Association Between Bone Mineral Density Around the Stem, Morphology of the Proximal Femur, and Effects of Osteoporosis Treatment in Patients with Femoral Neck Fracture

Abstract

Background/Objectives: The aim of this study was to evaluate changes in bone mineral density (BMD) around the stem in elderly patients with femoral neck fractures who underwent hemiarthroplasty using a collared and full hydroxyapatite coated cementless stem, as assessed using the Dorr classification and with anti-osteoporosis drug intervention. Methods: This study followed 85 older patients with femoral neck fractures classified by Dorr’s classification. We measured their BMD around the stem using dual-energy X-ray absorptiometry according to Gruen 7 zones classification and clinical scores. We compared the rate of BMD change based on Dorr’s classification and clinical scores. We also investigated the effect of osteoporosis treatment interventions on the rate of BMD change. The study followed up with the patients for one year after surgery. Results: After excluding patients with missing data, 40 patients were included in the analysis. The rate of change in BMD in zone 2 was significantly reduced in Dorr type C compared to Dorr type B. Clinical scores did not significantly differ between the three groups. Regarding the association between osteoporosis treatment and the rate of BMD change, the pre-injury intervention group had a significantly suppressed decline in the rate of BMD change in zones 1 and 7 compared to the post-injury intervention and no-intervention groups. Conclusions: Careful follow-up examination is crucial when performing hemiarthroplasty in patients with Dorr type C femoral neck fractures because the rate of BMD change may decline postoperatively. Pre-injury osteoporosis intervention therapy may suppress BMD loss around the stem.

1. Introduction

Hip replacement surgery is the most common treatment for displaced femoral neck fractures in older patients due to the poor results of osteosynthesis. Because of low activity and surgery invasiveness, hemiarthroplasty (bipolar hip arthroplasty: BHA) is frequently performed instead of total hip arthroplasty (THA) [1,2,3]. Poor initial fixation of the stem in surgical treatment affects postoperative pain, loosening and subsidence, and fractures around the stem. Initial fixation is also affected by the stem design and periprosthetic bone mineral density (BMD), particularly for cementless stems [4,5,6]. Therefore, stems are divided into cemented and cementless stems. Cementless stems account for approximately 70% of THA cases [1,7,8,9,10]. The ACTIS stem (Depuy Synthes, Warsaw, IN, USA) is a triple-tapered short stem with a medial collar that provides added stability and has a porous sintered titanium bead coating proximally, a grit-blasted surface distally, and an all-over HA coating to prevent subsidence. This prosthesis has been available since 2016 (Figure 1). Additionally, the implant is inserted using a hybrid compression and cutting broach, which allows the cancellous bone to be compressed towards the cortical edge until axial and rotational stability is achieved. Cancellous bone is maintained without the implant contacting the femoral cortex, and the HA-coated femoral stem is inserted into a sleeve of compressed cancellous bone. Although there are some treatment reports on THA using the ACTIS stem, there are few reports on BHA for femoral neck fractures [11,12,13]. Past reports have shown that the full HA-coated compaction stem has a good reputation as a stem with fewer complications in THA. On the other hand, there are reports that administering osteoporosis medications such as alendronate can prevent the decline in BMD around the stem after THA. However, most of these reports involved administering drugs after surgery [14,15,16,17]. We hypothesized that using the ACTIS stem in BHA for femoral neck fractures will result in reduced BMD loss, regardless of the medullary canal shape, and that administration of anti-osteoporosis drugs can mitigate BMD loss around the stem. As there have been no reports to investigate the BMD around the ACTIS stem for femoral neck fractures in older patients, we investigated changes in BMD around the stem over time, from the medullary canal shape to the osteoporosis treatment intervention. We also examined differences depending on the timing of the intervention with anti-osteoporosis treatment.

2. Materials and Methods

2.1. Study Design

This multicenter clinical trial began on 28 November 2018 (when the first patient provided informed consent). This is a prospective longitudinal study without a control group. It ended on 28 February 2021 (the day when the 1-year follow-up period of the last patient was completed). Eighty-five patients from 12 institutions were enrolled in this study. Of these, two patients died, and 22 withdrew due to various factors (10, 2, 2, and 8 patients were COVID-19 related events, withdrew their consent, were affected by new diseases, and withdrew their participation for unknown reasons, respectively). Data were missing for 21 out of the follow-up patients; therefore, data from 40 patients were included in the final analysis (Figure 2). Patients aged >65 years with displaced femoral neck fractures (Garden stages 3 and 4) and mobility before fracture were included. Those with an inability to walk, previous hip fractures, rheumatism, or dementia were excluded because these may affect BMD changes and postoperative movement recovery. Additionally, demographic factors (i.e., age, body mass index [BMI], and sex) that may have influenced the BMD and clinical outcomes were investigated.

