Global Trends in Blindness and Vision Impairment Resulting from Corneal Opacity 1984–2020

TopicWe provide global estimates of the prevalence of corneal blindness and vision impairment in adults 40 years of age and older and examine the burden by age, sex, and geographic region from 1984 through 2020.Clinical RelevanceCorneal opacities (COs) are among the top 5 causes of blindness worldwide, yet the global prevalence, regional differences, and risk factors are unclear.MethodsAbstracted data from the published literature and surveys were obtained from the Global Burden of Disease Vision Loss Expert Group. We supplemented this by an independent systematic literature search of several databases. Studies that provided CO vision impairment data based on population-based surveys for those 40 years of age or older were included, for a total of 244. For each of the 4 outcomes of blindness and moderate to severe vision impairment (MSVI) caused by trachomatous and nontrachomatous CO (NTCO), time trends and differences in prevalence by region, age, and sex were evaluated using a Poisson log-linear model with a generalized estimating equation method. Age-standardized estimates of global prevalence of blindness and MSVI were calculated using the 2015 United Nations standard populations.ResultsThe global prevalence of blindness resulting from NTCO in those 40 years and older was 0.081% (95% confidence interval [CI], 0.049%–0.315%); that of MSVI was 0.130% (95% CI, 0.087%–0.372%). A significant increase with age was found (prevalence rate ratio, 2.15; 95% CI, 1.99–2.32). Latin America and Europe showed the lowest rates, with 2- to 8-fold higher rates of blindness or MSVI in other regions. The global prevalence of blindness resulting from trachomatous CO in those 50 years and older was 0.0094% (95% CI, 0%-0.0693%); that from MSVI was 0.012% (95% CI, 0%–0.0761%). Blindness resulting from trachomatous CO and MSVI increased with age and female sex, and rates were significantly higher in the African regions. A decrease in trachomatous blindness rates over time was found (prevalence rate ratio, 0.91; 95% CI, 0.86–0.96).DiscussionAn estimated 5.5 million people worldwide are bilaterally blind or have MSVI resulting from CO, with an additional 6.2 million unilaterally blind. Blindness resulting from trachomatous CO is declining over time, likely because of the massive scaleup of the global trachoma elimination program and overall socioeconomic development.Financial Disclosure(s)The author(s) have no proprietary or commercial interest in any materials discussed in this article. We provide global estimates of the prevalence of corneal blindness and vision impairment in adults 40 years of age and older and examine the burden by age, sex, and geographic region from 1984 through 2020. Corneal opacities (COs) are among the top 5 causes of blindness worldwide, yet the global prevalence, regional differences, and risk factors are unclear. Abstracted data from the published literature and surveys were obtained from the Global Burden of Disease Vision Loss Expert Group. We supplemented this by an independent systematic literature search of several databases. Studies that provided CO vision impairment data based on population-based surveys for those 40 years of age or older were included, for a total of 244. For each of the 4 outcomes of blindness and moderate to severe vision impairment (MSVI) caused by trachomatous and nontrachomatous CO (NTCO), time trends and differences in prevalence by region, age, and sex were evaluated using a Poisson log-linear model with a generalized estimating equation method. Age-standardized estimates of global prevalence of blindness and MSVI were calculated using the 2015 United Nations standard populations. The global prevalence of blindness resulting from NTCO in those 40 years and older was 0.081% (95% confidence interval [CI], 0.049%–0.315%); that of MSVI was 0.130% (95% CI, 0.087%–0.372%). A significant increase with age was found (prevalence rate ratio, 2.15; 95% CI, 1.99–2.32). Latin America and Europe showed the lowest rates, with 2- to 8-fold higher rates of blindness or MSVI in other regions. The global prevalence of blindness resulting from trachomatous CO in those 50 years and older was 0.0094% (95% CI, 0%-0.0693%); that from MSVI was 0.012% (95% CI, 0%–0.0761%). Blindness resulting from trachomatous CO and MSVI increased with age and female sex, and rates were significantly higher in the African regions. A decrease in trachomatous blindness