1. Introduction
Arthroplasty has revolutionized the landscape of orthopedic surgery, providing effective solutions for enhancing mobility and alleviating pain in patients with debilitating shoulder and hip conditions. The evolution of shoulder and hip arthroplasties has been marked by continuous advancements in surgical techniques, materials, and implant designs. Despite these strides, implant failures remain a concern, prompting the need for a comprehensive examination to better understand the underlying causes and to establish criteria for implant success.
The history of hip arthroplasty has progressed significantly over the last century and a half with many different successes and failures along the way. It began in the early 19th century with Dr. Gluck’s ivory hemi hip replacement (1880s) and has since gone through key implant designs such as the Smith-Petersen mold arthroplasty (1923), McKee-Farrar total hip arthroplasty (THA) (1960), and the low friction THA by Dr. John Charnley (1962) to reach the systems we use today [
1,
2,
3]. Although many of these implants demonstrated successful outcomes, our goal was to identify described failure mechanisms that were improved upon in future designs and techniques, regardless of their incidence.
Similarly, the history of shoulder arthroplasty has gone through many stages of development. Since its inception in 1893 with Dr. Jules Emile Péan’s ivory total shoulder, shoulder arthroplasty has seen changes in materials, designs, and geometries that have led to the modern implants used today [
4,
5,
6].
The literature has shown that some developments in shoulder arthroplasty have been influenced by similar advancements in hip arthroplasty such is the case of Dr. Péan’s first shoulder arthroplasty, which utilized ivory implants, and Dr. Charles Neer’s hemiarthroplasty (HA) in 1951 that was likely inspired by the success of HA for the treatment of hip fractures [
4,
6,
7].
This literature review’s main objective is to identify common factors contributing to implant failures across shoulder and hip arthroplasties. By analyzing material, mechanical, and technical aspects as to why different prostheses failed, we aim to better understand how we arrived at implant designs that are currently in use and how developments in hip and shoulder arthroplasties have influenced their respective iterations. With this, we intend to create the first applicable criteria that classify failure methods based on the heuristic surgeons use to group failed arthroplasties. We also aim to emphasize the importance of communication amongst subspecialties during implant design processes, to optimize patient outcomes and avoid preventable failures for those performing shoulder and hip arthroplasties.
2. Materials and Methods
A review of the literature was conducted to elucidate the historical contexts, trends, and modes of failure associated with shoulder and hip arthroplasties. Articles were searched across different search engines and databases such as PubMed, Scopus, and Embase. Included studies provided historical insights into the evolution of designs, progressions in surgical techniques, and outcomes of shoulder and hip arthroplasties, with a specific focus on implant failures and complications. Once an article was identified to have pertinent material, its reference section was also reviewed for additional articles.
The search was not restricted by publication date in order to provide a broad historical overview. Two researchers (RA and JL) independently screened titles and abstracts, followed by a full-text assessment of selected articles to determine final inclusion. Data extraction involved retrieving publication details, historical contexts, and relevant outcomes.
All included failed prostheses were subcategorized based on their respective common failure modes into one of three broad failure categories: material, mechanical, or technical. Failure designation was decided by the initial reviewer (RA) and approved by an additional reviewer (MAF). Material failure was defined as a failure/wear of the implant materials (i.e., polyethylene wear, bearing surface wear particles, polytetrafluoroethylene (PTFE) wear, and foreign body reactions) that was deemed the primary cause of construct failure. Mechanical failure was defined as a violation of a biomechanical principle deemed the primary cause of construct failure (i.e., scapular notching and glenoid component failure secondary to prosthetic impingement). Technical failure was defined by failure due to surgical technique (i.e., greater tuberosity osteotomy, greater trochanteric osteotomy, and approach) that was deemed the primary cause of construct failure. Once classified, the total of implant failures by subcategory and their breakdowns in their respective subspecialties were calculated. The synthesized data aimed to provide a simplified overview of the historical trajectory and commonalities in failure modes associated with shoulder and hip arthroplasty implants.
4. Discussion
Surgical techniques, implant materials, and the understanding of joint biomechanics have significantly advanced since the inception of arthroplasty over 140 years ago as a treatment for debilitating joint diseases. However, ongoing innovation to mitigate common causes of failure underscores the importance of learning from historical missteps. Effective communication among arthroplasty designers is crucial as it ensures the exchange of expertise and insights, akin to how Dr. Charnley’s collaboration with dentists informed his decision to utilize PMMA bone cement [
18]. In the current era of orthopedics, implant designers often operate in isolated silos within their subspecialties. Lack of cross-disciplinary communication regarding historical failures and current insights into joint biomechanics, techniques, and materials may result in oversights and the repetition of failures evident in other subspecialized fields. In total hip arthroplasty, many of the failure modes have been addressed with improvements in implant design, biomaterials, and surgical technique.
