Total Hip Arthroplasty Stems and Femur Strain During Implant Insertion: A Biomechanical Study of Single Taper, Double Taper, and Collared Monolithic Stems

2.1 Specimen Characteristics

A total of 24 fresh-frozen human cadaveric femurs were procured. Specimens were excluded if they had evidence of trauma or a prior surgical procedure, which was inferred by direct inspection of the osseous specimens for deformity or implants. Dual X-ray absorptiometry (DEXA) scans measured the bone mineral density (BMD) of each femur at the greater trochanter and lesser trochanter. Femurs were then randomly assigned to three groups of stem geometry utilizing computer randomization: single (medial-to-lateral) taper, double taper (fit-and-fill), or collared stem.

2.2 Stem Characteristics

Broaches and implants were provided by Smith and Nephew (London, UK). Group 1 was randomized to the“Anthology” (Smith and Nephew, London UK) stem, serving as a proxy for Mont Type 1, single medial to lateral taper with proximal porous coating (Figure 1A) [12, 13]. Group 2 was randomized to the “Synergy” (Smith and Nephew, London UK) stem, representing typical double taper Mont Type 2 stems (Figure 1B) [14]. Finally, Group 3 was randomized to a collared stem, representing Mont Type 3C, for which the “Polar” (Smith and Nephew, London UK) stem was utilized (Figure 1C) [10, 15].

The three different monolithic femoral implants tested in this study are widely used in the field of arthroplasty. The “Anthology” stem used in Group 1 has demonstrated satisfactory clinical outcomes, with a reported stem survival rate of 99% over a follow-up period exceeding 20 years [12, 13] (Figure 1A). The “Synergy” stem used in Group 2, representing typical Mont Type 2 stems, is proximally porous coated with medial-to-lateral and anterior-to-posterior taper geometry [14] (Figure 1B). This design aims to achieve fixation by contacting the medial and lateral metaphyseal cortex, as well as the anterior and posterior metaphyseal cortex. In comparison to the Type 1 stem, it has a relatively thicker profile in the anterior–posterior plane and fills a larger surface area of the femoral metaphysis, giving rise to its “fit-and-fill” name. Clinical reports have indicated a Type 2 stem survival rate of approximately 95%–100% over a follow-up period of 15–20 years [12]. Lastly, Group 3 had the “Polar” stem implanted, which is a collared stem, representing a Mont Type 3C stem [10]. The Polar collared stem is fully titanium and hydroxyapatite coated with a proximal collar (Figure 1C). These stems have a rectangular cross-section that provides robust rotational stability with all four corners press-fitting into endosteal bone. As they extend distally, both the anterior–posterior and mediolateral planes taper in a wedge shape. The fixation of these Type 3C stems primarily occurs at the junction between the femoral metaphysis and diaphysis, as well as at the proximal part of the diaphysis, enabling three-point fixation in the sagittal plane. This type of stem has demonstrated satisfactory clinical outcomes with a stem survival rate of greater than 95% over a follow-up period exceeding 20 years [12, 15].

2.3 ImplantTemplating and Selection

All stems were implanted under the guidance of the senior author (O.S.), a fellowship-trained arthroplasty surgeon. X-rays of the femurs were obtained to digitally template the stems with OrthoView (Jacksonville, Florida, USA) software. All remaining soft tissue was removed from the femurs prior to instrumentation. The femoral neck cuts were performed 10 mm above the lesser trochanter and carried to the saddle of the greater trochanter with the assistance of the appropriate osteotomy guide. The femurs were then instrumented with implant-specific instrumentation sets. The decision of final broach size was made with a combination of the digitally templated stem sizes and clinical judgment. No fractures occurred during instrumentation or stem implantation.

