- Email: [email protected]
Outcomes After Arthroscopic Evaluation of Patients With Painful Medial Unicompartmental Knee Arthroplasty
Accepted Manuscript Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty Jason M. Hurst, MD, Riccardo Ranieri, Keith R. Berend, MD, Michael J. Morris, MD, Joanne B. Adams, BFA, Adolph V. Lombardi, Jr., MD, FACS PII:
S0883-5403(18)30515-1
DOI:
10.1016/j.arth.2018.05.031
Reference:
YARTH 56642
To appear in:
The Journal of Arthroplasty
Received Date: 10 April 2017 Revised Date:
16 May 2018
Accepted Date: 21 May 2018
Please cite this article as: Hurst JM, Ranieri R, Berend KR, Morris MJ, Adams JB, Lombardi Jr. AV, Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty, The Journal of Arthroplasty (2018), doi: 10.1016/j.arth.2018.05.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.
ACCEPTED MANUSCRIPT
Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty
RI PT
From: Joint Implant Surgeons, Inc., New Albany Ohio USA; White Fence Surgical Suites, New Albany, Ohio USA; and Mount Carmel Health System, Columbus, Ohio Short Title (10 words): Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty Jason M. Hurst, MDa, b, c
SC
Riccardo Ranieria, b, c, d
Keith R. Berend, MDa, b, c, e
M AN U
Michael J. Morris, MDa, b, c Joanne B. Adams, BFAa
Adolph V. Lombardi, Jr., MD, FACSa, b, c, e
a
b
TE D
Joint Implant Surgeons, Inc., 7277 Smith’s Mill Road, Suite 200, New Albany, Ohio USA 43054; (614) 221-6331 White Fence Surgical Suites, 7277 Smith’s Mill Road, Suite, Suite 300, New Albany, Ohio USA 43054; (855) 677-5005
c
d
EP
Mount Carmel Health System, 7333 Smith's Mill Road, New Albany, Ohio USA 43054; (614) 775-6600 Universitá Campus Bio-Medico Roma, Via Álvaro del Portillo, 21, Roma, Italia 00128; (+39) 06-225411
e
AC C
Department of Orthopaedics, The Ohio State University Wexner Medical Center, 376 W 10th Ave, Suite 725, Columbus, Ohio USA 43210; (614) 293-3715 Please direct correspondence to: Jason M. Hurst, MD Joint Implant Surgeons, Inc. 7277 Smith’s Mill Road, Suite 200 New Albany, Ohio USA 43054 Phone: (614) 221-6331 Fax: (614) 221-9042 Email: [email protected]
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty
AC C
1 2 3
1
ACCEPTED MANUSCRIPT
4
Abstract
5 Introduction: Persistent pain after medial unicompartmental knee arthroplasty (UKA) is a
7
prevailing reason for revision to total knee arthroplasty (TKA). Many of these pathologies can be
8
addressed arthroscopically. The purpose of this study is to examine the outcomes of patients who
9
undergo an arthroscopy for any reason after medial UKA.
SC
RI PT
6
10
Methods: A query of our practice registry revealed 58 patients who had undergone medial UKA
12
between October 2003 and June 2015 with subsequent arthroscopy. Mean interval from UKA to
13
arthroscopy was 22 months (1-101). Indications for arthroscopy were acute anterior cruciate
14
ligament tear (1), arthrofibrosis (7), synovitis (12), recurrent hemarthrosis (2), lateral
15
compartment degeneration including isolated lateral meniscus tears (11), and loose cement
16
fragments (25).
TE D
17
M AN U
11
Results: Mean follow-up after arthroscopy was 49 months (range, 1-143). Twelve patients have
19
been revised from UKA to TKA. Relative risk of revision after arthroscopy for lateral
20
compartment degeneration was 4.27 (6 of 11; 55%; p=0.002) and for retrieval of loose cement
21
fragments was 0.05 (0 of 25; 0%; p=0.03). Relative risk for revision after arthroscopy for
22
anterior cruciate ligament tear, arthrofibrosis, synovitis, or recurrent hemarthrosis did not meet
23
clinical significance secondary to the low number of patients in these categories.
AC C
EP
18
24 25
Conclusions: The results of this study suggest that arthroscopic retrieval of cement fragments
26
does not compromise UKA longevity. However, arthroscopy for lateral compartment 2
ACCEPTED MANUSCRIPT
27
degradation after UKA, while not the cause of revision, appears to be an ineffective treatment
28
and predicts a high risk of revision to TKA regardless of its relative radiographic insignificance.
RI PT
29 Keywords: medial unicompartmental knee arthroplasty, arthroscopy, lateral compartment
31
degeneration, loose cement fragment, pain
AC C
EP
TE D
M AN U
SC
30
3
ACCEPTED MANUSCRIPT
Introduction
33
Medial unicompartmental knee arthroplasty (UKA) is an increasingly utilized orthopedic
34
procedure to treat isolated medial compartment osteoarthritis, which has been reported to occur
35
in as many as 25% of patients with osteoarthritis of the knee [1]. Excellent long-term results
36
comparable to total knee arthroplasty (TKA) have been reported with medial UKA [2-6].
RI PT
32
37
Persistent pain is a common complication after UKA and is also an important reason for revision
39
of unicompartmental implants to TKA [7-9]. However it is suggested that the higher rates of
40
revision for pain in UKA reflect surgeon willingness to advise re-operation of a painful UKA
41
rather than a painful TKA because it is an easier and less hazardous procedure [10]. Furthermore
42
some studies show that revision of UKA to TKA for unexplained pain generally results in a less
43
favorable outcome than revision for a known cause of pain [11, 12].
M AN U
SC
38
TE D
44
Occasionally there are clear clinical issues that cause pain in the knee after UKA, and examples
46
include lateral compartment degeneration, loose bodies, synovitis, recurrent hemarthrosis, and
47
arthrofibrosis. Arthroscopy has been demonstrated to be a useful procedure for treating many of
48
these complications after UKA [13-23], and for diagnosing the etiology of unexplained pain after
49
UKA [14]. However, but there is a paucity of published data which analyze the clinical results of
50
arthroscopy after UKA. The purpose of this study is to examine the outcomes of patients who
51
undergo an arthroscopy after medial UKA.
53
AC C
52
EP
45
Materials and methods
4
ACCEPTED MANUSCRIPT
A query of our practice registry revealed 58 patients who had undergone medial UKA between
55
October 2003 and June 2015 with subsequent arthroscopy. These 58 patients (58 knees) treated
56
with arthroscopy after medial UKA represent 1.34% (58 or 4319) of medial UKA performed in
57
patients at our center during the study period. All patients signed a general research consent
58
approved and monitored by an independent institutional review board (Western IRB, Puyallup,
59
WA) and allowing inclusion of their records in retrospective reviews. No outside funding was
60
received in support of this study. Indications for medial mobile-bearing UKA were anteromedial
61
osteoarthritis or avascular necrosis of the medial condyle with full thickness cartilage loss
62
medially, intact lateral cartilage, intra-articular varus deformity correctible with valgus stress and
63
with functionally intact ligamentous stabilizers. Implant types used for medial UKA (all Zimmer
64
Biomet, Warsaw, IN) were Vanguard M fixed-bearing in 2 knees (3%), Oxford Phase III mobile-
65
bearing in 17 (29%), and Oxford Partial Knee Twin-Peg mobile-bearing in 39 (67%).
66
Medial UKA procedures were performed by 4 different surgeons (AVL, KRB, JMH, MJM), and
67
arthroscopy procedures were performed by two of the surgeons (JMH and KRB).
68
A retrospective database review was performed to evaluate reasons for arthroscopy and
69
outcomes after arthroscopy including need for revision surgery.
SC
M AN U
TE D
EP
70
RI PT
54
Patient demographics, including gender, age, height, weight, body mass index (BMI), and length
72
of follow-up were recorded. Patients were asked to return for clinical evaluation at 6 weeks
73
postoperative and annually thereafter or immediately if a problem with the operative extremity
74
arose. Clinical evaluations included assessment of range of motion (ROM), Knee Society
75
Clinical Rating System (KS) scores [24], the University of California at Los Angeles Activity
76
Scale (UCLA) [25]. The Knee Society pain subcomponent is on a 50-point scale with 0 points
AC C
71
5
ACCEPTED MANUSCRIPT
77
being “severe” pain and 50 points being “none”. All devices used have been approved by the
78
U.S. Food and Drug Administration, and were used according to the manufacturer’s labeling.
