Page 1
1
Bone marrow lesions, subchondral bone cysts and subchondral bone attrition are associated with histological synovitis in patients with
end-stage knee osteoarthritis: A cross-sectional study
Authors: Anwarjan Yusup1,2, Haruka Kaneko1, Lizu Liu1,3, Liang Ning2, Ryo Sadatsuki1, Shinnosuke Hada1, Koji Kamagata4, Mayuko Kinoshita1, Ippei Futami1, Yukio Shimura5, Masaru Tsuchiya5, Yoshitomo Saita1, Yuji Takazawa1, Hiroshi Ikeda1, Shigeki Aoki4, Kazuo Kaneko1,3 and Muneaki Ishijima1,3#
Affiliations: 1 Department of Medicine for Orthopaedics and Motor Organ, Juntendo University Graduate School of Medicine, Tokyo, Japan
2 Research Institute for Diseases of Old Age, Juntendo University Graduate School of Medicine, Tokyo, Japan 3 Sportology Center, Juntendo University Graduate School of Medicine, Tokyo, Japan 4 Department of Radiology, Juntendo University Graduate School of Medicine, Tokyo, Japan 5Department of Orthopedics, Juntendo Tokyo Koto Geriatric Medical Center, Tokyo, Japan
E-mail addresses: AY; [email protected] , HK; [email protected] , LL; [email protected] , LN; [email protected] , RS; [email protected] , SH, [email protected] , KOK, [email protected] , MK; [email protected] , IF; [email protected] , YS; [email protected] , MT; [email protected] , YS; [email protected] , YT; [email protected] , HI; [email protected] , SA, [email protected] , KAK; [email protected] , MI; [email protected] # Corresponding author: Muneaki Ishijima, M.D., Ph.D. Associate Professor, Department of Medicine for Orthopaedics and Motor Organ Juntendo University Graduate School of Medicine 2-1-1, Hongo, Bunkyo-ku, Tokyo 113-8421, Japan Tel; +81-3-3813-3111, Fax; +81-3-3813-3428 E-mail; [email protected]
Page 2
2
Abstract 1
Objective 2
The aim of this study was to examine the osteoarthritis (OA)-related structural changes 3
associated with histological synovitis in end-stage knee OA patients. 4
Methods 5
Forty end-stage knee OA patients (female: 88%, mean age: 71.8 y) were enrolled. All 6
participants underwent 3.0-Tesla MRI. The structural changes, such as cartilage morphology, 7
subchondral bone marrow lesion (BML), subchondral bone cyst (SBC), subchondral bone 8
attrition (SBA), osteophytes, meniscal lesion and synovitis, were scored using the 9
whole-organ MRI scoring (WORMS) method. Synovial samples were obtained from five 10
regions of interest of the knee joint during total joint replacement surgery. The associations 11
between the histological synovitis score (HSS) and WORMS or the synovial expression levels 12
of cyclooxygenase (COX)-2, interleukin (IL)-1β, IL-6 and transforming growth factor (TGF)–13
β were examined using Spearman’s correlation coefficient. 14
Results 15
Among the seven OA-related structural changes, the BML, SBC, SBA and synovitis were 16
significantly associated with the HSS (r = 0.33, 0.35, 0.48 and 0.36, respectively), while other 17
morphological changes were not. Although synovial COX-2, IL-1β or IL-6 expression levels 18
were not associated with the HSS, the synovial TGF-β expression levels were associated with 19
Page 3
3
the HSS. 20
Conclusion 21
The presence of BML, SBC and SBA was associated with histological synovitis in end-stage 22
knee OA patients. 23
24
Page 4
4
Introduction 25
There are no disease-modifying osteoarthritis (OA) drugs available for treating knee OA at 26
present, and symptom-modifying therapy is the only available treatment for knee OA [1-3]. 27
Despite the importance of the symptoms of the disease, much remains unknown regarding the 28
nature, causes and natural history of OA symptoms. 29
Pain is the most prominent and disabling symptom of knee OA [4, 5]. Because 30
OA-associated joint damage may be associated with clinical problems, the joint damage 31
associated with the pain and the pathogenesis of pain must be investigated [6]. However, the 32
severity of joint disease is only weakly related to the clinical findings according to classic 33
radiographic techniques [2]. 34
Although OA was considered to be a non-inflammatory condition, the role of synovitis in 35
OA has attracted particular attention [7, 8]. It was recently reported that synovial 36
inflammation could play an important role in the pathophysiology of OA [9-13]. Synovitis in 37
OA may be a secondary phenomenon related to the cartilage alterations induced by the release 38
of degenerative compounds from the extracellular matrix of articular cartilage in response to 39
the presence of microcrystals in the synovial fluid and in the synovium [11, 14]. However, 40
most of the previous studies focused on synovitis in OA were conducted on patients with 41
early- to advanced- stage knee OA. 42
