ABSTRACT
OBJECTIVE: To investigate the efficacy of corticosteroid (CS) injection methods for frozen shoulder. DATA SOURCES: PubMed, Embase, and Cochrane Library were searched up to May 6, 2023. STUDY SELECTION: Randomized controlled trials (RCTs) that investigated CS injection methods for frozen shoulder were included. DATA EXTRACTION: Data were extracted independently by 2 authors. Risk of bias was assessed using the RoB 2 tool. DATA SYNTHESIS: A random-effects network meta-analysis was performed within a frequentist framework. The certainty of evidence was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluations approach. A total of 66 RCTs involving 4491 patients were included. For short-term outcomes, 4-site injection (vs placebo [PLA]: standardized mean difference [SMD]=-2.20, 95% confidence interval [CI], -2.81 to -1.59 in pain; SMD=2.02; 95% CI, 1.39-2.65 in global function) was the most effective (low certainty). Rotator interval injection was the optimal treatment with moderate to high certainty (vs PLA: SMD=-1.07, 95% CI, -1.51 to -0.64 in pain; SMD=0.94, 95% CI, 0.49-1.40 in global function). For midterm outcomes, 4-site injection was most effective (vs PLA: SMD=-1.71, 95% CI, -2.41 to -1.01 in pain; SMD=2.22, 95% CI, 1.34-3.09 in global function; low certainty). Distension via rotator interval (D-RI) was the optimal treatment with moderate to high certainty (vs PLA: SMD=-1.10, 95% CI, -1.69 to -0.51 in pain; SMD=1.46, 95% CI, 0.73-2.20 in global function). Distension and intra-articular injection via anterior or posterior approaches produced effects equivalent to those of rotator interval injection and D-RI. CONCLUSIONS: Rotator interval injection, distension, and intra-articular injection had equivalent effects on symptom relief. More RCTs are required to validate the superiority of multisite injections.