Preoperative assessment of the existing implant aids the surgeon in selecting the appropriate revision implant. The revision total knee surgeon will encounter patients implanted with a myriad of different prosthetic devices. Some revision TKA procedures involve revising only a portion of the existing implant, and thorough knowledge of the implant system in place is necessary to prepare adequately for these cases. A thorough knowledge of the implant type with available modularity and revision options for the implant in place is necessary. In addition, awareness of the specific implant design in place allows the surgeon to order implant-specific extraction devices, should these be indicated. If the previous arthroplasty was performed by another surgeon, the operative note and purchasing order should be obtained to verify implant type and size.

Instability remains one of the most common causes leading to revision to primary and revision TKA.¹ As outlined by Vince et al,⁵ there are seven causes of knee instability following TKA: (1) component loosening, (2) bone loss, (3) prosthetic breakage, (4) errors in component size or
position, (5) fracture, (6) wear, and (7) collateral ligament failure. The authors emphasized that only incompetency of the collateral ligaments typically requires revision with an increased-constraint implant.

There are four recognized patterns of instability: recurvatum, varus-valgus (mediolateral or coronal), flexion, and global. The treatment for each frequently differs and is discussed below.

Recurvatum is the rarest form of instability⁶ and may be the most challenging to properly treat. In fact, it has been argued that the best treatment for recurvatum is prevention, as the deformity seldom develops following knee replacement except in situations of neuromuscular dysfunction.⁷ Intraoperatively, either an excessive extension gap or collateral ligament instability⁸ may result in knee hyperextension. Recurvatum deformity in the preoperative patient is typically due to substantial quadriceps deficiency, frequently from neuromuscular disorders (classically poliomyelitis⁹), but may also be seen secondary to spinal stenosis, in patients with a fixed valgus deformity with a contracted iliotibial band, or associated with rheumatoid arthritis with marked collateral ligamentous laxity.⁸ With quadriceps weakness, the knee displays compensatory hyperextension to stabilize the limb. This mechanism may also result in progressive recurvatum deformity following primary or revision TKA. Primary arthroplasty in patients with recurvatum deformity and quadriceps weakness should be approached with caution. One advocated surgical technique includes underresection of the distal femur, resulting in a deliberate slight flexion contracture.⁷ A second surgical technique, recommended by Krackow and Weiss,⁶ is to advance the femoral attachments of the collateral ligaments proximally, thereby attempting to recreate the normal tensioning effect of the collaterals during knee extension. Recurvatum presents with substantial treatment difficulty during revision TKA. Deformity secondary to true paralysis of the quadriceps is often best treated with arthrodesis rather than arthroplasty.⁵ Arthroplasty treatment in the setting of quadriceps weakness frequently requires the use of rotating hinged constraint⁹,¹⁰,¹¹; however, there is concern about excessive forces imposed with a hyperextension stop, potentially leading to early implant failure,⁵ and a “three-step” revision arthroplasty technique has been advocated⁵,¹² which involves reestablishing the tibial platform, then stabilizing the knee in flexion, followed by stabilization in extension.

Instability in the coronal plane (varus-valgus instability) is a much more commonly encountered clinical entity following primary TKA. This is typically due to ligament imbalance, component malpositioning, excessive bone loss, or component failure. When performing a clinical assessment of knee balance, it is important to examine the knee in both extension and flexion. As varus-valgus instability is often associated with preoperative deformity, it is important to analyze the original preoperative imaging studies. Intraoperatively, technical errors in ligament balancing, asymmetric bone resection, iatrogenic ligament injury may all lead to an unstable knee in the coronal plane. Postoperative attenuation of ligamentous soft tissues may result in a delayed diagnosis.

Flexion instability is another commonly encountered complication following primary TKA, where laxity is felt to be due to an excessively loose flexion gap. Flexion instability has historically been associated with the use of cruciate-retaining (CL) components¹³ and may have been underdiagnosed in patients with this construct.⁷ This pattern of instability may be symmetric involving both the medial and lateral aspects of the flexion gap or asymmetric, often associated with malrotation of the femoral component. Patients will typically complain of pain while navigating stairs (especially descent), have point tenderness over the pes anserinus bursa, have a positive posterior drawer test (or lag test), and yet demonstrate excellent flexion upon range-of-motion testing.⁵ Flexion instability may manifest with the knee positioned in 90° of flexion or in mid-flexion. As intuited from the descriptive title of this clinical complication, mid-flexion instability occurs at intermediate ranges of knee flexion between extension and 90°, with the knee perceived to be otherwise stable.