We previously revealed that the symptoms of patients with end-stage knee OA who required 43
Page 5
5
total knee arthroplasty (TKA) showed a significant correlation with the severity of synovitis 44
in the affected knee joint [15]. On the other hand, no joint structural changes evaluated by 45
classic radiography were associated with the pain and symptoms of these patients. In addition, 46
it has remained unclear whether synovitis is also related to cartilage alterations or other 47
features in patients with end-stage knee OA, where the articular cartilage has been nearly lost. 48
We also previously reported that the cytokine profiles of the synovium in patients with 49
end-stage knee OA were different from those in patients with early-stage knee OA [13]. 50
The interest in developing other treatments for OA has stimulated the search for more 51
sensitive indicators of OA for use in conjunction with the traditional radiographic outcomes. 52
Magnetic resonance imaging (MRI) measurements, in addition to biomarkers, have sufficient 53
sensitivity to detect OA-related structural changes [16, 17]. MRI is currently being optimized 54
for OA imaging and is more sensitive than radiographic techniques to detect bone and soft 55
tissue changes, which are features of OA [18]. MRI can also provide a wealth of information 56
on the pathology of knee OA, its natural history and the structure-pain relationships, which is 57
not obtainable using radiography [19]. The semi-quantitative whole-organ MRI scoring 58
(WORMS) method offers an initial instrument for performing multi-feature assessments of 59
the knee using conventional MRI [20]. 60
The aim of our present study was to investigate whether the MRI-detected structural changes 61
of the knee joint were associated with histological synovitis in patients with end-stage knee 62
Page 6
6
OA who underwent TKA. 63
64
Page 7
7
Patient and methods 65
Subjects 66
Forty patients who fulfilled the American College of Rheumatology criteria for knee OA 67
[21] and who received TKA were enrolled in this study. This study was approved by the 68
institutional ethical review committee of our university. Written informed consent was 69
obtained from all participating patients. 70
71
Clinical manifestations 72
The clinical manifestations were evaluated using the Japanese Knee Osteoarthritis Measure 73
(JKOM) score [22] and the pain was evaluated using the visual analog scale (VAS, 0–100). 74
The JKOM is a patient-based, self-answered evaluation score that includes four subcategories: 75
pain and stiffness, activities of daily living, social activities, and general health conditions 76
with 100 points as the maximum score. The JKOM score is higher in patients with more pain 77
and physical disabilities, and this evaluation modality is considered to have sufficient 78
reliability and validity for studies of the clinical outcomes of Japanese people with knee OA. 79
80
Radiographic evaluation of knee OA 81
The standing, extended and anteroposterior and lateral view radiographs and the 82
posteroanterior weight-bearing radiograph made with the knee in 45° of flexion 83
Page 8
8
(Rosenberg radiograph) were taken at the admission for the operation [23, 24]. In addition to 84
the evaluation of the Kellgren and Lawrence (K/L) grade [25] and the femorotibial angle 85
(FTA), the joint space width (JSW) was determined at the center point of the medial 86
femorotibial compartment on a radiograph. All radiographs were quantified independently by 87
two readers (AY, LL and RS) who were blinded to the baseline characteristics of the patients. 88
89
MRI-based evaluation of knee OA 90
All patients showed a K/L grade 4 [25] and were also examined with the MAGNETOM Verio 91
MR 3.0-Tesla MRI system (Siemens Medical Solutions, Erlangen, Germany) according to the 92
previously reported method [17]. The knee was scored using the WORMS method [20]. 93
Specifically, three regions (anterior, central and posterior) of the medial and lateral femoral 94
condyles and tibial plateaus, and two regions (medial and lateral) of the patella were each 95
scored separately for the cartilage morphology, bone marrow lesion (BML), subchondral bone 96
cyst (SBC), subchondral bone attrition (SBA) and osteophytes. Each region of a compartment 97
surface received its own score, following the method reported previously [17, 20]. At each 98
region, the cartilage morphology was given a score of 0-6. The BML and SBC were given a 99
score of 0-3. The SBA was scored 0-3. Osteophytes were scored 0-7. The anterior horn, 100
posterior horn, and body of the medial and lateral meniscus were each graded 0-4. A 101
