Hip resection arthroplasty as a primary treatment of displaced neck fracture in non-ambulatory and fragile patients

Abstract

Due to the increasing ageing population, femoral neck fracture (FNF) is a common and significant public health issue in the elderly, as it significantly impacts patients’ quality of life, frequently leading to severe disability. Undoubtedly, hip replacement is the standard current of care for displaced FNF in this population, as it can provide pain relief and allow immediate return to mobility. However, hip arthroplasty may present severe specific complications, such as implant dislocation and infection, which may increase mortality and morbidity, especially in more frail patients. Therefore, in this particular population, alternative treatments should be considered. Girdlestone resection arthroplasty, which includes excision of the femoral head, is a salvage procedure which was first described for the management of chronic tuberculous coxitis, and then widely used in uncontrolled infected hip replacements. This article provides an updated outcome analysis of hip resection arthroplasty as a primary definitive treatment for FNF in frail non-ambulatory patients.

Key Words: Hip resection arthroplasty; Excision arthroplasty; Girdlestone; Femoral neck fractures; Elderly; Frailty; Review; Definitive treatment; Non-ambulatory

Core Tip: Femoral neck fracture (FNF) is actually the most critical traumatic event in the elderly, because of the epidemic incidence and the high rate of mortality. Hip arthroplasty is the gold standard treatment for displaced FNFs, allowing early postoperative mobility, but severe complications may occur, increasing mortality and morbidity. Therefore, in most frail patients, alternative surgical options should be considered. Girdlestone resection arthroplasty is a salvage procedure which is currently used in various hip disorders. This article provides an updated analysis of outcomes of hip resection arthroplasty as a primary definitive treatment for acute FNF in frail non-ambulatory patients.

INTRODUCTION

Due to its epidemic incidence in the last decades, femoral neck fracture (FNF) is a common orthopaedic injury in the elderly, as it includes high rates of mortality and complications that has a significant impact on patients’ quality of life[1]. Furthermore, severe disability and impaired mobility may occur, with expensive economic consequences[2]. As the global population ages, the incidence has increased from 1.66 million in 1990 to 2.6 million expected in 2025, and it is estimated to reach 6.3 to 8.2 million cases in 2050[3].

Hip replacement, particularly hemiarthroplasty (HA), is the treatment of choice for Garden 3 and 4 fractures, as it proved to be superior to fixation regards to postoperative function, pain and complications, allowing to avoid the non-weight bearing restrictions[4]. However, hip arthroplasty is not free from complications (implant dislocation, infection, and failure, periprosthetic fracture, fat embolization syndrome), which require reinterventions that may increase morbidity and mortality[5].

Therefore, in more frail patients, alternative therapeutic options should be considered. Girdlestone resection arthroplasty (RA), which comprises excision of the femoral head, is a salvage procedure which was first described for the treatment of chronic tuberculous coxitis, and then widely used in uncontrolled infected hip replacements. This article provides an updated outcome analysis of hip RA as a primary definitive treatment for FNF in frail non-ambulatory patients.

METHODOLOGY

A comprehensive search was conducted in PubMed and Scopus databases between December 2024 and April 2025. The search strategy included combinations of the following keywords: “hip resection arthroplasty”, “girdlestone”, “excision arthroplasty”, ”frail”, “elderly”, “non-ambulatory”, “definitive treatment”, “complications”, and "femoral neck fracture”.

Inclusion criteria

Inclusion criteria comprised peer-reviewed articles published in English between 2018 and 2025 that reported the outcome of RA as a primary definitive treatment of FNF in frail non-ambulatory patients. Non-ambulatory status was defined as bedridden, or wheelchair-bound, with the ability to stand only for transfers. Moreover, special attention was given to the technical changes that the Girdlestone procedure received over time, and to the current indications in orthopedics. All studies were considered, regardless of sample size.

Exclusion criteria

Papers that were not written in English or in which full-text was not available; case reports and studies considering RA in revision surgery as a treatment of pyogenic hip infections or after hip replacement complications.

The first search yielded 106 papers, 26 of which were considered potentially relevant to this review. The initial selection was made by screening titles and abstracts of all the articles retrieved from the search. Full-text analysis was performed on all the articles identified to be suitable and in those cases in which the selection of the paper could not be adequate by reading only titles and abstracts.