rates over time was found (prevalence rate ratio, 0.91; 95% CI, 0.86–0.96). An estimated 5.5 million people worldwide are bilaterally blind or have MSVI resulting from CO, with an additional 6.2 million unilaterally blind. Blindness resulting from trachomatous CO is declining over time, likely because of the massive scaleup of the global trachoma elimination program and overall socioeconomic development. The author(s) have no proprietary or commercial interest in any materials discussed in this article. Corneal opacity (CO) is estimated to be the cause of 3.2% of all cases of blindness and 1.3% of all cases of moderate to severe vision impairment (MSVI)1Flaxman S.R. Bourne R.R.A. Resnikoff S. et al.Global causes of blindness and distance vision impairment 1990–2020: a systematic review and meta-analysis.Lancet Glob Health. 2017; 5: e1221-e1234Abstract Full Text Full Text PDF PubMed Scopus (1777) Google Scholar and is among the top 5 causes of blindness worldwide.2World Health OrganizationWorld health statistics.https://www.who.int/gho/publications/world_health_statistics/2015/en/Google Scholar The lifetime burden of corneal blindness is significant because it tends to affect younger people compared with other conditions such as cataract and glaucoma.3Ung L. Acharya N.R. Agarwal T. et al.Infectious corneal ulceration: a proposal for neglected tropical disease status.Bull World Health Organ. 2019; 97: 854-856Crossref PubMed Scopus (39) Google Scholar Major causes of corneal opacification include trachoma, infectious keratitis, xerophthalmia, use of traditional eye medicines, and ocular trauma.4Jeng B.H. Ahmad S. In pursuit of the elimination of corneal blindness: is establishing eye banks and training surgeons enough?.Ophthalmology. 2021; 128 ([published correction appears in Ophthalmology. 2021;128(8):1245]): 813-815Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar,5Whitcher J.P. Srinivasan M. Upadhyay M.P. Corneal blindness: a global perspective.Bull World Health Organ. 2001; 79: 214-221PubMed Google Scholar Overall, in low- and middle-income countries, infectious keratitis is reported to be the most common problem.3Ung L. Acharya N.R. Agarwal T. et al.Infectious corneal ulceration: a proposal for neglected tropical disease status.Bull World Health Organ. 2019; 97: 854-856Crossref PubMed Scopus (39) Google Scholar However, other conditions, such as trachoma or onchocerciasis, may dominate in other areas. Global trends in vision loss resulting from CO remain largely unknown, but regions that have the least capacity to manage CO are hypothesized to bear the greatest burden. Poor rural communities have limited access to treatment and higher prevalence of communicable diseases such as trachoma.3Ung L. Acharya N.R. Agarwal T. et al.Infectious corneal ulceration: a proposal for neglected tropical disease status.Bull World Health Organ. 2019; 97: 854-856Crossref PubMed Scopus (39) Google Scholar The epidemiologic features of vision loss resulting from CO have important implications for the global distribution of eye care provision, especially preventive public health interventions, treatments for infectious keratitis, and corneal transplantation. The purpose of this study was to use global data from published literature and surveys to provide global estimates of the prevalence of corneal blindness and vision impairment and to examine the burden by age, sex, and geographic region from 1984 through 2020. We included CO vision impairment data provided by the Global Burden of Disease Vision Loss Expert Group from 1984 through 2020. Their search strategy and data extraction methods have been well described.6Global Burden of Disease 2019 Blindness and Vision Impairment CollaboratorsVision Loss Expert Group of the Global Burden of Disease StudyTrends in prevalence of blindness and distance and near vision impairment over 30 years: an analysis for the Global Burden of Disease Study.Lancet Glob Health. 2021; 9: e130-e143Abstract Full Text Full Text PDF PubMed Scopus (379) Google Scholar The data included the numbers of cases of blindness and MSVI resulting from NTCO and, if included, from trachomatous CO, and total numbers surveyed by age group and sex, the latter if available. The Vision Loss Expert Group provided 276 studies, of which we used 239, including Rapid Assessment of Avoidable Blindness (RAAB) surveys.7Mactaggart I. Limburg H. Bastawrous A. et al.Rapid assessment of avoidable blindness: looking back, looking forward.Br J Ophthalmol. 