Is bone mineral density a risk factor for femoral stem subsidence in primary cementless total hip arthroplasty: a prospective study

Abstract

Background

The success of primary total hip arthroplasty (THA) in patiens with primary osteoarthritis (OA) of the hip is significantly influenced by stem osseointegration, migration and implantation. Additionally, the bone mineral density (BMD) of the proximal femur, the femoral morphology and stem design play important roles in these outcomes. This study aimed to evaluate the effect of BMD on subsidence who underwent primary cementless THA with a fit-and-fill stem design for primary OA of the hip, assuming that subsidence is not influenced by BMD.

Methods

This prospective study included 72 hips from 67 patients who underwent primary cementless THA for OA between May 2019 and March 2022. Two patients were excluded due to intraoperative complications, and the final analysis was conducted on 70 hips. Patients were grouped according to preoperative BMD and proximal femur anatomy based on Dorr classification. Harris Hip Score (HHS), radiological subsidence and Engh score values ​​were used for postoperative clinical evaluation.

Results

The mean age was 65.99 ± 8.10 years (range 52 to 83 years). The mean follow-up was 24.6 ± 5.8 months (range, 12 to 34 months). Amongst 70 OA of the hip cases, 22 (31.43%) were Dorr type A, 44 (62.86%) type B, and 4 (5.71%) type C. Forty-three patients (61.43%) had normal BMD, while 27 patients (38.57%) were osteopenic or osteoporotic. The mean axial migration at the latest follow-up was 1.04 ± 0.69 mm (range 0 to 3.1 mm). Total Engh score was 22.89 ± 3.28 (range 14 to 27), and in terms of stabilization bone ingrowth was seen in all of the femoral stems. There was no statistically significant effect of femoral T-score levels (p = 0.853) or femoral morphologies (p = 0.763) on femoral stem subsidence.

Conclusions

No statistically significant differences were observed in clinical or radiological outcomes between patients with normal and low BMD. These findings suggest that early stem migration and osseointegration are more closely related to initial mechanical stability than to BMD. Cementless femoral stems can therefore be considered reliable across different bone qualities, offering an innovative perspective that challenges the conventional emphasis on bone density in implant selection.

Introduction

The success and worldwide acceptance of total hip arthroplasty (THA) began with the use of polymethylmethacrylate (PMMA) cement, also known as acrylic, for low-friction arthroplasty and component fixation, developed and popularized by Sir John Charnley [1]. The emergence of problems with the use of acrylic cement led to the elimination of its use and the preference for cementless THA for biological fixation. Indeed, PMMA or bone cement, used in surgical procedures may lead to severe symptoms such as hypoxemia, hypotension, and unexpected loss of consciousness, which physicians are increasingly recognizing as bone cement implantation syndrome [2]. The bone quality of the proximal femur is important when the surgeon chooses cementless fixation in THA, as the stability of the cementless femoral component is achieved through bone growth into the pores of the component [3].

With increasing life expectancy, the number of THA procedures performed in older patients is also growing. Although many theoretical problems arise in hip arthroplasty with advancing age, the greatest concern is decreasing bone density [4]. Despite the fact that, cementless techniques are designed for individuals with normal bone density and healing capacity in THA applications, it is possible to use cementless components in patients with impaired bone quality and limited healing capacity via improved implant designs and bioactive coating materials [5].

In patients with primary osteoarthritis (OA) of the hip undergoing primary THA, stem insertion technique, stem osseointegration and stem migration have a significant effect on long-term functional results and pain. Although many factors affect bone and stem compatibility, proximal femur bone mineral density (BMD) and femoral morphology also significantly influence these variables [6]. Considering that early implant stability has a crucial role in the success of cementless THA, low BMD and age-related changes in proximal femoral anatomy may compromise femoral stem stabilisation and osseointegration [5].