Four out of the 26 full-text studies were assessed for eligibility, but 2 were then excluded as they did not fulfill the inclusion criteria. Two articles met all the criteria and were eventually analyzed in this article. References within included articles were also examined to identify additional relevant publications. Given the heterogeneity of study designs and outcomes, a formal meta-analysis was not feasible; instead, the results are presented in a descriptive and thematic synthesis in the following sections.

RA (GIRDLESTONE’S PROCEDURE)

RA was first described in 1928 by an English orthopedic surgeon, Gathorne Robert Girdlestone, who performed the procedure for the treatment of tuberculosis of the hip and septic arthritis in a pre-antibiotic era[6,7]. However, the first proximal resection of the femur was carried out more than one century before by Anthony White in a 9-year-old boy with septic pseudoarthrosis[7,8].

The technique originally involved excision of the greater trochanter (GT) and all lateral muscles and, if necessary, the acetabulum edges were flattened. The surgical incision, as long as 12 cm to 15 cm, was centered on the GT, exposing the deep fascia and the gluteal muscles, which were removed together with the GT, all damaged bone, and acetabular cartilage. Gauze wicks were then packed to fill the cavity[7].

Afterwards, between 1928 and 1950, the procedure was used for pyogenic infections of the hip arising from gunshot wounds or fractures[7,9]. Although it is no longer performed as originally described, the eponymous term “Girdlestone technique” is still used. Hip RA is actually a less invasive procedure, which leaves a rudimental coxofemoral joint. The consequent cranial migration of the proximal femur, leading to limb shortening (5-10 cm) and hip instability, causes an inevitable need for walking aids.

As well as surgical technique for head and neck resection, also indications have changed over the years. In the early 1960s, following the development of total hip arthroplasty (THA), RA was considered the last resort for managing infection or mechanical failure of the prosthesis. However, after removal of all hardwares, the gluteal muscles were spared.

In a recent case series including 91 hips which underwent femoral RA for refractory prosthesis infection, at a minimum follow-up of two years eradication was achieved in 65 cases (71%), and 6 postoperative deaths related to infection occurred (6/88, 7%). In 20 out of 91 hips (22%), one or more operations were performed for persistent infection, whose control was achieved in 10/20 (50%). The mean limb shortening was 5.7 ± 1.7 cm, and the average walking distance was 175 ± 350 m. The mean Merle d’Aubigné and Postel score at final follow-up was 9.3 ± 2.2 out of 18. The Authors concluded that definitive removal of the hip implant remains a viable therapeutic option in refractory infections, despite a substantial complication rate, and should be considered particularly when the prosthesis is no longer functional[10].

In 2012, Cordero-Ampuero[11] stated that absolute indications for RA were impossible reimplantations (due to unacceptable surgical or anesthetic risk, technical difficulties, patient rejection), and non-ambulatory subjects because of other problems. Furthermore, dementia, immunocompromise, and intravenous drug abuse were relative indications due to the increased risk of prosthesis dislocation[12].

Currently, RA is recommended in persistent infection, in severe periprosthetic bone loss, and in patients with multiple comorbidities[13,14]. From 1978, this technique has been applied in non-ambulatory cerebral palsy patients with chronic pain and hip dislocation, as it minimizes the risks of prosthetic surgery[15]. RA, usually including interposition myoplasty, leads to a mobile lower limb which allows an easily and comfortably positioning irrespective of pelvic obliquity or fixed contracture of the contralateral hip. The risk of failure associated with painful proximal femoral migration may take advantage of the interposition myoplasty, providing an improvement in quality of life and patient/carer satisfaction[16].

Shah and Parker[17] published a retrospective review of a prospectively collected database involving 36 Girdlestone procedures performed over a 30-year period as a primary treatment (3 cases) or as a result of a complications following surgery for a fracture of the hip. Overall, 41.7% of patients had died within 1 year after operation. In the remaining 19 cases, pain persisted universally in walking subjects (36.9%), and most of them experienced a significant loss of independence and a reduction in mobility status, always requiring aids. Therefore, due to the high degree of morbidity and mortality, it should only be used as a salvage procedure after all other surgical measures have been exhausted.