2.2. Sample Size Estimation

A power analysis was performed to plan the study, aiming to collect 100 samples in total, with at least 10 samples in each group. However, this did not work out as expected for the reasons shown in Figure 2.

2.3. Clinical and Radiographical Assessment

Based on radiographic findings, patients were divided into types A, B, and C according to the Dorr classification. Those with a narrow medullary canal size were classified as Type A, those with a normal size as Type B, and those with a wide medullary canal as Type C [18] (Figure 3). The surgical approach was posterior in all cases. The same stem model was used, but the size was selected by the surgeon for each case depending on the shape of the medullary canal (

Table 1. This is a list of selected stem sizes.

Stem Size	Number
1	1
2	3
3	9
4	8
5	8
6	7
7	3
8	1

). Stem sizes from 3 to 6 were commonly used. There was no set protocol for rehabilitation. Evaluations were conducted using radiographics (evaluation of implant loosening, subsidence, and fracture), dual-energy X-ray absorptiometry (DEXA), and clinical scores (Harris Hip Score [HHS] and EuroQoL 5-Dimension 5-Level [EQ-5D-5L]) [19,20,21]. Follow-up was at three points: at the time of discharge (within 1 month after surgery), 6 months, and 12 months.

2.4. DEXA Measurements

Seven Gruen zones were used to determine the BMD around each patient’s stem [22] using DEXA (Figure 4). Measurements were taken within 1, 6, and 12 months postoperatively. DEXA is a well-documented method to monitor bone quality after THA and BHA and detects even small changes in BMD around the stem. We referred to past reports for determining the timing of the DEXA exam [23,24].

The DEXA models used were GE (Lunar and Prodigy, Madison, WI, USA) and Hologic (Horizon, Explorer, and Discovery, Marlborough, MA, USA), each within six hospitals. Patients were positioned supine with their leg in a neutral position, with knee and foot support to allow scanning of the anterior–posterior projection of the proximal femur, including the area distal to the prosthesis, using an edge-detection technique.

Postoperative peri-prosthetic BMD was determined in the Gruen zone in seven regions of interest (ROIs) [22]. The seven Gruen zones were positioned on the distal tip and shoulder of the prosthesis. The two most proximal Gruen zones (1 and 7) were also combined to form a proximal femur ROI, representing the region where the porous coating of the uncemented stems is typically located. The BMD area was expressed as grams per square centimeter (g/cm²). A technician from the radiology department assessed the BMD around the stem at 1 month postoperatively (baseline BMD) and 6 and 12 months later. The BMD ratio was calculated by dividing the BMD values at 6 and 12 months by the BMD values obtained within 1 month after surgery. If the BMD change is more than 1, it indicates BMD increased, if less than 1, it indicates BMD decreased.

2.5. Osteoporosis Treatment

Based on previous reports that administration of anti-osteoporotic drugs after THA surgery suppresses BMD loss around the stem, we investigated the number of cases where osteoporosis treatment intervention was performed after femoral neck fracture (post-injury intervention group), where treatment intervention was performed before femoral neck fracture (pre-injury intervention group), and where no osteoporosis treatment was performed (no-intervention group) [14,15,16,17]. At the time of the final survey, we also investigated osteoporosis medications.

2.6. Statistical Analysis

The Kruskal–Wallis test was used to compare the change rate in the BMD around the stem over time and clinical scores in patients according to the Dorr classification and osteoporosis therapy intervention (Table 1 and

Table 2. Patient demographics were assessed using the Dorr classification and the association between the Dorr classification and BMD change rate.

