PURPOSE: To compare intravitreal bevacizumab monotherapy with intravitreal dexamethasone delayed delivery system monotherapy for persistent diabetic macular edema. METHODS: Single-center, randomized, subject-masked study of eyes with persistent diabetic macular edema, defined as central subfield thickness (CST) >340 μm despite ≥3 anti-vascular endothelial growth factors injections within 5 months. The intravitreal bevacizumab monotherapy (n = 23 eyes) and delayed delivery system monotherapy (n = 27 eyes) groups received treatments q1month and q3months, respectively. RESULTS: Baseline best-corrected visual acuity and CST were similar in the two groups. At Month 7, the mean final best-corrected visual acuity (mean ± SD) was 65 ± 16 letters (mean Snellen visual acuity 20/50) and 64 ± 11 letters (20/50) (P = 0.619), the mean change in best-corrected visual acuity was +5.6 ± 6.1 and +5.8 ± 7.6 letters (P = 0.785), the mean final CST was 471 ± 157 and 336 ± 89 μm (P = 0.001), and the mean change in CST was -13 ± 105 and -122 ± 120 μm (P = 0.005) in the intravitreal bevacizumab monotherapy and delayed delivery system monotherapy groups, respectively. The number of injections was 7.0 ± 0.2 and 2.7 ± 0.5 (P < 0.001) in the 2 groups. CONCLUSION: The two groups had similar best-corrected visual acuity gains. The delayed delivery system monotherapy group achieved a significantly greater reduction of CST compared with the intravitreal bevacizumab monotherapy group, with a q3month interval of treatment, and had no recurrent edema at any visit.