There is a paucity of outcome studies following revision arthroplasty for TKA instability. Functional improvement has been demonstrated to improve following revision for instability.¹⁴ There are a few investigations which report good outcomes following revision of CL TKA to posterior-stabilized TKA.¹³,¹⁵,¹⁶ However, the specific surgical treatment for instability must be carefully chosen. For example, polyethylene liner exchange alone has been reported to have a high incidence of postrevision failure.¹³,¹⁶,¹⁷

Most of the causes of TKA failure will result in some degree of pathologic bone loss, whether it be from infection, osteolysis, osteonecrosis, aseptic loosening, stress shielding, or mechanical failure.¹⁸ The management of bone loss in TKA is dictated by the extent and location of bone deficiency. In revision TKA, significant bone loss is often encountered, and the extent is usually more than preoperative radiographs would indicate.

While preoperative radiographs may provide some clues as to the extent of bone defect, accurate determination of bone deficiency typically occurs during revision TKA after the previous components have been removed, as the removal of well-fixed implants may accentuate the preoperative bone loss.

Infection following TKA is a devastating complication. As previously stated, it is the most common cause of early TKA failure, and one of the more common reasons
for revision TKA at all time points.¹ Treatment for infection is nearly universally surgical. While acute infections may be appropriately treated with irrigation, débridement, and polyethylene liner exchange, this treatment strategy is inappropriate for more chronic infections. For these chronic infections, there remains no clear consensus supporting treatment with single- or two-stage revisions. Two-stage revision procedures typically employ the use of high-dose antibiotic-impregnated bone cement as part of either a static or articulating spacer. Articulating spacers are typically either fashioned using primary TKA components with a liberal use of antibiotic cement or with the use of systems which allow for molded acrylic implants. During both single- and two-stage procedures, preservation of bone stock is of paramount importance for future reconstructive efforts. In general, implant selection following eradication of infection in TKA is similar as for other indications for revision TKA.

Although the principles of revision TKA are similar to those of primary arthroplasty, numerous additional difficulties are often encountered, including soft-tissue scarring, bone loss, flexion-extension gap imbalance, ligamentous instability, and disturbance of the anatomic joint line. To deal with these difficulties, use of a revision implant system, which includes various levels of prosthetic constraint, augmentations, and diaphysis-engaging stems, is imperative.³ The surgeon’s experience with the implant design system selected is paramount in ensuring the success of the revision arthroplasty.

In choosing a revision implant system, the surgeon must consider multiple variables. The coronal and sagittal geometry of the implant must be appropriate to provide proper kinematic function. Prostheses with increasing levels of constraint, ranging from posterior cruciate retention to posterior cruciate substitution, varus-valgus constraint, and hinged designs must be available. A vast array of implant designs is available which may have differences in eventual kinematic function. For example, cam-and-post mechanisms of posterior cruciate-substituting (PS) designs vary widely in size, shape, and position of the cam and post in addition to the degree of flexion at which cam-post engagement occurs. These differences are reflected in variable levels of implant performance. The locking mechanism for polyethylene bearing fixation to the tibial tray should be evaluated. A thorough knowledge of each implant type allows the surgeon to choose the appropriate implant for each individual patient. Because of the complexity of each procedure, no single implant type can be used for all cases.

In revision TKA, isolated single-component revision is typically reserved for situations of polyethylene liner exchange or patellar revision. Isolated polyethylene exchange has demonstrated good short-term success when performed for wear and early osteolysis. Griffin et al¹⁹ reported on the short-term outcomes following isolated polyethylene exchange in 68 press-fit condylar TKAs (PFC; DePuy Synthes, J&J, Warsaw, IN). At a minimum of 24 months, there were 16% failures, the majority being due to aseptic loosening; however, 97% of subjects in this study did not demonstrate any perceivable progression of osteolytic lesions. There is some evidence to suggest that isolated polyethylene exchange for severe wear, however, is fraught with complications.²⁰ Furthermore, isolated polyethylene liner exchange for reasons other than wear are frequently associated with poor outcomes. Babis et al¹⁷ demonstrated particularly poor outcomes of isolated polyethylene exchange in the setting of knee stiffness. All patients in this study where either severely painful or required revision at a mean follow-up of 4.2 years. Brooks et al²¹ reported nearly a 30% failure rate for isolated polyethylene exchange for situations of knee instability. Additionally, it appears that time to revision plays a role in the success of isolated polyethylene exchange. Willson et al²² demonstrated that patients with an isolated polyethylene exchange within 3 years of the primary procedure were 3.8 times more likely to undergo re-revision than patients with exchange occurring more than 3 years after the index procedure.

