Does joint effusion on T2 magnetic resonance images reflect

synovitis? Comparison of arthroscopic findings in internal

derangements of the temporomandibular joint
Natsuki Segami, DDS, PhD,a Masaaki Nishimura, DDS,b Keiseki Kaneyama, DDS, PhD,c
Masahisa Miyamaru, DDS, PhD,c Jun Sato, DDS,c and Ken-Ichiro Murakami, DDS, PhD,d
Ishikawa Prefecture and Kyoto, Japan
KANAZAWA MEDICAL UNIVERSITY AND KYOTO UNIVERSITY

Objective. The purpose of this study was to clarify the characteristics of joint effusion (JE) in the temporomandibular joint by
comparing T2-weighted magnetic resonance (MR) images with arthroscopic findings.
Study design. MR images of 47 symptomatic temporomandibular joints (47 patients) with internal derangement associated
with painful hypomobility were taken to evaluate the degree of JE on a scale of 0 to 3. Within 2 months after MR images,
arthroscopic findings with respect to the severity of synovitis, adhesion, and degenerative change in the superior compartment
were each quantitatively assessed on a scale of 0 to 10. The arthroscopic scores were compared among the 4 JE grades, as
well as between 2 groups, effusion-present (grades 2 and 3) and effusion-absent (grades 0 and 1), by using the Spearman
correlation coefficient and the Mann-Whitney U test.
Results. The distribution of JE was as follows: 10 joints had grade 0, 5 joints had grade 1, 19 joints had grade 2, and 13 joints
had grade 3. The synovitis score had a significant relationship to the degree of effusion (P = .0012) and was higher in the
effusion-present group (6.4 ± 2.0) than in the effusion-absent group (4.2 ± 1.6) (P = .0005) On the other hand, there was
no statistically significant correlation between either adhesion or degenerative change and effusion.
Conclusions. JE may reflect synovitis; however, the nature of JE itself needs to be further elucidated by additional studies.
(Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001;92:341-5)

Joint effusion (JE), detected by T2-weighted magnetic PATIENTS AND METHODS

resonance imaging (MRI), is prevalent in 30% to 80%
of temporomandibular joint (TMJ) disorders.1-8 For
example, Westesson and Brooks4 investigated 379
patients with TMJ disorders and found JE in 50% of
the joints with anterior disk displacement without
reduction and in 46% of the joints with painful sides.
JE is believed to reflect intra-articular pathosis;
however, to our knowledge, no study has linked the
presence of JE, detected by MRI, with direct observa-
tions of the inside of the TMJ. To better understand the
nature of JE, we studied the relationship between MRI
features and arthroscopic features in internally
deranged TMJs.
Partially supported by the Grant for Promoted Research of Kanazawa
Medical University (S00-11) and Grant-Aid for Scientific Research
(08457559) from the Ministry of Education, Science and Culture of
Japan.
aProfessor and Chairman, Department of Oral and Maxillofacial
Surgery, Kanazawa Medical University, Ishikawa Prefecture, Japan.
bPostgraduate Student, Department of Oral and Maxillofacial
Surgery, Kanazawa Medical University, Ishikawa Prefecture, Japan.
cAssistant Professor, Department of Oral and Maxillofacial Surgery,
Kanazawa Medical University, Ishikawa Prefecture, Japan.
dAssociate Professor, Department of Oral and Maxillofacial Surgery,
Kyoto University, Kyoto, Japan.
Received for publication Nov 3, 2000; returned for revision Apr 9,
2001; accepted for publication May 8, 2001.
Copyright © 2001 by Mosby, Inc.
1079-2104/2001/$35.00 + 0 7/16/117808
doi:10.1067/moe.2001.117808
Forty-seven consecutive patients (47 TMJs) under-
went arthroscopic surgery after nonsurgical treatment
was unsuccessful. The patients, 3 males and 44
females, had an average age of 39 years, ranging from
13 to 76 years. The patient population consisted of 42
unilateral cases with painful hypomobility and 5 bilat-
eral cases. In the bilateral cases, the more painful side
was used for this study. The mean duration of symp-
toms was 22.7 months (1-84 months), and the mean
maximum mouth opening was 31 mm, ranging from 20
mm to 39 mm.
MRI
By using a 1.5 Tesla Magnetom MRI system with 6-
inch surface coils (Siemens AG, Erlangen, Germany) and
continuous proton density (TR/TE, 2000/20 ms), T2-
weighted (TR/TE, 2000/80 ms) 3-mm–thick sagittal and
coronal images were taken in a spin-echo sequence. The
sagittal MR images were obtained in both closed-mouth
and maximum opening positions, but the coronal image
was obtained only in the closed-mouth position. The
degree of JE in the superior compartment was divided
into 4 grades: 0, no area of high signal intensity; 1, lining
or spot of high intensity along the articular surface; 2,
band of high intensity; and 3, collection with pooling in
the compartment (Figs 1-4). The presence of effusion
was indicated by a high signal intensity of 2 or 3,
according to the description by Westesson and Brooks.4

