- Email: [email protected]
Tibial Implant Fixation Behavior in Total Knee Arthroplasty: A Study With Five Different Bone Cements
Journal Pre-proof Tibial implant fixation behaviour in total knee arthroplasty – a study with five different bone cements Thomas M. Grupp, Christoph Schilling, Jens Schwiesau, Andreas Pfaff, Brigitte Altermann, William M. Mihalko PII:
S0883-5403(19)30868-X
DOI:
https://doi.org/10.1016/j.arth.2019.09.019
Reference:
YARTH 57523
To appear in:
The Journal of Arthroplasty
Received Date: 4 July 2019 Revised Date:
15 August 2019
Accepted Date: 13 September 2019
Please cite this article as: Grupp TM, Schilling C, Schwiesau J, Pfaff A, Altermann B, Mihalko WM, Tibial implant fixation behaviour in total knee arthroplasty – a study with five different bone cements, The Journal of Arthroplasty (2019), doi: https://doi.org/10.1016/j.arth.2019.09.019. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 The Author(s). Published by Elsevier Inc.
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
1 2
Tibial implant fixation behaviour in total knee arthroplasty – a study with five different bone cements
3 4 5
Abstract
6
The objectives of this study were to (1) evaluate if there is a potential difference of cemented implant fixation
7
strength between tibial components made out of cobalt-chromium (CrCoMo) and of a ceramic zirconium
8
nitride (ZrN) multi-layer coating and to (2) test their behaviour with five different bone cements in a
9
standardised in vitro model for testing of the implant-cement-bone interface conditions. We also analyzed (3)
10
wether initial fixation strength is a function of timing of the cement apposition and component implantation by
11
an early, mid-term and late usage within the cement specific processing window.
12
An in vitro study using a synthetic polyurethane foam model was performed to investigate the implant fixation
13
strength after cementation of tibial components by a push-out test. A total of 20 groups (n = 5 each) was
14
used: Vega PS cobalt-chromium tibia and Vega PS ZrN tibia with the bone cements BonOs R, Smart
15
Set HV, Cobalt HV, Palacos R, Surgical Simplex P, respectively, using mid-term cement apposition.
16
Three different cement apposition times, early, mid-term and late usage were tested with a total of 12 groups
17
(n = 5 each) with the bone cements BonOs R and Smart Set HV.
18
From the observations, it is concluded that there is no significant difference in implant-cement-bone fixation
19
strength between cobalt-chromium and ZrN multi-layer coated Vega tibial trays tested with five different
20
commonly used bone cements. Apposition of bone cements and tibial tray implantation in the early to mid of
21
the cement specific processing window is beneficial in regard to interface fixation in TKA.
®
®
®
®
®
®
®
®
®
®
22 23 24
Keywords
25
cements, timing of the cement apposition
total knee arthroplasty, tibial implant fixation, ceramic zirconium nitride (ZrN) multi-layer coating, five bone
26
27
I. Introduction
28
Total knee arthroplasty (TKA) can be considered as a successful clinical standard of care for the treatment of
29
endstage degenerative osteoarthritis, rheumatoid or other inflammatory arthritis and osteonecrosis, with an
30
increasing demand over the past two decades [1,2,3,4]. For primary TKA the cumulative percentages of
31
revision for all reasons are comparably low with 1-3 % early failures within 1 to 3 years post-operatively and
32
with 4-8 % revisions over a period of 10 to 15 years [2,4,5,6,7,8]. Sadoghi et al. [9] performed a 1
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
33
complication-based analysis of TKA revisions including 391,913 primary and 36,307 revision cases in a 30
34
years period entered in the joint registries of Sweden, Norway, Finland, Denmark, Australia and New
35
Zealand and found the most common causes to be aseptic loosening in 29.8 %, septic loosening in 14.8 %,
36
pain in 9.5 %, wear in 8.2 % and instability in 6.2 % of the cases. Niinimaeki [10] analysed the reasons for
37
knee revision based on the registries of Norway, Sweden, Australia, New Zealand and England & Wales in
38
2015 and found that the leading indication for revision was aseptic loosening (range 22.8 to 29.7 %).
39
Infection was the second leading indication for revision in Sweden, Australia and England & Wales (20.6 to
40
21.7 %), while pain was the second most common cause in Norway and New Zealand (27.4 & 22.0 %) [11].
41
Delanois et al. [12] used the National Inpatient Sample (NIS) database and identified all revision TKA
42
procedures performed between 2009 and 2013 in the United States. Collecting clinical and demographic
43
data for 337,597 procedures, infection (20.8 %), mechanical loosening (20.3 %) and instability (7.5 %) were
44
the most common specified etiologies for revision TKA [12].
45
In a multi-centre study including five centers in the United States, Lombardi et al. [13] reviewed 844 TKA
46
revisions which were performed in a period from 2010 to 2011. Aseptic loosening (31.2 %) was the
47
predominant failure mode, followed by instability (18.7 %), infection (16.2 %) and polyethylene wear (10 %).
48
The mean time to revision was 5.9 years (from 10 days to 31 years), whereas 36 % of all revisions occurred
49
within two years after primary surgery, 25 % between 2 and 5 years and 29 % between 5 and 15 years. The
50
remaining 10 % of late revisions (>15 years service in vivo) were mainly related to polyethylene wear [13].
51
Analyzing the incidence of early revision diagnosis for primary TKA in the National joint registries of Australia
52
and England, Wales & Northern Ireland in two years since primary procedure, the predominant reason for
53
revision was infection (0.6 % AUS; 0.6 % UK), followed by loosening (0.4 % AUS; 0.5 % UK). At five years
54
the cumulative incidence of loosening (0.8 % AUS; 0.9 % UK) is in an equal level to infection (0.8 % AUS;
55
0.8 % UK), whereas for the mid- to long-term (between 5 and 16 years) loosening is the most common
56
reason for revision [2,5].
57
Apart from infection, instability and sub optimal alignment, aseptic tibial component loosening remains a
58
major cause of TKA failure [8,14,15,16,17]. Based on 3,572 primary TKA revisions in the Swedish knee
59
arthroplasty registry, Sundberg et al. [18] reported the implant removal of only the tibial component in 7.1 %
60
and only the femoral component in 0.9 % of the cases, during a 10-year period from 2003 to 2012. Gothesen
61
et al. [18] evaluated the aseptic implant revisions within a cohort of 17,782 primary cemented TKA’s reported
62
to the Norwegian arthroplasty register during the years from 1994 to 2009, and found a ratio of 2.8 between
63
tibial and femoral component loosening, whereas Furnes et al. [20] extracted a ratio of 3.7 for the similar
2
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
64
database up to 2004. To answer the question if the causes of revision for TKA have changed during the past
65
two decades in Norway, Dyrhovden et al. [8] selected two cohorts of primary TKA’s implanted during 1994 to
66
2004 (Period 1; n = 17,404) and during 2005 to 2015 (Period 2; n = 43,219) and found that the relative risk
67
for revision (RR) has been unchanged for the tibia (RR 1.0), but substantially decreased for the femur (RR
68
0.3). This emphasizes the need for stable primary and secondary fixation of tibial trays, which is dependent
69
on multiple factors including implant design, bone interface preparation, surgical technique and cementation
70
[14,17,21,22,23,24].
71
Several studies have been undertaken to analyze the primary and secondary stability of tibial trays in vitro, in
72
vivo, in silico and ex vivo [14,15,16,21,25,26,27]. Various examinations have been undergone, such as
73
cement penetration depth analysis in the proximal tibia [28,29,30], radiographic short-term outcome
74
measurements by incidence of radiolucent lines [27], finite element analysis (FEA) to assess resulting
75
interface bone stresses and strains [15,31,32], static tension to measure the implant-cement-bone interface
76
bonding [21,26] and dynamic compression-shear loading [14,33] until implant-bone interface failure. In
77
previous studies [14,15,16,21,34] the influences of implant design, bone interface preparation, full or surface
78
cementing technique on the primary stability of bicompartmental tibial trays have been examined.
79 80
II. Objectives
81
The objectives of this study were to (1) evaluate if there is a potential difference of cemented implant fixation
82
strength between tibial components made out of cobalt-chromium (CrCoMo) and of a ceramic zirconium
83
nitride (ZrN) multi-layer coating and to (2) test their behaviour with five different bone cements in a
84
standardised in vitro model for testing of the implant-cement-bone interface conditions. We also sought to
85
analyze (3) wether initial fixation strength is a function of timing of the cement apposition and component
86
implantation by an early, mid-term and late usage within the cement specific processing window as reported
87
by the manufacturer.
88 89
III. Materials & Methods
90
An in vitro study using a synthetic polyurethane foam model (20 pcf cellular rigid foam, Sawbones, Sweden)
91
according to ASTM F1839-08 was performed to investigate the implant fixation strength after cementation of
92
tibial components by a push-out test. Size T0 tibial trays from the Vega knee system (Aesculap, Germany)
93
were chosen as they are available in both cobalt-chromium (CoCrMo) and zirconium nitride (ZrN) multi-layer
®
3
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
94
surfaces, and as the smallest standard size have the least surface area for fixation. Five different bone
95
cements were investigated to cover a diverse range of common clinically used bone cements.
96
For the evaluation of the research questions two lines with corresponding groups were defined: a) different
97
bone cements and b) cement apposition timing. For line a) a total of 10 groups (in each configuration n = 5;
98
as proposed by the FDA guidance document for quasi-static testing on orthopaedic implants) was used:
99
Vega PS cobalt-chromium tibia and Vega PS ZrN tibia with the five bone cements BonOs R (aap,
100
Germany), Smart Set HV (DePuy Synthes, UK), Cobalt HV (DJO Surgical, USA), Palacos R (Heraeus
101
Medical, Germany), Surgical Simplex P (Stryker, USA), respectively, using mid-term cement apposition. For
102
line b) a total of 12 groups (n = 5 each) was used: Vega PS cobalt-chromium tibia and Vega PS ZrN tibia
103
with two bone cements BonOs R (aap, Germany) and Smart Set HV (DePuy Synthes, UK) in three different
104
cement apposition times, early, mid-term and late usage, respectively. An additional group (n = 5) of the PFC
105
Sigma FB tibia tray (size 2) was integrated with Smart Set HV using the mid-term processing window for
106
cement apposition and component implantation. To enable a direct comparison to the smallest Vega size
107
T0 tibial tray under the same cementing conditions, a PFC Sigma FB tray size 1.5 – which has similar
108
antero-posterior and medio-lateral dimensions - has also been tested. We have chosen the PFC Sigma FB
109
tibial tray as a worldwide established predicate device (since 1996) with clinically promising long-term
110
behaviour, which has been well-documented and reviewed by Hopley & Dalury [35] based on registry data (n
111
= 233,843 TKA’s) and 19 survivorship related, peer-reviewed publications (n = 4,025 TKA’s) for a direct
112
comparison with the Vega PS tibial tray. For all tibial trays measurement of the surface roughness (Rmax,
113
Ra, Rz) according to DIN EN ISO 4287:1997 and DIN EN ISO 4288:1998 was conducted in the cement
114
pockets.
