Failure was defined as an increase of less than 20 points in the modified HSS knee score at the time of the review or the need for an additional operation related to the allograft. Thirteen knees were deemed to be failures giving a 75% success rate. Graft resorption occurred in five patients resulting in implant loosening. Four failed due to infection and non-union between the host bone and allograft was present in two. One patient with both knees grafted failed to gain a 20-point improvement. Survival analysis showed a 72% survival at 10 years. Clinically, the modified HSS score improved from a mean of 32.5 pre-operatively to 75.6 at the time of the review. Radiographic analysis of the surviving grafts showed no severe resorption, one moderate and two mild cases of resorption. Evaluation for loosening revealed one patient with a loose tibial component, while three patients had non-progressive tibial radiolucent lines. All four patients were asymptomatic.
The hips were approached posteriorly. A step cut was used to secure the host to allograft junction. The femoral component was cemented within the allograft and with a press-fit in the host bone. All but three cases had iliac crest bone graft and/or residual host bone chips added to the host-allograft site. The acetabulum was revised concurrently in 13 (two whole acetabular allografts).