Total knee arthroplasty with the original cemented posterior cruciate-sacrificing total condylar prosthesis has shown excellent results with good long-term durability. 7,12,13,16 With revision as the end point for series of 100 to 224 knees, survival rates of 91% to 92% at 10 to 15 years have been predicted. 12,13,16 Some subsequent condylar knee implant designs have provided for preservation or substitution of the posterior cruciate ligament and the addition of a metal-backed tibial component. Evaluation of the results of implants that incorporate these design changes are needed to determine how the changes affect clinical results and long-term durability.
In 1995, the 10-year results for a group of 168 total knee arthroplasties performed with insertion of the Kinematic-I condylar prosthesis (Howmedica, Rutherford, NJ) with cement were reported from the authors’ institution. 11 This prosthesis is designed for preservation of the posterior cruciate ligament, has a metal-backed tibial component, and a symmetric all-polyethylene patellar component (Fig 1). At the time of that study, six revisions had been performed. The survival rate of the prostheses was estimated to be 96% at 10 years with revision as the end point. The purpose of the current study is to provide additional long-term followup on this same group of patients.
MATERIALS AND METHODS
One hundred sixty-eight primary total knee arthroplasties in 118 patients were performed with insertion of the Kinematic-I condylar prosthesis with cement. Eleven surgeons performed the surgeries, but the majority (73.2%) were performed by one surgeon. There were 72 women (61%) and 46 men (39%) whose mean age was 65.1 ± 11.4 years (range, 21–88 years). The mean height of the patients was 166 ± 10.5 cm and the mean weight was 71.0 ± 9.8 kg. The prosthesis was inserted into 82 right knees (48.8%) and 86 left knees (51.2%). Fifty patients (42.4%) had bilateral total knee arthroplasties with the Kinematic-I condylar prosthesis either simultaneously or in a staged fashion. The tibial polyethylene insert was 6 or 8 mm in 146 knees (86.9%) and greater than 8 mm in 22 knees (13.1%). The underlying diagnosis that led to arthroplasty was osteoarthritis in 109 knees (64.9%), rheumatoid arthritis in 52 knees (31%), posttraumatic arthritis in five knees (3%), and avascular necrosis in two knees (1.2%). Thirty-two knees (16%) had been operated on previously: 16 upper tibial valgus osteotomies, three medial meniscectomies, three arthroscopic debridements, three cheilectomies and debridements, two patellectomies, one anterior cruciate ligament reconstruction, and various other procedures in five.
All surgeries were performed through a medial parapatellar approach to expose the knee. Varus and valgus deformities were corrected with soft tissue releases and bone resection. The position of the femoral and tibial components was determined with the use of an extramedullary system. Preparation of the patellas was done without the use of a cutting jig. Patellas routinely were resurfaced except in the two patients who had prior patellectomies. The femoral, tibial, and patellar components all were inserted with cement. A lateral retinacular release was performed in 64 of 168 knees (38.8%). Postoperatively, the knees were immobilized for 48 hours and physical therapy was started on the third day after surgery. Manipulation was performed with the patient under general anesthesia in 19 knees (11.3%). These patients had flexion less than 90° and motion had not improved during three consecutive days of postoperative physical therapy.
At the time of the current study, 61 patients (86 knees) died, one had an above-knee amputation, and three patients (five knees) were lost to followup. There have been 13 revision operations performed to date. Therefore, at the time of this review, 66 knees in 50 patients were available for review. The latest clinical followup was performed at a mean of 15.7 ± 1.1 years (range, 14–20 years) after arthroplasty. For the latest clinical followup, 16 patients (32%) were evaluated at the authors’ institution by questionnaire and examination, one patient (2%) was evaluated by a local physician, 17 patients (34%) were evaluated by detailed standardized telephone questionnaire, and 16 patients (32%) were evaluated by detailed standardized letter questionnaire. Patients who were unable or unwilling to return for examination answered the detailed telephone or letter questionnaire.
Of the 50 patients (66 knees) available for followup, there were 16 men (32%) and 34 women (68%) whose mean age was 60.8 ± 12.1 years (range, 21–81 years) at the time of surgery. The mean height and weight of these patients was 165.6 ± 10.9 cm and 70.8 ± 13.4 kg, respectively. Overall, 35 right knees and 31 left knees had a Kinematic-I condylar prosthesis inserted. Twenty-five knees (38.5%) had a lateral retinacular release. The mean preoperative range of motion (ROM) was from −9.36° ± 13.2° of extension to 105.9° ± 19.0° of flexion. The preoperative diagnosis was osteoarthritis in 31 patients (62%), rheumatoid arthritis in 17 patients (34%), and posttraumatic arthritis in two patients (4%). Nine patients (18%) had some type of prior knee surgery.
