Introduction:

Clinical instability is the main contributing factor for revision knee arthroplasties in 10 to 22 percent of cases (1). Of these, dislocations of the total knee replacement implants are rare, estimated to be 0.15 to 0.5 percent after TKA operations with posterior stabilized designs (2). Even for failure of polyethylene inserts, the most common location of fracture is at the tibial post following repetitive stresses (3). We present an unusual case of delayed fracture of isolated medial anterior locking wire and polyethylene tab of the tibial insert in a 66-year-old female patient. This case highlights the importance of careful polyethylene handling during insertion, maintaining optimal implant positioning and ligamental balancing, and early recognition of subtle clinical features of a dislocated total knee arthroplasty.

Case report:

A 66-year-old lady was followed up at our centre for right knee mechanical pain. There was no history of injury, knee surgery or knee injection in the past. She had a past history of old ischemic stroke with right-sided hemiplegia and was put on Aspirin. She could walk with stick for 15 minutes. On examination, her right knee active range of motion was 20 to 90 degrees, with 20 degrees varus deformity and varus thrusting. X-ray of right knee showed tricompartmental osteoarthritis. Her right quadriceps power was grade 3 of 5. Distal sensation and proprioception of right foot was preserved. Her activity of daily living was independent. She opted for right TKA after failing of conservative treatment.

Total knee arthroplasty under general anesthesia was performed. The operation time was 115 minutes. Intra-operative findings included tricompartmental osteoarthritis of right knee, with passive motion of 15 to 90 degrees and a flexion contracture of 15 degrees. Medial parapatellar incision was used. Superficial and deep medial collateral ligaments were released from tibial insertion by sharp dissection from proximal tibia, till posteromedial corner was reached. Stryker (Triathlon) system with Articular Surface Mounted (ASM) Knee Navigation was used to determine bone cut. Femur cut was 8mm from lateral condyle and tibial cut was 9mm from lateral plateau. Tibial tray rotation was determined by “best-fit” following insertion of femur and tibial trial with knee in extended position. Flexion and extension gaps were checked before final implant insertion. Cemented size 2 femoral and size 1 tibial posterior-stabilized implants were used. Tibial polyethylene thickness of 9mm was used. The patella was not replaced. Range of motion was 5 to 100 degrees before skin closure.

Continuous Passive Motion (CPM) was initiated after surgery. Physiotherapy with full-weight walking was commenced. The wound healed well. Patient could walk with quadripod with one assistant for 20 meters. Barthel Index improved from 43 to 66 after operation. Patient was discharged one month after surgery.

Upon initial follow-up at 2 months after surgery, the patient could walk with quadripod indoors, and remain wheelchair-bound outdoors. Passive range of motion of right knee was 20 to 90 degrees. No gross instability was observed. X-ray of right knee showed normal implant alignment and positioning. The patient continued to maintain regular follow-up. One year and two months after the operation, she developed insidious onset of right knee discomfort and stiffness. She became wheelchair-bound six months after the onset of symptom. There was no injury all along. There was no sudden jerking sensation of right knee. On examination of the right knee( 20 months after the operation), posterior subluxation of tibia was observed and it was not reducible with anterior drawer maneuver (Figure 1). Passive range of motion was 30 to 80 degrees with pain. Valgus-varus stress test showed gross instability at 30 and 90 degrees of knee flexion. X-ray of right knee showed anterior dislocation of femur implant and polyethylene insert over tibial tray, lateral tibial tray translation, with a broken anterior locking wire seen on anteromedial side of tibial tray (Figure 2). There were no radiolucent lines at bone-implant interface. The patient was admitted clinically for right knee revision operation.

Figure 1

Clinical photo showing posterior tibial subluxation of right knee, 20 months after total knee arthroplasty

Figure 2

X-ray showing Lateral and posterior translation of Tibial baseplate. Fractured metal locking wire and displaced polyethylene are also evident.

