Does an Extended Offset Stem Improve Hip Stability and Function in Elderly Patients with a Femoral Neck Fracture?

Abstract

Background

Hip fractures in elderly patients remain a major clinical challenge, with optimal surgical strategies still debated. Evidence regarding abductor strength recovery and functional outcomes after total hip replacement (THR) using an extended offset cemented stem in elderly patients with femoral neck fracture (FNF) remains limited. We evaluated functional outcomes and abductor muscle strength of THR for FNF in elderly patients, comparing an extended medial offset (MO) stem and a standard MO stem.

Methods

We retrospectively reviewed 120 patients undergoing THR with an extended MO Exeter 44-mm stem (group A) and 100 patients undergoing THR with a standard MO Exeter 37.5-mm stem (group B) for FNF. We enrolled patients with a Koval grade of 3 or higher, an American Society of Anesthesiologists (ASA) grade of 3 or lower, and a Mini-Mental State Examination score of 24 or higher. Postoperative MO of the affected hip was compared with the native hip. Abductor muscle strength was measured using a muscle dynamometer (KINVENT technologies). Complications including dislocation, thigh pain, and greater trochanteric pain (GTP) were investigated.

Results

No significant differences in ASA, body mass index, bone mineral density, or implant position were observed between the 2 groups. The mean MO of the affected hip was 42.6 mm in group A and 36.2 mm in group B, approximately 5 mm greater than the native hip in group A (37.8 mm) and 6 mm greater in group B (30.4 mm). Abductor muscle power of the affected hip recovered to approximately 86% of the native hip in group A and 81% in group B. No dislocations or thigh pain occurred. In group A, 5 patients had mild GTP treated with physical therapy. Group A demonstrated smaller differences between the affected and native hips in mean MO, mean vertical offset, and abductor muscle strength compared with group B (p < 0.05).

Conclusions

For healthy elderly patients with FNF undergoing THR, the extended MO stem had a positive effect on the recovery of abductor muscle strength, potentially contributing to improved joint stability.

The surgical treatment of hip fracture in elderly patients has long been debated. Orthopedic surgeons have extensive experience in managing this injury. In contrast to intertrochanteric fractures of the femur, femoral neck fractures (FNF) exhibit superior clinical outcomes with joint replacement compared to internal fixation.¹,²,³⁾ Total hip replacement (THR) is an uncommon procedure for the displaced FNF in elderly patients due to the high risk of mortality and morbidity associated with old age, osteoporosis, mental status, medical comorbidities, and postoperative complications.⁴,⁵⁾

The goal of treatment is to achieve complete rehabilitation and return to the individual’s pre-injury condition. However, one of the greatest concerns in the surgery of FNF with THR is the assumed increased risk of dislocation.⁶⁾ The development of femoral stem designs, including decreased neck-shaft angle and modular stems, along with new concepts for acetabular sides such as larger diameter heads and dual-mobility implants, may help increase construct stability and decrease dislocation.⁷,⁸,⁹⁾ Also, an extended medial offset (MO) stem, which is related to the abductor muscle moment arm, has an effect on reducing wear on bearing surfaces, implant loosening, and dislocation rates.¹⁰,¹¹⁾ However, the recovery of abductor muscle strength and functional outcomes after THR with an extended femoral offset (FO) cemented stem in FNF in elderly patients is not well documented. We conducted a retrospective study to verify the relationship between an extended FO cemented stem and abductor muscle strength, joint stability by increasing the abductor lever arm, leg length, and function following THR for FNF in elderly patients.

METHODS

Study Participants

This study was approved by the Institutional Review Board of Wonkwang University Hospital (IRB No. 2025-12-006). This study was conducted as a retrospective study, and the requirement for informed consent was waived by the Institutional Review Board.

We retrospectively reviewed 120 cases that underwent THR for FNF using a 44-mm extended MO stem (group A), 100 cases using a 37.5-mm standard MO stem (group B) with cement in our hospital between July 2010 and February 2019. Group A consisted of 42 men and 78 women with an average age of 81.5 years (range, 75–97 years). Group B consisted of 39 men and 61 women with an average age of 81.0 years (range, 75–94 years). We enrolled patients with a Koval mobility score of 3 or greater, an American Society of Anesthesiologists (ASA) physical status classification of class 3 or lower, and a Korean version of the Mini-Mental State Examination (K-MMSE) score of 24 or above to minimize the negative effect on rehabilitation after the operation (Table 1). Exclusion criteria were defined as a history of hip or knee surgery before THR, multiple fractures, and musculoskeletal disorders. All cases in group A used the Exeter 44-mm offset (Stryker) for the femoral side. We used the Secur-Fit acetabular cup (25 cases, metal on polyethylene) and Trident II cup (30 cases, ceramic on polyethylene; 20 cases, ceramic on ceramic, Stryker), and modular dual mobility acetabular cup system (45 cases, Stryker). The femoral head size was 28 mm in 25 cases with metal implants and in 45 cases with ceramic implants, and 32 mm in 50 cases with ceramic implants. All of group B underwent THR using the Exeter 37.5-mm offset (Stryker) for the femoral side. The acetabular cup used was the Secur-Fit (Stryker: metal on polyethylene in 11 cases, ceramic on polyethylene in 48 cases, and ceramic on ceramic in 17 cases) and modular dual mobility (24 cases, Stryker). The femoral head size was 28 mm in 11 cases with metal implants and in 37 cases with ceramic implants, and 32 mm in 52 cases with ceramic implants. The modified Hardinge approach in the lateral position was used in all cases. We performed THR using third-generation cementing techniques, including distal plugging, retrograde gunning, and proximal pressurization.

