Intra-articular injections have traditionally been performed "blind," guided only by anatomic palpation. Many may be placed inaccurately, and the use of imaging may significantly improve injection accuracy. This review analyzed the impact of ultrasound-guidance on injection accuracy and clinical efficacy in the knee and shoulder joints. Ultrasound-guided injections were significantly more accurate than blind injections in clinical studies of the knee joint (92.7% vs. 77.9%, p < 0.05). Shoulder injections into the glenohumeral and acromioclavicular joints showed modest benefit in cadaver studies, but improved accuracy has not been demonstrated in clinical studies. Insufficient data was available to show any effect in the subacromial bursa. Preliminary efficacy research shows that that ultrasound guidance may modestly improve the subjective pain relief and functional improvement provided by these injections.