Predictive factors associated with the safety and efficacy of refractive surgery for Myopia: a retrospective cohort...

Wavefront-guided refractive surgery delivers excellent visual outcomes for myopia, but the relative contribution of demographic, refractive, and corneal preoperative variables to safety and efficacy remains incompletely characterized in real-world single-surgeon cohorts. We aimed to identify independent predictors of safety and efficacy after wavefront-guided femtosecond LASIK (i-LASIK) and wavefront-guided PRK (wfPRK) for myopia, using a methodologically robust analytical approach.

Single-center, single-surgeon retrospective cohort of consecutive myopic patients operated between January 2020 and July 2024. After exclusion of hyperopia, ectatic disorders, incomplete follow-up, and amblyopia (preoperative best-corrected visual acuity [BCVA] logMAR ≥ 0.3), 329 patients (648 eyes) were analyzed. The primary outcomes were the safety index (postoperative BCVA / preoperative BCVA) and the efficacy index (postoperative uncorrected visual acuity / preoperative BCVA), modelled as continuous variables. To preserve the independence assumption while making full use of bilateral data, all inferential analyses - Spearman correlations, Mann-Whitney U / Kruskal-Wallis H tests, and multivariable linear regression with eight preoperative predictors entered simultaneously - were performed separately for right and left eyes. Standardized regression coefficients (β) and 95% confidence intervals are reported.

The cohort included 162 male (49.2%) and 167 female (50.8%) patients, mean age 30.0 ± 6.4 years; 411 eyes (63.4%) underwent i-LASIK and 237 (36.6%) wfPRK. Median follow-up was 101 days (range 23 days to 4.2 years). Overall, 99.4% of eyes had a safety index above 0.85 and 98.8% had an efficacy index above 0.80. Across all four multivariable models (safety and efficacy, right and left eyes), preoperative BCVA logMAR was the strongest independent predictor (β = +0.437 to +0.682, all p < 0.001). Preoperative astigmatism was independently associated with reduced left eye safety (β = -0.102, p = 0.028), and high myopia (SEQ ≤ -6.01 D) with reduced right eye safety (β = -0.096, p = 0.040) and efficacy (β = -0.113, p = 0.043). Age, sex, and surgical technique (i-LASIK vs wfPRK) were not independently associated with either index (all p > 0.05).

In a real-world wavefront-guided refractive surgery cohort, both i-LASIK and wfPRK delivered comparably high safety and efficacy. The strongest determinant of both indices was preoperative BCVA logMAR, while preoperative astigmatism and high myopia showed small but consistent independent contributions. The absence of an independent age, sex, or technique effect supports an indication-based rather than procedure-based decision framework and a more nuanced view of refractive surgery outcome prediction.