Biomechanical Comparison of Reconstruction Techniques for Disruption of the Acromioclavicular and Coracoclavicular Ligaments

Albert W. Pearsall IV, MD, J. Marcus Hollis, PhD, George V. Russell, Jr., MD, David A. Stokes, MD


J South Orthop Assoc. 2002;11(1) 

In This Article

Materials and Methods

Fourteen fresh frozen cadaveric human shoulders were used for testing. The average age of the specimens was 65 years. Each specimen consisted of a clavicle, the acromioclavicular joint, the coracoclavicular and coracoacromial ligaments, and the scapula. All specimens were free of significant acromioclavicular joint arthritis. All specimens were disarticulated at the sternoclavicular joint, leaving the entire clavicle attached to the shoulder joint for testing. The skin and subcutaneous tissues were dissected free. All muscles and tendons were removed after the humerus was disarticulated from the glenohumeral joint. Care was taken to preserve the coracoacromial, coracoclavicular, and acromioclavicular ligaments. All specimens were inspected for evidence of ligamentous or articular injury. All specimens were free of disease in the acromioclavicular supporting ligaments.

After each specimen was dissected, the scapula was embedded in an aluminum box secured by low melting point metal. The sternal end of the clavicle was embedded in an aluminum tube and secured with a low melting point metal. Each specimen was frozen at -20°C until the day of testing.

After thawing on the day of testing, each shoulder specimen had the scapula and clavicle fitted in a 6 degree-of-freedom testing apparatus (Fig 1). A linear potentiometer was attached between the clavicle and the coracoid to measure coracoclavicular displacement (Bourns Linear Potentiometer, Bourns Co, Riverside, Calif). The testing apparatus permitted any combination of load and torque to be applied. The 6 degrees of motion were used to align the specimen in the proper orientation for testing. During specimen loading, the joint was pulled linearly, using only 1 degree of freedom. All measurements were recorded in millimeters in the coronal plane.

Joint testing machine with 6 degrees of freedom and force feedback control.

After each specimen was secured in the testing apparatus, the specimen was displaced in line with the coracoclavicular ligaments at a rate of 1 newton per second. It was noted in preliminary testing that because of the age of the specimens, excessive displacement force frequently led to clavicular fracture before failure of the surgical construct. Therefore, displacement was continued only until a force of 50 N was reached. The load displacement curve was recorded in conjunction with the amount of displacement between the clavicle and coracoid. This loading was repeated 5 times, with data from the last loading used for comparison.

After specimen testing with the coracoclavicular and acromioclavicular ligaments intact, 3 surgical reconstruction procedures were tested. These procedures were designed to reapproximate the relationship between the clavicle and acromion. The following procedures were done: (1) Three groups of No. 1 PDS suture (Ethicon Inc, Summerville, NJ) were woven into a 9-strand braid. Each braid was placed around the corocoid process and through a 4 mm drill hole approximately 12.7 mm medial to the lateral end and 10 mm posterior to the anterior edge of the clavicle. The braid was tied with the AC joint reduced anatomically. The initial 2 knots of the braid were tied in bulk and the remaining 4 knots were tied individually. Coracoclavicular load-displacement testing was repeated. (2) Next, the medial one half of the coracoacromial ligament was dissected from the undersurface of the acromion. Two No. 0 Ethibond sutures were placed in the free end of the ligament and brought through two 2.0 mm drill holes in the clavicle at the insertion of the coracoclavicular (trapezoid) ligament. The coracoacromial ligament transfer was tied over a 12.7 mm diameter polypropylene button (Ethicon Inc) while the acromioclavicular joint was held anatomically reduced with the PDS suture weave and acromioclavicular ligament transfer secured. Load displacement testing was repeated. (3) Two 3.2 mm holes were drilled 10 mm from the anterior border of the acromion and clavicle. An additional two 3.2 mm drill holes were placed 3 mm posterior to the initial acromial and clavicular drill holes. Two strands of No. 5 Merselene tape (Ethicon Inc) were passed through the drill holes, crossing the AC joint in an X-shaped configuration.[14] The tape was tied in a square knot with 6 knots onto itself while the AC joint was held reduced. The medial half of the coracoacromial ligament was harvested and passed through clavicular drill holes and was tied over a polypropylene button as previously described. Load displacement testing was repeated. Stiffness was calculated for all groups.

Data were analyzed using a one-way analysis of variance (ANOVA) in addition to a Fisher's partial t test (StatView, Abacus Software, Berkley, Calif).