Total elbow arthroplasties1,2 may be categorized as unlinked3,4 or linked5. Commonly reported complications of total elbow arthroplasty include loosening, infection, triceps rupture, and ulnar nerve dysfunction1,3-5. Whereas unlinked prostheses tend to dislocate, linked components are prone to mechanical loosening6-8.
Various modes of failure of the axle assembly involving polyethylene wear and breakage of the locking mechanism have been documented8,9. There have been seven reported cases of failure of the axle assembly of the Coonrad-Morrey implant (Zimmer, Warsaw, Indiana) when the axle assembly was secured with a C-ring6. To the best of our knowledge, there has been no reported case of axle disassembly leading to disassociation of a Coonrad-Morrey elbow replacement since the pin-within-pin snap-fit articulation was introduced in 1998.
The subject of the present case report was informed that data concerning the case would be submitted for publication, and he consented.
A forty-nine-year-old unemployed schizophrenic man sustained a complex intra-articular fracture of the distal part of the right humerus and an ipsilateral displaced distal radial fracture. The humeral fracture was treated with open reduction and internal fixation. The wrist fracture was treated with closed reduction, percutaneous Kirschner wires, and application of an external fixator. A month later, the Kirschner wires and external fixator were removed. Six months following the injury, a distal humeral nonunion was evident (Fig. 1). The hardware was removed two years following the injury; however, the patient remained symptomatic with a painful flail elbow. Radiographs confirmed a nonunion of the medial and lateral epicondylar fragments. An uncomplicated Coonrad-Morrey elbow replacement was performed three years after the initial injury, and the radial head was excised (Fig. 2). There were no operative complications.
Three years after the arthroplasty, triceps insufficiency was evident. The patient could actively flex the elbow to 135°; however, he was unable to extend it against gravity, although passive extension was possible to 20°. Radiographs showed no evidence of loosening, but there was 22° of valgus angulation of the ulnar component (Fig. 3). The patient was asymptomatic, and no treatment was offered.
Four years after the arthroplasty, the patient presented with a one-month history of elbow pain, a decreased range of motion, and considerable swelling over the olecranon. There was no ulnar nerve dysfunction. Radiographs confirmed that the axle assembly had come apart as a result of a fracture of the tongs on the inner pin and that the fragmented hinge assembly was lying within the olecranon bursa (Fig. 4).
A month later, the elbow was re-explored, and the olecranon bursa was found to be stained blue. It was excised along with the fragmented hinge assembly and titanium debris. Most of the remaining stained capsule was excised as well. The humeral and ulnar components were well fixed and were not revised; however, new polyethylene bushings were inserted. The prosthesis was reassembled with a new pin-within-pin snap-fit articulation. The triceps was mobilized proximally, advanced, and reattached to the olecranon with use of intraosseous nonabsorbable sutures.
Postoperatively, the triceps reconstruction was protected with a hinge brace, which allowed flexion of no more than 90° for eight weeks. At two months, the triceps was found to be intact and the wound had healed. The active range of motion of the elbow was from 0° to 130°, and forearm rotation was full. Radiographs showed unremarkable findings and satisfactory coronal plane alignment.
One year after the hinge assembly revision, the patient was experiencing no pain. He stated that he was protective of the elbow, that he avoided heavy lifting, and that he experienced weakness of extension with the limb in elevation. On examination, extension power at the elbow was documented as 4/5, the active range of flexion was from 5° of hyperextension to 135° of flexion, pronation was to 75°, and supination was to 55°. In maximal extension, the free play of the elbow was from 5° to 20° of valgus. Radiographs demonstrated that fixation of the implants remained secure and the static coronal alignment measured 6° of valgus. Stress radiographs were not made.
Various types of total elbow arthroplasties have proved effective in relieving pain and restoring function1-5, albeit with a high rate of complications, which include loosening, instability, dislocation, and compromise of the ulnar nerve. While linked prostheses rarely dislocate, wear of the polyethylene bushings is recognized as a precursor to failure of the axle assembly6-9.
Lee et al. reported a bushing revision rate of 1.3% at a mean of 7.9 years after implantation in a study of 919 Coonrad-Morrey prostheses7. No cases of implant disassembly were reported. The authors concluded that younger patients with a posttraumatic condition or severe preexisting deformity are at greater risk for the development of excessive bushing wear. Hence, the recipient has to be cautioned about the risks of exceeding the recommended activity and lifting restrictions.
Wright and Hastings reported disassembly of the Coonrad-Morrey axle mechanism due to C-ring failure6. They postulated that repetitive valgus-varus and rotational loading leads to polyethylene wear, resulting in a greater valgus-varus arc and transmission of more force to the center pin and ultimately to the C-ring. The C-ring is consequently loaded beyond its limits, causing it to pop out and the axle mechanism to disassemble. Wright and Hastings suggested that the resulting particulate synovitis precipitates osteolysis. The C-ring device was replaced in 1998 by the pin-within-pin snap-fit articulation (Fig. 5), which is assumed to be more secure.
In the present case, risk factors for failure included the patient's young age, the fact that the replacement was consequent to failed fracture fixation, and the possibility that compliance was compromised by the psychiatric comorbidity. Bushing wear may have been accelerated by the attainment of a full range of movement, which is unusual when elbow replacement is performed for the treatment of arthritis or at a late stage after trauma2,5. The Coonrad-Morrey prosthesis is designed to accommodate only 7° of varus/valgus laxity10. Polyethylene bushing wear can be identified by measuring the angle from full varus to full valgus on an anteroposterior stress radiograph (with normal being 3.5° of valgus and 3.5° of varus [a total angle of 7°]). Free play from varus to valgus measuring >7° and <10° indicates partial wear of the bushings, and free play of >10° indicates complete wear of the bushings6,7. The radiographs of our patient made at three years after implantation (Fig. 3) demonstrated 22° of valgus angulation, indicative of bushing wear. The patient was asymptomatic at that time, and revision surgery was not offered.
Figgie et al. reported that polyethylene wear can lead to failure of the axle mechanism without symptoms8. In the study by Lee et al., all of twelve patients who underwent surgery because of polyethylene wear had had pain and/or crepitus7. The authors made no specific recommendations regarding follow-up; however, we suggest that, once polyethylene wear is identified, careful follow-up with radiographs is appropriate to minimize morbidity and the extent of revision surgery. In hindsight, the surgery in our patient could have been confined to bushing exchange, thereby potentially avoiding the need to perform an extensive débridement because of metallosis.
This case gives cause for concern that the inner pin of the new design is a potential flaw that can lead to early failure, particularly in high-demand patients. We hypothesize that wear and fragmentation of the protective polyethylene bushings precipitated cyclical valgus loading of the elbow, which led to repetitive tensile strain on the weakest point of the axle assembly (the junction of the snap-fit tongs with the body of the inner pin). Once all four of the tongs had succumbed to fatigue failure, the axle was free to disassemble.
Patients with radiographic evidence of polyethylene wear who are asymptomatic or refuse surgery need to be informed that, despite the pin-within-pin snap-fit articulation of the Coonrad-Morrey implant, it could disassemble and require revision surgery.
Note: The authors acknowledge Zimmer UK for provision of data and the image of the axle assembly.