Subsidence in the femoral stem is expected to some extent in the early stages of THA. Due to the use of cement in the initial stabilization, cemented femoral stems, compared to cementless stems, have less subsidence. However, it has been stated that both methods have similar results in the long-term follow-ups of THA [7]. In cementless hip arthroplasty, the fixation of the stem depends on osseointegration, and it is believed that this osseointegration depends on the bone density around the stem [8].

OA and osteoporosis are orthopedic conditions whose prevalence increases with advancing age. It is accepted that there is an inverse relationship between these two diseases; it has been suggested that a person with OA will be protected from osteoporosis [9]. Regarding THA application in patients with osteoporosis there are several important issues, especially initial osseointegration, intraoperative fractures, late periprosthetic fractures, and long-term survival of the implant. The use of cemented or cementless femoral stems in patients with low BMD is still controversial [10].

There are very few studies comparing local BMD and distal migration of femoral stems after cementless THA and the results are heterogeneous and still controversial [11, 12]. Some studies show that lower femoral BMD leads to increased subsidence, while other studies cannot demonstrate such a relationship [8]. Additionally femoral stem type and design as well as anatomical properties also may play a relevant role. Initial fixation of stem gains more importance on femoral stem subsidence rather than the bone mineral density [12–14]. Most of the studies included tapered femoral stems making metaphyseal fixation [12, 13]. As far as we know this is the first study with the use of a collarless fit-and-fill stem making metaphyseal and proximal diaphseal fixation. Thus we aimed to evaluate the effect of BMD, femoral morphology and stem design on clinical and radiological outcomes in patients who underwent primary cementless THA with a fit-and-fill stem design for primary hip OA.

Materials and methods

This prospective study included 72 hips from 67 patients who underwent primary cementless THA due to primary osteoarthritis of the hip between May 2019 and March 2022. Two patients were excluded from follow-up and analysis due to intraoperative complications, and the final evaluation was conducted on 70 hips. Preoperative BMD measurements of the lumbar spine and proximal femur were obtained in all eligible patients using dual-energy X-ray absorptiometry (DXA). Patients were categorized based on BMD (normal, osteopenic, or osteoporotic) and proximal femoral morphology according to the Dorr classification. Postoperative outcomes were assessed using the HHS, radiographic measurements of femoral stem subsidence, and the Engh score to evaluate osseointegration.

Ethical approval for this prospectively designed study was obtained from the Ethics Committee of Istanbul Training and Research Hospital (date: 12/04/2019, decision no: 1790). All procedures were conducted in accordance with the ethical standards of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to inclusion.

Inclusion and exclusion criteria

Patients aged between 50 and 85 years who were diagnosed with primary osteoarthritis of the hip, scheduled for primary total hip arthroplasty, and had no history of treatment related to osteoporosis or bone metabolism disorders were included. Only patients who were not receiving any pharmacological or non-pharmacological osteoporosis treatment at the time of enrollment were considered eligible. Volunteer patients who agreed to participate in the study and provided written informed consent were enrolled.

Patients were excluded if they had a history of previous hip surgery, severe deformity, revision hip arthroplasty, severe osteoporosis requiring or undergoing medical treatment, or if they had received or were currently receiving treatment affecting BMD. Additionally, patients with hereditary skeletal disorders, those receiving osteoporosis therapy and those who did not attend routine check-ups were also excluded.

Surgical procedure

All patients underwent the operation under either spinal or general anesthesia, based on the anesthesiologist’s preference and the patient’s health condition, following standard surgical antibiotic prophylaxis. All patients underwent surgery in the lateral decubitus position using a posterolateral approach performed by one experienced senior surgeon. Although perioperative template measurements were made, the stem size was left to the surgeon’s discretion during the case. Perioperative radiographic images were not routinely taken. The internal landmarks used during the free-hand placement of the acetabular component included the ischium, ilium, pubis and transverse acetabular ligament, with placement performed in proper anteversion and inclination. The largest femoral stem size that provided sufficient bone support and appropriate rotational stability was chosen for each patient. The surgeon made intraoperative decisions regarding the suitable femoral head and liner to ensure joint stability. For all stems, the goal was to achieve metaphyseal and proximal diaphyseal fixation. The same acetabular (Smith & Nephew EP-HIT PLUS^®) and femoral (Smith & Nephew SYNERGY^®) components were implanted in all patients.

