Introduction
In the treatment of trochanteric fractures (TF), stable fixation is mandatory to achieve a safe and early mobilization, because it is important for patients to return to their previous activity level (1,2). Although choice of the implant varies according to the type of TFs, many studies report that unstable pertrochanteric fractures of the femur, i.e. 31-A2 without medial support, and 31-A3, i.e. intertrochanteric fractures, can be treated successfully with intramedullary (IM) implants. IM fixation devices have become increasingly popular due to biomechanical advantages in the treatment of unstable TFs compared with extramedullary fixation (3-5).
TFs are common in the elderly population and its incidence increases twice in every decade after the age of 50 (6). Although increase in the anterior femoral bow by advanced age has been proved in the literature, the proximal femoral nails (PFN) in the market have still straight designs on the sagittal plane (7,8). Non-anatomic shapes of the PFNs sometimes make the surgeries difficult or can cause additional intraoperative fractures around the tip of the nail because they may impinge to the anterior cortex of the femur and increase the stress at this area due to increased femoral bowing (7-10). This problem is more apparent in shorter patients with shorter femurs, especially in the Asian or Caucasian population. We hypothesized that non-anatomical, straight PFNs impinge to anterior femoral cortex and cause anterior tight pain in patients with increased femoral bow due to increased age and osteoporosis.
The primary outcome of this study was to determine the rate of femoral anterior cortical encroachment after stabilization of the proximal femur in a consecutive series of patients using a short straight PFN. Secondary outcome was to discuss whether PFNs need some improvements in their design for the geriatric population.
Methods
This retrospective study was performed according to Declaration of Helsinki. One hundred and eleven patients with the diagnosis of TF (31-A2.1-3 or 31-A3.1-3 according to Arbeitsgemeinschaft für Osteosynthesefragen/Association for the Study of Internal Fixation’s classifications) (11), who had been treated using PFN between 2011 and 2015, were included in the study. The data were collected using files of the patients and the digital database of the hospital. Patients with high-energy trauma, a neoplastic reason for the fractures, open fractures, multiple fractures, the American Society of Anesthesiologists score of 5, inability to walk before the injury, degenerative osteoarthritis/arthritis in the injured hip were excluded from the study. The patients who could not be reached or lost to follow up were also excluded.
The mean age of the patients (59 male and 52 female) was 74.4 (65-95) years. The left hip was involved in 72 cases, and the right hip was involved in 39 cases. The etiologies were simple fall from standing position in 106 patients and pedestrian accident in five patients.
Experienced orthopedic trauma surgeons in a university hospital operated all patients in a standard way. The surgeries were performed using fracture table and under the fluoroscopy. After closed or open reduction (if acceptable reduction could not be achieved by closed reduction techniques), the PFN (InterTAN, Smith&Nephew, Memphis, Tennessee, USA) was inserted in all patients. Its proximal lag screws and distal static locking screws were placed with the appropriate sizes. Final positions of the fractured fragments and the implants were checked under the fluoroscopy before the patients left the operating rooms. On the first postoperative day, an antero-posterior (AP) and lateral X-ray was taken and the patients were allowed walking with weight bearing as they could tolerate.
The digital AP and lateral radiographs taken on the day after surgery and at the 6-month follow up were used for the radiographic measurements. The quality of reduction of the fracture (Garden alignment index) was classified as good, acceptable or poor (12). Position of the proximal screws in the femoral head was calculated by measuring the tip-apex distance (TAD), and the neck-shaft angle (NSA) to evaluate loss in reduction during the follow up (13). Clinical evaluations were performed using Harris Hip Score (HHS) and the visual analog scale (VAS) (14). Loss of reduction, implant failure and complications were recorded.
The angle between the anterior longitudinal axis of the nail and central anatomic axis of the femur at the level of the distal tip of the nail was measured digitally on the lateral radiographs to describe the relation between the anterior of the nail and anterior cortex of the femur. This angle was defined as “angle of distal axes (ADA)”. We divided the patients into two groups according to their ADA: group 1 included patients with ADA ≤4° and group 2 included patients with ADA >4°. All of the radiographs were measured by the same orthopedic surgeon.
Statistical Analysis
Group 1 and 2 were compared in terms of HHS and VAS scores with Kruskal-Wallis test. NSAs on the second postoperative day and at the 6-month follow up radiographs were compared using Wilcoxon signed-ranks test. P<0.05 was accepted as the level of significance.
Results
Quality of reduction was good in 32 patients, acceptable in 12 patients and poor in three patients in group 1, and good in 43 patients, acceptable in 15 patients and poor in six patients in group 2. The mean TAD were 18.3±8.25 mm and 19.7±7.41 mm in group 1 and group 2, respectively.
In group 1 and group 2, the mean early postoperative NSAs were 128.8°±4.5° and 130.1°±3.7° (p=0.08) and the mean final NSAs were 126.8°±6.8° and 128.7°±4.1° (p=0.078), respectively. The mean NSAs were similar initially and at final follow up in group 1 and 2.
The mean amount of ADA was found as 4.5°±1.5°. Forty-seven patients were classified in group 1 (Figure 1) and 64 patients were in group 2 (Figure 1). The mean HHSs were found as 80.6±2.17 (78-84) points in group 1 and 79.37±3.26 (72-86) points in group 2 (p=0.464). The mean VAS scores were 2.13±0.62 (1-3) and 5.35±0.78 (4-7) in groups 1 and 2, respectively (p<0.001). Although there was no difference between the groups in terms of the mean HHSs, the VAS scores were found to be significantly lower in group 1 than in group 2.
