Changes in axial length after orthokeratology lens treatment for myopia: a meta-analysis

Methodological quality of the review: Medium confidence

Author: Meng Guan, Weijia Zhao, Yu Geng, Yang Zhang, Jia Ma, Zonghan Chen Mingqian Peng, Yan Li

Region: China, East Asia, Japan, Spain, Korea

Sector: Myopia

Subsector: Myopia treatment

Equity focus: No

Study population: Younger than 18 years old

Type of programme: Hospital based

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: Myopia, known as shortsightedness or nearsightedness, is one of the most common eye disorders. It begins in early life and its frequency and severity evolve through childhood and adolescence into adulthood. As reported, the prevalence of myopia has increased worldwide. Myopia is responsible for increased health care costs and a number of degenerative eye diseases, including macular degeneration, cataract and glaucoma. The correction and control of children’s myopia has become an urgent issue. In recent years, orthokeratology (OK) lens has become more and more prevalent in myopia treatment. However, few studies have focused on the accommodative response, including the changes of axial length after receiving OK lens treatment, which is closely related to the refractive status of cornea and visual quality.

Objectives: To assess the effect of OK lens on axial length change compared with glasses.

Main findings:

Authors included 13 studies in the meta-analysis. After checking the three domains (selection of participants, comparability of study groups, and the ascertainment of exposure or outcome), all 13 studies got seven or greater scores. Two of the 13 studies were based on Spain and 11 studies concentrated on samples from East Asia, including Japan, Korea and China. All myopia patients from these studies were recruited at an early age and followed up at least one year later. The degree of myopia ranged from -0.5D to -10.0D among recruited children, and the degree of astigmatism was all under 2.00D. In terms of the axial length, all recruited studies chose IOL Master as the main measurement to ensure accuracy. A total of 408 children received OKs treatment and 396 children as control group across seven studies were analysed. According to the heterogeneity results (I²>50%, P<0.05), the random effect model was chosen to get the merged results. SMD (95% CI) of change in axial length at the end of one year was -0.857 (-1.146, -0.568), P<0.001; the variation in axial length in OK’s group was significantly smaller than the control group after one year’s treatment, indicating that OK’s treatment appears effective in controlling the axial extension in the first year. According to Begg’s test (P=0.592) and Egger’s test (P=0.618), together with the symmetrical plots in funnel plot, no publication bias was found in this meta-analysis. A total of 303 children received OKs treatment, and 308 children as a control group of seven studies were analysed. Heterogeneity analysis was carried out first. Due to I² value (50.4%) and the P value of Q test (0.041), the random effect model was chosen. At the end of two years, the merged SMD (95% CI) was -0.701 (-1.675, 0.272), P<0.001, which was consistent with the merged result of change in axial length after one year. The axial length change in OKs group was also significantly smaller than control group. Publication bias was carried out by both Begg’s test (P=0.466) and Egger’s test (P=0.010). Symmetrical funnel plot showed that little publication bias was found in this meta-analysis. Also, the sensitivity analysis indicated that this merged SMD was very stable.

Authors conclude that OK lens treatment appears more effective in slowing axial elongation than glasses during the early treatment of myopia in children.

Methodology:

Inclusion criteria consisted of: 1) study samples with patients with acquired myopia without organic lesion and younger than 18 years old; 2) studies which focused on the effect on axial length by orthokeratology lens treatment or ordinary glasses treatment; 3) information, including axial length measurements, myopic refraction, astigmatism degree and raw data was available; 4) intraocular lens master optical biometer (IOL Master) was used for axial length measurement; 5) English written articles.

Myopia, orthokeratology (OK) lens and axial length were used as keywords, and databases including PubMed, Web of Science and Embase were searched to retrieve the related articles up to December 2017. Manual search was also conducted as a complementary method by reviewing the reference lists and prospective citation search for all retrieved studies, and relevant review articles were finished in the meantime. Then, the titles, abstracts and full texts of included articles were screened step by step according to inclusion and exclusion criteria. All of the literature research and screening were conducted by two independent researchers. The methodological quality assessment of the included studies was conducted by two independent researchers according to the guidelines of the Newcastle-Ottawa Scale (NOS). Each question was answered with ‘‘yes’’, ‘‘no’’ or ‘‘unclear’’. A ‘‘yes’’ answer represents the low risk of bias and is assigned a score of 1, while a ‘‘no’’ or ‘‘unclear’’ answer represents a high risk of bias and is assigned a score of 0. Thus, the quality score of study ranged from 0 to 9 stars. Higher scores were indicative of lower risk of bias.

Due to the different measurements, standardised mean difference (SMD) and 95% confidence interval (CI) were chosen as effect size to combine and analyse the change in axial length. Cochran’s Q test and I² test were carried out to analyse the heterogeneity among the included articles. A P value of Cochran’s Q test above 0.05 or I² [50% indicated moderate or high levels of heterogeneity] and random effect model would be utilised to accomplish this meta-analysis. On the contrary, fixed effect model would be used. Begg’s test (P=0.131) and Egger’s test were used to assess publication bias. All these statistical analyses were carried out by STATA 12.0 software.

Applicability/external validity: The authors did not discuss the external validity of their results. Nonetheless, they acknowledge some limitations: first, only data of axial length change was merged to assess the therapeutic effect of OK lens; only three of the included studies have done three-year or longer follow-up researches; third, of the included 13 studies, only two studies were from Spain – others were all from East Asia.

Geographic focus: Not discussed.

Summary of quality assessment:

In conclusion, this review was attributed medium confidence. Although authors used rigorous methods to analyse data of included studies, the literature search was not rigorous enough that we are confident that relevant studies were not omitted in the review.