Peri-trochanteric fractures represent the pre-dominant type of hip fracture in elderly individuals, with an incidence that significantly increases with age and bone density deterioration (1). These fractures primarily encompass intertrochanteric and subtrochanteric fractures, characterized by complex load-bearing regions and dependence on optimal biomechanical conditions for successful healing (2). Femoral peri-trochanteric fractures of the hip in the elderly have devastating impacts on function and quality of life (3, 4). The primary therapeutic objective for peri-trochanteric fractures is to restore patient mobility as promptly as possible, thereby minimizing the risk of long-term complications associated with prolonged immobilization, such as pressure ulcers, pneumonia, and deep vein thrombosis (5). Consequently, surgical intervention is recommended for all such fractures unless contraindicated by the patient's overall health status or the inability to tolerate anesthesia (6). The most commonly used surgical strategies for peri-trochanteric fractures include closed reduction and internal fixation using intramedullary nails, femoral proximal plates, and sliding compression hip screws (7–9). Despite significant advancements in internal fixation devices and surgical techniques, which have substantially improved the success rate of surgery for peri-trochanteric fractures, implant failure still occurs in 6%−20% of cases (10). This is particularly evident in patients with unstable peri-trochanteric fractures or those with osteoporosis. Revision surgery is a significant challenge when internal fixation fails. Factors such as soft tissue adhesion, osteoporosis, bone deformities, and the presence of old internal fixation devices can increase the complexity and difficulty of revision surgeries (11). Consequently, patients undergoing a second operation often experience prolonged surgical time, increased blood loss, higher risk of complications, and extended hospital stay (12). The reasons for internal fixation failure in trochanteric fractures include inadequate fracture reduction, compromised blood supply, insufficient stability of the fracture ends, inappropriate choice of internal fixation devices, osteoporosis, and infections, among others (13). Therefore, revision surgeries should be performed with a comprehensive understanding of the underlying causes of failure, and appropriate revision measures should be implemented. Revision strategies include bone grafting, replacement of longer or thicker intramedullary nails, addition of supplementary plates, osteotomy, joint arthroplasty, and debridement, followed by external fixation (14).

In this study, we performed a detailed retrospective analysis of cases of failed internal fixation for peri-trochanteric fractures. Our primary objective was to thoroughly investigate the underlying causes of these failures, classify them into distinct categories, and develop personalized revision strategies to improve the overall management of peri-trochanteric fractures.

Hip fractures in elderly patients, often referred to as “the final fracture of one's life”, are associated with significant mortality, morbidity, and profound functional impairment (3). With an aging population, the number of hip fractures worldwide is expected to increase from 1.66 million in 1990 to an estimated 6.25 million by 2050, representing an annual growth rate of approximately 3% (12). Despite substantial improvements in patient outcomes achieved through timely surgical intervention, 6%−20% of patients continue to experience complications leading to internal fixation failure (10). Zhang et al. (15) conducted a retrospective analysis of 204 patients treated with proximal femoral nail anti-rotation II (PFNA II) for intertrochanteric fractures and identified 22 cases of internal fixation failure. This study revealed that factors, such as inadequate fracture reduction, advanced age, femoral neck reconstruction, and unstable fractures, were significantly associated with higher failure rates. Yamamoto et al. (16) noted that intramedullary malreduction of the anteromedial cortex was associated with failed internal fixation, emphasizing that immediate post-operative imaging alone was insufficient to predict internal fixation failure. Inui et al. (17) reported that inadequate anterior cortical reduction was significantly associated with screw cutout. Additionally, Konstantinidis et al. (18) highlighted that when bone mineral density falls below 250 mg/cm3, the risk of peri-trochanteric fracture fixation failure increases substantially. Similarly, Xu et al. (19) identified the Singh index as an independent predictor of internal fixation failure in peri-trochanteric fractures.

For orthopedic surgeons, selecting an appropriate revision strategy to minimize surgical trauma and enhance the success rate of operations is a critical clinical challenge in patients who have experienced failed primary surgeries for femoral trochanteric fractures (20). Key considerations in formulating these strategies include patient age, location of non-union (inter-trochanteric or sub-trochanteric), residual bone quality, and the condition of the internal implant. This study systematically analyzed 141 cases of failed internal fixation of femoral peri-trochanteric fractures, culminating in the development of a tailored revision strategy and workflow specifically designed to address the unique challenges associated with diverse types of initial treatment failures.

In this study, type I patients included those with the original implant that exhibited no signs of breakage or loosening and had a well-aligned position, accounting for approximately 32.6% of cases. Among the revision cases of sub-trochanteric fractures, 60% were classified as type I. The non-union in these patients may be attributed to insufficient blood supply or mechanical instability. Therefore, the primary approach to revision surgery involves autologous bone grafting, either alone or in combination with auxiliary plating. To enhance the probability of fracture union, we recommend employing bicortical plates and screws that traverse a minimum of six cortices (21). Three of the 46 type I patients in this study underwent simple autologous bone grafting, while the remaining 43 received combined treatment with auxiliary plates. During follow-up, the overall fracture healing rate was remarkably high, with only one patient failing to achieve union. The patient experienced plate failure and varus deformity, necessitating a subsequent revision according to the type II-B protocol. Ultimately, bony union was achieved after the intervention.

