RECONSTRUCTION OF THE CORACO-CLAVICULAR (CC) LIGAMENTS USING TENDON GRAFTS. A BIOMECHANICAL STUDY.

Lee SJ, Nicholas SJ, Akizuki KF, McHugh MP, Ben-Avi SS, Kremenic IJ

Nicholas Institute of Sports Medicine and Athletic Trauma and The Cooper Union School of Engineering, New York, NY

Shoulder separations and distal clavicle fractures are common injuries. Practically every conceivable operation has been described to treat shoulder separations and Type II distal clavicle fractures. Nevertheless, they can be grouped into two main types of treatments. The first are treatments that address the primary healing potential of the CC ligaments. Essentially, treatments such as sutures, suture anchors, screws and various acromioclavicular fixations have been developed to hold the relationship of the clavicle and the scapula reduced long enough to allow healing of the CC ligaments. However, these treatments all assume that the CC ligaments will heal not only at their pre-rupture strength, but also their pre-rupture length, neither of which has been proven, and neither of which is likely. The second main type are reconstructions of the CC ligaments. The most popular types include CA ligament transfers as popularized by Weaver and Dunn, and Conjoin tendon transfers as described by Cadenat in 1917. The most consistent problem that has been found however with these reconstructions is the weak fixation to the clavicle. The purpose of this study was to determine the structural properties of the normal, intact CC ligaments as well as coracoacromial ligament transfers; more importantly, to develop a better reconstruction technique using various tendon grafts, and whether this would result in an accelerated rehabilitation program, and ultimately, better outcomes. Ten fresh frozen human shoulder cadavers were dissected, leaving the CC and CA ligaments intact, while sectioning the AC ligaments. A custom jig and modified potting technique was devised to fix the clavicle. Unidirectional tensile loading to failure was performed on the normal, intact native CC ligaments on the MTS machine, at a rate of 25 mm/min. A CA ligament transfer was then performed, and this was then loaded to failure on the MTS.

Next, 10 fresh frozen whole legs were obtained. The following tendons were harvested: long toe extensors, gracilis, and semitendinosis. The reconstruction was performed by using the upper portion of the deltopectoral approach, drilling a 3.2mm hole in the anterior 1/3 of the clavicle, threading the graft through the hole and around the coracoid. This graft was then secured by tying it in a knot and then adding supplemental sutures. Several different fixation techniques were attempted, including a Pulver-Taft type weave, however, the only technique that did not fail at the fixation site was tying the tendon into a square knot and supplementing it with sutures. Frequently there was enough graft left over to make a second loop. Each tendon graft reconstruction was then loaded to failure on the MTS. The parameters studied included the ultimate strength, maximum strength within 2 cm (the average amount measured to approximate a grade 3 separation), stiffness, elongation at failure, and the mode of failure.

The results are as follows: the maximal forces within 2 cm for the native ligament averaged 641 N, while the CA ligament transfer had a peak load of only 70 N, and the Tendon grafts had 512, 493, and 618 N. Of note, all grafts were significantly stronger than the CA ligament transfer, and more importantly, were not significantly different from the native ligament. Similar results were found for the ultimate peak load. The tendon grafts were significantly stiffer than the CA ligament transfer, but also not as stiff as the native ligament. The native ligament predictably failed midsubstance, the CA ligament transfer failed at the sutures, and all tendon grafts failed through the midsubstance of the tendon, not at the fixation site. The advantages of a reconstruction are that it is a permanent biologic reconstruction, not some temporary synthetic material, or some hardware that must be removed at a later date. It requires a minimal surgical approach. The fixation is not the weak link because the tendon graft is long enough that it can be tied into a knot. And because the fixation is not the issue, supplemental fixation is not necessary. The strength and stiffness is much stronger than CA ligament transfer, and in fact, just as strong as the native CC ligament. Plus, because it does not affect the torn, native CC ligament, and does hold the clavicle reduced to the scapula, this native CC ligament should heal just as it would if you were to use a traditional suture or screw technique. Thus, the ultimate strength of the tendon graft reconstruction plus the healed native CC ligament may be stronger than the original, uninjured CC ligament. The significance of this is that the immobilization time could be significantly reduced, and that the rehabilitation may be significantly accelerated. This may lead to an earlier return to normal activities and sports. The strength of this reconstruction in addition to that of the healed native CC ligament may contribute to lower failure and recurrence rates, which is especially important considering that these procedures tend to be performed more for those patients who may be involved in heavy labor, or contact sports.

American Orthopaedic Society for Sports Medicine: Specialty Day. San Francisco, CA, March 3, 2001.