All patients received the same postoperative analgesia and antibiotic prophylaxis. Venous thromboembolism prophylaxis was administered based on risk assessment, using either acetylsalicylic acid or low-molecular-weight heparin.

All patients have received the same rehabilitation program determined for the study, which included early mobilization with a walker promptly, allowing weight-bearing as tolerated, along with range of motion and isometric exercises. Patients were mobilized with full weight-bearing on the first postoperative day, and the rest of the rehabilitation program was initiated accordingly.

Radiographic measurements

Bone density was measured only within three months prior to surgery in the proximal femur, femoral neck on the side to be operated, and lumbar vertebrae. Measurements were performed using a Hologic Discovery QDR A device (Hologic, Bedford, MA, USA). Based on the World Health Organization (WHO) criteria, patients were classified as either having normal bone density or low bone density (osteopenic/osteoporotic; T scores < -1) according to the lowest measurements. Bone density measurements of the proximal femoral region, covering approximately 25 cm of the proximal femur and defined as the ‘total femoral T-score’ (Horizon DXA System; Hologic^®), included the femoral neck, greater trochanter, intertrochanteric and subtrochanteric areas. These measurements were calculated based on the patients’ body mass index (BMI) and age data, as described in the literature [15].

Subsidence is defined as a femoral stem distalization in reference to the greater trochanter [13]. Antero–posterior (AP) radiographs of the pelvis were taken in a standardized standing position with the centralized beam focused on the symphysis. Postoperative radiological evaluations were performed on the 2nd to 3rd day after surgery, and patients were scheduled for follow-ups at 3, 6, 12, and 24 months. Subsidence measurements were conducted based on the radiographs obtained during the final follow-up. All radiographs were meticulously examined using the Picture Archiving and Communication System (PACS) provided by the institutional hospital information management system (PROBEL^®, İzmir, Turkey). To enable accurate measurement of femoral stem subsidence, a distance calibration was performed using the implanted femoral head as a radiographic reference. The known actual diameter of the implanted femoral head was divided by its measured length on radiographs obtained using the same X-ray device. This ratio was then applied to convert radiographic measurements into true values of femoral stem subsidence. The measurements were performed by two independent researchers (orthopedic surgeons, TSB and SA) after assessing the suitability of the pelvic AP radiographs. All measurements were made by two orthopedic surgeons who were blinded to the demographic and group information of the patients.

Parameters were measured according to the specified measurement methods:

1. The distance in ‘mm’ was measured using parallel lines passing through the greater trochanter and the femoral stem, matching femoral head size distance, the subsidence value was determined as ‘mm’ [16] (Figure-1).

2. The endosteal points at 3 cm and 10 cm distally were determined using the mid-lesser trochanteric line as a reference, as defined by Dorr et al. By comparing these lines to each other, the canal calcar ratio was found and classification was made according to these values [17] (Figure-2).

3. Osseointegration of the femoral component was evaluated using the comprehensive radiographic scoring system defined by Engh et al. [18]. Fixation was assessed based on the appearance of a porous interface and the presence of spot welds. Stability was evaluated through radiographic signs such as smooth interface appearance, pedestal formation, calcar modeling, interface deterioration, stem migration, and particle shedding. Fixation, stability and total scores were evaluated, and according to this scoring system, bone ingrowth for the prosthesis ( > + 10), suspicious bone ingrowth (0 to + 10), inadequate but stable (-10 to 0), and unstable (<-10) were defined as the 4 classes.

4. True pelvic AP radiographs of the patients in the preoperative period, early postoperative period and last follow-up were evaluated, and the ‘teardrop landmarks’, ‘ischial tuberosities’, and ‘lesser trochanters’ determined on the radiographs were used as guides and leg length differences were calculated [19] .