Material failures that were seen with the use of ivory, glass, PTFE, polyethylene, stainless steel, polyacetal, ceramics, and MoM bearing surfaces were decreased with the use of inert materials with favorable wear properties and low fracture rates [
1,
2]. Present-day femoral stems utilize stainless steel, cobalt–chromium alloys, and titanium alloys. Although the poor biocompatibility of stainless steel has decreased its use in THA, titanium (Ti-6Al-4V) or cobalt (CoCrMo) alloys have become the most predominant metals used in THA [
28]. Moreover, titanium alloys allow for bony ingrowth, making the use of uncemented implants more predictable to provide more durable fixation. Additionally, wear resistance has been improved with the introduction of highly crosslinked molecular weight (HXLE) and ceramic bearings that are being used more frequently in THA [
14,
25,
28]. Implant fracture rates have been decreased with the use of stronger metal alloys such as the ones mentioned above and zirconia-toughened alumina (ZTA) [
14,
28].
In total shoulder arthroplasty, the improvement in biomaterials was accelerated by some of the failures seen in total hip arthroplasty such as those in the case of the Judet brothers adopting acrylic in hips in 1948 and later Richards, Krueger, and deAnquin doing the same, leading to similar failure modes. The use of ivory, acrylic, and Vitallium have been replaced by more resistant materials that have decreased hardware-related complications [
29]. Similarly, shoulder arthroplasty has utilized these biomaterials as in hip arthroplasty, and in doing so, failures that were seen in early total hip designs have been avoided. Currently, humeral stems consist of combinations of metals with titanium (Ti-6Al-4V) or cobalt (CoCrMo), with cobalt being preferable because of improved wear resistance [
29]. Similar to hip arthroplasty, bearing surfaces used in shoulder arthroplasty include polyethylene (UHMWPE and highly crosslinked HXLPE), ceramic, and pyrolytic carbon (PyC) [
29,
30,
31]. Ceramic bearings, although widely used in hip arthroplasty, have been a challenge to implement in shoulder arthroplasty. Currently, the only ceramic bearing shoulder implant system on the market in the United States has a warning due to fracture at the coupling [
29,
30]. The wear properties of pyrolytic carbon have led to its increased popularity and have shown promising results in the hand and elbow [
29,
30]. The most popular bearing surface configuration consists of cobalt chrome on UHMWPE [
29,
30].
Mechanical failures such as acetabular wear, femoral component loosening, stress shielding, and stem fracture have all been reduced with advancements in implant design and patient selection. Previously observed acetabular erosions in the setting of hemiarthroplasty have been presently combated with acetabular component implantation, also known as total hip arthroplasty [
1]. Initial advancements in stem fixation were attributed to Dr. Charnley’s introduction of PMMA. However, high rates of aseptic loosening due to “cement disease” and “particle disease” set the groundwork for modern uncemented stems with different fixation methods [
8]. Femoral component loosening is now less frequent due to improved fixation methods such as refined cementing techniques including preparation under vacuum and the application of lower viscosity under pressure [
32]. Additionally, stem stability has improved with the utilization of varied stem geometry such as tapered, fluted, and wedged stems [
33]. The shape of the stem decides the pressurization of the cement during insertion and for rotational stability. Understanding this has led to the development of successful cementless stems [
34,
35,
36]. Additionally, the advent of porous coating and press-fit technique after appropriate canal preparation has made uncemented fixation gain popularity. Stress shielding was combated with greater metaphyseal fixation by upsizing the proximal body of the implant, converting a fully porous stems to metaphyseal coating only, using materials with a modulus of elasticity closest to bone, matching the geometry of the stem to that of the femoral canal, and decreasing the length of stems [
8,
33,
37].
