Adult Reconstruction, 1st Edition

Section II - Knee

Part C - Operative Treatment Methods

24

Unicompartmental Knee Arthroplasty

Mark W. Pagnano

Robert S. Rice

Unicompartmental knee arthroplasty (UKA) has experienced a resurgence of interest in the past decade. For a selected subgroup of patients with isolated advanced degenerative arthritis involving primarily the medial or lateral compartment of the knee, UKA may be the best surgical treatment option. Comparisons of the early outcomes of UKA with those of total knee arthroplasty (TKA) typically reveal faster recovery after UKA. In addition, often there is greater patient satisfaction with UKA because the knee feels more like a normal knee, possibly because of the preservation of both cruciate ligaments after UKA (Table 24-1). Comparisons of the early outcomes of UKA to those of osteotomy typically reveal faster recovery, more predictable pain relief, and fewer surgical complications after UKA. Progression of degenerative arthritis in the unresurfaced portions of the joint after UKA remains a mode of failure that is not faced after TKA. Controversy exists about the ability to predict, through physical exam or radiographs, those patients at risk for developing arthritis in other compartments after UKA. For that reason some surgeons remain reluctant to use UKA and instead rely on TKA for those patients who progress to require a knee arthroplasty. In large cohorts of patients, however, it is fair to conclude that the survivorship of modern UKA and modern TKA are essentially equivalent over the first decade after implantation.

Pathogenesis

Isolated advanced degenerative arthritis of the medial compartment of the knee is the most common indication for unicompartmental knee arthroplasty. The pathogenesis of isolated medial compartment disease is well recognized. Progressive loss of articular cartilage leads to varus malalignment of the limb, which then further overloads the articular cartilage and causes additional loss of articular cartilage over time. In most patients with an intact anterior cruciate ligament (ACL), the area of maximal articular cartilage loss is the anteromedial portion of the tibia. When the ACL is intact, most patients will have preservation of full-thickness articular cartilage on the posteromedial portion of the tibia. On the femoral side, almost all of the articular cartilage loss is from the distal femur, with the posterior femoral cartilage relatively well preserved. In patients without an ACL, the knee kinematics are altered substantially and the pattern of arthritis is markedly less predictable. In many, but not all, ACL-deficient patients, sufficient lateral compartment disease or patellofemoral compartment disease will be present such that a UKA is not appropriate.

Isolated advanced degenerative arthritis of the lateral compartment of the knee is decidedly less common than medial-sided disease. Most TKA studies suggest a 10-to-1 predominance of medial over lateral compartment disease, and most UKA studies suggest closer to 20-to-1 predominance of medial UKA versus lateral UKA. In many surgeons' experience, the patient with valgus deformity and lateral compartment disease often presents with concomitant anterior knee pain or patellofemoral radiographic findings that make UKA less appealing. Even in those patients with isolated lateral compartment disease, the pattern of degenerative change is less predictable than in patients with isolated medial disease. This likely reflects the more complex kinematics of the lateral compartment of the knee, which includes greater amounts of gliding and rolling than the medial side.

Diagnosis

Physical Examination and History

The ideal candidate for UKA is able to clearly pinpoint the medial (or lateral) joint line as the source of pain that prevents him or her from carrying out activities of daily living (Fig. 24-1). Those patients who have diffuse knee pain or who clearly identify anterior knee pain as substantially limiting likely will be served better with TKA. Specific anterior knee pain symptoms when squatting or standing from a seated position also would suggest TKA rather than UKA. As with any knee problem, care should be taken to exclude

P.165

hip disease or a neurologic cause for the pain. Patients with inflammatory arthritis are better suited to TKA than UKA. Considerable debate exists on how to factor age and body weight into the decision to proceed with UKA. Interestingly, at this time the available evidence suggests that weight does not affect early outcome or survivorship through the first decade. This may be because many obese patients are relatively sedentary. In distinction there is evidence from the Swedish Joint Registry that younger age is adversely correlated with survivorship; however, that applies to not just UKA but also TKA.

On physical exam the knee should flex >90 degrees and have no more than a 10- to 15-degree flexion contracture. More substantial flexion contractures typically can be corrected only partially with UKA. Varus or valgus deformity of >10 degrees is typically accompanied by degenerative changes in the other compartments of the knee that make UKA less predictable. Furthermore, varus/valgus deformity of >10 to 15 degrees often requires collateral ligament release at the time of surgery, which most authors have advised against during UKA. The stability of the ACL must be assessed preoperatively. A deficient ACL is a contraindication to the use of a mobile-bearing UKA design because the risk of bearing dislocation is substantial. Some authors suggest that a deficient ACL in a low-demand patient who has not experienced giving way episodes is not a contraindication to a fixed-bearing UKA. When UKA is selected for those low-demand ACL deficient patients, care should be taken not to increase the posterior slope of the tibial component. For active, high-demand patients and for those who have experienced symptomatic giving way episodes, an isolated UKA is contraindicated in the face of ACL deficiency. Some authors have described concomitant or sequential ACL reconstruction and UKA, but the data on that combination is limited.

