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 Maxiliofac Surg 5/: I 187-l 1 Y3, 1999

Relationship Between Joint Effusion, Joint Pain, and Protein Levels in Joint Lavage Fluid of Patients with Internal Derangement and Osteoarthritis of the Temporomandibular Joint Tetsu Takahashi, DDS, PbD, * Hirokazu Nagai, DDS, i” Hiroshi Seki, DOS,+ and Masayuki Fukuda, DDS, PbDJ Purpose: The purpose of this study was to investigate the relationship between the presence of joint effusion, joint pain, and protein levels in joint lavage fluid (JL) of patients with internal derangement (ID) and osteoarthritis (OA) of the temporomandibular joint (TMJ). Patients and Methods: Thirty-eight joints in 26 patients with ID and OA of the TMJ were studied. Magnetic resonance imaging (MRP) evidence of joint effusion was evaluated in T2-weighted images. Samples of JL were collected from the superior joint space during pumping manipulation, and the protein concentration was measured. The presence of pain was based on joint tenderness or a complaint of pain in the preauricular region during mouth opening or closing. Results: Joint effusion was demonstrated in 20 of 25 (80%) painful joints; a significantly higher incidence than in pain-free joints (5 of 13, 38.5%). The mean protein concentration (2.15 mg/mL) in JL from painful joints was significantly higher than in pain-free joints (1.22 mg/mL) (P < .05). Furthermore, the mean protein concentration (2.12 mg/mL) in JL from joints with elf&ion was significantly higher than in joints without joint effusion (1.27 mg/mL) (P < .05).

These data demonstrate that painful joints are more likely to show joint effusion on MRI, and the protein levels in JL recovered from these joints is higher than in pain-free joints. These data also suggested that joint effusion may be related to the inflammatory changes seen in patients with ID and OA.

Conclusions:

which is demonstrated on T2-weighted images as an increased signal in the joint compartment, has recently received considerable attention as a novel way to estimate inflammatory reactions accompanying arthralgia.2~5 Although several studies have shown that joint effusion may reflect inflammatory changes in the TMJ,@ its clinical as well as biologic significance remains unclear. Furthermore, there are no data concerning the dynamic changes in joint effusion in relation to the clinical signs and symptoms or treatment outcome. In this study, it was hypothesized that joint effusion represents an inflammatory change involving extravasation of protein molecules. Therefore, protein levels in joint lavage fluid (JL) collected from the superior joint space in patients with ID or OA were measured and correlated with the presence of joint effusion
Currently, magnetic resonance imaging (MFU) is the most reliable and useful imaging modality for internal derangement (ID) and osteoarthritis (OA) of the temporomandibular joint (TMJ) because of its lack of ionizing radiation, noninvasive characteristics, and superior soft-tissue contrast resolution.‘Joint effusion,

Received

from

University

School

the

Division of Medicine,

of Dentistry

and Oral

Surgery,

Akita

Akita, Japan.

*Lecturer. tResident. *Resident. gAssistant Supported the Ministry Address Division Medicine,

Professor. in part by a Grant-in-Aid for Scientific Research from of Education, Science, Sports, and Culture of Japan. correspondence

and reprint

requests

to Dr Takahashi:

of Dentistry and Oral Surgery, Akita University School of 1-1-l Hondo Akita City, Akita 010.8543, Japan; e-mail:

Patients

Methods

Thirty-eight joints in 26 patients with ID and OA of the TMJ (6 joints in 4 males and 32 joints in 22 females; mean age, 43.2 years; range, 16 to 67 years) were