115
The foam blocks were machined to provide a cavity, representing the contour of the osteodenser, for the
116
distal part of the Vega tibial tray (size 0) with a gap of at least 1 mm (Fig.1). For the cementation procedure
117
it was ensured that implants, cement and the processing materials (spatula, bowl etc.) were stored in the
118
airconditioned lab (19°C) for at least 24h prior to testing. The bone cement was manually mixed without
119
vacuum application according to the instructions for use taking into account the corresponding timing of
120
apposition for each cement at lab temperature (Table 1). One cement mixture was shared for the CoCrMo
121
and ZrN tibial tray to provide a paired comparison. When the cement reached the defined apposition time,
122
both tibial plateus were cemented (single layer) and inserted in a parallel fashion in the foam blocks followed
123
by a controlled pressurization of the cemented plateaus at 100 N and removal of the surplus bone cement.
124
After the cementation procedure, the specimens were stored according to ASTM F2118-14 for 7 days upside
®
®
®
®
®
®
®
®
®
®
®
®
®
®
®
4
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
125
down in a water bath at room temperature to ensure a defined hardening status and possible fluid immersion
126
(Fig. 2 - 5).
127
To test the fixation strength of the cemented tibiae a push-out test was performed for all specimens. The
128
load was applied displacement controlled (5 mm/min) with a push rod on the tibia keel using a quasi static
129
testing machine (Zwick Z010, Zwick/Roell, Germany). To support the foam blocks, a test frame with a stencil
130
for the sample was used (Fig. 6). From the load-/ displacement data the minimum failure load for each
131
specimen was analyzed.
132 133
For statistical analysis, an analysis of variance (ANOVA) was carried out (Statistica R13, Tibco, USA). To
134
determine differences regarding the parameters surface roughness (Rmax, Ra, Rz) and failure load of the
135
CoCrMo and ZrN tibial trays for the different used cements and cement apposition times, a post-hoc test
136
(Scheffe) was used. Prior to analysis the normal distribution (p-p plots) of the data and the homogeneity of
137
variance (Levene test) was proven. The level of significance was p = 0.05 for all analyses.
138
IV. Results
139 140
For the Vega PS cobalt-chromium tibial tray tested with five different bone cements, a mean push-out force
141
in a range of 2943 N to 3302 N and for the Vega PS ZrN multi-layer tibial tray of 2705 N to 3107 N was
142
measured showing a combined failure mode at the implant-cement and the cement-foam interface, whereby
143
no significant differences could be found regarding the used bone cements (p = 0.972 to p = 0.999) and
144
CoCrMo vs. ZrN tibia trays (p = 0.845 to p = 0.999) (Fig. 7 & 8). In contrast, the push-out force for the PFC
145
Sigma FB (size 1.5, n=1) was substantially lower with 1494 N, with a bonding failure at the implant-cement
146
interface as reported by Schlegel et al. [21]. The group of PFC Sigma FB (size 2, n=5) tested with Smart
147
Set HV, also showed substantially lower results of 1840 N ± 462 N in a range of 1377 N to 2452 N and a
148
characteristical failure mode at the implant-cement interface (Fig. 9).
149
In regards to cement apposition time two bone cements, BonOs R and SmartSet HV, were investigated.
150
The results showed a trend of decrease in failure load with increasing apposition time for both bone
151
cements. However, this was statistically not significant (p > 0.05). Additionally, no statistical difference for
152
CoCrMo and ZrN trays was found for the different bone cement apposition times (p = 0.104 to p = 0.995).
153
Only with Smart Set HV, the failure load for ZrN tibial tray decreased significantly (p = 0.003) between early
154
usage of the cement compared to end of processing window (Fig. 10 and Fig. 11).
®
®
®
®
5
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
155
From the surface roughness measurements, no statistical difference could be found for all analysed
156
parameters (Rmax, Ra and Rz) for the Vega tibial trays in CoCrMo (34.5, 3.88 and 29.2) and additional ZrN
157
multi-layer coating (33.7, 3.61 and 27.4) (p = 0.46 to p = 0.937), whereas the surface roughness values of
158
the PFC Sigma FB trays (5.49, 0.65 and 4.42) were on a significantly lower level (p < 0.001) (Fig. 12).
®
159 160 161
V. Discussion
162
The objectives of this study were to (1) evaluate if there is a potential difference of cemented implant fixation
163
strength between tibial components made out of cobalt-chromium and of a ceramic zirconium nitride (ZrN)
164
multi-layer coating and to (2) test their behaviour with five different bone cements in a standardised in vitro
165
model for testing of the implant-cement-bone interface conditions. We also analyzed (3) wether initial fixation
166
strength is a function of timing of the cement apposition and component implantation by an early, mid-term
167
and late usage within the cement specific processing window as reported by the manufacturer.
168
A limitation of our study may arise by the fact that uni-axial push-out does not reflect physiologic knee joint
169
loading conditions in vivo, where the tibial plateau is predominantly subjected to combined compression and
170
shear forces in a cyclic profile [14,36,37,38,39,40]. By applying dynamic compression-shear loading
171
conditions in a previous study [14] to evaluate the primary stability of Vega PS ZrN multi-layer coated TKA
172
tibial trays in 24 human tibiae, surface cementation (SC) with a keel length of 28 mm and 40 mm versus full
173
cementation (FC) was compared. No significant difference for the dynamic failure load between TibiaSC28
174
(4560 N), TibiaSC40 (4700 N) and TibiaFC40 (4920 N) as well as for the xyz-displacements of the tibial tray
175
relative to the bone was found, indicating that a keel length of 28 mm in surface cementation is able to create
176
a sufficient primary stability of the tibial plateau. In addition, the observed failure mode in all tested specimen
177
was migration into the metaphyseal head of the human tibiae, mechanically compromising the cement bone
178
interface under these highly demanding dynamic loading conditions [14], and not debonding between implant
179
and cement [22,23,41].
180
However, the intention of the current study was to establish a standardised in vitro model for testing of the
181
implant-cement-bone interface conditions, which reflects the characteristical failure modes as described by
182
Gebert de Uhlenbrock et al. [26] for their post mortem retrieval analysis on 22 bicompartmental tibial
183
plateaus from 17 patients. By applying quasi-static axial tension as a measure of the interfaces’s mechanical
184
capacity, Gebert de Uhlenbrock et al. [26] examined the influence of the time in situ on fixation strength.
185
Their specimens were retrieved after 5.3 years (range 0 to 11 years) in situ (mean human donor age 80.1
186
years; mean BMI 30.5 kg/m ) with pull-out forces ranging from 2751 N after only 2 days down to 231 N after
®
2
6
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
187
9 years in service and by trend, a decrease of fixation strength with the time in situ was found. To quantify
188
the failure mode in terms of the proportion of the tibial tray fixation failing at either the implant-cement or
189
cement-bone interface, Gebert de Uhlenbrock et al. [26] introduced a scoring system wereby the distal
190
surface of the tibial tray was divided into six regions. They reported that 29.4 % of the post-mortem retrieved
191
specimens failed exclusively at the implant-cement interface, 5.9 % failed exclusively at the cement-bone
192
interface and 64.7 % failed in a mixed mode [26]. The synthetic polyurethane foam model chosen for our
193
study mimics a failure mode at the implant-cement interface, as well as at the cement-synthetic bone
194
interface.
195
Comparing the cemented tibial tray fixation strength of components made out of cobalt-chromium and in a
196
variant covered by an additional zirconium nitride (ZrN) multi-layer coating, for the bone cements Surgical
197
Simplex P, Cobalt , Palacos R, Smart Set HV and BonOs HV, no statistically significant difference in the
198
five different cement groups was found for the two tibial tray materials. In the current study, push-out forces
199
above 2500 N were obtained in each of the ten test series with the cobalt-chromium and with the ZrN multi-
200
layer tibial trays showing substantial superiority (p < 0.01) in relation to the reported 1220 N of the PFC
201
Sigma FB tray – a clinically long-term successful implant [21,35]. On twelve paired, proximal thirds of human
202
donor tibiae Schlegel et al. [21] performed a similar pull-out test to determine the fixation strength of PFC
203
Sigma FB trays, using a similar surface cementing technique with Smart Set HV bone cement. In all pulsed
204
lavage tibial bone preparations (n = 6) failure was induced at the implant-cement interface with a mean pull-
205
out force of 1220 N (range 864 N to 1391 N), allowing for a direct comparison to our current results obtained
206
in a synthetic polyurethane foam model according to ASTM F1839-08. These results from Schlegel et al. [21]
207
fit well within the corridor of the current measurements on the PFC Sigma tibial tray and therefore they are
208
suitable to build a bridge from the current testing to a clinically long-term established predicate device.
209
Due to the fact that PMMA bone cement is not adhesive in interaction with metallic or ceramic implants and
210
the exact mechanism for implant fixation is a form-locking connection between the macro- and micro-
211
structured surface of the tibial tray and bone cement, the dominant factor influencing the implant-cement
212
interface bonding is the surface roughness [42]. From a biomechanical point of view the similar fixation
213
behaviour of the cobalt-chromium and the ZrN multi-layer coated tibial trays is reasonable, because of their
214
equivalent F22 corundium blast surface texture and roughness (Rz 25-35; Ra 3.5-4.5), whereas the PFC
215
Sigma FB has a visible and measureable smoother surface (Rz 5-7; Ra 0.5-1). These results on the PFC
216
Sigma FB tray fit very well into the corridor of roughness values (Ra 0.3-0.74), measured for PFC Sigma FB
217
tibial implants (n = 12 Ti6Al4V; n = 8 CoCrMo) in a comparative retrieval study by Cerquiglini et al. [43].
®
®
®
®
®
®
7
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1 ®
218
Performing the tests with five different bone cements for the Vega PS cobalt-chromium tibia tray a mean
219
push-out force in a range of 2943 N to 3302 N and for the Vega PS ZrN multi-layer tibia tray of 2705 N to
220
3107 N was measured. Between type of bone cement (Surgical Simplex P, Cobalt , Palacos R, Smart
221
Set HV and BonOs HV) for Vega PS cobalt-chromium & ZrN multi-layer no substantial differences were
222
found. These findings are in good accordance with the literature [42], where during characterization of
223
mechanical properties under quasi-static loading conditions typically no substantial differences are reported
224
between PMMA bone cements for 4-point-bending, compressive (ISO 5833:2002(E)) and tensile strength
225
(ISO 527-1:2012(E)). Under the aspect of clinical longterm behaviour, the fatigue properties of the bone
226
cements are of particular significance. Under flexural fatigue testing (ISO 16402:2008(E)) of bone cements,
227
e.g. Palacos R with a remaining fatigue strength of 16.6 MPa at 10 cycles shows a significantly better
228
fatigue performance than Surgical Simplex P (12.9 MPa at 10 cycles), whereas under quasi-static test
229
conditions both cements are in a range around 60 MPa [42,48]. The leading factor hereto is the molecular
230
weight of the polymer powder and the cured bone cement, whereby sterilization by ethylene oxide (EO) has
231
no influence on the molecular weight of the cement powder, but γ-irradiation significantly reduces it [42]. EO-
232
sterilised bone cements such as Cobalt , Palacos R, Smart Set HV and BonOs HV have a higher molecular
233
weight between 650,000 and 700,000 Dalton and subsequently a better fatigue performance, than bone
234
cements sterilised by γ-irradiation like Surgical Simplex P with a lower molecular weight between 250,000
235
and 300,000 Dalton [42,44].
