PURPOSE: The objective was to compare retinal morphology and function following intravitreal injections of bevacizumab (Avastin) or triamcinolone (Volon A) in patients with early diabetic macular edema (DME). PATIENTS AND METHODS: The study was planned as a randomized, prospective, interventional clinical trial. A total of 30 diabetic patients with treatment-naïve, clinically significant macular edema were included in this study and randomized to two equal groups. One group initially received three injections of 2.5 mg bevacizumab in monthly intervals. The second group received a single injection of 8 mg triamcinolone, followed by two sham interventions. Functional and anatomic results were evaluated monthly using ETDRS vision charts and spectral-domain optical coherence tomography. According to the study protocol, retreatment after 3 months was dependent on functional and anatomic outcome in a PRN regimen. RESULTS: Baseline best corrected visual acuity (BCVA) was 0.30 logMAR and central retinal subfield thickness (CSRT) was 505 µm in the bevacizumab group and 0.32 logMAR and 490 µm CSRT in the triamcinolone group. After 3 months, BCVA improved to 0.23 logMAR (bevacizumab) and 358 µm CRST and 0.26 logMAR (triamcinolone) and 308 µm CSRT. After 12 months, BCVA further recovered in the bevacizumab group (0.18 logMAR) but slightly decreased in the triamcinolone group (0.36 logMAR). CONCLUSION: Intravitreal bevacizumab and triamcinolone are both equally effective in reducing CSRT in early DME. After 6 months, rehabilitation of vision was comparable in both treatment arms, whereas at the final follow-up at month 12, BCVA was superior in the bevacizumab than in the triamcinolone sample. This may be related to cataract development following steroid treatment, as well as to substance-specific mechanisms within the angiogenic versus the inflammatory cascade.