Radiographic Features

A full-length standing radiograph including the hip-knee-ankle on a 3-foot film is useful. With that film the mechanical axis and anatomic axis can be calculated and the presence or absence of extra-articular bone deformity can be confirmed. On a standing anteroposterior (AP) view of the knee, the contralateral tibiofemoral compartment is examined for evidence of joint space narrowing or osteophyte formation. Some surgeons, particularly those in Europe, routinely obtain stress views of the knee in varus and valgus as part of the evaluation for UKA. These stress views can confirm the integrity of the opposite compartment and determine if adequate correction of the varus-valgus alignment can be obtained without collateral ligament release. On the lateral radiograph, superior and inferior pole patellar osteophytes can be observed. Axial views of the patella are used to grossly assess the patellofemoral articulation for evidence of subluxation or loss of articular cartilage. In the absence of symptoms, some surgeons will ignore the status of the patellofemoral joint; however most surgeons would regard the presence of bone-on-bone changes at the patellofemoral joint as a contraindication to UKA. The presence of diffuse chondrocalcinosis on x-ray films (particularly when accompanied by history or physical findings of recurrent effusion) is a contraindication to UKA.

Typically, plain radiographs and a targeted history and physical exam are sufficient to allow a definitive decision about whether UKA is appropriate. In rare circumstances MRI might be helpful in determining the status of the contralateral compartment or the ACL. MRI, however is helpful in patients with avascular necrosis for whom UKA is contemplated. Some surgeons make a distinction between patients with so-called spontaneous avascular necrosis (AVN) and patients with AVN secondary to corticosteroid use. Patients with spontaneous osteonecrosis typically have small areas of necrotic bone confined to the subchondral region, and those patients are often good candidates for UKA. Some patients with AVN secondary to steroid use have large, geographic avascular bone lesions that could compromise the fixation of the femoral or tibial component after UKA. MRI can be helpful in determining the depth and extent of that necrotic change. If it appears that after the predicted bone cuts a substantial portion of the UKA implant will rest on necrotic bone, then TKA may be a better choice (Fig. 24-2).

Treatment

Surgical Goals

Surgeons continue to debate what the appropriate postoperative limb alignment should be after UKA. Most, but not all, surgeons currently recommend that the limb remain slightly undercorrected after UKA. For the typical varus knee undergoing medial compartment UKA, this means leaving the limb with a mechanical axis that passes through the medial compartment just medial to the tibial spines. For most patients the postoperative anatomic femorotibial axis would thus measure 2 to 4 degrees of valgus as opposed to the normal 6 degrees of valgus. The rationale for slightly undercorrecting the mechanical axis is to avoid overloading the articular cartilage in the opposite compartment. Markedly undercorrecting the knee, however, is also inappropriate as

P.166

P.167

that will place excessive load on the UKA bearing and lead to failure owing to polyethylene wear. In both full extension and at 90 degrees of flexion, the femoral and tibial components should be parallel such that edge loading of the polyethylene does not occur. The knee should be balanced to incorporate 2 mm of laxity in both flexion and extension. The tibial component must not overhang medially where it can irritate the medial collateral ligament. The femoral component must not extend anteriorly beyond the edge of subchondral bone or it can impinge against the patella.

Techniques

Various techniques exist to implant contemporary UKA designs. Those techniques include noninstrumented, free-hand preparation through intramedullary, extramedullary, and computer-assisted instrumentation systems. Surgeons must understand the rationale for a given instrumentation system before using the system clinically. Contemporary UKA is often done through a so-called minimally invasive surgical approach (MIS). The MIS technique typically involves an 8- to 12-cm skin incision and a short medial arthrotomy that stops at the superior pole of the patella. A short split into the vastus medialis muscle can be made (mini midvastus approach) or alternatively the subvastus interval can be exploited if more exposure is needed. The patella does not need to be dislocated for UKA, and leaving the patella reduced in the trochlea helps the surgeon avoid some component orientation errors. When UKA is done using a traditional TKA approach with the patella everted and the knee flexed, the tibia tends to externally rotate and the medial flexion space tends to gap open, and that can lead to

