1 Introduction

Intra-prosthetic fractures (IPFs) of hip arthroplasty components present a rare but significant complication following total hip arthroplasty (THA) occurring in 0.1-3.4% of cases.^1,2 Improvements in stem design and cementing techniques have lowered rates in modern implants, however, implant failure can result in considerable morbidity and frequently require complex revision procedures.³ While much of the literature on prosthesis failure alludes to periprosthetic fractures and aseptic loosening, IPFs – particularly those involving femoral components – are less commonly reported.

Risk factors can be categorised into patient-, surgical-, and implant-related factors. Patient factors include obesity, younger age and male sex, osteoporosis and higher activity levels.^1,2,4 Fatigue related failure is the most widely accepted implant-related risk factor due to excessive mechanical overloading, as well as short stems and modular stems due to dissociation, and fretting and corrosion at the neck-stem junction.^4,5,⁶ Finally, surgical factors include varus alignment, previous revision arthroplasty and an inadequate proximal buttress support, leading to distal fixation of the stem and cantilever bending, whilst undersized stems and poor cementing techniques have also been linked to this critical mechanical failure event.^2,4

Identification of these failures can often be delayed due to the relatively atraumatic precipitating events. Stem failures often occur with minimal trauma with 87% occurring without significant precipitating event..² Symptoms can be non-specific but classical symptoms include sudden onset pain in thigh or groin, inability to bear weight on the affected limb or a grinding or clicking sensation during hip movement.⁸ Identification of the risks associated with this complication may guide prevention strategies to reduce the incidence of revision surgery, and the increased morbidity, mortality and financial burden associated with it.^4,9

We present the experience of a single academic institution with 14 intra-prosthetic hip arthroplasty implant failures between 2007 and 2024. The aim of this paper is to describe factors – surgical, implant, and patient – that may contribute to these rare but significant events. In doing so, we aim to contribute to the expanding literature and knowledge on the subject, inform prevention strategies and improve long-term outcomes in THA patients.

There remains a gap in the literature for a simple, clinically applicable classification system that allows surgeons to group femoral stem fractures based on fracture location and corresponding revision strategy. Therefore, this study also proposes and applies a pragmatic, location-based classification system for intra-prosthetic femoral stem fractures, with the aim of facilitating consistent reporting and assisting surgical decision-making.

2 Methods

A retrospective review was carried out to identify all intra-prosthetic fractures that were treated at our institution between 2007 and 2024. Data was collected from theatre logbooks and all patients who underwent revision arthroplasty for any reason were listed. Revision for periprosthetic fracture, infection, dislocation and loosening in the absence of intra-prosthetic fracture were excluded. Patients who had revision surgery for intra-prosthetic fracture were included.

Our primary outcome was to describe our experience with femoral stem fractures, including which stem designs and stem fixation were observed most, and the revision strategy for each fracture along the location. The secondary outcomes were to describe surgical-, patient- and implant-related factors that may have contributed to the device failure.

Patient records, including radiographs were reviewed to identify which patients had revision arthroplasty for fractured femoral stems, using the institutional system, Evolve (Kainos Evolve Ltd, Northern Ireland, UK). Patient records and radiographs were examined, by 2 independent reviewers, and details were included in the data collection sheet. Patient demographics were extracted and recorded.

Variables of interest included but were not limited to age, ASA, BMI, implant design, stem fixation (cemented vs cementless), location of fracture within the implant (Fig. 1), and time from index procedure to event.

Close

Fig. 1 Hip implant with labels.

Export to PPT

Each fracture was classified according to a proposed location-based system, developed by the authors following analysis of stem failure morphology within the cohort. The classification defines 4 fracture types based on the primary site of implant failure (Fig. 1):Type A: Neck/Trunnion regionType B: Proximal body/metaphyseal thirdType C: Middle thirdType D: Distal body/stem tip

Two independent reviewers assigned fracture types using standard AP and lateral radiographs. Disputes were resolved with a final assessment by a third reviewer. The classification was intended as descriptive and hypothesis-generating rather than predictive of outcome. We also propose a possible revision strategy for each fracture type, however, with a small sample size, and differences in stem designs, describing definitive surgical plans remains less robust.