236
Billi et al. [34] analyzed techniques for improving the initial strength of the tibial tray-cement interface bonding
237
and the effect of eight variables. Using a Triathlon tibial tray in surface cementation technique and an acrylic
238
material test block with high chemical affinity to bone cement, they found that late cementing reduced the
239
mean interface strength by 47% for Surgical Simplex P and by 73% for Palacos R compared to the normal
240
timing. Moreover, early cementing increased the mean interface strength by 48% for Surgical Simplex P and
241
by 72% for Palacos R versus cementing under normal conditions. In the present study, the influence of the
242
timing for cement apposition to the tibial tray was lower. For BonOs HV by trend an increase in fixation
243
strength by 2.3% for cobalt-chromium and by 6.3% for ZrN multi-layer was measured for the early
244
cementation in comparison to the mid of processing window, whereas for the late usage a decrease by
245
11.9% for cobalt-chromium and by 16.6% for ZrN multi-layer was found. For Smart Set HV, the early to mid
246
cementation increased the failure load by 12.9% for cobalt-chromium and by 16.3% for ZrN multi-layer,
247
whereas late usage decreased it by 4.3% for cobalt-chromium and by 23.1% for ZrN multi-layer, but without
248
statistically significant differences.
®
®
®
®
®
®
®
®
7
®
®
7
®
®
®
®
®
®
®
®
®
®
8
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
249
Possible reasons for the present different findings compared to Billi et al. [34] may be the different design,
250
surface texture and roughness of the Vega tibial tray, the storage for 7 days in water ensuring a defined
251
hardening status (ISO 5833) and possible fluid immersion versus 48 hours under dry conditions [34] and the
252
introduction of a synthetic polyurethane foam model allowing for clinically relevant fixation failure modes on
253
both interfaces [26].
®
254 ®
255
Examining the clinical behaviour of uncoated and ZrN multi-layer coated e.motion UC knee implants, a
256
mobile bearing, rotating platform design, Thomas et al. [45] performed a retrospective multi-center study on
257
196 TKA patients with an average follow-up of 5.7 years. In three participating centers they examined all
258
responding TKA patients, who received an e.motion UC in 2007 in uncoated (n = 99; age 67.0, BMI 29.8) or
259
a ZrN multi-layer coated version (n = 97; age 69.3, BMI 27.0) within a 5-year follow-up, including detailed
260
physical examination, knee pain & function (KSS), x-ray documentation (leg axis, positioning, radiolucencies)
261
and survival via Kaplan-Meier analysis. Thomas et al. [45] reported a favourable survivorship of 97 % for
262
uncoated and of 98 % for ZrN multi-layer coated TKA’s after 5.5 years, without significant difference between
263
the groups.
264
In a prospective randomized clinical trial on 120 primary TKA patients treated with a Columbus DD fixed
265
bearing knee implant either in cobalt-chromium uncoated (n = 59; age 68.6, BMI 30.5) or in ZrN multi-layer
266
coating (n = 61; age 66.6, BMI 31.3), Beyer et al. [46] published midterm clinical results with 5 year follow up.
267
Patient related outcomes (PROMS) were significantly improved for the uncoated as well as for the ZrN multi-
268
layer coated implant cohort. The Oxford Knee Score (OKS) improved substantially in both, the uncoated
269
group from 21.9 points (SD 7.6) preoperatively to 39.2 points (SD 7.9) four years after surgery and the ZrN
270
multi-layer coated group from 21.6 points (SD 6.2) to 39.5 points (SD 7.8), respectively [46]. Beyer et al. [46]
271
reported an excellent survival in both groups, with a 5-year survival rate of 98.1 % in the uncoated and of
272
100 % in the ZrN multi-layer coated TKA group.
273
For the ZrN multi-layer coated Vega PS knee system, Lionberger et al. [47] reported in a retrospective study
274
on a single surgeon case series an unacceptable failure rate of 6% prior to 1 year follow-up (15 aseptic
275
loosenings out of a cohort of 249 TKA’s implanted from 2015 to 2017) in comparison to a previously
276
implanted cohort of 850 Columbus cruciate retaining knees (2009 to 2014) with only two known revisions
277
due to aseptic loosening (0.24%). They hypothesized that the new design of the Vega PS tibial tray may be
278
a factor to be considered for their unfavourable clinical outcomes [47]. These findings are in contrast to the
279
promising short-term clinical behaviour of the Vega PS knee system reported by Jain et al. [48]. In three
®
®
®
®
®
9
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1 ®
280
cohorts they compared the clinical outcomes of 206 consecutive TKA’s using Vega PS with those of two
281
clinically long-term established posterior stabilized designs (e.motion PS n = 205; Genesis II n = 216) in a
282
two year follow-up study and found comparable or superior functional clinical performance of Vega PS
283
without incidence of implant-related adverse events [48]. Lionberger et al. [47] further hypothesized that the
284
ceramic hardened implant surface may be a potential factor for limited cement adhesion and early debonding
285
in the later loosenings and proposed further testing to ascertain the root cause for these failures, but his
286
cementation technique lists he was using Palacos R and cementing in the late window we investigated in
287
this study and he also lists two other cementation techniques that are specifically contrary to manufacturer
288
recommendations. The current study, including the results of this detailed evaluation and testing, showed a
289
high interface fixation strength of the cemented Vega PS tibial plateau without significant differences
290
between the cobalt-chromium and the ZrN multi-layer surface version.
®
®
®
®
291 292
Evaluating the clinical outcomes of primary TKA in the National Joint Registry of England, Wales, Northern
293
Ireland & Isle of Man with an observation time of 5 and 7 years, Porter et al. [5] reported for the uncoated
294
Columbus knee in cobalt-chromium (n = 10,659) a survivorship of 97.6 % at 5 years and of 97.2 % at 7
295
years which is comparable to ALL TKA (n = 977,488) with 97.8 % at 5 years and 97.3 % at 7 years. For the
296
ZrN multi-layer coated AS Columbus knee (n = 1,067) a survival rate of 98.6 % at 5 years and of 98.1 % at
297
7 years was calculated, which is significantly superior to the Columbus uncoated and to ALL TKA in the
298
National Joint Registry [5].
®
®
®
299
300
VI. Conclusion
301
From the observations reported in this study, it is concluded that there is no significant difference in implant-
302
cement-bone fixation strength between cobalt-chromium and ZrN multi-layer coated Vega tibial trays tested
303
with five different commonly used bone cements. When differences between types of bone cement (Surgical
304
Simplex P, Cobalt , Palacos R, Smart Set HV and BonOs HV) were investigated no substantial differences
305
were measured for either cobalt-chromium or ZrN multi-layer implant surface groups. Apposition of bone
306
cements and tibial tray implantation in the early to mid of the cement specific processing window is beneficial
307
in regard to interface fixation in TKA.
®
®
®
®
®
®
308 309
10
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
311
Acknowledgements: None Funding: None
312
Ethical approval: Not required
310
313 314
Tables
315 316 317 318
Table 1: Cement apposition and component implantation times for early, mid-term and late usage within the cement specific processing window at 19 °C early usage [min’ sec’’]
mid-term usage [min’ sec’’]
late usage [min’ sec’’]
intended
mean ± Std.
intended
mean ± Std.
intended
mean ± Std.
Surgical Simplex®P
---
---
7' 30''
7' 29'' ± 0' 05''
---
---
Cobalt®
---
---
6' 00''
5' 54'' ± 0' 10''
---
---
Palacos®R
---
---
4' 30''
4' 28'' ± 0' 15''
---
---
Smart Set®HV
1' 30''
1' 33'' ± 0' 10''
6' 30''
6' 23'' ± 0' 10''
9' 30''
9' 30'' ± 0' 01''
BonOs®HV
2' 00''
2' 12'' ± 0' 20''
6' 00''
6' 05'' ± 0' 18''
8' 30''
8' 30'' ± 0' 01''
319 320 321 322
Figures
323 324 325 326
Fig. 1: Polyurethane foam blocks (20 pcf cellular rigid foam, Sawbones, Sweden), representing the contour ® of the osteodenser, for the distal part of the Vega tibial tray (size 0) with a gap of at least 1 mm
327 328 329
Fig.2: Paired cementation of the CoCrMo and ZrN tibial trays
11
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
330 331 332
Fig. 3: Parallel cement application and insertion of both tibial plateaus
333 334 335
Fig. 4: Controlled pressurization and removal of surplus bone cement
336 337 338 339
Fig.5: Controlled hardening of both plateaus
12
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
test machine frame rod for load application Sawbone polyurethan boam block with sample in upside down position test frame
stencil for the sample size sample (tibia)
340 341 342 343 344
345 346 347 348
Fig. 6: Push-out testing after 7 day storage of the specimens in water for defined hardening status and possible fluid immersion.