		Dorr A	Dorr B	Dorr C	p Value	X2 Value
Number of Patients		6	24	10	-	-
Female, n (%)		3 (50%)	16 (66.7%)	9 (90%)	0.204	3.175
Age (years)		81.8 ± 4.5	80.0 ± 6.6	80.1 ± 7.3	0.173	-
BMI (kg/m2)		21.9 ± 3.1	20.8 ± 2.7	19.9 ± 1.8	0.338	-
HHS (points)		93.5 ± 6.0	81.9 ± 15.6	75.9 ± 19.0	0.087	-
EQ-5D-5L (score)		0.88 ± 0.11	0.76 ± 0.21	0.75 ± 0.18	0.287	-
Zone area	Time	BMD change rate
Zone 1	6 m	0.90 ± 0.09	0.89 ± 0.11	0.88 ± 0.12	0.218	-
	12 m	0.86 ± 0.12	0.86 ± 0.15	0.85 ± 0.13	0.943	-
Zone 2	6 m	0.91 ± 0.05	0.90 ± 0.12	0.83 ± 0.15	0.200	-
	12 m	0.88 ± 0.07	0.88 ± 0.16 *	0.77 ± 0.09 *	0.036 *	-
Zone 3	6 m	0.98 ± 0.03	0.96 ± 0.06	0.96 ± 0.09	0.673	-
	12 m	0.96 ± 0.02	0.95 ± 0.08	0.97 ± 0.07	0.771	-
Zone 4	6 m	0.99 ± 0.02	0.98 ± 0.05	0.97 ± 0.05	0.769	-
	12 m	0.98 ± 0.02	0.97 ± 0.05	0.97 ± 0.06	0.985	-
Zone 5	6 m	0.98 ± 0.03	0.97 ± 0.08	0.97 ± 0.08	0.948	-
	12 m	0.96 ± 0.03	0.92 ± 0.21	0.98 ± 0.09	0.429	-
Zone 6	6 m	0.91 ± 0.08	0.91 ± 0.12	0.88 ± 0.15	0.911	-
	12 m	0.90 ± 0.13	0.94 ± 0.10	0.82 ± 0.15	0.169	-
Zone 7	6 m	0.86 ± 0.15	0.78 ± 0.13	0.71 ± 0.16	0.298	-
	12 m	0.86 ± 0.21	0.80 ± 0.16	0.67 ± 0.10	0.113	-

* p < 0.05. Dorr A, B, and C were compared and the groups with significant differences were marked with *.

). The chi-square test was used to examine the Dorr classification, sex, and osteoporosis therapy intervention (Table 1 and Table 2). All statistical analyses were performed using IBM SPSS (IBM, Armonk, NY, USA) version 26 for Windows. A p-value < 0.05 was considered statistically significant.

2.7. Ethical Statement

This study was conducted in accordance with the principles embodied in the Declaration of Helsinki and ICMJE. It was approved by the institutional review boards of the 12 participating hospitals on 23 October 2018 (approval no. 180206). Written informed consent was obtained from all participants.

3. Results

The mean patient age at the time of operation was 80.3 ± 6.3 years. The mean BMI was 20.8 ± 2.5 kg/m². The HHS and EQ-5D-5L were 82.3 ± 15.8 and 0.78 ± 0.19 points, respectively. There were 6, 24, and 10 patients in the Dorr type A, B, and C groups. Table 1 shows patient backgrounds as assessed using the Dorr classification.

At 12 months, there was a significant difference in the BMD change rate at zone 2 between the Dorr B and C groups. The BMD change rate was significantly lower in Dorr C than in the Dorr B group (p = 0.036). There were no significant differences in any zone between the Dorr A and B groups and between the Dorr A and C groups. There was no significant difference between the three groups (Dorr A, B, and C) regarding HHS and EQ-5D-5L (Table 2).

Regarding osteoporosis treatment, there were 6, 19, and 15 patients in the pre-injury, post-injury, and no-intervention groups, respectively. There was no significant difference between the three groups in age, BMI, HHS, and EQ-5D-5L. Regarding the association between the osteoporosis treatment intervention and BMD, there was a significant difference between the pre-injury and no-intervention groups at 12 months in zones 1, 6, and 7. The decline in the BMD change rate was significantly suppressed in the pre-injury intervention group compared to the no-intervention group (between-group comparison at 12 months in zone 1: p = 0.039, at 6 months in zone 7: p = 0.005, and 12 months in zone 7: p = 0.029). Moreover, at 6 and 12 months in zone 7, there was a significant difference between the pre-injury and post-injury intervention groups. The decline in the BMD change rate was significantly suppressed in the pre-injury intervention group than in the post-injury intervention group (between-group comparison at 6 months in zone 7: p = 0.003 and 12 months in zone 7: p = 0.004). However, there was no significant difference in the BMD change rate between the post-injury and no-intervention groups (

Table 3. The patient background is classified based on osteoporosis therapy intervention and the association between the osteoporosis treatment intervention and BMD change rate.