Fig. 1.

Determination of subsidence distance a: postoperative and b: at follow up. The femoral stem axis (A) and the passing perpendicularly through the apex of greater trochanter B. The distance calibration has been performed based on the ratio of the ‘y’ values, and subsidence is defined as the change in the ‘x’ distances

Fig. 2.

Measurement of Canal calcar ratio (AA/CC)

The interobserver reliability of radiological parameters was evaluated using the intraclass correlation coefficient (ICC), based on a two-way random-effects model with absolute agreement (ICC(2,1)). Other coefficients such as kappa, Spearman rho, or Pearson were not used. ICC values ranged from 0.81 to 0.93 in this study, which were interpreted as indicating good to excellent reliability.

Outcome measures

Patients were evaluated in terms of pain, function, joint range of motion, and deformity with the ‘modified HHS,’ adapted to their native language, before surgery and at their last follow-up [20]. According to this scoring system, a score below 70 out of 100 is considered poor result, between 70 and 79 points is considered a fair result, between 80 and 89 points is considered a good result, and between 90 and 100 points is considered an excellent result [21].

Pain intensity was assessed using the visual analog scale (VAS), administered in the patients’ native language, both preoperatively and at the final postoperative follow-up. The scores were used to compare changes in pain levels over time.

Statistical analysis

Statistical analyses were carried out with the help of IBM SPSS Statistics for Windows, Version 23.0 (Released 2015; IBM Corp., Armonk, New York, United States). The suitability of the variables to normal distribution was examined using histogram graphics and the Kolmogorov-Smirnov/Shapiro-Wilk test. Mean, standard deviation, median, minimum and maximum values ​​were used when presenting descriptive analyses. Mann-Whitney U Test was used when evaluating non-normally distributed (nonparametric) variables between two groups, and Kruskal-Wallis Test was used when evaluating between more than two groups. While investigating the reason for the significant difference between the groups, Bonferroni multiple comparison test was used. While presenting categorical variables, the frequency and percentage values ​​of the variables were used and the analysis of categorical variables was carried out with the Chi-square (Exact) Test. Spearman Correlation test was used to evaluate the relationships between quantitative variables. Situations where the ‘p’ value was below 0.05 were considered statistically significant results.

Results

In our study, 65 patients were included, with 5 patients undergoing bilateral hip arthroplasty in different sessions, resulting in a total of 70 hips (Fig. 3). One patient experienced a fracture of the greater trochanter, and another sustained a distal femur fracture during the procedure. These fractures were treated in the same session. However, these two patients were excluded from follow-up because the established postoperative treatment and rehabilitation protocols could not be applied. Therefore, although a total of 72 hips were initially operated on, only 70 hips were included in the follow-up and subsequent statistical analysis.

Fig. 3.

Flow diagram of patient recruitment

Among the patients, 44 (62.86%) were female and 26 (37.14%) were male. The mean age of the patients was 65.99 ± 8.10 years (range 52 to 83 years). The mean follow-up was 24.6 ± 5.8 months (range, 12 to 34 months) for patients operated between May 2019 and March 2022. 40 (57.14%) of the operated hips were on the left side and 30 (42.86%) were on the right side (Table 1).

Table 1.

Descriptive statistics of demographic characteristics

		Min-Max	Median	Av.	±s.s.
Age		52-83	67	65.99	± 8.10
Gender, n (%)	Female			44(62.86)
	Male			26(37.14)
Follow-up period (Month)		12-34	24.5	24.62	± 5.81
Operated leg	Right			30	(42.86)
	Left			40	(57.14)

The ICC was determined to be 0.82 for subsidence measurements and 0.76 for Engh score assessments. During the follow-up periods, no stem revisions were performed in any of the patients. There is no statistically significant difference in clinical scores between males and females groups except stem subsidence. Subsidence values were found to be statistically significantly better in female group (p = 0.047) (Table 2).

Table 2.

