Purpose
Compare the response of the first injection of ranibizumab and bevacizumab in diabetic macular edema in patients without and after focal macular laser to determine the sample size needed to be able to either accept or reject the null hypothesis. The only prior study that directly compared ranibizumab to bevacizumab included only 29 patients.
Methods
48 eyes at the Williams Jennings Bryce Dorn VAMC in Columbia, SC were identified which had received injections of either bevacizumab or ranibizumab for diabetic macular edema from 1/01/08 - 03/31/12. Pre- and post-treatment central sub-foveal thickness, visual acuity, and when focal macular laser was applied were obtained. The mean and standard deviation of change of the central macular thickness of ranibizumab and bevacizumab with and without prior focal macular laser targeting injections given between 01/01/08 - 10/31/11 were compared by Student' s T-test. Exclusion criteria: exudative ARMD, vein occlusion, intravitreal or periocular injection within a year.
Results
There was a statistical difference in the pre-treatment central sub-foveal thickness of the ranibizumab without focal group when compared to the ranibizumab after focal group with the later being more edematous (p=0.02). There was no statistical difference between the ranibizumab without focal and bevacizumab without focal (p=0.24). There was no statistical difference between ranibizumab after focal and bevacizumab after focal (p=0.28). There was no statistical difference between ranibizumab without focal and ranibizumab after focal (p=0.57). There was no statistical difference between bevacizumab without focal and bevacizumab after focal (p=0.51).
Conclusions
Although statistical differences were not detected between any treatment groups, linear regression analysis for the ranibizumab groups, both without and after focal, followed a trend toward greater effectiveness when compared to both bevacizumab groups. This study suggests that there may be a difference in the mean reduction of central sub-foveal thickness of ranibizumab when compared to bevacizumab. More importantly, this study gives a information regarding reasonable means and standard deviations required to determine an adequate sample size. Based on these findings, approximately 45 patients per group are required to have sufficient power to either accept or reject the null hypothesis.
Keywords: 585 macula/fovea •
499 diabetic retinopathy •
561 injection