Tendons attached to prostheses by tendon-bone block fixation: an experimental study in dogs.

To develop a method of tendon attachment to a metallic endoprosthesis, we evaluated fixation strength, clinical function of the tendon, and morphological changes in an experimental model. The canine supraspinatus tendon was removed from the greater tubercle of the humerus and attached to a titanium prosthesis. In 12 animals, the bone block underlying the tendon insertion was preserved and attached in one limb; the soft part of the tendon was attached directly to the prosthesis in the contralateral limb. Fixation strength was evaluated after 16 weeks of in vivo implantation (12 specimens) and compared with the in vitro fixation strength (12 specimens) and with intact normal controls (six specimens from cadavera). Function of the tendon in vivo was evaluated by force-plate analysis (at 3-week intervals). All specimens were evaluated histologically. Sixteen weeks after surgery, the tendon-bone block attachment was significantly stronger (mean, 16%) than the direct tendon attachment and not significantly different from the normal control, and the direct tendon attachment was significantly weaker (mean, 68%) than the normal control. There was significantly more weight-bearing on the limbs with a tendon-bone block attachment than on the limbs with a direct tendon attachment at both 3 and 6 weeks postoperatively. Both front legs showed increased weight-bearing with time, but the differences were not statistically significant. Anchorage by tissue ingrowth to the titanium prosthesis was found consistently--there was bone ingrowth in the tendon-bone block attachments and fibrous tissue ingrowth in the direct tendon attachments. When a bone block was preserved, the strength and stiffness were comparable with those of a normal tendon insertion.(ABSTRACT TRUNCATED AT 250 WORDS)