Clinical outcomes of femoral shaft non-union: dual plating versus exchange nailing with augmentation plating

Mechanical stability improvement is the basis for treatment selection as well as the key to success in femoral shaft non-union cases [1‐4]. Nail dynamization, exchange lateral plating (EP), and exchange nailing (EN) have been generally applied in clinical practice [5‐9]. However, due to the mechanical instability of revision procedures mentioned above, the failure rate is as high as 20–50% [7, 10‐13]. Optimal mechanical stability and deformity correction can be achieved during exposure of the non-union site in the procedures of dual plating (DP) and exchange nailing with augmentation plating (EN/AP) [14‐18]. Evidence indicates that most patients who received either surgical procedure achieved fracture healing.

To avoid massive surgical trauma and complications, improve fracture healing and accelerate rehabilitation, evaluating the advantages and disadvantages of DP and EN/AP is critical. To our knowledge, no comparative study concerning both surgical procedures has been published.

Therefore, we retrospectively reviewed the medical data of patients with femoral shaft non-union who had been treated via DP or EN/AP in our trauma center for the first time. Surgical trauma, complications, fracture healing, and time to return to work were analyzed to evaluate the advantages and disadvantages of the two surgical procedures.

All patients were followed up, with a median follow-up of 2 years (range, 1–4 years). No statistically significant difference in the follow-up period (2.59 ± 1.16 years vs 2.03 ± 1.2 years, p = 0.206) was observed between the DP and EN/AP groups. Except for incision length (p = 0.000 < 0.05), no significant differences were noted in terms of the operative time and postoperative deformity between the two groups (see Table 2).

In the EN/AP group, the fracture healing rate was 100%, and the fracture healing time was 5.7 ± 1.7 months. In the DP group, the fracture healing rate was 94%, and the fracture healing time was 8.4 ± 4.1 months. The healing time and time to return to work were shorter in EN/AP group, with statistically significance (p = 0.024 in healing time, and p < 0.01 in time to return to work) (see Table 3). In the DP group, two patients did not return to work because one patient was followed up for only 1 year, while the other patient still exhibited non-union. Classic cases from the two groups are shown in Figs. 2 and 3.

In terms of complications, no significant difference was found between the two groups. In the EN/AP group, all patients with complications refused additional surgical procedures because they were basically satisfied with their function. In the DP group, one patient had a superficial soft tissue infection at 1 month postoperatively; therefore, debridement and vacuum-assisted closure were performed. After the infection was controlled, wound closure was carried out again. Additionally, intravenous antibiotics were administered for 3 weeks based on drug sensitivity testing. One patient showed no fracture healing 14 months after the operation and refused to undergo bone grafting again; this patient has been followed up for 2 years, and the fracture still has not healed. One patient received implant removal surgery at a local hospital 2 years postoperatively, and refracture occurred 3 months later. Subsequently, the patient underwent plating fixation and bone grafting in our hospital again, and the fracture healed at 12 months after this operation. Patients exhibiting knee stiffness and malunion did not receive additional surgical treatment (see Table 3).

The primary objective of femoral shaft non-union treatment is improved mechanical stability. The anti-rotation stability of previous nailing is reduced after dynamization procedures, leading to an unacceptable incidence of failure up to 40–60% [12, 13]. The intractability of rotational and bending instability remains unresolved, even when a larger nail is used. Up to 20~50% of patients who underwent EN did not achieve fracture healing [10, 20]. However, EN/AP is not only minimally invasive and can achieve axial mechanical stability via nailing but can also solve the problem of insufficient anti-rotation and anti-angulation stability with nailing (especially non-isthmus fractures). On the other hand, compared to EP, DP can solve the problem of eccentric fixation, build a more mechanically stable construct, and stabilize the autologous bone grafts around the non-union site [4, 21]. Therefore, DP and EN/AP, as three-dimensional fixation constructs that can provide sufficient multiplanar stability, are suitable for patients with femoral shaft non-union.

Publications show that the healing rate for femoral shaft non-union is as high as 90% [14‐18]. In our study, 29 patients (96%) finally achieved fracture healing, which is similar to previous reports. The fracture healing time with DP has been reported to be 5.2–6 months [14‐16], which is dramatically shorter than that with EN/AP (6–13 months) [17, 18] in previous reports. However, different results were observed in our study for the both healing time and the time to return to work, which were significantly shorter with EN/AP versus DP. Multiple factors are considered to be related to these results. First, excessive periosteal dissection is the most important reason for the prolonged fracture healing in the DP group [14]. Second, for patients in the EN/AP group, intramedullary debris can infiltrate the non-union site to achieve “autologous bone grafting” by reaming [8, 9]. Meanwhile, reaming can also increase the periosteal blood supply and stimulate callus formation [8, 9]. Finally, the fracture may heal faster due to less periosteal soft tissue detachment and better restoration of the endosteal blood supply [22].

Deformity correction is the main surgical technical difficulty in femoral non-union cases. Plating has been the main method of deformity correction for a long time [4, 14, 23]. With direct exposure of the non-union site, debridement, surface contact, deformity correction, and rigid fixation can be achieved simultaneously [23]. However, IMN has obvious shortcomings in the correction of rotation and angle deformity. Meanwhile, an augmentation plate with nailing in situ cannot correct deformities [24, 25]. Therefore, some surgeons have proposed the application of exchange nailing with augmentation plating (EN/AP) for the treatment of femoral shaft non-union with severe deformities. Wang et al. [17] described 12 patients (9 cases of shortening deformity and 3 cases of rotational deformity) with femoral shaft non-union who underwent the EN/AP procedure and achieved effective correction. Yang et al. [18] reported the application of EN/AP in two patients with severe deformities and obtained good results. In our study, 11 patients had severe deformities and underwent EN/AP. The EN/AP procedure achieved excellent deformity correction through limited exposure of the non-union site, which showed similarly good clinical results with those of DP, as reflected by the shorter incision length.

However, this study was a retrospective study with a relatively small sample. The advantages and disadvantages of both procedures require further investigation with a larger sample in well-designed clinical studies. Furthermore, some patients did not attend the scheduled appointment for the follow-up and radiological examinations, so the data might not truly reflect the healing time. More rigorous follow-up and more convenient diagnostic techniques are needed to improve the data accuracy. Finally, the choice of surgical techniques depends on the patient’s conditions and surgeon’s preference, the bias in treatment selection also is our trial limitation.

Overall, both EN/AP and DP can provide excellent mechanical stability to the non-union site and achieve effective deformity correction. Compared to the DP group, the EN/AP group showed a shorter incision length, faster fracture healing, and a shorter time to return to work. However, no significant differences were observed in terms of surgical trauma and complications in our study.