The effect of alignment of the knee on the outcome of unicompartmental knee replacement.

Many authors have recommended undercorrection of the deformity when carrying out unicompartmental knee arthroplasty (UKA). The isolated effect of alignment of the knee on the outcome of UKA has, however, received little attention. We reviewed 185 UKAs at a minimum of five years after surgery. They had been carried out by a single surgeon using metal-backed tibial components in the management of arthritis of the medial compartment. We measured the tibiofemoral angle (TFA) before and at four months after operation and at the most recent assessment. The amount of correction of the TFA and any subsequent loss were recorded. While adjusting for the effects of age, weight and gender of the patients and the type and thickness of the implants, the mean correction was significantly less for those with a Marmor rating of failure (6.8 degrees) than for those rated excellent (9.2 degrees). The mean correction was also significantly less for patients with a Marmor rating of failure (6.8 degrees) than for those rated poor (11.1 degrees). The mean correction for the UKAs which were revised (6.6 degrees) was significantly less than for those not revised (9.1 degrees). Additionally, revised UKAs had a significantly higher percentage (63%) of thinner tibial implants (<8 mm) than the surviving UKAs (27%). These findings suggest that undercorrection of the TFA in UKA of the medial compartment should be avoided, particularly if a thin tibial polyethylene insert is used.

Revision of failed unicompartmental knee arthroplasty.

Little information has been published on the salvage of failed unicompartmental knee arthroplasty. The current authors examined the failure mechanisms, complexity of surgery, and complications in such revisions. Of 39 consecutive unicompartmental knee revisions, two patients died within 2 years of surgery, four had inadequate followup, and one was excluded because of infection before revision. Therefore, 30 patients (32 knees) were followed up a mean 53 months. The patients were a mean age of 59 years at the time of unicompartmental knee arthroplasty and were predominantly male (60%). Revisions were done 9 to 204 months after the unicompartmental knee arthroplasty. The predominant failure mechanism was polyethylene wear; however, nine failed because of loosening. The mean polyethylene thickness of the failed arthroplasties was 7.3 mm. Revision procedures were straightforward. Ten patients required local autograft but no allografts were used. Primary femoral components were used in all patients; 25 were cruciate-retaining. Fourteen patients had stemmed tibial components; eight had wedge augments. The mean range of motion after revision was 111 degrees. Six patients had complications not requiring intervention. Three patients had rerevision surgery for polyethylene wear at 76, 77, and 102 months, respectively, two of whom required polyethylene insert and patellar component revision. The third patient also required tibial component revision for osteolysis. Although this is a small series, the simplicity of the procedure and the complications encountered compare favorably with those of total knee revision.

Does implant selection affect outcome of revision knee arthroplasty?

We reviewed 139 consecutive femoral or tibial revision knee arthroplasties to determine if the outcome of revision knee arthroplasty using revision implant systems was superior to revisions using primary implant systems. Group 1 (n = 42) consisted of revisions performed with implants designed for primary total knee arthroplasty. Group 2 (n = 42) consisted of revisions performed with modified primary components. Group 3 (n = 55) consisted of revisions performed with components specifically designed for revision arthroplasty. The implant status was known in 123 of 139 knees at a mean follow-up of 7 years (range, 5-12 years). The implant-related failure rate, defined as reoperation requiring component revision or removal, was 26% for group 1, 14% for group 2, and 6% for group 3 (P<.05). Revision implants exhibited superior performance and durability despite their use in more difficult reconstructions. The improved longevity of revision implants justifies the evolution of modular revision components.

Clinical manifestations of a sometimes silent disease.

Although the host tissue response to debris is rarely symptomatic, the clinical manifestations of implant wear are profound. Even the most experienced orthopedic surgeon can be challenged in managing this sometimes silent disease.

Bone loss with revision total knee arthroplasty: defect classification and alternatives for reconstruction.

Although the best method for managing large bone defects has not been established, the variables to consider are: (1) implant constraint (posterior-stabilized, varus-valgus-constrained, rotating hinge); (2) stem configuration (straight versus tapered, standard, or long-stemmed); (3) stem fixation (cement versus press-fit); and (4) method of bone-defect repair (cement, augments, bone graft). In principle, an implant with the least constraint required for satisfactory knee stability is selected to reduce stress on the implant-fixation interface with compromised bone. The severity of bone loss largely influences stem length. Canal-filling stems and cementless stem fixation are indicated when major structural allografts are used. Bone defects can be successfully and reliably repaired with metal augments, allograft bone, or cement, as long as long-stemmed components without excessive constraint are selected. Because structural allografts do not revascularize, the major advantage of the allograft, as compared to cement fill or augments, is the ability to unite to damaged host bone that has a poor cancellous structure.