In broad terms, cruciate-retaining (CR) implants provide for the least amount of mechanical constraint in TKA. Their use in revision TKA is limited and requires a surgeon skilled in appropriate balance of the posterior cruciate ligament (PCL). Preoperative assessment of both flexion and extension stability and the competence of the PCL is essential to the use of this type of implant. Principles of posterior cruciate retention are the same in revision TKA as in primary TKA. If present, the PCL may be retained if the operative surgeon can achieve flexion and extension balance with maintenance and competence of the PCL and restoration of the anatomic joint line. Advantages of the use of CR designs in revision knee arthroplasty are the preservation of bone stock and the theoretic advantages for retention of the PCL found in primary TKA. Disadvantages of the use of CR designs are the difficulty of balancing the PCL, restoring the joint line to its anatomic position, and obtaining adequate flexion stability in revision cases in which the competency of the existing PCL may be difficult to evaluate. Therefore, retaining or using a CR implant at the time of revision TKA is infrequently indicated. One possible indication would
be for a simple isolated polyethylene exchange (which, as described above, already has a very narrow range of indication) in the setting of a well-functioning CR knee with an intact PCL. The PCL is frequently observed to be attenuated or grossly absent at the time of revision TKA, which necessitates the conversion to a cruciate-substituting construct. Of particular note, patients with inflammatory arthritis such as rheumatoid arthritis have demonstrated higher rates of complications following CR TKA, and such implants should be utilized with caution in this population in the revision setting.²³

Considering the increased difficulty of determining PCL integrity (continuity, tension, and histologic damage) in revision TKA cases, surgeons favoring avoidance of conversion to a PCL-substituting design (cam and post) have utilized ultracongruent polyethylene inserts to enhance sagittal-plane stability²⁴ (Fig. 66-1). While medial pivot designs have been utilized in revision TKA, meaningful reports documenting their clinical value in revision TKA is lacking in the literature.

PS devices remain the authors’ implant choice for the majority of revision TKA cases. Advantages of these designs include reliable substitution for an absent or incompetent PCL, easier correction of deformity, increased flexion stability, and increased range of motion²⁵,²⁶ secondary to forced posterior femoral rollback. This can be beneficial in those cases in which preoperative stiffness is problematic. PS TKA designs incorporate a cam-and-post mechanism to enhance flexion stability and posterior femoral rollback. There are multiple types of cam-and-post mechanisms available that differ in post size (height and width), shape, and sagittal-plane position. As mentioned earlier, these design variances are reflected in differing patterns of kinematic function. The surgeon must be aware of these differences and select a design that optimizes patient function while providing long-term durability. Historically, in the most commonly implanted PS designs, the cam and post do not engage until approximately 70° of knee flexion.²⁷ Therefore, the cam and post are not engaged during lesser flexion activities such as walking. Other PS TKA designs allow cam-and-post engagement as early as 30° of knee flexion. Advocates of these designs report earlier posterior femoral rollback and enhanced quadriceps function due to an increased quadriceps lever arm.²⁸ Polyethylene post wear in traditional PS TKA designs has been limited. It is not yet known if post wear will become problematic in designs that permit earlier cam-post engagement.

Potential problems encountered with use of PS TKA implants include posterior dislocation of the cam relative to the post,²⁹,³⁰ intercondylar fracture due to increased bone resection,³¹ polyethylene post wear or fracture,³²,³³ and an increased incidence of patellar clunk syndrome.³⁴,³⁵ While mechanically enhancing flexion stability, the surgeon must still strictly adhere to the principle of obtaining flexion-extension gap balance to lessen the risk of dislocation. The risk of femoral condylar fracture is typically increased in the multiply revised TKA in which excessive distal femoral bone resection has previously been performed or in cases with substantial distal femoral osteolysis. The risk of condylar fracture is greatest medially because the medial metaphyseal contour transitions more abruptly to the diaphysis than the lateral contour. Because of this, the amount of bone remaining at the proximal margin of the intercondylar notch resection required for PS TKA designs is less medially. The risk of condylar fracture is enhanced in smaller subjects with reduced bone mass and those with significant osteopenia or osteolysis. This is of particular importance in certain TKA designs in which the intercondylar box size is not proportionally reduced with smaller implant sizes.