Fig. 7: Failure load of the tested groups with 5 different bone cements (mid-term cement apposition) in comparison to Schlegel et al. [21] (green doted line)
13
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
349 350
351 352
353 354 355 356 357 358
®
Fig. 8: Typical failure characteristics for the Vega PS cobalt-chromium (right marked with “C”) and ZrN multi-layer (left marked with “A”) tibial trays with a combination of failure at the implant-cement and at the cement-foam interface (mixed mode acc. to Gebert de Uhlenbrock et al. [26]) tested with 5 different bone cements
14
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
359 360 361 362 363 364 365
®
Fig. 9: Typical failure characteristics for the PFC Sigma FB tray tested with Smart Set HV with a predominant failure at the implant-cement interface as also reported by Schlegel et al. [21]
366 367 368 369 370
Fig. 10: Failure load of the tested groups with BonOs R for different cement apposition times in comparison to Schlegel et al. [21] (green doted line)
15
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
371 372 373 374 375 376
®
Fig. 11: Failure load of the tested groups with Smart Set HV for different cement apposition times in comparison to Schlegel et al. [21] (green doted line)
377
378 379 380
Fig. 12: Surface roughness Rmax, Ra and Rz of all tested specimens
381 382
16
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436
Literature 1 Kurtz SM, Ong KL, Lau E, Widmer M, Maravic M, Gómez-Barrena E, de Fátima de Pina M, Manno V, Torre M, Walter WL, de Steiger R, Geesink RGT, Peltola M, Röder C: International survey of primary and revision total knee replacement. Int Orthopaedics (SICOT) 2011;35(2):1783-9. 2 Graves S, de Steiger R, Lewis P, Harris I: Australian Orthopaedic Association National Joint Replacement Registry for Hip, Knee & Shoulder Arthroplasty. Annual Report 2018:235-63. 3 Carr AJ, Robertsson O, Graves S, Price AJ, Arden NK, Judge A, Beard DJ: Knee Replacement. The Lancet 2012;379:1331-1340. 4 Evans JT, Walker RW, Evans JP, Blom AW, Sayers A, Whitehouse MR: How long does a knee replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet 2019; 393: 655–63. 5 Porter M, Howard P, Lawrence S, Reed M, Stonadge J, Wilkinson M: National Joint Registry England, Wales, Northern Ireland & Isle of Man. Annual Report 2018:102-51. 6 Grimberg A, Jansson V, Liebs T, Melsheimer O, Steinbrueck A: Endoprothesenregister Deutschland (EPRD) Jahrensbericht 2017:33-42. 7 Gothesen O, Espehaug B, Havelin L, Petursson G, Lygre S, Ellison P, Hallan G, Furnes O: Survival rates and causes of revision in cemented primary total knee replacement – Norwegian Arthroplasty Register 19942009. Bone Joint J 2013;95B:636-42. 8 Dyrhovden GS, Lygre SHL, Badawy M, Gothesen O, Furnes O: Have the causes of revision for total and unicompartmental knee arthroplasties changed during the past two decades? Clin Orth Rel Res 2017;475(7):1874-87. 9 Sadoghi P, Liebensteiner M, Agreiter M, Leithner A, Böhler N, Labek G: Revision surgery after total joint arthroplasty: a complication-based analysis using worldwide arthroplasty registers. J Arthroplasty 2013;28:1329-1332. 10 Niinimaeki TT: The reasons for knee arthroplasty revisions are incomparable in the different arthroplasty registries. Knee 2015;22:142-4. 11 Lum ZC, Shieh AK, Dorr LD. Why total knees fail-A modern perspective review. World J Orthop 2018; 9(4):60-4. 12 Delanois RE, Mistry JB, Gwam CU, Mohamed NS, Choksi US, Mont MA: Current epidemiology of revision total knee arthroplasty in the United States. J Arthroplasty 2017;32:2663-8. 13 Lombardi Jr AV, Berend KR, Adams JB: The revision knee arthroplasty – Why knee replacements fail in 2013 patient, surgeon or implant? J Bone Joint Surg 2014;96B:101-4. 14 Grupp TM, Saleh KJ, Holderied M, Pfaff A, Schilling C, Schroeder C, Mihalko WM. Primary stability of tibial plateaus under dynamic compression-shear loading in human tibiae – influence of keel length, cementation area and tibial stem. J Biomechanics 2017;59:9-22. 15 Cawley DT, Kelly N, Simpkin A, Shannon FJ, McGarry JP: Full and surface tibial cementation in total knee arthroplasty – a biomechanical investigation of stress distribution and remodeling in the tibia. Clinical Biomechanics 2012;27:390-7. 16 Cawley DT, Kelly N, McGarry JP, Shannon FJ: Cementing techniques for the tibial component in primary total knee replacement. J Bone and Joint Surg 2013;95B(3):295-300. 17 Schlegel UJ, Bruckner T, Schneider M, Parsch D, Geiger F, Breusch SJ: Surface or full cementation of the tibial component in total knee arthroplasty: a matched-pair analysis of mid- to long-term results. Arch Orthop Trauma Surg 2015;135(5):703-708. 18 Sundberg M, Lidgren L, W-Dahl A, Robertsson O: Swedish knee arthroplasty register – Annual Report Lund University, Lund Sweden, 2014. 19 Gothesen O, Espehaug B, Havelin L, Petursson G, Lygre S, Ellison P, Hallan G, Furnes O: Survival rates and causes of revision in cemented primary total knee replacement – Norwegian Arthroplasty Register 19942009. Bone Joint J 2013;95B:636-42. 20 Furnes O, Espehaug B, Lie SA, Vollset SE, Engesaeter LB, Havelin L: Failure mechanism after unicompartmental and tricompartmental primary knee replacement with cement. J Bone and Joint Surg 2007;89A(3):519-25. 21 Schlegel UJ, Siewe J, Delank KS, Eysel P, Püschel K, Morlock MM, Gebert de Uhlenbrock A: Pulsed lavage improves fixation strength of cemented tibial components. International Orthopaedics (SICOT) 2011;35:1165-9. 17
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
22 Ries C, Heinichen M, Dietrich F, Jakubowitz E, Sobau C, Heisel C: Short-keeled cemented tibial components show an increased risk for aseptic loosening. Clin Orthop Relat Res 2013;471:1008-13. 23 Arsoy D, Pagnano MW, Lewallen DG, Hanssen AD, Sierra RJ: Aseptic tibial debonding as a cause of early failure in a modern total knee arthroplasty design. Clin Orthop Relat Res 2013;471:94-101. 24 Galasso O, Jenny JY, Saragaglia D, Miehlke RK: Full versus surface tibial baseplate cementation in total knee arthroplasty. Orthopedics 2013;36(2):e151-8. 25 Clarius M, Hauck C, Seeger JB, James A, Murray DW, Aldinger PR: Pulse lavage reduces the incidence of radiolucent lines under the tibial tray of Oxford unicompartmental knee arthroplasty. International Orthopaedics (SICOT) 2009;33:1585-90. 26 Gebert de Uhlenbrock A, Püschel V, Püschel K, Morlock MM, Bishop NE: Influence of time in-situ and implant type on fixation strength of cemented tibial trays – A post mortem retrieval analysis. Clinical Biomechanics 2012;27(9):929-935. 27 Staats K, Wannmacher T, Weihs V, Koller U, Kubista B, Windhager R: Modern cemented total knee arthroplasty design shows higher incidence of radiolucent lines compared to its predecessor. Knee Surgery, Sports Traumatology, Arthroscopy 2019;27:1148-1155. 28 Maistrelli GL, Antonelli L, Fornasier V, Mahomed N: Cement penetration with pulsed lavage versus syringe irrigation in total knee arthroplasty. Clin Orthop Rel Res 1995;312:261-65. 29 Clarius M, Seeger JB, Jaeger S, Mohr G, Bitsch RG: The importance of pulsed lavage on interface temperature and ligament tension force in cemented unicompartmental knee arthroplasty. Clinical Biomechanics 2012;27:372-76. 30 Jaeger S, Rieger JS, Bruckner T, Kretzer JP, Clarius M, Bitsch RG: The protective effect of pulsed lavage against implant subsidence and micromotion for cemented tibial unicompartmental knee components – an experimental cadaver study. J Arthroplasty 2014;29(4):727-32. 31 Completo A, Simoes JA, Fonseca F, Oliveira M: The influence of different tibial stem designs in load sharing and stability at the cement-bone interface in revision TKA. Knee 2008;15:227-32. 32 Kelly N: An experimental and computational investigation of the inelastic behaviour of trabecular bone. The National University of Ireland, Galway. Doctoral Thesis 2012:93-101. 33 Grupp TM, Pietschmann MF, Holderied M, Scheele C, Schröder C, Jansson V, Müller PE: Primary stability of unicompartmental knee arthroplasty under dynamic compression-shear loading in human tibiae. Clinical Biomechanics 2013;28:1006-13. 34 Billi F, Kavanaugh A, Schmalzried H, Schmalzried T: Techniques for improving the initial strength of the tibial tray-cement interface bond. Bone Joint J 2019;101B(1 Supple A):53-8. 35 Hopley CDJ, Dalury DF: A systematic review of clinical outcomes and survivorship after total knee arthroplasty with a contemporary modular knee system. J Arthroplasty 2014;29:1398-411. 36 Taylor WR, Heller MO, Bergmann G, Duda GN: Tibio-femoral loading during human gait and stair climbing. J Orthopaedic Research 2004;22:625-32. 37 Zhao D, Banks SA, D'Lima DD, Colwell CW, Fregly BJ: In vivo medial and lateral tibial loads during dynamic and high flexion activities. J Orthopaedic Research 2007;25(5):593-602. 38 Muendermann A, Dyrby CO, d'Lima DD, Colwell CW, Andriacchi TP: In vivo knee loading characteristics during activities of daily living as measured by an instrumented total knee replacement. J Orthopaedic Research 2008;26(9):1167-72. 39 Kutzner I, Heinlein B, Graichen F, Bender A, Rohlmann A, Halder A, et al.: Loading of the knee joint during activities of daily living measured in vivo in five subjects. J Biomechanics 2010;43:2164-73. 40 Bergmann G, Bender A, Graichen F, Dymke J, Rohlmann A, Trepczynski A, Heller MO, Kutzner I: Standardized Loads Acting in Knee Implants. PLoS ONE 2014;9(1) e86035.doi:10.1371/ journal.pone.0086035. 41 Foran JR, Whited BW, Sporer SM: Early aseptic loosening with a precoated low-profile tibial component a case series. J Arthroplasty 2011;26:1445-50. 42 Kuehn KD: PMMA cements – Are we aware what we are using? Springer Berlin Heidelberg 2014:1-14 & 183-230 DOI10.1007/978-3-642-41535-7.
18
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503
43 Cerquiglini A, Henckel J, Hothi H, Allen P, Lewis J, Eskelinen A, Skinner J, Hirschmann MT, Hart AJ: Analysis of the Attune tibial tray backside – a comparative retrieval study. Bone Joint Research 2019;8:136145. 44 Tepic S, Soltész U: Fatigue strength of bone cement with simulated stem interface porosity. J Mat Sciences Mater Med 1998;9:707-9. 45 Thomas P, Hisgen P, Kiefer H, Schmerwitz U, Ottersbach A, Albrecht D, Schinkel C: Blood cytokine pattern and clinical outcome in knee arthroplasty patients: comparative analysis 5 years after standard versus “hypoallergenic” surface coated prosthesis implantation. Acta Orthopaedica 2018;89 DOI:10.1080/17453674.2018.1518802 46 Beyer F, Luetzner C, Kirschner S, Luetzner J: Midterm results after coated and uncoated TKA: A randomized controlled study. Orthopedics 2016;39(3):S13-7. 47 Lionberger D, Conlon C, Wattenbarger L, Walker TJ: Unacceptable failure rate of a ceramic-coated posterior cruciate-substituting total knee arthroplasty. Arthroplasty Today 2019;5:1-6. 48 Jain NP, Lee SY, Morey VM, Chong S, Kang YG, Kim TK: Early clinical outcomes of a new posteriorly stabilized total knee arthroplasty prosthesis – comparison with two established prostheses. Knee Surg Relat Res 2017;29(3):180-8.