79 Results
RI PT
80 81
Mean interval from medial UKA to arthroscopy was 22 months (range, 1-101 months). In the
83
interval between UKA and arthroscopy 23 patients received corticosteroid injections and one
84
patient was treated with manipulation under anesthesia. Indications for arthroscopy were: 1 acute
85
anterior cruciate ligament tear (1.7%) treated with ACL reconstruction; 7 arthrofibrosis (12.1%)
86
treated with lysis of adhesion and/or excision of scar tissue and/or synovectomy; 12 synovitis
87
(20.7%) treated with synovectomy; 2 recurrent hemarthrosis (3.4%) treated with lysis of
88
adhesion and or/ synovectomy; 11 lateral compartment degeneration including isolated lateral
89
meniscus tears (19.0%) treated with chondroplasty and/or debridement and/or shaving and/or
90
meniscectomy; and 25 loose cement fragments (43.1%) treated with removal of loose or foreign
91
body and with partial synovectomy in some cases. The majority of these were clinical diagnoses
92
based on physical examination, including painful synovitis, arthrofibrosis, and hemarthrosis.
93
Loose cement fragments were diagnosed by a combination of acute pain with loose cement
94
visible on radiographs. Five of the 11 patients done for lateral compartment degeneration and 12
95
patients overall were diagnosed with MRI, and one with lateral progression was diagnosed with
96
CT. The remaining patients were diagnosed based on physical examination only. All of the
97
diagnoses that were mentioned were confirmed at the time of arthroscopic evaluation. In regard
98
to lateral compartment degeneration, several patients were documented preoperative to
99
arthroscopy to have meniscal injury based on MRI evaluation. The definition of lateral
AC C
EP
TE D
M AN U
SC
82
6
ACCEPTED MANUSCRIPT
compartment degeneration in this series is any meniscal damage requiring debridement and/or
101
the presence of chondral damage Grade 2 or worse. Valgus stress views were not performed for
102
patients in the current study, but this modality was added subsequently to our radiographic
103
evaluation of patients with painful UKA. MRI findings often times confirmed our clinical
104
suspicion in cases with lateral compartment pathology and ACL insufficiency.
RI PT
100
105
Of the 25 patients who underwent arthroscopy to remove loose fragments, size or general
107
description of the fragments extracted was documented in the dictated operative reports of all
108
patients, ranging from “two very small” in one patient to “10x15mm as the largest of several” in
109
another patient. While we did not routinely measure retrieved fragments at the time of
110
arthroscopy, all fragments were estimated to be greater than 5mm in length.
M AN U
SC
106
111
Patient demographics and outcomes measures by reason for arthroscopy are shown in Table 1. In
113
regard to demographics, patients requiring arthroscopy for treatment of arthrofibrosis were
114
younger (mean 54-years-old) than those patients treated with arthroscopy for removal of loose
115
cement (mean 64-years-old; P=0.003) or lateral compartment degeneration (mean 65-years-old;
116
P=0.027). Preoperatively, patients having arthroscopy for treatment of arthrofibrosis had less
117
pain (KS pain component 32.1) compared with those treated for removal of loose cement (KS
118
pain16.9; P=0.044) or synovitis (KS pain 10.8; P=0.0022), and higher KS clinical scores (78.7)
119
than patients treated for synovitis (57.1; P=0.0020). There were no other differences in patient
120
demographics or preoperative clinical measures between etiology groups.
AC C
EP
TE D
112
121
7
ACCEPTED MANUSCRIPT
Mean follow-up after arthroscopy was 49 months (range, 1-143, SD 34.1). Postoperatively,
123
patients treated for removal of loose bodies had greater improvement in ROM (2.8°) compared
124
with those treated for lateral compartment degeneration (-8.2°, P=0.0164). Patients treated for
125
arthrofibrosis had a deterioration in KS pain component (-2.5), KS clinical score (-1.8), and KS
126
functional score (-5.0) compared to their preoperative levels, with only a slight improvement in
127
ROM (2.5°). This was in contrast to patients treated with arthroscopy for removal of loose bodies
128
(KS pain improvement 21.7, P=0.041; KS functional improvement 22.0, P=0.041), and those
129
treated for synovitis (KS functional improvement 23.3, P=0.0297). Patients undergoing
130
arthroscopy for synovitis also had greater KS functional improvement (23.3) compared with
131
those treated with arthroscopy for lateral compartment degeneration (KS function improvement
132
1.8, P=0.008). Range of motion improvement was greater for patients treated for removal of
133
loose bodies (2.8°) compared with those treated for lateral degeneration (-8.2°; P=0.0164), as
134
was KS functional improvement (26.3 versus 1.8, P=0.023). There were no differences in KS
135
clinical improvement or postoperative UCLA scale between any of the of etiology groups.
SC
M AN U
TE D
136
RI PT
122
Thirteen patients have been revised from UKA to TKA (Table 2). When considering the reasons
138
for arthroscopy and the need for subsequent revision, 2 of 7 patients with arthrofibrosis (28.6%),
139
4 of 12 patients with synovitis (33%), 1 of 2 patients with recurrent hemarthrosis (50%) and 6 of
140
11 patients with lateral compartment degeneration (54.5%) underwent revision to TKA. No
141
patients who received arthroscopic procedure for ACL tear or loose cement fragment underwent
142
revision to TKA. Relative risk of revision to TKA after arthroscopy for lateral compartment
143
degeneration was 3.66 (6 of 11; 55%; 95% CI=1.53-8.75; p=0.0035) and for retrieval of loose
144
cement fragments was 0.05 (0 of 25; 0%; 95% CI=0.003-0.78; p=0.03). The relative risk for
AC C
EP
137
8
ACCEPTED MANUSCRIPT
145
revision after arthroscopy for acute anterior cruciate ligament tear, arthrofibrosis, synovitis, or
146
recurrent hemarthrosis did not meet clinical significance secondary to the low number of patients
147
in these categories.
RI PT
148
Two patients underwent a second arthroscopy procedure: one treated initially for lateral
150
compartment degeneration underwent a second debridement and removal of a chondral loose
151
body, and one treated initially for arthrofibrosis underwent a subsequent diagnostic arthroscopy
152
with synovectomy. One patient treated for arthrofibrosis subsequently underwent an open
153
excision of prepatellar fat pad fibrosis. One patient had pulmonary embolism successfully treated
154
with pharmacological anticoagulation after arthroscopy procedure performed for loose cement
155
fragment removal. One patient treated for synovitis underwent manipulation under anesthesia 2
156
years after arthroscopy.
M AN U
SC
149
TE D
157 Discussion
159
The results of this study suggest that in patients with persistent pain after medial UKA,
160
arthroscopy can be an effective procedure for some but not all underlying causes.
161
In the 25 patients with loose cement fragments after UKA, arthroscopic retrieval of cement does
162
not compromise UKA longevity and has a successful outcome. After a mean follow-up of 44
163
months none of these patients were revised to TKA.
AC C
164
EP
158
165
Even though symptomatic free cement fragments are not a common complication after UKA [9,
166
13, 26] some authors have noted that extrusion of cement and resulting loose cement fragments
167
are a problem in UKA especially with the use of minimal invasive techniques which reduce 9
ACCEPTED MANUSCRIPT
visibility with particular disadvantage for the posterior area of the knee. Particulate debris related
169
to third-body wear, specifically cement particles, is well known in TKA to lead to early failure of
170
ultra-high-molecular-weight polyethylene after TKA [18]. Some authors have reported damage
171
to UKA due to free cements bodies [17, 18]. Berger et al. first described one case of a revision
172
surgery to remove retained cement in UKA [15]. In the literature there are case reports of
173
symptomatic free cement fragments or fixed cement extrusion that were treated successfully with
174
arthroscopy [16, 19, 20]. This study confirms that arthroscopy is an effective procedure to
175
remove free cement and avoid revision. All of the UKA in the current series had cemented
176
fixation, and intuitively, loose cement fragments would not be an issue after cementless UKA.
177
Since we have not treated any patients with cementless UKA devices, we are unable to comment
178
further on the role of arthroscopy in patients with cementless UKA devices.
M AN U
SC
RI PT
168
179
After arthroscopy, our patients did not show a significant improvement of ROM (only 0.6°), but
181
there were substantial improvements in KS pain, clinical and functional scores compared with
182
scores before arthroscopy (respectively mean improvements of 34.3, 17.2 and 18.1 points). This
183
suggests that arthroscopy is not only a good procedure for avoiding revision to TKA in patients
184
with loose cement fragments, but also provides relief of pain and improvement in patient quality
185
of life.
