Purchase this article with an account.
Jose M. Herreras, Victoria De-Juan, Itziar Fernandez, Raul Martin, Inmaculada Perez, Guadalupe Rodriguez, Angela Morejon, Ana del Rio; Prediction Of Refractive Stabilization After Cataract Surgery. Invest. Ophthalmol. Vis. Sci. 2011;52(14):6243.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To develop a formula to predict the refractive stabilization (RS) after phacoemulsification with intraocular lens (IOL) implantation.
Automated refraction was measured before, one day and weekly for 4 weeks after surgery in a prospective study with 163 eyes undergoing uneventful cataract surgery in healthy eyes. Kaplan-Meier analysis was used to determine the probability of RS during follow-up. Several potential predictors were evaluated with a univariate Cox proportional hazards models and log-rank test. Preoperative predictors were age, eye surgery, sex, lens opacity (LOCS grading scale), optical biometry (IOLMaster V.5), ultrasonic biometry (Otiscan-1000) automated refraction, uncorrected visual acuity, best corrected visual acuity, anterior chamber depth and pachymetry (Orbscan II) and intraoperative predictors (surgical technique, ultrasound time, aspiration time and flow rate) were included.A multivariate Cox model was fitted with potential good predictors in univariate analysis; also it was optimised by stepwise variable selection. The performance of the formula was assessed by bootstrapping, according to stability of stepwise selected model, discrimination (c-index), and calibration (slope shrinkage).
Refraction was achieved at 8th day in 30% (CI95% 22-36); at 15th day in 60% (CI95% 52-68) and at 4th week in 95% (CI95% 88-98) of the cases.The predictive equation for RS was ln[ h(t)/h0(t)] = 0.11 Power LIO + 0.28 Axial length (multivariate Cox regression analysis), where h(t) is the probability of RS at time t and h0(t) is the baseline probability when all the explanatory variables are zero. These predictors were selected by over 40% of cases (1000 bootstrap samples). Discrimination of the final model was 0.602 (95%CI: 0.601-0.604) and calibration by the slope shrinkage was 0.901 (95%CI: 0.337-1.466). Rest of the preoperative and intraoperative predictors factors were no predictive effect of refractive.
Postoperative RS after cataract surgery can be estimated from preoperative biometry in healthy eyes and uneventul surgery. This information could be useful to surgeon to improve the information given to patients before surgery.
This PDF is available to Subscribers Only