Outcome Selection and Methodological Quality of Major and Minor Shoulder Surgery Studies: A Scoping Review

Background Core outcome sets aim to select and standardize the choice of important outcomes reported in clinical trials to encourage more effective data synthesis, increase the reliability of comparing results, and minimize reporting bias. A core outcome set for elective shoulder surgery has yet to be defined, and therefore a systematic assessment of outcomes and methodology is necessary to inform the development of a core outcome set. Questions/purposes The purpose of this study was to examine randomized controlled trials (RCTs) of patients having elective major or minor shoulder surgery to (1) identify the outcome domains reported; (2) determine specific outcome measurement tools that were utilized; and (3) assess the work for methodological quality and risk of bias. Methods We conducted a scoping review (a review that identifies the nature and extent of research evidence) to explore the reported outcome domains, outcome tools, and methodological quality from RCTs conducted in shoulder surgery. We considered both major shoulder surgery (defined as arthroplasty, rotator cuff repair, stabilization procedures, biceps tenodesis, or Bankart repairs) and minor shoulder surgery (simple arthroscopy, capsular plication, lateral clavicular excisions, or subacromial decompression). We queried 10 electronic databases for studies published between January 2006 and January 2015. Studies were included if they were prospective, randomized controlled, clinical trials enrolling patients who received an elective shoulder surgical intervention. We extracted data relating to trial characteristics, primary outcomes, tools used to measure these outcomes as well as methodological quality indicators. We assessed indicators of methodological quality by exploring (1) the reproducibility of power analyses; and (2) whether the primary outcomes were powered to minimum clinically important differences. Risk of bias was also assessed with the Jadad score with scores between 0 (very high risk of bias) and 5 (very low risk). Findings were qualitatively analyzed and reported according to systematic and scoping review guidelines. We included 315 studies involving 30,232 patients; 266 studies investigated anesthetic, analgesic, or surgical interventions. Results Of the 315 studies included, the most common outcome domains evaluated were analgesic (n = 104), functional (n = 87), anesthetic (n = 56), and radiologic (n = 29) outcomes, with temporal patterns noted. Studies of major shoulder surgery most commonly reported functional primary outcome domains, whereas minor shoulder surgery studies most frequently reported analgesic primary outcome domains. There were 85 different primary outcome tools utilized, which included 20 functional, 20 anesthetic, 13 analgesic, and 12 radiologic. A methodological quality assessment revealed that 24% of studies had reproducible power analyses, 13% were powered to minimum clinically important differences, and risk of bias assessment demonstrated a median (interquartile range [range]) Jadad score of 4 (3-5 [1-5]). Conclusions A wide range of outcome domains and outcome assessment tools are in common use in contemporary trials of patients undergoing elective surgery. Although some diversity is important to allow the assessment of patient populations that may have different goals, the large number of tools in common use may impair the ability of future meta-analyses to pool results effectively or even for systematic reviews to synthesize what is known. The limitations of methodological quality in RCTs may be improved by researchers following standard guidelines and considering the minimum clinically important differences in their trials to be of greater use to clinicians and their patients. Level of Evidence Level I, therapeutic study.