341
342 Segami et al
Table I. Scoring with intensity of synovitis10
Grade Findings
0 Normal pale, almost translucent, synovial lining with a
fine network of anastomosing small blood vessels
1 Increased vascularity and capillary hyperemia (mild)
2 Increased vascularity and capillary hyperemia
(moderate)
3 Increased vascularity and capillary hyperemia (severe)
4 Capillary dilatation and increasing network (mild to
moderate)
5 Capillary dilatation and increasing network (severe)
6 Contact bleeding occurs on probe palpation (mild to
moderate)
7 Contact bleeding occurs on probe palpation (severe)
8 Microbleeding and effusion
9 Granulative change, effusion, and debris (mild to
moderate)
10 Granulative change, effusion, and debris (severe)
Table II. Scoring with intensity of adhesion10
Grade Findings
0 No adhesion or fibrous change
1 Filmy adhesion (mild)
2 Filmy adhesion (moderate to severe)
3 Fibrosynovial band (mild to moderate)
4 Fibrosynovial band (severe)
5 Fibrous band (mild to moderate)
6 Fibrous band (severe)
7 Pseudowall formation (mild to moderate)
8 Pseudowall formation (severe)
9 Capsular fibrosis (mild to moderate)
10 Capsular fibrosis (severe)
Grade 1 was excluded from our subdivision of JE
into 4 grades, specifically from the effusion-present
group, because the high signal with a lining or spot
along the articular surface was seen in 7% of normal
volunteers.4 JE grading was determined by consensus
among 1 radiologist and 2 oral surgeons blinded to the
patients’ names and clinical information.
Arthralgia questionnaire
On the day the MR image was taken, the degree of
arthralgia from jaw movement during the previous
week was subjectively assessed by patients who used
the Visual Analogue Scale of pain (VAS, 0-10 points).
Diagnostic arthroscopy
After MRI, 31 of the 47 joints were treated unsuc-
cessfully by means of arthrocentesis and anti-inflam-
matory drug administration for 2 months until the
arthroscopic surgery. The other 16 joints underwent
arthroscopic surgery immediately (0-5 days) after the
MRI because of their nonresponse to previous conser-
vative treatments.
ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
September 2001
Table III. Scoring with intensity of degenerative
change10
Grade Findings
0 Normal articular cartilage
1 Softening in probe palpation (mild)
2 Softening in probe palpation (moderate to severe)
3 Convex-concave surface
4 Fibrillation (mild)
5 Fibrillation (moderate)
6 Fibrillation (severe)
7 Bone exposure (mild)
8 Bone exposure (moderate)
9 Bone exposure (severe)
10 Intracapsular fibrosis
Conventional diagnostic arthroscopy of the whole
area of the superior joint compartment9 was performed
and videotaped for approximately 10 minutes, followed
by arthroscopic lysis and lavage. The degree of
synovitis was evaluated according to the criteria of
Murakami et al10: Namely, the most pronounced degree
of synovitis in the superior compartment was given a
score on a 10-point scale (Table I). In addition, the
degree of intracapsular adhesion and degenerative
change was also evaluated in a similar manner (Tables
II and III). Each arthroscopic feature was scored retro-
spectively by means of the videotape by 2 oral surgeons
blinded to the patients’ names and condition(s).
Statistical analysis
Statistical analysis of the arthroscopic scores and
the VAS scores between the effusion-present and the
effusion-absent groups was performed by using the
Mann-Whitney U test. The Spearman correlation co-
efficient was used to compare the JE grades and the
arthroscopic scores. The occurrence of disk perforation
and JE were compared by using the Fisher exact test.
RESULTS
The proton density MR images clearly depicted ante-
rior disk displacement without reduction in all joints.
On T2-weighted MR images, there was positive JE in
32 joints (68.1%). The grading of all the joints is