3 Results

Fourteen fractured femoral implants were identified across the study period. The average age of the participants at the time of fracture was 68.5 [60.9,76.0 95%CI]. Five of the fourteen patients included were female (35.7%). Three patients had an ASA score of 1, five patients had an ASA score of 2 whilst there were 6 patients scored ASA 3. The average BMI of the patients included in the study was 31.3 [95% CI 27.8, 34.7] (Table 1).

Ten of the fourteen events included in the study occurred in the absence of trauma. The device failures occurred an average of 10.3 years (range 0.25-26 years) after the initial procedure. The average follow up was 37 months [95% 15.6, 58.3]. The most common site on the implant that fractured was the middle third, with 7 of the 14 implants showing fractures at this site. Other sites on the implant that were commonly affected were the head and neck junction (3/14), and the distal third (2/14). (Table 1)

Twelve of the fourteen episodes of stem failure occurred in cemented stems. Of those that were cemented, 8 were Exeter (Stryker™, MI, USA) stems, and 3 were Charnley (DePuy Synthes, Leeds, UK) stems. However, all patients with cemented had previously undergone hemiarthroplasty for traumatic hip fractures, indicating a likely poor bone stock even prior to insertion of the implant. (Table 1).

Using the proposed classification system, 4 fractures were type A (neck/trunnion) (Fig. 2), 1 was Type B (proximal body) (Fig. 3), 7 were type C (middle third) (Fig. 4), and 2 were type D (distal stem) (Fig. 5). Type C fractures were most common, making up 50%.

Close

Fig. 2 Fracture neck of implant type A

Export to PPT

Close

Fig. 3 Fracture middle ⅓ type B.

Export to PPT

Close

Fig. 4 Fracture proximal ⅓ type C.

Export to PPT

Close

Fig. 5 Subtle type C fracture.

Export to PPT

Two of the type A fractures occurred in Charnley stems, with one occurring in an Accolade stem, however, all 3 required extended trochanteric osteotomy to remove the well-fixed stem. The Accolade stem was revised with a cemented Exeter revision stem, with cables to fix the osteotomy. The Charnley stems were revised with a long, press fit stem with a hook plate to repair the osteotomy.

One of the Type B fractures occurred in a Charnley stem which was revised with a proximal femoral replacement (PFR), with the other occurring in an Exeter stem. The Exeter stem was revised with a long cementless stem, and the ETO was fixed with a plate and cables.

There were two fractures around cementless stems, one was a type C fracture, around an extended porous stem and the other a type D around a Taperloc stem. All the type C and D fractures required revision with long stems or PFRs and a plate and cables.

4 Discussion

This case series of fourteen intra-prosthetic fractures details the occurrence and complexity of intra-prosthetic fractures in hip arthroplasty, across a seventeen-year period at a single academic institution. We aimed to highlight the surgical-, implant- and patient-related factors that may contribute to femoral stem failure, including the involvement of specific implant designs, locations of fractures and associated mechanical or biological factors. Our proposed classification system is preliminary and requires external validation with large sample size studies.

Twelve of the fourteen prosthesis failures occurred in cemented stems, with seven of these occurring in polished taper slip stems (PTS). Historically, cementless stems have shown a higher correlation with IPFs, particularly in the presence of proximal bone loss, leading to a risk of distal fixation of the implant.² However, there is a growing body of evidence implicating PTS and a risk of periprosthetic femoral fractures^2,7 and whilst actual fracture of the stem itself is reportedly rare, some studies report a rate of 2.2% within 5 years. Factors contributing to this high incidence of stem failure include short femoral stems and poor cementing technique.¹⁰ Furthermore, the PTS included were inserted in the context of trauma and, whilst we were unable to extract any robust data on bone density, would lead us to believe that the quality of bone stock or adequate proximal fixation may also be a factor.

Several contributing patient factors were identified in our cohort, including male sex (64.3%), high body mass index (mean, 31.3), and high-demand physical activity coinciding with the patient related risk factors reported in the literature.² The majority of male patients involved in this case series were elderly farmers with a suspected, yet undocumented, high activity level.

