Clinical Results of Fixed-Bearing and Rotating-Platform Total Knee Prostheses

• Updated on: 2018-03-13
• Read original article here

The ultimate goals of total knee arthroplasty (TKA) are pain relief and restoration of knee motion and clinical performance. During the past 50 years, the designs of the prostheses used have been refined in an effort to improve clinical results after TKA. More than 20 years ago, the Optetrak cemented posterior-stabilized fixed-bearing and rotating-platform knee prostheses (Exactech, Gainesville, Florida) were introduced. They are a modification of the Insall-Burstein posterior-stabilized knee prosthesis (Zimmer, Warsaw, Indiana). This modification sought to reduce the localized stresses on the tibial polyethylene insert, improve patellar tracking, increase resistance to tibial subluxation, and provide flexion greater than 120° while remaining stable in the frontal and sagittal planes. Optetrak rotating-platform total knee prostheses were designed to provide dual-surface articulation at both upper and lower surfaces of the polyethylene insert.

Inconsistent results have been reported for the Optetrak stabilized fixed-bearing and rotating-platform total knee prostheses. Robinson reported that predicted fixed-bearing implant survival at 93 months was 97%. Furthermore, Robinson and Green reported that the survival rate of the fixed-bearing prosthesis was 98% at a mean follow-up of 11.6 years. On the other hand, Thelu et al reported poor results of these fixed and mobile prostheses after a mean follow-up of 25 months. They suggested cumulated survival rates of 81% at 36 months and 77% at 45 months. They believed that poor results of this prosthesis were attributable to early tibial component loosening and abnormal painful patellar contact on a highly restricted trochlea.

In this study, the authors sought to determine whether the design changes of the Optetrak posterior-stabilized fixed-bearing and rotating-platform knee prostheses had achieved their intended objectives at minimum 10-year follow-up. Furthermore, they intended to delineate the specific features of these knee prostheses to which good outcomes can be attributed.

