Knee Surg Sports Traumatol Arthrosc DOI 10.1007/s00167-013-2372-8

KNEE

Increased patellofemoral pressure after TKA: an in vitro study Ulf G. Leichtle • Markus Wu¨nschel • Carmen I. Leichtle • Otto Mu¨ller • Philipp Kohler Nikolaus Wu¨lker • Andrea Lorenz

•

Received: 2 July 2012 / Accepted: 4 January 2013 Ó Springer-Verlag Berlin Heidelberg 2013

Abstract Purpose Considering the discrepant results of the recent biomechanical studies, the purpose of this study was to simulate dynamic muscle-loaded knee flexion with a large number of specimens and to analyse the influence of total knee arthroplasty (TKA) without and with patellar resurfacing on the patellofemoral pressure distribution. Methods In 22 cadaver knee specimens, dynamic muscleloaded knee flexion (15°–90°) was simulated with a specially developed knee simulator applying variable muscle forces on the quadriceps muscles to maintain a constant ankle force. Patellofemoral pressures were measured with flexible, pressure-sensitive sensor foils (TEKSCAN) and patellofemoral offset with an ultrasound motion-tracking system (ZEBRIS). Measurements were taken on the native knee, after total knee arthroplasty and after patellar resurfacing. Correct positioning of the patellar implant was examined radiologically. Results The maximal patellofemoral peak pressure partly increased from the native knee to the knee with TKA with intact patella (35°–90°, p \ 0.012) and highly increased (twofold to threefold) after patellar resurfacing (20°–90°, p \ 0.001). Concurrently, the patellofemoral contact area decreased and changed from a wide area distribution in the native knee, to a punctate area after TKA with intact patella and a line-shaped area after patellar resurfacing. Patellar resurfacing led to no increase in patellar thickness and patellofemoral offset.

U. G. Leichtle
M. Wu¨nschel (&)
C. I. Leichtle
O. Mu¨ller
P. Kohler
N. Wu¨lker
A. Lorenz Department of Orthopaedic Surgery, University Hospital Tu¨bingen, Hoppe-Seyler-Straße 3, 72076 Tu¨bingen, Germany e-mail: [email protected]

Conclusions Despite correct implantation of the patellar implants and largely unchanged patellofemoral offset, a highly significant increase in pressure after patellar resurfacing was measured. Therefore, from a biomechanical point of view, the preservation of the native patella seems reasonable if there is no higher grade patellar cartilage damage. Keywords Knee simulator
Patellofemoral pressure
Total knee arthroplasty (TKA)
Patellar resurfacing

Introduction Total knee arthroplasty (TKA) is a common surgical procedure with excellent clinical results [7, 32]. However, there is controversy about whether additional patellar resurfacing should be performed. Persistent anterior knee pain is a problem in patients after TKA without patellar resurfacing, but also in patients after TKA with patellar resurfacing. Some authors report a higher incidence of anterior knee pain in patients without patellar resurfacing [33, 37], others a comparable [2, 6] and others a higher incidence of anterior knee pain in patients with patellar resurfacing [5, 27]. Several problems associated with patellar implants are still unsolved, including early aseptic loosening, patellar subluxation, abnormal tracking and increased or abnormal polyethylene wear. Therefore, surgeons may vary in treatment approach and may perform patellar resurfacing in all, none or only in selected cases. Current meta-analyses of the many available clinical studies of TKA support the use of a patellar implant [17, 29]. Patellar resurfacing may be associated with a lower relative risk of reoperation (relative risk 0.57, p = 0.004) and a lower incidence of postoperative anterior knee pain (12.9 % after

123

Knee Surg Sports Traumatol Arthrosc

patellar resurfacing and 24.1 % in the non-resurfacing group) [17]. Swan et al. described in a recent meta-analysis that the published literature seems to favour resurfacing of the patellar routinely. The authors postulated selective patellar resurfacing would be the ideal solution if sound pre-operative criteria could be established [29]. Patellar implants are small devices implanted into the spongy, often osteoporotic bone of the patella and they must resist high pressures. Therefore, it is important to determine whether a patellar implant results in increased patellofemoral pressure. Recent biomechanical studies suggest that there may be potential problems resulting from patellar resurfacing. These studies have used different testing devices and experimental arrangements, resulting in varied conclusions. Some studies have reported no significant increase in patellofemoral pressure after TKA without patellar resurfacing, but a significant increase after patellar resurfacing [21, 28]. Others have shown significantly increased patellofemoral pressure after TKA without patellar resurfacing, but no further significant change after patellar resurfacing [12, 13]. Further studies have compared different types of implants, reporting increased patellofemoral pressures or forces for most implants [23, 30, 31]. Most of these studies are quite old, performed only static measurements in few selected flexion positions [12, 13, 21, 23, 30, 31] and used Fuji films for the pressure measurements [12, 13, 30, 31] which have been shown to have a reduced accuracy [1, 36]. But also the type of the prosthesis [3], the design of the implants [30, 31], the design of the tibial inlay [14], the tibial joint line [11, 16] as well as femoral component rotation [22] and soft tissue balancing [25] have been shown as additional influence factors on patellofemoral pressures and/or kinematics. As different implants were used in different studies (Genesis I, Smith & Nephew, Memphis, USA) [12, 13]; (Oxford Knee; Biomet Ltd, Bridgend, UK; AGC Knee; Biomet Ltd, Bridgend, UK; Insall Burstein II; Zimmer, Swindon, UK) [23]; (Interax ISA, Stryker/Howmedica/Osteonics, Limerick, Ireland) [28]; and (large-sized Profix, Smith & Nephew, Memphis, USA; large-sized Low Contact Stress LCS, DePuy, Warsaw, IN, USA) [21], general statements must be made with caution because of potential differences in results with different prosthetic implants. Therefore, there is a need for a current biomechanical analysis using a dynamic measuring method with a proper method for the measurement of patellofemoral pressures. A complex dynamic knee simulator has been described that enables the performance of partial weight-bearing knee flexion and simulation of different force distributions of the quadriceps muscle [19, 24]. This experimental setting has been used in various studies about TKA [18, 39] or patellofemoral pressures [38] and kinematics [20].

