Etiology and classification of the unstable total knee arthroplasty

When bone loss occurs, this can lead to implant loosening or settling and a gap imbalance and instability; furthermore, bone loss can result from not just component loosening but also from overresection of the distal femur [15].

Although it is estimated that instability accounts for 20–30% [3, 13] of revision TKAs, McNabb et al [3] note that actual symptomatic instability is seen in less than 2% of patients after primary TKA. Garin explains, instability can be difficult to diagnose and some patients may be asymptomatic, so a good clinical examination is essential. The symptoms of instability are varied and the timing of the appearance of symptoms can provide some indication of the cause [3]; in Garin's experience, early instability is caused mainly by issues during the surgical procedure or acute trauma, and late instability is usually caused because of polyethylene wear or a trauma. Furthermore, acute instability presents in the initial postoperative period and chronic instability later [12]. When evaluating an unstable knee, it is important to understand whether the patient experienced any trauma to the knee after the TKA [3], or how the symptoms are perceived as this may also provide insight into the cause. Patients may experience instability—which can be reported as 'mild' or may be severe such as dislocation [13]—while climbing stairs [13] or transferring their position [3], which is seen particularly in flexion instability [13]. In such cases, pain may occur, the knee could buckle or give way, and over time there could be the appearance of a weight-bearing deformity [3]. In the case of flexion instability, the patient may have recurrent joint effusions, diffuse periretinacular tenderness, or pain at the site of tendinous attachment [13]. Instability may also be suspected, as discussed by Chang et al [16], if a patient presents with a stiff-legged or varus or valgus thrust gait, hyperextension of the knee during stance phase, or an abnormal or rotated foot progression angle during walking.

In the initial postoperative period, ie, weeks to months following the primary TKA [15], instability can have different causes. It may be due to asymmetric ligament balance or failure to balance the flexion-extension gaps, malalignment of the prosthetic components, or patellar fracture or ligament injury occurring during the treatment [3, 12, 15]. In one study, for instance, medial collateral ligament injury was reported in 8% of TKA patients [17]; additionally, rupture of the posterior cruciate ligament or patellar tendon rupture may occur [15]. If, however, the instability presents later and the primary TKA was initially stable, there may be pseudolaxity as a result of asymmetric polyethylene wear, which is a common cause, or attrition of the collateral or posterior cruciate ligaments may have led to true laxity [3]. In such cases the instability may appear suddenly after a stable period [12]. Polyethylene wear often occurs due to malalignment, and medial wear can lead to medial collateral ligament contracture resulting in varus deformity and instability [15]. An analysis of the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) [6] indicated that most revisions for instability took place 2 years after the primary procedure, implying that instability resulted from technical or prosthetic factors [11]. There is no clear timeline to determine an early or late instability, according to Garin, and some authors determine that early instability occurs weeks to months after TKA; but, the etiology of the instability can determine this timeline, as discussed above.

Risk factors for instability in total knee arthroplasty

Several risk factors for instability after TKA have been identified in various studies [3, 11, 13, 15]. Data from the AOANJRR show that women and patients older than 65 years were more at risk of revision for instability [11]. There is also a risk of instability in patients who have severe deformity before TKA and thus require a large surgical correction with associated aggressive ligament release [13, 15]. Patients who have neuromuscular pathology or recurvatum, and those with significant hip, foot, or ankle deformities [3], are also at risk of developing instability [15]. Such deformities occurring as a result of posterior tibial tendon rupture and pes planus cause valgus movement at the knee [15]. Other patient-specific risk factors for instability include rheumatoid arthritis, where ligamentous rupture during surgery may occur, connective tissue disease, or severe osteoporosis [13]. Additionally, patients with obesity are at risk of instability because of the complexities of surgery in these patients, such as surgical exposure, possible damage to the collateral ligaments, and difficulties in component positioning [3, 15].

Risk factors specifically for midflexion instability after TKA were assessed in a systematic literature review by Vajapey et al [18]. Similar to the categorization of the causes of instability, Vajapey et al also established the three categories of risk factors after examining the results of 18 studies: patient-specific, implant-specific, and technique-specific (Table 2) [18].

In the analysis of AOANJRR data by Lewis et al [11], it was noted that over half of the TKA revisions that took place for instability included insert exchanges; higher revision rates were seen when inserts thicker than 14 mm were used in the primary TKA, and in mobile-bearing prosthetic designs. A more detailed analysis of the data also revealed that minimally stabilized prosthetic designs, which have less inherent constraint, had a lower risk of revision for instability [11]. Furthermore, Lewis et al noted that there was more instability seen with medial pivot or posterior stabilized designs, the use of which requires resection of the posterior cruciate ligament, which may contribute to stability [11]. There was also a high rate of second revision seen in patients who underwent revision TKA, where the most common cause was recurrent instability [11].

Classification of the unstable total knee arthroplasty

Classification systems of instability have evolved over time [12] with different classification systems having been proposed [11, 12, 14, 19, 20]. The idea of midflexion instability is relatively modern and is still an evolving concept [21], and even instability as a diagnosis itself has caused some controversy [11]. Instability can be classified into extension instability (varus/valgus), flexion instability, midflexion instability, genu recurvatum, and global multiplanar instability [12, 20].

Extension instability

Extension instability or varus/valgus instability includes both symmetric and asymmetric instability, with the latter being more common [12, 20]. Patients may exhibit a balanced knee in flexion but a loose knee in extension [12]. Symmetric extension instability can be caused by excessive distal femoral bone resection leading to an extension gap that is larger than the flexion gap (Figure 2) [12, 20]. Asymmetric extension instability may be caused through incomplete correction of a deformity in the coronal plane in extension, which leads to ligamentous asymmetry [20], or a collateral ligament injury occurring during the primary TKA which was unnoticed [12].