P.168

component orientation problems. MIS techniques continue to be debated in the realm of TKA, but in UKA contemporary instruments are well suited to this approach. A portion of the retropatellar fat pad and the anterior horn of the medial meniscus can be excised for visualization early in the case. In midflexion the status of the ACL, the lateral compartment, and the patellofemoral joint are noted. Any intercondylar osteophytes can be removed from the notch to prevent impingement on the ACL, and patellar osteophytes can be debrided. The sequence of bone cuts is determined by the particular instrumentation system chosen by the surgeon. Typically, on the tibial side the emphasis is on minimal bone resection with at most 2 mm of bone removed from the most worn portion of the tibia. This cut is generally made perpendicular to the long axis of the tibia. The degree of posterior slope is most often 5 degrees but can vary based on patient and implant selection factors. For patients with ACL-deficient knees, less slope may be preferable. When an implant is used that has substantial sagittal plane conformity, then matching the posterior slope to the patient's anatomy is appropriate.

Most tibial instrumentation systems use a vertical free-hand cut from anterior to posterior, and this should be done as close to the medial tibial spine as possible without damaging the ACL. The surgeon should reference the tibial tubercle to avoid the tendency to internally rotate that sagittal plane cut. Typically, the largest tibial component that does not overhang should be selected. On the femoral side, most instrumentation systems resect the same thickness of bone (both distally and posteriorly) that will be replaced by the femoral implant. If an intramedullary cutting guide is used, the knee is brought to midflexion to facilitate access to the intramedullary canal. Care is taken to protect the patellar ligament and skin during this part of the procedure. The femoral component is sized from anterior to posterior such that 1 mm of subchondral bone is left exposed at the anterior edge of the component. That sizing will eliminate impingement of the femoral component with the patella even if the patient goes on to develop patellofemoral arthritis years later. In the medial-lateral direction, the femur should be centered over the tibial component without impingement into the notch and without overhang medially. The femoral component should be rotated at 90 degrees of flexion such that the femur and tibia are parallel, thus ensuring that edge loading of the femoral component will not occur. With a trial insert in place, the knee should be balanced with symmetric flexion and extension gaps of 2 mm. The overall mechanical alignment of the leg should be assessed with a cautery cord or long drop rod. If questions exist about component position or limb alignment, an intraoperative x-ray film or fluoroscopy can be used.

Complications

Complications after UKA can involve the entire spectrum of problems encountered with total knee arthroplasty including infection, bleeding, nerve injury, prosthetic loosening, wear, continued pain, thromboembolic disease, and bearing dislocation. The prevalence of infection after UKA has historically been equal to or slightly less than that after TKA. Substantial bleeding after contemporary UKA is uncommon, and it is rare for a patient to require blood transfusion after a single UKA. Injury to the peroneal or tibial nerves is rare after UKA and is substantially lower than that reported after upper tibial osteotomy. Prosthetic loosening, wear, or failure that requires revision surgery can be estimated to occur at a rate of 1% to 1.5% per year over the first decade. Slightly higher rates of failure have been observed in patients younger than 65 years of age compared with those older than 65 years according to the Swedish Joint Registry data and from the group in Oxford, England. Continued pain in the early period after UKA typically is the result of improper patient selection, although infection, early implant loosening, or tibial plateau fracture should be excluded. Late onset of pain can occur from progression of disease in the unresurfaced compartments of the knee, implant loosening, or from polyethylene wear with associated synovitis. Between 10 and 15 years after UKA, symptomatic patellofemoral arthritis has been reported in ≤10% of UKA patients in some series. The prevalence of deep venous thrombosis and pulmonary embolus has not been studied as well after UKA as after TKA, but the available evidence suggests lower prevalence of thromboembolic disease after UKA. For mobile-bearing designs, dislocation of the tibial bearing can occur, and the reported prevalence is 0.5% to 1.5%. Patients with a deficient ACL are at particular risk for bearing dislocation after mobile-bearing UKA. Fracture of the medial tibial plateau has been reported after UKA and is associated with the use of multiple pins to fix tibial cutting jigs to the proximal medial tibia. Similar fractures can occur after excessively deep tibial resections as well.