123

The hypothesis of the recent study was that the patellofemoral pressure would not increase after TKA (Genesis II) without and with patellar resurfacing because the examined Genesis II endoprosthesis has an S-shaped and lateralised trochlear groove which should allow an excellent patellar tracking throughout the whole range of motion. Furthermore, patellar resurfacing was done using a precision-reaming technique with special regard to restore the original height of the patella. The second hypothesis was that asymmetric muscle loading on the quadriceps muscle had no influence on patellofemoral pressures.

Materials and methods Dynamic muscle-loaded knee flexion was simulated in a cadaver model using an upright knee simulator in a closed kinetic chain experiment (Fig. 1), and patellofemoral pressure distribution was measured with a flexible resistive pressure sensor. Twenty-two fresh-frozen human cadaver knee specimens (age at death 70 ± 14 years) were examined. The specimens were thawed overnight at room temperature prior to the measurements. The tendons of the quadriceps muscles (vastus lateralis, vastus medialis and rectus femoris) and hamstring muscles (biceps femoris and semimembranosus) were exposed, and the knee joint capsule and the collateral ligaments remained intact. All other skin and soft tissues were removed. The fibula was secured to the tibia with cortical screws. The femur and tibia were cut 15 cm from the joint line and fixed within aluminium cylinders using a bone cement compound (PMMA: Technovit 2060, Heraeus Kulzer, Hanau, Germany) and multiple accurately positioned set screws. Custom-manufactured metal tendon clamps were used to connect the tendons to the muscle actuators (Fig. 1). The research was done in accordance with the Helsinki Declaration and local legislation and was approved by the Ethical Committee of the Medical Faculty of the University of Tu¨bingen. Knee simulator While moving the knee from 15° to 90° knee flexion, the knee simulator applied variable muscle forces on the 5 exposed muscles to maintain a preset value for the vertical force in its ankle bearing. This ankle force was equivalent to the simulated body weight or ground reaction force. Technical details and the validation of the simulator have been described previously [24]. We used an ankle force of 50 N for all trials, which required a total quadriceps force of C600 N at 90° knee flexion. The hamstring forces were kept constant at a low level (20 N) to maintain tension,

Knee Surg Sports Traumatol Arthrosc

reference points. The flexion angle of the tibia with respect to the femur was calculated using MATLAB (The MathWorks, Natick, MA, USA) and used for the description of all pressure-related quantities. Patellofemoral offset was defined as the distance between the origins of the femoral and patellar coordinate systems located at the midpoints of the respective reference points. Patellofemoral pressure

Fig. 1 Experimental arrangement of the knee simulator with a knee specimen in place. The femur and tibia were cut 15 cm from the joint line and fixed within aluminium cylinders using a bone cement compound. Custom-manufactured metal tendon clamps were used to connect the tendons to the muscle actuators. An ultrasonic motion capture system was used to determine flexion angle, and a flexible pressure-sensitive sensor foil was inserted between the femur and patella and fixed with several small sutures

according to the literature [8, 15]. To prevent the specimens from being hyperextended, knee movement was started at 15° flexion. The simulator then moved the specimens with an almost constant flexion speed of 1°/s to the final position of 90°, continuously adapting the muscle forces to maintain the preset ankle force. The applied muscle forces were recorded with a sampling rate of 5 Hz. Tibiofemoral kinematics and patellofemoral offset An ultrasonic motion capture system (CMS-H, Zebris, Isny, Germany, resolution: 0.085 mm, accuracy: 1 mm) was used to determine flexion angle and patellofemoral offset, recording the motion of femur, tibia and patella with a sampling rate of 1 Hz. Using an initial reference measurement in the starting position, segment-based moving coordinate systems were calculated from the recorded tracking data. For the reference measurement, the most prominent medial and lateral points at the femoral epicondyles, tibial plateau and patella were palpated, marked with screws and recorded with a stylus pointer. The flexion axes were defined along the connection lines of these