cumulative grade for each meniscus was also determined using a score of 0-6. No intravenous 102
Page 9
9
contrast was injected in this study, which thus precluded us from differentiating synovial 103
thickening and joint effusion. Thickening and effusion were therefore graded collectively as 104
per the WORMS protocol from with a score of 0-3 for synovitis [20]. 105
The intra-observer reproducibility (AY) of the MRI evaluation of OA was measured at 106
separate times for twenty patients [inter-class correlation coefficient (ICC): 0.81 (95 % CI: 107
0.79 - 0.84)]. AY and RS independently conducted the MRI evaluation. To confirm their 108
accuracy for the MRI evaluation, the musculoskeletal radiologist (KOK) also independently 109
conducted the MRI evaluation. The inter-observer reproducibility was measured between 110
these three observers (AY, RS and KOK) who conducted 20 examinations [ICC between AY 111
and RS: 0.71 (95 % CI: 0.68 - 0.75), ICC between AY and KOK: 0.72 (95 % CI: 0.68 - 0.75) 112
and ICC between RS and KOK: 0.69 (95 % CI: 0.61 - 0.75)]. 113
114
Histological examination: 115
Sample preparation 116
Synovial tissue samples were obtained from the patients at the time of the operation from 117
five regions of interest (ROIs) [26]. They included three in the suprapatellar recess [lateral 118
recess (ROI 1), medial recess (ROI 2)], and just above the trochlear groove (ROI 5), and one 119
each in the lateral and medial distal femoral gutters (ROIs 3 and 4, respectively). The synovial 120
tissue samples were fixed in 10% neutral buffered formalin, and subsequently processed by 121
Page 10
10
standard histological techniques, followed by mounting in paraffin blocks for sectioning. The 122
5-µm paraffin sections were stained with hematoxylin and eosin for a microscopic analysis 123
[13, 15]. The staining procedures were performed consecutively at one time for all the 124
sections. Ten sections were randomly selected from 30 sections per sample from one ROI. For 125
one section, five examination fields (EFs) which included synovial articular surfaces were 126
randomly selected using a special eye-piece with a 10×10-grid frame. Six inflammatory 127
parameters described below were studied in each of the EFs under 200x magnification [26]. 128
129
The histological parameters 130
The six histological parameters were: (1) the thickness of the synovial lining layer, (2) 131
subsynovial infiltration by lymphocytes and plasma cells, (3) surface fibrin deposition, (4) 132
blood vessel vasodilation and blood vessel proliferation, (5) fibrosis, and (6) perivascular 133
edema. The parameters were graded as follows: 0, none; 1, mild; 2, moderate; and 3, severe. 134
Grade 0 corresponded to normal synovial tissue, and grade 3 corresponded to the most severe 135
pattern observed in the OA samples. 136
The histological examiner (AY) was blinded to all data. The mean of each of the 137
parameters for five EFs was used to represent the section. The procedures were undertaken 138
for 10 sections for each parameter, and the mean of each of these 10 sections was used to 139
represent each ROI. The procedures were repeated for the five ROI specimens. The mean of 140
Page 11
11
each parameter from the five ROIs was considered to represent the histological parameters of 141
the patients. The average of the six histological parameters was calculated to give a mean total 142
histological synovitis score (HSS) for the patient. 143
144
Immunohistochemical staining 145
Similar to chondrocytes, the synovial cells of OA patients produce inflammatory cytokines, 146
such as interleukin (IL)-1β, IL-6 and tumor necrosis factor (TNF)-α, which in turn decrease 147
anabolic collagen synthesis and increase catabolic mediators, such as matrix 148
metalloproteinase (MMP)-1 and MMP-13, via their subsequent intracellular signaling through 149
nuclear factor kappa B (NF-κB) and cyclooxygenase (COX)-2 [27]. Therefore, among these 150
molecules, IL-1β, COX-2 and IL-6 expression levels in the synovium were examined in the 151
present study. We also examined the synovial expression levels of transforming growth factor 152
(TGF)-β, which is also known to be expressed in the synovium in knee OA [7]. The 5-µm 153
paraffin sections (the same samples used for HE staining) were stained for TGF-β (TB21, 154
AbD Serotec, Kidlington; 1:5,000 dilution) [28], COX-2 (33/COX-2, BD Biosciences, 155
Franklin Lakes, USA; 1:50 dilution) [10], IL-1β (MAB601; R&D Systems, Wiesbaden, 156
Germany; 1:20 dilution) [28] and IL-6 (MAB2061; R&D Systems, Wiesbaden, Germany; 157
1:300 dilution) [29]. A semi-quantitative analysis of the stained sections was conducted, as 158