Cut-out of the lag screws were observed in five patients as complication. In three of these patients, the nails were removed and the fractures were re-reduced and fixed with new PFNs. However, in two patients, the fractures were treated with total hip arthroplasty (Figures 2 and 2b). During the follow up period, union was achieved in all patients including PFN revisions, except two patients with athroplasty. Surgical site infection was not observed in any patient. Intraoperative or postoperative fracture of the femur around the distal tip of the nail did not occur in any patient.
Discussion
This retrospective study evaluated 111 patients with TFs who were treated with non-anatomic, straight PFNs. The main goal of the study was to investigate whether straight nails impinged with the anterior femoral cortex due to its anatomic bowing or not. Secondary aim of the study was to evaluate the clinical relevance of this possible impingement. The mean ADA was found as 4.9±1.5, and it was >4° (grade 3 to 5) in 64 patients (57.6%) which was considered as anterior cortical impingement.
Success of the IM short nails in the treatment of TFs has already been proved. Especially in unstable fractures, they protect proximal femoral anatomy and reduction against the deforming forces (15,16). It is more preferred than the sliding hip screws in the treatment of unstable fractures, including reverse oblique fractures [A3-the AO Foundation and Orthopaedic Trauma Association (AO-OTA)], because the PFN provides higher stability and its use is easier (16,17). Successful results with PFN in large series have been published although it has some complications such as intraoperative fractures, cut-out of the lag screws and varus collapse, nonunion and malunion (3,5,15,16,18). In the treatment of proximal femoral fractures, anterior cortical encroachment of the cephalo-medullary nails have been described before, however, to our knowledge, the relation between anterior cortical impingement and short PFN have not been reported before (19). Similar to the literature, in the current study, successful results with PFN were obtained in 111 patients with unstable TFs (31-A2.1-3 and 31-A3.1-3 AO-OTA) with a total complication rate of 4.5% (11).
In the treatment of TFs with proximal nailing, the most frequent complications are varus collapse of the proximal femur, cut-out of the lag screw, shortening of the femur, nonunion, secondary fracture of the femur or greater trochanter, thigh pain, screw fracture, and Z-effect or reverse Z-effect of nails with two lag screws (20,21). In our series, cut-out of the lag screw was seen in five patients, two of them were treated with arthroplasty and in three patients the nails were revised. Although the amount of varus collapse at the final radiograph compared to initial radiograph, which was about 2°, was statistically significantly different (p<0.001) in both groups, it did not have clinical or radiological impact on the results. According to alignment index of Garden, quality of reduction was good in 89% of the patients but poor in 9% of the patients (12). ADA was correlated with tight pain and VAS scores, which was higher in group 2 (p<0.001). This result shows the clinical importance of the anterior impingement of the PFN.
It has been well documented that PFN antirotation (PFNA) (Synthes, Switzerland) may cause femoral fractures or valgus impingement of the lateral cortex, especially in the Asian population (9,22-24). Because of these data in the literature, its proximal diameter, lateral bending angle and lateral surface have been revised and improved, and reproduced as PFNA-II which has been shown to be more appropriate for the Asian population. In the treatment of unstable TFs, biomechanical and clinical studies resulted in superiority of InterTan PFN because of its rotational strength and low rate of malunion (25-28). Anterior cortical impingement or destruction of the femoral nails can lead to potential complications such as thigh pain and disability and serve as a stress riser for future fractures (19,29). Similar problems may occur with straight non-anatomic nails on the sagittal plane (PFNA-II or InterTAN) due to mismatch of the anterior femoral bowing. It has been proved that anterior femoral bowing increases with the increased age and osteoporosis in the elderly; therefore short, straight nails can impinge with the anterior cortex, however the design of the PFNs have not been changed, yet (7-10).
Chang et al. (29) treated 158 patients with unstable TFs using PFNA-II and found encroachment of the distal tip of the nail to the anterior cortex in 55 patients (34.8%). Hwang et al. (9) reported mismatch of PFN/PFNA with the anterior and lateral cortices of the femur in four patients in their series. Radiological studies have also reported mismatch between the femoral nails and increased femoral bowing with advanced age (7,8). Our study showed anterior cortical impingement of the nail in 57.6% of the patients and statistically significantly higher rate of tight pain in this group. In the literature, groin or tight pain after hip nailing is not rare in this group of patients. Inappropriate stress distribution between the femur and the implant is thought to be a reason fortight pain (29-32). Similarly, in our patients, tight pain could be related with increased stress over the anterior cortex of the femur. For this reason, we recommend using shorter or curved nails in order to decrease the stress between the implant and the anterior cortex, which can decrease tight pain and possible fractures around the distal tip of the nail.
Study Limitations
In this study, we had some limitations. First, evaluation of the anterior tight pain could be more objective. Second, stress distribution over the anterior cortex could be evaluated with a biomechanical study to report a measurable data. Third, standardization of the radiological measurements was difficult. ADA was measured on the lateral radiographs, which might change with the rotation of the femur.
Conclusion
Straight, non-anatomic femoral nails impinges anterior cortex of the femur due to increased sagittal bowing of the femur by age and these nails are not appropriate for the shorter people, especially in the Caucasian population. New design proximal femoral nails with anatomic bow can solve these problems.
Ethics
Ethics Committee Approval: Retrospective study.
Informed Consent: Retrospective study.
Peer-review: Externally peer-reviewed.
Authorship Contributions
Surgical and Medical Practices: F.Y., N.M.E., D.K., Concept: F.Y., N.M.E., Y.G., T.E., Design: F.Y., N.M.E., Y.G., Data Collection or Processing: T.E., D.K., Analysis or Interpretation: F.Y., N.M.E., Y.G., Literature Search: F.Y., N.M.E., D.K., T.E., Writing: F.Y., N.M.E., Y.G.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.