19
S0883-5403(19)30868-X
DOI:
https://doi.org/10.1016/j.arth.2019.09.019
Reference:
YARTH 57523
To appear in:
The Journal of Arthroplasty
Received Date: 4 July 2019 Revised Date:
15 August 2019
Accepted Date: 13 September 2019
Please cite this article as: Grupp TM, Schilling C, Schwiesau J, Pfaff A, Altermann B, Mihalko WM, Tibial implant fixation behaviour in total knee arthroplasty – a study with five different bone cements, The Journal of Arthroplasty (2019), doi: https://doi.org/10.1016/j.arth.2019.09.019. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 The Author(s). Published by Elsevier Inc.
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
1 2
Tibial implant fixation behaviour in total knee arthroplasty – a study with five different bone cements
3 4 5
Abstract
6
The objectives of this study were to (1) evaluate if there is a potential difference of cemented implant fixation
7
strength between tibial components made out of cobalt-chromium (CrCoMo) and of a ceramic zirconium
8
nitride (ZrN) multi-layer coating and to (2) test their behaviour with five different bone cements in a
9
standardised in vitro model for testing of the implant-cement-bone interface conditions. We also analyzed (3)
10
wether initial fixation strength is a function of timing of the cement apposition and component implantation by
11
an early, mid-term and late usage within the cement specific processing window.
12
An in vitro study using a synthetic polyurethane foam model was performed to investigate the implant fixation
13
strength after cementation of tibial components by a push-out test. A total of 20 groups (n = 5 each) was
14
used: Vega PS cobalt-chromium tibia and Vega PS ZrN tibia with the bone cements BonOs R, Smart
15
Set HV, Cobalt HV, Palacos R, Surgical Simplex P, respectively, using mid-term cement apposition.
16
Three different cement apposition times, early, mid-term and late usage were tested with a total of 12 groups
17
(n = 5 each) with the bone cements BonOs R and Smart Set HV.
18
From the observations, it is concluded that there is no significant difference in implant-cement-bone fixation
19
strength between cobalt-chromium and ZrN multi-layer coated Vega tibial trays tested with five different
20
commonly used bone cements. Apposition of bone cements and tibial tray implantation in the early to mid of
21
the cement specific processing window is beneficial in regard to interface fixation in TKA.
®
®
®
®
®
®
®
®
®
®
22 23 24
Keywords
25
cements, timing of the cement apposition
total knee arthroplasty, tibial implant fixation, ceramic zirconium nitride (ZrN) multi-layer coating, five bone
26
27
I. Introduction
28
Total knee arthroplasty (TKA) can be considered as a successful clinical standard of care for the treatment of
29
endstage degenerative osteoarthritis, rheumatoid or other inflammatory arthritis and osteonecrosis, with an
30
increasing demand over the past two decades [1,2,3,4]. For primary TKA the cumulative percentages of
31
revision for all reasons are comparably low with 1-3 % early failures within 1 to 3 years post-operatively and
32
with 4-8 % revisions over a period of 10 to 15 years [2,4,5,6,7,8]. Sadoghi et al. [9] performed a 1
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
33
complication-based analysis of TKA revisions including 391,913 primary and 36,307 revision cases in a 30
34
years period entered in the joint registries of Sweden, Norway, Finland, Denmark, Australia and New
35
Zealand and found the most common causes to be aseptic loosening in 29.8 %, septic loosening in 14.8 %,
36
pain in 9.5 %, wear in 8.2 % and instability in 6.2 % of the cases. Niinimaeki [10] analysed the reasons for
37
knee revision based on the registries of Norway, Sweden, Australia, New Zealand and England & Wales in
38
2015 and found that the leading indication for revision was aseptic loosening (range 22.8 to 29.7 %).
39
Infection was the second leading indication for revision in Sweden, Australia and England & Wales (20.6 to
40
21.7 %), while pain was the second most common cause in Norway and New Zealand (27.4 & 22.0 %) [11].
41
Delanois et al. [12] used the National Inpatient Sample (NIS) database and identified all revision TKA
42
procedures performed between 2009 and 2013 in the United States. Collecting clinical and demographic
43
data for 337,597 procedures, infection (20.8 %), mechanical loosening (20.3 %) and instability (7.5 %) were
44
the most common specified etiologies for revision TKA [12].
45
In a multi-centre study including five centers in the United States, Lombardi et al. [13] reviewed 844 TKA
46
revisions which were performed in a period from 2010 to 2011. Aseptic loosening (31.2 %) was the
47
predominant failure mode, followed by instability (18.7 %), infection (16.2 %) and polyethylene wear (10 %).
48
The mean time to revision was 5.9 years (from 10 days to 31 years), whereas 36 % of all revisions occurred
49
within two years after primary surgery, 25 % between 2 and 5 years and 29 % between 5 and 15 years. The
50
remaining 10 % of late revisions (>15 years service in vivo) were mainly related to polyethylene wear [13].
51
Analyzing the incidence of early revision diagnosis for primary TKA in the National joint registries of Australia
52
and England, Wales & Northern Ireland in two years since primary procedure, the predominant reason for
53
revision was infection (0.6 % AUS; 0.6 % UK), followed by loosening (0.4 % AUS; 0.5 % UK). At five years
54
the cumulative incidence of loosening (0.8 % AUS; 0.9 % UK) is in an equal level to infection (0.8 % AUS;
55
0.8 % UK), whereas for the mid- to long-term (between 5 and 16 years) loosening is the most common
56
reason for revision [2,5].
57
Apart from infection, instability and sub optimal alignment, aseptic tibial component loosening remains a
58
major cause of TKA failure [8,14,15,16,17]. Based on 3,572 primary TKA revisions in the Swedish knee
59
arthroplasty registry, Sundberg et al. [18] reported the implant removal of only the tibial component in 7.1 %
60
and only the femoral component in 0.9 % of the cases, during a 10-year period from 2003 to 2012. Gothesen
61
et al. [18] evaluated the aseptic implant revisions within a cohort of 17,782 primary cemented TKA’s reported
62
to the Norwegian arthroplasty register during the years from 1994 to 2009, and found a ratio of 2.8 between
63
tibial and femoral component loosening, whereas Furnes et al. [20] extracted a ratio of 3.7 for the similar
2
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
64
database up to 2004. To answer the question if the causes of revision for TKA have changed during the past
65
two decades in Norway, Dyrhovden et al. [8] selected two cohorts of primary TKA’s implanted during 1994 to
66
2004 (Period 1; n = 17,404) and during 2005 to 2015 (Period 2; n = 43,219) and found that the relative risk
67
for revision (RR) has been unchanged for the tibia (RR 1.0), but substantially decreased for the femur (RR
68
0.3). This emphasizes the need for stable primary and secondary fixation of tibial trays, which is dependent
69
on multiple factors including implant design, bone interface preparation, surgical technique and cementation
70
[14,17,21,22,23,24].
71
Several studies have been undertaken to analyze the primary and secondary stability of tibial trays in vitro, in
72
vivo, in silico and ex vivo [14,15,16,21,25,26,27]. Various examinations have been undergone, such as
73
cement penetration depth analysis in the proximal tibia [28,29,30], radiographic short-term outcome
74
measurements by incidence of radiolucent lines [27], finite element analysis (FEA) to assess resulting
75
interface bone stresses and strains [15,31,32], static tension to measure the implant-cement-bone interface
76
bonding [21,26] and dynamic compression-shear loading [14,33] until implant-bone interface failure. In
77
previous studies [14,15,16,21,34] the influences of implant design, bone interface preparation, full or surface
78
cementing technique on the primary stability of bicompartmental tibial trays have been examined.
79 80
II. Objectives
81
The objectives of this study were to (1) evaluate if there is a potential difference of cemented implant fixation
82
strength between tibial components made out of cobalt-chromium (CrCoMo) and of a ceramic zirconium
83
nitride (ZrN) multi-layer coating and to (2) test their behaviour with five different bone cements in a
84
standardised in vitro model for testing of the implant-cement-bone interface conditions. We also sought to
85
analyze (3) wether initial fixation strength is a function of timing of the cement apposition and component
86
implantation by an early, mid-term and late usage within the cement specific processing window as reported
87
by the manufacturer.
88 89
III. Materials & Methods
90
An in vitro study using a synthetic polyurethane foam model (20 pcf cellular rigid foam, Sawbones, Sweden)
91
according to ASTM F1839-08 was performed to investigate the implant fixation strength after cementation of
92
tibial components by a push-out test. Size T0 tibial trays from the Vega knee system (Aesculap, Germany)
93
were chosen as they are available in both cobalt-chromium (CoCrMo) and zirconium nitride (ZrN) multi-layer
®
3
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
94
surfaces, and as the smallest standard size have the least surface area for fixation. Five different bone
95
cements were investigated to cover a diverse range of common clinically used bone cements.
96
For the evaluation of the research questions two lines with corresponding groups were defined: a) different
97
bone cements and b) cement apposition timing. For line a) a total of 10 groups (in each configuration n = 5;
98
as proposed by the FDA guidance document for quasi-static testing on orthopaedic implants) was used:
99
Vega PS cobalt-chromium tibia and Vega PS ZrN tibia with the five bone cements BonOs R (aap,
100
Germany), Smart Set HV (DePuy Synthes, UK), Cobalt HV (DJO Surgical, USA), Palacos R (Heraeus
101
Medical, Germany), Surgical Simplex P (Stryker, USA), respectively, using mid-term cement apposition. For
102
line b) a total of 12 groups (n = 5 each) was used: Vega PS cobalt-chromium tibia and Vega PS ZrN tibia
103
with two bone cements BonOs R (aap, Germany) and Smart Set HV (DePuy Synthes, UK) in three different
104
cement apposition times, early, mid-term and late usage, respectively. An additional group (n = 5) of the PFC
105
Sigma FB tibia tray (size 2) was integrated with Smart Set HV using the mid-term processing window for
106
cement apposition and component implantation. To enable a direct comparison to the smallest Vega size
107
T0 tibial tray under the same cementing conditions, a PFC Sigma FB tray size 1.5 – which has similar
108
antero-posterior and medio-lateral dimensions - has also been tested. We have chosen the PFC Sigma FB
109
tibial tray as a worldwide established predicate device (since 1996) with clinically promising long-term
110
behaviour, which has been well-documented and reviewed by Hopley & Dalury [35] based on registry data (n
111
= 233,843 TKA’s) and 19 survivorship related, peer-reviewed publications (n = 4,025 TKA’s) for a direct
112
comparison with the Vega PS tibial tray. For all tibial trays measurement of the surface roughness (Rmax,
113
Ra, Rz) according to DIN EN ISO 4287:1997 and DIN EN ISO 4288:1998 was conducted in the cement
114
pockets.