Table 1. Demographic Characteristics.

Variable		Group A	Group B	p-value
Age (yr), mean (range)		81.5 (75 to 97)	81.0 (75 to 94)	0.412*
Sex (male : female)		42 : 78	39 : 61	0.637†
Follow-up duration (yr), mean (range)		5.3 (3.4 to 7)	5.1 (3.0 to 7)	0.200*
BMI (kg/m2), mean (range)		23.4 (20 to 30)	23.1 (19 to 29)	0.431*
BMD (T-score), mean (range)		–2.57 (–3.1 to –2.2)	–2.49 (–3.4 to –1.5)	0.201*
ASA class
	Grade 1	65	52	0.750‡
	Grade 2	41	35	0.897‡
	Grade 3	14	13	0.765‡
K-MMSE, mean (range)		25.8 (24 to 30)	25.6 (24 to 30)	0.567*

Group A: extended MO exeter 44-mm stem group, Group B: standard MO exeter 37.5-mm stem group.

BMI: body mass index, BMD: bone mineral density, ASA: American Society of Anesthesiologists, K-MMSE: Korean version of the Mini-Mental State Examination.

^*t-test. ^†Chi-square test. ^‡Two-proportion z-test.

The postoperative exercise protocol began with bedside quadriceps exercises on the first day after surgery, and wheelchair and walker walking exercises were started on the second to third days after surgery if the pain was tolerable. In addition, continuous muscle strengthening exercises were performed with the goal of walking for 30 minutes 3 times a day for up to 6 weeks after surgery.

Research Methods

The anteroposterior (AP) and lateral radiographs were taken before and after surgery. Hip AP radiographs with internal rotation of 15° were calibrated using the 5-cm metal block. For the radiological measurement, the hip joint center was defined as the geometrical center of the femoral and prosthetic femoral head (Fig. 1). MO was calculated as the distance from the longitudinal axis of the femur to the center of the femoral head. Vertical offset (VO) was the distance from the center of the femoral head to a perpendicular line passing through the medial border of the ipsilateral teardrop point of the pelvis. Leg length was determined by drawing a line through the inferior borders of the ischial tuberosity and measuring the distance to the superior margin of the lesser trochanter (Fig. 2). Radiographic parameters were measured on postoperative radiographs for both the affected hip (THR state) and the native hip.¹⁰,¹²⁾

Fig. 1. Radiographic templating illustrating the method to evaluate variations in femoral offset and limb length between the center of rotation of the native femoral head and that of the prosthetic femoral head.

Fig. 2. Radiologic measurements obtained. Femoral offset: red line, vertical height: blue line, leg length discrepancy: difference between 2 black lines.

We measured isometric abductor muscle strength by the peak force of both legs (injured and normal legs) using a hand-held dynamometer (KFORCE device and software, KINVENT). Before the strength test, the patients underwent 2 to 3 minutes of passive stretching and active-assistive exercises of the lower extremities with a physical assistant. The patients lay on a table in the lateral position with the test leg up and the resistance arm of the hand-held dynamometer placed above the lateral malleolus. The test leg was positioned straight, whereas the non-test leg was positioned in about 30° of hip and knee flexion against the table for comfort and stability. The native leg was tested first. Each leg was tested 3 times and the highest recorded maximum isometric muscle strength for 3 seconds was selected (Fig. 3). We also observed complications during and after surgery such as periprosthetic fracture, dislocation, thigh pain, and greater trochanteric pain.

Fig. 3. Repetitions analysis for the maximum abduction strength in kilograms.

