Article Text

Download PDFPDF

Prevalence of corneal diseases in the rural Indian population: the Corneal Opacity Rural Epidemiological (CORE) study
  1. Noopur Gupta1,
  2. Praveen Vashist1,
  3. Radhika Tandon1,
  4. Sanjeev K Gupta2,
  5. Sadanand Dwivedi3,
  6. Kalaivani Mani3
  1. 1Dr Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India
  2. 2Centre for Community Medicine, All India Institute of Medical Sciences, New Delhi, India
  3. 3Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India
  1. Correspondence to Dr Praveen Vashist, Dr Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, Room No. 787, 7th floor, Ansari Nagar, New Delhi 110029, India; PRAVEENVASHIST{at}YAHOO.COM

Abstract

Objective The present population-based study was undertaken to estimate the prevalence, determinants and causes of corneal morbidity and blindness in a rural North Indian population.

Design Population-based study in India with 12 899 participants of all ages.

Methods Participants were recruited from 25 village clusters of district Gurgaon, Haryana, India using random cluster sampling strategy. All individuals were examined in detail with a portable slit lamp for evidence of any corneal disease during the door-to-door examination. Comprehensive ocular examination including logMar visual acuity, slit lamp biomicroscopy, non-contact tonometry and dilated retinal evaluation was performed at a central clinic site in the respective villages.

Results Overall, 12 113 of 12 899 people (93.9% response rate) were examined during the household visits. Prevalence of corneal disease was 3.7% (95% CI 3.4% to 4.1%) and that of corneal blindness was 0.12% (95% CI 0.05% to 0.17%). Multivariable analysis demonstrated that corneal disease was significantly higher in the elderly (p<0.0001) and illiterates (p<0.0001). Common causes of corneal opacity in the study population were pterygium (34.5%), ocular trauma (22.3%) and infectious keratitis (14.9%). Corneal diseases contributing to blindness were post-surgical bullous keratopathy (46.2%) and corneal degenerations (23.1%).

Conclusions The study findings demonstrate that currently ocular trauma, infectious keratitis, post-surgical bullous keratopathy, and corneal degenerations are responsible for the major burden of corneal blindness and morbidity in the Indian population. The prevalence of corneal morbidity due to vitamin A deficiency and trachoma was low in this rural population.

  • Cornea
  • Epidemiology
  • Infection
  • Public health

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Corneal blindness, including corneal opacity and trachoma, contribute to 7% of the blindness burden globally.1 ,2 This translates to nearly 2.7 million blind people worldwide due to corneal diseases.2 The global burden of corneal blindness is concentrated in emerging and developing countries. The major causes of corneal blindness include corneal ulceration, ocular trauma, trachoma, bullous keratopathy, corneal degenerations and vitamin A deficiency. It has been reported that nearly 90% of the global cases of ocular trauma and corneal ulceration leading to corneal blindness occur in developing countries.3 Being a developing country with a predominantly rural population, India faces a significant challenge in eliminating corneal blindness. A meta-analysis of various population-based blindness studies conducted in India demonstrated that the prevalence of corneal blindness was 0.45% (95% CI 0.27% to 0.64%) in adults.4

There is an absence of representative community-based data on the magnitude of corneal diseases in the general population, especially in North India. To the best of our knowledge, this study is the first of its kind to reflect the magnitude of corneal opacity in a large sample of the Indian population. Available literature on corneal blindness in India is either hospital-based5 or extrapolated data from blindness surveys that were not specifically designed for studying corneal morbidity.6–12 Hence, this population-based study provides new insights into the prevalence, risk factors and causes of corneal blindness and morbidity across all age groups in a rural Indian population.

Methods

Study design and population

The Corneal Opacity Rural Epidemiological (CORE) study is a cross-sectional study involving cluster random sampling. It was conducted from July 2011 to January 2013 to determine the prevalence and determinants of corneal opacity in the rural regions of Gurgaon district, state of Haryana, India. A sample size of 9640 was calculated assuming 4% prevalence of corneal opacity,13 design effect of 2, 0.6% absolute precision, and response rate of 85%. It was planned to cover a sample of 10 000 people from 25 randomly selected rural clusters. A computerised simple random sampling approach was used to select 25 clusters from a sampling frame of 910 clusters generated from the total rural population of 472 085 from 229 villages of District Gurgaon.14 The study protocol received ethical approval from the Institutional Ethics Committee, All India Institute of Medical Sciences, New Delhi in April 2010. Written informed consent for enrolment and examination was obtained from all adults for their own participation and from parents or appropriate guardians of all eligible children before they were included in the study in accordance with the principles of the Declaration of Helsinki.

A house-to-house visit was made to each family in the selected village cluster. All eligible individuals were examined by an experienced ophthalmologist for the presence of corneal opacity using a portable slit-lamp. Individuals detected with corneal opacification during the house-to-house examination were invited for a detailed ocular examination at a temporary makeshift clinic utilising the local infrastructure of the village that was easily accessible to the enumerated study population of the concerned village. People who were bedbound or those who failed to attend the central clinical facility in the village were examined at home.