EP
AC C
186
TE D
180
187
Lateral compartment degeneration is one of the major causes of revision for UKA. Lewold et al.
188
[27] reported a 1.4% rate of revision of Oxford UKA for contralateral arthrosis, accounting for
189
20% of all revisions. National registries have shown that progression of arthritis constitutes 25%
190
of the reasons for revision of UKA in Sweden and 33% in New Zealand [8, 28]. In our patients
10
ACCEPTED MANUSCRIPT
who developed lateral compartment degeneration after medial UKA, the use of arthroscopy
192
predicted a high risk of revision to TKA regardless of relative radiographic insignificance. Fifty-
193
five percent of patients underwent revision to TKA within a mean of 37 months regardless of
194
whether or not there was significant joint space narrowing on pre-arthroscopy radiographs.
195
Moreover, after arthroscopy these patients showed a slight mean worsening of ROM (-6.4°)
196
compared to ROM before arthroscopy. However, there were moderate improvements of KS pain,
197
clinical and functional scores (respectively mean improvements of 9, 8.3 and 7 points). These
198
data suggest that arthroscopy in patients with lateral compartment degeneration after medial
199
UKA is not a very effective procedure and does not lead to clinical improvement. A published
200
case report describes a single patient with unexplained pain after UKA where arthroscopy
201
showed severe osteoarthritis regardless of relative radiographic insignificance [29]. This
202
highlights the importance of specialized preoperative radiographs to evaluate the lateral
203
compartment, like posteroanterior (PA) flex or valgus stress view radiographs.
SC
M AN U
TE D
204
RI PT
191
Other indications for arthroscopy after UKA found in this study were for acute anterior cruciate
206
ligament tear, arthrofibrosis, synovitis, or recurrent hemarthrosis. In the literature these
207
complications are very rare causes of revision of UKA to TKA [7]. Our rates of revision to TKA
208
in patients who received arthroscopy for these complications did not demonstrate significance
209
secondary to the low number of patients in these categories. However, patients with synovitis
210
showed significant improvement of KS pain, clinical and functional scores after arthroscopy
211
(respectively mean improvements of 26.8, 27.1 and 16.4 points) and a slight mean improvement
212
of ROM (4°), indicating that these patients can benefit from an arthroscopic procedure. In a case
213
report of two patients in whom persistent pain after UKA was attributed to synovial
AC C
EP
205
11
ACCEPTED MANUSCRIPT
impingement, treatment with arthroscopy provided successful pain relief [23]. For patients with
215
arthrofibrosis after medial UKA, arthroscopy can provide improved ROM (mean improvement
216
was 10°), but KS pain, clinical and functional scores were unchanged after the arthroscopic
217
procedure.
RI PT
214
218
This study has some major limitations. First, the mean time of follow-up is short at only 37
220
months, and longer follow-up should be considered in future studies. Second, the number of
221
patients in different etiology groups (acute anterior cruciate ligament tear, arthrofibrosis,
222
synovitis, and recurrent hemarthrosis) was inadequate for purpose of statistical comparisons.
223
Additional studies are required to investigate the rates of revision after arthroscopy in patients
224
who develop these complications. Third, several patients with lateral degeneration had MRI
225
preoperative to arthroscopy, while others did not. None of the patients had valgus stress
226
radiographs after their UKA prior to arthroscopy, but this additional imaging is currently a
227
routine part of our diagnostic algorithm since undertaking the current review. It is important to
228
thoroughly re-evaluate the lateral compartment with auxiliary radiographic views, such as valgus
229
stress test and posteroanterior (PA) flex views, in the postoperative period after medial UKA.
230
EP
TE D
M AN U
SC
219
Conclusions
232
In literature there are studies which mention arthroscopy as a potential useful procedure in order
233
to treat some painful complications after UKA, but this is the first study which specifically
234
reports the outcome of a series of patients who underwent arthroscopy after UKA for multiple
235
reasons. The results of this study suggest that arthroscopic retrieval of cement fragments does not
236
compromise UKA longevity. However, arthroscopy for lateral compartment degradation after
AC C
231
12
ACCEPTED MANUSCRIPT
237
UKA predicts a high risk of revision to TKA regardless of its relative radiographic
238
insignificance. While the arthroscopy does not cause the revision, it appears to be an ineffective
239
treatment for arthritic progression after UKA.
RI PT
240
AC C
EP
TE D
M AN U
SC
241
13
ACCEPTED MANUSCRIPT
242
References
243
1.
245
2003;85(7):937-942. 2.
246
Price AJ, Svärd U. A second decade lifetable survival analysis of the Oxford
unicompartmental knee arthroplasty. Clin Orthop Relat Res 2011;469:174-9. 3.
Berger RA, Meneghini M, Jacobs JJ, Sheinkop MB, Della Valle CJ, Rosenberg AG,
SC
247
RI PT
244
Ackroyd CE. Medial compartment arthroplasty of the knee. J Bone Joint Surg Br.
Galante JO. Results of unicompartmental knee arthroplasty at a minimum of ten years of
249
follow-up. J Bone Joint Surg Am. 2005 May;87(5):999-1006.
250
4.
251 252
M AN U
248
Svärd UCG, Price AJ. Oxford medial unicompartmental knee arthroplasty: a survival analysis of an independent series. J Bone Joint Surg Br 2001;83:191-4.
5.
O'Rourke MR, Gardner JJ, Callaghan JJ, Liu SS, Goetz DD, Vittetoe DA, Sullivan PM, Johnston RC. The John Insall Award: unicompartmental knee replacement: a minimum
254
twenty-one-year followup, end-result study. Clin Orthop Relat Res 2005;440:27-37.
255
6.
TE D
253
Zuiderbaan HA, van der List JP, Khamaisy S, Nawabi DH, Thein R, Ishmael C, Paul S, Pearle AD. Unicompartmental knee arthroplasty versus total knee arthroplasty: Which type
257
of artificial joint do patients forget? Knee Surg Sports Traumatol Arthrosc.2015 Nov 21. 7.
National Joint Registry for England and Wales. 10th Annual Report 2013.
AC C
258
EP
256
259
http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/Reports/10th_annual_r
260
eport/NJR%2010th%20Annual%20Report%202013%20B.pdf Accessed 2013 June
261
8.
New Zealand Orthopaedic Association. The New Zealand Joint Registry. Thirteen year
262
report: January 1999 – December 2011.
263
http://nzoa.org.nz/system/files/NJR%2013%20Year%20Report.pdf. Accessed 2012 Oct.
14
ACCEPTED MANUSCRIPT
264
9.
Baker PN, Petheram T, Avery PJ, Gregg PJ, Deehan DJ. Revision for unexplained pain
265
following unicompartmental and total knee replacement. J Bone Joint Surg Am. 2012 Sep
266
5;94(17):e126. 10.
Goodfellow JW, O’Connor JJ, Murray DW. A critique of revision rate as an outcome
RI PT
267 268
measure. Re-interpretation of knee joint registry data. J Bone Joint Surg Br.
269
2010;92(12):1628-31. 11.
Kerens B, Boonen B, Schotanus MG, Lacroix H, Emans PJ, Kort NP. Revision from
SC
270
unicompartmental to total knee replacement: the clinical outcome depends on reason for
272
revision. Bone Joint J. 2013 Sep; 95-B(9):1204-8.
273
12.
M AN U
271
Châtain F, Richard A, Deschamps G, Chambat P, Neyret P. [Revision total knee
274
arthroplasty after unicompartmental femorotibial prosthesis: 54 cases]. Rev Chir Orthop
275
Reparatrice Appar Mot. 2004 Feb;90(1):49-57. 13.
277 278
Kim KT, Lee S, Lee JI, Kim JW. Analysis and treatment of complications after
TE D
276
unicompartmental knee arthroplasty. Knee Surg Relat Res. 2016 Mar;28(1):46-54. 14.
Hannaoui S, Lustig S, Servien E, Aït Si Selmi T, Neyret P. Arthroscopy of the knee after unicompartmental arthroplasty. Rev Chir Orthop Reparatrice Appar Mot. 2008
280
Nov;94(7):678-84. 15.
Berger RA, Nedeff DD, Barden RM, Sheinkop MM, Jacobs JJ, Rosenberg AG, Galante JO.
AC C
281
EP
279
282
Unicompartmental knee arthroplasty. Clinical experience at 6- to 10-year followup. Clin
283
Orthop. 1999;367:50-60.
284 285
16.
