Introduction

Cataract is defined as any opacity of the crystalline lens of the eye, which impedes the image clarity causing reduced visual acuity and impaired contrast sensitivity. Cataract in children may be congenital or acquired, unilateral or bilateral1 and in the majority of cases is treatable. Although it is rare, childhood cataract is one of the most important causes of blindness and severe visual impairment in children and is responsible for 5–20% of pediatric blindness worldwide.2 It is estimated that 200 000 children worldwide are blind due to cataract, and that 20 000–40 000 children are born each year with congenital cataract.3 Cataract blindness in children presents an enormous problem to developing countries in terms of human morbidity, economic loss, and social burden.4

Studies conducted in schools for the blind have investigated the various causes of childhood blindness. Previous reports from West Africa, South India, and Chile showed that lens abnormalities accounted for 15.5%, 7.4%, and 9.2% of blindness in such schools.5 Similar studies conducted in Malawi, Kenya and Uganda found that blindness was caused by unoperated cataract in 13.1%, 9.1%, and 27.6% of children, respectively.6 In Ethiopia, unoperated cataract or aphakia accounted for 9.2% of blindness in schools for the blind.7 With significant reductions in some of the preventable causes of blindness such as measles and vitamin A deficiency, cataract has become the major cause of treatable blindness in children in developing countries.8

Reliable region-specific data on the prevalence and incidence of childhood cataract is important as a basis for policy decisions, including the evidence-based allocation of resources. Cost and logistics limit the feasibility of the large-scale data gathering required for prevalence estimates. The key informant method, in which key community members are trained to identify people within the community with a given health condition, was introduced to calculate prevalence based on a ratio of cases identified and an estimate of the total number at risk in a particular geographical area. However, few studies have used this method to date,9, 10 and there is a paucity of epidemiological information about cataract in children globally. Thus, there is a lack of evidence to guide policy related to childhood cataract. Currently, there are no systematic reviews on the question of prevalence and incidence of childhood cataract. The aim of this study is to systematically review existing research to determine a reliable estimate of global prevalence and incidence of congenital (from birth) and acquired (due to trauma or disease) cataract in children.

Methods

Search strategy and selection criteria

We followed the Preferred Reporting of Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The Cochrane Library, Medline, and Embase were searched (the date of last search was January 2015 via OVID and EBSCOHOST) using the following search terms formatted for OVID search: (‘Child*’ [All Fields] OR ‘infan*’ [Title] OR ‘p?ediatric*’ [Title] OR ’ adolescen*’ [Title] OR ‘teenage*’ [Title] OR ‘juvenile*’ [Title] OR ‘minor’ [Title] OR ‘young people’) AND (‘Cataract’ [Abstract] OR ‘lens*’ [Abstract] OR ‘near opacity*’ [Abstract]) AND (‘prevalence’ [Abstract] OR ‘incidence’ [Abstract] OR ‘epidemiology’ [Abstract]. No restrictions were imposed based on language or year of publication. Bibliographies of related articles were checked to identify additional potentially relevant reports. The World Health Organisation website was searched for program reports and government documentation. The protocol for this review has been registered and published on the Prospero database (http://www.crd.york.ac.uk/PROSPERO/prospero.asp; reference number CRD42014014909).

Inclusion and exclusion criteria

We included all studies at any global location which estimated the prevalence and/or incidence of cataract among children (aged less than 18 years). In this context, prevalence indicates the number of children in a population that have cataract at a given point of time divided by those at risk (the total number of children in the population sample). Incidence indicates how many new cases of cataract occur in children under 18 years within a defined period of time. For estimating prevalence, data from non-random samples (eg, from schools for the blind) or based on self-report were excluded. For incidence studies, no exclusion criteria were imposed.

Quality assessment and data extraction

One reviewer (SS) conducted the search and all of the studies derived from the search were independently assessed by two reviewers (SS and CMS) for inclusion initially based on title and abstract content followed by full-text review of potentially eligible studies, using the criteria outlined above. Any discrepancies were discussed and resolved by consensus. After this process the included studies were assessed for methodological quality based on the full-published paper independently by both SS and CMS using the prevalence critical appraisal instrument developed by Munn et al.11 Criteria used to judge quality are provided in Figure 1. Data were independently extracted from eligible studies by two reviewers (SS and CMS), and the resulting data were verified by a third reviewer (JGL). All the quantitative data synthesis was carried out using Open Meta Analyst.12

Figure 1
figure 1

Quality assessment of the 24 included studies.