115
The foam blocks were machined to provide a cavity, representing the contour of the osteodenser, for the
116
distal part of the Vega tibial tray (size 0) with a gap of at least 1 mm (Fig.1). For the cementation procedure
117
it was ensured that implants, cement and the processing materials (spatula, bowl etc.) were stored in the
118
airconditioned lab (19°C) for at least 24h prior to testing. The bone cement was manually mixed without
119
vacuum application according to the instructions for use taking into account the corresponding timing of
120
apposition for each cement at lab temperature (Table 1). One cement mixture was shared for the CoCrMo
121
and ZrN tibial tray to provide a paired comparison. When the cement reached the defined apposition time,
122
both tibial plateus were cemented (single layer) and inserted in a parallel fashion in the foam blocks followed
123
by a controlled pressurization of the cemented plateaus at 100 N and removal of the surplus bone cement.
124
After the cementation procedure, the specimens were stored according to ASTM F2118-14 for 7 days upside
®
®
®
®
®
®
®
®
®
®
®
®
®
®
®
4
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
125
down in a water bath at room temperature to ensure a defined hardening status and possible fluid immersion
126
(Fig. 2 - 5).
127
To test the fixation strength of the cemented tibiae a push-out test was performed for all specimens. The
128
load was applied displacement controlled (5 mm/min) with a push rod on the tibia keel using a quasi static
129
testing machine (Zwick Z010, Zwick/Roell, Germany). To support the foam blocks, a test frame with a stencil
130
for the sample was used (Fig. 6). From the load-/ displacement data the minimum failure load for each
131
specimen was analyzed.
132 133
For statistical analysis, an analysis of variance (ANOVA) was carried out (Statistica R13, Tibco, USA). To
134
determine differences regarding the parameters surface roughness (Rmax, Ra, Rz) and failure load of the
135
CoCrMo and ZrN tibial trays for the different used cements and cement apposition times, a post-hoc test
136
(Scheffe) was used. Prior to analysis the normal distribution (p-p plots) of the data and the homogeneity of
137
variance (Levene test) was proven. The level of significance was p = 0.05 for all analyses.
138
IV. Results
139 140
For the Vega PS cobalt-chromium tibial tray tested with five different bone cements, a mean push-out force
141
in a range of 2943 N to 3302 N and for the Vega PS ZrN multi-layer tibial tray of 2705 N to 3107 N was
142
measured showing a combined failure mode at the implant-cement and the cement-foam interface, whereby
143
no significant differences could be found regarding the used bone cements (p = 0.972 to p = 0.999) and
144
CoCrMo vs. ZrN tibia trays (p = 0.845 to p = 0.999) (Fig. 7 & 8). In contrast, the push-out force for the PFC
145
Sigma FB (size 1.5, n=1) was substantially lower with 1494 N, with a bonding failure at the implant-cement
146
interface as reported by Schlegel et al. [21]. The group of PFC Sigma FB (size 2, n=5) tested with Smart
147
Set HV, also showed substantially lower results of 1840 N ± 462 N in a range of 1377 N to 2452 N and a
148
characteristical failure mode at the implant-cement interface (Fig. 9).
149
In regards to cement apposition time two bone cements, BonOs R and SmartSet HV, were investigated.
150
The results showed a trend of decrease in failure load with increasing apposition time for both bone
151
cements. However, this was statistically not significant (p > 0.05). Additionally, no statistical difference for
152
CoCrMo and ZrN trays was found for the different bone cement apposition times (p = 0.104 to p = 0.995).
153
Only with Smart Set HV, the failure load for ZrN tibial tray decreased significantly (p = 0.003) between early
154
usage of the cement compared to end of processing window (Fig. 10 and Fig. 11).
®
®
®
®
5
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
155
From the surface roughness measurements, no statistical difference could be found for all analysed
156
parameters (Rmax, Ra and Rz) for the Vega tibial trays in CoCrMo (34.5, 3.88 and 29.2) and additional ZrN
157
multi-layer coating (33.7, 3.61 and 27.4) (p = 0.46 to p = 0.937), whereas the surface roughness values of
158
the PFC Sigma FB trays (5.49, 0.65 and 4.42) were on a significantly lower level (p < 0.001) (Fig. 12).
®
159 160 161
V. Discussion
162
The objectives of this study were to (1) evaluate if there is a potential difference of cemented implant fixation
163
strength between tibial components made out of cobalt-chromium and of a ceramic zirconium nitride (ZrN)
164
multi-layer coating and to (2) test their behaviour with five different bone cements in a standardised in vitro
165
model for testing of the implant-cement-bone interface conditions. We also analyzed (3) wether initial fixation
166
strength is a function of timing of the cement apposition and component implantation by an early, mid-term
167
and late usage within the cement specific processing window as reported by the manufacturer.
168
A limitation of our study may arise by the fact that uni-axial push-out does not reflect physiologic knee joint
169
loading conditions in vivo, where the tibial plateau is predominantly subjected to combined compression and
170
shear forces in a cyclic profile [14,36,37,38,39,40]. By applying dynamic compression-shear loading
171
conditions in a previous study [14] to evaluate the primary stability of Vega PS ZrN multi-layer coated TKA
172
tibial trays in 24 human tibiae, surface cementation (SC) with a keel length of 28 mm and 40 mm versus full
173
cementation (FC) was compared. No significant difference for the dynamic failure load between TibiaSC28
174
(4560 N), TibiaSC40 (4700 N) and TibiaFC40 (4920 N) as well as for the xyz-displacements of the tibial tray
175
relative to the bone was found, indicating that a keel length of 28 mm in surface cementation is able to create
176
a sufficient primary stability of the tibial plateau. In addition, the observed failure mode in all tested specimen
177
was migration into the metaphyseal head of the human tibiae, mechanically compromising the cement bone
178
interface under these highly demanding dynamic loading conditions [14], and not debonding between implant
179
and cement [22,23,41].
180
However, the intention of the current study was to establish a standardised in vitro model for testing of the
181
implant-cement-bone interface conditions, which reflects the characteristical failure modes as described by
182
Gebert de Uhlenbrock et al. [26] for their post mortem retrieval analysis on 22 bicompartmental tibial
183
plateaus from 17 patients. By applying quasi-static axial tension as a measure of the interfaces’s mechanical
184
capacity, Gebert de Uhlenbrock et al. [26] examined the influence of the time in situ on fixation strength.
185
Their specimens were retrieved after 5.3 years (range 0 to 11 years) in situ (mean human donor age 80.1
186
years; mean BMI 30.5 kg/m ) with pull-out forces ranging from 2751 N after only 2 days down to 231 N after
®
2
6
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
187
9 years in service and by trend, a decrease of fixation strength with the time in situ was found. To quantify
188
the failure mode in terms of the proportion of the tibial tray fixation failing at either the implant-cement or
189
cement-bone interface, Gebert de Uhlenbrock et al. [26] introduced a scoring system wereby the distal
190
surface of the tibial tray was divided into six regions. They reported that 29.4 % of the post-mortem retrieved
191
specimens failed exclusively at the implant-cement interface, 5.9 % failed exclusively at the cement-bone
192
interface and 64.7 % failed in a mixed mode [26]. The synthetic polyurethane foam model chosen for our
193
study mimics a failure mode at the implant-cement interface, as well as at the cement-synthetic bone
194
interface.
195
Comparing the cemented tibial tray fixation strength of components made out of cobalt-chromium and in a
196
variant covered by an additional zirconium nitride (ZrN) multi-layer coating, for the bone cements Surgical
197
Simplex P, Cobalt , Palacos R, Smart Set HV and BonOs HV, no statistically significant difference in the
198
five different cement groups was found for the two tibial tray materials. In the current study, push-out forces
199
above 2500 N were obtained in each of the ten test series with the cobalt-chromium and with the ZrN multi-
200
layer tibial trays showing substantial superiority (p < 0.01) in relation to the reported 1220 N of the PFC
201
Sigma FB tray – a clinically long-term successful implant [21,35]. On twelve paired, proximal thirds of human
202
donor tibiae Schlegel et al. [21] performed a similar pull-out test to determine the fixation strength of PFC
203
Sigma FB trays, using a similar surface cementing technique with Smart Set HV bone cement. In all pulsed
204
lavage tibial bone preparations (n = 6) failure was induced at the implant-cement interface with a mean pull-
205
out force of 1220 N (range 864 N to 1391 N), allowing for a direct comparison to our current results obtained
206
in a synthetic polyurethane foam model according to ASTM F1839-08. These results from Schlegel et al. [21]
207
fit well within the corridor of the current measurements on the PFC Sigma tibial tray and therefore they are
208
suitable to build a bridge from the current testing to a clinically long-term established predicate device.
209
Due to the fact that PMMA bone cement is not adhesive in interaction with metallic or ceramic implants and
210
the exact mechanism for implant fixation is a form-locking connection between the macro- and micro-
211
structured surface of the tibial tray and bone cement, the dominant factor influencing the implant-cement
212
interface bonding is the surface roughness [42]. From a biomechanical point of view the similar fixation
213
behaviour of the cobalt-chromium and the ZrN multi-layer coated tibial trays is reasonable, because of their
214
equivalent F22 corundium blast surface texture and roughness (Rz 25-35; Ra 3.5-4.5), whereas the PFC
215
Sigma FB has a visible and measureable smoother surface (Rz 5-7; Ra 0.5-1). These results on the PFC
216
Sigma FB tray fit very well into the corridor of roughness values (Ra 0.3-0.74), measured for PFC Sigma FB
217
tibial implants (n = 12 Ti6Al4V; n = 8 CoCrMo) in a comparative retrieval study by Cerquiglini et al. [43].
®
®
®
®
®
®
7
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1 ®
218
Performing the tests with five different bone cements for the Vega PS cobalt-chromium tibia tray a mean
219
push-out force in a range of 2943 N to 3302 N and for the Vega PS ZrN multi-layer tibia tray of 2705 N to
220
3107 N was measured. Between type of bone cement (Surgical Simplex P, Cobalt , Palacos R, Smart
221
Set HV and BonOs HV) for Vega PS cobalt-chromium & ZrN multi-layer no substantial differences were
222
found. These findings are in good accordance with the literature [42], where during characterization of
223
mechanical properties under quasi-static loading conditions typically no substantial differences are reported
224
between PMMA bone cements for 4-point-bending, compressive (ISO 5833:2002(E)) and tensile strength
225
(ISO 527-1:2012(E)). Under the aspect of clinical longterm behaviour, the fatigue properties of the bone
226
cements are of particular significance. Under flexural fatigue testing (ISO 16402:2008(E)) of bone cements,
227
e.g. Palacos R with a remaining fatigue strength of 16.6 MPa at 10 cycles shows a significantly better
228
fatigue performance than Surgical Simplex P (12.9 MPa at 10 cycles), whereas under quasi-static test
229
conditions both cements are in a range around 60 MPa [42,48]. The leading factor hereto is the molecular
230
weight of the polymer powder and the cured bone cement, whereby sterilization by ethylene oxide (EO) has
231
no influence on the molecular weight of the cement powder, but γ-irradiation significantly reduces it [42]. EO-
232
sterilised bone cements such as Cobalt , Palacos R, Smart Set HV and BonOs HV have a higher molecular
233
weight between 650,000 and 700,000 Dalton and subsequently a better fatigue performance, than bone
234
cements sterilised by γ-irradiation like Surgical Simplex P with a lower molecular weight between 250,000
235
and 300,000 Dalton [42,44].