2019; 103: 1549-1552Crossref PubMed Scopus (28) Google Scholar The Institutional Review Boards approved the study and waived the requirement for informed consent because of the retrospective nature of the study. All research adhered to the tenets of the Declaration of Helsinki. Following their data extraction methods, we performed a parallel literature search to identify additional published studies containing data on vision loss resulting from CO that were published between 1990 and 2020 by searching the following online databases: PubMed, Embase, Web of Science, Scopus, and Cochrane (Appendix A, available at www.aaojournal.org). Studies that provided CO vision impairment data (blindness, MSVI, or both) based on population-based surveys for those 40 years of age or older were included. Studies that were based on school surveys, case reports, or key informants were not eligible. The complete inclusion and exclusion criteria and data sources are given in Appendix B (available at www.aaojournal.org). We included 5 additional studies based on these criteria (Appendix C, available at www.aaojournal.org).8He Y. Nie A. Pei J. et al.Prevalence and causes of visual impairment in population more than 50 years old: the Shaanxi Eye Study.Medicine (Baltimore). 2020; 99e20109Crossref Scopus (15) Google Scholar, 9Guo C. Wang Z. He P. et al.Prevalence, causes and social factors of visual impairment among Chinese adults: based on a national survey.Int J Environ Res Public Health. 2017; 14: 1034Crossref PubMed Scopus (26) Google Scholar, 10Tafida A. Kyari F. Abdull M.M. et al.Poverty and blindness in Nigeria: results from the national survey of blindness and visual impairment.Ophthalmic Epidemiol. 2015; 22: 333-341Crossref PubMed Scopus (21) Google Scholar, 11AlSawahli H. McCormick I. Mpyet C.D. et al.Population-based rapid assessment of avoidable blindness survey in Sohag governorate in Egypt.BMJ Open. 2020; 10e036337Crossref PubMed Scopus (3) Google Scholar, 12Melese M. Alemayehu W. Bayu S. et al.Low vision and blindness in adults in Gurage Zone, central Ethiopia.Br J Ophthalmol. 2003; 87: 677-680Crossref PubMed Scopus (49) Google Scholar Moderate to severe vision impairment was defined as presenting visual acuity of less than 6/18 but 3/60 or more in the better eye (presenting refers to "as the patient presents, which can be with or without corrective eyewear"13Global Burden of Disease 2019 Blindness and Vision Impairment CollaboratorsVision Loss Expert Group of the Global Burden of Disease StudyCauses of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global Burden of Disease Study.Lancet Glob Health. 2021; 9 ([published correction appears in Lancet Glob Health. 2021;9(4):e408]): e144-e160Abstract Full Text Full Text PDF PubMed Scopus (836) Google Scholar). Blindness was defined as presenting visual acuity of less than 3/60 in the better eye. For both outcomes, best-corrected visual acuity was used if presenting visual acuity was not reported. Unilateral blindness or MSVI was defined as presenting acuity in the worse eye of less than 6/18 but 3/60 or more, where the better-eye acuity was not blindness or MSVI. Nontrachomatous CO (NTCO) was determined in each study by examination of the cornea by trained personnel and observation of an opacity deemed not resulting from trachoma and sufficiently dense and extensive to produce vision loss as severe as MSVI or blindness. If scarring and trichiasis or trichiasis surgery and other signs of trachoma such as Herbert's pits were present and the survey was in a known endemic trachoma region, the CO was attributable to trachoma. Data on blindness and MSVI resulting from CO were sparse, and we could not use regions as constructed previously.1Flaxman S.R. Bourne R.R.A. Resnikoff S. et al.Global causes of blindness and distance vision impairment 1990–2020: a systematic review and meta-analysis.Lancet Glob Health. 