Introduction: This study compared the accuracy of reduction of intra-medullary nailed femoral shaft fractures, comparing conventional and computer navigation techniques.

Methods: Twenty femoral shaft fractures were created in human cadavers, with segmental defects ranging from 9–53mm in length. All fractures were fixed with antegrade 9mm diameter femoral nails on a radiolucent operating table. Five fractures (control) were fixed with conventional techniques. Fifteen fractures (study) were fixed with computer navigation, using fluoroscopic images of the normal femur to correct for length and rotation. The surgeon was blinded to defect size. Two landmark protocols were used in the study group referencing the piriform fossa (Group A, n=10) or proximal shaft axis (Group B, n=5). Postoperative CT scans, blindly reported by a musculoskeletal radiologist, were used to compare femoral length and rotation with the normal leg.

Results were analysed using ANOVA with 95% Confidence Intervals.

Results: The control and study groups were not statistically different with respect to age of cadaver or size of femoral defect. Results: The mean leg length discrepancy in the study groups were significantly less (3.6mm (95% CI 1.072 – 6.128) and 4.2mm (95% CI 0.63–7.75), compared with 9.8mm (95% CI 6.225 – 13.37) in the control group (p< 0.023). The mean torsional deformities in the study groups were 8.7 degrees (95% CI 4.282 – 13.12) and 5.6 degrees (95% CI -0.65 – 11.85), compared with 9 degrees (95% CI 2.752 – 15.25) in the control group (p=0.650). Within the navigated study group, length discrepancy was similar in subgroups 1 (3.6mm) and 2 (4.2mm). Torsion appeared more accurate in group 2 (5.6 degrees) than group 1 (8.7 degrees), although this was not statistically significant.

Discussion: Computer navigation significantly improves the accuracy of femoral shaft fracture fixation with regard to length. With further modifications to improve reduction of rotational deformity, it may be a useful technique in the treatment of femoral fractures.

This study compared the accuracy of reduction of intra-medullary nailed femoral shaft fractures, comparing conventional and computer navigation techniques.

Twenty femoral shaft fractures were created in human cadavers, with segmental defects ranging from 9–53mm in length. All fractures were fixed with antegrade 9mm diameter femoral nails on a radiolucent operating table. Five fractures (control) were fixed with conventional techniques. Fifteen fractures (study) were fixed with computer navigation, using fluoroscopic images of the normal femur to correct for length and rotation. The surgeon was blinded to defect size. Two landmark protocols were used in the study group referencing the piriform fossa (n=10) or proximal shaft axis (n=5). Postoperative CT scans, blindly reported by a musculoskeletal radiologist, were used to compare femoral length and rotation with the normal leg. Results were analysed using the Wilcoxon two-sample test.

The mean leg length discrepancy in the study group was 3.8mm (range 1–9), compared with 9.8mm(range 0–17) in the control group (p=0.076). The mean torsional deformity in the study group was 7.7 degrees (range 20–2) compared with 9 degrees (range 0–22) in the control group (p=0.86). Within the navigated study group, length discrepancy was similar in subgroups A (3.6mm) and B (4.2mm). Torsion appeared more accurate in group B (5.6 degrees) than group A (8.7 degrees), although not significantly.

Computer navigation appears to improve leg length discrepancy following femoral nailing. Technique modification during the study improved rotational accuracy, and with further improvement, will make this technique applicable to femoral fracture fixation.