Routine plain radiographs in the setting of atraumatic shoulder pain: Are they useful?

Introduction: Plain radiographs of the shoulder are routinely obtained for patients presenting with atraumatic shoulder pain. The diagnostic utility of these radiographs is debatable. Methods: Patients presenting for the first time to a shoulder clinic with atraumatic shoulder pain received a plain radiographic series. After considering their medical history and physical examination, we made a preliminary diagnosis and formulated a tentative treatment plan. We then reviewed the radiographs and determined whether the addition of these radiographs altered the diagnosis and/or treatment. Results: A total of 343 patients met the inclusion criteria. Age .50 years (odds ratio, 3.3; 95% confidence interval, 1.9 to 5.8; P, 0.00001) and weakness in forward elevation (odds ratio, 2.9; 95% confidence interval, 1.6 to 5.4; P = 0.0003) were associated with the presence of radiographic findings. Plain radiographs altered the proposed diagnosis in only 14.9% (51/343) of cases. However, clinical management was changed in only 1.7% of cases. Plain radiographs had no effect on clinical management in 97.4% of patients younger than 50 years. Discussion: Plain radiographs rarely alter the diagnosis or affect management in the setting of atraumatic shoulder pain, particularly in patients younger than 50 years. Patients may be exposed to unnecessary radiation, and the healthcare system as a whole may be burdened with the increased cost of this imaging modality. Level of Evidence: Level II (diagnostic).