236
Billi et al. [34] analyzed techniques for improving the initial strength of the tibial tray-cement interface bonding
237
and the effect of eight variables. Using a Triathlon tibial tray in surface cementation technique and an acrylic
238
material test block with high chemical affinity to bone cement, they found that late cementing reduced the
239
mean interface strength by 47% for Surgical Simplex P and by 73% for Palacos R compared to the normal
240
timing. Moreover, early cementing increased the mean interface strength by 48% for Surgical Simplex P and
241
by 72% for Palacos R versus cementing under normal conditions. In the present study, the influence of the
242
timing for cement apposition to the tibial tray was lower. For BonOs HV by trend an increase in fixation
243
strength by 2.3% for cobalt-chromium and by 6.3% for ZrN multi-layer was measured for the early
244
cementation in comparison to the mid of processing window, whereas for the late usage a decrease by
245
11.9% for cobalt-chromium and by 16.6% for ZrN multi-layer was found. For Smart Set HV, the early to mid
246
cementation increased the failure load by 12.9% for cobalt-chromium and by 16.3% for ZrN multi-layer,
247
whereas late usage decreased it by 4.3% for cobalt-chromium and by 23.1% for ZrN multi-layer, but without
248
statistically significant differences.
®
®
®
®
®
®
®
®
7
®
®
7
®
®
®
®
®
®
®
®
®
®
8
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
249
Possible reasons for the present different findings compared to Billi et al. [34] may be the different design,
250
surface texture and roughness of the Vega tibial tray, the storage for 7 days in water ensuring a defined
251
hardening status (ISO 5833) and possible fluid immersion versus 48 hours under dry conditions [34] and the
252
introduction of a synthetic polyurethane foam model allowing for clinically relevant fixation failure modes on
253
both interfaces [26].
®
254 ®
255
Examining the clinical behaviour of uncoated and ZrN multi-layer coated e.motion UC knee implants, a
256
mobile bearing, rotating platform design, Thomas et al. [45] performed a retrospective multi-center study on
257
196 TKA patients with an average follow-up of 5.7 years. In three participating centers they examined all
258
responding TKA patients, who received an e.motion UC in 2007 in uncoated (n = 99; age 67.0, BMI 29.8) or
259
a ZrN multi-layer coated version (n = 97; age 69.3, BMI 27.0) within a 5-year follow-up, including detailed
260
physical examination, knee pain & function (KSS), x-ray documentation (leg axis, positioning, radiolucencies)
261
and survival via Kaplan-Meier analysis. Thomas et al. [45] reported a favourable survivorship of 97 % for
262
uncoated and of 98 % for ZrN multi-layer coated TKA’s after 5.5 years, without significant difference between
263
the groups.
264
In a prospective randomized clinical trial on 120 primary TKA patients treated with a Columbus DD fixed
265
bearing knee implant either in cobalt-chromium uncoated (n = 59; age 68.6, BMI 30.5) or in ZrN multi-layer
266
coating (n = 61; age 66.6, BMI 31.3), Beyer et al. [46] published midterm clinical results with 5 year follow up.
267
Patient related outcomes (PROMS) were significantly improved for the uncoated as well as for the ZrN multi-
268
layer coated implant cohort. The Oxford Knee Score (OKS) improved substantially in both, the uncoated
269
group from 21.9 points (SD 7.6) preoperatively to 39.2 points (SD 7.9) four years after surgery and the ZrN
270
multi-layer coated group from 21.6 points (SD 6.2) to 39.5 points (SD 7.8), respectively [46]. Beyer et al. [46]
271
reported an excellent survival in both groups, with a 5-year survival rate of 98.1 % in the uncoated and of
272
100 % in the ZrN multi-layer coated TKA group.
273
For the ZrN multi-layer coated Vega PS knee system, Lionberger et al. [47] reported in a retrospective study
274
on a single surgeon case series an unacceptable failure rate of 6% prior to 1 year follow-up (15 aseptic
275
loosenings out of a cohort of 249 TKA’s implanted from 2015 to 2017) in comparison to a previously
276
implanted cohort of 850 Columbus cruciate retaining knees (2009 to 2014) with only two known revisions
277
due to aseptic loosening (0.24%). They hypothesized that the new design of the Vega PS tibial tray may be
278
a factor to be considered for their unfavourable clinical outcomes [47]. These findings are in contrast to the
279
promising short-term clinical behaviour of the Vega PS knee system reported by Jain et al. [48]. In three
®
®
®
®
®
9
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1 ®
280
cohorts they compared the clinical outcomes of 206 consecutive TKA’s using Vega PS with those of two
281
clinically long-term established posterior stabilized designs (e.motion PS n = 205; Genesis II n = 216) in a
282
two year follow-up study and found comparable or superior functional clinical performance of Vega PS
283
without incidence of implant-related adverse events [48]. Lionberger et al. [47] further hypothesized that the
284
ceramic hardened implant surface may be a potential factor for limited cement adhesion and early debonding
285
in the later loosenings and proposed further testing to ascertain the root cause for these failures, but his
286
cementation technique lists he was using Palacos R and cementing in the late window we investigated in
287
this study and he also lists two other cementation techniques that are specifically contrary to manufacturer
288
recommendations. The current study, including the results of this detailed evaluation and testing, showed a
289
high interface fixation strength of the cemented Vega PS tibial plateau without significant differences
290
between the cobalt-chromium and the ZrN multi-layer surface version.
®
®
®
®
291 292
Evaluating the clinical outcomes of primary TKA in the National Joint Registry of England, Wales, Northern
293
Ireland & Isle of Man with an observation time of 5 and 7 years, Porter et al. [5] reported for the uncoated
294
Columbus knee in cobalt-chromium (n = 10,659) a survivorship of 97.6 % at 5 years and of 97.2 % at 7
295
years which is comparable to ALL TKA (n = 977,488) with 97.8 % at 5 years and 97.3 % at 7 years. For the
296
ZrN multi-layer coated AS Columbus knee (n = 1,067) a survival rate of 98.6 % at 5 years and of 98.1 % at
297
7 years was calculated, which is significantly superior to the Columbus uncoated and to ALL TKA in the
298
National Joint Registry [5].
®
®
®
299
300
VI. Conclusion
301
From the observations reported in this study, it is concluded that there is no significant difference in implant-
302
cement-bone fixation strength between cobalt-chromium and ZrN multi-layer coated Vega tibial trays tested
303
with five different commonly used bone cements. When differences between types of bone cement (Surgical
304
Simplex P, Cobalt , Palacos R, Smart Set HV and BonOs HV) were investigated no substantial differences
305
were measured for either cobalt-chromium or ZrN multi-layer implant surface groups. Apposition of bone
306
cements and tibial tray implantation in the early to mid of the cement specific processing window is beneficial
307
in regard to interface fixation in TKA.
®
®
®
®
®
®
308 309
10
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
311
Acknowledgements: None Funding: None
312
Ethical approval: Not required
310
313 314
Tables
315 316 317 318
Table 1: Cement apposition and component implantation times for early, mid-term and late usage within the cement specific processing window at 19 °C early usage [min’ sec’’]
mid-term usage [min’ sec’’]
late usage [min’ sec’’]
intended
mean ± Std.
intended
mean ± Std.
intended
mean ± Std.
Surgical Simplex®P
---
---
7' 30''
7' 29'' ± 0' 05''
---
---
Cobalt®
---
---
6' 00''
5' 54'' ± 0' 10''
---
---
Palacos®R
---
---
4' 30''
4' 28'' ± 0' 15''
---
---
Smart Set®HV
1' 30''
1' 33'' ± 0' 10''
6' 30''
6' 23'' ± 0' 10''
9' 30''
9' 30'' ± 0' 01''
BonOs®HV
2' 00''
2' 12'' ± 0' 20''
6' 00''
6' 05'' ± 0' 18''
8' 30''
8' 30'' ± 0' 01''
319 320 321 322
Figures
323 324 325 326
Fig. 1: Polyurethane foam blocks (20 pcf cellular rigid foam, Sawbones, Sweden), representing the contour ® of the osteodenser, for the distal part of the Vega tibial tray (size 0) with a gap of at least 1 mm
327 328 329
Fig.2: Paired cementation of the CoCrMo and ZrN tibial trays
11
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
330 331 332
Fig. 3: Parallel cement application and insertion of both tibial plateaus
333 334 335
Fig. 4: Controlled pressurization and removal of surplus bone cement
336 337 338 339
Fig.5: Controlled hardening of both plateaus
12
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
test machine frame rod for load application Sawbone polyurethan boam block with sample in upside down position test frame
stencil for the sample size sample (tibia)
340 341 342 343 344
345 346 347 348
Fig. 6: Push-out testing after 7 day storage of the specimens in water for defined hardening status and possible fluid immersion.
