Steroid Injection and Nonsteroidal Anti-inflammatory Agents for Shoulder Pain: A PRISMA Systematic Review and Meta-Analysis of Randomized Controlled Trials

Medicine (Baltimore). 2015 Dec;94(50):e2216. doi: 10.1097/MD.0000000000002216.

Abstract

Advantages and possible risks associated with steroid injection compared with nonsteroidal anti-inflammatory drugs (NSAIDs) for shoulder pain are not fully understood. To compare the efficiency and safety of steroid injection versus NSAIDs for patients with shoulder pain. PubMed, Embase, and the Cochrane Library were searched through July 2015. Study eligibility criteria, participants, and interventions: randomized controlled trials (RCTs) that assessed steroid injection versus NSAIDs for patients with shoulder pain. Study appraisal and synthesis methods: predefined primary efficacy outcome was functional improvement; and secondary efficacy outcomes included pain relief and complications. Relative risks (RRs) and standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated using a random-effects model accounting for clinical heterogeneity. Eight RCTs involving 465 participants were included in the meta-analysis. Five trials compared steroid injection with oral NSAIDs, and 3 compared steroids injection with NSAIDs injection. Compared with steroid injection, oral NSAIDs were less effective in 4 or 6 weeks for functional improvement (SMD 0.61; 95% CI, 0.08-1.14; P = 0.01), while there was no significant difference in pain relief (SMD 0.45; 95% CI, -0.50-1.40; P < 0.00001) or complication rate (RR 1.10; 95% CI, 0.26-4.58; P = 0.29). Meta-analysis was not performed for NSAIDs injection due to considerable heterogeneity. Conflicting results were observed in favor of either steroid or NSAIDs injection. Not all diseases that can lead to shoulder pain were included, detailed intervention protocols were inconsistent across studies, and some estimated data were input into comparison while some data were lost, which could exert an influence on pooled results. Steroid injection, compared with oral NSAIDs, provides slightly more improvement in shoulder function without superiority in pain relief or risk of complications at 4 to 6 weeks. Treatment decision should be made based on diseases. NSAIDs injection might be a treatment method for shoulder pain.

Publication types

  • Meta-Analysis
  • Research Support, Non-U.S. Gov't
  • Review
  • Systematic Review

MeSH terms

  • Anti-Inflammatory Agents, Non-Steroidal / therapeutic use*
  • Glucocorticoids / therapeutic use*
  • Humans
  • Injections, Intra-Articular
  • Randomized Controlled Trials as Topic
  • Shoulder Pain / drug therapy*

Substances

  • Anti-Inflammatory Agents, Non-Steroidal
  • Glucocorticoids