The Knee 21 (2014) 1254–1257

Contents lists available at ScienceDirect

The Knee

Radiographic evaluation of factors affecting bearing dislocation in the domed lateral Oxford unicompartmental knee replacement A. Gulati a, S. Weston-Simons a, D. Evans a, C. Jenkins b, H. Gray a, C.A.F. Dodd b, H. Pandit a,b, D.W. Murray a,b,⁎ a b

The Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Windmill Road, Headington, Oxford OX3 7LD, UK The Nuffield Orthopaedic Centre, Headington, Oxford, UK

a r t i c l e

i n f o

Article history: Received 2 April 2014 Received in revised form 4 July 2014 Accepted 11 August 2014

Keywords: Dislocation Domed bearing Unicompartmental knee arthroplasty Component positioning

a b s t r a c t Background: The rate of bearing dislocation with the domed lateral Oxford Unicompartmental Knee Replacement (OUKR) in different series varies from 1% to 6% suggesting that dislocation is influenced by surgical technique. The aim of this study was to identify surgical factors associated with dislocation. Methods: Aligned post-operative antero-posterior knee radiographs of seven knees that had dislocated and 87 control knees were compared. Component alignment and position and the alignment of the knee were assessed. All bearing dislocations occurred medially over the tibial wall. Results: Knees that dislocated tended to be overcorrected: Compared with those that did not dislocate, they were in 2° less valgus (p = 0.019) and the tibial components were positioned 2 mm more proximal (p b 0.01). Although the relative position of the centre of the femoral component and the tibial component was the same (p = 0.8), in the dislocating group the gap between the edge of the femoral component and the top of the wall in flexion was 3 mm greater (p = 0.019) suggesting that the components were internally rotated. Conclusions: To minimise the risk of dislocation it is recommended that the knee should not be overstuffed. This is best achieved by selecting the bearing thickness that just tightens the ligaments in full extension, and re-cutting the tibia if necessary. In addition to minimise the gap between the femoral and tibial components through which the bearing dislocates, the femoral component should be implanted in neutral rotation and should not be internally rotated. Level of evidence: Level IV © 2014 Elsevier B.V. All rights reserved.

1. Introduction The number of Unicompartmental Knee replacements (UKRs) implanted annually is continuing to increase [1]. The reduced recovery time, improved functional outcome, and decreased mortality and morbidity compared with Total Knee Replacement (TKR) have contributed to this [1–4]. UKRs can either have mobile or fixed bearings. The mobile bearing devices have lower linear wear rates, but they have the potential complication of bearing dislocation. On the medial side the dislocation rate is low, about 0.5%, principally because the medial collateral ligament is tight so the compartment only opens 2 mm [5,6]. However, it is more common on the lateral side as in flexion the lateral collateral ligament is slack so the compartment opens 7 mm on average [5]. The original flat bearing lateral Oxford Unicompartmental Knee Replacement (OUKR) had an unacceptable five-year survival of 82%, primarily due to the high dislocation rate of 10% [7]. To address this, changes to the operative technique were introduced. These included a lateral para-patella approach, internal rotation of the tibial component

⁎ Corresponding author. E-mail address: [email protected] (D.W. Murray).

http://dx.doi.org/10.1016/j.knee.2014.08.008 0968-0160/© 2014 Elsevier B.V. All rights reserved.

and measures to avoid elevation of the joint line (such as the avoidance of over-stuffing or over-milling the femur) which had previously been shown to be associated with dislocation [6,8,9]. Although these changes reduced the dislocation rate, it was still unacceptably high, so the components were redesigned. In order to increase bearing entrapment and improve kinematics an anatomic domed tibial component with a biconcave bearing was introduced [8]. These alterations have reduced the overall dislocation rate to 1.7% in the designers' hands and the primary dislocation rate to 0.8% [5,9]. Other surgeons have, however, had a higher dislocation rate. For example, Streit et al. [11] have reported a dislocation rate of 6.2% in a series of 50 patients. The aim of this study was to review the post-operative Anteroposterior (AP) radiographs of a series of Domed OUKRs to compare the radiographic features in patients with and without a bearing dislocation. It is hoped that this analysis will help identify key surgical causes of bearing dislocation in patients with domed lateral OUKR. 2. Methods We were able to identify seven knees with domed lateral OUKR (Biomet, Swindon, UK) that had a history of bearing dislocation from our institution. As controls, we identified 94 knees with domed lateral

