Polyethylene wear after total knee arthroplasty may predispose an implant to loosening and subsequent failure. In many cases, the revision operation is made more difficult by the varying degree of periprosthetic bone loss that occurs as a direct result of this wear. Unfortunately, the etiology of polyethylene wear in total knee replacements is multifactorial, and may be in part, attributable to prosthetic design, articular conformity, knee alignment, the technique of operation, component fixation, third body wear, the manufacturing technique of the polyethylene, and polyethylene thickness. 10 Generally, knee replacements with at least 8 mm to 10 mm of tibial polyethylene are preferred because the increased contact stresses associated with thinner polyethylene may predispose to implant failure. 14 Based on finite-element analysis, Bartel et al 1 suggested that polyethylene thickness should be maximized and “. . . A minimum thickness of the layer of polyethylene of eight millimeters should be maintained whenever possible.”
Is 8 mm of tibial polyethylene necessary in total knee arthroplasty? The authors hypothesized that, despite using less than the preferred 8 mm of tibial polyethylene, excellent mid-term results can be obtained using compression molded, flat-on-flat, nonconstrained tibial components.
MATERIALS AND METHODS
Between January 1983 and December 1989, a consecutive series of 1247 primary AGC (Anatomic Graduated Components, Biomet, Warsaw, IN) posterior cruciate-retaining total knee arthroplasties was performed at the authors’ institution. From this study base, 387 AGC total knee prostheses were implanted with 8-mm thick tibial components (combined metal and polyethylene). All tibial prostheses were manufactured with 4.4 mm of polyethylene directly compression molded to a 3.6-mm CoCr metal baseplate and central stem. Although the coronal plane articulation between the femoral and tibial components was conforming anteriorly, the sagittal plane articulation essentially was flat-on-flat and nonconstrained (Fig 1).
The diagnosis primarily was osteoarthritis (84.7%). The average age of the patients at surgery was 70.6 years. There were 204 women (53%) and 109 men (47%). The remaining demographic data are listed in Table 1.
During this same period, 232 patients underwent simultaneous bilateral AGC total knee replacements. One hundred sixteen of these patients received an 8-mm tibial prosthesis on one side (4.4 mm of polyethylene) and at least a 10-mm thick tibial component on the other side. Ten-millimeter tibial components (6.4 mm of polyethylene) were used in 87 knees. Twelve-millimeter thick tibial components (8.4 mm of polyethylene) were used in 26 knees. Three knees received a 16-mm tibial component (12.4 mm of polyethylene). The average age of these 116 patients at surgery was 70.1 years (range, 31–89 years). Fifty-seven percent (66 patients) were women.
All knees were approached through a standard medial parapatellar incision. All tibial and patellar components and all but five femoral components were cemented. Jet lavage and suction drying was used in every case. Between January 1983 and July 1987, metal-backed patellas were implanted. After July 1987, only all-polyethylene patellas were used. Ambulation was started on the first postoperative day and range of motion (ROM) was started on postoperative Day 2. Patients were allowed immediate full weightbearing as tolerated.
Patients were evaluated preoperatively and postoperatively at 8 weeks, 6 months, 1 year, and then every 2 to 3 years according to the Knee Society clinical 5 and radiographic 4 scoring system. These data were collected prospectively starting at the latest preoperative visit and through each clinical followup including the most recent examination. Furthermore, radiographs were scrutinized retrospectively for polyethylene thickness changes (of at least 2 mm) and for osteolytic lesions. As suggested by Colizza et al, 2 changes in the thickness of the polyethylene on the anteroposterior (AP) radiograph obtained with the patient standing were considered evidence of polyethylene wear. The current study represents a retrospective review of these data.
Followup of the 313 patients averaged 10.7 years (range, 2–17 years). No patient was lost during this followup period. However, 54 patients (17%) died during the followup period, all from causes unrelated to the arthroplasty. Followup of the patients who had bilateral knee replacements averaged 10.3 years (range, 2–16 years). Statistical analysis was performed including Student’s t test, Log-rank test, and Kaplan-Meier survival rates at 5, 10, and 15 years with the aid of the Statistical Analysis System (Cary, NC). 6 Failure was defined as revision of any component for any reason or loosening of any component.