[email protected] 0 1999 American Associabn of Oral and Max~llofcmal 0278.2391/99/T O-o005$3 00/O

and

Surgeons

1187

1188

JOINT

involved in this study. All patients had been treated with nonsurgical modalities for at least 3 months without satisfactory improvement before intra-articular pumping and lavage were performed. These modalities included medication (nonsteroidal anti-inflammatory drugs and muscle relaxants), use of a bite appliance, and physical therapy. A clinical examination was performed in a routine manner, included palpation of the masticatory muscles and TMJ for tenderness and measurement of the range of motion. The radiographic examination was routinely performed using panoramic and transcranial views (open and closed mouth) and tomography for determination of the bony changes such as osteophyte formation and erosion. MRI was performed in all patients using a 1.5-T MRI scanner (General Electric, Milwaukee, WI) with bilateral 3-inch dual-surface coils after initial nonsurgical treatment and before the pumping manipulation. Four pulse sequences were obtained: 1) axial localizer (300/16 [TR/TE]. 1 excitation, l&cm field of view [FOVJ, and a 256 X 128 matrix); 2) sagittal closedmouth (2,000/20, 80, 2 excitations, lo-cm FOV 256 X 256 matrix); 3) sag&al open-mouth (2,000/20,

FIGURE 1. A, Positive joint effusion on MRI. Arrows Negative joint effusion on MRI. The arrow indicates was considered not to be joint effusion.

indicate the hyperintense a point-like hyperintense

EFFUSION

AND

SYNOVL4L

FLUID

PROTEIN

LEVELS

80, 2 excitations, 12cm FOV, 256 X 256 matrix); and 4) coronal closed-mouth (350/17, 1 excitation, l&cm FOV, 256 X 256 matrix). Tl-weighted, protonweighted, and T2-weighted views were obtained in 3-n-m thickness. On T2-weighted images, joint efision was identified as an area of high signal intensity in the region of the superior or inferior joint space (Fig 1A). On proton-weighted images, the signal intensity in these areas was decreased. When more than a line of high signal was evident in at least 2 consecutive sagittal sections, it was considered positive for joint effusion (Fig 1A). Otherwise, it was considered as negative (Fig 1B). Joint effusion in the inferior joint space was excluded from this study because JL was collected only from the superior joint space. The clinical status of each patient was evaluated independently by 2 of the authors (T.T. and H.N.), who each determined the presence or absence of joint effusion. Clinical criteria for internal derangement were locking of the TMJ, pronounced impairment of joint mobility, joint soreness, or a history of clicking and intermittent locking. Clinical criteria for OA of the TMJ were impaired joint mobility, joint soreness, and degenerative changes in the osseous surfaces seen on

signal above the anteriorly signal above the anteriorly

displaced displaced

disc in the superior disc in the superior

joint space. 5, joint space that

TAKAHASHI

1189

ET AL

tomography or on arthroscopy. Arthroscopy was performed in 13 joints of 10 patients (one joint in a male, and 12 joints in 9 females). The arthroscopic findings, included superficial fibrillation of the articular surface of the temporal component and the disc, exposure of subchondral bone, and disc perforation, were also supportive of the diagnosis of OA of the TMJ. On Tl-weighted MFU, all 38 TMJs showed anterior disc displacement without reduction. Eight of the 38 joints also showed bone changes on tomography or on arthroscopy (4 joints in 3 subjects) and were diagnosed as OA of the TMJ. The presence of pain was determined by joint tenderness or a complaint of pain in the preauricular region on mouth opening or closing. The TMJs were divided into 2 groups: a painful joint group and a pain-free joint group. In 12 subjects (2 males and 10 females), the lavage samples were collected bilaterally; otherwise, the samples were collected from 1 joint. In 9 subjects (2 males and 7 females), 1 joint was classified in the painful group and the other in the pain-free group. In 2 female subjects, both joints were classified in the pain-free group. In one female subject, both joints were classified in the painful group. Therefore, the painful group comprised 25 samples from 24 subjects; 14 samples from 14 subjects whose JL was collected unilaterally and 11 sample from 10 subjects whose JL was collected bilaterally (unilaterally painful in 9 subjects and bilaterally painful in 1 subject). The pain-free group comprised 13 samples from 11 subjects (unilaterally in 9 subjects and bilaterally in 2 subjects). There were no significant differences in sex distribution, age distribution, mean age, and mean maximal mouth opening (interincisal distance) between the 2 groups (Table 1). The patients were given no medication for at least 2 weeks before JL sampling was performed. JOINT LAVAGESAMPLE PREPARATION Samples of JL from the superior joint space were obtained from those patients who underwent pumping manipulation for therapeutic purposes. The samples were prepared as previously described.‘O