Howe DJ, Taunton OD Jr, Engh GA. Retained cement after unicondylar knee arthroplasty. A report of four cases. J Bone Joint Surg Am. 2004 Oct;86-A(10):2283-6.
15
ACCEPTED MANUSCRIPT
286
17.
Elmadağ M, Imren Y, Erdil M, Bilsel K, Tuncay I. Excess retained cement in the
287
posteromedial compartment after unicondylar knee arthroplasty. Acta Orthop Traumatol
288
Turc 2013;47(4):291-294. 18.
Hauptmann SM, Weber P, Glaser C, Birkenmaier C, Jansson V, Müller PE. Free bone
RI PT
289 290
cement fragments after minimally invasive unicompartmental knee arthroplasty: an
291
underappreciated problem. Knee Surg Sports Traumatol Arthrosc. 2008 Aug;16(8):770-5. 19.
Karataglis D, Agathangelidis F, Papadopoulos P, Petsatodis G, Christodoulou A.
SC
292
Arthroscopic removal of impinging cement after unicompartmental knee arthroplasty.
294
Hippokratia. 2012 Jan;16(1):76-9.
295
20.
M AN U
293
Kim WY, Shafi M, Kim YY, Kim JY, Cho YK, Han CW. Posteromedial compartment
296
cement extrusion after unicompartmental knee arthroplasty treated by arthroscopy: a case
297
report. Knee Surg Sports Traumatol Arthrosc. 2006 Jan;14(1):46-9. 21.
299 300
Yamakado K. Dissecting a popliteal cyst after failed unicompartmental knee arthroplasty.
TE D
298
Arthroscopy. 2002 Nov-Dec;18(9):1024-8. 22.
Yamakado K, Arakawa H, Hayashi S. Arthroscopic observation was useful to detect loosening of the femoral component of unicompartmental knee arthroplasty in a recurrent
302
hemoarthrosis. Sports Med Arthrosc Rehabil Ther Technol. 2012 Feb 21;4:8. 23.
304 305 306
Jung KA, Lee SC, Hwang SH. Pseudomeniscal synovial impingement after unicondylar
AC C
303
EP
301
knee arthroplasty. Orthopedics. 2009 May;32(5):361. 24.
Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res. 1989 Nov;(248):13-4.
16
ACCEPTED MANUSCRIPT
307
25.
Amstutz HC, Thomas BJ, Jinnah R, Kim W, Grogan T, Yale C. Treatment of primary
308
osteoarthritis of the hip. A comparison of total joint and surface replacement arthroplasty. J
309
Bone Joint Surg Am. 1984 Feb;66(2):228-41. 26.
Lewold S, Robertsson O, Knutson K, Lidgren L. Revision of unicompartmental knee
RI PT
310 311
arthroplasty: outcome in 1,135 cases from the Swedish Knee Arthroplasty study. Acta
312
Orthopedica. 1998 Oct;69(5):469-74. 27.
Lewold S, Goodman S, Knutson K, Robertsson O, Lidgren L. Oxford meniscal bearing
SC
313
knee versus the Marmor knee in unicompartmental arthroplasty for arthrosis. A Swedish
315
multicenter survival study. J Arthroplasty. 1995 Dec;10(6):722-31.
316
28.
317 318
M AN U
314
Lidgren L, Knutson K, Robertsson O. Swedish Knee Arthroplasty Register: Annual Report 2004. Lund: Swedish Knee Arthroplasty Register, 2004.
29.
Hama S, Hamada D, Goto T, Tsutsui T, Tonogai I, Suzue N, Matsuura T, Sairyo K. Revision total knee arthroplasty for unexplained pain after unicompartmental knee
320
arthroplasty: a case report. J Med Invest. 2015;62(3-4):261-3.
EP AC C
321
TE D
319
17
ACCEPTED MANUSCRIPT
Table 1. Patient demographics and outcomes measures by reason for arthroscopy.
Overall
ACL Tear
RI PT
Characteristic
Reason for Arthroscopy Lateral Loose Arthrofibrosis Compartment Cement Degeneration Fragment
Recurrent Hemarthrosis
Synovitis
AC C
EP
TE D
M AN U
SC
Number of 58 1 7 11 25 2 12 patients/knees Males:Females 29:29 1:0 4:3 5:6 14:11 2:0 3:9 Age (years) 61.7 ±8.8 58.7 53.9 ±8.3 64.7 ±9.7 64.1 ±7.2 58.2 ±3.0 59.2 ±9.8 Height (inches) 67.6 ±4.9 70 69.9 ±5.4 66.8 ±5.5 67.8 ±4.8 71.0 ±5.7 65.7 ±4.5 Weight (pounds) 207.6 ±45.4 255 205.7 ±52.0 200.4 ±57.8 217.2 ±43.4 209.5 ±50.2 191.1 ±32.6 BMI (kg/m2) 31.9 ±6.0 36.6 29.7 ±7.9 31.1 ±5.9 33.3 ±6.3 28.9 ±2.4 31.1 ±4.5 ROM preop 114.7 ±12.1 125 114.3 ±11.7 118.2 ±7.2 115.2 ±10.2 92.5 ±24.7 113.8 ±15.4 ROM postop 115.4 ±10.1 120 115.8 ±13.2 110.0 ±11.0 118.7 ±8.6 110.0 ±7.1 113.8 ±1.1 ∆ROM 0.6 ±12.7 -5.0 2.5 ±10.4 -8.2 ±13.2 2.8 ±11.2 17.5 ±17.7 1.1 ±12.7 KS pain preop 17.9 ±17.1 45 32.1 ±15.0 17.7 ±18.6 16.9 ±17.3 0.0 10.8 ±10.8 KS pain postop 34.3 ±19.4 50 28.3 ±24.8 30.0 ±20.2 37.0 ±18.1 50.0 ±0.0 31.7 ±20.3 ∆KS pain 17.2 ±24.5 5.0 -2.5 ±30.0 12.3 ±20.7 21.7 ±24.7 50.0 20.8 ±24.4 KS clinical preop 64.5 ±18.7 90 78.7 ±14.6 65.5 ±19.8 63.6 ±20.0 38.0 57.1 ±11.2 KS clinical postop 81.6 ±20.2 99 75.2 ±27.4 77.0 ±21.0 85.9 ±17.3 96.0 77.5 ±22.1 ∆KS clinical 17.2 ±26.0 9.0 -1.8 ±21.4 11.5 ±21.1 22.0 ±27.2 58.0 20.6 ±26.9 KS function preop 58.0 ±24.0 60 74.3 ±22.3 55.0 ±20.6 55.8 ±24.5 45.0 56.7 ±27.4 KS function postop 76.2 ±22.9 100 73.3 ±19.7 56.8 ±14.5 80.8 ±22.0 100.0 ±0.0 80.0 ±25.2 ∆KS function 18.1 ±28.8 40 -5.0 ±28.8 1.8 ±13.1 26.3 ±32.5 55.0 23.3 ±21.0 UCLA postop 5.4 ±1.6 7 5.0 ±1.4 4.8 ±1.6 5.8 ±1.6 8.0 ±1.4 4.8 ±1.4 ACL = anterior cruciate ligament; BMI = body mass index; ROM = range of motion; KS = Knee Society; preop = preoperative to arthroscopy; postop = most recent postoperative to arthroscopy
ACCEPTED MANUSCRIPT
Table 2. Frequency of subsequent revision to total knee arthroplasty (TKA) by reason for arthroscopy.
Overall
ACL Tear
RI PT
Characteristic
Reason for Arthroscopy Lateral Loose Cement Arthrofibrosis Compartment Fragment Degeneration
AC C
EP
TE D
M AN U
SC
Number of 58 1 7 11 25 patients/knees Number revised to TKA, any part for 13 (22%) 0 (0%) 2 (29%) 6 (55%) 0 (0%) any reason Relative risk 1.0741 1.3247 3.6623 0.0484 95% CI 0.09 to 12.39 0.37 to 4.78 1.53 to 8.75 0.003 to 0.78 Significance P=0.9543 P=0.6675 P=0.0035 P=0.0326 Odds ratio 1.0988 1.4545 6.8571 0.0298 95% CI 0.04 to 28.57 0.25 to 8.54 1.64 to 28.74 0.002 to 0.53 Significance P=0.9548 P=0.6783 P=0.0085 P=0.0169 ACL = anterior cruciate ligament; TKA = total knee arthroplasty; CI = confidence interval.