Full size image

Statistical analysis

We intended to calculate a pooled estimate of the global prevalence of childhood cataract (congenital and acquired) and the prevalence of congenital cataract only. In addition, we obtained an estimate of the prevalence of childhood cataract according to the country’s economic status across included studies. Heterogeneity was assessed using the Cochran's Q chi-squared statistic and by calculating the I2.13, 14 Prevalence was assessed for geographical location according to income status, defined according to the gross national income per capita per year and calculated using the World Bank Atlas method (http://data.worldbank.org/about/country-and-lending-groups). Correlation tests were used to correlate variables with P<0.05 considered as statistically significant.

Results

Out of a total of 677 potentially relevant titles/abstracts, 44 full-text articles were identified from searches of bibliographic databases, with 24 of these meeting the inclusion criteria. The PRISMA flow diagram is shown in Figure 2. The majority of the studies reviewed were in English (n=41), two in Mandarin and one in Portuguese. Reasons for exclusion of the other 20 studies are reported in Appendix 1 (available as Supplementary Information).

Figure 2
figure 2

Summary of review strategy—PRISMA Flow Diagram.

Full size image

Half of the included studies (n=13) were published between 2004 and 2014 and all of the included studies were published between 1988 and 2014. Twenty studies reported data on prevalence15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 and four studies reported incidence.35, 36, 37, 38

The included studies represented five geographical regions including Europe and Central Asia (n=8), South Asia (n=3), East Asia and Pacific (n=8), Sub-Saharan Africa (n=3), and North America (n=2).

Sample sizes in the included studies varied greatly, ranging from small samples in regional cross-sectional studies to analyses of large data sets derived from national registries. The methods used for case definition also varied between studies: from lens opacities detected following an ocular examination to cataract causing varying degrees of unilateral or bilateral visual impairment. The characteristics of included studies are presented in Table 1 and the results of the quality assessment summarised in Figure 1. Studies were generally of moderate to good methodological quality, although they were often poorly reported.

Table 1 Characteristics of the included study
Full size table

A considerable degree of heterogeneity was found between the 20 studies reporting prevalence of childhood cataract (Cochran’s Q-test, P<0.01; I2 =94%; see Figure 3). Given the heterogeneity in prevalence estimates and differences in study design and methods of case ascertainment, we did not perform a meta-analysis. The overall prevalence of childhood cataract ranged from 0.32 to 22.9 per 10 000 (median 1.03/10 000) and 0.63 to 9.74 per 10 000 (median 1.71/10 000) for congenital cataract based on 13 studies that reported congenital cataract.

Figure 3
figure 3

Forest Plot on prevalence of Childhood cataract in low- and lower middle-income countries compared with high- and higher middle-income countries (proportions with 95% confidence interval). For each study, the size of the symbol corresponds to the sample size.

Full size image

The prevalence in low-income and lower middle-income economies ranged from 0.42 to 2.05 per 10 000 and 0.32 to 8.49 per 10 000, respectively; in upper middle-income economies, it was from 0.74 to 22.7 per 10 000; and in high-income economies, it was from 0.63 to 13.6 per 10 000.

Prevalence by laterality (unilateral or bilateral) was reported in four studies22, 30, 32, 34 and three studies reported data on traumatic cataract.25, 28, 30 Overall, the reported prevalence of unilateral and bilateral cataract was similar (P=0.21). Prevalence was reported according to gender in five studies22, 24, 30, 31, 34 and there was no difference in prevalence of childhood cataract by gender (P=0.48).

Incidence was reported in four studies35, 36, 37, 38 and ranged from 1.8 to 3.6 per 10 000 per annum. Laterality was reported in two of these studies35, 37 and gender breakdown was reported in three studies.35, 36, 37 The incidence of cataract by laterality (P=0.35) and gender (P=0.76) was similar.

Discussion

To our knowledge, this is the first systematic review of prevalence and incidence studies of childhood cataract. The review included 20 prevalence studies and four incidence studies from five different geographical regions that were published between 1988 and 2014. The median prevalence of childhood cataract was 1.03 per 10 000 (range 0.32–22.9/10 000) children. Over 90% of cataracts were classified as congenital or developmental.

It is not clear whether the wide range in reported prevalence values reflects true variances between populations or whether this is due to differences in methodology and/or case definitions used in the included studies. For example, birth cohort studies would have missed developmental cataracts; studies using visual acuity of the better-seeing eye to identify those requiring further evaluation would have missed unilateral cataract and those who have successfully undergone cataract surgery. Moreover, studies classifying cataract as any lens opacity would have a higher prevalence than those using a definition of visual impairment or blindness because of cataract. Reliability of diagnosis is of fundamental importance in a prevalence study. In the studies we have reviewed, a detailed description of the diagnostic method was often lacking. For example, some studies indicated that slit lamp biomicroscopy was used, but did not explain on what basis (eg, grading scheme) cataract was diagnosed. It has been suggested that both subjective and objective evaluations of infantile cataracts are important to predict its effects on visual performance.39

Various methods have been developed and validated for the assessment of vision in infants and young children.40 In most of the studies included here it was unclear whether the methods used would provide a reliable assessment of vision, and in general basic methods such as infants’ detection of small objects, or perception of light were used. These methods cannot provide an accurate indication of acuity, and simple, affordable methods such as preferential looking cards would provide a better means of gauging the severity of vision loss in prevalence studies on childhood cataract.