		Pre-Injury Intervention	Post-Injury Intervention	No Intervention	p Value	X2 Value
Number of Patients		6	19	15	-	-
Dorr type		A, 2; B, 3; C, 1	A, 1; B, 13; C, 5	A, 3; B, 8; C, 4	0.523	-
Female, n (%)		5 (83.3%)	13 (68.4%)	10 (66.7%)	0.737	0.610
Age (years)		84.2 ± 3.3	79.9 ± 6.8	79.2 ± 6.6	0.129	-
BMI (kg/m2)		21.1 ± 2.1	21.5 ± 2.7	19.7 ± 2.3	0.902	-
HHS (points)		83.7 ± 16.3	79.6 ± 18.3	85.2 ± 12.9	0.832	-
EQ-5D-5L (score)		0.82 ± 0.18	0.77 ± 0.22	0.77 ± 0.16	0.743	-
Zone area	Time	BMD change rate
Zone 1	6 m	0.94 ± 0.10	0.86 ± 0.11	0.87 ± 0.11	0.351	-
	12 m	0.95 ± 0.11 +	0.88 ± 0.13	0.80 ± 0.14 +	0.039 +	-
Zone 2	6 m	0.93 ± 0.08	0.86 ± 0.16	0.89 ± 0.07	0.699	-
	12 m	0.92 ± 0.09	0.86 ± 0.18	0.83 ± 0.10	0.357	-
Zone 3	6 m	1.01 ± 0.04	0.95 ± 0.06	0.97 ± 0.06	0.920	-
	12 m	0.98 ± 0.03	0.96 ± 0.09	0.94 ± 0.05	0.336	-
Zone 4	6 m	1.00 ± 0.04	0.97 ± 0.04	0.98 ± 0.05	0.446	-
	12 m	0.99 ± 0.02	0.97 ± 0.05	0.97 ± 0.04	0.552	-
Zone 5	6 m	0.98 ± 0.08	0.97 ± 0.07	0.96 ± 0.08	0.974	-
	12 m	0.97 ± 0.04	0.96 ± 0.08	0.89 ± 0.26	0.700	-
Zone 6	6 m	0.98 ± 0.10	0.90 ± 0.13	0.87 ± 0.10	0.076	-
	12 m	0.94 ± 0.10	0.92 ± 0.15	0.89 ± 0.09	0.387	-
Zone 7	6 m	0.94 ± 0.09 #+	0.73 ± 0.14 #	0.76 ± 0.12 +	0.003 #, 0.005 +	-
	12 m	0.93 ± 0.11 #+	0.74 ± 0.19 #	0.78 ± 0.14 +	0.004 #, 0.029 +	-

⁺p < 0.05, ^# p < 0.05. Pre-injury intervention of anti-osteoporotic drugs, post-injury intervention, and no intervention were compared and the groups with significant differences were marked with ⁺ and ^#.

).

Table 4. Type and number of anti-osteoporotic drug.

Type of Drug	Number of Patients (n = 39)
Vitamin D	17 (44%)
Bisphosphonate	16 (41%)
Denosumab	3 (8%)
Teriparatide	2 (5%)
SERM	1 (2%)

shows the type of anti-osteoporotic drugs administered. Vitamin D and bisphosphonate preparations were frequently used, accounting for over 80% of the total.

Regarding radiographic evaluation, there were no peri-implant fractures, evident stem subsidence, or major complications during the follow-up period in the 40 patients analyzed.

• Case presentation

A 79-year-old woman fell and sustained a displaced left femoral neck fracture. The medullary canal shape was Dorr C. BHA was performed the day after the injury. The radiograph immediately after surgery showed that a slightly smaller stem was selected. There was no significant subsidence, but the radiograph taken 12 months later showed stress shielding around the stem. Stress shielding was observed especially in Gruen zone 2 and 7 (Figure 5). Compared to immediately after surgery, 12 months later, there was a decrease in BMD of 18.6% in zone 2 and 32.7% in zone 7.