Right knee TKA revision operation was performed. Intra-operatively, the polyethylene insert was found protruding anteriorly from the tibial tray (Figure 3). Part of the locking wire and the medial polyethylene tab housing the locking wire were broken. The lateral locking mechanism was intact.

Figure 3

Figure 3a: Polyethylene insert found protruding anteriorly from tibial tray upon surgical exploration. Figure 3b: Balanced extension gap. Figure 3c: Tight lateral gap and loose medial gap during knee flexion. Figure 3d: Polyethylene insert hinged out from the medial tibial baseplate as a result of imbalanced flexion gap. Blue arrows: forces directed to polyethylene insert, causing an anti-clockwise moment arm which hinges out and causes failure of medial locking tab.

The undersurface of the polyethylene insert was found seated over the anterior rim of the tibial baseplate, causing backside wear especially over the medial side (Figure 4). On the contrary, the tibial post, polyethylene articulating surface, and the metal components including tibial baseplate locking tabs were intact. The metal components were stable. There was no evidence of infection. Removal of the polyethylene insert, the broken polyethylene piece and the locking wire was performed, followed by debridement of any abundant meniscal rims, irrigation and exchange to a new liner of same 9mm height. The extension gap after reduction was acceptable and balanced over both medial and lateral side. Knee flexion showed a widened medial gap compared to the lateral side (Figure 3). The femur metal implant was externally rotated. Both collateral ligaments were intact.

Figure 4

Figure 4a: Superior view of retrieved polyethylene insert, broken locking tab and wire. Figure 4b: Anterior view of polyethylene insert. Note the medial locking tab was broken. Figure 4c: Inferior view of polyethylene insert, showing increased backside wear especially medial side.

A hinged knee brace at 0 to 90 degrees was given for post-operative support. The patient could walk with quadripod for 10 meters under slight support. Barthel Index was 77. She was discharged one month after surgery.

Figure 5

Figure 5. Alignment restoration in coronal and sagittal plane, after polyethylene revision.

She was last seen in outpatient clinic 2 months after the revision operation. There was no tenderness around the right knee. She was able to walk with quadripod for 30 minutes. On physical examination, right knee active range of motion was 10 to 90 degrees. No clinical instability or posterior sagging was observed. X-ray of right knee showed aligned implants (Figure 5). She refused further revision surgery for the femoral component.

Discussion:

Metal-backed tibial components were introduced since 1980s. The metallic baseplate reduces tensile stresses and the maximum compressive stresses on the bone under load. Isolated liner exchange is possible in revision operations, providing ease while refraining from disturbing the bone stock and environment at implant-bone interface. Drawbacks of this design include having one more interface to failure, wear and dislocation.

In order to avoid dislocation at polyethylene-baseplate interface, different designs have been utilized. There are two commonly used liner locking systems: peripheral locking and dovetail locking mechanism. Implants with peripheral locking mechanism, e.g. Stryker Triathlon, utilizes rim lock along the periphery of the tibial tray, allowing a snap fit of the inlay by a tongue and groove lock. On the other hand, the dovetail locking mechanism, e.g. Zimmer NexGen and S&N Genesis II, consists of a beveled lip within the polyethylene inlay, which fits into the posterior wall of the tibial component, and blocks the inlay against a less prominent anterior metal rim. There was no literature review on the locking mechanism stability in view of the case rarity. However, polyethylene inserts with peripheral locking mechanisms appear to have reduced backside wear from a consecutive series of 102 retrieved inlays in vivo. (4) The Stryker Triathlon system offers additional locking mechanism with anteromedial and anterolateral locking wires and tabs, an anti-rotation island posteriorly, with peripheral relief for ease of insertion assembly.

