Traditional evaluation of non-traumatic shoulder pain (NTSP) involves four radiographic views of the joint: AP, Grashey (G), axillary (AX), and lateral scapular (LS). The LS view has historically been useful in diagnosing various glenohumeral injuries. However, some studies have questioned the clinical utility of the LS view. PURPOSE: To determine if including the LS view makes meaningful contribution to accurate diagnosis and treatment plan for NTSP. METHODS: Two sets of fifty clinical vignettes were developed based on patients who presented with NTSP. Each set included identical patient history and physical exam findings, but different radiographs. The first set included AP, G, and AX views. The second set included AP, G, AX, and LS views. Four independent raters initially evaluated the first set of vignettes and provided diagnoses, need for further imaging, and treatment recommendations. Each rater repeated the evaluation one month later with the second set of vignettes. Cohen's kappa was used to assess intra-observer reliability across both sets of vignettes. A survey was also created and distributed to members of American Shoulder and Elbow Surgeons (ASES) to evaluate how often surgeons utilize the LS view, and for what reason. RESULTS: Cohen's kappa demonstrated substantial intra-rater reliability between the two sets of vignettes for diagnosis (0.702), x-ray findings (0.645) and further studies (0.620). Percent agreement was calculated and showed that raters maintained the same diagnosis in 77.5% of cases across both sets of vignettes. Furthermore, none of the raters recommended further x-rays when evaluating the first set of vignettes, suggesting they did not feel a need for the LS view prior to developing a diagnosis and treatment plan. Of surgeons who responded to the ASES survey, 82.6% routinely order a LS radiograph to evaluate patients with NTSP. Indications included: better characterization of acromion morphology (75.4%), better appreciation of humeral head relative to glenoid (38.2%), and better evaluation of scapular morphology (28.1%). CONCLUSION: Most surgeons who responded to the ASES survey report routinely ordering a LS radiograph in the initial evaluation of NTSP. However, responses to clinical vignettes suggest that a LS view may not provide a meaningful contribution to diagnosis.

No abstract