A flexible pressure-sensitive sensor foil (K-Scan, TekScan, Boston, MA, USA; thickness 0.08 mm) was inserted between the femur and patella and fixed with several small sutures. When the patella was large, 2 foils were glued together, partially overlapping, to cover the complete patellar surface. Before the measurements, the pressure foils were calibrated using a material testing machine (858 MiniBionix II, MTS, Eden Prairie, MN, USA), and a linear calibration method with a repeatability error below 10 % in similar in vitro experiments [35, 36] was applied. During the flexion motion, 1 pressure frame (a sensor matrix of 572 sensor elements) per second was captured by the corresponding application software (IScan 5.83, TekScan, Boston, MA, USA). After the measurements, the path of the centre of pressure (COP), the maximal peak pressure and the patellofemoral contact area were determined from the measured data using MATLAB. Experimental protocol Each series of measurements began with an initial reference measurement to determine the segment coordinate systems. Then, 2 trials were performed with each of 3 muscle conditions: (1) a central (symmetric) quadriceps force distribution (33 % vastus lateralis; 34 % rectus femoris; 33 % vastus medialis); (2) a mainly lateral distribution (67 % vastus lateralis; 33 % rectus femoris; 0 % vastus medialis); and (3) a mainly medial distribution (0 % vastus lateralis; 33 % rectus femoris; 67 % vastus medialis). The entire procedure was performed with 3 knee joint conditions: (1) the native (unmodified) knee joint after arthrotomy; (2) the knee joint after implantation of a TKA prosthesis (Genesis II, Smith & Nephew, London, United Kingdom) without patellar resurfacing; and (3) the knee joint with the TKA and the patella resurfaced. Surgical procedure For implantation of the Genesis II endoprosthesis, an anteromedial approach to the knee joint was used. The medial patellar retinaculum was divided near the patella leaving enough tissue for later re-attachment. The posterior cruciate-retaining Genesis II endoprosthesis was implanted

123

Knee Surg Sports Traumatol Arthrosc

by an experienced orthopaedic surgeon according to the company surgical instructions. Bone preparation was done using an intramedullary femoral and an extramedullary tibial cutting guide. Patellar resurfacing was done using a precision-reaming technique with special regard to an optimal positioning of the implant and restoring of the original height of the patella. To detect possible bony deformities and to document the correct implant position, pre- and postoperative X-rays (in anteroposterior and lateral as well as patellar tangential direction) were performed. The radiographs were analysed with special regard to a correct implant position, the classification by Wiberg [34] and the thickness of the patella prior and after patella resurfacing.

Table 1 Wiberg classification and changes of the patellar thickness during patellar resurfacing Change of patellar thickness during resurfacing Decreased

Total number

Unchanged

Increased

0 mm

?1 mm

?2 mm

-2 mm

-1 mm

Wiberg type 1a

0

0

1

0

0

1

Wiberg type 2a

8

1

9

1

0

19

Wiberg type 3a

0

0

1

0

1

2

Total number

8

1

11

1

1

22

Number of specimens in each category as well as the total numbers is shown a

The measured data were resampled in 1° intervals based on the flexion angle and averaged over the 2 measured repetitions. A repeated measures analysis of variance (ANOVA) of the data was performed for the 3 knee joint conditions and 3 muscle conditions at the 95 % confidence level (p = 0.05) using MATLAB (anova_rm.m, MATLAB File Exchange). Post hoc t tests were performed using the Bonferroni adjustment which resulted in a significance level of p = 0.0167 for the 3 examined conditions. Error bars indicating the standard deviation (SD) and the statistical significance were computed and illustrated in intervals of 5°. The Pearson’s correlation coefficient was computed of maximal peak pressure, contact area and patellar thickness.

Results

Wiberg classification of patellae [34]: type 1: almost equal size of medial and lateral facets; type 2: lateral facet slightly larger than medial facet; type 3: lateral facet much larger than medial facet

75

native 70

patellofemoral offset (mm)

Statistical analysis

TKA TKA+PR

65 60 55 50 45 40 35

The classification by Wiberg resulted in mainly type 2 patellae (lateral facet slightly larger than medial facet, Table 1). An excellent postoperative preservation of the patella thickness was achieved (mean deviation -0.6 ± 1.2 mm, Table 1). For the patellofemoral offset, similar results were attained, and there was only marginal influence of the operation condition (Fig. 2). After patellar resurfacing, a small but significant decrease in the offset was found for some smaller flexion angles (knee flexion 20°–55°, 1–2 mm). Maximal peak pressure increased with increased flexion angle of the knee joint for all 3 joint conditions, except for the situation after TKA without patellar resurfacing, where the maximum was found at 75° of flexion. Patellofemoral contact area increased with increased knee flexion angle for all 3 joint conditions (Fig. 3). A distinct negative correlation was found between the maximal pressure and the contact area for the knee after TKA without (R = -0.56)

