Abstract
To assess the outcome of periprosthetic femoral fractures (Vancouver B3 type) around loose stems treated by impaction grafting revision, 106 B3 fractures were reviewed. To assess the association between those who healed and those who did not for the factors of interest such as stem length, impaction grafting and the combination of the above, odd ratios along with their 95% CI and their p-values were reported. Logistic regression in STATA version 7.0 employed.
75 out of 89 fractures treated with long stem revision and 10 out of 17 with sort stem healed. 74 out of 89 fractures treated with impaction grafting and 11 out of 17 treated without impaction grafting healed.
66 out of 75 fractures treated with long stem and impaction grafting and 9 out of 14 treated with with long stem but no impaction grafting healed. 8 out of 14 fractures treated with short stem and impaction healed. Average healing 8.5 months.
Those treated with long stem are almost four times more likely to heal than those treated with short stem (odds ratio = 3.75 95%CI: 1.21–11.6 p=0.022) and those with impaction grafting are also more likely, but not statistically significant, to heal than those without impaction grafting (odds ratio = 2.69, 95%CI: 0.86– 8.45 p=0.090). Furthermore, those with long stem and impaction are significantly more likely to heal than those without impaction grafting and those with short stem and impaction grafting (odds ratios = 4.07, 95%CI: 1.10 – 15.0 p=0.035 and 5.5, 95%CI: 1.54 – 19.6 p=0.009 respectively).
Impaction grafting is an increasingly popular technique for the restoration of femoral bone stock. It can successfully be applied to periprosthetic femoral fractures but a long stem should be used to bypass the distal fracture line.
The abstracts were prepared by Mr Tim Briggs. (Editoral Secretary 2003/4) Correspondence should be addressed to him at Lane Farm, Chapel Lane, Totternhoe, Dunstable, Bedfordshire LU6 2BZ, United Kingdom
