What Do You Do When You Have a Loose Screw?

Purpose: This study seeks to compare the pullout strength of various anchor configurations in an osteoporotic bone model. We have tested and present here a technique designed to augment the pullout resistance of an anchor in poor-quality bone with the use of a second anchor as an interference fit; this report describes our in vivo results with this procedure. Methods: Four groups of suture anchor constructs were tested. These included a single 5.0-mm Bio-Corkscrew (Arthrex, Naples, FL) (group I), a single 5.5-mm Bio-Corkscrew FT (fully threaded; Arthrex, Naples, FL) (group II), a single 6.5-mm Bio-Corkscrew (Arthrex, Naples, FL) (group III), and an interference fit of two 5.0-mm Bio-Corkscrew suture anchors (group IV). Anchors were secured in a 10-lb/ft3 polyurethane foam block to simulate osteoporotic bone. Each construct was cycled, then was pulled to failure with an Instron testing device (Instron, Canton, MA); measurements regarding cyclic displacement, yield load, and extension at yield load were recorded. During the in vivo portion of the study, the interference fit technique was performed in 18 shoulder arthroscopy cases in which a loose screw was a matter of concern. After the technique was performed, both anchors were pulled so their security could be assessed; cuff repair then proceeded normally. Results: Biomechanical study: In terms of yield load, every anchor construct was significantly different from every other construct. Specifically, pullout strength increased significantly as follows: group I was the weakest against pullout (176 ± 13 N), group III (223 ± 17 N) was significantly stronger than group I, group II (247 ± 12 N) was significantly stronger than group III, and, finally, group IV (305 ± 16 N) was significantly stronger than group II. The only statistically significant difference in terms of cyclic displacement was that group IV (1.4 mm ± 0.2) had significantly less displacement than group III (1.9 mm ± 0.3). No significant differences in extension at yield load were observed among any of the groups. In vivo study: The interference anchor technique was used in 18 of 24 loose screw situations over a 6-month period. In all 18 of these cases (100%), a stable dual-anchor construct was achieved. All anchors were stable to the tug test, and none failed during knot tying or at any time during the procedure. Conclusions: From the perspective of strength against pullout, the strongest suture construct of those tested in the osteoporotic bone model was the dual-anchor-against-an-anchor interference fit construct. The next strongest anchor tested was the 5.5-mm Bio-Corkscrew FT, followed by the 6.5-mm Bio-Corkscrew, and, finally, the 5.0-mm Bio-Corkscrew. Each group was statistically different from every other group in terms of pullout strength. The interference fit construct was not only the strongest in vitro, but it performed well in the in vivo setting, offering the added benefit of additional sutures to be used for securing a cuff defect. This study gives the arthroscopic surgeon important data for use in planning what to do when a loose screw is encountered. Clinical Relevance: Data from this study may be useful for the arthroscopic surgeon in choosing the proper anchor construct for osteoporotic bone. This study also lends support to the technique of press-fitting an anchor against an anchor in the loose screw situation.