Abstract
Purpose::
Numerous anecdotal risk factors for CME have been proposed, with no evidence-based studies to date. Our objective was to elucidate predictors of clinically significant CME after cataract surgery in patients with and without glaucoma.
Methods::
This was a retrospective observational study whereby a chart review was conducted on a glaucoma practice at a tertiary care setting in London Ontario Canada. Clinical information was compiled for patients who underwent cataract surgery (posterior chamber phacoemulsification) done by one surgeon between June 1997 - September 2006. Univariate analyses and multivariate logistic regression modeling was performed.
Results::
697 patient files were reviewed and the rate of CME was 1.3% and 2.3% for the entire data set and the glaucoma medication subgroup respectively.For the entire data set, multivariate logistic regression revealed four variables that were strongly associated with clinically significant CME to a p-value of 0.05. These independent predictors, along with their adjusted odds ratio (OR) are as follows:1. Pre-operative topical fixed combination beta-blocker & carbonic anhydrase inhibitor (i.e. Cosopt) -- OR 10.72. Intra-operative pupil stretching -- OR 20.43. Intra-operative PMMA intraocular lens implant -- OR 65.74. Post-operative topical prostaglandin -- OR 8.5Based upon these predictors, an integer scoring system was derived which allowed for determination of absolute risk and risk level (i.e. low, moderate, and high risk).For the glaucoma medication subgroup, multivariate logistic regression revealed two variables that were strongly associated with clinically significant CME to a p-value of 0.05. These independent predictors, along with their adjusted odds ratio (OR) are as follows:1. Pre-operative topical fixed combination beta-blocker & carbonic anhydrase inhibitor (i.e. Cosopt) -- OR 6.92. Intra-operative pupil stretching -- OR 14.7
Conclusions::
This was a retrospective chart review of almost 700 charts. Predictors of clinically significant CME were identified in the multivariate analyses for the entire data set and for the glaucoma medication subgroup. An integer scoring system was then generated for the entire data set, which lead to absolute risk determination and risk stratification.
Keywords: clinical (human) or epidemiologic studies: risk factor assessment