LITERATURE REVIEW

In 2022, Steelman et al[18] were the first to describe RA for acute FNFs in five patients (6 hips), who were non-ambulatory for severe clinical conditions not including cerebral palsy or spinal cord injury. All fractures occurred due to mechanical falls during transfers. They were retrospectively compared with the most recent 10 consecutive patients treated with HA. A standard anterior (Smith-Peterson) access was performed in all RAs, while prostheses were implanted by the same group of surgeons through both anterolateral (Watson-Jones) and posterior approaches. At 1-year follow-up, RA resulted in significantly reduced postoperative vs preoperative pain scores, and decreased operative times, allowing immediate return to functional baseline status with the ability to sit up in bed or in a chair. Due to infection, one prosthesis required explantation, irrigation, debridement, and placement of an antibiotic cement spacer. The Authors concluded that RA may be an acceptable treatment for FNFs in the non-ambulator, preventing possible arthroplasty-specific complications.

At a mean follow-up of 1.5 years, a retrospective study comparing two matched cohorts (21 RA and 42 HA, with a mean age at surgery of 79 years and 81 years, respectively) was recently published by Bellova et al[19]. Both operations were performed using a direct lateral approach. Although 1-month mortality did not differ, 1-year mortality was higher in the RA group (71% vs 49%). The mean surgery duration was 42 minutes and 84 minutes following resection and hemiarthroplasty, respectively. Functional scores were lower after Girdlestone procedure, with 82% vs 19% bedridden patients. One HA was converted to a dual-mobility cup for recurrent dislocation. Furthermore, an intraoperative periprosthetic fracture of the GT occurred, which was treated conservatively with partial weightbearing. Consequently, due to higher survival and better functional outcomes of hemiarthroplasty, RA for FNF should be selected restrictively to subjects with remarkable comorbidities and limited mobility or life expectancy (Table 1).

Table 1 Published data concerning hip resection arthroplasty as a definitive treatment for femoral neck fracture in elderly.

Ref.	Design	Cases	Mean age	Mean CCI	Follow-up (years)	Outcomes	Surgical approach	Mean surgical time (minutes)	Complications (HA only)	Results	Conclusions
Steelman et al[18], 2022	Retrospective case series with comparison group	6 RA	NA	NA	1	(1) Post- vs preoperative pain scores; (2) Operative time; and (3) Function	RA: Standard anterior	RA: 59.2	1 infection (revision)	RA: Decreased VAS pain scores (7.7 vs 3.3) and operative time, immediate return to baseline function	RA: Acceptabletreatment in the non-ambulator
		10 HA					HA: Anterolateral or posterior	HA: 111.8
Bellova et al[19], 2024	Retrospective matched cohort study	21 RA	RA 79	7.1	1.5	(1) Survival; and (2) Function	Direct lateral	RA: 42	1 recurrent dislocation (revision), 1 greater trochanter fracture	RA: Mortality 71%, bedridden 82%	RA: Higher short-term mortality and poorer functional results
		42 HA	HA 81					HA: 84		HA: Mortality 49%, bedridden 19%

RA: Resection arthroplasty; HA: Hemiarthroplasty; CCI: Charlson comorbidity index; VAS: Visual Analogue Scale; NA: Not available.

ANALYSIS

FNFs are the most common traumatic injuries in the elderly, and are associated with low energy falls. As society ages, the annual number of cases is constantly increasing[2]. Consequently, a significant public health issue is occurring, as FNF frequently leads to severe disability, which impacts patients’ quality of life, and increases considerably the mortality rate[1,2].

Hip replacement is the gold standard procedure for displaced fractures in elderly population. In a systematic review and meta-analysis of randomized control trials, THA proved to be superior to HA in terms of total and pain Harris Hip Score, Euroqol-5-dimensions, and acetabulum erosion, showing a lower mortality rate within 6 months after surgery[20]. The average operative time and dislocation rate were higher in the THA group, as well as general complication rate. No significant difference in terms of reoperation rate, postoperative infection, peri-prosthetic fracture, and venous thromboembolism prevalence was found. Consequently, THA may be an advisable treatment option for active elderly patients over 75 years old, as it can provide better hip function and life quality with acceptable risks. Particularly in the first 6 months, a careful management is required to prevent dislocation following a THA.

