Total Elbow Replacement in Dogs

Loïc Déjardin and Reunan P. Guillou

During the past two decades, functional limitations of conservative management and nonreplacement surgeries for the treatment of end-stage canine elbow osteoarthritis have fueled growing interest in total elbow replacement. Clinical use of a total elbow replacement was first reported by Whittick, who in 1964 used a custom-made spherical hinged prosthesis to treat a gunshot-induced comminuted elbow fracture in a cat.¹⁷ To our knowledge, the first clinical canine total elbow replacement was implanted in 1989 by Chancrin, who used a prototype cemented hinged prosthesis to treat a Labrador Retriever afflicted with end-stage osteoarthritis and severe ankylosis (Figure 54-1, A).⁵ Subsequently, in 1996, Lewis reported on the first clinical results of a hinged total elbow replacement implanted in 10 dogs.¹⁵ In these first-generation systems, Chancrin and Lewis used cemented, fully constrained hinged designs (linked systems). Because of the rigid mechanical link between the humeral and radioulnar (RU) components, most of the forces across the joint were transmitted through the implant to the cement and its interfaces.⁴ The high complication rates encountered with these initial designs quickly led to a paradigm shift to unlinked total elbow replacement designs (Figure 54-1, B through E). Vasseur,¹⁶ Lewis (second and third generations),¹⁵ Cook,¹² and Conzemius^6,10 developed the first unlinked designs in the late 90s (Figure 54-1, B and C). All encountered unacceptable postoperative morbidity that led to termination or further refinement of the respective designs. Following iterations of his earlier designs (Figure 54-1, D), Conzemius reported encouraging results after total elbow replacement in 6 normal dogs, then 2 years later in 20 dogs afflicted with naturally occurring osteoarthritis.^7,8 Minor modifications, including the addition of porous coated surfaces to the lateral and medial aspects of the humeral component, as well as a curvilinear humeral articular profile, led to Conzemius’ fifth total elbow replacement generation, which became commercially available in 2005 (Iowa State system, BioMedtrix, Boonton, NJ).

Figure 54-1 Synopsis of past and current linked (A) and unlinked, semi-constrained total elbow replacement (TER) systems (B–E). From left to right: Chancrin, Lewis (third generation), Cook, Iowa State (Conzemius fourth generation), and TATE (Acker/Van Der Muelen). Chancrin’s prosthesis (A) was a pure linked, hinged system. The current Lewis prosthesis (B) is a hybrid three-component system in which the humeral component and a radial ultra-high-molecular-weight-polyethylene (UHMWPE) button (allowing pro-supination) are cemented, and a radioulnar (RU) shelf is screwed to the ulna. This system is still in limited use today. Cook’s prosthesis (C) used a hybrid cemented/screwed design. This system has been abandoned because of severe complications. The current Iowa State (Conzemius) prosthesis differs slightly from the one depicted here (fully cemented [D]). It now uses a hybrid design, allowing bone ingrowth at the level of the lateral and medial condylar surfaces. Finally, the TATE (Acker) prosthesis was designed as a resurfacing cementless cartridge unit. (Please see text for additional details.) These drawings were either generated from actual photographs or modified from original U.S. patent applications by Reunan P. Guillou.

Recently, a radically novel total elbow replacement design (TATE Elbow system, BioMedtrix) was developed by Acker and Van Der Meulen (Figure 54-1, E).^2,3 Similar to Conzemius’ prosthesis, the TATE uses an unlinked, semiconstrained design. However, several fundamental differences exist between these two systems. Unlike previous conventional stemmed and cemented prostheses, the cementless TATE implant was designed to use a novel resurfacing concept, as well as less invasive surgical approaches. In 2008, Acker and Van Der Meulen reported satisfactory results 6 months after implantation of this resurfacing prostheses in 6 dogs affected with end-stage osteoarthritis.¹

Current Systems—Design Rationale and Evolution

Both currently commercially available prostheses (i.e., fifth-generation Iowa State and the recently released second-generation TATE Elbow systems) use unlinked semi-constrained two-component designs. In the absence of a rigid mechanical link between the humeral and radioulnar components, the stability of unlinked prostheses is provided by the matching geometry of the prosthetic articulating surfaces and the surrounding soft tissue envelope.⁴ Specifically, current total elbow replacement systems are designed so that some of the transarticular forces (e.g., the forces in varus-valgus) are almost exclusively counteracted by passive soft tissue constraints (i.e., collateral ligaments), while other forces (e.g., internal/external rotation and mediolateral translation) are also controlled by the geometry of the prosthetic articular surfaces. Although contributing to joint stability, prosthetic constraint has been shown to influence stresses at the bone-implant interface. Interfacial stress distribution is further complicated by the fact that both systems feature a single radioulnar component that eliminates motion between the radius and the ulna. Although the ideal prosthesis should allow near-normal joint kinematics of humeroantebrachial and radioulnar joints, such a design would likely require more complex three-component prostheses. The current designs were likely chosen as a compromise between optimal joint kinematics and decreased prosthetic complexity and related risk of implant failure.

Although physiologic radioulnar motion is limited in arthritic elbows, residual motion could be detrimental to the long-term stability of the radioulnar component. To reduce the risk of interfacial failure at the level of the radioulnar component, the surgical procedure in either system includes a radioulnar synostosis. The current Iowa State system is a composite fixation (cemented and porous) system featuring a cobalt chrome humeral stemmed component and a 120 degree arc ultra-high-molecular-weight-polyethylene (UHMWPE) radioulnar component. The articular surfaces of both components are symmetric with respect to the sagittal plane and therefore do not differ between right and left implants. The humeral stem can be used in both left and right elbows, whereas right and left radioulnar components feature side-specific peg locations. The humeral stem and radioulnar posts are cemented in the corresponding medullary cavities for primary implant fixation. In addition, the lateral and medial aspects of the humeral stem feature a porous surface for subsequent osteointegration and long-term fixation. The humeral and radioulnar components are implanted individually in sequence following preparation of the respective bones.

In contrast, the TATE Elbow system uses a cementless resurfacing design consisting of a cobalt chrome humeral component and a 175-degree arc ultra-high-molecular-weight-polyethylene radioulnar component featuring a cobalt chrome metal backing. Both humeral component and radioulnar metal backing feature two mediolateral posts for primary fixation and a porous surface for long-term stability via bone ingrowth. The TATE Elbow system was designed to use a minimally invasive approach via osteotomy of the medial humeral epicondyle. During implantation, both components are linked by a set plate and are inserted simultaneously as a “cartridge implant.”

Design modifications have been recently devised for both systems in response to their respective (documented or potential) shortcomings. In an attempt to reduce the rate of post-operative elbow luxation, Conzemius’ team is currently developing a highly constrained, metal on metal, total elbow replacement system Figure 54-2, A and B.^15a Although this new design has been shown to offer greater intrinsic stability than the fifth generation Iowa system, to the author’s knowledge, this system has not been used in live dogs. Similarly, in an effort to optimize bone ingrowth a second generation TATE Elbow system was released early 2010 (Figure 54-2, C and E). This refined design has been used in ~30 clinical cases worldwide (Figure 54-2, D and F). Design modifications included hollow primary fixation posts and hydroxyapatite coating. In addition, reduced prosthetic constraint in rotation and mediolateral translation was achieved through modification of the radioulnar articular profile.^13b,13c Although no cases of implant loosening had been observed with either TATE generation, these design iterations, were mainly intended to promote implant osteointegration.