The clinical rating system of the Knee Society was used to calculate the knee scores. 6 Two scores are assigned to each knee. The pain score is based on pain, ROM, and stability. The functional score considers walking ability, stair climbing, and the use of any walking aids. The percentage of knees in which the patient reported moderate or severe pain, the use of any walking aids, and ambulating distances are reported preoperatively and at latest followup. Survival of the prosthesis was estimated using the Kaplan-Meier method. 8
The pain scores, as determined by the system of the Knee Society, improved from a mean of 32.8 ± 16 points (N = 61) preoperatively to a mean of 87.9 ± 14.2 points at latest clinical followup. The percentage patients reporting moderate or severe pain decreased from 94% preoperatively to 4% at latest clinical followup. Of the 50 patients, 84% reported no pain and 12% reported mild pain. The functional scores improved from a mean of 48.7 ± 16.5 points preoperatively to a mean of 51.3 ± 32.9 points (N = 66) at latest evaluation at which time the mean patient age was 76.5 years. Preoperatively, only 32% of patients were able to walk greater than three blocks and 12% could ambulate indoors only. At latest clinical followup, 62% of patients could walk greater than three blocks and 14% could ambulate indoors only. At latest clinical followup, 54% of patients did not use an aid to ambulate and 18% used only a cane compared with 66% and 20%, respectively preoperatively. The preoperative ROM was from −9.36° ± 13.2° extension to 105.9° ± 19° flexion. The ROM at 1 year after arthroplasty was from −2.0° ± 5.8° extension to 102.7° ± 13.4° flexion and at latest followup, the ROM was from −0.15° ± 1.23° extension to 106.1° ± 14.8° flexion.
Thirteen revisions were performed in the 168 primary total knee arthroplasties. Four isolated patellar component revisions were performed for aseptic patellar component loosening at 3, 5, 10, and 10 years (Fig 2). There were four revisions exclusively for an indication of tibial polyethylene wear (one tibial and femoral revision at 8 years, two tibial and patellar revisions at 12 years, and a revision of all components at 13 years) (Fig 3). Three of the four patients who underwent revision surgery for polyethylene wear had an initial tibial polyethylene thickness of 8 mm or less. There were five revisions of the femoral or tibial components for reasons other than polyethylene wear. One revision was performed at 6 years after a traumatic medial femoral condyle fracture. The femoral and patellar components were found to be loose at the time of surgery and all components were revised. One was revised for aseptic loosening of the femoral component in a patient who had a prior patellectomy; the femoral and tibial components were revised at 8 years. Two knees in two patients were revised at 12 and 13 years primarily for aseptic loosening of the tibial component, but both also had associated tibial polyethylene wear and one had associated tibiofemoral instability. One patient had only the tibia and patella revised and the other patient had all components revised. One patient had all components revised at 14 years for tibiofemoral instability. At the time of surgery, the tibial component was loose and there was some associated polyethylene wear.
Sixteen perioperative complications occurred in the entire cohort of 168 knees. A postoperative hematoma developed in seven knees (seven patients); three of the hematomas were evacuated. There were five superficial wound infections in four patients who were treated with local wound care, dressing changes, and antibiotics; two of the five knees with a superficial wound infection also had a postoperative hematoma. One of the patients with a hematoma had a superficial skin slough develop that became secondarily infected and required eventual split thickness skin grafting for wound coverage. This patient had a deep infection develop secondary to urosepsis 10 years after arthroplasty in an otherwise well functioning knee. Because of the patient’s comorbidities and prior skin graft, the deep knee infection was treated with above-knee amputation. A second patient had a deep knee infection develop 5 years after arthroplasty secondary to bacteremia from a liver abscess. Treatment consisted of debridement, retention of components, and 4 weeks of intravenous antibiotics. There was no recurrence of infection at latest followup. One patient (bilateral arthroplasties) had a sacral decubitus ulcer develop that required debridement and local rotational myocutaneous flap for coverage. Two patients (four knees) sustained nonfatal pulmonary embolisms postoperatively that were treated successfully with anticoagulation. Seven years after arthroplasty, one patient had a below-knee amputation secondary to vascular disease. This same patient sustained a supracondylar femur fracture 12 years after arthroplasty that was treated with open reduction and internal fixation but went on to nonunion. At 13 years, a repeat open reduction and internal fixation with bone graft resulted in union. Another patient sustained a supracondylar femur fracture 6 years after arthroplasty and was treated with open reduction and internal fixation, which resulted in solid bony union. There was one nondisplaced supracondylar femur fracture that healed with conservative treatment. Four traumatic patella fractures occurred in three knees (three patients). Two fractures (two knees) were nondisplaced and were treated conservatively with retention of patellar components. The third fracture was treated with open reduction and internal fixation and retention of component. This same patient sustained a second patella fracture that was treated in the same fashion 11 years later. There was one retained drain that required surgical removal. Sixteen patients (11 knees; 11.5% of knees) required manipulation under anesthesia because of limited ROM postoperatively. These patients did not respond to physical therapy.