Mechanical failures in shoulder arthroplasty have decreased because of iteratively improved implant designs that were driven by applied biomechanical data. Currently, surgeons prevent instability by addressing patient-specific variations in glenoid size and version, glenoid lateralization, humeral head height, soft tissue balancing, and rotator cuff integrity [
38]. Glenoid loosening has been addressed by cementing the glenoid component, the introduction of the baseplate and the inlay design in 1985, the use of a central peg and peripheral screw fixation, and the use of augmentation in patients with bone loss [
4,
17,
22,
31,
39]. Humeral stem loosening has been combated with the use of cement, porous coating to improve press-fit fixation, variations in stem size, stemless humeral implants, inlay versus onlay models, and most recently, the use of convertible platforms to facilitate conversion between a TSA and an RSA in the revision setting [
4,
31,
40]. Unlike the weight bearing hip joint, load distribution across the shoulder is primarily concentrated in the proximal aspect of the humerus. This load distribution in the shoulder leads to the ability to achieve fixation in the humeral metaphysis as opposed to the diaphysis [
41]. This has led to the development of different proximal stem geometries and cylindrical stems that engage the endosteal canal to achieve appropriate alignment and fixation [
41,
42]. Moreover, stem diameter, modularity, and geometry have been shown to be important features to increase fixation and rotational stability in both shoulder and hip arthroplasties [
34,
35,
36,
43,
44,
45]. However, due to the different load distributions across the shoulder, component breakage has different causes when compared to hip arthroplasty. The mechanical breakage of components that was observed mainly in constrained prostheses has ushered in implant designs with less constraints. Component impingement, particularly scapular notching, has decreased because of a reduced neck shaft angle in RSA and by achieving adequate glenoid lateralization and positioning [
46,
47]. Scapular fractures in RTSA are becoming less frequent by improving bone health, by correcting glenoid baseplate screw length, and by avoiding excessive deltoid tension [
48,
49]. The introduction of reverse shoulder arthroplasty led to decreased mechanical complications seen in the past, but component geometry is still evolving. The more common “anatomic style reverses” have led to a lower rate of mechanical failures and the improved preservation of functions [
50].
Improvements in surgical techniques have minimized technical failures. Practices such as avoiding excessive reaming component undersizing and accurate component positioning have led to a decrease in failures. These advancements have led to a reduction in issues such as component loosening, periprosthetic fracture, and the accelerated wear of bearing surfaces [
3,
11,
26].
Technical failures in shoulder arthroplasty have been reduced by the better understanding of surgical site infections, avoiding osteotomies and excessive bone resection, and favoring muscle-sparing approaches [
51,
52]. Instability has been mitigated with component upsizing and increasing offset. However, achieving the ideal balance to avoid increased joint reactive forces that will accelerate wear rates is still evolving [
53]. Greater tuberosity osteotomies, particularly in the fracture setting, have been associated with lower functional results and more postoperative complications [
54]. Currently, the subscapularis sparing technique has been shown to reduce instability risk, but is more technically complex, and limited exposure may lead to the incorrect fixation of the humeral head [
52]. These advancements have led to alternatives to address instability and a reduction in infection rates, rotator cuff failures, and implant failures.
The benefit of shoulder arthroplasty emerging behind the developments in total hip replacement surgery is demonstrated by several innovations in the shoulder that were developed using evidence from failed total hip. For example, impingement-free range of motion, a feature that led to some failures of total hips, was then minimized with the introduction of larger femoral head implants. This was then expanded in RSA with the introduction of lateralized glenosphere to increase impingement-free ROM. Additionally, the reduced amount of material failures in shoulders compared to hips, as evident by the 10% failures in shoulders compared to the 48% failures in hips, validates this benefit.
The future benefit of hip arthroplasty and shoulder arthroplasty surgeons collaborating could be the use of preop planning software that may lead to improvements in placing implants to maximize fixation and optimize function such as impingement-free ROM in both total shoulder and hip arthroplasties. Moreover, the introduction of the reverse hip emulating the concept of the reverse total shoulder shows the inspiration taken from other subspecialties [
55,
56]. Although there has been little utilization of this type of prosthesis, the development of the reverse hip demonstrates that collective research between specialties can increase treatment options for recurrent complications and ultimately improve patient outcomes [
55,
56,
57].
This review has several limitations. One limitation is the oversimplification of each modality of implant failure. The authors acknowledge that most implant failures are multifactorial in nature. Another limitation is that the categorization of the failures themselves is subjective to the authors and are not precise but more of a generalization. An additional limitation is that failed implants were mostly extracted from historical articles; thus, the information available is limited by what is reported in those publications. It should also be known that the named failures were not meant to target any brand but rather state a failure that was described in the literature. However, a strength of our article is that this is the first time a classification of failure modes has been proposed. By grouping similar failure mechanisms, it simplifies the way surgeons and implant designers can address issues in a collaborative form with other subspecialties in orthopedics.