Painful

TMJ (n) Subject (n) Male/female Mean age (age range) MM0

25 24 4/20 47.2 (16-67) 31.3 2 5.9

(mm)

Disc displacement OA change Abbreviations: MMO, anterior disc displacement

ADDw/oR 6/25 maximal without

mouth opening; reduction.

Pain-Free 13 11 2/9 42.6 (19-67) 30.5 -c 8.2 ADDw/oR 2/13 ADD

w/o

R,

Briefly, after local anesthesia, 2 mL saline solution was injected into the superior joint space and the patient was asked to open and close his or her mouth to mix the saline solution with the synovial fluid (SF). The mixture of SFand saline was aspirated and reinjected a total of 10 times. The JL sample was then collected. The sample was centrifuged (SOOg for 5 minutes) to remove cells and stored at -85°C until assay was performed. JL samples collected from the inferior joint space or samples containing blood resulting from active bleeding during the sampling procedure were excluded from the study. JL samples whose recovery was less than 50% of the injected volume were also excluded because our preliminary data suggested that these JL samples sometimes showed extremely low levels of protein (data not shown). MEASUREMENT OF TOTAL PROTEIN The measurement of total protein concentration was performed using the micro-BCA assay method (Pierce, Rockford, IL), according to the instructions provided by the manufacturer. The total amount of the TMJ JL sample analyzed was 100 uL. STATISTICS The relationship between the incidence of joint effusion on MIU and pain in the joint area was assessed using Fisher’s exact test. The mean protein concentration in JL from joints with and without effusion and in JL from painful and pain-free joints was compared and analyzed using Student’s t-test or Welch’s t-test.

Results Joint effusion on MRI was seen in 25 of 38 (65.8%) TMJs of patients with ID and OA; 19 of 30 joints (63.3%) with ID, and 6 of 8 joints (75%) with OA. Joint pain also was seen in 25 of 38 (65.8%) joints; 19 of 30 (63.3%) with ID, and 6 of 8 (75%) with OA. Joint effusion was shown in 20 of 25 (80%) painful joints, representing a significantly higher incidence than in pain-free joints (5 of 13; 38.5%) (Table 2). The mean total protein concentration was 1.22 mg/mL in pain-free joints and 2.15 mg/mL in painful joints. Although there was no difference in the yield of JL aspirates from painful joints and pain-free joins (84.3% and 86.9%, respectively), the total protein concentration in JL from painful joints was significantly higher than that from pain-free joints (Table 2). Furthermore, although the total protein concentration was distributed over a wide range, the highest value (5.36 mg/mL) in the painful group was twice that in the pain-free group (2.74 mg/mL) (Table 2). When the level of protein in joints with and without effusion was cornpaired, the concentration in JL from joints with effusion (2.12 mg/mL) was significantly

1190

JOINT EFFUSIONAND SYNOVIAL FLUID PROTEIN LEVELS

Protein Concentration (w/mL)

1

Incidence of Joint Effusion (%)

TMJ Painful (n = 25) Pain-free (n = 13) Total

20/25(80)

Yield of JL (%) (range) 84.3

*

5/13

(38.5)

25/38

(65.8)

Mean?SEM

Range

2.15 2 0.26

-c 16.9

(54.5-l 14) 86.9 + 16.8 (54.5-100) 85.2 + 16.6 (55.4-114)

1.22

-c 0.23

1.83

? 0.2

1

0.56-5.36

*

0.46-2.74 0.46-5.36

*P < .05.