Recurrent Hemarthrosis
Synovitis
2
12
1 (50%)
4 (33%)
2.333 0.53 to 10.19 P=0.2598 3.6667 0.21 to 63.03 P=0.3706
1.7037 0.63 to 4.59 P=0.2924 2.0556 0.21 to 8.37 P=0.3145
S0883-5403(18)30515-1
DOI:
10.1016/j.arth.2018.05.031
Reference:
YARTH 56642
To appear in:
The Journal of Arthroplasty
Received Date: 10 April 2017 Revised Date:
16 May 2018
Accepted Date: 21 May 2018
Please cite this article as: Hurst JM, Ranieri R, Berend KR, Morris MJ, Adams JB, Lombardi Jr. AV, Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty, The Journal of Arthroplasty (2018), doi: 10.1016/j.arth.2018.05.031. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.
ACCEPTED MANUSCRIPT
Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty
RI PT
From: Joint Implant Surgeons, Inc., New Albany Ohio USA; White Fence Surgical Suites, New Albany, Ohio USA; and Mount Carmel Health System, Columbus, Ohio Short Title (10 words): Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty Jason M. Hurst, MDa, b, c
SC
Riccardo Ranieria, b, c, d
Keith R. Berend, MDa, b, c, e
M AN U
Michael J. Morris, MDa, b, c Joanne B. Adams, BFAa
Adolph V. Lombardi, Jr., MD, FACSa, b, c, e
a
b
TE D
Joint Implant Surgeons, Inc., 7277 Smith’s Mill Road, Suite 200, New Albany, Ohio USA 43054; (614) 221-6331 White Fence Surgical Suites, 7277 Smith’s Mill Road, Suite, Suite 300, New Albany, Ohio USA 43054; (855) 677-5005
c
d
EP
Mount Carmel Health System, 7333 Smith's Mill Road, New Albany, Ohio USA 43054; (614) 775-6600 Universitá Campus Bio-Medico Roma, Via Álvaro del Portillo, 21, Roma, Italia 00128; (+39) 06-225411
e
AC C
Department of Orthopaedics, The Ohio State University Wexner Medical Center, 376 W 10th Ave, Suite 725, Columbus, Ohio USA 43210; (614) 293-3715 Please direct correspondence to: Jason M. Hurst, MD Joint Implant Surgeons, Inc. 7277 Smith’s Mill Road, Suite 200 New Albany, Ohio USA 43054 Phone: (614) 221-6331 Fax: (614) 221-9042 Email: [email protected]
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
Outcomes after Arthroscopic Evaluation of Patients with Painful Medial Unicompartmental Knee Arthroplasty
AC C
1 2 3
1
ACCEPTED MANUSCRIPT
4
Abstract
5 Introduction: Persistent pain after medial unicompartmental knee arthroplasty (UKA) is a
7
prevailing reason for revision to total knee arthroplasty (TKA). Many of these pathologies can be
8
addressed arthroscopically. The purpose of this study is to examine the outcomes of patients who
9
undergo an arthroscopy for any reason after medial UKA.
SC
RI PT
6
10
Methods: A query of our practice registry revealed 58 patients who had undergone medial UKA
12
between October 2003 and June 2015 with subsequent arthroscopy. Mean interval from UKA to
13
arthroscopy was 22 months (1-101). Indications for arthroscopy were acute anterior cruciate
14
ligament tear (1), arthrofibrosis (7), synovitis (12), recurrent hemarthrosis (2), lateral
15
compartment degeneration including isolated lateral meniscus tears (11), and loose cement
16
fragments (25).
TE D
17
M AN U
11
Results: Mean follow-up after arthroscopy was 49 months (range, 1-143). Twelve patients have
19
been revised from UKA to TKA. Relative risk of revision after arthroscopy for lateral
20
compartment degeneration was 4.27 (6 of 11; 55%; p=0.002) and for retrieval of loose cement
21
fragments was 0.05 (0 of 25; 0%; p=0.03). Relative risk for revision after arthroscopy for
22
anterior cruciate ligament tear, arthrofibrosis, synovitis, or recurrent hemarthrosis did not meet
23
clinical significance secondary to the low number of patients in these categories.
AC C
EP
18
24 25
Conclusions: The results of this study suggest that arthroscopic retrieval of cement fragments
26
does not compromise UKA longevity. However, arthroscopy for lateral compartment 2
ACCEPTED MANUSCRIPT
27
degradation after UKA, while not the cause of revision, appears to be an ineffective treatment
28
and predicts a high risk of revision to TKA regardless of its relative radiographic insignificance.
RI PT
29 Keywords: medial unicompartmental knee arthroplasty, arthroscopy, lateral compartment
31
degeneration, loose cement fragment, pain
AC C
EP
TE D
M AN U
SC
30
3
ACCEPTED MANUSCRIPT
Introduction
33
Medial unicompartmental knee arthroplasty (UKA) is an increasingly utilized orthopedic
34
procedure to treat isolated medial compartment osteoarthritis, which has been reported to occur
35
in as many as 25% of patients with osteoarthritis of the knee [1]. Excellent long-term results
36
comparable to total knee arthroplasty (TKA) have been reported with medial UKA [2-6].
RI PT
32
37
Persistent pain is a common complication after UKA and is also an important reason for revision
39
of unicompartmental implants to TKA [7-9]. However it is suggested that the higher rates of
40
revision for pain in UKA reflect surgeon willingness to advise re-operation of a painful UKA
41
rather than a painful TKA because it is an easier and less hazardous procedure [10]. Furthermore
42
some studies show that revision of UKA to TKA for unexplained pain generally results in a less
43
favorable outcome than revision for a known cause of pain [11, 12].
M AN U
SC
38
TE D
44
Occasionally there are clear clinical issues that cause pain in the knee after UKA, and examples
46
include lateral compartment degeneration, loose bodies, synovitis, recurrent hemarthrosis, and
47
arthrofibrosis. Arthroscopy has been demonstrated to be a useful procedure for treating many of
48
these complications after UKA [13-23], and for diagnosing the etiology of unexplained pain after
49
UKA [14]. However, but there is a paucity of published data which analyze the clinical results of
50
arthroscopy after UKA. The purpose of this study is to examine the outcomes of patients who
51
undergo an arthroscopy after medial UKA.
53
AC C
52
EP
45
Materials and methods
4
ACCEPTED MANUSCRIPT
A query of our practice registry revealed 58 patients who had undergone medial UKA between
55
October 2003 and June 2015 with subsequent arthroscopy. These 58 patients (58 knees) treated
56
with arthroscopy after medial UKA represent 1.34% (58 or 4319) of medial UKA performed in
57
patients at our center during the study period. All patients signed a general research consent
58
approved and monitored by an independent institutional review board (Western IRB, Puyallup,
59
WA) and allowing inclusion of their records in retrospective reviews. No outside funding was
60
received in support of this study. Indications for medial mobile-bearing UKA were anteromedial
61
osteoarthritis or avascular necrosis of the medial condyle with full thickness cartilage loss
62
medially, intact lateral cartilage, intra-articular varus deformity correctible with valgus stress and
63
with functionally intact ligamentous stabilizers. Implant types used for medial UKA (all Zimmer
64
Biomet, Warsaw, IN) were Vanguard M fixed-bearing in 2 knees (3%), Oxford Phase III mobile-
65
bearing in 17 (29%), and Oxford Partial Knee Twin-Peg mobile-bearing in 39 (67%).
66
Medial UKA procedures were performed by 4 different surgeons (AVL, KRB, JMH, MJM), and
67
arthroscopy procedures were performed by two of the surgeons (JMH and KRB).
68
A retrospective database review was performed to evaluate reasons for arthroscopy and
69
outcomes after arthroscopy including need for revision surgery.
SC
M AN U
TE D
EP
70
RI PT
54
Patient demographics, including gender, age, height, weight, body mass index (BMI), and length
72
of follow-up were recorded. Patients were asked to return for clinical evaluation at 6 weeks
73
postoperative and annually thereafter or immediately if a problem with the operative extremity
74
arose. Clinical evaluations included assessment of range of motion (ROM), Knee Society
75
Clinical Rating System (KS) scores [24], the University of California at Los Angeles Activity
76
Scale (UCLA) [25]. The Knee Society pain subcomponent is on a 50-point scale with 0 points
AC C
71
5
ACCEPTED MANUSCRIPT
77
being “severe” pain and 50 points being “none”. All devices used have been approved by the
78
U.S. Food and Drug Administration, and were used according to the manufacturer’s labeling.
