Methodological quality of the review: Low confidence

Author: Suffee Bibi Shaminah, Skipkolye Mohammad Ashiff, Lyu Fan

Region: Taiwan, China, Singapore

Sector: Myopia

Subsector: Treatment: Atropine controlling myopia progression

Equity focus: School-age children (5 to 15 years)

Study population: Children (5 to 15 years)

Type of programme: School based

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: Myopia has become a global public health issue affecting a significant percentage of the population. It is a serious condition in both children and adults which may lead to blindness. High myopia may induce irreversible vision loss, such as nuclear cataract, open angle glaucoma, peripheral retinal changes, maculopathy, choroidal neovascularisation and also retinal detachment. Due to the prominent rise of myopia, effective treatments have been emphasised. Several methods are available to control the myopic progression and correct the refractive error, with different favourable outcomes. The options are classified either into conservative or pharmacological interventions. Initially, researchers suggested that the spasm of accommodation is the primary mechanism of atropine in myopic patients to reduce myopia progression. Some studies found that conservative agents, such as bifocals, eliminate accommodation. Nevertheless, they cannot slow the progress of myopia. Due to the rapid increase in the prevalence of myopia in recent years, many vision scientists are investigating ways of controlling the progression of myopia, and atropine has become a major focus of attention due to its aspect of ameliorating the progression of myopia.

Objectives: To examine the published evidence on the effectiveness of topical atropine for the prevention of childhood myopia progression and the myopic rebound occurring after cessation of atropine.

Main findings: Eleven randomised controlled trials met the selection criteria and were included in this meta-analysis. The articles included a total of 3,162 children, aged between 5 and 15 years. The initial baseline spherical equivalent varies from -50 D to -6.75 D. The children received atropine dosage ranging from 0.01% to 1%. The overall quality of the randomized controlled trials (RCTs) studies was moderate. Five studies were conducted in Singapore, three studies were conducted in Taiwan, and three studies were conducted in China.

The pooled data showed a significantly low myopic progression in all different dosage of atropine compared to the control groups. The MD of myopic progression were 0.95 D/year (95% CI: 0.69-1.22, P<0.001，I²:86%), 0.93 D/year (95% CI: 0.50-1.36，P<0.001，I²:98%) 0.82 D/year (95% CI: 0.68-0.96，P<0.001) and 0.46 D/year (95% CI: -0.02-0.94, P=0.06, I²：94%) for 1.0%，0.5%，0.1% and 0.01% atropine，respectively.

The meta-analysis indicated the occurrence of a myopic rebound after cessation of atropine treatment. After 12 months of atropine cessation，the analysis indicated that the MD between the different dosage of atropine，and the control group was -0.36 D/year (95% CI: -0.70- -0.02，P＝0.04，I²: 98%). Furthermore，after 36 months of atropine treatment cessation，meta-analysis showed that the MD between the different concentrations of atropine and the control group was -0.32 D/year (95% CI: -0.69-0.05，I²：97%) ，and the difference was not statistically significant (P＝0.09).

Methodology: Inclusion criteria: Randomised controlled trials (RCTs) were selected and included if they 1) were school-aged children (between 5 and 15 years old) with myopia，both sexes, all nationality and ethnicity; 2) used atropine in at least one treatment arm and nonatropine treatment or placebo in another as control; 3) reported an annual progression of myopia assessed as the mean change in refractive error (spherical equivalent) from baseline and measured using any methods. The data extracted included author, year of publication, location, population size, age, study duration，follow-up，intervention and control，mean change in spherical value and methodology information. The search was done on PubMed，Embase and the Cochrane Library to identify RCTs studies published from inception to 9 April, The electronic searches were only restricted only to English language. Data was extracted independently by two reviewers. Jadad scoring for RCTs was used independently by both reviewers to determine the quality of the included trials. Analysis was done using a mean difference in spherical value and 95% confidence interval (95% CI). The heterogeneity was statistically evaluated by I².

Applicability/external validity: The authors did not discuss how generalised the results are and whether it can be applied in other settings.

Geographic focus: The meta-analysis just focused on Asian countries.

Summary of quality assessment:

Low confidence was attributed to the conclusions about the effects of this study, as important limitations were identified. Authors did not conduct thorough searches of the literature to ensure that all relevant studies were included in the review; in addition, inclusion of studies was restricted to those written in English only. Furthermore, grey literature searching was not part of the search strategy. Authors did not specify if two authors independently screened studies for inclusion.