Retrospective comparison of freehand and ultrasound-guided shoulder steroid injections

The objective of this study was to assess the clinical benefits and financial feasibility of using ultrasound for steroid injections of the shoulder. A retrospective chart review and telephone survey of patients in a clinical shoulder practice were performed. ICD-9 codes and CPT codes identified patients who received shoulder injections without (2006) and with (2007) ultrasound guidance during 2 consecutive years. Results were assessed by patient assessment of relief and duration of that relief via telephone survey or from chart review. Financial data was assessed by reviewing the patients' accounts. One hundred fifty-seven injections were given in 2006 and 159 in 2007. In 2006, 126 unique patients were injected compared to 99 unique patients in 2007 (P<.001). Clinical data was collected on 272 patients (86.1%). Ultrasound had no significant effect on the amount of pain relief following injection (P=.468). One hundred thirty-six patients (50.0%) reported significant pain relief, 72 (26.5%) reported moderate, 39 (14.3%) reported mild, and 25 (9.2%) reported no pain relief following injection. During both years, 92.4% of patients had subacromial injections with no significant difference in injection sites (subacromial vs glenohumeral) between the 2 years (P=.252). Neither the injection site (subacromial or glenohumeral, P=.152) nor diagnosis (P=.540) had a significant effect on pain relief. Financial collections from injections increased as expected due to the use of ultrasound.Ultrasound guidance did not change the efficacy of steroid injections, the number of injections, or the type of injections. Steroid injections are useful for managing pain in several shoulder conditions and ultrasound guidance may not be necessary.