79 Results
RI PT
80 81
Mean interval from medial UKA to arthroscopy was 22 months (range, 1-101 months). In the
83
interval between UKA and arthroscopy 23 patients received corticosteroid injections and one
84
patient was treated with manipulation under anesthesia. Indications for arthroscopy were: 1 acute
85
anterior cruciate ligament tear (1.7%) treated with ACL reconstruction; 7 arthrofibrosis (12.1%)
86
treated with lysis of adhesion and/or excision of scar tissue and/or synovectomy; 12 synovitis
87
(20.7%) treated with synovectomy; 2 recurrent hemarthrosis (3.4%) treated with lysis of
88
adhesion and or/ synovectomy; 11 lateral compartment degeneration including isolated lateral
89
meniscus tears (19.0%) treated with chondroplasty and/or debridement and/or shaving and/or
90
meniscectomy; and 25 loose cement fragments (43.1%) treated with removal of loose or foreign
91
body and with partial synovectomy in some cases. The majority of these were clinical diagnoses
92
based on physical examination, including painful synovitis, arthrofibrosis, and hemarthrosis.
93
Loose cement fragments were diagnosed by a combination of acute pain with loose cement
94
visible on radiographs. Five of the 11 patients done for lateral compartment degeneration and 12
95
patients overall were diagnosed with MRI, and one with lateral progression was diagnosed with
96
CT. The remaining patients were diagnosed based on physical examination only. All of the
97
diagnoses that were mentioned were confirmed at the time of arthroscopic evaluation. In regard
98
to lateral compartment degeneration, several patients were documented preoperative to
99
arthroscopy to have meniscal injury based on MRI evaluation. The definition of lateral
AC C
EP
TE D
M AN U
SC
82
6
ACCEPTED MANUSCRIPT
compartment degeneration in this series is any meniscal damage requiring debridement and/or
101
the presence of chondral damage Grade 2 or worse. Valgus stress views were not performed for
102
patients in the current study, but this modality was added subsequently to our radiographic
103
evaluation of patients with painful UKA. MRI findings often times confirmed our clinical
104
suspicion in cases with lateral compartment pathology and ACL insufficiency.
RI PT
100
105
Of the 25 patients who underwent arthroscopy to remove loose fragments, size or general
107
description of the fragments extracted was documented in the dictated operative reports of all
108
patients, ranging from “two very small” in one patient to “10x15mm as the largest of several” in
109
another patient. While we did not routinely measure retrieved fragments at the time of
110
arthroscopy, all fragments were estimated to be greater than 5mm in length.
M AN U
SC
106
111
Patient demographics and outcomes measures by reason for arthroscopy are shown in Table 1. In
113
regard to demographics, patients requiring arthroscopy for treatment of arthrofibrosis were
114
younger (mean 54-years-old) than those patients treated with arthroscopy for removal of loose
115
cement (mean 64-years-old; P=0.003) or lateral compartment degeneration (mean 65-years-old;
116
P=0.027). Preoperatively, patients having arthroscopy for treatment of arthrofibrosis had less
117
pain (KS pain component 32.1) compared with those treated for removal of loose cement (KS
118
pain16.9; P=0.044) or synovitis (KS pain 10.8; P=0.0022), and higher KS clinical scores (78.7)
119
than patients treated for synovitis (57.1; P=0.0020). There were no other differences in patient
120
demographics or preoperative clinical measures between etiology groups.
AC C
EP
TE D
112
121
7
ACCEPTED MANUSCRIPT
Mean follow-up after arthroscopy was 49 months (range, 1-143, SD 34.1). Postoperatively,
123
patients treated for removal of loose bodies had greater improvement in ROM (2.8°) compared
124
with those treated for lateral compartment degeneration (-8.2°, P=0.0164). Patients treated for
125
arthrofibrosis had a deterioration in KS pain component (-2.5), KS clinical score (-1.8), and KS
126
functional score (-5.0) compared to their preoperative levels, with only a slight improvement in
127
ROM (2.5°). This was in contrast to patients treated with arthroscopy for removal of loose bodies
128
(KS pain improvement 21.7, P=0.041; KS functional improvement 22.0, P=0.041), and those
129
treated for synovitis (KS functional improvement 23.3, P=0.0297). Patients undergoing
130
arthroscopy for synovitis also had greater KS functional improvement (23.3) compared with
131
those treated with arthroscopy for lateral compartment degeneration (KS function improvement
132
1.8, P=0.008). Range of motion improvement was greater for patients treated for removal of
133
loose bodies (2.8°) compared with those treated for lateral degeneration (-8.2°; P=0.0164), as
134
was KS functional improvement (26.3 versus 1.8, P=0.023). There were no differences in KS
135
clinical improvement or postoperative UCLA scale between any of the of etiology groups.
SC
M AN U
TE D
136
RI PT
122
Thirteen patients have been revised from UKA to TKA (Table 2). When considering the reasons
138
for arthroscopy and the need for subsequent revision, 2 of 7 patients with arthrofibrosis (28.6%),
139
4 of 12 patients with synovitis (33%), 1 of 2 patients with recurrent hemarthrosis (50%) and 6 of
140
11 patients with lateral compartment degeneration (54.5%) underwent revision to TKA. No
141
patients who received arthroscopic procedure for ACL tear or loose cement fragment underwent
142
revision to TKA. Relative risk of revision to TKA after arthroscopy for lateral compartment
143
degeneration was 3.66 (6 of 11; 55%; 95% CI=1.53-8.75; p=0.0035) and for retrieval of loose
144
cement fragments was 0.05 (0 of 25; 0%; 95% CI=0.003-0.78; p=0.03). The relative risk for
AC C
EP
137
8
ACCEPTED MANUSCRIPT
145
revision after arthroscopy for acute anterior cruciate ligament tear, arthrofibrosis, synovitis, or
146
recurrent hemarthrosis did not meet clinical significance secondary to the low number of patients
147
in these categories.
RI PT
148
Two patients underwent a second arthroscopy procedure: one treated initially for lateral
150
compartment degeneration underwent a second debridement and removal of a chondral loose
151
body, and one treated initially for arthrofibrosis underwent a subsequent diagnostic arthroscopy
152
with synovectomy. One patient treated for arthrofibrosis subsequently underwent an open
153
excision of prepatellar fat pad fibrosis. One patient had pulmonary embolism successfully treated
154
with pharmacological anticoagulation after arthroscopy procedure performed for loose cement
155
fragment removal. One patient treated for synovitis underwent manipulation under anesthesia 2
156
years after arthroscopy.
M AN U
SC
149
TE D
157 Discussion
159
The results of this study suggest that in patients with persistent pain after medial UKA,
160
arthroscopy can be an effective procedure for some but not all underlying causes.
161
In the 25 patients with loose cement fragments after UKA, arthroscopic retrieval of cement does
162
not compromise UKA longevity and has a successful outcome. After a mean follow-up of 44
163
months none of these patients were revised to TKA.
AC C
164
EP
158
165
Even though symptomatic free cement fragments are not a common complication after UKA [9,
166
13, 26] some authors have noted that extrusion of cement and resulting loose cement fragments
167
are a problem in UKA especially with the use of minimal invasive techniques which reduce 9
ACCEPTED MANUSCRIPT
visibility with particular disadvantage for the posterior area of the knee. Particulate debris related
169
to third-body wear, specifically cement particles, is well known in TKA to lead to early failure of
170
ultra-high-molecular-weight polyethylene after TKA [18]. Some authors have reported damage
171
to UKA due to free cements bodies [17, 18]. Berger et al. first described one case of a revision
172
surgery to remove retained cement in UKA [15]. In the literature there are case reports of
173
symptomatic free cement fragments or fixed cement extrusion that were treated successfully with
174
arthroscopy [16, 19, 20]. This study confirms that arthroscopy is an effective procedure to
175
remove free cement and avoid revision. All of the UKA in the current series had cemented
176
fixation, and intuitively, loose cement fragments would not be an issue after cementless UKA.
177
Since we have not treated any patients with cementless UKA devices, we are unable to comment
178
further on the role of arthroscopy in patients with cementless UKA devices.
M AN U
SC
RI PT
168
179
After arthroscopy, our patients did not show a significant improvement of ROM (only 0.6°), but
181
there were substantial improvements in KS pain, clinical and functional scores compared with
182
scores before arthroscopy (respectively mean improvements of 34.3, 17.2 and 18.1 points). This
183
suggests that arthroscopy is not only a good procedure for avoiding revision to TKA in patients
184
with loose cement fragments, but also provides relief of pain and improvement in patient quality
185
of life.
EP
AC C
186
TE D
180
187
Lateral compartment degeneration is one of the major causes of revision for UKA. Lewold et al.