123

15

30

45

60

75

90

knee flexion (°) Fig. 2 Patellofemoral offset for the native knee, after total knee arthroplasty (TKA) with intact patella, and after TKA with patellar resurfacing (PR). Significant differences are denoted by an asterisk

and with resurfacing (R = -0.58). While the native knee had a wide area of pressure distribution, the TKA with intact patella had a central punctuate area of localised pressure, and the TKA with patellar resurfacing had a lineshaped distribution of higher pressure. Comparing the native knee with the knee after TKA without resurfacing, the mean maximal peak pressure at the patellofemoral joint significantly increased within a reduced flexion range (knee flexion 35°–85°, p \ 0.012), while the mean patellofemoral contact area decreased (Table 2; Fig. 3). Large variations between the specimens

600

12

native TKA TKA+PR

500

maximal peak pressure (MPa)

Fig. 3 Patellofemoral contact area and maximal peak pressure (with central, symmetric quadriceps muscle force distribution) in the native knee, after total knee arthroplasty (TKA) with intact patella and after TKA with patellar resurfacing (PR). Significant differences are denoted by an asterisk

patellofemoral contact area (mm2)

Knee Surg Sports Traumatol Arthrosc

400 300 200 100 0 15

30

45

60

75

10 8 6 4 2 0 15

90

30

knee flexion (°)

occurred. The greatest increase in the maximal pressure was found for the specimen with the Wiberg type 1 patella (Fig. 4). The range of motion of the COP was reduced after TKA (6 mm) compared with the native knee (12 mm). While the COP motion in mediolateral direction did not change, it was significantly more cranial for the native knee than after TKA with intact patella with higher knee flexion (knee flexion 50°–90°, p \ 0.005, Fig. 5). After additional patella resurfacing, the mean maximal peak pressure was further increased, and patellofemoral

45

60

75

90

knee flexion (°)

contact area was further decreased (Fig. 3). Compared to TKA without resurfacing, the maximal peak pressure was 1.5 to 2.5-fold greater after the resurfacing (knee flexion 20°–90°, p \ 0.001). The range of motion of the COP was reduced again (3 mm), and the motion was slightly cranial to the motion after TKA without resurfacing for lower knee flexion (knee flexion 20°–45°, p \ 0.013, Fig. 5). Comparing the knee after TKA with patellar resurfacing to the native knee, a twofold to threefold increase in the maximal peak pressure was found (knee flexion 20°–90°,

Table 2 Effect of total knee arthroplasty (TKA), patellar resurfacing and different muscle loading conditions on patellofemoral pressure, contact area and COP motion in human cadaver specimens (N = 22) Condition

Native knee

Muscle load

Central

TKA (intact patella) Lateral

Medial

Central

TKA with patellar resurfacing

Lateral

Medial

Central

Lateral

Medial

Maximal peak pressure (MPa)a 1.9 ± 0.8

2.0 ± 1.0

2.1 ± 1.0

2.5 ± 1.1

2.2 ± 0.8

2.9 ± 1.4

5.1 ± 1.5

4.6 ± 1.5

5.4 ± 1.5

Range (minimum to maximum)

Mean ± SD

0.8–3.0

0.8–3.3

0.7–3.6

1.1–3.9

1.2–3.1

1.0–4.7

2.0–7.0

1.8–6.6

2.0–7.0

Variation coefficient (%)b

36

39

35

41

39

43

46

53

44

Flexion angle of maximum (°)c

90

90

90

78

90

81

90

90

89

Flexion angle of minimum (°)c

15

16

17

15

21

15

15

17

15

Contact area (mm2)d

230 ± 120

190 ± 90

230 ± 130

170 ± 110

140 ± 70

160 ± 110

70 ± 50

70 ± 40

70 ± 50

COP, ML motion (mm)e

2.9 ± 1.9

7.5 ± 2.2

-0.2 ± 1.3

1.9 ± 3.0

5.0 ± 3.7

-0.3 ± 2.2

2.3 ± 3.2

6.1 ± 2.7

-0.3 ± 2.7

COP, CC motion (mm)f

11.7 ± 4.3

11.4 ± 4.0

11.7 ± 4.6

9.5 ± 2.2

8.4 ± 2.1

9.3 ± 2.4

11.5 ± 1.5

10.0 ± 1.6

11.8 ± 1.8

a

Maximal peak pressure of patellofemoral joint; data reported as mean ± SD over entire flexion range (15°–90°)

b

Variation coefficient is defined as SD/mean

c

Flexion angle of maximum or minimum values of maximal peak pressure

d

Patellofemoral contact area; data reported as mean ± SD over entire flexion range (15°–90°) Mediolateral motion of the centre of pressure; data reported as mean ± SD over entire flexion range (15°–90°)

e f

Craniocaudal motion of the centre of pressure; data reported as mean ± SD over entire flexion range (15°–90°)

123

300

8 type 1 type 2 type 3

200

maximal peak pressure (MPa)