Comparison of clinical scores of patients by gender

	Male			Female
	Mean	SD	Max-Min	Mean	SD	Max-Min	p
Subsidende(mm)	1.25	0.86	3.10–0.00	0.91	0.54	1.80–0.00	0.047*µ
Engh score	23.7	3.5	27.0–14.0	22.3	3.0	27.0–16.5	0.088¥
Harris Hip Score postoperative	94.03	5.05	100.0 -83.85	91.63	6.30	100.0–75.0	0.126¥
VAS score postoperative	0.6	1.0	2.0–0.00	1.0	1.2	4.0–0.00	0.204¥

¥: Mann Whitney U test

µ: Independent sample t test

Patients were classified according to T scores and Dorr classification (Table 3).

Table 3.

Descriptive statistics of clinical characteristics

		Min	Max	Medyan	Ort.	±s.s.
Femoral neck T-score		-5.2	5.2	-0.9	-0.42	± 1.93
Total femoral T-score		-2.5	2.6	-0.0	-0.23	± 1.28
T-score(Lowest value)		-6	0.7	-1.95	-2.00	± 1.27
Dorr type	A				22	(31.43)
	B				44	(62.86)
	C				4	(5.71)
Femoral neck T-score	Osteopenic/ osteoporotic				33	(47.14)
	Normal Bone Density				37	(52.86)
Total femoral T-score	Osteopenic/ osteoporotic				27	(38.57)
	Normal Bone Density				43	(61.43)
T-score (Lowest value)	Osteoporotic				26	(37.14)
	Osteopenic				30	(42.86)
	Normal Bone Density				14	(20.00)

When examining whether clinical measurements of patients differ according to femoral T-score levels, no statistically significant difference was found between the measurement levels for low and normal bone density (p = 0.853) (Table 4).

Table 4.

Comparison of clinical measurement scores in terms of T score

	Total femoral T score						p
	Low bone density			Normal bone density
	Av.	±s.s.	Median	Av.	±s.s.	Median
VAS score preop	8.74	± 0.98	8.00	8.70	± 0.96	8.00	0.751
VAS score postop	0.72	± 0.98	0.00	0.98	± 1.18	0.00	0.157
Harris Hip Score preop	42.86	± 7.95	42.60	43.57	± 10.13	43.40	0.795
Harris Hip Score postop	93.4	± 5.44	93.85	92.42	± 6.22	93.80	0.745
Subsidence(mm)	1.08	± 0.73	1.10	1.02	± 0.69	1.10	0.853
Engh score(fixation)	7.00	± 3.06	5.00	6.28	± 2.91	5.00	0.360
Engh score(stability)	16.02	± 2.37	17.00	16.65	± 1.31	17.00	0.427
Engh score(Total)	22.82	± 3.99	22.00	22.91	± 2.89	22.00	0.832

Evaluating whether clinical scores and subsidence values differ based on femoral neck T-score and the lowest T-score levels revealed that the results were statistically similar between the low and normal bone density groups (p = 0.758).

Since the number of patients in the Dorr C group was insufficient for statistical evaluation, subsidence measurement levels were assessed between the Dorr A and Dorr B groups. No statistically significant difference was found between the subsidence measurement levels of Dorr B type patients (mean 1.10 ± 0.66) and those of Dorr A type patients (mean 0.81 ± 0.69) (p = 0.763) .

Relationships between subsidence and Engh scores with T-scores were investigated; however, no statistically significant correlation was found (p = 0.926) .

To identify potential predictors of femoral stem subsidence, both logistic regression and robust regression analyses were performed. Independent variables included Dorr classification, Engh score, total femoral T-score, preoperative HHS, stem size, patient age and sex.

In the logistic regression analysis, a higher Engh stability score was significantly associated with decreased subsidence (β = − 0.0575, p = 0.018). Other variables, including T-score, femoral morphology, age, preoperative functional status, and stem size, did not show statistically significant associations in this model.