The comprehensive ocular examination of patients with corneal opacity included visual acuity estimation with the logMAR tumbling E chart for distance and near, non-cycloplegic refraction and prescription (cycloplegic refraction for children), non-contact tonometry, lens examination and cataract assessment by the Lens Opacity Classification System III, slit lamp biomicroscopy with fluorescein staining, and detailed retinal evaluation. The corneal examination was performed under high magnification and emphasis was laid on the aetiopathogenesis, morphological characteristics and management of the corneal disease. In the present study, blindness was defined as visual acuity <3/60 in the better eye with available correction.

Quality assurance and standardisation of all study procedures and equipments was maintained throughout the period of the study. A single cornea specialist conducted the ophthalmic examination on the entire study population. The items of ophthalmic equipment were calibrated and standardised at regular intervals. Any potential observer or measurement bias was thus reduced.

Statistical analysis

The data were entered into a specially designed Microsoft Access based platform with inbuilt validation and consistency checks. Data analysis was carried out using Stata V.12.0 (Stata, College Station, Texas, USA). Qualitative data has been described as number (%) and quantitative data has been described as mean±SD and median (range) as appropriate. The multivariable logistic regression analysis was done to find the association between sociodemographic factors and the presence of corneal opacity and corneal blindness. The results for the same were reported as OR (95% CI). A value of p<0.05 was considered statistically significant.

Results

The total enumerated population in the 25 clusters of Rural Gurgaon was 12 899. Out of these, 12 113 individuals completed all the study procedures, with a coverage of 93.9%. The synopsis of the study participants recruited at each step has been summarised in figure 1.

Figure 1

Flow chart summarising number of participants at various steps of the study.

Sociodemographic profile

The age and gender distribution of the enumerated and the covered population was similar and it matched the national demographic profile of the Indian population. Although 6.1% of the sample population could not be covered, the enumerated population and the covered population had comparable sociodemographic characteristics. Males comprised 51% of the study population (table 1).

Table 1

Sociodemographic profile of the enumerated and covered study population in rural Gurgaon

Prevalence of corneal opacity

The prevalence of corneal opacity in this study population was 3.7% (95% CI 3.4% to 4.1%). During house-to-house examination, 452 people were detected with corneal opacity and comprehensive ocular examination was possible for 435 participants (96.2%). The prevalence increased with age and was highest for elderly patients (26.1%, 95% CI 23.4% to 28.8%). The prevalence of corneal opacity was higher among females (4.1% 95% CI 3.6% to 4.6%) compared to males (3.4%, 95% CI 2.9% to 3.8%), the difference being statistically significant (p=0.025).

With multivariable analysis, the odds of having corneal opacity increased with age and decreasing literacy status, and were higher for females and unemployed participants (table 2). After adjustment, females were at higher odds of having corneal opacity, though the difference was not statistically significant (p=0.53).

Table 2

Effect of sociodemographic factors on the prevalence of corneal opacity by logistic regression analysis

Prevalence of corneal blindness

The overall prevalence of corneal blindness in our study population, including both unilateral and bilateral cases, was 0.55% (95% CI 0.42% to 0.69%); the prevalence of bilateral cases was 0.12% (95% CI 0.05% to 0.17%) and that of unilateral corneal blindness was 0.45% (95% CI 0.33% to 0.56%). The majority of the corneal blind patients were females (61.5%) and were aged ≥60 years (84.6%).

Logistic regression analysis was conducted for individuals who had corneal blindness in any (54) or both eyes (13). In this subgroup, comprising 67 people with either unilateral or bilateral corneal blindness, the odds of having corneal blindness increased with age and were higher for females and unemployed participants (table 3). Similar results were obtained after adjustment for age, gender and other demographic parameters (table 3). The elderly had 4.7 times more risk of developing corneal blindness when compared to the younger age group after adjustment for other demographic factors.

Table 3

Effect of sociodemographic factors on the prevalence of corneal blindness by logistic regression analysis

Causes of corneal opacity

Common causes of corneal opacity included pterygium (34.5%), trauma (22.3%) and infectious keratitis (14.9%). In the 150 cases with pterygium, 89 (59.3%) had normal vision and 61 (40.7%) had mild to moderate visual impairment. Table 4 lists the various corneal diseases prevalent in the study population. Corneal involvement due to trachomatous infection and its sequelae was seen in 33 eyes of 21 participants (table 4).

Table 4

Causes and age distribution of corneal opacities and corneal blindness in the study population (n=435)

Causes of corneal blindness

Corneal diseases contributing to blindness were aphakic and pseudophakic bullous keratopathy following cataract surgery, corneal degenerations, and trauma. Unilateral corneal blindness was mostly caused by ocular trauma, infectious keratitis, and post-surgical bullous keratopathy (table 4).

Discussion

The CORE study is the first detailed, population-based study on corneal diseases in the Indian population. To the best of our knowledge, our study is the largest epidemiological study on corneal opacities worldwide. This study provides reliable epidemiological data on the prevalence and risk factors of corneal blindness and morbidity across all age groups in a rural North Indian population.

