Artisan versus Artiflex phakic intraocular lens implantation in the treatment of moderate to high myopia: meta-analysis

Authors: Hou C, Li H, Li J, Peng H and Wang Q.

Geographical coverage: Iran, Netherland, Spain, Korea, Japan, USA and France.

Sector: Biomedical

Sub-sector: Treatment

Equity focus: None

Study population: Patients older than 18 years of age with myopia whose refraction was stable within 1-2 years

Review type: Effectiveness review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: Phakic intraocular lens (PIOL) implantation is a safer method for correcting myopia between -6.00D and -20.00D compared to excimer laser techniques. Among various PIOLs, Artisan and Artiflex lenses are widely used due to their safety and efficacy. Both lenses are implanted in the same location and attached directly to the iris tissue, but they differ in material properties and surgical incision size. However, there is currently no consensus on the best technique due to insufficient studies and high heterogeneity in outcome measures and duration.

Objectives: To compare the postoperative safety, efficacy, predictability, visual quality and biomechanics after implantation of Artisan vs. Artiflex phakic intraocular lenses (PIOLs).

Main findings:

To summarize, the authors concluded that both the Artisan and Artiflex PIOL techniques are safe and effective for treating myopia. However, when compared to the Artisan PIOL, the Artiflex PIOL showed notable enhancements in terms of efficacy, predictability, contrast sensitivity, and higher-order aberrations (HOA). The only areas where Artiflex did not show improvement were safety and complications in the treatment of moderate to high myopia.

A total of 11 studies that met the eligibility criteria were included in the meta-analysis.

Selected trials were published from 2000 to 2020. The number of participants ranged between 30 to 628 in both PIOL groups of the studies. The number of all patients meeting the criteria was 1,164 (610 participants treated with Artisan PIOL, and 554 participants treated with Artiflex PIOL). For experimental type, the analysis included 8 retrospective studies, 2 prospective studies and 1 historical cohort study. Ten studies used lenses with Artisan model 204; Artisan model 204 and 206 were both used in the Shin et al study.

The authors determined that reporting and randomisation biases in all studies were low risk. However, due to insufficient reporting, selection, implementation, and measurement biases related to allocation concealment were deemed uncertain in most studies. The Coullet et al trial, which used a double-blind method, was assessed as low risk. Only the Aerts et al study reported incomplete outcome data, resulting in a high risk of bias assessment.

The pooled Weighted Mean Difference (WMD) and Odds Ratio (OR) showed statistical significance in terms of postoperative Best Corrected Visual Acuity (BCVA), efficacy, postoperative Spherical Equivalence (SE), predictability, contrast sensitivity, and mean intraocular Higher-Order Aberrations (HOA) for a 6mm pupil. This indicates that the Artiflex PIOL demonstrated a clear beneficial effect in correcting myopia compared to the Artisan PIOL. However, there was no significant difference in the incidence of complications between the two groups.

The study also found a high degree of heterogeneity. This remained the case even when studies were removed during sensitivity analysis and when subgroup analysis was conducted by age and aberrometer type.

The authors emphasize the necessity for high-quality, multicentre research. They also highlight the importance of gathering more data to minimize bias, thereby enhancing the accuracy and reliability of clinical follow-up outcomes and analysis results.

Methodology:

The study included English-language retrospective or prospective comparative studies involving patients over 18 years old with stable refraction for 1-2 years, excluding those with a history of ophthalmic diseases. The participants were treated with either Artisan or Artiflex phakic intraocular lenses, and the follow-up period was more than six months. The study excluded patients without moderate to high myopia (equivalent spherical lens: >− 3.00), those with contraindications or previous ophthalmic surgeries, and non-quantitative studies such as case reports, comments, editorials, and proceedings. Duplicate data from previously reported cases were also excluded.

Authors conducted a systematic retrieval to select clinical controlled trials using electronic databases, including PubMed, EMBASE and the Cochrane Library, published from January 2000 to June 2020. The following key words were used to retrieve the relevant literature: (“Artisan” OR “fixed phakic intraocular lens” OR “Artisan PIOL”) AND (“Artiflex” OR “foldable” OR “flexible phakic intraocular lens” OR “Artiflex PIOL”). Additionally, the references of the original articles were screened to prevent omissions for this meta-analysis.

Two investigators independently assessed the methodological quality of the included studies according to the risk of bias evaluation criteria recommended by the Cochrane Manual. Any disputes over evaluation results were resolved by focused discussion. Data collection was conducted independently by two researchers. In case of uncertainty about the eligibility of articles, the authors discussed to resolve the discrepancies.

Primary and secondary outcome data were analyzed using Stata version 16.0. Random-effects models were used for continuous and dichotomous variables. Weighted Mean Differences (WMDs) and Odds Ratios (ORs) were calculated for continuous and binary outcomes, respectively. Positive results indicated beneficial effects if WMD >0 or ORs >1, while negative results indicated beneficial effects if ORs <1 or WMD <0. Statistical heterogeneity was assessed using I2 statistics, with I2 >50% indicating significant heterogeneity. In case of significant heterogeneity, sensitivity or subgroup analysis was performed to ensure result stability. Due to the limited number of studies reporting certain outcomes, sensitivity analysis was primarily used. A P value less than 0.05 was considered statistically significant.

Applicability/external validity: The authors acknowledge several limitations in their study, such as the varying outcome durations in the included studies and the persistent high degree of heterogeneity. These factors may constrain the generalisability of their results.

Geographic focus: Three of the 11 included studies were undertaken in LMICs, however authors do not discuss how findings may differ in different geographical contexts.

Summary of quality assessment: The methodologies used to identify, include, and critically evaluate studies had several limitations. The search was confined to English-language publications, and only one author undertook the selection of studies. Although the data analysis approach was generally solid, there was no clear effort to categorise the included studies based on varying bias risks. The potential impact of unit of analysis errors on the reported results was not discussed. Despite numerous attempts to minimise it, a high degree of data heterogeneity persisted, as noted by the authors. Due to these factors, our confidence in this review’s findings is limited.