Continuous variables are presented as mean (range) and were compared between groups using the independent samples t-test. For side-to-side comparisons, we calculated the within-subject difference (Δ) as affected minus native for each patient, and the Δ values were compared between groups. Because the variances of Δ were not equal, Welch’s t-test was used. Categorical variables are presented as counts and were compared using the chi-square test, and Fisher’s exact test was used when expected cell counts were small. For comparisons of proportions across specific categories, a 2-proportion z-test was performed. Statistical analysis was performed using IBM SPSS version 27.0 (IBM Corp.). The p-values of < 0.05 were considered significant.

RESULTS

The baseline demographic characteristics were generally comparable between group A and group B, with no significant differences in age, sex distribution, follow-up duration, body mass index, bone mineral density (T-score), ASA class distribution, or K-MMSE score (all p > 0.05) (Table 1).

In group A, the mean MO of the native hip was 37.8 mm (range, 31–42 mm) and the mean MO of the affected hip was 42.6 mm (range, 40–48 mm) after THR. The mean VO was 22.5 mm (range, 18–25 mm) on the affected hip and 20.7 mm (range, 15–22 mm) on the native hip. Postoperative leg length discrepancy (LLD) was found to be less than 5 mm in all cases, with 100 cases exhibiting equal lengths, 16 cases having longer lengths, and 4 cases having shorter lengths. Stem alignment was neutral in 104 cases, varus (< 3 mm) in 10 cases, and valgus (< 3 mm) in 6 cases. In all cases, the acetabular cup was positioned in the safety zone (anteversion: 10°–20°, inclination: 30°–50°). The mean abductor muscle strength in the affected legs was 16.0 (range, 4.5–21.5), while in the native legs, it was 18.6 (range, 5.2–26.6) (Table 2). The abductor muscle power of the affected legs recovered to as much as 86% of the non-operated legs.

Table 2. Radiographic Results.

		Group A	Group B	p-value
Mean inclination (°), mean (range)		37.5 (30–47)	38.1 (31–48)	0.402*
Mean anteversion (°), mean (range)		15.9 (10–19)	15.3 (11–19)	0.480*
Femoral stem position
	Neutral	104	88	0.366†
	Varus (< 3 mm)	10	5	0.732‡
	Valgus (< 3 mm)	6	7	0.799‡
Femoral offset (mm)
Mean medial offset (affected : native)		42.6 : 37.8	36.2 : 30.4	0.005§
Mean vertical offset (affected : native)		22.5 : 20.7	19.3 : 16.2	0.010§
Leg length discrepancy		All < 5 mm	All < 7 mm	-
Mean abductor muscle strength (kgf) (affected : native)		16.0 : 18.6	13.0 : 16.0	0.045§

Group A: extended MO exeter 44-mm stem group, Group B: standard MO exeter 37.5-mm stem group.

^*t-test. ^†Two-proportion z-test. ^‡Fisher’s exact test. ^§Welch’s t-test.

In group B, the mean MO of the native hip was 30.4 mm (range, 26–36 mm). The mean MO of the affected hip was 36.2 mm (range, 34–39 mm) after THR. The mean VO was 19.3 mm (range, 17.1–23.2 mm) on the affected hip and 16.2 mm (range, 13–20.4 mm) on the native hip. Postoperative LLD was found to be less than 7 mm in all cases, with 75 cases exhibiting equal lengths, 12 cases having longer lengths, and 13 cases having shorter lengths. Stem alignment was neutral in 88 cases, varus (< 3 mm) in 5 cases, and valgus (< 3 mm) in 7 cases. In all cases, the acetabular cup was positioned in the safety zone (anteversion: 10°–20°, inclination: 30°–50°). The mean abductor muscle strength in the affected legs was 13.0 (range, 3.7–18.9), while in the native legs, it was 16.0 (range, 4.8–23.6) (Table 2). The abductor muscle power of the affected legs recovered to as much as 81% of the native legs.

Differences in measurements between the affected hip and the native hip were compared between group A and group B. Group A showed significantly smaller mean MO differences compared with group B (p = 0.005, Welch’s t-test). Likewise, the mean VO difference between the affected and native hips was significantly smaller in group A than in group B (p = 0.01, Welch’s t-test). The difference in mean abductor muscle strength between the affected and native hips was significantly reduced in group A compared with group B (p = 0.045, Welch’s t-test).

There were no other intraoperative or postoperative complications, such as periprosthetic fracture, infection, or dislocation. There were 5 patients in group A, who complained of pain around the greater trochanter with a visual analog scale score of < 5. Symptoms improved with medication and physical therapy and did not require additional surgery.