2.4 Experimental Setup

The femurs were potted with polymethyl methacrylate (PMMA) into 3-in. diameter polyvinylchloride (PVC) tubes so that the tip of the greater trochanter was 12 in. above the base of the PVC tube for all test subjects (Figure 2). The testing apparatus was assembled to reproducibly create an impaction force upon the insertion handle similar to that in previous biomechanical literature [16]. After manually broaching to the correct size, guided by digital templating and clinical judgment, theone of the three matching stem designs (Figure 1) was hand-inserted until the top of its ingrowth surface stood 5 mm above the neck cut. A 48-in. polyvinylchloride tube housed a 1-kg weight attached to a pulley that was dropped from height of 1 meterto produce the final stem impaction force (Figures 2 and 3). To maintain a consistent 1-m drop height for all tests, the weight's drop height was adjusted based on slight variations in the length of the inserter handles for each stem type. Strain gauges (rosette 350 Ohm strain gauges (Vishay, Raleigh, NC)) were affixed to the anterior, medial, posterior, and lateral proximal femur to measure the strain generated byfinal stem impaction in the horizontal and vertical directions (Figure 2). Strain was recorded with data acquisition software. The measurement of strain is unitless. In our experimental setup, a positive value indicated a tensile strain while a negative value indicated a compressive strain. Immediately prior to testing, the potted femurs with attached strain gaugeswere allowed to come to room temperature and then placed into a base and affixed with bolts to ensure a secure fit.

2.5 Data Analyses

Baseline demographic characteristics, including age, sex, race, and BMD, were compared between stem groups using the Kruskal–Wallis and chi-square tests, where appropriate. Peak strain, calculated as the difference between maximum strain at final impaction and baseline strain, was compared between stem designs for each vector orientation (anterior, posterior, medial, and lateral) utilizingthe Kruskal–Wallis test. For significant differences identified from global testing with Kruskal–Wallis, wethen conducted pairwise comparisonsusing the Deans–Steel–Critchlow–Fligner test. p-values < 0.05 were considered statistically significant. Data management and analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

3.1 Demographics

Of the 24 fresh-frozen human cadaveric femurs used in this study, the mean donor age was 63 years, with a standard deviation of ±10 years, and ages ranged from 45 to 80 years. There were 14 specimens from men and 10 from women. There were 18 Caucasian specimens, and six were African American. There were no differences in age, sex, and race between the three groups. There were no differences between groups in BMD (Table 1).

3.2 Horizontal Strain on Medial Proximal Femur

At the medial proximal femur, the collared stem produced compressive strain (−276) in the horizontal vector at final impaction, differing significantly from the tensile strain generated by the single (41, p = 0.009) and double taper stems (218.5, p = 0.003) (Figure 4). No significant difference in strain existed between the single and double taper stems at the medial proximal femur in the horizontal vector (41 vs. 218.5, p = 0.545).

3.3 Horizontal Strain on Lateral Proximal Femur

At the lateral proximal femur, double taper stem impaction resulted in compressive strain (−69), significantly differing from the tensile strain produced by single taper (221, p = 0.024) and collared stem impaction (462, p = 0.009). There was no difference in strain observed between the collared and singletaper stems at the lateral proximal femur in the horizontal vector (462 vs. 221, p = 0.859).

3.4 Anterior and Posterior Horizontal Strain and Vertical Strain

There was no difference in proximal femur strain generated by the three stems in the horizontal vector at either the anterior or posterior femur (Figure 4). Additionally, there were no differences in strain between the three stems in the vertical vector at any of the 4 locations evaluated in this study (Figure 5).

4.1 Main Findings

Collared stems generated a compressive strain at the medial proximal femur in the horizontal plane, which was significantly different from the tensile strain produced by single and double taper stems. At the lateral proximal femur in the horizontal plane, a compressive strain was produced by the double tapered stems, while the collared and single taper stems produced a tensile strain. There was no significant difference between stems at the anterior and posterior cortex in the horizontal vector and in any area in the vertical vector.

4.2 Study Purpose and Stem Design Overview

This study aimed to investigate the strain on the proximal femur during implantation of different types of primary cementless THA stems and its potential association with the risk of intraoperative periprosthetic femoral fracture (IOPFF). IOPFF during THA is a major contributing factor to reducing overall survivorship of uncemented stems and increasing overall morbidity and mortality [2, 3, 17, 18]. These findings provide insights into the biomechanical behavior and associated relative fracture risk of the three stem geometries examined: single (medial-to-lateral) taper, double taper, and collared stems.