It has been previously reported that the prevalence of blindness due to childhood cataract is 10 times higher in low-income economies compared with high-income economies.3 This is primarily due to inadequate health-care systems, malnutrition and higher rates of perinatal infections, for example, rubella. The present findings do not agree with this, and suggest higher prevalence estimates in high-income than lower income economies. This may reflect the fact that the majority of included studies in high-income countries did not use visual acuity as part of the case definition of cataract. Studies using visual acuity to define cases were mostly focused on children with blindness or visual impairment, and would identify cases with severe vision loss, missing those with unilateral or moderate vision loss. Such studies may therefore underestimate cataract prevalence. In addition, the relatively low estimate in low-income economies may be due in part to the association between conditions causing blindness and high under 5 mortality rates in these regions. For example, the survival rate of children with blinding conditions such as vitamin A deficiency is lower in countries with high under-five mortality rates.41 As outlined above, our prevalence estimates do not show higher prevalence in low income economies and these findings suggest that more studies with adequate, representative samples are needed with a common case definition to more accurately estimate the prevalence of childhood cataract. This is particularly challenging in low-income countries due to costs and the logistics involved, compared with high-income economies where national registries and surveillance systems facilitate epidemiological data collection.26, 29, 36

It is worth noting that heterogeneity of reported prevalence varies considerably within as well as between regions. If we take China (an upper middle-income economy) as one example, prevalence studies included in this review were carried out in Beijing (prevalence 1.7/10 000),25 South-Eastern China (0.7/10 000),34 South-Western China (5.6/10 000),23 across all states (1.5/10 000)21 and in Western China (22.7/10 000).28 The authors of the latter study commented that Western China is relatively undeveloped compared with other regions in the country, and this may illustrate the existence of a range of health-care provision and prevalence within one country.

Incidence studies included in this review were conducted in Sweden,35 Denmark,36 the UK,37 and Australia.38 These are all high-income economies; we found no incidence studies based in low- to middle-income economies.

Another important finding from this review is that both bilateral and unilateral cataract have similar prevalence, so about half of the cases are bilateral and about half are unilateral. Both have significant impact on vision in different ways. Unoperated bilateral cataract has the obvious impact of reducing vision in both eyes, thus causing severe visual impairment and blindness. Unilateral cataract, on the other hand, has seemingly less impact, since it affects vision in only one eye, leaving the fellow eye able to provide unimpeded vision. However, it is important to note that bilateral visual deprivation during early childhood has a less severe impact on visual system development than unilateral deprivation.42 In particular, amblyopia is a condition in which vision is abnormal (eg, reduced acuity in one eye and poor binocular depth perception) as a result of abnormal visual input during early life. Treatment to correct visual abnormality is more successful in early childhood,43 during a period of visual system plasticity, than later, so early diagnosis and management is important for any childhood condition in which vision is impeded.44, 45 Thus, early treatment in both cases is important, to remove the cataract as an impediment to vision and provide refractive correction. Consistent with this, the appropriate provision of surgery for congenital cataracts is one of the specific disease control objectives in the Vision 2020 program to control blindness in children.2, 46

To conclude, this review highlights substantial gaps in the epidemiological knowledge of childhood cataract worldwide, particularly from low- and lower middle-income economies, where the burden of childhood cataract is presumed to be high. Using the median prevalence of 1.03/10 000 children and an estimated 26% of the global population aged <15 years (http://www.worldometers.info/world-population/world-population-gender-age.php) (1.86 billion children in this age group), this would translate to approximately 191 000 cases of childhood cataract worldwide. Similarly, using the median incidence of 1.69 per 10 000, which translates to around 314 000 new childhood cataract (both congenital and developmental) cases every year. Future studies should report age, gender and ethnicity-specific estimates of incidence and prevalence, and attempt to standardise epidemiological methods and case definitions (particularly incorporating visual impairment). These estimates could then inform policy decisions to prioritise funding of programs to reduce visual impairment and blindness due to childhood cataract at regional and global levels. Delivering timely surgical intervention6 and appropriate follow-up after surgery would avoid blindness in children due to cataract, as emphasised and advocated by the Vision 2020 initiative: The Right to Sight Initiative.47