The role of implant age also appeared significant with fractures tending to occur in implants older than 16 years, suggesting a time-dependent risk of fatigue failure. Ten of the fourteen events occurred in the absence of trauma, highlighting the relative innocuity of this presentation. The risk of missed or misdiagnosis is an important consideration, particularly due to the often-subtle signs on X-ray (Fig. 6), not to mention that patient and implant outcomes post all-cause revision is more favourable than revision post-PFF^11–13

Close

Fig. 6 Fracture distal ⅓ type D.

Export to PPT

The small sample size in this study makes it impossible to draw inferences about overall risk factors for femoral stem failure. However, we do see a correlation between the variables and demographics in our cohort and the documented risk factors in the literature. Whilst we were unable to collate robust data on sizing of the stems, Thompson et al., in their systematic review and single centre cases series on Exeter V40 femoral stem failures, suggest that short stems (<125 mm) are associated with higher rates of fracture, earlier time to fracture and younger patient age at time of fracture.¹⁴

The literature suggests that surgical management of IPFs often proves technically demanding. In some cases, extended trochanteric osteotomy, and proximal or total femoral replacement is required, depending on the extent of component damage and surrounding bone integrity.¹ Our experience with femoral stem fractures reflected this with almost all cases requiring ETO and fixation of the bone with plate and cables as part of the revision procedure. Outcomes varied based on preoperative bone stock and patient comorbidities, though follow-up indicated satisfactory restoration of function in most cases.²

A key finding of this study is that intra-prosthetic fractures cluster anatomically. The proposed classification provides a simple, clinically reproducible method to describe implant fractures by their primary failure zone (Table 2). Importantly, each fracture type suggests different revision strategies: Type A fractures, involving the neck/trunnion may be amenable to modular neck exchange, whereas type B and C failures, involving the proximal body and the middle third, respectively, typically require full stem revision. Type D fractures often necessitate diaphyseal fixation or extended trochanteric osteotomy. However, in relation to Type A fractures in modular stems, while revision of the neck is possible, several factors can impact the likelihood of this, most notably, corrosion and metal debris at the neck-taper interface, and damage to the taper itself in femoral stem fractures.^5,15

Preventive measures for this catastrophic failure, including rigorous surgical planning, accurate sizing and position of stems must be considered by all orthopaedic surgeons to reduce the risk of such an event that can have such significant implications on patient morbidity and mortality. The literature recommends avoidance of varus position and undersized stems, both of which have been proven to impact the survivability of femoral implants.² Furthermore, revision surgery for stem fractures shows higher complication rates with 24% subsidence and 33% mortality in some cohorts.¹

5 Conclusion

This observational study reports on a single institution's experience with femoral implant stem failures. We noted a proportionally higher rate of stem failures in cemented stems, particularly PTS implants. Most stem failures in our study occurred in elderly male patients in keeping with the evidence in the literature, and the high proportion of atraumatic stem fractures highlights the relative innocuity associated with the incidence of femoral stem implant fractures. We also propose a simple, location-based classification system that may facilitate clearer communication and guide surgical planning in cases of intra-prosthetic stem fracture.

Credit author statement

Robert Woods: Data curation, Investigation, Writing – Original draft preparation, writing – Reviewing & Editing.

Alexander Price: Investigation, Data curation, Writing – Reviewing & Editing.

Tomas Donovan: Investigation.

Stephen Kearns: Supervision, Validation.

Colin Murphy: Supervision, Validation.

Gerard Sheridan: Conceptualisation, Investigation, Visualisation, Supervision, Writing – Reviewing & Editing.

Ethical statement

Ethical approval for the completion of this research project, Failure of the Femoral Stem: A Case Series on Intra-Prosthetic Fractures and Proposed Classification System, was granted by the Institutional Research Board at our hospital.

Funding statement

The authors of this paper Failure of the Femoral Stem: A Case Series on Intra-Prosthetic Fractures and Proposed Classification System received no funding for the research, authorship and/or publication of this of this manuscript.