shown in Table IV. The VAS score averaged 5.8 ± 2.1
and had no correlation with the presence of JE. The
mean arthroscopic scores were 5.7 ± 2.1 for synovitis,
5.6 ± 2.2 for adhesion, and 1.8 ± 0.2 for degenerative
change. Synovitis was commonly observed in the
posterior and the posterior-lateral aspects of the
compartment, especially in the retrodiskal tissue.
A significant relationship between the degree of JE
and the synovitis score was found from grade 0 to
grade 3 (P = .0012) (Fig 5). The joints with effusion
had significantly higher synovitis scores than the joints
ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
Volume 92, Number 3
Fig 1. Sagittal T2 MRIs of grade 0 in closed-mouth position
(left) and in maximum opening (right) position. High signal
intensity is not visible in joint space.
Fig 2. Sagittal T2 MRIs of grade 1 in closed-mouth position
(left) and in maximum opening (right) position. High spot
signals are identified in anterior recesses (arrowheads).
without effusion (6.4 ± 2.0 vs 4.2 ± 1.6; P = .0005).
Neither the degree of adhesion (5.8 ± 2.0 vs 5.9 ± 2.7,
P = .97) nor the degree of degenerative change (1.3 ±
1.8 vs 3.1 ± 3.4; P = .09) had a significant correlation
with JE. Nine joints had disk perforation, identified
through arthroscopy; however, disk perforation was
found to have no significant correlation with JE (P =
.16) (Table IV).
The results of the 16 TMJs that immediately under-
went arthroscopy after MRI demonstrated similar
significant correlation between synovitis scores and
presence/absence of JE (6.0 ± 2.0 vs 3.2 ± 1.0; P =
.011) as well as between synovitis scores and grades of
JE (P = .014).
DISCUSSION
To our knowledge, no study of the amount of JE and
the degree of synovitis in the TMJ or any other human
joint has been reported. In our study, the joints with
effusion had more pronounced synovitis than the joints
with no effusion. Moreover, the amount of JE increased
on the basis of the severity of synovitis, assessed by
arthroscopy. Therefore, the presence of JE on T2 MR
images generally reflects synovitis.
Segami et al 343
Fig 3. Sagittal T2 MRIs of grade 2 in closed-mouth position
(left) and in maximum opening (right) position. Band shape
high signals are indicated in anterior and posterior recesses
(arrowheads).
Fig 4. Sagittal T2 MRIs of grade 3 in closed-mouth position
(left) and in maximum opening (right) position. High signal is
depicted in wide area in joint space as pooling (arrowheads).
It is of particular interest whether the effusion
detected by means of the MRI in this study is actually
synovitis. If it is, we would expect to detect effusion
whenever synovitis is clearly observed. However,
discrepancies between the MR images and arthroscopy
occurred in 5 of 47 TMJs: MR images did not reveal
effusion even though the synovitis score was greater
than 5. Nevertheless, this does not necessarily mean
that effusion is not the same as synovitis. One reason
for this seeming contradiction is that most of the JE
may be caused by liquid-containing secondary prod-
ucts produced by inflamed synovium. Another reason
is that synovitis may be overdiagnosed when there is a
mild to moderate degree of hyperemia or vascular
dilatations, because arthroscopic features of the normal
appearance of synovial vasculature have not been very
specific (ie, arthroscopic examination reveals the artic-
ular surface, not the deep structure). Therefore, to
make a confident conclusion, it might be helpful, first,
to evaluate synovitis at the histologic level in biopsy
specimens because the histologic grading system for
the quantitative evaluation of synovitis in biopsy spec-
imens11,12 will resolve this discrepancy. Furthermore,
arthroscopic inspection normally has a blind area in the
344 Segami et al ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
September 2001

