2017 ISAKOS Biennial Congress ePoster #2129

 

Intra-Articular Steroid Injection For Frozen Shoulder: A Systematic Review And Meta-Analysis Of Randomized Controlled Trials With Trial Sequential Analysis

Yaying Sun, MD, Shanghai CHINA
Peng Zhang, PhD, Shanghai CHINA
Hong Li, MD, Shanghai CHINA
Jia Jiang, PhD, Shanghai CHINA
Shiyi Chen, MD, PhD, Prof., Shanghai CHINA
Jiwu Chen, MD, PhD, Shanghai CHINA

Department of Sports Medicine, Shanghai, CHINA

FDA Status Not Applicable

Summary

Compared with placebo, the effect of steroid injection lasts for at least 12-16 weeks, longer than the previously acknowledged six weeks.

Abstract

Backgroud:
Intra-articular steroid injection is a common intervention for frozen shoulder with, however, inconsistency in the effect.

Purpose

To illustrate the effect of steroid injection for frozen shoulder.
Study design:
Systematic review and meta-analysis

Method

Pubmed, Embase, and Cochrane library were searched for randomized controlled trials (RCTs) comparing intra-articular steroid injection with no or sham injections for frozen shoulder. Primary outcome was pain visual analog score (VAS), and secondary outcomes were passive external rotation, abduction, flexion, and internal rotation/hand behind back, and functional scores. Safety outcome was complication rate. Comparisons were calculated by mean difference (MD). Trial sequential analysis (TSA) was used to verify the pooled results.

Results

Eight RCTs with 416 patients were included. All injections were performed only once. Compared with control, intra-articular steroid injection significantly reduces VAS at 4-6 weeks (MD 1.28cm, 95% CI 0.75, 1.82) and 12-16 weeks (MD 1.00 cm, 95% CI 0.47, 1.52) after injection, which was confirmed by TSA, and might maintain to 24-26 weeks (MD 0.65cm, 95% CI 0.19, 1.10), which was not confirmed by TSA. Steroid injection was beneficial regarding passive external rotation, abduction, flexion, and Shoulder Pain and Disability Index at 4-6, 12-16, and 24-26 weeks, and Association of Shoulder and Elbow Surgery score at 12-16 weeks (MD 12.20, 95% CI 2.55, 21.85). Steroid injection was not superior to control regarding Constant score (MD 5.70, 95% CI -0.59, 11.99), and internal rotation/hand behind back except for 12-16 weeks (MD 0.81, 95% CI 0.18, 1.44) and 24-26 weeks (MD 3.88cm, 95% CI 0.51, 7.25). The complication rate was 3.9%.

Conclusion

A single steroid injection is effective and safe for frozen shoulder, relieving pain, improving functional performance, and increasing range of motion. The effect is significant at 4-6 and 12-16 weeks after injection, and maybe last to 24-26 weeks.