188
[27] reported a 1.4% rate of revision of Oxford UKA for contralateral arthrosis, accounting for
189
20% of all revisions. National registries have shown that progression of arthritis constitutes 25%
190
of the reasons for revision of UKA in Sweden and 33% in New Zealand [8, 28]. In our patients
10
ACCEPTED MANUSCRIPT
who developed lateral compartment degeneration after medial UKA, the use of arthroscopy
192
predicted a high risk of revision to TKA regardless of relative radiographic insignificance. Fifty-
193
five percent of patients underwent revision to TKA within a mean of 37 months regardless of
194
whether or not there was significant joint space narrowing on pre-arthroscopy radiographs.
195
Moreover, after arthroscopy these patients showed a slight mean worsening of ROM (-6.4°)
196
compared to ROM before arthroscopy. However, there were moderate improvements of KS pain,
197
clinical and functional scores (respectively mean improvements of 9, 8.3 and 7 points). These
198
data suggest that arthroscopy in patients with lateral compartment degeneration after medial
199
UKA is not a very effective procedure and does not lead to clinical improvement. A published
200
case report describes a single patient with unexplained pain after UKA where arthroscopy
201
showed severe osteoarthritis regardless of relative radiographic insignificance [29]. This
202
highlights the importance of specialized preoperative radiographs to evaluate the lateral
203
compartment, like posteroanterior (PA) flex or valgus stress view radiographs.
SC
M AN U
TE D
204
RI PT
191
Other indications for arthroscopy after UKA found in this study were for acute anterior cruciate
206
ligament tear, arthrofibrosis, synovitis, or recurrent hemarthrosis. In the literature these
207
complications are very rare causes of revision of UKA to TKA [7]. Our rates of revision to TKA
208
in patients who received arthroscopy for these complications did not demonstrate significance
209
secondary to the low number of patients in these categories. However, patients with synovitis
210
showed significant improvement of KS pain, clinical and functional scores after arthroscopy
211
(respectively mean improvements of 26.8, 27.1 and 16.4 points) and a slight mean improvement
212
of ROM (4°), indicating that these patients can benefit from an arthroscopic procedure. In a case
213
report of two patients in whom persistent pain after UKA was attributed to synovial
AC C
EP
205
11
ACCEPTED MANUSCRIPT
impingement, treatment with arthroscopy provided successful pain relief [23]. For patients with
215
arthrofibrosis after medial UKA, arthroscopy can provide improved ROM (mean improvement
216
was 10°), but KS pain, clinical and functional scores were unchanged after the arthroscopic
217
procedure.
RI PT
214
218
This study has some major limitations. First, the mean time of follow-up is short at only 37
220
months, and longer follow-up should be considered in future studies. Second, the number of
221
patients in different etiology groups (acute anterior cruciate ligament tear, arthrofibrosis,
222
synovitis, and recurrent hemarthrosis) was inadequate for purpose of statistical comparisons.
223
Additional studies are required to investigate the rates of revision after arthroscopy in patients
224
who develop these complications. Third, several patients with lateral degeneration had MRI
225
preoperative to arthroscopy, while others did not. None of the patients had valgus stress
226
radiographs after their UKA prior to arthroscopy, but this additional imaging is currently a
227
routine part of our diagnostic algorithm since undertaking the current review. It is important to
228
thoroughly re-evaluate the lateral compartment with auxiliary radiographic views, such as valgus
229
stress test and posteroanterior (PA) flex views, in the postoperative period after medial UKA.
230
EP
TE D
M AN U
SC
219
Conclusions
232
In literature there are studies which mention arthroscopy as a potential useful procedure in order
233
to treat some painful complications after UKA, but this is the first study which specifically
234
reports the outcome of a series of patients who underwent arthroscopy after UKA for multiple
235
reasons. The results of this study suggest that arthroscopic retrieval of cement fragments does not
236
compromise UKA longevity. However, arthroscopy for lateral compartment degradation after
AC C
231
12
ACCEPTED MANUSCRIPT
237
UKA predicts a high risk of revision to TKA regardless of its relative radiographic
238
insignificance. While the arthroscopy does not cause the revision, it appears to be an ineffective
239
treatment for arthritic progression after UKA.
RI PT
240
AC C
EP
TE D
M AN U
SC
241
13
ACCEPTED MANUSCRIPT
242
References
243
1.
245
2003;85(7):937-942. 2.
246
Price AJ, Svärd U. A second decade lifetable survival analysis of the Oxford
unicompartmental knee arthroplasty. Clin Orthop Relat Res 2011;469:174-9. 3.
Berger RA, Meneghini M, Jacobs JJ, Sheinkop MB, Della Valle CJ, Rosenberg AG,
SC
247
RI PT
244
Ackroyd CE. Medial compartment arthroplasty of the knee. J Bone Joint Surg Br.
Galante JO. Results of unicompartmental knee arthroplasty at a minimum of ten years of
249
follow-up. J Bone Joint Surg Am. 2005 May;87(5):999-1006.
250
4.
251 252
M AN U
248
Svärd UCG, Price AJ. Oxford medial unicompartmental knee arthroplasty: a survival analysis of an independent series. J Bone Joint Surg Br 2001;83:191-4.
5.
O'Rourke MR, Gardner JJ, Callaghan JJ, Liu SS, Goetz DD, Vittetoe DA, Sullivan PM, Johnston RC. The John Insall Award: unicompartmental knee replacement: a minimum
254
twenty-one-year followup, end-result study. Clin Orthop Relat Res 2005;440:27-37.
255
6.
TE D
253
Zuiderbaan HA, van der List JP, Khamaisy S, Nawabi DH, Thein R, Ishmael C, Paul S, Pearle AD. Unicompartmental knee arthroplasty versus total knee arthroplasty: Which type
257
of artificial joint do patients forget? Knee Surg Sports Traumatol Arthrosc.2015 Nov 21. 7.
National Joint Registry for England and Wales. 10th Annual Report 2013.
AC C
258
EP
256
259
http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/Reports/10th_annual_r
260
eport/NJR%2010th%20Annual%20Report%202013%20B.pdf Accessed 2013 June
261
8.
New Zealand Orthopaedic Association. The New Zealand Joint Registry. Thirteen year
262
report: January 1999 – December 2011.
263
http://nzoa.org.nz/system/files/NJR%2013%20Year%20Report.pdf. Accessed 2012 Oct.
14
ACCEPTED MANUSCRIPT
264
9.
Baker PN, Petheram T, Avery PJ, Gregg PJ, Deehan DJ. Revision for unexplained pain
265
following unicompartmental and total knee replacement. J Bone Joint Surg Am. 2012 Sep
266
5;94(17):e126. 10.
Goodfellow JW, O’Connor JJ, Murray DW. A critique of revision rate as an outcome
RI PT
267 268
measure. Re-interpretation of knee joint registry data. J Bone Joint Surg Br.
269
2010;92(12):1628-31. 11.
Kerens B, Boonen B, Schotanus MG, Lacroix H, Emans PJ, Kort NP. Revision from
SC
270
unicompartmental to total knee replacement: the clinical outcome depends on reason for
272
revision. Bone Joint J. 2013 Sep; 95-B(9):1204-8.
273
12.
M AN U
271
Châtain F, Richard A, Deschamps G, Chambat P, Neyret P. [Revision total knee
274
arthroplasty after unicompartmental femorotibial prosthesis: 54 cases]. Rev Chir Orthop
275
Reparatrice Appar Mot. 2004 Feb;90(1):49-57. 13.
277 278
Kim KT, Lee S, Lee JI, Kim JW. Analysis and treatment of complications after
TE D
276
unicompartmental knee arthroplasty. Knee Surg Relat Res. 2016 Mar;28(1):46-54. 14.
Hannaoui S, Lustig S, Servien E, Aït Si Selmi T, Neyret P. Arthroscopy of the knee after unicompartmental arthroplasty. Rev Chir Orthop Reparatrice Appar Mot. 2008
280
Nov;94(7):678-84. 15.
Berger RA, Nedeff DD, Barden RM, Sheinkop MM, Jacobs JJ, Rosenberg AG, Galante JO.
AC C
281
EP
279
282
Unicompartmental knee arthroplasty. Clinical experience at 6- to 10-year followup. Clin
283
Orthop. 1999;367:50-60.
284 285
16.
Howe DJ, Taunton OD Jr, Engh GA. Retained cement after unicondylar knee arthroplasty. A report of four cases. J Bone Joint Surg Am. 2004 Oct;86-A(10):2283-6.