Fig. 4 Differences of patellofemoral contact area and maximal peak pressure between the knee after TKA without resurfacing and the native knee for all 22 specimens. The different colours denote the three types of patellae according to the classification by Wiberg

patellofemoral contact area (mm²)

Knee Surg Sports Traumatol Arthrosc

100 0 −100 −200 −300

15

30

45

60

75

6

4

2

0

−2 15

90

30

knee flexion (°) 25

lateral

10

5

0

native TKA TKA+PR

−5

medial

−10 15

30

45

60

75

knee flexion (°)

p \ 0.001). No correlation was found between increased patellar thickness and increased patellofemoral pressure (R = 0.20) or decreased patellofemoral contact area (R = 0.28). For lower knee flexion, the COP of the knee with resurfaced patellar was moving more cranial (knee flexion 25°–40°, p \ 0.004), while for higher flexion angles, the COP of the native knee was moving more cranial (knee flexion 80°–90°, p \ 0.009). Application of 3 different force distributions of the quadriceps muscles resulted in only a small effect on maximal peak pressure (Fig. 6), contact area (Table 2) and craniocaudal motion of the COP (Table 2). However, there were significant differences in mediolateral COP motion between the 3 different muscle conditions for all 3 joint conditions (knee flexion 20°–90°, p \ 0.004, Fig. 6). With mainly lateral muscle loading, significant differences were found between the 3 joint conditions (knee flexion 15°–35°, p \ 0.04).

Discussion The most important finding of our study was that compared with the native knee maximal peak pressure at the

123

COP − CC movement (mm)

15

COP − ML movement (mm)

Fig. 5 Movement of the centre of pressure (COP) in the mediolateral (ML) and craniocaudal (CC) directions in the native knee, after total knee arthroplasty (TKA) with intact patella and after TKA with patellar resurfacing (PR) (with central symmetric quadriceps force distribution). Significant differences are denoted by an asterisk

45

60

75

90

75

90

knee flexion (°) cranial

20

15

10

5

caudal

90

0 15

30

45

60

knee flexion (°)

patellofemoral joint was increased after TKA with intact patella and further increased after patellar resurfacing (Table 2; Fig. 3). Therefore, our first hypothesis had to be rejected. Furthermore, the large patellofemoral contact area of the native knee joint was changed by TKA, with a concentration of greater peak pressure to a smaller area on the patella or patellar implant. These results suggest that TKA and patellar resurfacing result in potentially clinically important changes in the biomechanical properties of the patellofemoral joint. Our results varied partially from those previously reported. No significant changes of the measured patellofemoral contact stresses after TKA without patella resurfacing but a significant increase after additional patellar resurfacing were found by Matsuda et al. [21] and Stukenborg-Colsman et al. [28]. On the contrary, significantly increased average and maximal pressures were reported by Fuchs et al. [13] in their measurements following TKA without resurfacing. After further patellar resurfacing, unchanged patellofemoral pressures but a reduced contact area were found by the same group [12]. These studies differ from our study in particular regarding the experimental setting. Only static measurements at

Knee Surg Sports Traumatol Arthrosc 10

COP − ML movement (mm)

maximal peak pressure (MPa)

8 7 6 5 4 3 2 1 0 15

lateral

8

central−native central−TKA central−TKA+PR lateral−native lateral−TKA lateral−TKA+PR medial−native medial−TKA medial−TKA+PR

6 4 2 0

−2 −4

medial

30

45

60

75

90

−6 15

knee flexion (°)

30

45

60

75

90

knee flexion (°)

Fig. 6 Effect of asymmetric muscle loading on maximal peak pressure and centre of pressure movement. Muscle loads were applied with 3 muscle conditions: central (symmetric), mainly lateral and mainly medial quadriceps force distribution. The maximal peak

pressure and the movement of the centre of pressure (COP) in mediolateral direction (ML) are plotted against knee flexion angle for the native knee, after total knee replacement (TKA) with intact patella and after TKA with patellar resurfacing (PR)

different angles of knee flexion with constant [12, 13] or adapted [21] muscle loading were performed by Matsuda et al. and Fuchs et al. while a constant flexion torque was simulated by Stukenborg-Colsman et al. [28]. In contrast to many available testing devices, our knee simulator enables the simulation of a continuos flexion motion with an almost physiological muscle force progression during flexion movement. This innovative setting is considered a major advancement compared with the simulation of a constant quadriceps force or static measurements. Compared with our study, similar results concerning the increasing patellofemoral pressure in TKA with patellar resurfacing were achieved only by Matsuda et al. [21] and Stukenborg-Colsman et al. [28]. In contrast to these studies, an additional significant difference between the native knee and the knee after TKA without resurfacing was detected in our study within a flexion range of 35°–85° of flexion. Besides the different experimental settings, this finding is partly attributed to the larger number of specimens analysed. While previous studies were performed with only 5–6 specimens [12, 13, 21, 23, 28], our study included 22 human knee joints. As the patellofemoral pressure in TKA without resurfacing is highly depending on the individual knee anatomy, the variation is high, and therefore, a larger number of specimens are necessary to detect a statistically significant difference. The measurement technique used for the patellofemoral pressure measurement might also influence the results. As in the present study, many newer studies were using flexible pressure sensor foils (TekScan) [3, 14, 21, 25, 26, 28, 38], while most of the older studies used pressure-sensitive films (Fuji Prescale type ‘super low’ film) [12, 13, 30, 31]. Not the patellofemoral pressure distribution, but the patellofemoral force was measured by Miller [23] using a