A. Gulati et al. / The Knee 21 (2014) 1254–1257

OUKR (Biomet, Swindon, United Kingdom) with good quality postoperative radiographs. The control group comprised of all patients with domed lateral OUKR who had screened radiographs with radiographic beam parallel to the tibial component, thereby allowing an accurate assessment. The knees that had the X-ray beam excessively rotated relative to the components were excluded. These radiographs were identified by comparing the width of the vertical wall of the tibial component with the actual width. The radiographs were deemed to be ‘perfectly aligned’ if this measurement was within ±5% of the actual tibial component vertical wall width. All patients met the criteria for having a lateral OUKR: 1) isolated bone-on-bone osteoarthritis in the lateral compartment; 2) an intact anterior cruciate ligament; and 3) a valgus deformity that was correctable [13]. Seven of these knees had a history of bearing dislocation. The previously described standard surgical technique for domed lateral UKR was used [6,10]. The tibial resection was done just above Gerdy's tubercle and was recut if a size 3 bearing could not be accommodated. No ligament releases were undertaken. The aim was to restore normal ligament tension. This was done by implanting the femoral component anatomically and selecting the bearing that just tightened the lateral collateral ligament. In flexion the lateral collateral ligament was slack. The post-operative radiographs were non-weight bearing and were taken with the X-ray beam aligned with the tibial components [12]. One author, (AG), who was blinded to patient outcomes, performed radiological reviews. All analyses were performed in MATLAB version 7.4 (MathWorks, Cambridge, UK). Prior to radiograph evaluation, the sizes of the femoral component and the bearing implanted were retrieved from the patient's primary operation note. The radiographs were analysed in the following manner.

Fig. 1. An AP radiograph of a left knee with a lateral domed OUKR implanted and illustrating how the alignment of components was measured.

1255

Joint evaluation The long axes of the femur and tibia were drawn on the AP view (Fig. 1). Full-length radiographs of the limb were not available, so the anatomical axes were defined as the lines joining the centre of the femur and tibia, 10 cm from the knee joint surfaces, and the centre of the knee [13,14]. The centre of the knee was defined as single-point at the centre and base of the tibial spines as this has been shown to correlate better with the hip–knee angle [13]. The anatomic angle of the knee was the angle between these two axes. The varus/valgus alignments of the femoral and tibial components were measured relative to the long axis of the tibia (Fig. 1). For all angular measurements neutral was considered to be 0° (with varus values designated as being negative). Nine additional measurements were taken (Fig. 2). These measurements were chosen as they provided the most consistent points from which to evaluate the component position relative to each other and bony landmarks. A circle matching the inferior border of the femoral component was drawn. In order to calculate the magnification of the radiograph, the radius of this circle was measured and was compared with the known radius of the femoral component. A line (the tibial line) was drawn perpendicular to the tibial component along the lateral part of the vertical wall of the component. A further line was drawn parallel to this from the tip of the lateral tibial spine (the spine line). All measurements were expressed in millimetres (mm). Measurement 1: The shortest distance between the centre of the circle around the femoral component and the tibial line. This assesses the relationship between the centre of the femoral component and tibial component and thus the distance between the centre of the bearing and the tibial wall. Measurement 2: The distance between the closest part of the femoral component to the tibial line. Measurement 3: The distance from the lower corner of the femoral component to the tibial line. This gives an indication of the position of the distal part of the femoral component relative to the tibial component. Measurement 4: The amount of tibial component lateral underhang. Negative values represent overhang.

Fig. 2. An AP radiograph of a left knee with a lateral domed OUKA implanted and illustrating the measurements.