4- to 10-year results with the anatomic modular total knee.

The outcome of 186 consecutive total knee arthroplasties performed with the Anatomic Modular Knee is reported. One hundred forty-two knees had followup of 4 to 10 years (mean, 6.9 years). Two knees required revision of all components, one because of infection and the other for instability. Seven patients underwent exchange of the modular tibial polyethylene insert at an average of 84 months postoperatively. The mean age of these seven patients at primary arthroplasty was 54 years. No implant showed clinical or radiographic evidence of loosening and there were no complications related to patellofemoral articulation. The revision rate of the femoral, tibial, and patellar components was two of 142 or 1.4%. The revision rate including exchange of the polyethylene insert was nine of 142 or 6.3%. This minimally constrained posterior cruciate retaining modular design performed well at intermediate followup. The absence of patellofemoral complications and aseptic loosening was notable. Wear related phenomena were the most common indications for reoperation and these occurred in younger, active individuals with relatively thin polyethylene bearings. The relative value of modular tibial inserts and the outcome of isolated polyethylene exchanges warrant additional study.

Classification and preoperative radiographic evaluation: knee.

Preoperative radiographic planning for revision total knee arthroplasty begins with obtaining excellent quality AP and lateral radiographs that permit: 1. Evaluation of the extent of bone loss in the metaphyseal region of the femur and tibia. 2. Full visualization of the patient's intramedullary canal for determining appropriate stem size and length. The surgeon should then determine the appropriate bone defect classification, keeping in mind the provisions needed to address Type 2 and 3 defects. These provisions include any augments or allografts, stemmed components, and the degree of component constraint needed in the patient's revision surgery. Through preoperative templating, the surgeon can determine whether a particular implant system provides the options necessary to achieve an optimal surgical result. Whenever templating leaves unanswered questions regarding the extent of bone damage or the degree of knee instability, the surgeon must prepare for the worst case scenario to ensure that the appropriate components and graft material are made available.

Influence of surgical approach on lateral retinacular releases in total knee arthroplasty.

The rate of lateral retinacular releases in 2 groups of patients who underwent knee surgery performed by the senior author is compared and the factors that contribute to these differing rates are discussed. The first group of subjects consists of 88 patients who had total knee arthroplasties performed with a standard medial parapatellar approach between August 1987 and August 1988. The second group is comprised of 88 patients who had total knee arthroplasties through a midvastus surgical arthrotomy that splits the fibers of the vastus medialis muscle. This method is used by the senior author for all primary total knee arthroplasties. Lateral retinacular releases were performed in 50% of the cases in the medial parapatellar group versus only 3% of the cases in the midvastus group. Patellofemoral instability occurs as a consequence of incising the quadriceps tendon in the medial parapatellar approach and results in the need to perform lateral retinacular releases. The reduction in lateral retinacular releases is attributed to the fact that the midvastus approach leaves the connection of the vastus medialis to the quadriceps tendon intact.

Tibial osteolysis in cementless total knee arthroplasty. A review of 25 cases treated with and without tibial component revision.

Revision total knee arthroplasty was performed in 25 knees with tibial osteolysis that was identified radiographically and confirmed histologically. Eighteen knees had focal osteolysis in the medial tibial plateau, 2 in the lateral plateau, and 5 in both plateaus. Three surgical treatment options were used: (1) exchange of the modular tibial polyethylene insert with screw removal; (2) revision of the tibial component with curettage and optional grafting of the defect; (3) and full revision of all components. The clinical and radiographic results of partial and full revision remain excellent 2 to 6 years (mean, 41 months) postrevision arthroplasty for 23 knees. Clinical failure with pending revision occurred in 1 patient with polyethylene exchange without the removal of tibial fixation screws, and in 1 patient with progression of osteolysis adjacent to the femoral component 2 years after tibial component revision. Tibial defects have not progressed, and new lesions have not developed with simple curettage of the defect and removal of sources of particulate. Tibial baseplates remain radiographically stable. Component revision of the tibial component, femoral component, or both, required the insertion of long-stemmed revision components and frequently the use of major structural allografts. These revisions have faired well, although the long-term stability of components supported with major structural allografts is unknown.