In cases of chronic acromioclavicular joint separation, the biomechanical properties of anatomical reconstructions are closer to the native configuration than the Weaver–Dunn procedure. Consequently, the radiological and clinical outcomes are better. However, an additional incision is needed to harvest the graft, which increases the procedure's morbidity.

Triple-bundle reconstruction can be performed with the coracoacromial ligament (CAL) and the semi conjoined tendon (SCT).

Bilateral dissection was performed on the upper limb of six fresh-frozen cadavers. Measurements useful to the procedure were taken on one limb, specifically the minimum graft length needed and the available length. The surgical procedure was performed on the other limb. The proximally based SCT was passed through the base of the coracoid process, then divided into two strips tightened from the superior aspect of the coracoid process to the clavicular insertion points of the conoid and trapezoid ligaments. The CAL was detached from the coracoid process and then secured in the medullary canal of the clavicle, after its lateral one-quarter was resected (i.e., 10mm).

The mean length of the SCT was 101.7±7.6mm (95.1–114.5) and the mean length of the CAL was 35.3±4.7mm (28.7–42.5). The SCT length needed for this reconstruction was 58±4.3mm (51.5–62) medially and 60.3±4.6mm (54.3–66.3) laterally. The procedure was feasible in all six cadavers with an average excess length of 39.9±5.7mm (32.2–47) for the conoid bundle, 37.6±5mm (31–45.1) for the trapezoid ligament and 6±2.7mm (3–9.5) for the CAL.

Triple-bundle anatomical reconstruction using the SCT and CAL is feasible. However, the strength of this construct must be evaluated biomechanically before it can be used clinically.

Not applicable – cadaver study.