Photorefractive keratectomy (PRK) in patients with thin corneas: A systematic review and meta-analysis of clinical outcomes and complications.

Topic: To systematically investigate the safety and efficacy of photorefractive keratectomy (PRK) or transepithelial photorefractive keratectomy (transPRK) in patients with thin corneas.

Clinical relevance: Concerns exist regarding the outcomes of refractive surgeries including PRK/trnasPRK in patients with thin corneas.

Methods: This study was registered prospectively in PROSPERO with an ID of CRD42023457156. It was conducted according to the Cochrane Handbook for Systematic Reviews of Interventions and reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 Consensus Statement. We searched six databases and three registries from January 1, 1985, until November 30, 2024: PubMed, Web of Science, Scopus, Google Scholar, Embase, ProQuest, Cochrane Central Register of Controlled Trials (CENTRAL), ClinicalTrials.gov, and International Clinical Trials Registry Platform (ICTRP). Our primary outcomes include the safety index (SI), efficacy index (EI), and best-corrected visual acuity (BCVA) changes, while our secondary outcomes include predictability and the rate of complications (including corneal ectasia and haze). We included any observational or interventional study reporting data on the primary or secondary outcomes of PRK or transPRK in patients with a mean central corneal thickness (CCT) <500 μm, with no history of previous ocular pathology or surgery. The outcome estimates and 95% confidence intervals (CI) were derived from a random-effects model meta-analysis to account for possible heterogeneity.

Results: A total of 946 eyes from 10 studies were included. All included studies were case series, either retrospective (n=4) or prospective (n=6). Four studies (n=4) reported the SI and EI, with pooled estimates of 1.01 (95% CI: 0.73 to 1.29) and 1.01 (95% CI: 0.98 to 1.04) after sensitivity analysis, respectively. The mean post-operative change in BCVA was -0.01 (95% CI: -0.07 to 0.05). Compared to baseline values, a mild yet non-significant improvement in BCVA was observed (Standardized mean difference: -0.20 95% CI: -0.89 to 0.49). The pooled predictability within ±0.5 D of emmetropia, based on data from six studies, was 0.83 (95% CI: 0.98 to 1.04). No study reported any cases of corneal ectasia after surgery.

Conclusion: PRK/transPRK appears to be a safe and effective refractive surgery for thin corneas with simple or astigmatic myopia and without any other ocular pathologies/topographic abnormalities. However, the power analysis and insufficient follow-up call for future studies to draw definitive conclusions about safety and efficacy.