Session III - Femur/Knee

*A Prospective Comparison of Retrograde and Antegrade Femoral Intramedullary Nailing

David M. Huebner, MD; Robert F. Ostrum, MD; Brian L. Davison, MD; Attila Poka, MD, Grant Medical Center, Columbus, OH (all authors ­ a-Biomet)

Purpose: We prospectively compared the results of antegrade and retrograde reamed femoral intramedullary nailing with respect to fracture union, knee and hip function, operative time, blood loss, and secondary procedures required.

Methods: All 238 femoral fractures treated with an intramedullary nail between February 1, 1998 and December 31, 2000 at a Level I trauma center were randomized by medical record number to receive either an antegrade (Group A) or retrograde (Group R) implant. The femurs were reamed until cortical chatter was encountered or to a maximum of 13.5 mm. We selected the diameter of the implant to be 1 or 1.5 mm less than the last reamer. All nails (except one) were 10, 11, or 12 mm in diameter and were locked both proximally and distally. The following data were collected prospectively on all patients: operative time, blood loss, nail characteristics, presence and grade of open fractures, degree of comminution (Winquist), associated injuries, injury severity score, knee pain, thigh pain, hip pain, symptomatic implants, implant removal, knee range of motion at last follow-up, date of fracture union, occurrence of complications, and secondary procedures required to achieve union.

Results: One hundred ten fractures were treated with antegrade nailing and 128 were treated with retrograde nailing. We were unable to follow up 18 antegrade and 25 retrograde nail patients due to patient demise or failure to complete follow-up, leaving 92 patients in Group A and 103 patients in Group R. The average operative time was 72.5 minutes for Group A and 65.8 minutes for Group R. The average blood loss (closed fractures) was 234.0 ml for Group A and 147.9 ml for Group R. Statistical analysis revealed a significant difference between the two groups in the amount of blood loss (P = 0.00001) and operative time (P = 0.0363). Union rates after the index procedure were 94.6% in Group A and 95.1% in Group R (P = 0.867). The estimated average time to union was 14.1 weeks in Group A and 13.2 weeks in Group R (P = 0.410). Four dynamizations, one bone graft, and one exchange nailing were required to achieve union in the fractures of Group A. Two dynamizations (and one pending dynamization), one bone graft, and one exchange nailing were required to achieve union in the fractures of Group R. Knee pain in patients with more than 6 weeks of follow-up and no intraarticular fractures was seen in 20.6% of Group A patients and 16.2% of Group R patients (P = 0.645). Hip pain in patients with more than 6 weeks of follow-up was 14.7% in Group A compared with 3.4% in Group R (P = 0.023). More patients from Group R had symptomatic distal locking screws (23.4% vs. 11.9%, P = 0.072) which often required removal (18.1% vs. 10.7%, P = 0.237).

Discussion: Retrograde femoral intramedullary nailing has become a useful technique in the treatment of femoral shaft fractures. The results of this series of 195 patients shows no significant difference in union rates, time to union, knee pain, or knee range of motion. Blood loss and operative time were significantly lower in the retrograde group. Hip pain was more common in the patients with antegrade nails (P = 0.023), and pain from a distal locking screw necessitating removal seemed to be more common in the retrograde nail group, although this difference was not statistically significant.

Conclusions: Both antegrade and retrograde nailing have their assets and liabilities. Although certain fractures may be better suited to one technique or the other, we have used retrograde nailing on fractures from the supracondylar to the subtrochanteric region with excellent results. Given the equal performance of these two techniques, we believe that retrograde nailing represents a viable alternative to antegrade nailing in the treatment of femoral shaft fractures.