higher than that from joints without effusion (1.27 mg/mL) (Table 3). To further investigate the relationship between joint elfusion, joint pain, and total protein levels in JL, joints were divided into 4 groups based on the presence of joint elfusion and joint pain, and the protein levels were compared (Fig 2). Among the 4 groups, the mean protein concentration in painful joints with effusion showed the largest value (2.23 mg/mL), which was significantly greater than that in pain-free joints without effusion (0.94 mg/mL). The mean protein concentration in painful joints (1.8 mg/mL) was also significantly greater than that in pain-free joints (0.94 mg/mL) in the group without joint effusion. Although considerable numbers of painful joints with effusion showed a greater total protein value than those in the other 3 groups, there were no significant differences in the mean protein concentration between any of the other groups.

Discussion

TMJ was first described by Harms et aL6 using Tl- and T2-weighted images with a spin-echo sequence. Schellhas et al7 showed that small fluid collections within the joint spaces might be detected with Tl- and T2-weighted and short gradient refocused acquisition in a steady state (GRASS) scans. They speculated that serous and proteinaceous effusions would exhibit a high intrinsic signal on both GRASS and SE long TR/long TE images, and bloody etfusions would exhibit a varying appearance.’ Westesson and Brook@ showed that there was a strong correlation between joint pain and joint effusion on MRL Although these studies suggested that joint effusion would reflect inflammatory changes occurring in the joint space, data were lacking concerning the biologic or clinical significance of joint effusion. In this study, JL was collected and the total protein level was measured and compared with the incidence of joint effusion on MRl and clinical symptom of pain in the joint area. This is, to our knowledge, the first study investigating the biochemical significance of joint effusion in the TMJ.

Because of the long T2 of free water, most body fluids are easily detected by T2-weighted images in which fluids have an intense signal relative to solid tissues.11,12Joint fluid is known to have a higher signal than fat on long TR and TE images.12 Clinically, joint effusion on MRl has been shown to indicate inflammation in other joints, including the knee and hi~.~-s,~~,~*MRI evidence of effusion in the

** 6

I A

Protein (mdmL)

TMJ Effusion + (n = 25) Effusion (n = 13) *P < .05.

Yield of JL (%) (raw> 84.0

2 16.3

Protein Concentration GWmL) Mean+SEM 2.12

2 0.27

1.27

+ 0.18

(54.5-l 14) 87.2

+ 17.6

(54.5-109)

1 *

-s 0.56-5.36

I

I

0’

8

I

Painful

Pain-free

Painful

Pain-free

Effusion+

Effusion+

Effusion-

Effusion-

0.46-2.23 FIGURE 2. The level of protein in the joint lavage fluid of the 4 groups divided by the presence or absence of joint effusion on MRI and joint pain. [**, P< .Ol.)