Fig. 7: Failure load of the tested groups with 5 different bone cements (mid-term cement apposition) in comparison to Schlegel et al. [21] (green doted line)
13
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
349 350
351 352
353 354 355 356 357 358
®
Fig. 8: Typical failure characteristics for the Vega PS cobalt-chromium (right marked with “C”) and ZrN multi-layer (left marked with “A”) tibial trays with a combination of failure at the implant-cement and at the cement-foam interface (mixed mode acc. to Gebert de Uhlenbrock et al. [26]) tested with 5 different bone cements
14
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
359 360 361 362 363 364 365
®
Fig. 9: Typical failure characteristics for the PFC Sigma FB tray tested with Smart Set HV with a predominant failure at the implant-cement interface as also reported by Schlegel et al. [21]
366 367 368 369 370
Fig. 10: Failure load of the tested groups with BonOs R for different cement apposition times in comparison to Schlegel et al. [21] (green doted line)
15
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
371 372 373 374 375 376
®
Fig. 11: Failure load of the tested groups with Smart Set HV for different cement apposition times in comparison to Schlegel et al. [21] (green doted line)
377
378 379 380
Fig. 12: Surface roughness Rmax, Ra and Rz of all tested specimens
381 382
16
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436
Literature 1 Kurtz SM, Ong KL, Lau E, Widmer M, Maravic M, Gómez-Barrena E, de Fátima de Pina M, Manno V, Torre M, Walter WL, de Steiger R, Geesink RGT, Peltola M, Röder C: International survey of primary and revision total knee replacement. Int Orthopaedics (SICOT) 2011;35(2):1783-9. 2 Graves S, de Steiger R, Lewis P, Harris I: Australian Orthopaedic Association National Joint Replacement Registry for Hip, Knee & Shoulder Arthroplasty. Annual Report 2018:235-63. 3 Carr AJ, Robertsson O, Graves S, Price AJ, Arden NK, Judge A, Beard DJ: Knee Replacement. The Lancet 2012;379:1331-1340. 4 Evans JT, Walker RW, Evans JP, Blom AW, Sayers A, Whitehouse MR: How long does a knee replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet 2019; 393: 655–63. 5 Porter M, Howard P, Lawrence S, Reed M, Stonadge J, Wilkinson M: National Joint Registry England, Wales, Northern Ireland & Isle of Man. Annual Report 2018:102-51. 6 Grimberg A, Jansson V, Liebs T, Melsheimer O, Steinbrueck A: Endoprothesenregister Deutschland (EPRD) Jahrensbericht 2017:33-42. 7 Gothesen O, Espehaug B, Havelin L, Petursson G, Lygre S, Ellison P, Hallan G, Furnes O: Survival rates and causes of revision in cemented primary total knee replacement – Norwegian Arthroplasty Register 19942009. Bone Joint J 2013;95B:636-42. 8 Dyrhovden GS, Lygre SHL, Badawy M, Gothesen O, Furnes O: Have the causes of revision for total and unicompartmental knee arthroplasties changed during the past two decades? Clin Orth Rel Res 2017;475(7):1874-87. 9 Sadoghi P, Liebensteiner M, Agreiter M, Leithner A, Böhler N, Labek G: Revision surgery after total joint arthroplasty: a complication-based analysis using worldwide arthroplasty registers. J Arthroplasty 2013;28:1329-1332. 10 Niinimaeki TT: The reasons for knee arthroplasty revisions are incomparable in the different arthroplasty registries. Knee 2015;22:142-4. 11 Lum ZC, Shieh AK, Dorr LD. Why total knees fail-A modern perspective review. World J Orthop 2018; 9(4):60-4. 12 Delanois RE, Mistry JB, Gwam CU, Mohamed NS, Choksi US, Mont MA: Current epidemiology of revision total knee arthroplasty in the United States. J Arthroplasty 2017;32:2663-8. 13 Lombardi Jr AV, Berend KR, Adams JB: The revision knee arthroplasty – Why knee replacements fail in 2013 patient, surgeon or implant? J Bone Joint Surg 2014;96B:101-4. 14 Grupp TM, Saleh KJ, Holderied M, Pfaff A, Schilling C, Schroeder C, Mihalko WM. Primary stability of tibial plateaus under dynamic compression-shear loading in human tibiae – influence of keel length, cementation area and tibial stem. J Biomechanics 2017;59:9-22. 15 Cawley DT, Kelly N, Simpkin A, Shannon FJ, McGarry JP: Full and surface tibial cementation in total knee arthroplasty – a biomechanical investigation of stress distribution and remodeling in the tibia. Clinical Biomechanics 2012;27:390-7. 16 Cawley DT, Kelly N, McGarry JP, Shannon FJ: Cementing techniques for the tibial component in primary total knee replacement. J Bone and Joint Surg 2013;95B(3):295-300. 17 Schlegel UJ, Bruckner T, Schneider M, Parsch D, Geiger F, Breusch SJ: Surface or full cementation of the tibial component in total knee arthroplasty: a matched-pair analysis of mid- to long-term results. Arch Orthop Trauma Surg 2015;135(5):703-708. 18 Sundberg M, Lidgren L, W-Dahl A, Robertsson O: Swedish knee arthroplasty register – Annual Report Lund University, Lund Sweden, 2014. 19 Gothesen O, Espehaug B, Havelin L, Petursson G, Lygre S, Ellison P, Hallan G, Furnes O: Survival rates and causes of revision in cemented primary total knee replacement – Norwegian Arthroplasty Register 19942009. Bone Joint J 2013;95B:636-42. 20 Furnes O, Espehaug B, Lie SA, Vollset SE, Engesaeter LB, Havelin L: Failure mechanism after unicompartmental and tricompartmental primary knee replacement with cement. J Bone and Joint Surg 2007;89A(3):519-25. 21 Schlegel UJ, Siewe J, Delank KS, Eysel P, Püschel K, Morlock MM, Gebert de Uhlenbrock A: Pulsed lavage improves fixation strength of cemented tibial components. International Orthopaedics (SICOT) 2011;35:1165-9. 17
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
22 Ries C, Heinichen M, Dietrich F, Jakubowitz E, Sobau C, Heisel C: Short-keeled cemented tibial components show an increased risk for aseptic loosening. Clin Orthop Relat Res 2013;471:1008-13. 23 Arsoy D, Pagnano MW, Lewallen DG, Hanssen AD, Sierra RJ: Aseptic tibial debonding as a cause of early failure in a modern total knee arthroplasty design. Clin Orthop Relat Res 2013;471:94-101. 24 Galasso O, Jenny JY, Saragaglia D, Miehlke RK: Full versus surface tibial baseplate cementation in total knee arthroplasty. Orthopedics 2013;36(2):e151-8. 25 Clarius M, Hauck C, Seeger JB, James A, Murray DW, Aldinger PR: Pulse lavage reduces the incidence of radiolucent lines under the tibial tray of Oxford unicompartmental knee arthroplasty. International Orthopaedics (SICOT) 2009;33:1585-90. 26 Gebert de Uhlenbrock A, Püschel V, Püschel K, Morlock MM, Bishop NE: Influence of time in-situ and implant type on fixation strength of cemented tibial trays – A post mortem retrieval analysis. Clinical Biomechanics 2012;27(9):929-935. 27 Staats K, Wannmacher T, Weihs V, Koller U, Kubista B, Windhager R: Modern cemented total knee arthroplasty design shows higher incidence of radiolucent lines compared to its predecessor. Knee Surgery, Sports Traumatology, Arthroscopy 2019;27:1148-1155. 28 Maistrelli GL, Antonelli L, Fornasier V, Mahomed N: Cement penetration with pulsed lavage versus syringe irrigation in total knee arthroplasty. Clin Orthop Rel Res 1995;312:261-65. 29 Clarius M, Seeger JB, Jaeger S, Mohr G, Bitsch RG: The importance of pulsed lavage on interface temperature and ligament tension force in cemented unicompartmental knee arthroplasty. Clinical Biomechanics 2012;27:372-76. 30 Jaeger S, Rieger JS, Bruckner T, Kretzer JP, Clarius M, Bitsch RG: The protective effect of pulsed lavage against implant subsidence and micromotion for cemented tibial unicompartmental knee components – an experimental cadaver study. J Arthroplasty 2014;29(4):727-32. 31 Completo A, Simoes JA, Fonseca F, Oliveira M: The influence of different tibial stem designs in load sharing and stability at the cement-bone interface in revision TKA. Knee 2008;15:227-32. 32 Kelly N: An experimental and computational investigation of the inelastic behaviour of trabecular bone. The National University of Ireland, Galway. Doctoral Thesis 2012:93-101. 33 Grupp TM, Pietschmann MF, Holderied M, Scheele C, Schröder C, Jansson V, Müller PE: Primary stability of unicompartmental knee arthroplasty under dynamic compression-shear loading in human tibiae. Clinical Biomechanics 2013;28:1006-13. 34 Billi F, Kavanaugh A, Schmalzried H, Schmalzried T: Techniques for improving the initial strength of the tibial tray-cement interface bond. Bone Joint J 2019;101B(1 Supple A):53-8. 35 Hopley CDJ, Dalury DF: A systematic review of clinical outcomes and survivorship after total knee arthroplasty with a contemporary modular knee system. J Arthroplasty 2014;29:1398-411. 36 Taylor WR, Heller MO, Bergmann G, Duda GN: Tibio-femoral loading during human gait and stair climbing. J Orthopaedic Research 2004;22:625-32. 37 Zhao D, Banks SA, D'Lima DD, Colwell CW, Fregly BJ: In vivo medial and lateral tibial loads during dynamic and high flexion activities. J Orthopaedic Research 2007;25(5):593-602. 38 Muendermann A, Dyrby CO, d'Lima DD, Colwell CW, Andriacchi TP: In vivo knee loading characteristics during activities of daily living as measured by an instrumented total knee replacement. J Orthopaedic Research 2008;26(9):1167-72. 39 Kutzner I, Heinlein B, Graichen F, Bender A, Rohlmann A, Halder A, et al.: Loading of the knee joint during activities of daily living measured in vivo in five subjects. J Biomechanics 2010;43:2164-73. 40 Bergmann G, Bender A, Graichen F, Dymke J, Rohlmann A, Trepczynski A, Heller MO, Kutzner I: Standardized Loads Acting in Knee Implants. PLoS ONE 2014;9(1) e86035.doi:10.1371/ journal.pone.0086035. 41 Foran JR, Whited BW, Sporer SM: Early aseptic loosening with a precoated low-profile tibial component a case series. J Arthroplasty 2011;26:1445-50. 42 Kuehn KD: PMMA cements – Are we aware what we are using? Springer Berlin Heidelberg 2014:1-14 & 183-230 DOI10.1007/978-3-642-41535-7.
18
Tibial implant fixation behaviour in TKA_Journal of Arthroplasty 2019_Rev.1
488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503
43 Cerquiglini A, Henckel J, Hothi H, Allen P, Lewis J, Eskelinen A, Skinner J, Hirschmann MT, Hart AJ: Analysis of the Attune tibial tray backside – a comparative retrieval study. Bone Joint Research 2019;8:136145. 44 Tepic S, Soltész U: Fatigue strength of bone cement with simulated stem interface porosity. J Mat Sciences Mater Med 1998;9:707-9. 45 Thomas P, Hisgen P, Kiefer H, Schmerwitz U, Ottersbach A, Albrecht D, Schinkel C: Blood cytokine pattern and clinical outcome in knee arthroplasty patients: comparative analysis 5 years after standard versus “hypoallergenic” surface coated prosthesis implantation. Acta Orthopaedica 2018;89 DOI:10.1080/17453674.2018.1518802 46 Beyer F, Luetzner C, Kirschner S, Luetzner J: Midterm results after coated and uncoated TKA: A randomized controlled study. Orthopedics 2016;39(3):S13-7. 47 Lionberger D, Conlon C, Wattenbarger L, Walker TJ: Unacceptable failure rate of a ceramic-coated posterior cruciate-substituting total knee arthroplasty. Arthroplasty Today 2019;5:1-6. 48 Jain NP, Lee SY, Morey VM, Chong S, Kang YG, Kim TK: Early clinical outcomes of a new posteriorly stabilized total knee arthroplasty prosthesis – comparison with two established prostheses. Knee Surg Relat Res 2017;29(3):180-8.
19