Fig 5. Distribution grade of JE and arthroscopic synovitis score. Significant correlation is indicated by means of the
Mann-Whitney U test.

Table IV. Distribution grade of JE, VAS, arthroscopic scores, and number of disk perforations
Arthroscopic scores
Degenerative No of joints with
Grade No. of joints VAS Synovitis* Adhesion change disk perforation
0 10 6.5 ± 3.0 4.6 ± 1.7 4.8 ± 2.7 2.0 ± 3.3 2
1 5 2.5 ± 3.1 3.6 ± 1.5 7.0 ± 1.8 4.2 ± 3.0 2
2 19 5.5 ± 2.5 6.2 ± 2.3 6.2 ± 1.5 2.3 ± 0.3 3
3 13 6.7 ± 2.3 6.7 ± 1.6 5.0 ± 2.5 0.5 ± 1.4 2
Total number and average score 47 5.8 ± 2.1 5.7 ± 2.1 5.6 ± 2.2 1.8 ± 0.2 9
*Only synovitis score had a significant correlation in 4 grades of JE (P = .0012) and between positive/negative groups (P = .0005).

puncture site, from the lateral capsule to the lateral disk
attachment.
Interestingly, Takaku et al5 reported that in 30 of 43
TMJs with JE, the retrodiskal tissue histologically
showed marked synovitis with synovial cell prolifera-
tion and edematous synovial connective tissue. They
suggested that JE might indicate the presence of
synovitis due to disk damage or degeneration.
However, in our study, 9 joints with disk perforation
did not appear to have JE. Therefore, synovitis appears
to be more closely related to JE than does degenerative
change.
In an interesting study on knee joints with rheuma-
toid arthritis and osteoarthritis, Ostergaard et al13
reported that the synovial volume detected with intra-
venous gadopentetate dimeglumine was significantly
correlated with histologically evaluated synovial
inflammatory activity in biopsy specimens at the
macroscopic synovitis level. A gadopentetate dimeglu-
mine injection may be useful to distinguish the actual
effusion from hypertrophic synovial tissue, which
might be enhanced through blood supply.14 In these
instances, orthopedic researchers commonly interpret
JE as being both synovial fluid and hypertrophic
synovium caused by synovitis.13,15,16 In TMJ disor-
ders, the etiologic factors of JE may be similar to those
of synovial fluid and hypertrophic synovitis.
Through the evaluation of recent biochemistry
research on synovial fluid and synovial tissue,17-19
Takahashi et all8 detected higher levels of protein
concentration in painful TMJs with internal derange-
ment and osteoarthritis and suggested that JE is related
to inflammatory changes. On the other hand, Sandler et
al20 reported that the concentration level of interleukin-
6 in synovial fluid had a close correlation with the level
of acute synovitis, which was demonstrated arthroscop-
ically. Therefore, it is possible that joints with effusion
may contain high concentrations of secondary products
subsequent to inflammation of the synovium. Although
arthralgia may be accelerated with inflammatory prod-
ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
Volume 92, Number 3
ucts in JE, the VAS of pain did not correlate either with
the presence/absence of JE or the degree of synovitis.
Similarly, Murakami et al6 reported that JE did not
directly correlate with the presence of either TMJ pain or
increased pain level, which was subjectively assessed in
19 patients through questionnaires.
Lastly, the prevalence of JE, 68.1% (32/47 joints) in
TMJs with anterior disk displacement without reduc-
tion and arthralgia is comparable with those reported in
previous studies.4-8
In conclusion, T2-weighted MRI can be a highly
clinically significant and useful diagnostic tool for
predicting the existence of synovitis—in its ability to
detect JE. Our finding that there is a strong relationship
between JE and arthroscopic synovitis may help to
resolve the clinical puzzle of whether JE is really a sign
of synovitis. Further investigation of histologic and
histochemical findings in synovial tissue samples will
be required to clarify the real nature of JE.
REFERENCES
1. Katzberg RW. Temporomandibular joint imaging. Radiology
1989;170:297-307.
2. Schellhas KP, Wilkes CH, Baker CC. Facial pain, headache, and