15
ACCEPTED MANUSCRIPT
286
17.
Elmadağ M, Imren Y, Erdil M, Bilsel K, Tuncay I. Excess retained cement in the
287
posteromedial compartment after unicondylar knee arthroplasty. Acta Orthop Traumatol
288
Turc 2013;47(4):291-294. 18.
Hauptmann SM, Weber P, Glaser C, Birkenmaier C, Jansson V, Müller PE. Free bone
RI PT
289 290
cement fragments after minimally invasive unicompartmental knee arthroplasty: an
291
underappreciated problem. Knee Surg Sports Traumatol Arthrosc. 2008 Aug;16(8):770-5. 19.
Karataglis D, Agathangelidis F, Papadopoulos P, Petsatodis G, Christodoulou A.
SC
292
Arthroscopic removal of impinging cement after unicompartmental knee arthroplasty.
294
Hippokratia. 2012 Jan;16(1):76-9.
295
20.
M AN U
293
Kim WY, Shafi M, Kim YY, Kim JY, Cho YK, Han CW. Posteromedial compartment
296
cement extrusion after unicompartmental knee arthroplasty treated by arthroscopy: a case
297
report. Knee Surg Sports Traumatol Arthrosc. 2006 Jan;14(1):46-9. 21.
299 300
Yamakado K. Dissecting a popliteal cyst after failed unicompartmental knee arthroplasty.
TE D
298
Arthroscopy. 2002 Nov-Dec;18(9):1024-8. 22.
Yamakado K, Arakawa H, Hayashi S. Arthroscopic observation was useful to detect loosening of the femoral component of unicompartmental knee arthroplasty in a recurrent
302
hemoarthrosis. Sports Med Arthrosc Rehabil Ther Technol. 2012 Feb 21;4:8. 23.
304 305 306
Jung KA, Lee SC, Hwang SH. Pseudomeniscal synovial impingement after unicondylar
AC C
303
EP
301
knee arthroplasty. Orthopedics. 2009 May;32(5):361. 24.
Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res. 1989 Nov;(248):13-4.
16
ACCEPTED MANUSCRIPT
307
25.
Amstutz HC, Thomas BJ, Jinnah R, Kim W, Grogan T, Yale C. Treatment of primary
308
osteoarthritis of the hip. A comparison of total joint and surface replacement arthroplasty. J
309
Bone Joint Surg Am. 1984 Feb;66(2):228-41. 26.
Lewold S, Robertsson O, Knutson K, Lidgren L. Revision of unicompartmental knee
RI PT
310 311
arthroplasty: outcome in 1,135 cases from the Swedish Knee Arthroplasty study. Acta
312
Orthopedica. 1998 Oct;69(5):469-74. 27.
Lewold S, Goodman S, Knutson K, Robertsson O, Lidgren L. Oxford meniscal bearing
SC
313
knee versus the Marmor knee in unicompartmental arthroplasty for arthrosis. A Swedish
315
multicenter survival study. J Arthroplasty. 1995 Dec;10(6):722-31.
316
28.
317 318
M AN U
314
Lidgren L, Knutson K, Robertsson O. Swedish Knee Arthroplasty Register: Annual Report 2004. Lund: Swedish Knee Arthroplasty Register, 2004.
29.
Hama S, Hamada D, Goto T, Tsutsui T, Tonogai I, Suzue N, Matsuura T, Sairyo K. Revision total knee arthroplasty for unexplained pain after unicompartmental knee
320
arthroplasty: a case report. J Med Invest. 2015;62(3-4):261-3.
EP AC C
321
TE D
319
17
ACCEPTED MANUSCRIPT
Table 1. Patient demographics and outcomes measures by reason for arthroscopy.
Overall
ACL Tear
RI PT
Characteristic
Reason for Arthroscopy Lateral Loose Arthrofibrosis Compartment Cement Degeneration Fragment
Recurrent Hemarthrosis
Synovitis
AC C
EP
TE D
M AN U
SC
Number of 58 1 7 11 25 2 12 patients/knees Males:Females 29:29 1:0 4:3 5:6 14:11 2:0 3:9 Age (years) 61.7 ±8.8 58.7 53.9 ±8.3 64.7 ±9.7 64.1 ±7.2 58.2 ±3.0 59.2 ±9.8 Height (inches) 67.6 ±4.9 70 69.9 ±5.4 66.8 ±5.5 67.8 ±4.8 71.0 ±5.7 65.7 ±4.5 Weight (pounds) 207.6 ±45.4 255 205.7 ±52.0 200.4 ±57.8 217.2 ±43.4 209.5 ±50.2 191.1 ±32.6 BMI (kg/m2) 31.9 ±6.0 36.6 29.7 ±7.9 31.1 ±5.9 33.3 ±6.3 28.9 ±2.4 31.1 ±4.5 ROM preop 114.7 ±12.1 125 114.3 ±11.7 118.2 ±7.2 115.2 ±10.2 92.5 ±24.7 113.8 ±15.4 ROM postop 115.4 ±10.1 120 115.8 ±13.2 110.0 ±11.0 118.7 ±8.6 110.0 ±7.1 113.8 ±1.1 ∆ROM 0.6 ±12.7 -5.0 2.5 ±10.4 -8.2 ±13.2 2.8 ±11.2 17.5 ±17.7 1.1 ±12.7 KS pain preop 17.9 ±17.1 45 32.1 ±15.0 17.7 ±18.6 16.9 ±17.3 0.0 10.8 ±10.8 KS pain postop 34.3 ±19.4 50 28.3 ±24.8 30.0 ±20.2 37.0 ±18.1 50.0 ±0.0 31.7 ±20.3 ∆KS pain 17.2 ±24.5 5.0 -2.5 ±30.0 12.3 ±20.7 21.7 ±24.7 50.0 20.8 ±24.4 KS clinical preop 64.5 ±18.7 90 78.7 ±14.6 65.5 ±19.8 63.6 ±20.0 38.0 57.1 ±11.2 KS clinical postop 81.6 ±20.2 99 75.2 ±27.4 77.0 ±21.0 85.9 ±17.3 96.0 77.5 ±22.1 ∆KS clinical 17.2 ±26.0 9.0 -1.8 ±21.4 11.5 ±21.1 22.0 ±27.2 58.0 20.6 ±26.9 KS function preop 58.0 ±24.0 60 74.3 ±22.3 55.0 ±20.6 55.8 ±24.5 45.0 56.7 ±27.4 KS function postop 76.2 ±22.9 100 73.3 ±19.7 56.8 ±14.5 80.8 ±22.0 100.0 ±0.0 80.0 ±25.2 ∆KS function 18.1 ±28.8 40 -5.0 ±28.8 1.8 ±13.1 26.3 ±32.5 55.0 23.3 ±21.0 UCLA postop 5.4 ±1.6 7 5.0 ±1.4 4.8 ±1.6 5.8 ±1.6 8.0 ±1.4 4.8 ±1.4 ACL = anterior cruciate ligament; BMI = body mass index; ROM = range of motion; KS = Knee Society; preop = preoperative to arthroscopy; postop = most recent postoperative to arthroscopy
ACCEPTED MANUSCRIPT
Table 2. Frequency of subsequent revision to total knee arthroplasty (TKA) by reason for arthroscopy.
Overall
ACL Tear
RI PT
Characteristic
Reason for Arthroscopy Lateral Loose Cement Arthrofibrosis Compartment Fragment Degeneration
AC C
EP
TE D
M AN U
SC
Number of 58 1 7 11 25 patients/knees Number revised to TKA, any part for 13 (22%) 0 (0%) 2 (29%) 6 (55%) 0 (0%) any reason Relative risk 1.0741 1.3247 3.6623 0.0484 95% CI 0.09 to 12.39 0.37 to 4.78 1.53 to 8.75 0.003 to 0.78 Significance P=0.9543 P=0.6675 P=0.0035 P=0.0326 Odds ratio 1.0988 1.4545 6.8571 0.0298 95% CI 0.04 to 28.57 0.25 to 8.54 1.64 to 28.74 0.002 to 0.53 Significance P=0.9548 P=0.6783 P=0.0085 P=0.0169 ACL = anterior cruciate ligament; TKA = total knee arthroplasty; CI = confidence interval.
Recurrent Hemarthrosis
Synovitis
2
12
1 (50%)
4 (33%)
2.333 0.53 to 10.19 P=0.2598 3.6667 0.21 to 63.03 P=0.3706
1.7037 0.63 to 4.59 P=0.2924 2.0556 0.21 to 8.37 P=0.3145