2017; 5: e1221-e1234Abstract Full Text Full Text PDF PubMed Scopus (1777) Google Scholar We required at least 3 surveys per region, which resulted in 7 regions: Europe and Central Asia, East Asia, Latin America and the Caribbean, North Africa and the Middle East, South Asia, Southeast Asia and Oceania, and Sub-Saharan Africa. No data on CO from North America were available. Figure 1 shows the makeup of each region. Forty-five studies only provided data on the total number of individuals with blindness or MSVI resulting from CO (no age or sex data) but did provide an age breakdown of the total study population. For each of these studies, we calculated age group-specific prevalence risk ratios derived from all other data in the region of the study using the youngest age group as the reference. Then, using the risk ratios, the observed total number of patients in the study, and the population studied in each age group, we were able to impute the fraction of number of patients we expected for each age group such that the sum would equal the total observed in the study. Duplicate data sources were cleaned up. These were largely RAAB surveys that were sent as raw data but that also were used in an article in the published literature. If the country or region, year of the survey, and population matched, the published article was excluded, and the RAAB data were used. For each of the 4 outcomes of blindness and MSVI caused by NTCO and blindness and MSVI caused by trachomatous CO, we assessed whether a significant time trend existed in the prevalence using a univariable Poisson log-linear model. The analysis unit was each survey study. The total number of people surveyed was modeled as an offset in the Poisson model, and year of the survey was the primary exposure variable. Because the same country may have multiple surveys, a generalized estimating equation (GEE) method was applied with the Poisson regression model to account for potential correlation between surveys conducted in the same country. Univariable Poisson log-linear models with GEE also were used to assess whether prevalence of an outcome varied by risk variables including region, age group, and sex (where data on sex were available). The unit of analysis in the models for age group and sex was age group- and sex-specific survey results. A multivariable Poisson model with GEE was used to estimate adjusted associations with an outcome including risk variables that had P values of less than 0.05 from their univariable regression models with the outcome. The 95% confidence intervals (CIs) were calculated from the Poisson log-linear models with GEE. Too few trachoma surveys were from the individual regions to examine the effect of each region independently, so a comparison was made of the high-risk regions (North Africa and the Middle East and Sub-Saharan Africa) versus all others. All analyses were implemented in SAS software (SAS Institute). A sensitivity analysis to evaluate whether there was a time trend for each outcome was conducted by restricting the analysis to countries that had data from at least 2 surveys conducted in different years. Age–standardized estimates of prevalence of blindness and MSVI resulting from CO were calculated using the United Nations standard population by region in 2015.14Population Division, Department of Economic and Social Affairs, United Nations. World population prospects 2019. Custom data acquired via website: https://population.un.org/wpp/Download/Standard/MostUsed/ Accessed June 1, 2022.Google Scholar Because most of the studies were from 2000 to 2020, we chose 2015 as the midpoint year. We applied the region- and age-specific rates to the standard population of the region to derive the expected numbers of patients in that region, which then were summed and divided by the estimate of the global population (minus the population of North America because no data were available for North America). We used the same approach for trachomatous CO but with the inclusion of the population of North America and Europe, presuming no individuals with CO resulting from trachoma were from these two regions. The 95% CIs around the estimated prevalences were calculated as follows: using the age- and region-specific lower bound of the 95% CI for the prevalence rate, we calculated the expected minimum number of cases globally; these cases divided by the global population as above became the lower 95% CI. Similarly, using the age- and region-specific upper bound of the 95% CI, we derived the expected maximum number of cases and thus the upper bound of the 95% CI. By using the age-specific rates within the region applied to the regional population distribution, we account for potential differences in the sampled population age structure of the studies within the region. The National Eye Institute and the benefactor of the El Maghraby Chair had no role in the study design, data collection, data analyses, data interpretation, or the writing of this report. The corresponding author had full access to the data in the study and final responsibility for the decision to submit the manuscript for publication.