Moreover, the mid- and long-term outcomes of hip replacement in elderly patients were compared by Liu et al[21] in 147 patients (46 THA and 101 HA) using a propensity score-matched analysis. The THA group demonstrated significantly longer operative times, greater blood loss, and higher hospitalization costs compared to the HA group. No significant difference was observed in postoperative hospital stay, but the incidence of pain was significantly higher in the HA group. At 6 months and 12 months after surgery, Harris Hip Scores were significantly higher in the THA group, and no significant differences were observed in the overall rate of postoperative complications. They concluded that THA offers superior long-term outcomes, although HA may be more appropriate for patients with poor general health.

However, general and local complications may occur, including fat embolization syndrome, prosthesis dislocation and infection, implant failure, and periprosthetic fracture, which require reoperations that may increase morbidity and mortality[5].

Girdlestone RA, which includes excision of the femoral head, is a salvage procedure that may be considered an acceptable option of treatment for non-ambulatory and high-risk surgical patients. First described in 1928 as a treatment for tuberculosis of the hip, RA subsequently became widely used to manage septic arthritis[17,22,23].

Twenty-four cases of RA performed as a definitive treatment in various complex hip problems showed elevated reoperation and early mortality rates, as well as low functional outcomes. Consequently, it should be considered only in fragile patients with severe medical comorbidities who underwent multiple revision surgeries[24].

Conversely, at a median follow-up of 70 months, Macaux et al[25] reported adequate sepsis and pain control in 30 cases of multi-operated hip infections, concluding that the Girdlestone procedure may be a valuable last-line rescue strategy for patients with complex or recurrent periprosthetic infections.

In a retrospective analysis conducted on 65 patients, 25 paraplegic and 40 nonparaplegic, RA proved out to be a definitive solution that successfully treats infections in both groups, with higher rates of reoperations in paraplegic cohort[26].

In an observational cadaveric study, Spuehler et al[27] found that RA performed through a direct anterior approach causes a 3.8 cm and 1.8 cm shift of the femur in vertical and dorsal direction, respectively. Moreover, a 36.5° external shift was observed. The subsequent changes in the anatomy of the femoral nerve may cause stress at the lateral oblique femoral nerve division leading to poor functional outcomes.

RA is now recommended in persistent and unresolvable infections, in cases of recurrently dislocating prostheses or repeated failed revisions, and in patients with multiple and severe comorbidities[17,28]. Hip RA certainly has faster surgical times than hip arthroplasty, significantly reducing odds of complications related to anesthesia. The primary target of the procedure is pain relief, as well as return to partial activities, according to the limited functional demands and lower expectations of the older population[29-31].

The majority of patients undergoing RA experienced satisfactory pain control with acceptable daily life. It is a quite simple surgical procedure that allows to regain a functional and painless hip joint, although significant limb shortening constantly occurs. Grauer et al[32] showed that subjects with less shortening gained better walking ability and function. Furthermore, the inevitable need for walking aids is not only related to the leg length discrepancy, but also to the associated gluteus medius insufficiency[12].

Currently, RA is rarely indicated as a primary procedure following hip trauma, and most papers consider this technique basically to manage failure of previous surgeries[33]. In a cohort of 290 patients who underwent RA, FNF or malunion/nonunion were the second common indication for the procedure (26%) after infection (39.8%). Moreover, short-term complication rates were comparable between RA and revision hip arthroplasty[34].

In a retrospective descriptive study including a mixed population of 24 cases followed up to five years, Sawadogo et al[8] found a mean shortening of 3.5 cm, with an average hip flexion of 102°, and Trendelenburg lameness. The mortality rate was 20%, and the functional outcome was not excellent, but all patients were autonomous with walking aids. Therefore, although RA cannot be a first intention indication, it could be considered as an acceptable option for low-demand patients with high surgical and anesthetic risk.

The main limitations of this review are lack of randomized controlled trials, and retrospective design and limited quality of the available studies, which included small sample sizes.

CONCLUSION

Hip replacement remains the optimal surgical treatment for displaced FNFs in elderly. However, in very selected cases, including non-ambulatory and frail patients with severe medical comorbidities, RA could be considered as a primary and definitive procedure to control pain and allow immediate mobilization. Further studies with a larger sample size are required to confirm this possibility.