The authors prospectively followed and retrospectively reviewed 111 patients (222 knees) who underwent bilateral simultaneous primary TKAs from January 2005 to February 2007 using an Optetrak posterior-stabilized fixed-bearing prosthesis in one knee and an Optetrak rotating-platform prosthesis in the other knee. No other implant was used during this study period. The study protocol received institutional review board approval. Each patient provided written informed consent. Unilateral TKAs were performed in only 8 patients during the study period. Eleven patients (22 knees) were lost to follow-up. The remaining 100 consecutive patients (200 knees) formed the study cohort. There were 88 women and 12 men. They had a mean age of 66.2±7.5 years (range, 35–81 years) at the time of the index surgery. The high prevalence of end-stage knee osteoarthritis in the female patients of this study might be attributable to the inherent varus deformity of the knee. The mean height of the patients was 156.7±8.6 cm (range, 145–188 cm). Their mean weight was 63.4±8.9 kg (range, 48–84 kg), and their mean body mass index was 27.1±2.9 kg/m (range, 20.2–41.1 kg/m). Seventy-one patients (71%) had a body mass index between 20 and 29 kg/m, 27 (27%) had a body mass index between 30 and 39 kg/m, and 2 (2%) had a body mass index of more than 40 kg/m. All of the patients had the preoperative diagnosis of osteoarthritis of the knee (Table 1). The mean preoperative overall anatomical (femorotibial) knee alignment was 9°±6.1° varus (range, 8°–20°) in the fixed-bearing group and 9.8°±6.1° varus (range, 5°–20°) in the rotating-platform group on the basis of long-leg radiographs including the femoral head and ankle. All of the TKAs were performed by the senior author (Y.-H.K.). The procedure was carried out through a midline skin incision of 10 to 12 cm in length using a medial parapatellar arthrotomy. Extramedullary instrumentation was used for the tibial component, and intramedullary instrumentation was used for the femoral side. The femoral valgus angle for the intramedullary guide, which was between 5° and 7°, was determined preoperatively on standardized long-leg weight-bearing radiographs. The anterior and posterior cruciate ligaments were resected in all knees. Ligamentous balance was restored and 10 mm of tibial bone was resected from the least affected side to achieve a surface that was perpendicular to the shaft of the tibia in the coronal plane with a posterior slope of 3° in the sagittal plane. Resection of the distal femur (9 mm) and the posterior femoral condyles (9 mm) was attempted to remove a thickness of bone that was equal to that of the femoral component to be implanted. During femoral and tibial resection, the tibia was prepared first in all knees. Anterior cortical reference was used for the anterior–posterior cut of the distal femur. Femoral component rotation was determined using 3 reference axes: (1) the transepicondylar axis, (2) the mid-trochlear line (Whiteside line), and (3) 3° of external rotation relative to the posterior aspect of the condyles. Tibial component rotation was determined using the medial one-third of the tibial tuberosity and the attachment site of the posterior cruciate ligament. Ligamentous balance was established first in knee extension and then in knee flexion with the use of a tensor. Metal-backed tibial components were used in all knees. Randomization to a fixed-bearing or a mobile-bearing TKA was accomplished using a sealed study number envelope, which was opened in the operating room before the skin incision was made. A computer program was used to assign all patients to receive a fixed-bearing TKA in one knee and a mobile-bearing TKA in the contralateral knee. The Optetrak fixed-bearing and rotating-platform prostheses include a cemented femoral implant in cobalt-chromium alloy that is anatomic and asymmetrical. The trochlear groove is deep, oblique, and outward (7°) to facilitate patellar tracking. The surface of the tibial component is a microbead porosity. The keel is small, short, and round and has wings. The tibial post was moved anteriorly to reduce the impingement between cam and post. The patella is a symmetrical and spherical dome with a slightly concave slope matching the shape of the prosthetic trochlea and has a choice of 4 patellar implant sizes. All femoral, tibial, and patellar implants were cemented (Figure 1). Optetrak posterior-stabilized fixed-bearing and rotating-platform prostheses (Exactech, Gainesville, Florida). Anterior view of the fixed-bearing prosthesis. The femoral component in cobalt-chromium alloy is anatomic and asymmetrical. The tibial tray with fixed-bearing polyethylene is made of titanium alloy. The trochlear groove is deep and oblique upward and outward (7°). The femoral condyles present a symmetrical curve radius. The tibial keel is small, short, and round and has wings (A). Anterior view of the rotating-platform total knee prosthesis. Femoral and tibial components are in cobalt-chromium alloy. The polyethylene component can turn around the axis of the proximal surface of the tibial tray, which is collinear with the plug axis. The tibial surface of this rotating-platform tray is adapted to the geometry of the polyethylene, having a wave-shaped design so that the tibial polyethylene thickness remains constant with the joint surface (B). The knee was placed in a continuous passive motion after the splint was removed on the second day postoperatively. All patients began walking with crutches or a walker and started active and passive range of knee motion exercise on the second day postoperatively. Patients used the crutches or walker, with full weight bearing, for 6 weeks and a cane when necessary thereafter. Two of the authors (J.-W.P., J.-S.K.) assessed the patients via both physical examination and knee score preoperatively and at 3 months, 1 year, and 2 or 3 years postoperatively. The Knee Society clinical rating system, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) questionnaire, and the University of California at Los Angeles (UCLA) activity scale were used. Mean duration of follow-up was 10.4 years (range, 10–11 years). At each follow-up visit, radiographic data were analyzed and recorded by another author (Y.-H.K.), who was not part of the operating team. To ascertain instability of the knee, patients were asked whether they had a “giving way” sensation. The stability of the knee was examined in full extension, midrange (60°) flexion, and full flexion to ascertain any instability of the tibial bearing. Patients were specifically asked about patellar pain. With the patient in the supine position, the active arc of motion of the knee was measured twice using a standard (60-cm) goniometer preoperatively and at each follow-up visit by 2 observers (J.-W.P., J.-S.K.). The chance-corrected kappa coefficient for interobserver agreement ranged from 0.81 to 0.91. All clinical data were compiled and collected by a research associate. Standing anteroposterior hip-to-ankle radiographs, supine anteroposterior and lateral radiographs, and skyline patellar radiographs of the knee were obtained preoperatively and at each follow-up visit. One observer (J.-S.K.), who was not a member of the operating team, evaluated the radiographs to determine the anatomic axis of the limb, alignment of the components, posterior tibial slope, posterior femoral condylar offset, level of the joint line, presence and location of radiolucent lines at the bone–cement or cement–implant interface, and patellar tilt or dislocation using the Knee Society clinical rating system. All radiographs were obtained under fluoroscopic guidance to control rotation of the knee. Changes in Knee Society knee scores, WOMAC scores, and UCLA activity scores were evaluated using a 2-tailed Student's t test. Chi-square tests were used to analyze radiological data and complication rates. All statistical analyses were performed with SPSS version 14.0 software (SPSS Inc, Chicago, Illinois), with P<.05 being significant.