1256

A. Gulati et al. / The Knee 21 (2014) 1254–1257

Measurement 5: The distance between the medial part of the vertical wall of the tibial component and the spine line, this gives an indication of the medio/lateral position of the tibial component relative to the tibia. Measurement 6: The distance between the tip of the vertical wall of the tibial component to the intersection point of tibial line's intersection with the circle around the femoral component. Measurement 7: The distance between the top of the vertical wall of the tibial component and the top of the vertical tibial cut. Measurement 8: The distance between the top of posterior tibial spine and the top of the tibial component's vertical wall. Measurement 9: The distance between the medial part of the femoral component at the level of the femoral component centre and the tibial line. If the knee is flexed this part of the femoral component is close to the tibia. While measurement 3 provided information about the relative position of the femoral component compared to the tibial component with knee in extension, this measurement, provides an estimate of the same information with the knee in flexion. Statistical analysis was performed using SPSS for Windows (version 18, Chicago, IL, USA). A significant difference was indicated by a p-value b 0.05. Non-parametric analysis was performed using the Mann–Whitney U test for comparison between groups and one-way and multi-variant ANOVAs were used to examine normally distributed data. Intra-observer error was assessed by repeating the measurement of the alignment of the knee in 20 patients (by the same observer with the measurements taken two weeks apart to minimise recall bias). Inter-observer error was assessed between two authors (AG and DE) by analysing measurements recorded in 20 knee radiographs.

3. Results Inter-observer error, was calculated as 0.9° and intra-observer error was 0.4°. The mean age at surgery was 63 years (range: 39–86 years) in those that did not dislocate and 64 years (range 56–74) in those that did. Knees that dislocated had a median bearing thickness of 4 mm (range: 3–5), which was the same as that of non-dislocated knees (range: 3–7). There were no significant differences between the groups for age, bearing size and gender (Table 1). All bearings dislocated medially over the tibial wall. There was a significant difference (p = 0.019) in the overall alignment of the leg between those that did not dislocate (4.6° (SD: 2.41)) and those that did (2.23° (SD: 2.21)). Knees that dislocated had less valgus than those that did not dislocate. There were no significant differences in component alignment between the groups (Table 1). There were significant differences between the groups for measurements 3, 5, 7, 8 and 9. This suggests that relative to the tibial spine, the tibial component was implanted more medially (measurement 5, mean difference 1.75 mm SED 0.67, p = 0.038) and higher Table 1 Demographics of knees between the two groups and alignment data, in degrees (mean (SD)), for all bearings (negative values indicate varus). Non-dislocators

Dislocators

Demographics Number Mean age at surgery (range) M:F% Median bearing thickness (range)

87 64 (56–74) 32:68 4 (3–7)

7 68 (39–86) 14:86 4 (3–5)

Alignment variable AP femoral alignment mean (SD) AP tibial alignment mean (SD) AP knee alignment mean (SD)

2.17 (5.38) 0.57 (2.86) 4.61 (2.41)

2.46 (3.43) 1.35 (3.05) 2.23 (2.21)

(measurement 7, mean difference 2.20 mm, SED 0.31, p = 0.001; measurement 8, mean difference 2.36 mm, SED 0.62, p = 0.008) in knees that dislocated compared to those that did not dislocate. In addition, in the knees that dislocated the femoral component was further from the top of the tibial vertical wall in both extension (measurement 3, mean difference 1.63 mm, SED 0.71, p = 0.040) and flexion (mean 9, mean difference 2.61 mm, SED 1.08, p = 0.019). For none of the other measurements, differences between the two groups were significant or of near significance. 4. Discussion This radiographic analysis demonstrates that the aetiology of lateral mobile bearing dislocation is multi-factorial. Factors identified by this study that contribute to dislocation are overcorrection of the valgus deformity and improper placement of the tibial and femoral components. Overcorrection of the valgus angle has previously been shown to be associated with dislocation of flat lateral meniscal bearings [15]. This study demonstrates that overcorrection is also associated with dislocation of domed lateral bearings. Normal knee alignment is 6° valgus [16,17] (with a range of 2° to 12°) [14]. The alignment was normal (mean 5°) in those that did not dislocate. Knees that dislocated were in significantly (p = 0.019) less valgus and would be considered to be over corrected (mean 2°) (Table 2). This overcorrection is a manifestation of “over stuffing” of the lateral compartment. Ideally once the femoral and tibial components are fixed, a bearing of appropriate thickness that just tightens the lateral collateral ligament (LCL) in extension is inserted. If a bearing 2 or 3 mm thicker is inserted, the knee will be overstuffed. With time the LCL and other soft tissues will stretch to accommodate this “over stuffing”. This may lead to marked laxity in flexion, which in turn may increase the risk of dislocation. To prevent this, the surgeon must select the bearing thickness in full extension, when the lateral soft tissues are tight, and ensure that the bearing is just gripped, and is not tight. In both flexion and extension the distance between the edge of the femoral component and the top of the tibial component wall was significantly larger in patients that dislocated than those that did not (measurements 3 and 9). The difference was more marked in flexion (2.6 mm) than extension (1.6 mm). All the dislocations occurred medially with the bearings subluxing over the tibial wall. The further the edge of the femoral component from the top of the wall the larger the gap through which the bearing has to pass to dislocate, and the more likely it is for the bearing to dislocate. Bearings tend to dislocate in flexion when the lateral ligament is slack. It is therefore important that surgeons should avoid there being a large gap between the components in flexion. There was no relationship between dislocation and the relative medio-lateral position of the centre of the femoral component and the tibial component (measurement 1). Therefore, the most likely reason for there to be a large gap between the femoral component and tibial component in flexion is if the femoral component is