Survivorship analysis was performed using revision as the end point and included the failure modes of aseptic loosening of any component, tibial polyethylene wear, and tibiofemoral instability. Revisions for infections or periprosthetic fractures without implant loosening were excluded. The 5-, 10-, and 15-year survivorships free of any component revision (for the above noted indications) were 98.7% (confidence interval, 97°–100°), 96.5% (confidence interval, 93.5°–99.6°), and 88.7% (confidence interval, 82°–95°), respectively (Fig 4). The 5-, 10-, and 15-year survivorships free of any revision (using the same failure modes as noted above) for the individual tibial, femoral, and patellar components are shown in Table 1.
Survivorship analysis also was performed using radiographic loosening and/or revision as the end point. Failure modes for this end point included revision for aseptic loosening, tibial polyethylene wear, tibiofemoral instability, and radiographic loosening. Revisions for infections or periprosthetic fractures without loosening were excluded. The 5-, 10-, and 15-year survivorship free of radiographic loosening and/or revision of any component (for the above noted indications) were 98.1% (confidence interval, 95.9%–100%), 95.1% (confidence interval, 91.5%–98.7%), and 85.1% (confidence interval, 78%–92%), respectively (Fig 5). Using the same end point for the individual components, the 5-, 10-, and 15-year survivorships for the tibial, femoral and patellar components are shown in Table 2.
The original total condylar prosthesis was designed for insertion with cement and sacrifice of the posterior cruciate ligament. The implant had an all-polyethylene tibial component. Survival rates of 91% to 92% at 10 to 15 years have been predicted with revision as the end point. 12,13,16 Preservation, sacrifice, or substitution of the posterior cruciate ligament and the addition of a metal-backed tibial component all are possible with implant designs developed since the original total condylar prosthesis.
Retention of the posterior cruciate ligament in total knee arthroplasty has been advocated as a way to transmit load through the ligament to the tibia, to encourage femoral component rollback to increase flexion, and to assist in maintaining the joint line. Retention of the posterior cruciate ligament results in a central contact area of the femur on the tibia that helps to distribute load evenly on the tibial component. 17 Ritter et al 15 reported on 394 knees treated with a total condylar prosthesis with sparing of the posterior cruciate ligament and predicted a 95% survival rate at 10 years with revision as the end point. Metal backing was added to the tibial component in an attempt to decrease loosening by distributing the load more evenly across the proximal tibia. 14,18 Malkani et al 11 reported on the current group of patients who had the kinematic condylar prosthesis inserted with cement, preservation of the posterior cruciate ligament, and a metal-backed tibial component. 11 With revision as the end point, the rate of survival of the prostheses was estimated to be 96% at 10 years. 11 Weir et al 19 also reported on the kinematic condylar total knee arthroplasty. They reviewed 208 consecutive total knee arthroplasties with insertion of the kinematic condylar components with cement, preservation of the posterior cruciate ligament, and a metal-backed tibial component. None of the patellas were resurfaced. The diagnosis was rheumatoid arthritis in 63.9% of patients, osteoarthritis in 34.6% and other in 1.5%. With revision as the end point, cumulative estimates of survival of 92% at 10 years and 87% at 12 years were reported. 19 Five of 22 revisions were done for fractured base plates, 12 were done for polyethylene wear, and three were done for aseptic loosening.
The posterior-substituting implant designs of total knee arthroplasty provide anteroposterior stability in extension and early flexion. As flexion proceeds, the central post produces posterior rollback of the femoral condyles on the tibial surface thus substituting for the posterior cruciate ligament. Scuderi et al 16 reported a 97% survivorship at 10 years for a posterior-stabilized prosthesis. Emmerson et al 5 reported on the Kinematic Stabilizer prosthesis (Howmedica, Rutherford, NJ), a posterior cruciate-substituting design inserted with cement. Although no patellas were resurfaced, a cumulative survival rate of 95% at 10 years and 87% at 13 years were reported with revision as the end point. 5
Various implant design changes have been developed with features that provide for sacrifice, preservation, and substitution of the posterior cruciate ligament and the addition of a metal-backed tibial component. The difference in survivorship between different designs has been shown to be relatively small with revision as the end point at 10-to 12-years followup. The current study gives longer-term followup of a group of 168 previously reported total knee arthroplasties that had the Kinematic-I condylar prostheses inserted. The underlying diagnosis was osteoarthritis in 62% of patients, rheumatoid arthritis in 34%, and other in 4%. The 15-year survivorship free of any component revision for any reason was 88.7% (confidence interval, 82%–95%).