In both groups, the preoperative Knee Society knee scores, WOMAC scores, and UCLA activity scores had improved significantly at the final follow-up (Table 2). Mean preoperative Knee Society knee score was 26 points (SD, 9 points; range, 16–35 points) in the fixed-bearing group and 27 points (SD, 9 points; range, 15–36 points) in the rotating-platform group. Mean postoperative Knee Society knee score was 90 points (SD, 12 points; range, 70–100 points) in the fixed-bearing group and 92 points (SD, 11 points; range, 75–100 points) in the rotating-platform group. Mean preoperative total WOMAC score was 58 points (SD, 17 points; range, 36–94 points) and mean postoperative total WOMAC score was 16 points (SD, 8 points; range, 5–46 points) in the fixed-bearing group. Mean preoperative total WOMAC score was 58 points (SD, 17 points; range, 36–96 points) and mean postoperative total WOMAC score was 16 points (SD, 8 points; range, 6–49 points) in the rotating-platform group. Mean pre-operative UCLA activity score was 2.2 points (range, 1–4 points) and mean postoperative UCLA activity score was 6.5 points (range, 4–10 points) in the 2 groups. Mean preoperative range of knee motion was 125°±9° (range, 95°–140°) and mean postoperative range of knee motion was 123°±9° (range, 75°–140°) in the fixed-bearing group. Mean preoperative range of knee motion was 128°±9° (range, 85°–140°) and mean postoperative range of knee motion was 125°±8° (range, 75°–140°) in the rotating-platform group. At the final follow-up, 81 patients (81%) had no pain, 12 patients (12%) had mild pain, and 7 patients (7%) had moderate pain in the fixed-bearing group. In the rotating-platform group, 79 patients (79%) had no pain, 15 (15%) had mild pain, and 6 (6%) had moderate pain. No patient had severe pain in either group. No patient in either group reported patellar pain. The alignment of the knee improved substantially. Mean preoperative knee alignment was 9.3°±5° (range, 5°–20°) varus, which improved to a mean of 5.5° (range, 4°–8°) valgus in the fixed-bearing group. Mean preoperative knee alignment was 9.8°±5.8° (range, 5°–20°) varus, which improved to a mean of 5.5° (range, 4°–7°) valgus in the rotating-platform group. The positions of the femoral and tibial components were satisfactory in all knees. Preoperative joint line and posterior femoral condylar offset were well restored postoperatively. The prevalence of a radiolucent line (<1 mm) was 14% (14 knees) around the anterior femoral condyle and 4% (4 knees) around the posterior femoral condyle in the fixed-bearing group. The prevalence of a radiolucent line (<1 mm) around the tibial component was 6% (6 knees) in the fixed-bearing group. The prevalence of a radiolucent line (<1 mm) was 8% (8 knees) around the anterior femoral condyles and 6% (6 knees) around the posterior femoral condyles in the rotating-platform group. The prevalence of a radiolucent line was 8% (8 knees) around the tibial component in the rotating-platform group. No knee in either group had a radiolucent line wider than 1 mm. No knee in either group had aseptic loosening of the component or osteolysis (Table 3). The predicted implant survival rate at 10.4 years was 99% as the endpoint of any reoperation (Figure 2). Radiographic Results for the 100 Patients at Final Follow-up Radiographic evaluation of the knees of a 61-year-old woman with an end stage of osteoarthritis following total knee arthroplasty. Anteroposterior radiograph of both knees obtained 11 years postoperatively showing that the fixed-bearing prosthesis (right knee, shown in the left image) and the rotating-platform prosthesis (left knee, shown in the right image) are fixed in a satisfactory position. There is no evidence of a radiolucent line or osteolysis (A). Lateral radiograph of both knees obtained 11 years postoperatively showing that the fixed-bearing prosthesis (right knee, shown in the left image) and the rotating-platform prosthesis (left knee, shown in the right image) are embedded in a satisfactory position. There is no evidence of a radiolucent line or osteolysis (B). One knee (1%) in each group had early deep infection. These 2 knees were treated with open debridement, tibial polyethylene liner change, and intravenous antibiotics for 6 weeks. There was no further recurrence of infection.

