Abstract
OBJECTIVE: Local corticosteroid injections, commonly accepted by rheumatologists to be effective treating painful shoulder, have shown controversial results. High frequency ultrasonography is an accurate and safe imaging modality for guiding musculoskeletal injections. We prospectively compared the short term response to randomized blind injection versus sonographic-guided injection of local corticosteroid in patients with painful shoulder. METHODS: We studied 41 consecutive patients with painful shoulder. Patients with previous trauma or chronic inflammatory arthritis were excluded. No patient had received previous physiotherapy or local steroid injection in the shoulder. Patients were randomized to receive either a blind subacromial injection of 20 mg triamcinolone (Group 1, n = 20) or a sonographic guided injection of 20 mg triamcinolone (Group 2, n = 21) by the same rheumatologist blinded to the clinical evaluation. In both groups we recorded shoulder abnormalities and the location of the steroid postinjection by ultrasound. Each patient was clinically assessed within 5 days before injection and 6 weeks after injection by another rheumatologist without knowledge of the injection technique performed. Clinical assessment included demographic and clinical data, a visual analog scale (VAS) for pain (0-100), the Shoulder Function Assessment (SFA) scale (0-70), and postinjection adverse effects. No patient received physical therapy during the followup period. Initially, demographic, clinical, and ultrasonographic findings in both groups showed no significant differences. RESULTS: Six weeks after injection, the VAS and the SFA score showed a significantly greater improvement in Group 2 compared with Group 1 (mean VAS score change 34.9 for Group 2 vs 7.1 for Group 1, p < 0.001; and mean SFA score change 15 for Group 2 vs 5.6 for Group 1, p = 0.012). One patient in Group 1 reported mild postinjection adverse effects. CONCLUSION: We suggest that sonographic-guided corticosteroid injections should be indicated, at least, in patients with poor response to previous blind injection to ensure accurate medication placement in order to improve therapeutic effectiveness.