Biomechanical evaluation of anatomical double-bundle coracoclavicular ligament reconstruction secured with knot fixation versus screw fixation.

Methods to reconstruct the coracoclavicular ligaments anatomically have been described. No clear advantage of 1 technique has been elucidated. The authors' hypothesis was that the biomechanical properties of a modified knot fixation technique would be similar to the anatomical double-bundle technique. Sixteen matched cadaveric shoulders were used for this study, and 1 additional shoulder was used in the knot fixation group only. Shoulders were randomly assigned to the anatomical double-bundle coracoclavicular ligament reconstruction technique (n=8) or a knot fixation technique (n=9). The intact coracoclavicular ligaments were tested to failure with superior displacement at a rate of 2 mm/s. Reconstruction was performed using a semitendinosus tendon allograft, and load to failure was repeated for each construct. Ultimate failure load, stiffness, and failure mode were compared using a paired t test (P<.05). No significant difference existed in load to failure between native and reconstructed ligaments or between reconstruction techniques. Stiffness decreased significantly after reconstruction in the double-bundle group (from 32.5 to 22.5 N/mm; P=.035) and in the modified knot fixation group (from 35.5 to 21.9 N/mm; P=.043). No significant difference existed in stiffness between the 2 reconstruction groups. A significant difference (P=.003) existed between failure modes between the 2 reconstruction techniques. Although less stiff than the native ligament, either technique used to reconstruct the coracoclavicular ligament can be performed to yield a load to failure similar to the intact ligament. The majority of failures in the double-bundle group were by means of the graft slipping at the screw-tendon interface at 1 of the clavicular drill holes. The modified knot fixation technique failed the majority of the time by graft elongation.