previously reported [13, 30]. The number of positively-stained cells was estimated in ten 159
Page 12
12
high-powered fields (400x) chosen at random. In each high-powered field, three histological 160
findings of the synovium, including the vasculature, synovial sublining layer and the synovial 161
lining layer, were scored separately according to the scoring system. Each section was scored 162
on a scale from 0-3 to reflect the degree of specific staining as follows: 0 represented 0-5% 163
positive staining, 1 represented 6-29%, 2 indicated 30-59% and 3 indicated ≥60% staining. 164
The cumulative staining scores were calculated by summing the mean score of these four 165
parameters for each case. 166
The intra-observer reproducibility (AY) of the histological scores was measured at separate 167
times for ten sections [ICC: 0.97 (95% CI: 0.94 - 0.98)]. The inter-observer reproducibility 168
was measured by two observers [AY and LL for HE, AY and LN for immunohistochemistry] 169
who conducted 10 examinations ([ICC: 0.94 (95% CI: 0.89 - 0.97)] and [ICC: 0.93 (95% CI: 170
0.86 - 0.96)], respectively). 171
172
Statistical analysis 173
Spearman’s correlation coefficient was used to investigate the correlation of either the 174
MRI-detected OA-related joint structural changes or the inflammatory cytokines and growth 175
factor with histological synovitis in the patients. A p-value < 0.05 was considered to be 176
statistically significant. The SPSS 19.0 software program (SPSS Institute, Chicago, IL, USA) 177
was used for the statistical analysis. 178
Page 14
14
Results 180
Patient characteristics 181
The characteristics of the patients in the present study are shown in Table 1. All patients 182
had medial type knee OA with a K/L grade of 4 for the radiographic OA severity, and most of 183
the patients were female (88%). The HSSs of the patients were significantly associated with 184
the symptoms evaluated by the JKOM score (r = 0.345, p = 0.03), consistent with the findings 185
of a previous study [15]. 186
187
The MRI-detected joint structural changes associated with synovitis in the patients 188
Among the OA joint changes evaluated by the WORMS analysis, the BML score, SBC 189
score, SBA score and synovitis scores were associated with the HSSs of the patients (Figure 190
1), while other joint structural changes, such as the cartilage morphology score, osteophyte 191
score and meniscal pathology scores, were not (Table 2). 192
193
Associations between synovitis and the synovial expression levels of inflammatory 194
mediators and growth factors in the patients 195
No associations were observed between the synovial IL-1β expression levels and the HSSs 196
of the patients (Table 3). Similarly, neither synovial COX-2 expression levels nor synovial 197
IL-6 expression levels were associated with the HSSs of the patients (Table 3). Although the 198
Page 15
15
TGF-β expression in the synovial lining and sublining layers were not associated with the 199
HSSs of the patients, the TGF-β expression in the vasculature of the synovium were 200
associated with the HSSs of the patients (Table 3). 201
202
Page 16
16
Discussion 203
We previously reported that synovitis in the knee joint was associated with the symptoms in 204
patients with end-stage knee OA who received TKA [15]. However, the OA-related structural 205
changes associated with synovitis have remained unclear. In the present study, we examined 206
the OA-related structural changes using MRI to determine which changes were associated 207
with histological synovitis in patients with end-stage knee OA who underwent TKA. We 208
found that the presence of subchondral pathologies, such as BML, SBC and SBA, in addition 209
to MRI-detected synovitis, was significantly associated with the presence of histological 210
synovitis in patients with end-stage knee OA. The inflammatory mediators known to be 211
involved in the pathophysiology of OA [7], such as COX-2, IL-1β and IL-6, were not 212
associated with the histological synovitis in patients with end-stage knee OA. On the other 213
hand, the synovial expression of TGF-β, which is a growth factor known to be involved in the 214
pathophysiology of OA [31], was associated with histological synovitis in patients with 215
end-stage knee OA. Our study demonstrated the associations between BML, SBC and SBA, 216
synovial TGF-β expression and histological synovitis in patients with end-stage knee OA, 217
which were not previously well established in the literature. 218
219
Because synovitis in knee OA occurs locally in areas of cartilage loss [11], the medial type 220
of knee OA shows medial local synovitis, which is associated with cartilage degradation. In 221
