Purpose: Retrograde nailing is emerging among methods proposed or stabilisation of femoral fractures above total knee arthroplasties (TKA).
Material and methods. Between June 1994 and may 2000, 12 fractures above TKA were treated by retrograde nailing. These fractures occurred 43 months (4–51) after implantation of the TKA in three men and women aged 74 years (43–88). The fracture was situated just above the prosthetic trochlea in ten, and distant from the implant in two. The posterior cruciate ligament was preserved in six TKA and six were posterior stabilised prostheses. Indications for arthroplasty were degenerative joint disease in nine and rheumatoid polyarthritis in three. Four patients had proximal implants (one fixation and three prostheses). A percutaneous approach was used except for three cases in order protect the tibial component. Closed reduction was achieved, but required an open reduction for completion in two cases. the nail was advanced just to the trochlea in patients with a preserved posterior cruciate ligament and beyond the posterior stabilisation cage for the posterior stabilised implants. The knee was mobilised immediately after surgery and total weight-bearing was encouraged four to six weeks later.
Results: There was one error in the proximal aiming, one metastatic infection from a leg ulcer at three months and one tibial loosening in a polyarthritic woman 66 months after arthroplasty, i.e. 51 months after the fracture. Bone healing was achieved at two to four months. Frontal deviation was less than 5°. Recurvatum was less than 5° in eight cases, between 5° and 10° in two cases and between 10° and 20° in two others. At mean follow-up of 23 months (3–60), maximal moss of mobility was 10°. There was not worsening of pain.
Discussion: Retrograde nailing leads to bone healing with satisfactory frontal alignment and minimal loss of mobility. The approach uses the initial incision, facilitating complementary procedures or revision if needed. The main problem is controlling recurvatum, even though at the follow-up reported there was no clinical impact or loosening. The limitations of this method are well defined: free medullary canal, sufficient knee flexion, compatible femoral component. The tibial obstacle in posterior stabilised implants is less well known; It should be protected during the nailing if it is high. The polyethylene insert may have to be removed temporarily in certain cases.
Conclusion: The two principal problems with retrograde nailing are recognising implants compatible with this technique and controlling recurvatum. Results are acceptable with a minimally invasive technique.