non-invasive method. In his study, a decreased patellofemoral force in extension and a 20 % increase in force in flexion after TKA were described. The use of pressuresensitive foils for the measurement of patellofemoral pressure has the advantage of continuous measurement of the pressure distribution which prevents the influence of shear forces [36] and it is less sensitive to changes of temperature and humidity. Tekscan foils are much thinner than Fuji films, and they have shown to be more accurate [1, 36]. However, an exact placement sometimes is difficult, and a careful calibration is necessary to achieve accurate pressure data [4]. The design of the TKA prostheses also contributes to the patellofemoral pressure distribution, which had been shown, for example, in static pressure measurements on different designs [30, 31] or in finite element analyses [9, 10]. Besides the design of the patellofemoral joint surface itself, this might be attributed to a possibly changed patellofemoral offset. Compared with other implants, the Genesis II total knee endoprosthesis has a relatively flat S-shaped and lateralised trochlear groove with a high congruence to the patellar implant. The aim of this special design is to preserve the biomechanics of the native knee and to restore the original height of the patella and avoid overstuffing. In our study, an almost unchanged height of the patella during resurfacing and only a marginal effect on the patellofemoral offset, which also includes height variations of the femur during the operation procedure, were shown. Despite the avoidance of overstuffing, a highly significant increase in maximal peak pressures was found. The highly congruent implant design of the patellar and femoral components could not prevent a clear reduction in the patellofemoral contact area compared with the native patella. The high correlations of contact areas and

123

Knee Surg Sports Traumatol Arthrosc

maximum pressures indicate that these decreasing contact areas can be considered as a main reason for the pressure increase. Most of the specimens included in our study had a patella with a slightly larger lateral facet (Wiberg type 2). The one patella with equal lateral and medial facets (Wiberg type 1) led to the greatest increase in maximum patellofemoral pressure from the native knee to the knee after TKA without resurfacing (Fig. 4). Maybe, the tested implant designs are more beneficial for slightly asymmetric patellae. In addition to the effect on the maximal pressure, in our recent study, an interesting influence of TKA on the movement of the centre of pressure (COP) was shown. In mediolateral direction, no significant influence of TKA and patellar resurfacing was found which might be attributed to the S-shaped trochlear design. The range of motion of the COP in the craniocaudal direction was found to be much larger in the native knee than after TKA without and with patellar resurfacing (Fig. 5). This effect might be explainable by the reduced patellofemoral contact area in this direction during the observed rolling motion of the patella on the femur from caudal to cranial. The observation that the COP of the TKA with resurfaced patella was slightly cranial to that of the TKA with intact patella (Fig. 5) could be explained by the mostly slightly cranial implant position and the ball-shaped design of the patellar implant compared to the more cylindrical design of the native patella. The variation of forces applied at the different parts of the quadriceps muscle had only a small effect on the contact area, maximal peak patellofemoral pressure (Fig. 6) and craniocaudal motion of the centre of pressure (COP), but it had significant effects on the mediolateral COP motion (Fig. 6). Therefore, our second hypothesis was rejected as well. Interestingly, for flexion angles of more than 35°, different quadriceps loading had a similar effect on the mediolateral COP motion for the native knee and the knee after TKA without and with resurfacing. Only for small flexion angles, a varying influence was found, especially for mainly lateral muscle loading. In this region, the guidance of the unresurfaced patellar within the femoral implant seems to be greater than for the native knee and for the resurfaced patella. As we have shown before in the native knee [20, 38], the varying motion of the COP does not have to be accompanied by a greater mediolateral shift of the patella. Small changes in patellar tilt and rotation may be sufficient. Regarding the results of our biomechanical in vitro study, general statements on total knee arthroplasty must be made carefully, because our examinations were done with only one TKA design (Genesis II). There also remain some limitations concerning the experimental setting, including the reduced simulated flexion range of 15°–90°, the

123

constant hamstring forces or the small part of simulated bodyweight. These limitations are similar to many in vitro studies and have been discussed previously [19, 24, 38]. Nevertheless, our knee simulator is considered as an advanced setting, permitting the simulation of an almost physiological muscle-loaded knee flexion, as far this is possible in in vitro studies. Concerning the clinical relevance, our data contribute to the current discussion whether or not to use a patellar implant and support our philosophy of doing patellar resurfacing only in cases of a symptomatic patellofemoral arthrosis.