TAKAHASHI

ET AL

The major findings of this study were 1) joint effusion was demonstrated by MRI in 20 of 25 (80%) painful joints; a significantly higher incidence than for pain-free joints (5 of 13 joints, 38.5%); 2) protein levels were higher in the JL from painful joints than from pain-free joints; and 3) protein levels were greater in JL from joints with effusion than in joints without effusion. On rare occasions, a large volume of SF was obtainable from TMJs by direct aspiration. Cooksey and Girard14 studied joint effusion from in patients with TMJ synovitis due to familial Mediterranean fever. They aspirated 6 mL of a sterile, yellow, slightly cloudy fluid containing a large number of white blood cells. However, they did not report any evidence of joint effusion on MRI. Schellhas and Wilkes8 in their series, found that SFwas directly aspirated from only 2 joints percutaneously, although they did observe SF aspiration from 28 joints during surgery in a study of 100 consecutive TMJs. However, they did not examine any biologic properties of the SF. Kopp et all5 first reported a high protein concentration in SF collected from 7 of 3 1 TMJs with temporomandibular disorders using a direct aspiration technique. Ideally, fluid directly aspirated from the TMJ should be more appropriate for the biochemical analysis. However, direct aspiration seems to be an extremely difficult and unreliable method of acquiring a sulficient amount of SF for biologic analyses because of the small volume.16 Aghabeig i et all7 estimated the average SF volume of the superior joint space of the normal human TMJ to be 37 uL (0.9 to 185 yL). Therefore, JL aspirates in this study were obtained using a saline injection and aspiration technique (SF sampling by a dilution method) instead of by direct aspiration. Thus, a limitation of this study was its sampling procedure. Moreover, there was no external standard such as vitamin B,, (as used by Alstergren et ails), and it therefore was difhcult to assess how much of the saline or synovial fluid was lost during the process of JL sampling. One could speculate that the protein level depends on the JL sampling procedure and dilution method. Also, active bleeding during the sampling procedure can alter the total protein level of JL samples. However, these objections may be countered as follows: First, the volume of normal saline injected was constant (2 mL) and the yield of aspirate after the pumping procedure with 10 repetitions was about 85%. Thus, the concentration of the components in the diluted JL was representative of their amount in the JL and was independent of the amount of aspirate obtained. Second, JL samples that represented less than 50% of the injected volume were excluded. Furthermore, JL samples that contained fresh blood resulting from the trauma of the sampling procedure were also excluded.

1191 Care should be taken when considering the differences in the biochemical components of SF obtained by direct aspiration and by JL as used in this study. Kubota et al19 showed that the level of interleukin (IL)-lp in JL (a diluted SF) was higher than that in SF collected by direct aspiration. They pointed out the possibility that additional IL-lj3 retained in the inflamed synovium or cartilage was released into the JL during the aspiration procedure. Therefore, the protein level obtained in this study may not represent the intrinsic concentration in the joint space. The protein level shown in this study has to be considered as representing that contained in the SFplus that washed out/released into the JL from the synovium and cartilage during the sampling procedure. Nonetheless, it is of biochemical or clinical importance to compare the protein levels and joint effusion. Future studies will be necessary to compare the biochemical contents of JL collected by our method and that obtained by direct aspiration. Recently, it was reported that variable degrees of inflammation exist in some temporomandibular disorders (TMD). Gynther et a120noted that arthroscopic signs of synovitis (capillary hyperemia and synovial hyperplasia) correlated with microscopic findings in synovial biopsy specimens. Quinn and Bazan21 noted that the levels of prostaglandin E2 detected in the synovial fluid of inflamed, dysfunctional TMJs had a strong correlation with synovitis and was an index of clinical joint pathology. Takahashi et a122noted that several proinflammatory cytokines, including IL-l@, IL-6, and tumor necrosis factor CY(TNFa) were detectable in patients with ID and OA of the TMJ and also found a strong correlation between the detection of IL-lp and pain in the joint area. These data suggest that many inflammatory mediators are contained in SF from patients with ID and OA of the TMJ and also suggest the involvement of these mediators in the pathogenesis of ID and OA of the TMJ. Takaku et al9 surgically confirmed serous joint elhtsion in 30 joint spaces that showed high signal intensities on MRI and demonstrated that the finding of joint effusion suggests the presence of synovitis due to disc damage or degeneration. Israel et al23 showed that the protein concentration in JL was significantly higher in joints with synovitis diagnosed by arthroscopy than in those without synovitis using the same JL sampling procedure as this study. Therefore, elevated levels of protein, especially in painful joints, suggest the presence of inflammatory changes such as synovitis (capillary hyperemia and synovial hyperplasia). Capillary hyperemia and synovial hyperplasia will lead to an increase in vascular permeability, resulting in the exudation of white blood cells as well as molecules such as inflammatory mediators and various proteins into the superior and inferior joint spaces of

