Some patients are prone to instability. Those who have greater preoperative deformity, especially if compounded by extra-articular deformity or dynamic aberrations of gait, will require large surgical corrections and aggressive ligament releases and may be difficult to stabilize [8].

Several factors can produce instability after TKA. The most important are the following [5]: ligament imbalance, component disalignment, component failure, implant design, mediolateral instability, bone loss from overresection of the distal femur, bone loss from femoral or tibial component loosening, soft tissue laxity of the medial and lateral collateral ligaments, connective tissue disorders (rheumatoid arthritis or Ehlers-Danlos syndrome), inaccurate femoral or tibial bone resection, and collateral ligament imbalance (underrelease, overrelease, or traumatic disruption).

Specific patient-related risk factors are a large surgical correction including an aggressive ligament release, general or regional neuromuscular pathology (quadriceps weakness inducing recurvatum or weak hip abductors that impart a medial thrust to the knee), and hip or foot deformities typified by tibial posterior rupture and pes planus, inducing valgus moments at the knee. Clinical obesity is also a risk factor because it complicates surgical exposure, jeopardizes the collateral ligaments (8 % incidence of avulsion of the MCL in obese patients), and makes it difficult to appreciate component position [1, 8].

Instability of the knee can be prevented in most cases with an adequate selection of implants and a good surgical technique. Preoperative physical examination will allow us to evaluate the state of the LCL, MCL, and PCL in order to select the adequate implant for each patient.

PS implants should be used in those patients with PCL insufficiency and in those with increased risk of posterior instability (rheumatoid arthritis, previous patellectomy, or the need to liberate the PCL to correct a ligamentous imbalance, flexion contracture, or previous tibial osteotomy).

If the choice is made to preserve the PCL, it is important to take special care in maintaining its integrity when the tibial cut is made. In case of doubt, it is preferable to convert the arthroplasty to a PS design. Careful attention to the balance of soft tissue and the correct implantation of the components in every plane, including the rotation of the femoral component, is paramount to achieve symmetric spaces on flexion and extension.

In some patients with marked instability (knee with valgus and complete insufficiency of the PCL, poliomyelitis, or Charcot arthropathy), a primary constrained or linked implant can be indicated.

Conservative treatment can be useful in a small percentage of patients with knee instability. Orthotics utilization and rehabilitation programs are effective to strengthen the quadriceps and the hamstring and reduce the symptoms of some patients with mild and moderate instability. However, in many cases, it is necessary to turn to surgical treatment, especially if other alterations are noted such as disalignment of the components, deterioration, or loosening [2].

Most of the patients with KPI require surgical treatment, and the use of preoperative planning is paramount. An implant with the required constraint can be determined preoperatively [10]. Planning for a stable revision TKA must include not only how to “stabilize” the knee but how to eliminate the forces of destruction: disalignment and gap imbalance. Unchecked, these forces will ultimately destroy any constrained device, hinged or nonhinged by breakage or loosening.

Revision surgery for instability requires control over the mechanical axis of the lower limb, equalization of the flexion and extension gaps, assessment of ligament integrity, and access to constrained implants if necessary. As ever, diagnosis precedes successful treatment [8].

As a general rule, it is recommended that the minimum amount of constraint necessary to achieve stability should be used. With many choices of component designs and levels of constraint, it can be a very difficult process to select the optimum implant for a given patient [11].

CR designs represent the least amount of component constraint. This translates to the presence of good quality bone with minimal defects, intact soft tissues, and a PCL that remains functional and balanced. In most revision situations, CR implants are not indicated.

The next level in constraint is cruciate substitution (PS designs); this design mechanically substitutes for PCL function. Many people find this option easier and more forgiving because all the technical and judgment issues of balancing the PCL are eliminated. There is no gain in varus-valgus stability and, realistically speaking, minimal rotational stability. Thus, for a PS implant to succeed, a functional soft tissue envelope is needed to provide varus-valgus stability. However, the need for good flexion-extension balancing is also important, because a residually loose flexion space can result in posterior tibiofemoral dislocation.

The next level of constraint is nonlinked hinge implant such as the VVC (varus-valgus constrained) design or the CCK (constrained condylar knee) implant. Such components provide a significant degree of rotational control and more significantly a great deal of constraint to varus-valgus angulation. The trade-off is the theoretical disadvantage of increased stress transmission to the component-bone interfaces. Because these implants limit varus-valgus angulation between the femoral and tibial components, it would seem intuitive that they could be used in cases of severe medial or lateral instability. One must not forget that severe flexion instability is still a limitation for these implants [12].

Less constrained components have severe limitations in the absence of collateral soft tissue support or in the presence of gross flexion-extension instability. Unfortunately, going to the highest degree of constraint (a hinged or linked implant) has historically produced disappointing results, predominantly because of implant loosening, significant patellar pain, and high infection rates. However, newer rotating hinge designs have produced more encouraging clinical results [12] (Fig. 23.1).

Barrack et al. found no revision or radiographic failure at longer follow-up [13, 14]. Similar promising results have been reported by Westrich and associates using a different modern hinged implant [15]. The current potential indications for the use of a rotating hinge are the following [14]: MCL disruption, massive bone loss for the distal femur, proximal tibia (including collateral ligament origin or insertion), comminuted distal femur fracture in the elderly, distal femoral nonunion or malunion, extensor mechanism disruption requiring reconstruction in an unstable knee, and ankylosis requiring a femoral peel exposure with moderate or severe residual flexion-extension gap imbalance.