Materials and Methods: Retinal vascular caliber of 42 patients with diabetic macular edema aged between 41-78 were measured before, 1 week and 4 weeks after the intravitreal injection. 21 (50.0%) patients received 2 mg/0.05 ml intravitreal triamcinolone acetonide and 21 (50.0%) patients received 1.25 mg/0.05 ml intravitreal bevacizumab. Retinal vascular caliber assessment was performed manually on retinal nerve fiber layer imaging mode scan by using semi-automated program of Spectralis optical coherence tomography. Statistical analyses of differences between retinal vascular caliber were determined using repeated measures of analysis of variance and Bonferroni adjustment.

Results: Mean retinal arteriolar diameter of healthy control group, triamcinolone group, and bevacizumab group before the injection was 88.55±11.29, 87.19±10.00 and 91.67±16.69 μm, respectively. Mean retinal venular diameter of healthy control group, triamcinolone group, and bevacizumab group before the injection was 127.00±16.91, 130.38±24.01 and 132.52±18.74 μm, respectively. In triamcinolone group, there was statistically significant decline in retinal arteriolar and venular diameter during the study period (p values were 0.047 and <0.001, respectively). In bevacizumab group, retinal venular diameter was significantly decreased (p<0.001) but there was no significant change in retinal arteriolar diameter during the study period (p=0.209).

Conclusions: In short-term, intravitreal triamcinolone acetonide led to more significant changes than intravitreal bevacizumab in retinal vascular caliber. Keywords : Bevacizumab, diabetic macular edema, macular volume, retinal vascular caliber, triamcinolone acetonide