To verify the robustness of these findings and to mitigate the influence of outliers, a robust regression analysis using Huber’s M-estimation was conducted. Results were consistent with the logistic regression model. Engh stability score remained significantly and inversely associated with subsidence (β = − 0.171, p < 0.001). Additionally, a significant negative association was observed between preoperative Harris Hip Score and subsidence (β = − 0.0194, p = 0.045).

Discussion

No statistically significant relationship was found between subsidence and femoral BMD in this study. From our perspective, initial stability is a more crucial factor than bone density for successful osseointegration, as it is for preventing stem subsidence. Additionally, our findings suggest no significant association between BMD and HHS or VAS scores for clinical outcomes. However, in the literature, a study evaluating the outcomes of THA in osteoporotic patients, irrespective of cement use in femoral stem fixation, found that patients with low BMD had lower clinical scores. Although inadequate osseointegration was suggested as a possible reason in this study, no evaluation of osseointegration was conducted [10].

Osteoporosis is common and insufficiently treated in patients with primary OA assessed prior to arthroplasty [10]. Furthermore, it is still unclear whether BMD measurement with DXA is helpful in the decision-making process for choosing between cemented and uncemented total joint arthroplasty. In fact, this decision is more significantly influenced by factors such as age, bone geometry, and the Dorr classification of the proximal femur in standard radiographs. While specific patient groups may be better suited for either cemented or uncemented options, studies have shown that survival rates for both cemented and uncemented hip and knee arthroplasties are generally comparable in patients with primary OA. Nevertheless, the broader connections between bone health -especially BMD- and the choice between cemented and uncemented total joint arthroplasty remain unclear [22].

Cementless femoral components were used to reduce cement-related complications such as loosening and osteolysis when cementing techniques were inadequate. However, neither the use of cementless femoral components nor the development of cementing techniques have been fully successful to solve these problems [23]. Throughout historical development, implant designs and fixation preferences have varied greatly among countries, and no clear consensus has been reached regarding these discussions [24].

In cemented femoral stems, less subsidence is expected initially due to the use of cement for initial stabilization compared to cementless stems; however, it has been reported that both methods yield similar long-term results [8]. In hip arthroplasty where the cementless method is preferred, the fixation of the stem depends on osseointegration, which is believed to be influenced by the bone density surrounding the stem [8].

Osteoporosis presents several key challenges during THA, including initial osseointegration, intraoperative fractures, late periprosthetic fractures, and the long-term survival of the implant. Research addressing these issues in both cemented and cementless techniques remains limited [10]. A comprehensive study investigated the prevalence of osteoporosis in elderly patients undergoing THA, retrospectively evaluating 268 patients over the age of 70. Of these, 159 patients were found to be osteopenic or osteoporotic [25].

Factors influencing subsidence in patients undergoing cementless THA were investigated in postmenopausal women with a high prevalence of osteoporosis. It was suggested that reduced BMD after menopause complicates axial and rotational stability of the femoral stem. However, a prospective study involving 65 postmenopausal women with an average age of 68 found no statistically significant difference in subsidence between groups with normal and low BMD. Moreover, although antiresorptive osteoporosis treatments such as denosumab and bisphosphonates have demonstrated positive effects on proximal femoral cortical bone in postmenopausal women with low BMD, randomized clinical trials have shown that these agents do not effectively prevent early migration of uncemented femoral stems [26].

In a prospective study involving 211 patients with 218 hips, the subsidence measurements of cementless femoral stems were evaluated over a 5-year period. Factors such as age, gender, BMI, stem type, surface coating material, and femoral head size were assessed. The average subsidence value was found to be 0.75 mm, with only 16 patients reporting subsidence greater than 2 mm. No differences were observed between groups based on age and gender [27]. However, the findings of this study indicated statistically significant lower subsidence values in the female group. The average BMD in the female group included in this study was found to be lower compared to the male group. Considering that elderly women generally have lower BMD compared to the general population, the authors propose two potential explanations for the more pronounced initial femoral stem subsidence observed in the male group, specifically in cases of metaphyseal-engaging femoral stems, which exhibited higher bone density. One potential explanation is that patients with osteoporosis exhibit diminished cancellous bone, resulting in the stem being primarily anchored and stabilized within the cortical bone during the surgical procedure. In contrast, patients with normal BMD maintain intact cancellous bone, which is compacted using rasps during the preparation of the stem. This process leaves a cancellous bone layer between the cortical bone and the stem, which could contribute to subsidence. Another potential explanation is that; surgeons, upon observing poor bone quality during the operation, might exercise increased caution in stem preparation and fixation. This heightened vigilance could influence the outcome and contribute to the observed subsidence patterns [28].

