Intramedullary nailing has sufficient durability for metastatic femoral fractures

The patient background is summarized in Table 1. Eighty IMN procedures were performed in 75 patients, 37 males and 38 females. Five patients underwent bilateral IMN in separate procedures at least 8 days apart. Seventy-four cases underwent radiotherapy after IMN procedure; however, six out of 80 cases were unable to undergo radiotherapy due to postoperative poor medical problems (Table 1).

The 2- and 3-year postoperative survivals were 14.2 and 8.4 %, respectively (Fig. 2a). In contrast, the implant survival rate was 94.0 % at both 2 and 3 years; however, it dropped to 62.8 % at 50 months (Fig. 2b).

In this study, three intramedullary nails broke through their proximal parts (Table 2). For all three patients, the fracture site was subtrochanteric and the implant failures showed a similar pattern of breakage at the proximal part of the IM rod. The IM rods were removed in all three cases, and the proximal part of the femurs were resected and then replaced with EPRs. The three patients did not require any further operations until death. Figure 3 shows the clinical course of case 2 patient who was a 49-year female and had a 5-year history of breast cancer, who presented with severe left femoral pain, and the patient Mirels’ score was 10. The patient underwent IMN procedure for impending fracture. Fifty months after the first surgery, she underwent EPR as a revision surgery due to implant breakage. The patient did not require any further operations until she died after 1.5 months. Table 3 shows three early postoperative deaths on postoperative days 3, 7, and 12. Two of the three patients (cases 1 and 3) underwent palliative therapy before the surgery, but complained of severe femoral pain with Mirels’ score of 10. One patient (case 2) was transferred to the emergency department due to pathological fracture. All the patients had several comorbidities including lung, liver, adrenal, brain, and multiple bone metastases, and the cause of death in all three cases were respiratory failure (Table 3).

The goal of surgical treatment of femoral metastatic fractures is not only to internally fix or prevent pathological fracture but also to reduce pain and optimize recovery, mobility, or care for the patient with minimal invasiveness and complications [3‐5, 14]. The strategy is to ensure the durability of treatment, and in this regard, implant survival should exceed patient survival after surgery.

Our study has several potential limitations. Firstly, this study is a single-center retrospective study with all the limitations inherent to such design. Secondly, there were some metastatic femoral fracture cases that were not treated according the indication pathway because of surgeon and patient preferences. Thirdly, patients with metastases underwent not only surgery but also adjuvant therapies, such as chemotherapy, radiotherapy, and hormonal therapy. Postoperative survival may be affected by those therapies. However, in this study, there were 186 consecutive procedures for metastatic femoral fractures, and 80 IMN cases, suggesting a degree of external validity and robustness.

Surgical treatment strategy towards femoral metastases still remains unclear. Both IMN and EPR for femoral metastases are widely performed. IMN has some advantages over EPR, such as lower cost, less blood loss, less muscle wasting, shorter operation time, and shorter hospitalization. On the other hand, some studies have reported that EPR is associated with higher patient survival [9], lower mechanical failure rate, and more durability than IMN [10, 11]. However, those studies did not consider the differences in patient demographics, and thus comparing outcomes between IMN and EPR in these reports may not be appropriate.

Treatment indications vary between institutions. Table 4 shows peer-reviewed articles from 2008 through 2013 and our current study, describing IMN, EPR, plating and cementation, and other surgical procedures in femoral metastases [7‐13]. These studies can be divided into two groups. Among four studies, the proportion of IMN/EPR is approximately 1:2 [7, 10‐12]. Nilsson at al. reported that they did not perform IMN for trochanteric or subtrochanteric metastases [7]. Steensma et al. and Alvi at el. avoided IMN for the patients with intertrochanteric lesions of the femur [11, 12]. In the other group, the proportions of IMN and EPR were nearly equal [8, 9, 13]. Sarahrudi et al. reported that their IMN group included inter- and subtrochanteric metastases [8], and Mavrogenis et al. reported also performing IMN for fractures in the intertrochanteric part of the femur except for metastases invading the articular surface [9]. These reports included a broader indication for IMN. Regarding the superiority of EPR to IMN and vice versa, while Mavrogenis at al. reported a significantly higher survival in patients with EPR [9], Sarahrudi et al. reported that EPR and IMN were equivalent in terms of safety [8]. A few other articles have also mentioned good outcomes with IMN [15, 16].

As far as we know, there is no strong evidence concerning the indication of IMN for subtrochanteric metastases. The evidence grade of EPR for femoral neck fracture is grade B which indicates consistent, fair (level II or III) evidence, while IMN is grade C which indicates conflicting or poor-quality (level IV or V) evidence, including inter- and subtrochanteric fractures, except for intertrochanteric impending fracture whose grade is B [14]. In our institution, the postoperative survival was 14.2 and 8.4 % at 2 and 3 years, respectively, and the implant survival rate was 94.0 % at both 2 and 3 years. Harvey et al. noted that the IMN implant survival rate was 85 % at 2 years [10], and Steensma et al. noted that the IMN implant survival rate was 88 % at 3 years [11]. Compared to these reports, our IMN implant survival rate of 94 % at 3 years is comparable and can be regarded as appropriate for a patient population whose survival at 3 years is only 8.4 %.

There were three patients who underwent revisions due to implant breakage. In all such cases, the site was in the subtrochanteric part of the femur, and failure occurred in the proximal part of the nail. EPR was chosen as the revision procedure in order to prevent further complications after revision. Forsberg et al. recommended the use of EPR as a salvage procedure even at the end of life [3], and we concur with their recommendation.

In our study, three cases had early fatal complications after surgery (3.8 %), all of which had lung metastases at the time of IMN procedure. The role of IMN should be carefully considered in patients who have pre-existing pulmonary dysfunction. Moon et al. reported that prophylactic IMN did not appear to be safer than curative IMN for femoral fractures [18]. Barwood et al. reported that acute oxygen desaturation and hypotension occurred in 24.4 % of patients during IMN procedures for metastatic femoral fractures and 6.6 % of patients died from cardiorespiratory dysfunctions during the perioperative period [19].

Cost-effective treatment for bone metastases is important because of the already high cost of treating malignancy [20]. Schulman et al. noted that total medical cost for patients with bone metastases was significantly higher than that for patients without bone metastases [21]. A cost-effectiveness analysis between IMN and EPR should be studied further.