Radiolucency and Migration After Oxford Unicompartmental Knee Arthroplasty

Abstract

Radiolucent lines frequently appear on radiographs around unicompartmental knee arthroplasties. It is unknown why this occurs, although the lines usually appear during the first year following the procedure. Knee arthroplasty implants are also known to migrate during the first year. Considering the similarity in the timing of appearance of radiolucency and implant migration, it seems that they could be related. The aim of this study was to determine whether there is a correlation between presence of radiolucency and the migration of Oxford unicompartmental knee arthroplasty implants.

Tantalum marker balls, 0.8 mm in diameter, were inserted around the components of Oxford unicompartmental knee implant (Biomet Orthopedics, Inc, Warsaw, Ind) during surgery. Implant migration and radiolucency associated with Oxford unicompartmental knee arthroplasty were measured at postoperative years 1 and 2 in eight knees (seven patients) using fluoroscopically assisted radiograph and roentgen stereophotogrammetric analysis. During the first year, the tibial component migrated distally and anteriorly, whereas the femoral component migrated proximally and anteriorly. After the first year, there were no further significant migrations of either component. Radiolucencies were noted on 1-year follow-up radiographs in six of the eight tibial components considered. The maximum width of radiolucency was 0.90 mm, and the average width was 0.37 mm. The radiolucencies remained static in extent and thickness after initial appearance. There was no correlation between presence of radiolucency and migration of the tibial component (r=0.08).

This study suggests that radiolucencies are not caused by early movement of the components and confirms that they are not likely to lead to loosening.

Radiolucent lines usually occur at the bone-cement interface after joint arthroplasty. In total hip arthroplasty, evidence clearly shows that radiolucency is associated with loosening. Ritter et al¹ found the incidence of acetabular loosening to be 28% at a mean of 11.7 years if radiolucency was present in zone I² on the postoperative radiograph and 0.7% if radiolucency was not present. With knee arthroplasties, the situation is different, and the clinical significance and etiology of radiolucent lines remain unclear. Although all loose knee implants are associated with radiolucency, all implants with radiolucencies are not loose.³

Radiolucent lines are commonly observed under the tibial component of Oxford unicompartmental knee implants.⁴ The reported high incidence is partly due to the technique of fluoroscopically aligned radiographs; this technique is routinely used at our center for radiologic assessment.³ Previous studies show that there are two distinct types of radiolucencies: the common, or physiologic, type and the rare, or pathologic, type.³Physiologic radiolucency is narrow and nonprogressive and is usually surrounded by a radiodense line.⁵ It usually occurs at approximately postoperative year 1 and has no influence on the long-term clinical outcome; in particular, physiologic radiolucency is not indicative of loosening. Pathologic radiolucency, however, is progressive, broad, poorly defined, and not surrounded by a radiodense line. Its presence can be suggestive of implant loosening. The distinction between physiologic and pathologic radiolucencies is important, because misinterpretation of physiologic lucencies can result in unnecessary revision by surgeons unfamiliar with their benign nature.³

It is not clear why physiologic radiolucencies appear. The reason could be that if the component is initially implanted inadequately, it will move. This early micromovement will result in formation of soft tissue rather than bone at the interface, producing a radiolucent line. It is possible to measure the movement of an implant relative to the bone in three dimensions accurately with roentgen stereophotogrammetric analysis.⁶

This study is aimed at observing the occurrence and progression of radiolucency at the bone-cement interface beneath the tibial component of Oxford unicompartmental knee implants and examining its correlation to component migration using roentgen stereophotogrammetric analysis.

Materials and Methods

Figure 1: The three areas of the tibial component evaluated to determine the extent of radiolucency.

This study was approved by the Oxfordshire local ethics committee. All patients gave informed consent for participation in the study. Seven patients (eight knees replaced with unicompartmental knee arthroplasty) were recruited and followed prospectively. The study group included two men, and the average patient age was 63.2 years (range: 56-75 years, standard deviation [SD]: 6.5 years). Tantalum marker balls, 0.8 mm in diameter, were inserted in the femur and the tibia at the time of surgery. Four to six balls were inserted into each bone. Patients were assessed before surgery and then annually for the first 2 years after surgery. The assessment included clinical scores using the Oxford Knee Score scoring system and the Knee Society scoring system (objective and functional) as well as radiologic evaluation including fluoroscopically screened radiographs performed postoperatively and then at 1 and 2 years after surgery. Migration was measured using roentgen stereophotogrammetric analysis at postoperative years 1 and 2 years, relative to baseline measurements made within 1 week after surgery. Short et al⁷ previously provided a detailed description of the roentgen stereophotogrammetric analysis system, protocol, and accuracy.