The Optetrak cemented posterior-stabilized fixed-bearing and rotating-platform prostheses were a modification of the Insall-Burstein posterior-stabilized prosthesis to improve patellar tracking, provide flexion greater than 120°, and increase resistance to tibial subluxation. Little published evidence exists regarding whether these prostheses further improve the clinical and radiographic results of TKA. In addition, the available studies of these fixed-bearing and rotating-platform TKAs have had short-term follow-up and have used different tibial keel models. Therefore, the clinical benefits after long-term follow-up are not clear. The current authors made assessments regarding clinical results, radiographic results, revision rate, and complication rate at 4 time points. Clinical function and survivorship of the knee components were excellent in both groups. Furthermore, the complication rate was very low in both groups. Conflicting results have been reported for Optetrak TKAs. One study reported poor results after a mean follow-up of 25 months. Other studies have reported excellent results at 5- and 11-year follow-up. At 25-month follow-up, Thelu et al reported a mean Knee Society knee score of 83.7 points (range, 13–100 points) and a mean Knee society knee function score of 82.6 points (range, 30–100 points). On the contrary, Robinson reported that the average Knee Society knee score was 92 points (range, 59–100 points) at 5-year follow-up. Thelu et al reported a high incidence of a radiolucent line around the tibial component (38%). They insisted that the fixation mode (keel geometry) and the high level of stress (high congruency) played a role in the occurrence of a high incidence of a radiolucent line. They found that 25 tibial components (22%) had aseptic loosening. By contrast, Robinson and Robinson and Green did not observe any aseptic loosening of the tibial components at 5- and 11-year follow-up, respectively. The current authors found no aseptic loosening of the tibial components in either group at 10.4-year follow-up. Thelu et al performed 13 revisions for either resistant patellofemoral pain (4 knees) or early aseptic loosening of the tibial components (9 knees). They further observed that, at the end of their analyses, 10 more patients remained with pain and presented worrisome radiological and clinical signs indicative of tibial component loosening or patellofemoral impingement. Therefore, with the endpoint of tibial component loosening, the cumulative mean Kaplan–Meier survival rate was 89%±4% at 25 months, 81%±9.1% at 36 months, and only 77%±12% at 45 months postoperatively. Robinson reported that the only revision was the patellectomy due to avascular necrosis and fragmentation at 5-year follow-up. Therefore, using revision of any component as an endpoint, 99% of the implants were predicted to survive at 93 months. Robinson and Green reported that the survival rate with revision for aseptic loosening as an endpoint was 100% at 11-year follow-up. In the current study, the survival rate of the implant was 99% at 10.4-year follow-up. The current authors believe that the good results of these TKAs in both groups are attributable to the small stature and low weight of the patients; preponderance of female patients; good cementing technique with pulsed lavage and cement; pressurization; correct flexion and extension gaps; and well-balanced ligaments. To avoid bearing dislocation or instability of the knee, a good surgical technique, especially balancing of flexion and extension gaps, is mandatory in a rotating-platform TKA. Correct flexion and extension gaps led to no bearing dislocation or instability in the current series. Several factors may be related to the absence of osteolysis in both groups in the current study: a high percentage (88%) of the patients were female and had a relatively low body weight (mean, 62 kg); the polyethylene inserts used were sterilized with gamma irradiation in a vacuum; the inserts had a short shelf-life; and follow-up was not sufficiently long to reveal osteolysis. Thelu et al reported postoperative complications including thromboembolic disease (4 phlebitis and 1 pulmonary embolism) and 4 early revisions (1 for hematoma and 3 for deep infection). Robinson observed that 2 knees had mild crepitation, 1 had patellar avascular necrosis with fragmentation, and 1 had infection. In the current study, each group had 1 early deep infection, which was treated with open debridement, tibial polyethylene liner change, and intravenous antibiotics for 6 weeks. There was no further recurrence of infection. Limitations of this study included the relatively small sample and the lack of interobserver comparisons, which could lead to bias in interpreting the radiological results. Further limitations included the low body weight of the patients, good preoperative range of knee motion, and the relatively young age of these patients, which might limit the general applicability of these results to other groups of patients. Although these patients had low body weight, their daily activities, including farming, squatting, and lifting, were vigorous. Finally, it is difficult for patients who have undergone bilateral TKA to distinguish the independent function of each knee. Pain, support, and motion were easily differentiated, but distance walked and stair-climbing ability were more difficult to distinguish. If the patients had difficulties in these areas, they could always identify the knee that limited their activities more.