Conclusion In our biomechanical in vitro examination of 22 human knee specimens, significant differences of maximal peak pressure for a flexion range of [35° after TKA without patellar resurfacing compared to the native knee were found. For smaller flexion angles, no significant difference was detected. Despite correct implantation of the patellar implants without overstuffing and a largely unchanged patellofemoral offset, further patellar resurfacing resulted in a highly significant 1.5- to 2.5-fold increase in patellofemoral peak pressure. The relatively flat design of the S-shaped trochlear groove of the Genesis II endoprosthesis with a high congruence between the patellar and the femoral component could not prevent a clear reduction in the patellofemoral contact area and a huge increase in maximal peak pressure compared with the native patella. Therefore, from a biomechanical point of view, patellar resurfacing is not recommended if no higher grade patellar cartilage damage is present. Acknowledgments We gratefully acknowledge the financial support of Smith & Nephew. Conflict of interest This study was partly funded by Smith & Nephew, but the sponsor had neither been involved in design and execution of the experiments nor in the preparation of the manuscript.

References 1. Bachus KN, DeMarco AL, Judd KT, Horwitz DS, Brodke DS (2006) Measuring contact area, force, and pressure for bioengineering applications: using Fuji Film and TekScan systems. Med Eng Phys 28(5):483–488 2. Barrack RL, Bertot AJ, Wolfe MW, Waldman DA, Milicic M, Myers L (2001) Patellar resurfacing in total knee arthroplasty. A prospective, randomized, double-blind study with five to seven years of follow-up. J Bone Joint Surg Am 83-A(9):1376–1381 3. Becher C, Heyse TJ, Kron N, Ostermeier S, Hurschler C, Schofer MD et al (2009) Posterior stabilized TKA reduce patellofemoral contact pressure compared with cruciate retaining TKA in vitro. Knee Surg Sports Traumatol Arthrosc 17(10):1159–1165

Knee Surg Sports Traumatol Arthrosc 4. Brimacombe JM, Wilson DR, Hodgson AJ, Ho KC, Anglin C (2009) Effect of calibration method on Tekscan sensor accuracy. J Biomech Eng 131(3):034503 5. Burnett RS, Haydon CM, Rorabeck CH, Bourne RB (2004) Patella resurfacing versus nonresurfacing in total knee arthroplasty: results of a randomized controlled clinical trial at a minimum of 10 years’ followup. Clin Orthop Relat Res 428:12–25 6. Campbell DG, Duncan WW, Ashworth M, Mintz A, Stirling J, Wakefield L et al (2006) Patellar resurfacing in total knee replacement: a ten-year randomised prospective trial. J Bone Joint Surg Br 88(6):734–739 7. Emmerson KP, Moran CG, Pinder IM (1996) Survivorship analysis of the Kinematic Stabilizer total knee replacement: a 10- to 14-year follow-up. J Bone Joint Surg Br 78(3):441–445 8. Escamilla RF, Fleisig GS, Zheng N, Barrentine SW, Wilk KE, Andrews JR (1998) Biomechanics of the knee during closed kinetic chain and open kinetic chain exercises. Med Sci Sports Exerc 30(4):556–569 9. Fitzpatrick CK, Baldwin MA, Laz PJ, FitzPatrick DP, Lerner AL, Rullkoetter PJ (2011) Development of a statistical shape model of the patellofemoral joint for investigating relationships between shape and function. J Biomech 44(13):2446–2452 10. Fitzpatrick CK, Rullkoetter PJ (2012) Influence of patellofemoral articular geometry and material on mechanics of the unresurfaced patella. J Biomech 45(11):1909–1915 11. Fornalski S, McGarry MH, Bui CN, Kim WC, Lee TQ (2012) Biomechanical effects of joint line elevation in total knee arthroplasty. Clin Biomech (Bristol, Avon) 27(8):824–829 12. Fuchs S, Schu¨tte G, Witte H, Rosenbaum D (2000) Retropatellar contact characteristics in total knee arthroplasty with and without patellar resurfacing. Int Orthop 24(4):191–193 13. Fuchs S, Schu¨tte G, Witte H, Rosenbaum D (2000) What retropatellar changes result by implantation of a superficial knee joint prosthesis? Unfallchirurg 103(11):972–976 14. Heyse TJ, Becher C, Kron N, Ostermeier S, Hurschler C, Schofer MD et al (2010) Patellofemoral pressure after TKA in vitro: highly conforming vs. posterior stabilized inlays. Arch Orthop Trauma Surg 130(2):191–196 15. Isear JA Jr, Erickson JC, Worrell TW (1997) EMG analysis of lower extremity muscle recruitment patterns during an unloaded squat. Med Sci Sports Exerc 29(4):532–539 16. Ko¨nig C, Sharenkov A, Matziolis G, Taylor WR, Perka C, Duda GN et al (2010) Joint line elevation in revision TKA leads to increased patellofemoral contact forces. J Orthop Res 28(1):1–5 17. Li S, Chen Y, Su W, Zhao J, He S, Luo X (2011) Systematic review of patellar resurfacing in total knee arthroplasty. Int Orthop 35(3):305–316 18. Lo J, Mu¨ller O, Dilger T, Wu¨lker N, Wu¨nschel M (2011) Translational and rotational knee joint stability in anterior and posterior cruciate-retaining knee arthroplasty. Knee 18(6):491–495 19. Lo J, Mu¨ller O, Wu¨nschel M, Bauer S, Wu¨lker N (2008) Forces in anterior cruciate ligament during simulated weight-bearing flexion with anterior and internal rotational tibial load. J Biomech 41(9):1855–1861 20. Lorenz A, Mu¨ller O, Kohler P, Wu¨nschel M, Wu¨lker N, Leichtle UG (2012) The influence of asymmetric quadriceps loading on patellar tracking—an in vitro study. Knee 19(6):818–822 21. Matsuda S, Ishinishi T, White SE, Whiteside LA (1997) Patellofemoral joint after total knee arthroplasty. Effect on contact area and contact stress. J Arthroplast 12(7):790–797 22. Merican AM, Ghosh KM, Iranpour F, Deehan DJ, Amis AA (2011) The effect of femoral component rotation on the