Type II failures were primarily attributed to mechanical instability, accounting for 39.7% of all cases. Among these, patients with type II-A malocclusion exhibit discontinuity or loosening of the original internal fixation devices. For these patients, our revision strategy involved removing the original internal fixation and replacing it with stronger devices, such as longer and thicker intramedullary nails, or substituting the locking plates with intramedullary nails/dynamic hip screws, in combination with autologous bone grafting. It is important to note that the replacement of the internal fixation can be technically challenging owing to differences in device design, particularly concerning the neck-shaft and valgus angles, which can significantly impact the biomechanical stability of the hip joint. Coxa vara is a common complication following peri-trochanteric fractures, often leading to hip and knee pain, limb length discrepancy, muscle imbalance, and abnormal gait (22, 23). For type II-B patients who not only have non-union but also suffer from coxa vara, our approach involves removing the original internal fixation, correcting the hip varus deformity, performing iliac structural bone grafting, and placing a robust internal fixation to ensure long-term stability and facilitate healing. Age (particularly < 65 years) and the extent of femoral head damage are the core criteria for considering joint preservation. In younger patients, if imaging meets all criteria for “minimal femoral head damage”, the case is classified as type II-C, and joint-preserving revision is strongly indicated. Conversely, if there is evidence of femoral head collapse, articular cartilage violation, or acetabular involvement, the case is classified as type III, necessitating arthroplasty. For younger patients, the indications for joint preservation can be appropriately liberalized. The recommended approach for type II-C includes removing the original internal fixation, correcting the hip varus deformity, and using a strong internal fixation along with autologous bone transplantation. This strategy aims to maintain joint integrity and restore function, especially in patients with high activity demands and a relatively preserved hip anatomy.

Type III patients exhibit significant femoral head and acetabular joint surface destruction. In managing this complex condition, it is essential to conduct a thorough and meticulous evaluation of the patient's age, functional demands, and the extent of damage to both the femoral head and acetabulum (24, 25). In the present study, arthroplasty was conducted as a revision procedure in 32 cases: 24 THA and eight HA. Among the HA cases, seven involved patients aged >80 years. The remaining patient was 62 years old; this individual was selected for HA due to poor general health status and significant comorbidities, including hearing impairment and hemiparesis. In general, HA may be an appropriate option for elderly patients with limited mobility and minimal acetabular damage. Conversely, THA is more suitable for patients with higher activity demands or significant acetabular joint surface damage.

Infection-related non-union is a significant cause of failure in peri-trochanteric fractures (13, 26). In our study, this condition accounted for 5.0% of cases. If the pre-operative white blood cell count, CRP, and ESR levels are elevated and intraoperative pathology reveals more than five neutrophils per high-power field, infection should be suspected (27). After confirming the diagnosis of infection, we removed the internal fixation device, performed thorough debridement, applied antibiotic-impregnated cement rods to fill the medullary cavity, and used external fixation to temporarily stabilize the fracture site. After the CRP level and ESR normalize, typically within 6 weeks, revision internal fixation surgery or joint arthroplasty is performed. The primary complications associated with type IV cases include persistent or recurrent infection, femoral head necrosis, bone defects, and joint dysfunction. In this study, one patient required a second debridement and cement rod exchange due to an uncontrolled infection following the initial antibiotic-impregnated cement rod implantation. Two patients required hip arthroplasty due to femoral head necrosis, whereas the other five patients achieved bony union. All seven patients exhibited varying degrees of bone defect. Post-revision functional outcomes, as evaluated by the HSS score, were the lowest among the other three groups.

This study systematically compared the healing times and clinical outcomes of different types of non-union fractures. The results demonstrated that type IV non-union exhibited significantly higher intraoperative blood loss than the other types. Regarding healing time, although type IV showed a longer duration than the other groups, this difference was not statistically significant, likely due to the relatively small sample size (n = 7) of type IV cases. Additionally, revision surgery for type II non-union was significantly longer, which may be attributed to the increased difficulty in repositioning old fractures. Furthermore, we evaluated the surgical conditions and post-operative function of patients who underwent joint replacement and internal fixation revision surgery. Patients in the joint replacement group were significantly older, had shorter surgery times, and experienced lower intraoperative blood loss than those with type II and type IV non-union; however, no significant differences were observed when compared to type I patients. Post-operative functional scores were not significantly different between patients who underwent joint replacement and those who underwent internal fixation revision surgery. These findings are consistent with prior research (11). Therefore, in our treatment protocol, young adults with mild femoral head injuries were classified as type II, with the objective of preserving hip joint function.

This study had some limitations that should be considered when interpreting the findings. First, as a retrospective analysis conducted at a single center, it is inherently susceptible to selection bias, which could affect the generalizability of our results. Second, the assessment of functional outcomes primarily relied on the HSS score, which was not performed under blinded conditions; this may have introduced observer bias. Third, despite our analysis, unmeasured confounding factors, such as detailed patient comorbidities, bone mineral density, and post-operative compliance, could have influenced both the risk of mechanical failure and the eventual clinical outcomes. Fourth, all patients in this study underwent bone grafting procedures, although some failures due to mechanical instability theoretically may not have required such interventions. Finally, given the relatively low incidence of this condition, the sample size was modest, potentially limiting the statistical power and robustness of our conclusions. Future research should benefit from larger-scale randomized controlled trials to further validate these findings.

Despite the relatively low failure rate of internal fixation of peri-trochanteric fractures, addressing such failures can be particularly challenging. It is crucial to carefully analyze the underlying causes and formulate precise individualized revision strategies. We categorized peri-trochanteric fracture internal fixation failures into four types based on age, presence of infection, condition of the femoral head and acetabulum, varus deformity, and implant loosening or rupture. We developed a corresponding treatment flowchart for each type. By adhering to our treatment protocol, all patients achieved satisfactory radiographic outcomes and functional recovery. Our treatment strategy provides a valuable reference for standardizing the revision of peri-trochanteric fractures.