Purpose: This study compared the safety and efficacy of prostatic arterial embolization (PAE) with that of trisacryl gelatin microspheres of different sizes for treatment of benign prostatic hyperplasia (BPH). Materials and Methods: This study consisted of a single-center, randomized controlled clinical trial in 138 patients who underwent PAE for BPH between July 2015 and December 2016. Patients were randomized to PAE using microspheres of different sizes: group A patients were treated with microspheres 100–300 μm, group B with 300–500 μm, and group C with 100–300 μm followed by 300–500 μm. All patients were evaluated before and at 1, 3, 6, 12, and 18 months after PAE. Baseline data were comparable across the 3 groups, using the following mean International Prostate Symptom Score/quality of life (IPSS/QoL); prostate volume (PV) scores, respectively: 23.0/4.14; 87.9 cm 3 (group A); 23.0/4.09; 89.0 cm 3 (group B); and 24.2/4.29; 81.0 cm 3 (group C) (P > 0.05). Results: Mean IPSS/QoL scores; PV after PAE were: 9.98/2.49; 65.1 cm 3 (group A); 8.24/2.26; 63.1 cm 3 (group B); and 10.1/2.69; 53.1 cm 3 (group C) (P = 0.23; P = 0.39; P = 0.24). There were 26 clinical failures. The cumulative probabilities of clinical success at 18 months were 76.7% in group A, 82.6% in group B, and 83.3% in group C (P = 0.68). Nontarget embolization was prevented in 6 patients by coil embolization. All adverse events were mild and self-limited with rates of 86.0% in group A (37 of 43); 41.3% in group B (19 of 46); and 58.3% in group C (28 of 48) (P < 0.001). Dysuria was the most frequent adverse event (28 of 137 [20.4%]). Conclusions: PAE outcomes were not significantly different among microspheres of different sizes. The use of 100- to 300-μm microspheres was associated with an increased risk of minor adverse events.