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

kinematics of the tibiofemoral and patellofemoral joints after total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 19(9):1479–1487 Miller RK, Goodfellow JW, Murray DW, O’Connor JJ (1998) In vitro measurement of patellofemoral force after three types of knee replacement. J Bone Joint Surg Br 80(5):900–906 Mu¨ller O, Lo J, Wu¨nschel M, Obloh C, Wu¨lker N (2009) Simulation of force loaded knee movement in a newly developed in vitro knee simulator. Biomed Tech (Berl) 54(3):142–149 Peretz JI, Driftmier KR, Cerynik DL, Kumar NS, Johanson NA (2012) Does lateral release change patellofemoral forces and pressures? A pilot study. Clin Orthop Relat Res 470:903–909 Saranathan A, Kirkpatrick MS, Mani S, Smith LG, Cosgarea AJ, Tan JS et al (2012) The effect of tibial tuberosity realignment procedures on the patellofemoral pressure distribution. Knee Surg Sports Traumatol Arthrosc 20(10):2054–2061 Smith AJ, Wood DJ, Li MG (2008) Total knee replacement with and without patellar resurfacing: a prospective, randomised trial using the profix total knee system. J Bone Joint Surg Br 90(1):43–49 Stukenborg-Colsman C, Ostermeier S, Burmester O, Wirth CJ (2003) Dynamic in vitro measurement of retropatellar pressure after knee arthroplasty. Orthopade 32(4):319–322 Swan JD, Stoney JD, Lim K, Dowsey MM, Choong PF (2010) The need for patellar resurfacing in total knee arthroplasty: a literature review. ANZ J Surg 80(4):223–233 Szivek JA, Anderson PL, Benjamin JB (1996) Average and peak contact stress distribution evaluation of total knee arthroplasties. J Arthroplast 11(8):952–963 Takeuchi T, Lathi VK, Khan AM, Hayes WC (1995) Patellofemoral contact pressures exceed the compressive yield strength of UHMWPE in total knee arthroplasties. J Arthroplast 10(3):363–368 Vessely MB, Whaley AL, Harmsen WS, Schleck CD, Berry DJ (2006) The Chitranjan Ranawat Award: long-term survivorship and failure modes of 1000 cemented condylar total knee arthroplasties. Clin Orthop Relat Res 452:28–34 Waters TS, Bentley G (2003) Patellar resurfacing in total knee arthroplasty. A prospective, randomized study. J Bone Joint Surg Am 85-A(2):212–217 Wiberg G (1941) Roentgenographs and anatomic studies on the femoropatellar joint: with special reference to chondromalacia patellae. Acta Orthop Scand 12(1–4):319–410 Wilson DC, Niosi CA, Zhu QA, Oxland TR, Wilson DR (2006) Accuracy and repeatability of a new method for measuring facet loads in the lumbar spine. J Biomech 39(2):348–353 Wirz D, Becker R, Li SF, Friederich NF, Muller W (2002) Validation of the Tekscan system for statistic and dynamic pressure measurements of the human femorotibial joint. Biomed Tech (Berl) 47(7–8):195–201 Wood DJ, Smith AJ, Collopy D, White B, Brankov B, Bulsara MK (2002) Patellar resurfacing in total knee arthroplasty: a prospective, randomized trial. J Bone Joint Surg Am 84-A(2):187–193 Wu¨nschel M, Leichtle U, Obloh C, Wu¨lker N, Mu¨ller O (2011) The effect of different quadriceps loading patterns on tibiofemoral joint kinematics and patellofemoral contact pressure during simulated partial weight-bearing knee flexion. Knee Surg Sports Traumatol Arthrosc 19(7):1099–1106 Wu¨nschel M, Lo J, Dilger T, Wu¨lker N, Mu¨ller O (2011) Influence of bi- and tri-compartmental knee arthroplasty on the kinematics of the knee joint. BMC Musculoskelet Disord 12:29

123