In a study by Dyreborg et al. [8], the relationship between proximal femoral BMD and subsidence was examined in 62 patients. The patients were divided into two groups based on T-score values: >-1 and ≤-1. Subsidence measurements were assessed using postoperative radiostereometric analysis (RSA). No significant relationship was found between subsidence values and BMD; however, it was noted that the study was conducted in patients under 75 years of age with a T-score above − 2.5. On the other hand, a study involving 39 female patients aimed to evaluate early stability and osseointegration in patients with low bone density. The study assessed the proximal femoral bone density and femoral morphology of the patients and categorized them based on their bone density. Stem migration and osseointegration were evaluated using RSA and clinical outcomes were assessed using the HHS. Although stem migration was numerically higher in the low bone density group during the first 3 months, the 2-year follow-up showed similar results in both groups, consistent with our study [5]. Comparing these results with patients T-score above − 2.5, this study included also patients with scores lower than − 2.5 which shows the strength of this study [8].

In a study conducted by Rhyu et al. [4], early subsidence values were assessed in a cohort of 320 patients. Patients over the age of 70 with proximal femur T-scores below − 2.5 were compared to a control group of patients under the age of 50. At the 1-year follow-up, no statistically significant differences in subsidence values were found between the two groups. While no patients in the study group experienced subsidence greater than 3 mm, 3 patients in the control group had subsidence exceeding 3 mm; however, this difference was not statistically significant. The study reported that low subsidence and excellent fixation were achieved after 1 year of follow-up in osteoporotic patients undergoing cementless hip arthroplasty. The study concluded that, consistent with our findings, cementless femoral stems are not contraindicated in osteoporotic patients.

Although early osseointegration is valuable for long-term prosthesis survival and subsidence, studies related to osseointegration in the literature are limited. In one study evaluating osseointegration, the prosthesis design was examined and Engh scores were assessed. The results indicated that prosthesis design was considered more valuable for osseointegration than BMD [29]. Another study found that osseointegration scores and clinical outcomes were related to canal fill ratio (CFR) rather than femoral morphology. It was even reported that CFR was lower in Dorr type A femurs [30]. According to the literature the maximum of femoral component subsidence occurs within the first 6–8 weeks postoperatively. It is assumed that subsidence usually occurs within the first weeks of weight bearing and osseointegration takes four to 12 weeks but rarely may take up to three years [11, 13]. However, recent studies investigating the impact of therapist-patient relationship continuity on orthopedic rehabilitation outcomes have shown that changes in therapists during the rehabilitation process do not have a significant effect on functional outcomes and early recovery [31]. This suggests that the osseointegration process may be more closely related to surgical stabilization and mechanical loading rather than individual patient factors. These literature data was important to choose the mean follow-up time as 24.6 + 5.8 months (range 12 to 34 months) and also not to evaluate the influence of rehabilitation on clinical results. Also most of the studies with RSA have a 2-year follow-up time in the literature [5, 8]. To prevent subsidence of the femoral stem an adequate interference fit, mechanical stability and initial press-fit fixation against the cortical bone are required [13, 14]. Metaphyseal and proximal diaphyseal fixation achieved with the chosen femoral stem which facilitates the fixation against the cortical bone in proximal diaphysis maybe effective on the results in femurs with low BMD in this study.