1192 the TMJ. Clinically, these inflammatory changes result in pain, soft tissue swelling, crepitus, and disability.7 Therefore, the elevated levels of protein in joints with effusion seen in this study suggest that the efl?usion may, at least in part, reflect molecules that include protein extravasated into the joint space. Likewise, elevated levels of protein in painful joints also suggest the presence of inflammatory changes. Osteoarthritis, or degenerative joint disease, has been emphasized in the pathogenesis of temporomandibular disorders.z4,z5 It is hypothesized that overloading of the joint induces degradation of the molecular components of the cartilage, resulting in excessive release of proteoglycan degradation products into the synovial fluid and subsequent cartilage breakdown. Israel et alz6 showed that the increased levels of keratan sulfate, a product of proteoglycan degeneration, are related to OA of the TMJ. Although statistically not significant in this study, the mean protein concentration in JL from joints with OA (2.39 mg/mL) was higher than the protein level in JL from joints without OA (1.68 mg/mL). Therefore, increased levels of protein seen in joints with effusion also may represent degradation of the molecular components of the cartilage matrix. Westesson and Brooks13 reported that the incidence of effusion ln joints with arthrosis (27%) was lower than in those with disc displacement without reduction (50%). In our study, the incidence of effusion in joints with OA (75%) was higher than in those without OA (63.3%). The difference between the 2 studies may reflect the severity of the disease activity or the presence of synovitis. In fact, the incidence of pain in joints with OA (75%) was higher than in those without OA (63.3%) in our series. It should be pointed out that the differences in the mean protein concentration between painful and pain-free joints were comparable to those between the joints with and without MRI evidence of joint effusion. Although statistically not significant, the mean protein concentration in pain-free joints with effusion was higher than in pain-free joints without effusion Fig 2). Conversely, the mean protein concentration in painful, effusion-free joints was significantly higher than that in pain-free joints that were effusionfree (Fig 2). Therefore, the elevated levels of protein, which correlated with MRI evidence of joint effusion, may be correlated with joint pain rather than joint elision. A future, well-designed study using a larger number of TMJ samples will be required to determine which factors contribute more to the level of protein in synovial fluid. Although joint effusion was more frequently seen in painful joints than in pain-free joints, there was no pain in 5 joints with effusion. Among these joints, the protein concentration in 2 joints was relatively high (both 2.73 mg/mL). However, in the 2 other joints, the

JOINT EFFUSION AND SYNOVL4L

FLUID PROTEIN LEVELS

protein concentration was very low (0.87 mg/mL and 0.67 mg/mL). In addition, effusion was not observed in 5 joints with clinical evidence of joint pain. The mean concentration of protein from these joints was relatively high (1.80 mg/mL). Conversely, even in painful joints with effusion, the protein levels were relatively low in several joints (Fig 2). A possible explanation for the discrepancy between joint pain, joint effusion, and protein level in some joints shown in this study is that joint effusion might be caused by components or molecules in SF other than protein. Alternatively, pain in the joint area may be independent of the inflammatory changes or effusion in some joints. It is well recognized that effusion is sometimes observed in joints without pain.1s,27 Even joints in asymptomatic volunteers sometimes show MFU evidence of joint effusion. 28 The difference in the incidence of joint effusion in this study compared with other studies may be related to the method of evaluation or the characteristics of the subjects such as age, duration of symptoms, and disease activity. In this study, the subjects all had anterior disc displacement without reduction and advanced ID or OA. Furthermore, these patients showed resistance to conservative therapy, and all underwent pumping manlpulation. Therefore, a controlled study to examine the relationship between the levels of protein in JL and joint effusion in asymptomatic volunteers is definitely needed before one can conclude that joint effusion or protein levels reflect inflammatory changes occurring in the TMJ. Among of the advantages of MBI as a diagnostic modality is its lack of ionizing radiation and its noninvasive characteristics. If joint effusion seen on MIU represents an inflammatory response, it can serve as a marker for the detection of early, inflammatory arthralgia in the TMJ. Further intensive studies at both the molecular and clinical levels will be necessary to determine what kind of protein(s) or other molecule(s) in SF increase the signal intensity noted in MRI with joint effusion, and whether this could serve as a marker of inflammatory changes occurring in the TMJ. Acknowledgment The authors thank Drs Noriaki Tomura Department of Radiology, Akita University their contribution to this project.

