suffered from an acute angle 3. Arvind H, Paul PG, Raju P, Baskaran M, George R, Balu S, et al.
closure attack during the five years period. Methods and design of the Chennai Glaucoma Study. Ophthalmic
Twenty five of the twenty nine ocular hypertensives Epidemiol 2003;10(5):337-48.
4. Ramakrishnan R, Nirmalan PK, Krishnadas R, Thulasiraj RD,
diagnosed at baseline were re-examined.23 After correction for
Tielsch JM, Katz J, et al. Glaucoma in a rural population of
CCT they were reclassified as ocular hypertensive if the southern India: the Aravind comprehensive eye survey.
corrected IOP was above 21 mm Hg. Among untreated ocular Ophthalmology 2003;110(8):1484-90.
hypertensives 17.4% (95% CI: 1.95 to 32.75 %) progressed to 5. Dandona R, Dandona L, Naduvilath TJ, Nanda A, McCarty
develop POAG (optic disc and visual field changes) over a five CA. Design of a population-based study of visual impairment in
years period. One of the normals controls had progressed to India: The Andhra Pradesh Eye Disease Study. Indian J
develop POAG. The relative risk of POAG among those with Ophthalmol 1997;45(4):251-7. Review.
untreated OHT was 19.1(95% CI: 2.2. 163.5).Those who 6. Raychaudhuri A, Lahiri SK, Bandyopadhyay M, Foster PJ,
progressed had wider diurnal fluctuations in IOP. Reeves BC, Johnson GJ. A population based survey of the
The numbers in each cohort are small with consequently prevalence and types of glaucoma in rural West Bengal: the
wide confidence intervals. However, this is the only incidence West Bengal Glaucoma Study. Br J Ophthalmol 2005;89(12):
1559-64.
study from India and gives us an estimate of the risk of disease
7. Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The definition
among those with risk factors for glaucoma. and classification of glaucoma in prevalence surveys. Br J
The population based studies that report glaucoma Ophthalmol 2002;86(2):238-42. Review.
prevalence provide important information. The Indian 8. Dandona L, Dandona R, Srinivas M, Mandal P, John RK,
population has substantial risk for POAG with the additional McCarty CA, et al.Open-angle glaucoma in an urban population
risk of PACG and large burden of angle closure disease. This in southern India: the Andhra Pradesh eye disease study.
assumes greater importance as the risk of all forms of glaucoma Ophthalmology 2000;107(9):1702-9.

10
 Prevalence of Glaucoma in India: A Review

9. Krishnadas R, Nirmalan PK, Ramakrishnan R, Thulasiraj RD, 19. Vijaya L, George R, Arvind H, Baskaran M, Raju P, Ramesh SV,
Katz J, Tielsch JM, et al. Pseudoexfoliation in a rural population et al. Prevalence and causes of blindness in the rural population
of southern India: the Aravind Comprehensive Eye Survey. Am of the Chennai Glaucoma Study. Br J Ophthalmol
J Ophthalmol 2003;135(6):830-7. 2006;90(4):407-10.
10. Arvind H, Raju P, Paul PG, Baskaran M, Ramesh SV, George 20. Thulasiraj RD, Nirmalan PK, Ramakrishnan R, Krishnadas R,
RJ, et al. Pseudoexfoliation in South India. Br J Ophthalmol Manimekalai TK, Baburajan NP, et al. Blindness and vision
2003;87(11):1321-3. impairment in a rural south Indian population: the Aravind
11. Vijaya L, George R, Baskaran M, Arvind H, Raju P, Ramesh SV, Comprehensive Eye Survey. Ophthalmology 2003;110(8):1491-
8.
et al. Prevalence of Primary Open-angle Glaucoma in an Urban
21. Thomas R, George R, Parikh R, Muliyil J, Jacob A. Five year
South Indian Population and Comparison with a Rural
risk of progression of primary angle closure suspects to primary
Population the Chennai Glaucoma Study. Ophthalmology 2007;
angle closure: a population based study. Br J Ophthalmol
[Epub ahead of print]. 2003;87(4):450-4.
12. Vijaya L, George R, Paul PG, Baskaran M, Arvind H, Raju P, et 22. Thomas R, Parikh R, Muliyil J, Kumar RS. Five-year risk of
al. Prevalence of open-angle glaucoma in a rural south Indian progression of primary angle closure to primary angle closure
population. Invest Ophthalmol Vis Sci 2005;46(12):4461-7. glaucoma: a opulation-based study. Acta Ophthalmol Scand
13. Dandona L, Dandona R, Mandal P, Srinivas M, John RK, 2003;81(5):480-5.
McCarty CA, et al. Angle-closure glaucoma in an urban 23. Thomas R, Parikh R, George R, Kumar RS, Muliyil J. Five-year
population in southern India. The Andhra Pradesh eye disease risk of progression of ocular hypertension to primary open
study. Ophthalmology 2000;107(9):1710-6. angle glaucoma. A population-based study. Indian J Ophthalmol
14. Vijaya L, George R, Arvind H, Baskaran M, Paul PG, Ramesh 2003;51(4):329-33.
SV, et al. Prevalence of angle-closure disease in a rural southern
Indian population. Arch Ophthalmol 2006;124(3):403-9.
15. Vijaya L, George R, Arvind H, Baskaran M, Ve Ramesh S, Raju
P, et al. Prevalence of Primary Angle-Closure Disease in an
Urban South Indian Population and Comparison with a Rural
Population the Chennai Glaucoma Study. Ophthalmology 2007;
[Epub ahead of print].
16. George R, Paul PG, Baskaran M, Ramesh SV, Raju P, Arvind H,
et al. Ocular biometry in occludable angles and angle closure
glaucoma: a population based survey. Br J Ophthalmol
2003;87(4):399-402.
17. Thomas R, Nirmalan PK, Krishnaiah S. Pseudoexfoliation in
southern India: the Andhra Pradesh Eye Disease Study. Invest
Ophthalmol Vis Sci 2005;46(4):1170-6.
18. Arvind H, George R, Raju P, Ramesh SV, Baskaran M, Paul PG,
et al. Glaucoma in aphakia and pseudophakia in the Chennai Dr Ronnie George
([email protected])
Glaucoma Study. Br J Ophthalmol 2005;89(6):699-703.

11