DISCUSSION

Hip fracture is a clinical concern for surgeons due to the high morbidity and mortality rates in elderly patients. Unlike trochanteric fractures, displaced FNF are treated with hemiarthroplasty (HA) in elderly patients who are institutionalized, have cognitive impairment, and limited physical activity. HA is usually performed when the acetabular cartilage is intact. HA provides similar outcomes with satisfactory stability but includes inguinal pain and a risk of acetabular wear, which may eventually require conversion to THR. THR is associated with better functional outcomes and a lower risk of revision surgery in self-supporting, physically active patients.¹³,¹⁴⁾

The goal of treatment is full rehabilitation and a return to the pre-injury functional state. However, pain, decreased muscle strength recovery, LLD after surgery, and complications such as intraoperative fracture, infection, and dislocation have an impact on rehabilitation. Another study reported that the restoration of leg length, acetabular offset, and FO reduces postoperative trochanteric pain and improves clinical outcomes.

The cementless femoral stem has a higher incidence of intraoperative femur fractures compared to the cemented stem. The cemented stem is a good option for avoiding thigh pain and minimizing the risk of intraoperative fracture.¹⁵,¹⁶,¹⁷,¹⁸⁾ Avoiding intraoperative femur fracture can facilitate early bedside exercise and ambulation. LLD is also easily controlled with a cemented stem, due to its combined option for adjusting the depth of stem insertion and head-neck size, regardless of the size difference (37.5 mm vs. 44 mm), with the only requirement being an increase in MO. In our study, there were no instances of thigh pain, intraoperative fracture, or LLD of more than 7 mm in any of the patients.

Recovery of muscle strength by restoring the abductor lever arm after hip arthroplasty is an important factor in rehabilitation and clinical outcomes postoperatively.¹²,¹⁵,¹⁹⁾ Decreased FO is a concern because it may prevent improvement of gait and muscle recovery and decrease the patient’s satisfaction due to limping. The advantage of high offset is that it lateralizes the center of the femoral stem and therefore places greater tension on the abductor mechanism, giving greater stability to the implant. The disadvantage of high offset is that it can place excessive tension on the abductors and cause trochanteric bursitis. In group A, we performed THR using a 44-mm extended offset cemented stem. Five cases with an MO increase of more than 15 mm experienced mild to moderate pain around the greater trochanter, but it was well treated. In our study, although dislocation and major complications were absent, trochanteric pain occurred in some patients. It would be useful to expand on whether the magnitude of offset increase correlated with this complication.

Vertical height can be increased if the stem does not sit properly, referred to as a “potted stem.” Vertical height can be increased if the stem is placed in valgus (where the tip of the stem touches the medial cortex of the femur). In our cases, there was no potted stem and a valgus position was observed in only 13 cases (< 3 mm), which was not related to LLD due to the use of a cemented stem. The cemented stem only increased the MO without altering the vertical height.

Dislocation plays an important role in clinical outcomes by delaying rehabilitation and resulting in muscle weakness. Multiple factors such as surgical approach, patient factors, implant malposition, and abductor muscle weakness are associated with dislocation.²⁰,²¹,²²⁾ The anterior and anterolateral approaches have a lower dislocation rate than a posterior approach.²³⁾ But a concern with the lateral approach is the weakness of the abductor muscles due to possible muscle damage or injury to the superior gluteal nerve. Downing et al.²⁴⁾ reported that there was no difference in hip abductor strength recovery at 3 and 12 months between the lateral approach and the posterior approach. We performed THR using the anterolateral (modified Hardinge) approach, and there was no limping or dislocation. And abductor muscle strength in the operated hip of elderly patients was found to have recovered to more than 80% of that on the native side. Although this difference in strength recovery was statistically significant, further study is needed to determine whether this magnitude of difference translates into a meaningful functional benefit for the patient. We enrolled patients with a Koval mobility score of 3 or greater, ASA physical status classification of class 3 or lower, and a K-MMSE score of 24 or above to minimize the negative effect on rehabilitation after the operation. There was no malposition of the acetabular cup and stem, and a positive abductor lever arm was achieved using an extended FO stem.

Our study has several limitations. First, it is a retrospective, single-center study with a relatively small sample size. Second, we did not present patient-reported outcome measures, such as the Harris Hip Score or Western Ontario and McMaster Universities Osteoarthritis Index, which limits our ability to definitively conclude that radiographic and strength improvements translate directly into superior clinical function. Third, there was heterogeneity in the implants used, particularly the inclusion of dual-mobility cups in some cases, which could serve as a confounding factor regarding joint stability and dislocation rates. Finally, while a statistically significant difference in abductor strength recovery was observed, its direct clinical impact on gait quality remains to be further validated in larger prospective trials.

The goal of treatment for elderly patients with FNF is full rehabilitation and a return to their pre-injury functional state. If they are self-supporting and physically active and have normal cognitive function, THR using the Exeter^® 44-mm offset stem with cement had a positive effect on the recovery of abductor muscle strength, potentially contributing to improved joint stability.