and Ryuji Sashi at the School of Medicine, for

References 1. Larheim TA: Current trends in temporomandibular joint imaging. Oral Surg Oral Med Oral Path01 Oral Radio1 Endod 80:555, 1995 2. Turner DA, Prodromos CC, Petasnick JP, et al: Acute injury of the ligaments of the knee: Magnetic resonance evaluation. Radiology 154:717,1985 3. Beltran J, Caudilf JL, Herman IA, et al: Rheumatoid arthritis: MR imaging manifestations. Radiology 165: 153, 1987

G&LAN

1193

W. GYNTHER

4. Schweitzer ME, Falk A, Pathria M, et al: MR imaging of the knee: Can changes in the intracapsular fat pads be used as a sign of synovial proliferation in the presence of an effusion? AJR Am J Roentgen01 160:823, 1993 ME, Magbalon MJ, Fenlm JM, et al: Effusion criteria 5. Schweitzer and clinical importance of clenohumeral joint fluid; MR imaging evaluation. Radiology 194:821, 1995 6. Harms SE, Wilk RM, Wolford LM, et al: The temporomandibular joint: Magnetic resonance imaging using surface coils. Radiology 157:133,1985 CM, et al: The diagnosis of 7. Schellhas KP, Wilkes CH, Peterson temporomandibular joint disease: Two-compartment arthrography and MR. AJNR Am J Neuroradiol9:579, 1988 8. Schellhas KP, Wilkes CH: Temporomandibular joint inflammation: Comparison of MR fast scanning with Tl- and TZ-weighted imaging techniques. AJRAm J Roentgen01 153:93, 1989 S, Toyoda T, Sano T, et al: Correlation of magnetic 9. Takaku resonance and surgical findings in patients with temporomandibular joint disorders. J Oral Maxillofac Surg 53:1283, 1995 10. Takahashi T, Kondoh T, Kamei K, et al: Elevated levels of nitric oxide in synovial fluid from patients with temporomandibular disorders. Oral Surg Oral Med Oral Path01 Oral Radio1 Endod 82:505, 1996 11 Beltran J, Noto AM, Herman LJ, et al: Joint effusions: MR imaging. Radiology 158:133, 1986 DG, Rao V, Dalinka M, et al: MRI of joint fluid in the 12 Mitchell normal and ischemic hip. AJR Am J Roentgen01 146: 12 15, 1986 P-L, Brooks SL: Temporomandibular joint: Relation13 Westesson ship between MR evidence of effusion and the presence of pain an disk displacement. AJRAm J Roentgen01 159:559, 1992 14 Cooksey DE, Girard K: Temporomandibular joint synovitis with effusion in familial Mediterranean fever. Oral Surg 47: 123, 1979 B, Clemmensson E: Clinical, microscopi15 Kopp S, Wennenberg cal, and biochemical investigation of synovial fluid from temporomandibular joints. Stand J Dent Res 91:33, 1983 16 Zardeneta G, Milam SB, Schmitz JP: Elution of proteins by continuous temporomandibular joint arthrocentesis. J Oral Maxillofac Surg 55:709, 1997 17 Aghabeigi B, Henderson B, Hopper C, et al: Temporomandibular joint synovial fluid analysis. Br J Oral Maxillofac Surg 3 1: 15, 1993

