C.A. Roth
M.H. Santare, L. Snyder-Mackler, M.G. Ciccotti, A.R. Bartolozzi
Department of Mechanical Engineering
University of Delaware
Newark, Delaware 19176
INTRODUCTION: Suture anchors in anterior shoulder repairs have been introduced during the past several years to simplify classical standard Bankart-type or capsule shift procedures. The implantation of two or more anchors in the middle and inferior anterior glenoid has been suggested to secure the reattachment of the soft tissue to bone. The ultimate pullout strengths of various anchors have been reported recently [1], and results indicate that this property might vary for each quadrant of the anterior glenoid [2]. During postoperative rehabilitation, however, the suture anchor construct is subjected to repeated sub-maximal loads. This loading should, in theory, reduce the fixation strength of the anchors and could eventually lead to anchor failure. There is little information available to date on fatigue properties of suture anchors. The purpose of this study is to investigate the ultimate pullout strength and fatigue properties of two commercially available anchors, the Mitek G2 (Mitek Surgical Products, Norwood, MA) and the Statak 3.5 (Zimmer, Warsaw, IN), between the 2:00 and 5:30 locations on the anterior glenoid.
MATERIALS AND METHODS: Forty-four cadaveric glenoids with an average age of 68 years were harvested 3 cm from the joint surface and embedded in an epoxy resin (Epo-Kwick, Buehler, Lake Bluff, IL) with the anterior half exposed. The anchors were threaded with steel wire and up to three anchors were implanted in each glenoid at various locations between 2:00 and 5:30. Manufacturer s specifications for insertion were followed. The specimens were mounted on an Instron servo-hydraulic fatigue testing machine (Model 1331, Instron Corp., Canton, MA) with the longitudinal axis of the anchor aligned with the actuator. The anchors were cycled at 2 Hertz sinusoidal loading between preselected minimum and maximum loads. The stress ratio was kept constant at 0.15 for all tests. The maximum load was varied among samples and locations to obtain a number of cycles versus maximum load relationship for each location. All of the specimens were tested to failure, the total loss of tension in the system. If an anchor pulled out during the first cycle before reaching the preselected maximum load, the load at pullout was recorded as the ultimate pullout strength. After failure occurred, the cortical thickness at each implantation site was measured using a caliper. A total of 50 Mitek G2 and 57 Statak 3.5 anchors were tested and the results were analyzed using Student s t-test and least square regression analysis procedures.
RESULTS: The cortical thickness of the anterior glenoid at the implantation sites decreased steadily from 1.3 mm at 2:00 to 0.7 mm at 5:30 with a standard deviation at each location of less than 20%. All anchors pulled out, no wires broke. Sixteen Mitek G2 and twenty-one Statak 3.5 anchors pulled out prior to reaching the preselected maximum load. Using a linear regression analysis, a model was developed for each anchor predicting the ultimate pullout strength as a function of cortical thickness. There is a significant (P<0.05) difference in ultimate pullout strength between the two anchors and each anchor s strength varies significantly (P<0.05) with decrease in cortical thickness. The Mitek G2 failed at 241N when implanted through a 1.3 mm thick cortex and at 152N when implanted through a 0.7 mm thick cortex, whereas the Statak 3.5 failed at 128N and 59N, respectively. During cycling at sub-maximal loads, all anchors cut through the soft cancellous bone and came to rest against the inner cortex during the initial 10 to 100 cycles. After settling, the anchors were exposed between 1 and 2 mm, depending on the cortical thickness at the implantation site. The fatigue life of each anchor was greatly dependent on the magnitude of the load applied and the location of implantation. To reduce the influence of cortical thickness on the prediction of the fatigue life of an anchor, the theoretical ultimate pullout strength for every anchor at its implantation site was estimated from the aforementioned ultimate pullout strength versus cortical thickness model. The actual applied maximum load was then normalized by the theoretical ultimate pullout strength, and plotted versus the number of cycles to failure. For the Mitek G2, at each location, the maximum load which can be applied to guarantee a life of 1000 cycles is less than 50% of the ultimate pullout strength at this location. A similar behavior for the Statak 3.5 can be expected, but could not be proven with statistical significance.
DISCUSSION: The Mitek G2 and Statak 3.5 currently represent the most commonly used metallic suture anchors. Earlier studies [1] have determined that, for a given size of implantation hole, metallic anchors generally have a higher pullout strength than nonmetallic anchors. Since the resistance against pullout is also important for fatigue, nonmetallic anchors would theoretically show a similar decrease in fixation strength when cyclically loaded as metallic anchors.
The decrease in cortical thickness from superior to inferior locations on the anterior rim of glenoid has a significant influence on the ultimate pullout strength and fatigue life for both anchors tested. Interiorly placed anchors perform generally worse than superiorly placed anchors. This result emphasizes the importance of correct placement of the inferior anchor during anterior shoulder refixations. Furthermore, since abduction and external rotation of the arm put a high load on the inferior glenohumeral ligament, the interiorly placed anchor will be very vulnerable during this humeral rotation.
Usually, number 2 suture is used in combination with the suture anchors. The failure strength of this suture is below the ultimate pullout strength of both the Mitek G2 anchor and the superiorly placed Statak 3.5 anchor, which minimizes the possibility of anchor pullout. During sub-maximal cyclical loading, however, the pullout strength of both anchors decreases significantly, and failure could be shifted from suture breakage to anchor pullout.
The initial settling of all anchors suggests that clinically, after only a few sessions of rehabilitation, the reconstruction would loosen, resulting in a reduced soft tissue to bone contact and reduced tension of the capsule. Hence, the authors suggest manually precycling each anchor up to ten times immediately after implantation before the soft tissue is reattached.
CONCLUSION: Cortical thickness directly affects the failure properties of suture anchors. The more interiorly an anchor is placed, the lower its ultimate pullout strength and the less the load which can be applied to guarantee fixation during the period of rehabilitation. The settling of the anchors suggests a weakening of the soft tissue reattachment after only a few cycles. The degree of settling is determined by the initial depth of implantation and the cortical thickness. Fatigue properties of suture anchors must be considered when selecting an anchor and determining the minimum number of anchors required to guarantee a successful shoulder reconstruction.
(1) Barber et al.: The ultimate pullout strength of suture anchors. Arthroscopy 11: 21-28, 1995.
[Return to Table of Contents, Back to Research Symposium Home Page, Orthopaedic Department Home Page]