Purpose: To determine the limitations of randomized sham surgery-controlled trials in orthopaedic sports medicine and fidelity of the trials' conclusions.
Methods: Randomized placebo surgery-controlled trials in orthopaedic sports medicine were included in this Preferred Reporting Items for Systematic Reviews and Meta-Analyses-guided systematic review. Several aspects of investigation design and conduct were analyzed: genetic analysis for propensity to placebo response, equipoise of informed consent process, geography of trial subjects, percentage of eligible subjects willing to be randomized, changes from protocol publication to results publication, and perioperative complications.
Results: Seven sham surgery-controlled trials (845 subjects [370 knees, 449 shoulders, 26 elbows]; 5 from Europe, 1 from North America, and 1 from Australia; all superiority model, efficacy design) were analyzed. There were consistent methodologic deficiencies across studies. No studies reported genetic analysis of susceptibility to placebo response. Three studies (43%) were underpowered. Crossover rates ranged from 8% to 36%, which led to un-blinding in up to 100% of subjects. There were low enrollment rates of eligible subjects (up to 57% refused randomization). Follow-up was short term (2 years or less in all but one study). Complication rates ranged from 0% to 12.5%, with complications occurring in both groups (no significant difference between groups in any study).
Conclusions: Randomized sham-controlled studies in orthopaedic sports medicine have significant methodologic deficiencies that may invalidate their conclusions. Randomized trial design (with or without placebo control) may be optimized through the inclusion of per-protocol analysis, blinding index, equivalence or noninferiority trial design, and a nontreatment group.
Level of evidence: Level II Systematic Review of Level II studies.
Copyright © 2020 Arthroscopy Association of North America. Published by Elsevier Inc. All rights reserved.