Page 17
17
the medial type of knee OA, the number of anti-type II collagen-positive fragments, which is 222
one of the causes of synovitis in the medial compartment of the knee, was higher than that 223
found in the lateral compartment [32]. In the present study, the presence of subchondral 224
pathologies, such as BML, SBC and SBA, was associated with synovitis. These data suggest 225
that synovitis is present not only in early-stage but also end-stage knee OA and is possibly 226
induced by not only articular cartilage destruction but also, at least in part, due to the 227
subchondral pathologies. 228
229
We previously reported that the factors associated with the pain severity in patients with 230
knee OA varied with the radiographic disease severity [33]. The serum levels of IL-6, which 231
is related to synovitis, were mainly associated with the pain severity in early-stage knee OA, 232
while the anatomical axis angle, which represents the detrimental mechanical loading across 233
the joint, was mainly associated with that in advanced- to end-stage knee OA. 234
As the expression of COX-2, IL-1β and IL-6 in synovial tissues is thought to be activated by 235
degenerated articular cartilage, the expression of these inflammatory mediators would be 236
expected to be enhanced in the synovial tissues dependent on the severity of the disease. 237
However, the expression levels of these cytokines in patients with end-stage knee OA were 238
decreased in comparison to those in patients with early- to advanced-stage knee OA in 239
previous studies [12, 13]. In contrast, the expression levels of TGF-β in the synovium were 240
Page 18
18
increased depending on the severity of the disease in patients with the medial type of knee OA 241
[13]. 242
243
BMLs are suggested to be one of the pathological features related to detrimental mechanical 244
loading across the joint [18, 34-36]. Prevalent and incident SBA is strongly associated with 245
BMLs in the same subregion [37]. BML has also been revealed to be associated with the 246
progression of OA [18, 34, 36, 38]. The enlargement of BML was associated with the 247
progression of the disease [34]. The malalignment of the lower limbs was associated with 248
both the incident risk and the enlargement of BML in the more loaded compartments of the 249
knee joint [35]. As all patients in the present study showed the medial type of knee OA, 250
BMLs were observed in the medial compartments of the knee joint in all patients who showed 251
varus lower-limb alignment (FTA: 184.9º on average). Although COX-2, IL-1β and IL-6 may 252
be produced by and/or related to the cartilage destruction in knee OA, limited articular 253
cartilage remained and the subchondral bone showed eburnation, especially in the medial 254
compartment of the knee joint of patients with end-stage medial type of knee OA in the 255
present study. Consistent with this phenomenon, the HSSs were associated with the 256
subchondral pathologies, such as BML, SBC and SBA, and the TGF-β expression levels in 257
the vasculature of the synovium in the patients, suggesting the presence of a relationship 258
between histological synovitis and subchondral pathologies, and presumably synovial TGF-β 259
Page 19
19
expression in patients with end-stage knee OA. 260
However, there are still many unsolved questions remaining, such as how to explain the 261
relationship between synovial TGF-β and subchondral pathologies, such as BML, SBC and 262
SBA, whether TGF-β is released when subchondral pathologies develop in subchondral 263
lesions, and if so, how TGF-β in subchondral lesions is transferred to the joint, i.e., directly or 264
indirectly (such as via the blood flow), or, if not, how BMLs in subchondral lesions induce 265
TGF-β expression in the synovium. In addition, whether TGF-β expressed in the synovium in 266
patients with end-stage knee OA is activated remains unclear. However, TGF-β is the present 267
studies suggest that TGF-β is involved in the association between synovitis and the 268
subchondral pathologies in patients with end-stage knee OA, although further investigation is 269
required. 270
271
There were several limitations associated with the current study. As the present study is 272
cross-sectional, the causal relationship between the MRI-detected OA-related structural 273
changes and histological synovitis is still obscure. In this context, the results obtained in this 274
study may be understood to be indirect evidence. The MRI-based WORMS assessment was 275
performed by non-expert readers with limited experience in the semi-quantitative MRI 276
