Purpose: To evaluate the effects of posterior subtenon triamcinolone acetonide (TA) injection on clinical, angiographic, and optical coherence tomographic (OCT) features in refractory diabetic macular edema (DME).Method: In a randomized double-masked placebo-controlled clinical trial, 38 consecutive patients (64 eyes) with DME refractory to previous laser therapy or not suitable for such treatment were studied. An equal number of eyes (32) were allocated in the treatment and control groups. Posterior subtenon injection of 40 mg TA in the treatment group and subconjnctival injection of 0.1 ml of lidocaine 2% in the control group was performed and repeated after two months. Clinical, angiographic, and OCT features were evaluated before intervention and after 4 months. Quantitative measurement of hard exudates (HE), size of foveal avascular zone (FAZ), and leakage was done using Photoshop 7-0 software.Results: Before injection, best corrected visual acuity (BCYA) was 0.93±0.39 LogMAR in the control group and 0.75±0.38 LogMAR in the treatment group (P= 0.078). At the last follow up, BCYA was 0.88±0.48 LogMAR in the control group and 0.71±0.42 LogMAR in the treatment group (P= 0.136). Before injection, central macular thickness (CMT) was 388.2±119.1 μm in controls and 392.2±153.6 μm in cases (P= 0.9). At the end of the study, CMT was 375.4±154.66 μm and 377.37±180.04 μlm in cases and controls, respectively (P=0.5). There was also no significant difference in HE, FAZ, and leakage changes in the angiograms. In 2 eyes of the treatment group, intraocular pressure exceeded 20 mmHg which was controlled by medication.Conclusion: Posterior subtenon injection of TA in the eyes with severe DME seems to have no effect on visual acuity and macular thickness.