In the present study, the prevalence of corneal opacity was 3.7% (95% CI 3.4% to 4.1%). There are a few published studies on ocular morbidity conducted in the Indian population, which report the prevalence rate of corneal opacity as 4.2% in people aged ≥20 years and 3% in people aged ≥50 years.13 ,15 These studies involved a very small sample of the population and hence do not represent accurate estimations of the prevalence of corneal opacity. In a hospital-based study in Haryana, 11.6% of patients with a visual acuity of 6/9 or worse were found to have corneal disease.16 As these studies did not include all age groups, they failed to provide comprehensive and representative data on the burden of corneal diseases in the general population.

The prevalence of corneal blindness in this rural population was 0.12%, and 0.45% of people were unilaterally blind due to corneal disease. This amounts to a significant burden on the health system in general, and the blindness programme in particular. In another Indian study, the prevalence of corneal blindness (defined as presenting visual acuity <6/60 in the better eye in this study) was reported to be 0.10% and that of unilateral corneal blindness was 0.56%.17 A Chinese study also reported a similar prevalence of corneal blindness of 0.1%.18

The common causes of corneal opacity in this rural population were pterygium (34.5%) followed by trauma (22.3%) and infectious keratitis (14.9%). The aetiological factors responsible for corneal disease were reported to be different in the late 19th century, wherein trachoma was the most common corneal disease reported (30%) in a hospital-based study.16 The impact of community trachoma control programmes and nutritional education together with a measles immunisation programme has resulted in a reduction of the burden of corneal blindness due to trachoma and vitamin A deficiency.

The main aetiological diseases responsible for corneal blindness in our study population were post-surgical bullous keratopathy (46.2%), corneal degenerations (23.1%) and trachoma (15.4%). In the South Indian population, corneal blindness was predominantly due to infectious keratitis (59.5%) and trauma (23.2%).17 Studies from Thailand and Gambia also report the common causes of corneal blindness.19 ,20 In Thailand, these were infections (35.6%), surgical bullous keratopathy (27.8%), and trauma (14%).19 In Gambia, vitamin A deficiency (7.8%) was an important cause of corneal blindness in addition to infectious keratitis (22.7%) and trauma (14.3%).20 Thus, contrary to popular belief and previous studies,17 ,19 ,20 the predominant cause of corneal blindness in the current study was not infectious keratitis but corneal decompensation following cataract surgery. Addressing the issue of improving the quality of cataract surgical services in the country will help reduce the requirement for corneal transplantation. There is a need for implementing monitoring mechanisms so that good quality cataract surgery is available at all levels of healthcare delivery, thereby reducing the load of corneal complications after cataract surgery.

The study highlights the changing trends in the pattern of corneal diseases in the rural Indian population. The success of prevention programmes related to keratomalacia and trachoma have led to marked reductions in their prevalence, as exemplified in the present study. We need to be cautious about the increasing prevalence and occurrence of corneal complications due to high-volume cataract surgery, which is shifting the burden from cataract blindness to corneal blindness in a developing country like India.

It is important to interpret the results of our study keeping in view the few limitations. Rural data cannot be directly extrapolated to urban populations and further surveys in urban regions are required to achieve a comprehensive picture. Though the overall participation rate in the present study was high, the coverage of individuals aged 40–59 years and the male population was comparatively lower than other sub-groups. This could be attributed to the non-availability of males and people aged 40–59 years in their households during daytime, as they constitute the productive population and were out at work when the house-to-house visits were conducted. Although the present study involved a large sample population which was larger than any previously published study on corneal blindness, the sample size was not adequate for ascertaining the true prevalence of corneal blindness.17 ,18 However, the sample size was appropriate for assessing the primary objective of the study which was to determine the population prevalence of corneal opacity.

The major strengths of this study are the high response rate (93.9%) in a large sample population covering all age groups, and that it is the first reported population-based study, specifically designed to study the prevalence of corneal opacity in India, outlining all determinants and aetiological factors responsible for its occurrence. This study will prove useful in planning blindness programme initiatives and estimating resources required to provide comprehensive corneal services for the community, thereby helping to eliminate the causes of avoidable corneal blindness and visual impairment.

References

Footnotes

  • Contributors NG: Literature search and compilation, study conception and design, data collection and acquisition, data analysis and interpretation, drafting and writing of manuscript. PV: Study conception and design, critical revision of manuscript, administrative, technical and material support, supervision. RT: Study conception and design, data interpretation, revision of manuscript for important intellectual content, supervision. SKG: Study conception and design, data collection and acquisition, revision of manuscript for important intellectual content, technical support. SD: Statistical expertise, data analysis and interpretation, revision of manuscript for important intellectual content. KM: Study design, statistical expertise, data analysis and interpretation.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Institute Ethics Committee, All India Institute of Medical Sciences, New Delhi, India.

  • Provenance and peer review Not commissioned; externally peer reviewed.