The incidence of spontaneous polyethylene dislocation from its tibial baseplate remains low. We found only 2 case series and a number of case reports on polyethylene dislocation in posterior-stabilized (PS) implants (5-12). External trauma, repetitive flexion movements, implant design including a shallow tibial base and narrow track for polyethylene fitting, and technical errors including incomplete seating and retainment of posterior osteophytes have all been postulated. On the contrary, polyethylene dislocation appears more common in cruciate retaining (CR) designs. A case series of 22 painful primary posterior CR total knee arthroplasties revised have their flexion instability improved after revising to PS implant. The author proposed 3 reasons: 1. Knees with over-resected posterior condyles or excessive tibial slope can have excessive flexion gap. 2. Knees that were too tight in flexion early after CR implant may later rupture the posterior cruciate ligament. 3. Flexion instability may result from a chronic attritional rupture of the posterior cruciate ligament (13). 

Fracture-dislocation of the polyethylene inserts has uncommonly been described. Of these, fracture tibial posts in posterior stabilized TKA are the most common, with reasons including component mal-alignment, inappropriate soft tissue balancing, trauma, and suboptimal component designs being postulated. On a thorough search of published literatures, there is only one case series and one case report on fracture anterior locking flange of polyethylene liner following primary knee replacement, using Depuy Press-fit Conylar (PFC) prosthesis (14,15). We describe what we believe is the first report of asymmetrical unilateral breakage of the anterior locking tab-wire mechanism in the Stryker Triathlon TKA implant.

Similar to other case reports, there is no conclusive evidence causing the polyethylene locking mechanism failure. We believe the reason is multifactorial, with the primary reason being the imbalanced flexion gap seen during revision operation. Judging from the backside wear of the polyethylene undersurface, it is clear that the medial joint space is subjected to more frequent and larger stresses (Figure 4c). With a tight lateral and loose medial gap during knee flexion, the polyethylene insert experiences repetitive valgus stress, causing a coronal ‘see-saw’ effect and subsequent weakening of the locking mechanism, and with flexion causes an anterior translation force. The tibial post acts as a lever arm to external rotate and hinge out the insert from the medial tibial baseplate (Figure 3). This explains why the medial locking system fails but the lateral side is still intact. Failure of only one side of the locking system can lead to complete polyethylene dislocation. From the integrity of the tibial post and the relative sedentary lifestyle of our patient, it appears that the polyethylene failure is due to long term repetitive stress rather than a single trauma.

Failure of the anterior flange may also have occurred during initial polyethylene insertion, with the anterior medial tab being impacted against the metal locking tab when the insert is not properly positioned. Improper polyethylene positioning occurs especially with inadequate surgical exposure or with entrapment of soft tissues like the remaining meniscus. This leads to initial weakening and subsequent fracture of the locking mechanism after repetitive stress. On the contrary, manufacturing fault is less likely in our case, in view of the integrity of the articular surface and the lateral locking tab and wire. Subsurface delamination, which presents as shear cracks that propagate parallel to, but is confined to 2mm below the articular surface, is also not evident in our case.

Signs of tibial polyethylene dislodgement can be subtle. Clinically, persistent or increased pain, swelling, sensations of instability and gait difficulty have been reported. A useful clinical sign is the presence of posterior sagging, which also appears in other modes of TKA dislocations. It is important to assess patients’ neurovascular status as there were cases reporting concurrent neurovascular injuries requiring vascular interventions (16). Radiographic features include tibial implant subluxation with reference to the femur implant in both coronal and sagittal planes, and evidence of polyethylene subluxation shown by the lucent area without soft tissue shadow. In our particular case, the fractured metal locking wire provides a useful clue to polyethylene locking system fracture (Figure 2).

Good results achieved by revising the polyethylene insert only have been reported. However, revision of tibial and femoral components is required upon metal-on-metal contact, or failure of locking mechanism on the tibial baseplate. This is demonstrated by intra-operative movement between the pre-existing baseplate and the new polyethylene insert, which signifies the weakening of the locking mechanism. An author has also advocated the polyethylene insert fixation be enhanced by screw or metal clip (17).

Conclusion:

We report the first case of dislocation of polyethylene insert associated with fracture of its anterior locking flange in Stryker Triathlon TKA system. Proposed reasons include improper positioning of the polyethylene insert as well as imbalanced flexion gap created during bone cuts.