Cylindric titanium rods with different surfaces were axially implanted into the femora of sheep. The three surfaces were grit-blasted titanium, plasma-sprayed titanium and plasma-sprayed hydroxyapatite (HA). After 2 months, a 2-cm segment of the femoral shaft was completely resected to load the implant, and the animals were allowed full weight-bearing for 9 months. Biomechanical and histological evaluation of the implants was undertaken 2 months after implantation and 9 months after the segmental resection. The mechanical testings of well-fixed implants were performed 9 months after segmental resection. Loosening of 45% of the titanium-coated implants was observed in the first 3 weeks, but thereafter, no further loosening occurred. The HA-coated implants remained entirely fixed for 3 weeks, but thereafter, a progressively increasing incidence of loosening up to 55% after 9 months of loading was detected as subsidence on X-radiographs. The maximum push-out strength of the titanium-coated implants was 4.9 MPa compared with 2.3 MPa for HA-coated ones. No significant mechanical interlock between the grit-blasted surface and bone was observed. The HA coating was found to be delaminated in all unstable implants, whereas the titanium coating remained completely intact. Morphometric analyses of well-fixed rods showed complete bony ingrowth onto the HA surface, whereas the contact area between the bone and the two titanium surfaces was less than 40%. Concerning clinical significance bony ingrowth with long-term mechanical interlock between the implant surface and the bone cannot be achieved by grit-blasting or HA-coating. The titanium plasma-coating, however, can induce a bone-implant interface which resists the mechanical stress resulting from continuous cyclic loading in vivo.