In a study where the use of cementless femoral stems in Dorr type C femurs was reported to be controversial, no statistically significant relationship was found between femoral stem subsidence and femoral morphology or bone mineral density in the current cohort. This finding is consistent with the results of Rattanaprichavej et al. [32], who performed multivariate regression analysis in a large cohort and reported that femoral morphology exhibited only a marginal, non-significant influence on subsidence. In their analysis, the only factor significantly associated with reduced subsidence was the canal-fill ratio at the metadiaphyseal junction, emphasizing the importance of optimal implant fit rather than proximal bone morphology alone. These findings support the notion that femoral morphology and bone density may not independently predict early stem migration when adequate primary stability and surgical technique are achieved.

This study has several limitations. Only one type of prosthesis was used, and patients who underwent cemented THA were not evaluated. But we also aimed to evaluate the stem design with a metaphyseal and proximal fixation in patients with a low BMD. Due to the limited resources and financial constraints of our hospital, radiological assessments were performed using standard AP and lateral X-rays instead of more advanced computer-assisted systems. The error margin of these methods is reported as 2 mm in the literature, and recent studies suggest that RSA would provide more accurate results [33]. However RSA has been used as a valuable clinical tool to measure the migration of implants in reference to the bone in 3 dimensions [26, 34]. RSA also shows translational and rotational migration of the implant. But this study aimed to evaluate only the subsidence which is defined as a femoral stem distalization in reference to the greater trochanter. The small number of patients in the Dorr Type C group did not allow for meaningful statistical analysis. Therefore, this group was excluded from the final evaluation to ensure the validity and reliability of the results. Further research with larger sample sizes and different stem designs is required to strengthen the hypothesis regarding the safety of cementless prosthesis use in patients with low bone density.

Conclusion

Based on the findings of this study, BMD was not a determining factor for femoral stem subsidence or osseointegration in patients undergoing cementless THA. Comparable clinical scores and radiographic outcomes were observed in patients with normal, osteopenic, and osteoporotic bone profiles. These results suggest that cementless femoral stems can provide reliable fixation across varying bone qualities when adequate initial mechanical stability is achieved. The absence of a significant association between BMD or proximal femoral morphology and stem migration underscores the importance of implant design and primary fixation technique in early postoperative outcomes. Preoperative BMD did not significantly affect radiological or functional outcomes following THA. However, postoperative improvement in BMD may be influenced by concurrent osteoporosis management, which warrants further prospective evaluation. These findings support a shift in clinical focus from bone density-based risk stratification to individualized implant selection strategies prioritizing initial stability. This approach represents an innovative perspective that challenges traditional assumptions regarding the role of BMD in the success of cementless femoral stems.

Abbreviations

AP

Antero-posterior

BMD

Bone mineral density

BMI

Body mass index

CFR

Canal fill ratio

DXA

Dual-energy X-ray absorptiometry

HHS

Harris hip score

ICC

Intraclass correlation coefficient

OA

Osteoarthritis

PMMA

Polymethylmethacrylate

RSA

Radiostereometric analysis

THA

Total hip arthroplasty

VAS

Visual analog scale

WHO

World Health Organization

Authors’ contributions

All authors contributed to the study. Conceptualization/Methodology: Ahmet Sinan KALYENCI and Yusuf OZTURMEN. Data Collection: Saltuk Bugra TEKIN and Zana OZMEN. Statistical analyses: Ahmet SENEL and Saltuk Bugra TEKIN. Literature research: Ahmet Sinan KALYENCI, Ibrahım DOGAN and Zana OZMEN. Writing - Original Draft: Ahmet Sinan KALYENCI. Writing - Review & Editing: Yusuf OZTURKMEN. Visualization: Ahmet Sinan KALYENCI and Ibrahım DOGAN. Supervision: Ahmet SENEL and Yusuf OZTURKMEN. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support was received during the preparation of this manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. The study involved human participants and approved by the Ethics Committee of Istanbul Training and Research Hospital. (Date: 12/04/2019 No: 1790). Informed consent was obtained from all participants prior to their involvement in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.