Assessment of Radiolucencies

Anteroposterior (AP) radiographs were taken in a standardized manner using fluoroscopic screening. Fluoroscopic screening allows the knee or radiograph tube to be moved until the horizontal surface of the tibial prosthesis in the AP projection lies parallel to the radiography beam. For the purpose of evaluating the extent of lucent zones, the tibial prosthesis was divided into three areas: the medial zone, the keel, and the lateral zone (Figure 1). The thickness of the radiolucency was measured at three different places in each area and the mean thickness of radiolucency calculated. A correction was made for the magnification based on the known curvature radius of the implanted femoral component. Radiographs taken immediately postoperatively were used as baseline, and appearances of the radiolucencies on the subsequent follow-up radiographs were compared.

Femoral radiolucencies were not studied, because they could not be accurately assessed. The inner surface of the femoral component is concave and, therefore, radiolucencies tend to be obscured by the implant.

Results

Clinical assessment. The mean preoperative scores were 12 (Oxford Knee Score), 30.9 (Knee Society score [objective]), and 47.5 (Knee Society score [functional]), and improved to 35.9 (Oxford Knee Score) 87.8 (Knee Society score [objective]), and 82.5 (Knee Society score [functional]) at postoperative year 2 (Table 1).

Preoperative and 2-year Clinical Assessment

Radiolucencies

Six of eight knees had radiolucent lines under the tibial tray at 24 months postoperatively. The average width of the radiolucent lines was 0.37 mm (range: 0.2-0.9 mm). All the radiolucencies appeared during the first year and thereafter remained static in extent and thickness.

Implant Migration

During the first 12 postoperative months, the tibial components migrated distally by an average of 0.43 mm (range: 0.01-1.04 mm; SD: 0.42) and anteriorly by an average of 0.47 mm (range: 0.06-1.16 mm; SD: 0.39). Also, during the first 12 postoperative months, the femoral components migrated proximally by an average of 0.61 mm (range: 0.32-1.16 mm; SD: 0.35) and anteriorly by an average of 0.35 mm (range: 0.0-0.76 mm; SD: 0.29). Neither the tibial nor femoral components displayed any rotational migration. Between postoperative months 12 and 24, no additional statistically significant migration was measured for either component (Tables 2 and 3).

Migration of Tibial Component at 12 and 24 Months Postoperatively

Migration of Femoral Component at 12 and 24 Months Postoperatively

Radiolucency and Migration

There was no correlation (R²= 0.08) between presence and/or extent of radiolucency and migration of the tibial component (Figure 2). Two patients with radiolucency around the tibial component did not show any evidence of tibial component migration and, conversely, two patients with tibial component migration did not have any radiolucency.

Discussion

Figure 2: Graph showing the correlation between radiolucency around the tibial component and its migration.

This study confirms that physiologic radiolucencies frequently occur under the tibial component of the Oxford unicompartmental knee implant. They tend to appear during the first year and remain static thereafter. Because no correlation existed between the presence of radiolucency and implant migration, it is unlikely that the radiolucencies are a result of poor fixation and micromovement.

The precise etiology for the radiolucency occurrence is unknown. We believe that there are good mechanical reasons for the presence of physiologic radiolucency, and its presence indicates that a state of equilibrium exists between mechanics and biology. At implantation, the bone at the implant-bone interface is damaged either by direct trauma or by heat from the curing of the cement, and during the first year it is resorbed. The damaged bone is subsequently replaced by soft tissue, bone, or both. During this period, the implant tends to stabilize by a variable amount. After 1 year, the interface becomes stable, little additional migration occurs, and the tissue type remains static. Two separate previous histologic studies showed fibrocartilaginous connective tissue layer in the radiolucent area.^5,8 The fibrocartilaginous layer sits on a layer of bone, which resembles a subchondral plate with trabeculae terminating in it, and is the sclerotic margin of the radiolucency. The fibrocartilaginous layer is highly compliant and can accommodate the different strains, which inevitably develop when a rigid implant transmits load to a relatively flexible bone.

In a published series, the incidence of radiolucencies around knee replacement varied greatly (32% to 96%).⁵ Tibrewal et al⁵ believed that the reported difference in occurrence of radiolucency was due to differences in radiographic technique. On fluoroscopically assisted radiographs, which are well aligned with the component, the incidence of physiologic radiolucency observation was high. The radiolucency was parallel to the radiograph beam and, thus, was clearly visible. On poorly aligned radiographs, however, the incidence was low because the radiolucent lines were obscured by the component. Figure 3 illustrates this point further. Both radiographs in Figure 3 were taken on the day of the implantation. Figure 3A shows an AP radiograph in which the radiograph tube is inclined 5° cephalad; with this deliberate misalignment, only part of the radiolucency surrounding tibial component is visible. The same joint imaged using the correct fluoroscopically assisted technique in which the radiograph beam is parallel to the tibial implant displays markedly extensive radiolucency (Figure 3B).