4.3 Collard Stem Design and Medial Proximal Femoral Strain

In this biomechanical model of skeletally mature specimens with a range of bone densities, the impaction of a collared stem produced significantly less strain in the horizontal vector at the medial proximal femur compared to the strain produced from the impaction of single taper and double taper implants. This finding suggests that the design of the stem may play a critical role in modulating strain distribution in the proximal femur. The use of a collared stem may shield the medial proximal femur from circumferential strain as the collar of the device contacts the relatively strong calcar bone during implantation [19, 20]. Furthermore, no difference in strain was observed between the single taper and double taper stems at the medial cortex, challenging the often-remarked adage that the single (medial-to-lateral) taper stem may have a greater propensity for IOPFF than its metaphyseal filling counterpart. These findings are particularly relevant as the cortex of the medial proximal femur is a common location for intraoperative fracture, and increased strain in this region has been associated with a higher risk of IOPFF in previous studies [21, 22].

4.4 Evidence Supporting Collard Stems as Protective Against Periprosthetic Fracture

Indeed, other studies have touted that collared stems are protective against periprosthetic femoral fractures (PFF). In their paired cadaver biomechanical study, Lamb et al. concluded that the addition of a collar to an uncemented stem increases the torque required to cause periprosthetic femur fracture [19]. The authors postulate that the collar mitigates the rotation of the femoral stem with respect to the endosteal geometry, thereby decreasing the risk of fracture [20]. Similarly, Demey et al. demonstrated in their biomechanical model that collared uncemented stems have significantly greater immediate stability compared to collarless stems, enabling them to withstand greater forces before subsidence and fracture [23]. Furthermore, in a multicenter, retrospective study of 3433 primary cementless THAs performed from 2014 to 2021, Calkins et al. reported that compaction-collared stems had a significantly lower risk of postoperative PFF requiring operative treatment than single-taper and double-taper stems [24].

4.5 Implant Selection Based on Bone Density

For such reasons, some authors have advised selecting the appropriate implant design based on the bone density observed in preoperative radiographs. Carli et al. conducted a systematic review of 34 out of 596 initially identified articles and found that cementless implants, particularly single taper and double taper types, were linked to a significantly higher number of PFF when compared to anatomical, fully coated, and tapered/rounded stems [25]. The authors proposed reducing the utilization of single taper and double taper uncemented stems in older patients who have osteoporotic bone [25].

4.6 Double Taper Stem Design and Lateral Proximal Femur Strain

The present study, however, found that impaction of the double taper stem resulted in lower strain in the horizontal vector at the lateral proximal femur compared to the single taper and collared stems. Compared to the single taper and collared stems used in this study, the double taper stem has a relatively thicker profile in the anterior–posterior plane and fills a significant portion of the femoral metaphysis, giving rise to its “fit-and-fill” name. The metaphyseal-filling design with greater fixation in the meta-diaphysis may explain the decreased strain observed at the lateral proximal femur during impaction of the double taper stem compared to the single taper and collared designs.

4.7 Considerations for Cemented and Dual-Modular Stems

It has been suggested that surgeons should consider returning tocementing femoral stems as a solution to address the risk of IOPFF [26, 27]. However, it should be noted that simply transitioning to a cemented stem does not necessarily guarantee a low rate of periprosthetic fractures. Several studies have indicated that smooth, polished, tapered cemented femoral stems (which do not have a collar)actually have a relatively high incidence of fractures following the surgery [28, 29]. Despite these findings, cementation of femoral THA stems should remain within the contemporary arthroplasty surgeon's armamentarium for patients who have Dorr C bone, poor bone quality, and those with femoral neck fractures [26]. Lastly, dual-modular stems have largely fallen out of use during primary THA due to complications such as higher rates of fretting corrosion and fatigue failure at the modular neck-stem junction, associated adverse local tissue reactions, and need for revision surgery [30]. Despite these complications, dual-modular stems still have unique applicability in specific cases to address hip dysplasia and deformities and may be useful on a case-by-case basis [30].