Table 2 lists the clinical results at the most recent followup. Knee Society knee scores average 81.4 points with pain scores averaging 47.2 points (50 points equates to no pain). Thirty-five knees had a residual flexion contracture of no more than 5°. Flexion averaged 115°.
No polyethylene thickness change consistent with measurable tibial component wear was identified in any knee. Furthermore, no focal osteolytic lesion was seen radiographically at final followup. Postoperative alignment averaged 6° valgus (range, 0–10° valgus). In no case was a change of AP alignment greater then 2° observed. This change was thought to be within the error of measurement. Excluding infection, tibial radiolucencies were identified in four knees (1%). All of these radiolucent lines were found at the bonecement interface and were measured between 1 and 2 mm. Zone 1 tibial radiolucencies were found in all four of these knees. Two of these four knees also had Zone 2 radiolucencies. Another one of these four knees had radiolucent lines in Zones 6 and 7. Interestingly, all four of these patients were pain-free at final followup. No clinical or radiographic tibial or femoral component aseptic loosening was identified.
At the time of the most recent followup, three prostheses were removed because of late (greater than 6 months) infection (0.7%). Five knees required revision because of metallosis and polyethylene failure of metal-backed patellas (1.3%). One other loose metal-backed patella was identified (0.3%). The overall prosthetic failure rate of the entire study group was 2.3% (nine cases). No knee was revised for tibial or femoral component loosening. Table 3 shows the 5-, 10-, and 15-year Kaplan– Meier survival rates of 98.7%, 95.4%, and 94.3%, respectively.
Table 4 compares the Knee Society knee and pain scores at final followup in those patients with an 8-mm thick tibial prosthesis on one side and a 10-mm thick or greater tibial component on the other side. Knee and pain scores in both groups were remarkably similar to the group of 387 knee arthroplasties. More importantly, no statistical differences were seen in these scores between the two groups.
One late infection (0.8%) occurred in each group. Eight metal-backed patellas failed in these 116 patients. Five patellas (4.3%) failed in patients with 8-mm components and three patellas (2.6%) failed in patients with 10-mm to 16-mm components. All but one patient required revision surgery. The overall failure rate of these 232 knee arthroplasties was 4.3% (10 cases). No tibial or femoral component was revised for reason other than infection or patellar failure in either group. Furthermore, no tibial or femoral component aseptic loosening occurred. Thus, 94.9% of the 8-mm tibial components and 96.6% of the 10-mm to 16-mm tibial components have continued to function well at an average followup of 10.3 years. This difference was not statistically significant (p = 0.4472, log-rank test).
Excellent midterm results were obtained in the current series of knee arthroplasties using tibial components with only 4.4 mm of polyethylene. The success of this implant may be related, at least in part, to the manufacturing technique of the polyethylene and the tibial component design. Both of these variables are important factors that may contribute to polyethylene wear in total knee arthroplasty. The material strength of ultrahigh molecular weight polyethylene is directly related to the randomness of its crystalline lamellae. Oriented polyethylene lamellae decreases its material strength whereas random polyethylene lamellae increases its material strength. This increase in randomness tends to resists loads. 7 The manufacturing process of machining polyethylene is known to increase polyethylene crystalline lamellar alignment and, therefore, decrease its strength and wear characteristics. 11 By shearing the polyethylene, this polymer structural change results in increased surface roughness and lamellar orientation at the subsurface level. 11 Similar changes in the ultrastructural characteristics of the polyethylene occur during reciprocating sliding 7 and wear simulation. 3 Thus, machining creates changes in the polyethylene that may serve as a precursor to abrasive wear and crack propagation 3,7,11 before the prosthesis even is implanted. The factory process of compression molding avoids surface machining, therefore, eliminating this initial surface damage.