analysis of knee OA features. To overcome this limitation, we calculated the inter-observer 277
reproducibility for the MRI evaluation between the observers and the musculoskeletal 278
Page 20
20
radiologist, as described in the Method section. 279
280
In conclusion, the presence of BML, SBC and SBA, but not other MRI-detected structural 281
changes, was associated with local histological synovitis in end-stage knee OA patients who 282
received TKA. 283
284
Page 21
21
Acknowledgement 285
The authors wish to thank Dr. H. Kurosawa for giving us a chance to conduct this study. The 286
authors would also like to thank Dr. K. Hayashi for his helpful comments and suggestions in a 287
statistical analysis and Dr. T. Sueyoshi for technical assistance. This study was funded in part 288
by a High Technology Research Center Grant from the Ministry of Education, Culture, Sports, 289
Science and Technology, Japan (to MI and KAK). 290
291
Page 22
22
Contributions 292
AY, HK and MI conceived and designed the study. HK, RS, SH, KOK, MK, IF, YS, MT, YS, 293
YT, HI and MI collected and registered patients data. AY, HK, LL, LN, RS, SH, MK, YS, MT, 294
SA, KAK and MI had the major role in analysis and interpretation of the data, and contributed 295
to drafting the report. KAK also supervised the statistical analysis. All authors have read and 296
approved the final manuscript. 297
298
Page 23
23
Role of the funding source 299
This study was funded in part by a High Technology Research Center Grant from the Ministry 300
of Education, Culture, Sports, Science and Technology, Japan (to MI and KAK). 301
302
Page 24
24
Competing interests 303
All authors declare that they have no competing interests. 304
305
Page 25
25
References 306
1. Felson DT. Clinical practice. Osteoarthritis of the knee. N Engl J Med 2006; 354: 307
841-848. 308
2. Dieppe PA, Lohmander LS. Pathogenesis and management of pain in osteoarthritis. 309
Lancet 2005; 365: 965-973. 310
3. Ishijima M, Kaneko H, Kaneko K. The evolving role of biomarkers for osteoarthritis. 311
Therapeutic Advances in Musculoskeletal Disease 2014; 6: 144-153. 312
4. Felson DT. The sources of pain in knee osteoarthritis. Curr Opin Rheumatol 2005; 17: 313
624-628. 314
5. Ishijima M, Watari T, Naito K, Kaneko H, Futami I, Yoshimura-Ishida K, et al. 315
Relationships between biomarkers of cartilage, bone, synovial metabolism and knee 316
pain provide insights into the origins of pain in early knee osteoarthritis. Arthritis Res 317
Ther 2011; 13: R22. 318
6. Carr AJ, Robertsson O, Graves S, Price AJ, Arden NK, Judge A, et al. Knee 319
replacement. Lancet 2012; 379: 1331-1340. 320
7. Goldring MB, Goldring SR. Osteoarthritis. J Cell Physiol 2007; 213: 626-634. 321
8. Scanzello CR, Goldring SR. The role of synovitis in osteoarthritis pathogenesis. Bone 322
2012; 51: 249-257. 323
9. Fernandez-Madrid F, Karvonen RL, Teitge RA, Miller PR, An T, Negendank WG. 324
Page 26
26
Synovial thickening detected by MR imaging in osteoarthritis of the knee confirmed by 325
biopsy as synovitis. Magn Reson Imaging 1995; 13: 177-183. 326
10. Crofford LJ. COX-2 in synovial tissues. Osteoarthritis Cartilage 1999; 7: 406-408. 327
11. Ayral X, Pickering EH, Woodworth TG, Mackillop N, Dougados M. Synovitis: a 328
potential predictive factor of structural progression of medial tibiofemoral knee 329
osteoarthritis -- results of a 1 year longitudinal arthroscopic study in 422 patients. 330
Osteoarthritis Cartilage 2005; 13: 361-367. 331
12. Benito MJ, Veale DJ, FitzGerald O, van den Berg WB, Bresnihan B. Synovial tissue 332
inflammation in early and late osteoarthritis. Ann Rheum Dis. 2005; 64: 1263-1267. 333
13. Ning L, Ishijima M, Kaneko H, Kurihara H, Arikawa-Hirasawa E, Kubota M, et al. 334
Correlations between both the expression levels of inflammatory mediators and growth 335
factor in medial perimeniscal synovial tissue and the severity of medial knee 336
osteoarthritis. Int Orthop 2011; 35: 831-838. 337
14. Scanzello CR, McKeon B, Swaim BH, DiCarlo E, Asomugha EU, Kanda V, et al. 338
Synovial inflammation in patients undergoing arthroscopic meniscectomy: molecular 339
characterization and relationship to symptoms. Arthritis Rheum 2011; 63: 391-400. 340
15. Liu L, Ishijima M, Futami I, Kaneko H, Kubota M, Kawasaki T, et al. Correlation 341
between synovitis detected on enhanced-magnetic resonance imaging and a histological 342
analysis with a patient-oriented outcome measure for Japanese patients with end-stage 343
Page 27
27
knee osteoarthritis receiving joint replacement surgery. Clin Rheumatol 2010; 29: 344
1185-1190. 345
16. Menashe L, Hirko K, Losina E, Kloppenburg M, Zhang W, Li L, et al. The diagnostic 346