Table 2 A table presenting all measurement values, in mm (mean (SD)), for the total knees and their bearing types.

p-Value

Measurement

Non-dislocators Mean (SD)

Dislocators Mean (SD)

Difference Mean (SED)

p-Value

0.988 0.121 0.019

1 2 3 4 5 6 7 8 9

12.46 (1.97) 0.26 (2.03) 0.92 (2.30) 0.92 (1.69) 3.97 (2.31) 3.70 (1.61) 2.97 (1.20) 4.95 (3.12) 2.40 (2.64)

12.48 (1.94) 0.40 (1.65) 2.55 (1.77) 0.86 (1.10) 2.21 (1.65) 3.73 (1.70) 0.77 (0.75) 2.60 (1.40) 5.01 (2.77)

−0.02 (1.96) −0.14 (2.02) −1.63 (0.71) 0.06 (0.45) 1.75 (0.67) −0.04 (0.67) 2.20 (0.31) 2.36 (0.62) -2.61 (1.08)

0.878 0.757 0.040 0.966 0.038 0.999 0.001 0.008 0.019

A. Gulati et al. / The Knee 21 (2014) 1254–1257

internally rotated. There is a tendency to internally rotate the femoral component as in internal rotation it appears to be aligned with the lateral condyle. This tendency should be avoided and the femoral component should be implanted in neutral rotation. External rotation should be avoided as it may result in anterolateral overhang of the front of the femoral component. One way to achieve neutral rotation is to use the new microplasty instrumentation. The tibial components were implanted about 2 mm more medially in patients that dislocated compared to those that did not. This might increase the risk of dislocation by increasing the gap between the upper part of the wall and the femoral component in flexion, and so should be avoided. However, the tibial component should also not be placed too far lateral as the bearing will impinge against the vertical wall. Therefore a sensible approach from a surgical point of view would be to do the vertical cut in the standard manner, through the centre of the patellar tendon beside the femoral condyle, and then before making the keel cut to do a trial reduction to ascertain if the bearing impinges against the tibial component vertical wall. If it does, the vertical cut should be repeated slightly more medially. The tibial components were implanted about 2 mm more proximal in the knees that dislocated compared to those that did not dislocate (measurements 7 and 8). There are two possible reasons why this might contribute to dislocation. As there was no difference in bearing thickness between the groups, a higher tibial cut would be associated with overcorrection of the valgus deformity as discussed above. Overcorrection would be recognised if the bearing is too tight in full extension so a smaller bearing should be selected. If the thinnest bearing (3 mm) is too tight then it will be necessary to recut the tibia. Alternatively with a high tibial cut, there will be less bone in the region of the tibial spines above the wall so a medial dislocation might be more likely to occur. There are limitations associated with this study. The most important is that there were only seven dislocations. A larger multicentre study might give more insight into the causes of dislocation. Ideally for analysis of lower limb alignment, long leg rather than short leg films should be used. The study by Kraus et al. [18], however, found good correlation between the standard knee radiographs and long leg films. We did not analyse lateral radiographs as in a previous study, these were not found to be as useful as the AP views [8]. In conclusion, this study demonstrates that overcorrecting the valgus deformity in the lateral domed OUKA leads to a significant increase in the risk of dislocation. Assessing the bearing thickness in full extension and avoiding over-tightening the ligaments in this position should prevent this. In addition, to minimise the risk of dislocation, the gap between the top of the tibial vertical wall and the edge of the femoral component in flexion should be as small as possible without the femoral component overhanging or the bearing impinging against the wall. This is best achieved by implanting the

