Methodological quality of the review: Medium confidence

Author: Yuan Sun, Fan Xu, Ting Zhang1, Manli Liu, Danyang Wang, Yile Chen, Quan Liu

Region: Hong Kong, Japan, Spain, USA

Sector: Myopia

Subsector: Myopia control

Equity focus: No

Study population: Children

Type of programme: Hospital based

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: The extensive and growing prevalence, relevant ocular morbidity, and substantial costs associated with myopia have made it a significant public health issue. Walline et al. compared various interventions, including multifocal lenses, rigid and soft contact lenses, timolol drops, and muscarinic receptor antagonists, and found that the latter was the most effective in slowing myopia progression. However, they failed to assess the effectiveness of orthokeratology (ortho-k); this technique involves reshaping the epithelium to correct ametropia and was first described in the early 1960s. Its efficacy was improved by the introduction of new materials and a reverse geometry lens design. In 2002, the U.S. Food & Drug Administration approved an overnight-wear contact lens by Paragon Vision Sciences, which led to improved patient compliance. Due to the fact that studies with large sample sizes were not available, we reviewed a number of relevant trials to obtain more precise estimates of the myopic control following overnight use of ortho-k lenses in pediatric subjects.

Objectives: The aim of this study was to evaluate the clinical treatment effects of orthokeratology to slow the progression of myopia.

Main findings: A total of seven eligible studies were included in the meta-analysis. These studies included a total of 546 individuals (age range from 6 to 16 years old), and 435 subjects completed the 24-month follow-up visit. The dropout rate ranged from 12.4% to 46.2%. Only two trials were randomised, and the studies employed different recruitment criteria. The pooled estimates indicated that change in axial length in the ortho-k group was 0.27 mm (95% confidence interval [CI]: 0.22, 0.32) less than the control group. Myopic progression was reduced by approximately 45%. The combined results revealed that the difference in vitreous chamber depth between the two groups was 0.22 mm (95% confidence interval [CI]: 0.14, 0.31). None of the studies reported severe adverse events. Overall, authors suggest that ortho-k can slow myopia progression in school-aged children.

Methodology: Inclusion criteria consisted of studies with: 1) measurement of axial length (AL) between the baseline and end of the study; 2) at least two comparison groups (intervention and control); and 3) the follow-up period was more than one year. The research was performed with Medline and Embase databases to identify relevant studies using the following keywords: corneal reshaping, CRT, OK, orthokeratology, ortho-k, ametropia and myopia. The final search was performed on 25 January 2014. The results were not limited by language. Authors hand-searched the references of the relevant articles and reviews to identify additional studies that may have been missed. Two researchers separately scanned the identified titles and abstracts to determine if they might meet the inclusion criteria. The following information was collected for each study: first author’s name, country or area of the research, year of publication, and AL and vitreous chamber depth (VCD) at the baselines and endpoints. Two authors separately performed the quality assessments of all the included studies. Authors used the Jadad scale to access the evidence quality of the randomised controlled trials (RCTs), with scores of 0 and 5 indicating the lowest and highest qualities, respectively. The checklist included methods of randomisation, masking and withdrawal. For the remaining observational controlled trials (CTs), a methodological index for non-randomised studies (MINORS) was used to perform the assessments. All extracted data was imported into a database, and all statistical analyses were performed using STATA 11.0 (Stata Corporation, College Station, TX). The weighted mean differences (WMDs) and 95% confidence intervals (CIs) were calculated for each study to assess AL and VCD changes between the two groups (intervention and control). Statistical heterogeneity was evaluated using the chi-square-based Q statistic and the I²

Applicability/external validity: Regarding external validity, the authors acknowledge some limitations which need to be considered when reading the results: two studies used historical comparisons; the control and intervention groups were not treated and assessed during the same time period. No studies reported the use of double-blinding design, and it was hard to hide the allocation for researchers who performed the examinations. The authors did not assess for a possible publication bias. Although this kind of bias could be tested using funnel plots and statistical methods (Begg’s and Egger’s test), these results might not be conclusive with such a small number of trials.

Geographic focus: Not discussed.

Summary of quality assessment:

Medium confidence was attributed to the conclusions about the effects of this review, as literature searches were not comprehensive enough to ensure that relevant studies were omitted in the review.