Background
To determine whether treatment decision-making in acromioclavicular joint injuries is possible using only Alexander views with diagnostic reliability comparable to conventional pre-operative imaging.
Methods
Patients with acute first-time acromioclavicular joint injuries (Rockwood II-V) at 2 specialized centers (January 2019-June 2024) were retrospectively reviewed. Inclusion required complete pre-operative imaging consisting of either a panoramic view or bilateral Zanca views, plus at least one pre-operative modified Alexander stress view. A reference classification was established using coracoclavicular distance measurements, supplemented by the modified Alexander projection. Patients were stratified into a nonoperative group (Rockwood II-IIIA) and a potentially operative group (Rockwood IIIB-V). Modified circle measurements were performed for Alexander views. Three independent surgeons first assessed all cases using anteroposterior radiographs plus Alexander view and again after a 2-week washout period using only the Alexander view and recommended operative or nonoperative treatment. Inter-rater and intrarater reliability, diagnostic accuracy relative to the reference classification, and the correlation between circles values and injury severity were calculated. Receiver operating characteristic analyses were conducted to determine cutoff values for distinguishing higher-grade injuries.
Results
54 patients met inclusion criteria. Using complete imaging, raters demonstrated a sensitivity of 95.1% and specificity of 90.1% for treatment allocation with substantial inter-rater agreement (κ = 0.653). Using only Alexander views, diagnostic performance remained high (sensitivity 87.7%, specificity 90.1%) with substantial inter-rater (κ = 0.753) and substantial to almost perfect intrarater reliability (κ = 0.779-0.852). Modified circles values correlated strongly with injury severity (τ = 0.753, P < .001). An 18-mm threshold differentiated Rockwood IIIB from V injuries with high sensitivity (92.9%) and negative predictive value (97.2%).
Conclusion
The Alexander view alone enabled accurate identification of high-grade instability as defined by Rockwood's classification and supported treatment decisions comparable to standard anteroposterior imaging. These findings suggest that routine panorama or Zanca views may not be required for decision-making in acute AC joint injuries.