A systemic review and network meta-analysis of accuracy of intraocular lens power calculation formulas in primary angle-closure conditions

Author: Lu W, Hou Y, Yang H, Sun X.

Geographical coverage: Not reported

Sector: Primary angle-closure conditions

Sub-sector: Treatment

Equity focus: Not reported

Study population: Patients with glaucoma

Review type: Effectiveness review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: Primary angle-closure disease (PACD), the leading cause of irreversible blindness in East Asia, is projected to affect 32 million adults aged 40–80 by 2040. Cataract surgery has become the preferred treatment due to its benefits, including improved vision and reduced intraocular pressure (IOP). However, PACD patients undergoing cataract surgery often have less accurate refractive outcomes than those with uncomplicated cataracts, largely due to challenges in predicting intraocular lens (IOL) power. Existing IOL calculation formulae also vary in their accuracy when applied to PACD cases.

Objectives

To evaluate the accuracy of IOL power calculation formulae in patients with PACD.

Main findings

Overall, the review found no definitive evidence to support the superiority of any specific IOL power calculation formula for eyes with PAC or PACG.

The search identified 2,752 records, of which six studies involving 419 eyes and eight IOL calculation formulae were included. All included studies were retrospective observational in design, and most had a predominantly female population. The formulae assessed included Barrett Universal II (BUII), Kane, SRK/T, Hoffer Q, Haigis, Holladay I, RBF 3.0, and LSF. SRK/T served as the reference as it was used in all studies.

Direct comparisons showed no statistically significant differences in mean absolute error (MAE) or the percentage of refractive outcomes within ±0.50 or ±1.00 dioptres when comparing the other formulae to SRK/T (all P > 0.05). Network comparisons indicated that BUII, Kane, and RBF 3.0 had marginal but statistically insignificant advantages. Network funnel plots showed no evidence of publication bias.

Methodology

Two independent reviewers conducted comprehensive searches in PubMed, the Cochrane Database of Systematic Reviews, the Cochrane Central Register of Controlled Trials (CENTRAL), and Web of Science up to August 2022. Eligible studies included randomised controlled trials, cross-sectional studies, and prospective or retrospective observational studies on eyes with PACS undergoing uncomplicated cataract surgery or clear lens extraction with one-piece IOL implantation. Searches were limited to English-language publications. Reference lists of relevant articles were also screened to identify additional eligible studies.

Two reviewers independently screened articles, extracted data, and resolved any discrepancies through discussion or by consulting a third reviewer. Study quality was assessed using a modified Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. The level of evidence was graded according to the Oxford Centre for Evidence-based Medicine.

Two reviewers independently screened the articles based on the inclusion criteria. Any disagreements were resolved through discussion or with the assistance of a third reviewer. Two reviewers also independently extracted the relevant data. The quality of all included studies was assessed using a modified QUADAS-2 tool, and the level of evidence was evaluated following the Oxford Centre for Evidence-Based Medicine guidelines.

Findings were synthesised using either fixed-effect or random-effects meta-analysis models, depending on the level of heterogeneity. Heterogeneity was assessed using I² statistics and the P-value of Cochran’s Q test or the Mantel–Haenszel Q test. Publication bias was evaluated using network funnel plots. Network meta-analysis was conducted using Markov Chain Monte Carlo (MCMC) methods to derive pooled effect sizes, and results were summarised and ranked via bar plots.

Applicability / External validity

The review acknowledged limited external validity due to the small number of included studies, which may not reflect a diverse range of patient populations. The authors emphasised the need for further well-designed research to improve generalisability.

Geographic focus

The review did not restrict inclusion by geographic location, and individual study locations were not specified.

Summary of quality assessment

This review’s conclusions are based solely on studies published in English, resulting in a medium level of confidence in its findings.