Methodological quality of the review: Medium confidence
Author: Fang Li, Zhi-Xin Jiang, Peng Hao, and Xuan Li
Region: USA, Australia, China, Japan, New-Zealand, Brazil
Subsector: Corneal thickness (CCT) in the treatment of orthokeratology of myopia
Equity focus: None specified
Study population: Patients aged from 7 to 37 years
Type of programme: Hospital based
Review type: Other review
Quantitative synthesis method: Meta-analysis
Qualitative synthesis method: Not applicable
Background: Orthokeratology (OK) is a clinical technique that uses specially designed rigid gas-permeable contact lenses to reshape the cornea to temporarily modify, reduce or eliminate refractive errors and astigmatism to achieve a transient improvement in unaided visual acuity. Nowadays, OK is usually regarded as an effective treatment of myopia, not as visual aids. Central corneal thickness (CCT), which is also very important for the planning of refractive surgery and diagnosis of glaucoma, is being widely studied to monitor the retardation of myopia and OK-related complications, such as regression, keratectasia, and so on. Different studies have demonstrated different CCT changes through the programmed application of rigid contact lenses. Therefore, it is not entirely clear when and how OK affects CCT and whether short-term wear and long-term wear produce the same change.
Objectives: To assess the change of central corneal thickness (CCT) in the treatment of orthokeratology in patients with myopia.
Main findings: A total of 10 published studies (239 patients (339 eyes)) which satisfied the eligibility criteria were included in this meta-analysis. Four studies compared the change of CCT between day 1 and 1 week, and four studies compared CCT change between day 1 and 1 month. Of the 10 studies, three studies were conducted in Australia, two studies each were conducted in the USA and China, and a study each was conducted in Japan, New Zealand and Brazil.
The results showed that the CCT reduced significantly (P=0.005) by 5.73 mm from day 1 to 1 week, with 95% CI: 1.75 to 9.7. Sensitivity analysis was performed by excluding every study. The results demonstrated that a significant CCT reduction of 5.89 mm occurred from day 1 to 1 month, with 95% CI: 3.5 to 8.29 (P<0.001). Sensitivity analysis also showed no change in the overall results (all P<0.05).
The results showed that there was no significant CCT reduction from 1 week to 1 month, with 95% CI: -2.79 to 8.5 (P=0.32). No significant CCT reduction from 1 to 2 weeks, with 95% CI: -0.83 to 6.2 (P,0.001). Only 2 studies reported CCT change after six months of OK wear. No significant publication bias was shown for CCT change between 1 day and 1 week, 1 day and 1 month, 1 week and 1 month, or 1 week and 2 weeks by Begg rank correlation method, all P values were larger than 0.05 (P=1.00, 0.308, 1.00, 0.089, respectively).
Based on findings from the review, authors conclude that most reduction of CCT occurred during the first week and remained thinner for 1 month. They also recommend further randomised controlled trials with larger sample sizes and standardised outcome measurements; different follow-up periods are warranted to find the precise change.
Studies were included based on the following criteria: articles from peer-reviewed medical journals; the subjects should be patients who had a low to moderate amount of myopia (-0.75 to -6.0 diopter [D]) and with or without with-the-rule corneal astigmatism of up to 1.5 D; the follow-up period should include at least 1 day; sufficient data should be provided to gain or calculate mean value and standard deviation of CCT. Authors defined central corneal thickness as the total cornea thickness averaged centrally over a circular area of 2 mm or obtained across a horizontal meridian from specific corneal centre locations by equation or visually. For each study, the following data were extracted: titles, authors, year of publication, location of the trial, number of subjects/eyes included in the statistical analysis, patient age, sex, and the baseline spherical equivalent (in diopters).
Authors conducted a systematic search of Ovid, Embase, Medline, and ScienceDirect to identify published reports of clinical trials comparing CCT change after overnight OK for patients with myopia. Searches were restricted to English language reports published before April 30 2014. The keywords were corneal thickness or corneal pachymeter, orthokeratology, ortho-k, or OK lens. In addition, the reference lists of original reports and review articles retrieved through the search were reviewed for additional studies. Two reviewers independently screened the titles and abstracts for the eligibility criteria. Authors assessed sources of systematic bias and the quality of the trials according to the methods described by Jadad.
As CCT is a type of continuous variance, authors calculated the weighted mean difference. I2 was used to quantify the inconsistency among the results of studies. The Mantel-Haenszel method for fixed effects and the Der Simonian and Laird method for random effects were used to estimate pooled effects.
Applicability/external validity: The authors do not discuss the generalisability of the results. However, the authors acknowledge some limitation to consider if the results want to be used in another context: first, different measurements were used to evaluate CCT, and the effects of instrument evolution in precision or patient compliance during the evaluation are not negligible. Second, the statistical efficacy may be inadequate because the number of studies included is finite. The limited number of studies or the limited sample size of the pooled results may influence the stability of the results. In addition, different studies chose different time points to observe changes in CCT, whereas the meta-analysis was limited to the time points of one day, one week, and one month. The change in CCT within one day is still unclear.
Geographic focus: No focus on low or middle income countries, or all countries.
Summary of quality assessment:
Overall, medium confidence was attributed to the conclusions about the effects of this study. Authors did not conduct thorough searches of the literature to ensure that all relevant studies were included in the review, which can impact on the validity of findings. In addition, authors did not avoid language bias as studies written in English only were included. The table summarising the quality assessment of each study was not included and data extraction was done by one reviewer.