performance of MRI in osteoarthritis: a systematic review and meta-analysis. 347
Osteoarthritis Cartilage 2012; 20: 13-21. 348
17. Hada S, Kaneko H, Sadatsuki R, Liu L, Futami I, Kinoshita M, et al. The degeneration 349
and destruction of femoral articular cartilage shows a greater degree of deterioration 350
than that of the tibial and patellar articular cartilage in early stage knee osteoarthritis: a 351
cross-sectional study. Osteoarthritis Cartilage 2014; 22: 1583-1589. 352
18. Felson DT, Niu J, Guermazi A, Roemer F, Aliabadi P, Clancy M, et al. Correlation of 353
the development of knee pain with enlarging bone marrow lesions on magnetic 354
resonance imaging. Arthritis Rheum. 2007; 56: 2986-2992. 355
19. Wenham CY, Conaghan PG. Imaging the painful osteoarthritic knee joint: what have 356
we learned? Nat Clin Pract Rheumatol 2009; 5: 149-158. 357
20. Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D, et al. Whole-Organ 358
Magnetic Resonance Imaging Score (WORMS) of the knee in osteoarthritis. 359
Osteoarthritis Cartilage 2004; 12: 177-190. 360
21. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of 361
criteria for the classification and reporting of osteoarthritis. Classification of 362
Page 28
28
osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the 363
American Rheumatism Association. Arthritis Rheum 1986; 29: 1039-1049. 364
22. Akai M, Doi T, Fujino K, Iwaya T, Kurosawa H, Nasu T. An outcome measure for 365
Japanese people with knee osteoarthritis. J Rheumatol 2005; 32: 1524-1532. 366
23. Buckland-Wright JC, Macfarlane DG, Williams SA, Ward RJ. Accuracy and precision 367
of joint space width measurements in standard and macroradiographs of osteoarthritic 368
knees. Ann Rheum Dis 1995; 54: 872-880. 369
24. Ravaud P, Auleley GR, Chastang C, Rousselin B, Paolozzi L, Amor B, et al. Knee joint 370
space width measurement: an experimental study of the influence of radiographic 371
procedure and joint positioning. Br J Rheumatol 1996; 35: 761-766. 372
25. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 373
1957; 16: 494-502. 374
26. Loeuille D, Chary-Valckenaere I, Champigneulle J, Rat AC, Toussaint F, 375
Pinzano-Watrin A, et al. Macroscopic and microscopic features of synovial membrane 376
inflammation in the osteoarthritic knee: correlating magnetic resonance imaging 377
findings with disease severity. Arthritis Rheum 2005; 52: 3492-3501. 378
27. de Lange-Brokaar BJ, Ioan-Facsinay A, Yusuf E, Visser AW, Kroon HM, Andersen SN, 379
et al. Degree of synovitis on MRI by comprehensive whole knee semi-quantitative 380
scoring method correlates with histologic and macroscopic features of synovial tissue 381
Page 29
29
inflammation in knee osteoarthritis. Osteoarthritis Cartilage 2014; 22: 1606-1613. 382
28. Mussener A, Funa K, Kleinau S, Klareskog L. Dynamic expression of transforming 383
growth factor-betas (TGF-beta) and their type I and type II receptors in the synovial 384
tissue of arthritic rats. Clin Exp Immunol 1997; 107: 112-119. 385
29. Fiocco U, Oliviero F, Sfriso P, Calabrese F, Lunardi F, Scagliori E, et al. Synovial 386
biomarkers in psoriatic arthritis. J Rheumatol Suppl 2012; 89: 61-64. 387
30. Soden M, Rooney M, Cullen A, Whelan A, Feighery C, Bresnihan B. 388
Immunohistological features in the synovium obtained from clinically uninvolved knee 389
joints of patients with rheumatoid arthritis. Br J Rheumatol 1989; 28: 287-292. 390
31. van der Kraan PM, van den Berg WB. Osteophytes: relevance and biology. 391
Osteoarthritis Cartilage 2007; 15: 237-244. 392
32. Saito I, Koshino T, Nakashima K, Uesugi M, Saito T. Increased cellular infiltrate in 393
inflammatory synovia of osteoarthritic knees. Osteoarthritis Cartilage 2002; 10: 394
156-162. 395
33. Shimura Y, Kurosawa H, Sugawara Y, Tsuchiya M, Sawa M, Kaneko H, et al. The 396
factors associated with pain severity in patients with knee osteoarthritis vary according 397
to the radiographic disease severity: a cross-sectional study. Osteoarthritis Cartilage 398
2013; 21: 1179-1184. 399
34. Kubota M, Ishijima M, Kurosawa H, Liu L, Ikeda H, Osawa A, et al. A longitudinal 400
Page 30
30
study of the relationship between the status of bone marrow abnormalities and 401
progression of knee osteoarthritis. J Orthop Sci 2010; 15: 641-646. 402
35. Hayashi D, Englund M, Roemer FW, Niu J, Sharma L, Felson DT, et al. Knee 403
malalignment is associated with an increased risk for incident and enlarging bone 404
marrow lesions in the more loaded compartments: the MOST study. Osteoarthritis 405
Cartilage 2012; 20: 1227-1233. 406
36. Felson DT, Parkes MJ, Marjanovic EJ, Callaghan M, Gait A, Cootes T, et al. Bone 407