1257

femoral component in a neutral position in flexion, with no internal or external rotation. Conflict of interest statement The author or one of more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other nonprofit organisation with which one or more of the authors are associated. References [1] National Joint Registry, 8th Annual Report; 2011 [http://www.njrcentre.org.uk/ NjrCentre/Portals/0/Documents/NJR%208th%20Annual%20Report%202011.pdf]. [2] Price AJ, Svard U. A second decade lifetable survival analysis of the Oxford unicompartmental knee arthroplasty. Clin Orthop Relat Res 2011;469(1):174–9. [3] Newman J, Pydisetty RV, Ackroyd C. Unicompartmental or total knee replacement: the 15-year results of a prospective randomised controlled trial. J Bone Joint Surg Br 2009;91(1):52–7. [4] Price AJ, Dodd CA, Svard UG, Murray DW. Oxford medial unicompartmental knee arthroplasty in patients younger and older than 60 years of age. J Bone Joint Surg Br 2005;87(11):1488–92. [5] Tokuhara Y, Kadoya Y, Nakagawa S, Kobayashi A, Takaoka K. The flexion gap in normal knees. An MRI study. J Bone Joint Surg Br 2004;86(8):1133–6. [6] Pandit H, Jenkins C, Beard DJ, Price AJ, Gill HS, Dodd CA, Murray DW. Mobile bearing dislocation in lateral unicompartmental knee replacement. Knee 2010; 17(6):392–7. [7] Gunther TV, Murray DM, Miller R, Wallace DA, Carr AJ, O’Connor JJ, Mclardy-Smith P, Goodfellow JW. Lateral unicompartmental arthroplasty with Oxford meniscal knee. Knee 1996;3:33–9. [8] Robinson BJ, Rees JL, Price AJ, Beard DJ, Murray DW, McLardy Smith P, Dodd CA. Dislocation of the bearing of the Oxford lateral unicompartmental arthroplasty. A radiological assessment. J Bone Joint Surg Br 2002;84(5):653–7. [9] Bare JV, Gill HS, Beard DJ, Murray DW. A convex lateral tibial plateau for knee replacement. Knee 2006;13(2):122–6. [10] Weston-Simons JS, Pandit H, Kendrick BJ, Jenkins C, Barker K, Dodd CA, Murray DW. The mid-term outcomes of the Oxford domed lateral unicompartmental knee replacement. Bone Joint J 2014;96-B(1):59–64. [11] Streit MR, Walker T, Bruckner T, Merle C, Kretzer JP, Clarius M, Aldinger PR, Gotterbarm T. Mobile-bearing lateral unicompartmental knee replacement with the Oxford domed tibial component: an independent series. J Bone Joint Surg Br 2012;94(10):1356–61. [12] 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(4):523–8. [13] McDaniel G, Mitchell KL, Charles C, Kraus VB. A comparison of five approaches to measurement of anatomic knee alignment from radiographs. Osteoarthritis Cartilage 2010;18(2):273–7. [14] Moreland JR, Bassett LW, Hanker GJ. Radiographic analysis of the axial alignment of the lower extremity. J Bone Joint Surg Am 1987;69(5):745–9. [15] Robinson BJ, Rees JL, Price AJ, Beard DJ, Murray DM, Oxford Hip and Knee Group. A kinematic study of lateral unicompartmental arthroplasty. Knee 2002;9(3):237–40. [16] Chao EY, Neluheni EV, Hsu RW, Paley D. Biomechanics of malalignment. Orthop Clin North Am 1994;25(3):379–86. [17] Krackow K. The measurement and analysis of the axial deformity at the knee; 2008. [18] Kraus VB, Vail TP, Worrell T, McDaniel G. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum 2005;52(6):1730–5.

Radiographic evaluation of factors affecting bearing dislocation in the domed lateral Oxford unicompartmental knee replacement.

The rate of bearing dislocation with the domed lateral Oxford Unicompartmental Knee Replacement (OUKR) in different series varies from 1% to 6% sugges...
392KB Sizes 0 Downloads 7 Views