Another factor that may contribute to the longevity of this prosthesis is the lack of modularity. Micromotion between the tibial insert and the metal baseplate has been shown to be an important additional source of polyethylene debris contributing to osteolysis after total knee replacement. 13 Urban et al 12 evaluated polyethylene wear particles disseminated to the liver, spleen, and abdominal lymph nodes of patients with hip and knee replacements and found that the principle source of these particles came from secondary, nonbearing surfaces, rather than the two primary bearing surfaces. Furthermore, Parks et al 8 evaluated nine contemporary modular tibial insert designs and found that none of the current designs eliminated micromotion and that even under relatively low loads, sufficient motion occurred in all implants to create fretting at the modular tibial interface. The successful results presented in the current review may be a reflection of the absence of tibial insert micromotion and the elimination of the deleterious effects of backside wear.
The difficulties in measuring even minor changes in the polyethylene thickness on standard radiographs are well recognized. 2 Although no such changes were identified in the current study, and although no gross tibial component wear was identified in those knees requiring revision, the probability that polyethylene wear did occur during a 17-year period definitely exists. Clinically, however, such wear was not apparent.
A multicenter report has evaluated previously 2001 AGC total knee replacements in 1351 patients. Excluding infection, 10-year prosthetic survival was 98% and tibial component survival was 99%. 9 The results of the current report parallel the results of this earlier study.
With thinner polyethylene (4.4 mm) and lower conformity total knee replacement no tibial component radiographic wear, loosening, or revision for reason other than infection occurred at an average followup of greater than 10 years. Excellent midterm results were obtained and 15-year Kaplan–Meier survivorship was greater than 94%. Furthermore, similar clinical results and survival rates were observed in knees implanted with thicker polyethylene (6.4 mm or greater). Direct compression molding of the polyethylene seems to be advantageous in minimizing the higher rates of failure associated with using thinner polyethylene in lower conforming knee arthroplasties. The problems associated with tibial insert micromotion and backside wear are eliminated. The significance of polyethylene design and manufacturing and tibial component modularity in total knee arthroplasty cannot be overemphasized.
1. Bartel DL, Bicknell VL, Wright TM: The effect of conformity, thickness, and material on stresses in ultra-high molecular weight components for total joint replacement. J Bone Joint Surg 68A: 1041–1051, 1986.
2. Colizza WA, Insall JN, Scuderi GR: The posterior stabilized total knee prosthesis. J Bone Joint Surg 77A: 1713–1720, 1995.
3. Edidin AA, Pruitt L, Jewett CW, et al: Plasticity-induced damage layer is a precursor to wear in radiation: Cross-linked UHMWPE acetabular components for total hip replacement. J Arthroplasty 14: 616–627, 1999.
4. Ewald FC: The Knee society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop 248: 9–12, 1989.
5. Insall JN, Dorr LD, Scott RD, et al: Rationale of the Knee Society clinical rating system. Clin Orthop 248: 13–14, 1989.
6. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457–481, 1958.
7. Klapperich C, Komvopoulos K, Pruitt L: Tribological properties and microstructure evolution of ultra-high molecular weight polyethylene. Trans Am Soc Mech Eng 121: 394–402, 1999.
8. Parks NL, Engh GA, Topolski LDT, et al: Modular tibial insert micromotion: A concern with contemporary knee implants. Clin Orthop 356: 10–15, 1998.
9. Ritter MA, Worland R, Saluki J, et al: Flat-on-flat, nonconstrained, compression molded polyethylene total knee replacement. Clin Orthop 321: 79–85, 1995.
10. Schmalzried TP, Callaghan JJ: Wear in total hip and knee replacements. J Bone Joint Surg 81A: 115–136, 1999.
11. Song J, Liu P, Cremens M, et al: Effects of machining on tribological behavior of ultra high molecular weight polyethylene (UHMWPE) under dry reciprocation sliding. Wear 225–229: 716–723, 1999.
12. Urban RM, Jacobs JJ, Tomlinson MJ, et al: Dissemination of wear particles to the liver, spleen, and abdominal lymph nodes of patients with hip or knee replacements. J Bone Joint Surg 82A: 457–477, 2000.
13. Wasielewski RC, Parks N, Williams I, et al: Tibial insert undersurface as a contributing source of polyethylene wear debris. Clin Orthop 345: 53–59, 1997.
14. Wright TM, Bartel DL: The problem of surface damage in polyethylene total knee components. Clin Orthop 205: 67–74, 1986.
Merrill A. Ritter, MD; and John B. Meding, MD, Guest Editors