Abstract
Purpose: :
To determine predictive factors for poor refractive errors, defined as >±0.50 diopters refractive difference between target refraction and postoperative spherical equivalent.
Methods: :
872 cataract cases at the Massachusetts Eye and Ear Infirmary from August 2001 through July 2005 were retrospectively reviewed for postoperative one month data. Patient data included age, gender, diabetes, and history of age–related macular degeneration, glaucoma, and pseudoexfoliation. Surgical data included laterality, pre and post operative refractive parameters, target refraction, IOL power used, refractive error, and intraoperative events. Fisher exact test and paired t–tests were used to compare controls with the poor–refractive–error group. Multivariate logistic regression analysis was also performed to investigate predictive factors for poor refractive errors.
Results: :
492 female (56%) and 380 male (44%) patients underwent cataract surgery in the time frame. The mean patient age for this group was 72 ± 11 years. Among the variables tested, posterior capsule tear, vitreous loss, and preoperative best corrected visual acuity (BCVA) were significantly different between both groups (p=0.001, p=0.002, and p=0.002 respectively). However, preoperative BCVA was the only predictive factor for poor refractive error on multiple logistic regression analysis (p=0.03, OR 1.38, 95% Confidence interval, 1.02–1.87).
Conclusions: :
Poor pre–operative BCVA appears to be an independent predictor of poor refractive outcome in patients undergoing cataract extraction and IOL placement. Posterior capsule tear and vitreous loss are significantly higher in the poor refractive error group; however, this effect is decreased when analyzed in conjunction with other patient and surgery dependent variables.
Keywords: cataract • clinical (human) or epidemiologic studies: outcomes/complications