Implant migration is difficult to assess on plain radiographs. Roentgen stereophotogrammetric analysis is a proven technique to assess implant position accurately and reliably.⁹ Ryd et al⁶ noted that virtually all knee replacement tibial components migrate initially and, in most cases, this movement stops after 1 year. They used the concept of maximum total point motion; this value is the maximum amount of migration occurring in any direction. Any patient with nominal maximum migration > 200µm at any follow-up after 1 year was considered to be at risk of developing subsequent implant loosening.


Figure 3: Radiograph showing partially obliterated radiolucency due to misalignment of the radiograph tube (A). Radiograph of the same knee as in A, showing the full extent of radiolucency with a correctly aligned radiograph tube (B).

Our results of implant migration after unicompartmental knee arthroplasty are comparable with results from previously published work. Ryd et al¹⁰ measured migration of the tibial component in six successful Marmor unicompartmental knee arthroplasties using roentgen stereophotogrammetric analysis. In their study, migration had rotational, as well as translatory, components in all knees measured. The tibial component tended to translate distally (0.1-2.7 mm at 2 years). They also noted tibial component migration in the first year and no additional significant migration in the second year. The typical migratory pattern for tibial components was a tendency to translate distally and anteriorly. Distal migration is expected, because this is the direction of main loading. Because Oxford unicompartmental knee arthroplasty uses a mobile bearing, anterior load on the tibial component should not occur, except for load from the effects of friction. Therefore, the investigators cannot explain the anterior migration of a tibial component. The femoral component also migrated in a proximal and anterior direction. This migration is to be expected, because it is the direction of the main forces acting on the component in extension and flexion, respectively. Absence of significant migration in the second year (for both the femoral and tibial components) suggests that equilibrium has been reached, leading to secure fixation of the components. None of the components in this study migrated >200µm in the second year, suggesting that a stable interface was established.

Conclusion

No correlation was found between the presence of radiolucency and implant migration. This suggests that poor fixation and early micromovement does not cause radiolucency. In addition, the finding suggests that static physiologic radiolucencies do not lead to loosening. Stronger evidence for this is that although physiologic radiolucencies are common, the 10-year survival for the tibial component is nearly 100%.¹¹

References

Ritter MA, Zhou H, Keating CM, et al. Radiological factors influencing femoral and acetabular failure in cemented Charnley total hip arthroplasties. J Bone Joint Surg Br. 1999; 81:982-986.
DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res. 1976; 121:20-32.
Goodfellow JW, O’Connor JJ, Dodd CAF, Murray DW. Postoperative management and radiography. In: Goodfellow JW, O’Connor JJ, Dodd CAF, Murray DW, eds. Unicompartmental Arthroplasty with the Oxford Knee. Oxford: Oxford University Press; 2006:117-128.
Li MG, Yao F, Joss B, Ioppolo J, Nivbrant B, Wood D. Mobile vs. fixed bearing unicondylar knee arthroplasty: a randomized study on short term clinical outcomes and knee kinematics. Knee. 2006; 13:365-370.
Tibrewal SB, Grant KA, Goodfellow JW. The radiolucent line beneath the tibial components of the Oxford meniscal knee. J Bone Joint Surg Br. 1984; 66:523-528.
Ryd L, Albrektsson BE, Carlsson L, et al. Roentgen stereophotogrammetric analysis as a predictor of mechanical loosening of knee prostheses. J Bone Joint Surg Br. 1995; 77:377-383.
Short A, Gill HS, Marks B, et al. A novel method for in vivo knee prosthesis wear measurement. J Biomech. 2005; 38:315-322.
Ryd L, Linder L. On the correlation between micromotion and histology of the bone-cement interface. Report of three cases of knee arthroplasty followed by roentgen stereophotogrammetric analysis. J Arthroplasty. 1989; 4:303-309.
Karrholm J, Gill RH, Valstar ER. The history and future of radiostereometric analysis. Clin Orthop Relat Res. 2006; 448:10-21.
Ryd L, Boegard T, Egund N, Lindstrand A, Selvik G, Thorngren KG. Migration of the tibial component in successful unicompartmental knee arthroplasty. A clinical, radiographic and roentgen stereophotogrammetric study. Acta Orthop Scand. 1983; 54:408-416.
Ritter MA, Berend ME, Meding JB, Keating EM, Faris PM, Crites BM. Long-term followup of anatomic graduated components posterior cruciate-retaining total knee replacement. Clin Orthop Relat Res. 2001; 388:51-57.

Authors

Mr Rea, and Drs Short, Pandit, Price, Kyberd, Beard, Gill, and Murray are from the Nuffield Orthopaedic Centre, Oxford, England.

Mr Rea and Drs Short, Kyberd, and Pandit have no financial interests in the materials mentioned herein. Drs Price, Beard, and Gill have received research grants from the sponsor. Dr Murray is a consultant for the sponsor and has received compensation from them in the past 12 months.