Results and Outcomes

Multiple studies demonstrate faster recovery after UKA than after TKA. Most studies reveal that the mean range of motion after UKA is substantially better than that after TKA even when accounting for differences in preoperative motion. One prospective randomized trial of UKA versus TKA demonstrated more excellent results after UKA, and those superior results were maintained at 5 years follow-up. Early series of UKA reported survivorship of 85% to 88% at 10 years. More recent series suggest 90% to 98% survivorship at 10 years, which may be attributable to the combination of more appropriate patient selection and improved instrumentation and technique. Most of these studies, however, have been done on elderly patients with a predominance of females over males, and that makes extrapolation of these data to the younger active middle-aged patient difficult. Several recent studies in younger, more active patients have been encouraging with 10 year survivorship of 90% to 92%. Those UKAs that require revision typically are divided equally between patients who fail because of disease progression in the unresurfaced compartments and those who fail because of loosening or wear of the UKA components. Early reports of conversion of the failed UKA to TKA suggested that substantial bone loss was encountered commonly and that these were difficult reoperations. In contrast, many authors now suggest that conversion of

P.169

the failed contemporary UKA to TKA is relatively straightforward. There are data to suggest that revision of a UKA to TKA is more reliable than revision of UKA to another UKA. Surgeons continue to disagree on whether conversion of a failed UKA to TKA is more or less difficult than conversion of a failed upper tibial osteotomy to TKA.

Postoperative Management

Postoperative pain can be improved by the injection of local anesthetic into the capsule and subcutaneous tissues prior to closing the wound. Patients can typically begin weight bearing as tolerated early after surgery and progress with activities as tolerated. Although some surgeons will perform UKA as an outpatient procedure, most patients are hospitalized for 1 to 3 days. Most surgeons now use some form of rapid rehabilitation protocol such that patients use ambulatory aids for a short period of time after surgery. Just as in TKA, these patients should work diligently early after surgery to regain maximal knee extension and flexion.

Patellofemoral Arthroplasty

The decidedly poor results with early patellofemoral arthroplasty designs has had a lasting influence on surgeons and has resulted in the continued limited use of patellofemoral arthroplasty. Nonetheless, there likely is a small subgroup of patients for whom contemporary patellofemoral arthroplasty is a good treatment option. For older patients with advanced patellofemoral arthritis, TKA has proved to be a reliable, reproducible, and durable procedure. For patients younger than 55 years of age who have substantial primary or posttraumatic patellofemoral degenerative arthritis without patellar malalignment, patellofemoral arthroplasty may be considered. Furthermore, patellofemoral arthroplasty can be considered in patients with arthritis secondary to trochlea dysplasia. Because a patellofemoral arthroplasty allows retention of both cruciate ligaments, the knee kinematics are better preserved as compared with TKA, and thus patients may perceive the knee to feel more normal. Although contemporary implant designs do offer improvements over historical designs, patellofemoral arthroplasty remains a technically demanding operation. Implant malposition can result in prosthetic impingement, pain, and extensor mechanism instability problems. Implant loosening with contemporary cemented patellofemoral arthroplasty has not proved to be common. With longer-term follow-up, however, a substantial number of patients (25% at 15 years) will go on to develop symptomatic degenerative arthritis in the tibiofemoral articulation. Conversion of patellofemoral arthroplasty to TKA typically is not particularly difficult.

Suggested Readings

Argenson JN, Komistek RD, Aubaniac JM, et al. In vivo determination of knee kinematics for subjects implanted with a unicompartmental arthroplasty. J Arthroplasty. 2002;17:1049–1053.

Cartier P, Sanouiller JL, Khefacha A. Long-term results with the first patellofemoral prosthesis. Clin Orthop Relat Res. 2005;436:47–54.

Langdown AJ, Pandit H, Price AJ, et al. Oxford medial unicompartmental arthroplasty for focal spontaneous osteonecrosis of the knee.Acta Orthop. 2005;76:688–692.

Lonner JH. Patellofemoral arthroplasty: pros, cons, and design considerations. Clin Orthop Relat Res. 2004;428:158–165.

Newman JH, Ackroyd CE, Shah NA. Unicompartmental or total knee replacement? Five year results of a prospective randomized trial of 102 osteoarthritic knees with unicompartmental arthritis. J Bone Joint Surg Br. 1998;80:862–865.

Pandit H, Beard DJ, Jenkins C, et al. Combined anterior cruciate reconstruction and Oxford unicompartmental knee arthroplasty. J Bone Joint Surg Br. 2006;88:887–892.

Pennington DW, Swienckowski JJ, Lutes WB, et al. Unicompartmental knee arthroplasty in patients sixty years of age or younger. J Bone Joint Surg Am. 2003;85:1968–1973.

Price AJ, Dodd CA, Svard UG, et al. Oxford medial unicompartmental knee arthroplasty in patients younger and older than 60 years of age. J Bone Joint Surg Br. 2005;87:1488–1492.

Springer BD, Scott RD, Thornhill TS. Conversion of failed unicompartmental knee arthroplasty to TKA. Clin Orthop Relat Res. 2006;446:214–220.

Walton NP, Jahroni I, Lewis PL, et al. Patient-perceived outcomes and return to sport and work: TKA versus mini-incision unicompartmental knee arthroplasty. J Knee Surg. 2006;19:112–116.