marrow lesions in knee osteoarthritis change in 6-12 weeks. Osteoarthritis Cartilage 408
2012; 20: 1514-1518. 409
37. Roemer FW, Neogi T, Nevitt MC, Felson DT, Zhu Y, Zhang Y, et al. Subchondral bone 410
marrow lesions are highly associated with, and predict subchondral bone attrition 411
longitudinally: the MOST study. Osteoarthritis Cartilage 2010; 18: 47-53. 412
38. Felson DT, McLaughlin S, Goggins J, LaValley MP, Gale ME, Totterman S, et al. Bone 413
marrow edema and its relation to progression of knee osteoarthritis. Ann Intern Med 414
2003; 139: 330-336. 415
416
Page 31
31
Tables 417
Table 1 418
The characteristics of the patients in this study 419
Subject (n) 40
Age (y) 71.8 (6.9)
Gender (F / M) 35 / 5
JKOM score (Range; 0 - 100) 49.2 (20.0)
Pain VAS score (0 - 100) 61.2 (26.9)
FTA (º) 185.0 (5.6)
WORMS (0 - 332) 145.0 (27.0)
Cartilage morphology (0 - 84) 61.1 (8.6)
BML (0 - 45) 16.1 (9.3)
SBC (0 - 45) 2.9 (3.0)
SBA (0 - 42) 4.3 (3.2)
Osteophyte (0 - 98) 48.0 (13.7)
Menisci Medial (0 – 6) 5.5 (0.5)
Lateral (0 – 6) 3.8 (1.0)
Synovitis (Range; 0 – 3) 2.0 (0.9)
HSS (0 – 15) 5.5 (0.8)
The data indicates the means (SD). JKOM: Japanese knee osteoarthritis measure, VAS: visual 420
analog scale, FTA: femoro-tibial angle (lateral FTA; 185 degrees indicate 5 degrees of 421
anatomical varus), WORMS: whole-organ magnetic resonance imaging score, BML: bone 422
marrow lesion, SBC: subchondral bone cyst, SBA: subchondral bone attrition, HSS: 423
histological synovitis score. 424
425
Page 32
32
Table 2 426
Correlation between the total HSS and the WORMS in patients with end-stage knee OA 427
receiving TKA 428
HSS
r p
Cartilage morphology -0.006 0.97
BML 0.325 0.04*
SBC 0.350 0.03*
SBA 0.482 <0.01*
Osteophyte -0.100 0.54
Menisci Medial 0.155 0.34
Lateral 0.266 0.10
Synovitis 0.358 0.02*
HSS: histological synovitis score, WORMS: whole-organ magnetic resonance imaging score, 429
BML: bone marrow lesion, SBC: subchondral bone cyst, SBA: subchondral bone attrition. * 430
indicates p<0.05. 431
432
Page 33
33
Table 3 433
Correlation between the total HSSs and the inflammatory cytokines and growth factor in 434
patients with end-stage knee OA receiving TKA 435
HSS
r p
IL-1β Lining layer
Sublining layer
0.223
0.277
0.19
0.10
COX-2
Lining layer
Sublining layer
Vasculature
0.018
0.080
0.231
0.92
0.67
0.22
IL-6
Lining layer
Sublining layer
Vasculature
0.043
0.211
0.122
0.81
0.24
0.50
TGF-β
Lining layer
Sublining layer
Vasculature
0.073
0.151
0.394
0.70
0.43
0.03*
HSS: histological synovitis score, IL-1β; interleukin-1β, COX-2; cyclo-oxygenase (COX)-2, 436
TNF-α; tumor necrosis factor-α, TGF-β; transforming growth factor, *p<0.05 437
438
Page 34
34
Figure 1. Association between the HSS and the BML score (A), SBC score (B) and SBA 439
score (C) in patients with end-stage knee OA receiving TKA 440
The sagittal T2-weighted MR image (D, E and F) and the histological sections (G, H and I) of 441
the three representative cases (J, K and L, respectively) those showed the presence of BML, 442
SBC and SBA. J: Cartilage loss: grade 6 in MFc (arrow with head) and MTc, grade 5 in MFa 443
and MTa, grade 4 in MFp and MTp, BML: grade 3 in MTa, grade 2 in MFc (arrow head), 444
SBC: grade 1 in MFc (thin arrow), SBA: grade 1 in MTc (thick arrow), osteophyte: grade 7 in 445
MFa (open arrow head), grade 5 in MFp, MTa and MTp, meniscus: grade 4 in the medial 446
meniscus (open arrow). K: Cartilage loss: grade 6 in MFa, MFc, MTa and MTc, grade 5 in 447
MFp (arrow with head) and MTp, BML: grade 3 in MFa, MFc (arrow head), MTa and MTc, 448
grade 2 in MTp, SBC: grade 1 in MFa and MFc (thin arrow), SBA: grade 2 in MTc (thick 449
arrow), osteophyte: grade 7 in MFa (open arrow head) and MFp, grade 6 in MTa, grade 5 in 450
MTp, meniscus: grade 3 in the medial meniscus (open arrow), effusion: grade 1 (star). L: 451
Cartilage loss: grade 6 in MFc (arrow with head), MFp and MTc, grade 4 in MFa and MTa, 452
BML: grade 3 in MFa, MFc (arrow head), MFp, MTa, MTc and MTp, SBC: grade 3 in MFc 453
(thin arrow), grade 2 in MFp, SBA: grade 3 in MFc and MTc (thick arrow), osteophyte: grade 454
4 in MTa (open arrow head) and MTp, grade 2 in MFa, meniscus: grade 3 in the medial 455
meniscus (open arrow), effusion: grade 3 (star). BML: bone marrow lesion, SBC: subchondral 456
bone cyst, SBA: subchondral bone attrition, HSS: histological synovitis score, TKA: total 457
Page 35
35
knee arthroplasty, MF: medial femoral plateau, MT: medial tibial plateau, a: anterior, c: 458
central, p: posterior. 459
Page 36
Figure 1
50µm
HSS
BML score
J
K
L
HSS
0
4
5
6
7
0 10 20 30 40
A
SBC score
J K
L H
SS
0
4
5
6
7
0 2.5 5 7.5 10 12.5
B
SBA score
0
4
5
6
7
0 2.5 5 7.5 10 12.5
J K
L
C
G H I 50µm 50µm
D E F
J K L