temporomandibular joint inflammation. Headache 1989;29:229-32.
3. Schellhas KP, Wilkes CH. Temporomandibular joint inflamma-
tion: comparison of MR fast scanning with T1- and T2-weighted
imaging techniques. AJR Am J Roentgenol 1989;53:93-8.
4. Westesson PL, Brooks SL. Temporomandibular joint: relation-
ship between MR evidence of effusion and the presence of pain
and disk displacement. AJR Am J Roentgenol 1992;159:559-63.
5. Takaku S, Toyoda T, Sano T, Heishiki A. Correlation of
magnetic resonance imaging and surgical findings in patients
with temporomandibular joint disorders. J Oral Maxillofac Surg
1995;53:1283-8.
6. Murakami K, Nishida M, Bessho T, Iizuka T, Tsuda Y, Konishi
J. MRI evidence of high signal intensity and temporomandibular
arthralgia and relating pain. Does the high signal correlate to the
pain? Br J Oral Maxillfac Surg 1996;34:220-4.
7. Adame CG, Monje F, Offnoz M, Martin-Granizo R. Effusion in
magnetic resonance imaging of the temporomandibular joint: a
study of 123 joints. J Oral Maxillofac Surg 1998;56:314-8.
8. Takahashi T, Nagai H, Seki H, Fukuda M. Relationship between
joint effusion, joint pain, and protein levels in joint lavage fluid of
patients with internal derangement and osteoarthritis of the
temporomandibular joint. J Oral Maxillofac Surg 1999;57:1187-93.
Segami et al 345
9. Murakami K, Ono T. Temporomandibular joint arthroscopy by
inferolateral approach. Int J Oral Maxillofac Surg 1986;15:410-7.
10. Murakami K, Segami N, Fujimura K, Iizuka T. Correlation
between pain and synovitis in patients with internal derange-
ment of the temporomandibular joint. J Oral Maxillofac Surg
1991;49:1159-61.
11. Gynther GW, Dijkgraaf LC, Reinholt F P, Holmlund AB, Liem
RS, de Bont LG. Synovial inflammation in arthroscopically
obtained biopsy specimens from the temporomandibular joint: a
review of the literature and a proposed histologic grading
system. J Oral Maxillfac Surg 1998;56:1281-6.
12. Holmlund A, Hellsing G. Arthroscopy of the temporomandibular
joint: occurrence and location of osteoarthritis and synovitis in a
patient material. Int J Oral Maxillofac Surg 1988;17:36-40.
13. Østergaard M, Stoltenberg M, Løvgreen-Nielsen P, Volck B,
Jensen CH, Lorenzen I. Magnetic resonance imaging–deter-
mined synovial membrane and joint effusion volumes in
rheumatoid arthritis and osteoarthritis: comparison with the
macroscopic and microscopic appearance of the synovium.
Arthritis Rheum 1997;40:1856-67.
14. Smith HJ, Larheim TA, Aspestrand F. Rheumatic and
nonrheumatic disease in the temporomandibular joint:
gadolinium-enhanced MR imaging. Radiology 1992;185:229-34.
15. Leitch R, Walker SE, Hillard AE. The rheumatoid knee before
and after arthrocentesis and prednisolone injection: evaluation
by Gd-enhanced MRI. Clin Rheum 1996;15:358-66.
16. Beltran J, Noto AM, Herman LJ, Mosure JC, Burk JM,
Christoforidis AJ. Joint effusions: MR imaging. Radiology
1986;158:133-7.
17. Quinn JH, Bazan NG. Identification of prostaglandin E2 and
leukotriene B4 in the synovial fluid of painful, dysfunctional
temporomandibular joints. J Oral Maxillfac Surg 1990;48:968-71.
18. Takahashi T, Kondoh T, Fukuda M, Yamazaki Y, Toyosaki T,
Suzuki R. Proinflammatory cytokines detectable in synovial
fluid from patients with temporomandibular joint disorders. Oral
Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85:135-41.
19. Suzuki T, Segami N, Kaneyama K, Nishimura M, Nojima T.
Specific expression of interleukin-1β in temporomandibular
joints with internal derangement: correlation with clinical find-
ings. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
1999;88:413-7.
20. Sandler NA, Buckley MJ, Cillo JE, Braun TW. Correlation of
inflammatory cytokines with arthroscopic findings in patients
with temporomandibular joint internal derangements. J Oral
Maxillfac Surg 1998;56:534-43.
Reprint requests:
Natsuki Segami, DDS, PhD
Professor and Chairman
Department of Oral and Maxillofacial Surgery
Kanazawa Medical University
Uchinada, Ishikawa Prefecture 920-0293
Japan
n-segami@kanazawa-med.ac.jp