Myopic open-angle glaucoma prevalence in Northeast Asia: a systematic review and meta-analysis of population-based studies

Author: Jeong Y, Ha A, Shim SR, Kim YK.

Geographical coverage: Northeast Asia

Sector: Burden of disease

Sub-sector: Epidemiology

Equity focus: None

Study population: General population of Northeast Asia

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: Myopia is extremely widespread, especially in Asia, and glaucoma is the second leading cause of blindness globally, with open-angle glaucoma (OAG) being its most common manifestation. However, whereas the respective prevalence of myopia and glaucoma are well-known, little is confirmed as to their prevalence in combination.

Objectives: To investigate myopic open-angle glaucoma (OAG) prevalence in Northeast Asia by systematic review and meta-analysis.

Main findings:

Overall, this review found that in a myopic population, OAG prevalence was 4.10% (95% CI, 3.00-5.70; I² = 93%); in a general population, myopic OAG prevalence was 1.10% (95% CI, 0.60-1.70; I² = 94%).

The final analysis of this review included a total of five studies. Five population-based studies involving a total of 26,661 individuals had been conducted in four countries: two in China, one in Japan, one in South Korea and one in Singapore (Malay population). The number of participants varied from 2,528 to 13,831; the median sample size was 3,108. The mean age in each study was 55.8 ± 10.3 years in Beijing, China; 63.5 ± 8.8 years in Shanghai, China; 58.4 ± 11.8 years in Japan; 55.1 ± 0.2 years in South Korea; and 58.2 ± 10.9 years in Singapore Malay population). In three studies, myopia was defined according to an SE of -0.5 D [20,22,23], and one study reported myopia of less than -1.0 D SE. One other study provided no definition of myopia. In the South Korean study, the number of OAG patients was 710, the highest among the studies. In the Chinese, Japanese and Singaporean (Malay population) studies, the OAG patient numbers ranged between 72 and 119. The number of myopic patients, like the number of OAG patients, was highest in South Korea (4,039), followed by Beijing, China (1,837). The myopic patient numbers in the three remaining studies ranged between 239 and 877.

The meta-analysis indicated that, in a myopic population, OAG prevalence was 4.10% (95% CI, 3.00-5.70; I² = 93%); in a general population, myopic OAG prevalence was 1.10% (95% CI, 0.60-1.70; I² = 94%). The percentage of OAG among patients with low myopia was 2.24% to 4.37%, while that with high myopia was 6.07% to 8.68%. The estimated OAG prevalence among myopia patients ranged from 2.7% to 6.3%. In South Korea, the prevalence was 6.3%, the highest among the studies. The prevalence was the lowest in China (2.7%).

A visual examination of funnel plot symmetry raised a suspicion of publication bias. Notwithstanding, Begg and Mazumdar’s adjusted rank correlation test showed no such evidence (p = 0.6242).

Authors noted that further research involving more qualified studies from multiple countries, ethnicities and continents should be conducted for more generalised conclusions regarding any ethnic predisposition to myopic OAG.

Methodology:

The study inclusion criteria were as follows: (1) population-based; and (2) evaluation of both OAG and myopia prevalence. The exclusion criteria were as follows: (1) not conducted with humans or adults; (2) narrative and/or systematic review, commentary, case report; (3) involving either secondary glaucoma or angle-closure glaucoma; (4) not conducted in Northeast Asia; (5) published in a language other than English; and (6) lacking any detailed definition of OAG.

Study authors systematically searched for relevant studies in the PubMed, EMBASE and Cochrane Library databases. Two investigators conducted the search in an independent and masked manner, and any inconsistencies were resolved by discussion and consensus, or if needed, by third-party adjudication. The retrieved articles’ reference lists were also manually reviewed and additional relevant studies thereby identified. The databases were searched for any and all pertinent reports published through 30 November  2022.

Two investigators extracted data in both an independent and masked fashion using a standardised data extraction method based on those employed by the Cochrane Library’s Database of Systematic Reviews. To assess the methodological quality of the studies, the Newcastle-Ottawa Scale (NOS) for assessment of comparative nonrandomised study quality was applied. Studies were additionally evaluated to determine the risks of bias related to selection, comparability, exposure/outcome, or any other factor.

Because of anticipated high levels of heterogeneity, a random-effect model was used by the authors, to estimate the pooled OAG prevalence in a myopic population and the pooled myopic OAG prevalence in a general population. Inter-study heterogeneity was quantified by the I² statistic representing the interstudy variation prevalence that can be attributed to heterogeneity rather than sampling error. To qualitatively synthesise myopia degree, the prevalence of myopic OAG, and percentage of OAG among myopia, myopia was stratified into low, moderate, moderate-to-high, and high degree categories, as based on spherical equivalent (SE) refractive error up to -3 dioptres (D), between -6 D and -3 D (or -4 D), lower than -3 D (or -4 D), and lower than -6 D, respectively. Publication bias was evaluated in two ways: (1) qualitatively by funnel plot, and quantitatively by Begg and Mazumdar’s adjusted rank correlation test. All of the 95% confidence intervals (CIs) and p-values were two-sided; p <0.05 was considered to represent statistical significance. All of the statistical analyses were performed with R.

Applicability/external validity: Review only includes data from four countries, and authors note that some of population studies were conducted in urban areas only (where prevalence of myopia is known to be higher). This potentially limits the applicability of this review.

Geographic focus: Authors indicate how prevalence of co-occurrence of myopia and OAG varies in different countries in Northeast Asia – and suggest that further work is needed to determine how far this may be due to ethnicity.

Summary of quality assessment:

The approaches used to identify, include and critically appraise studies were generally robust, with two authors undertaking all key tasks. However, the search was limited to material published in English, which may mean that some relevant studies were excluded. While the reported meta-analysis was robust, the authors did not consider in the analysis how the quality and risk of included studies might have affected the reported results (although we appreciate that this would have been challenging, given that only five studies were ultimately included). For these reasons, and the high degree of heterogeneity among studies noted by authors, we would have “low confidence” in the findings of this review.