Abstract
This multi-center cohort study included 3401 myopic laser in situ keratomileusis (LASIK) procedures conducted in 1756 myopia patients between 2002 and 2005. Pre- and postoperative uncorrected visual acuity (UCVA), best corrected visual acuity (BCVA), and manifest refraction spherical equivalent (SE) were recorded. Factors predicting low postoperative efficacy (defined as a postoperative UCVA < 0.5) were identified using univariate and multivariate logistic regression analysis. Compared with 1 month postoperatively, logMAR UCVA at 3 months postoperatively was significantly decreased (p = 0.002) and that at 2 and 3 years was significantly increased (p < 0.001). LogMAR BCVA at 2 years postoperatively was significantly decreased compared with 1 month postoperatively (p = 0.008). Over the 3-year postoperative period, overall refractive predictability within ±1.00 D and ±0.50 D ranged from 69.0% to 86.2% and from 43.3% to 67.8%, respectively. This also decreased from 1 month to 6 months postoperatively (p < 0.005). Multivariate logistic regression analysis using generalized estimating equations, revealed that higher preoperative SE (odds ratio [OR], 2.58 and 7.23; p < 0.001) and lower preoperative BCVA (OR, 2.44; p = 0.003) were predictive of a low postoperative efficacy. In summary, myopic LASIK can be effective and safe with a high refractive predictability in a Korean population, but myopic regression occurs over time. Higher preoperative SE and lower preoperative BCVA are predictive of a low postoperative efficacy.