Purpose: Plain radiographic evaluation of acute shoulder trauma in adults requires a minimum of two projections, commonly the anteroposterior (AP) and lateral scapular projections, with additional projections taken for diagnosis. The aim of this retrospective study was to determine whether the addition of the apical oblique (AO) projection to the AP and lateral scapular projections increases the number and/or alters the types of abnormalities detected in the examination of acute shoulder trauma.
Methods: Examinations of 56 adults who had undergone three-projection (AP, lateral scapular, AO) radiographic shoulder examination for acute trauma were allocated into two-projection (AP, lateral scapular) and three-projection cases and assessed by a radiologist. The differences in number and types of abnormalities between the two-projection and three-projection cases were quantified using the one-tailed t test and chi-square goodness-of-fit test, respectively.
Results: Test-retest reliability was moderate (intra-class correlation coefficient [95%CI], 0.56 [0.15 to 0.80]) for number, and almost perfect (kappa [95%CI], 0.94 [0.85 to 1.00]) for types, of abnormalities detected. There was a significant increase in the number of abnormalities detected across all three-projection versus two-projection cases (difference in means [95%CI], 0.20 [0.01 to 0.39]) and for fractures (difference in means [95%CI], 0.30 [0.11 to 0.49]), but no difference in the types of abnormalities detected (χ 2 = 4.7, p = 0.19).
Conclusion: This study suggests that adding the AO projection to two-projection examination of acute shoulder trauma increases the number of abnormalities detected; this has potential implications for patient management. Further research investigating differences in types of abnormalities detected between two-projection and three-projection cases is warranted.
Keywords: Apical oblique; Emergency; Projection; Radiography; Shoulder; Trauma.