Thirteen revisions were performed. The most common mechanisms of failure were patellar component loosening and tibial polyethylene wear. Four failures were attributable to early aseptic loosening of the patellar component. Numerous studies have identified extensor mechanism complications, including patellar component loosening as one of the most common causes of failure of condylar total knee arthroplasty. 1,2,10 The kinematic condylar total knee arthroplasty patellar component had one central peg. This design now is known to be at risk of failure caused by loosening associated with central peg fracture. Four revisions were performed exclusively for tibial polyethylene wear and three of these had a tibial polyethylene thickness of 8 mm or less. Metal-backed tibial components with polyethylene thickness less than 8 mm have been shown to be at increased risk for wear. 3,4,9 Only three knee replacements were revised specifically for aseptic loosening of the tibial or femoral components.
The current study shows good function and survivorship of a cruciate-retaining total knee arthroplasty inserted with cement, a metal-backed tibial component, and resurfacing of the patella (Fig 6). Excellent pain relief was obtained with only 4% of patients reporting moderate pain and none reporting severe pain at a mean followup of 15.7 ± 1.1 years compared with 94% preoperatively. There has been some age-related decline in function as would be expected as the length of followup increases and the cohort ages. Overall 15-year survivorship free of any component revision for any reason was 88.7%. The current study provides long-term information on the performance of an early designed posterior cruciate retaining cemented total knee arthroplasty with a metal-backed tibial component against which other more modern implants may be compared.
1. Boyd Jr AD, Ewald FC, Thomas WH, et al: Long-term complications after total knee arthroplasty with or without resurfacing of the patella. J Bone Joint Surg 75A: 674–681, 1993.
2. Brick GW, Scott RD: The patellofemoral component of total knee arthroplasty. Clin Orthop 231: 163–178, 1988.
3. Chillag KJ, Barth E: An analysis of polyethylene thickness in modular knee components. Clin Orthop 273: 261–263, 1991.
4. Collier JP, Mayor MB, McNamara JL, et al: Analysis of the failure of 122 polyethylene inserts from uncemented tibial knee components. Clin Orthop 273: 232–242, 1991.
5. Emmerson KP, Moran CG, Pinder JM: Survivorship analysis of the kinematic stabilizer total knee replacement: A 10-to 14-year follow-up. J Bone Joint Surg 78B: 441–445, 1996.
6. Insall JN, Dorr LD, Scott RD, et al: Rationale of The Knee Society clinical rating system. Clin Orthop 248: 13–14, 1989.
7. Insall JN, Kelly M: The total condylar prosthesis: A report on 220 cases. J Bone Joint Surg 61A: 173–180, 1979.
8. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457–481, 1958.
9. Kilgus DJ, Morland JR, Finerman GA, et al: Catastrophic wear of tibial polyethylene insets. Clin Orthop 273: 223–231, 1991.
10. Lynch AF, Rorabeck CH, Bourne RB: Extensor mechanism complications following total knee arthroplasty. J Arthroplasty 2: 135–140, 1987.
11. Malkani AL, Rand JA, Bryan RS, et al: Total knee arthroplasty with the kinematic condylar prosthesis: A 10-year follow-up study. J Bone Joint Surg 77A: 423–431, 1995.
12. Ranawat CSA, Boachie-Adjei O: Survivorship analysis and results of total condylar knee arthroplasty: Eight-to 11-year follow-up period. Clin Orthop 226: 6–13, 1988.
13. Ranawat CS, Hansraj KK: Effect of posterior cruciate sacrifice on durability of the cement-bone interface: A nine-year survivorship study of 100 total condylar knee arthroplasties. Orthop Clin North Am 20: 63–69, 1989.
14. Reilly D, Walker PS, Ben-Dov M, et al: Effects of tibial components on load transfer in the upper tibia. Clin Orthop 165: 273–282, 1982.
15. Ritter MA, Campbell E, Faris PM, et al: Long-term analysis of the posterior cruciate condylar total knee arthroplasty: A 10-year evaluation. J Arthroplasty 4: 293–296, 1989.
16. Scuderi GR, Insall JN, Windsor RE, et al: Survivorship of cemented knee replacements. J Bone Joint Surg 71B: 798–803, 1989.
17. Soudry M, Walker PS, Reilly DT, et al: Effects of total knee replacement design on femoral-tibial contact conditions. J Arthroplasty 1: 35–45, 1986.
18. Walker PS, Greene D, Reilly D, et al: Fixation of tibial components of knee prostheses. J Bone Joint Surg 63A: 258–267, 1981.
19. Weir DJ, Moran CG, Pinder JM: Kinematic condylar total knee arthroplasty: 14-year survivorship analysis of 208 consecutive cases. J Bone Joint Surg 78B: 907–911, 1996.
Merrill A. Ritter, MD; and John B. Meding, MD, Guest Editors© 2001 Lippincott Williams & Wilkins, Inc.