P, Appelgren A, Appelgren B, et al: Determination of 18. Alstergren temporomandibular joint fluid concentrations using vitamin B~z as an internal standard. Eur J Oral Sci 103:214, 1995 E, Imamura H, Kubota T, et al: Interleukin lp and 19. Kubota stromelysin (MMP3) activity of synovial fluid as possible markers of osteoarthritis ln the temporomandibular joint. J Oral Maxillofac Surg 55:20, 1997 20. Gynther GW, Holmulund AB, Reinholt FP: Synovitis in internal derangement of temporomandibular joint: Correlation between arthroscopic and histologic findings. J Oral Maxillofac Surg 52:913, 1994 21. Quinn JH, Bazan NG: Identification of prostaglandin Ez and leukotriene Bq in the synovial fluid of painful, dysfunctional temporomandibular joints. J Oral Maxillofac Surg 48:968, 1990 22. Takahashi T, Kondoh T, Fukuda M, et al: Proinflammatory cytokines detectable in synovial fluids from patients with temporomandibular disorders. Oral Surg Oral Med Oral Path01 Oral Radio1 Endod 85:135, 1998 concentra23. Israel HA, Ellis P, Furgang D: Synovial fluid protein tion in patients undergoing temporomandibular arthroscopy. J Dent Res 69:296, 1990 (abstr) 24. Quinn JH: Pathogenesis of temporomandibular joint chondromalacia and arthralgia. Oral Maxillofac Surg Clin North Am 1:47, 1989 B, DeBlont LGM, Boering G: Osteoarthrosis as the 25. Stegenga cause of craniomandibular pain and dynfunction. J Oral Maxillofat Surg 47:249, 1989 26. Israel HA, Saed-Nejad F, RatclitTe A: Early diagnosis of osteoarthrosis of the temporomandibular joint: Correlation between arthroscopic diagnosis and keratan sulfate level in the synovial fluid. J Oral Maxillofac Surg 49:708, 1991 K, Nishida K, Bessho T, et al: MRI evidence of high 27. Murakami signal intensity and temporomandibular arthralgia and relating pain: Does the high signal correlate to the pain? Br J Oral Maxillofac Surg 34:220,1996 28. Katzberg RW, Westesson PL, Tallents RH, et al: MR assessment of joint fluid in the temporomandibular joint of asymptomatic volunteers. Presented at the Annual Meeting of American Society of TMJ Surgeons, 1993 (abstr)

J Oral Maxillofac Surg 57: 1 193-l 194, 1999

Discussion Relationship Between Joint Effusion, Joint Pain, and Protein levels in Joint Lavage Fluid of Patients With Internal Derangement and Osteoarthritis of the Temporomandibular Joint G&an

W. Gynther, DDS, PhD

Associate Professor and Head, Department of Oral and Maxillofacial Surgery, Visby Hospital, Sweden; Associate Professor, Department of Oral and Maxillofacial Surgery, Huddinge University Hospital, Karolinska Institute, Sweden; [email protected]

e-mail:

This study reflects the increased interest and research during the last few years on the biochemical events associated with temporomandibular joint (TMJ) disorders.‘” The authors conclude that high protein levels in joint lavage fluid

of the TMJ may be related to inflammatory changes in patients with TMJ internal derangement and osteoarthritis. The same relation with inflammatory changes was found respect to joint effusion on magnetic resonance imaging. So far, TMJ arthroscopy, with its unique possibilities for simultaneous biopsy for histologic analysis, is the gold standard for classification of inflammatory changes.9.l1 Also, for the diagnosis of degenerative changes in the TMJ, arthroscopy seems to be a reliable method, with extremely rare complications.10,12 Therefore, I was disappointed that arthroscopy was not performed in all patients to correlate the arthroscopic and histologic classification with the protein levels in joint lavage fluid and the presence of joint effusion on magnetic resonance imaging. In only 13 of 38 joints was arthroscopy performed. The arthroscopic findings included superficial fibrillation of the cartilage and the disc, exposure of subchondral bone, and disc perforation, and this supported the diagnosis of osteoarthritis. Surpris-