Rishi P. Singh, MD (Shaker Heights, OH), Oswaldo FM Brasil, MD (Cleveland, OH), Anat Galor, MD (Cleveland, Ohio), Scott D Smith, MD, MPH (Cleveland, OH), Careen Y. Lowder, MD (Cleveland, OH), Jonathan E. Sears, MD (Cleveland, OH), Peter K. Kaiser, MD (Cleveland, OH)


To evaluate the predictive factors for visual outcome after intravitreal triamcinolone acetonide injection in the treatment of diabetic macular edema.


Retrospective chart review of patients with diabetic macular edema who met the following inclusion criteria: clinically significant diabetic macular edema, receipient of a 4mg/0.1ml intravitreal triamcinolone acetonide injection, and optical coherence tomography of the macula performed up to ten days prior to injection. All patients received a full ophthalmic examination including best-corrected Snellen visual acuity. The main outcome measure was mean change in vision three months after injection.


Multivariate linear regression analysis included 52 eyes of 42 patients. Mean change in vision three months after  injection was significantly related to baseline visual acuity (-0.267 logMAR, P=0.002), presence of cystoid macular edema (0.146 logMAR, P=0.03) and interaction of epiretinal membrane and baseline visual acuity (0.282 logMAR, P=0.0008). The presence of subretinal fluid was highly suggestive of clinical correlation, but not statistically
significant (-0.170 logMAR, P=0.06). Vision change three months after injection was statistically independent (P>0.1) of all other studied variables.


OCT characteristics and baseline visual acuity is useful in predicting visual outcomes 3 months after intravitreal triamcinolone acetonide injection for DME.