4. Discussion

In this study, we evaluated older patients with femoral neck fractures who underwent BHA using a full hydroxyapatite-collared compaction short stem according to Dorr classification to assess the change in BMD around the stem. The present results differ from our hypothesis that BMD around the stem can be maintained without being affected by the shape of the medullary canal, even in femoral neck fractures in older adults. BMD around the stem was preserved regardless of the medullary canal shape in a case series of THA with ACTIS stems, which included patients with hip osteoarthritis undergoing THA who were relatively young (mean age, 65 years) [11]. Our cohort of patients with femoral neck fractures was older (mean age, 80 years); thus, advanced osteoporosis was predicted, with a more significant decrease in BMD over time, especially in Dorr type C cases and cases with a wider medullary canal.

In this study, the significant BMD loss in Dorr C zone 2 can be attributed to a mismatch between the stem size and the medullary canal in patients with a wide medullary canal, such as Dorr type C cases. We believe surgeons may opt for a smaller stem size due to concerns about intraoperative femoral fractures (case presentation). According to the stem concept, this may have prevented load transfer, resulting in proximal BMD loss. In this study, fortunately, there were no cases of peri-stem fracture or implant subsidence in patients with Dorr type C; nonetheless, careful follow-up is necessary. We believe that BMD loss is less of a problem in Dorr A and B, but for medullary canals of Dorr type C, a cemented stem may be considered because BMD loss may progress.

Regarding the second hypothesis that anti-osteoporotic drug interventions can suppress BMD loss around the stem, there was a significant difference between the pre-injury and no-intervention groups and the pre-injury and post-injury intervention groups. The use of anti-osteoporotic drugs and changes in BMD around the stem were critical. Patients who took medication before the femoral neck fracture had suppressed BMD loss around the stem. When older people are diagnosed with osteoporosis based on DEXA scan results and previous fragility fractures, proactively intervening with anti-osteoporotic drugs may prevent secondary fractures and limit BMD reduction around the stem. It is possible to suppress the decrease in BMD around the stem, implying that postoperative periprosthetic fractures and stem subsidence and loosening can be prevented. Some studies have revealed that using anti-osteoporotic drugs after THA suppressed the decreased BMD around the stem [14,15,16,17]. However, no reports on drug intervention for osteoporosis before the injury and BMD around the stem have been published. We believe that proactively diagnosing osteoporosis and providing treatment will not only prevent femoral neck fractures but will also lead to reduced bone density around the stem after BHA.

Regarding clinical scores, Hayashi et al. found a correlation between daily activity and periprosthetic BMD in THAs [25]. In contrast, Murat et al. found no correlation between the clinical scores and peri-stem BMD in femoral neck fractures [26]. Regarding our clinical scores studied in femoral neck fractures, no significant differences were found among the three medullary canal shapes, suggesting that the shape of the medullary canal did not affect the clinical scores during our follow-up period. This may be due to the older age of the neck fracture patients compared to the age of the THA patients. In addition, the short follow-up period of this study (12 months) may have also influenced the results.

This study had some limitations. First, the sample size was small. Before this study, we performed a power analysis; at least 10 cases in each group were needed. However, the statistical power was insufficient because of dropout cases and missing data. Second, the postoperative follow-up period was only one year. However, this study was conducted in an older population, so long-term follow-up was difficult for various reasons. Third, patient data could not be collected for one-quarter of the study participants. Fourth, only one type of stem was investigated; thus, cemented or other cementless stems could have affected our study results, as it is known that stem design affects bone density around the stem [27]. Fifth, we could not investigate the rate of change in BMD by type of osteoporosis medication. Finally, the duration of osteoporosis treatment prior to injury was not investigated. To confirm the changes in BMD around the stem and the outcomes of anti-osteoporotic drugs in BHA patients, a comparative study with a longer follow-up period with a larger number of participants and a different stem design is desirable.

5. Conclusions

The BMD change rate in zone 2 was significantly reduced in Dorr type C compared to Dorr type B. Regarding the association between osteoporosis treatment and the BMD change rate, the pre-injury intervention group had significantly suppressed BMD change rate decline relative to the post-injury intervention and no-intervention groups in zones 1 and 7. A careful follow-up examination is essential when performing BHA in older patients with femoral neck fractures of Dorr type C because BMD is likely to decline postoperatively in the proximal area. It has been proposed that administering anti-osteoporotic drugs before femoral neck fracture can suppress the loss of BMD around the stem.