Int. J. Oral Maxillofac. Surg. 1998; 27:268-273 Printed in Denmark. All rights reserved

Copyright �9 Munksgaard 1998

lnm*'nauonalJoumal of

Ord Max ofacial Surgery

ISSN 0901-5027

Aesthetic

Temporomandibular joint osteoarthritisand crystal deposition diseases: a study of crystals in synovial fluid lavages in osteoarthritic temporomandibular joints

and reconstructive surgery

Leonore C. Dijkgraaf 1, Robert S. B. Liem =, Lambert G. M. de Bont 1 1TMJ Research Group, Department of Oral and Maxillofacial Surgery, University Hospital Groningen; PI'MJ Research Group, Laboratory for Cell Biology and Electron Microscopy, University of Groningen, Groningen, The Netherlands

L. C. Dijkgraaf, R. S. B. Liem, L. G. M. de Bont: Temporomandibular joint osteoarthritis and crystal deposition diseases: a study of crystals in synovial fluid lavages in osteoarthritic temporomandibular joints. Int. J. Oral Maxillofac. Surg. 1998; 27." 268-273. �9 Munksgaard, 1998

Abstract. To study the presence of crystals in synovial fluid lavages of osteoarthritic temporomandibular joints (TMJs), in order to evaluate the possible role of these crystals in the osteoarthritic (OA) process, synovial fluid lavage samples of the upper joint compartment from 44 TMJs were obtained prior to arthroscopy. The OA group consisted of 32 TMJs. The control group consisted of 12 TMJs that had been diagnosed with other nonosteoarthritic conditions. The lavage samples were analysed as wet preparations, unstained and stained, with ordinary light, polarized light and compensated polarized light for the presence of crystals and white blood cells. One sample was prepared for subsequent electron microscopic (EM) examination. Synovial fluid lavage analysis of osteoarthritic TMJs did not show any monosodium urate monohydrate or calcium pyrophosphate dihydrate crystals. However, in three lavages, particles which possibly contained calcium were identified with alizarin red S staining. White blood cells were occasionally seen. Synovial fluid analysis of the lavages of the control TMJs did not reveal any crystals. EM examination of synovial fluid lavage from an osteoarthritic TMJ failed to clearly show crystal formation. Concurrence of TMJ crystal deposition and OA appears less prominent than in other synovial joints. We conclude that crystals probably do not play an important role in TMJ OA.

Key words: temporomandibular joint; osteoarthritis; crystals; synovial fluid.

Accepted for publication 6 January 1998

Concurrence of crystal deposition and osteoarthritis (OA) in synovial joints, in general, is well known 10,13,52. It is still unknown whether crystal deposition in OA is a result, a contributing factor, or possibly both. Three crystal types have been predominantly found: monoso- dium urate monohydrate (MSUM), cal-

cium pyrophosphate dihydrate (CPPD), and basic calcium phosphates (BCPs). BCPs encompass hydroxyapatite, trical- cium phosphate, and octacalcium phos- phate. Deposits of MSUM, CPPD, and BCPs can also form in the absence of OA. These are clinically associated with acute or chronic gouty arthritis

(MSUM deposition), acute pseudogout or chronic pyrophosphate arthropathy (CPPD deposition), and acute periar- thritis or chronic destructive joint dis- ease (BCP deposition) 14'43. However, mixtures of several different types of crystals are frequently found in crystal deposits 15. MSUM deposition is highly

T M J osteoarthritis and crystals 269

related to hyperuricemia, i.e. a super- saturation for urate in serum, which is caused by a generalized metabolic ab- normality 14,43. Urate is an endproduct of human purine metabolism. Although hyperuricemia predisposes individuals to gout, it should be noted that hyper- uricemia may remain asymptomatic and does not necessarily result in MSUM deposition or clinical symp- toms. Because hyperuricemia is a gener- alized disorder, precipitation of MSUM could occur anywhere. The site is de- cided by secondary local tissue factors that promote crystallization ]a. In con- trast to MSUM deposition, deposition of other crystal types is predominantly caused by a localized disorder, fre- quently a local increase in metabolic ac- tivity or loss of inhibition of crystalliza- tion. CPPD deposition is probably caused by an increased production of inorganic pyrophosphate by local chon- drocytes. Inorganic pyrophosphate is produced in many intracellular reac- tions, including the pyrophosphorylysis of nucleotide triphosphates during bio- synthesis of cell components and ma- trix-destined molecules 43. Relatively little is known of BCP deposition. Re- search has been hampered by the small size of the crystals and their strong ten- deney t o occur in mixtures with other crystals 14.

In the temporomandibular joint (TMJ), crystal deposition appears to be quite rare. Only few reports on MSUM deposition 4"23'26'27,32'39,48,54,55 and CPPD deposition t 1,12,16,17,30,32,36-3s,40,42,44,49-51,

57,58 have appeared in the literature. Re- cently, hydroxyapatite deposition has been described in the TMJ in a patient with renal failure 5. In synovial mem- brane of osteoarthritic TMJs, a few hy- droxyapatite-like crystals were found in an ultrastructural study Is. Obvious concurrence of crystal deposition dis- eases and OA of the TMJ has never been reported.

In synovial fluids of osteoarthritic knee joints, CPPD and BCP crystals have frequently been demon- strated 9,24,25,29,33,35,41,46,56. A mixture of

CPPD and BCP crystals was found more often than either alone 33. CPPD and BCP crystals correlated with the patient's age and joint degeneration, re- spectively 9'29'33'35'46. Although TMJ OA occurs at a young age (20s and 30s), arthroscopically and microscopically it is characterized by degeneration ]9,2~ The aim of this study was to study the presence of crystals in synovial fluid

lavages of osteoarthritic TMJs, in order to evaluate the concurrence of crystal deposition and OA and the possible role of these crystals in the osteoarthrit- ic process.

Material and methods SynovlM fluld 18v~ sample u~ecth~n

Synovial fluid lavage samples were obtained prior to arthroscopy of the upper joint com- partment from 44 TMJs, including both os- teoarthritic and nonosteoarthritic (control) TMJs. The OA group consisted of 32 TMJs (26 women, 6 men; mean age, 31.2_+9.3 years) that had been diagnosed with TMJ OA. The control group consisted of 12 TMJs (9 women, 3 men; mean age, 29.7• years) that had been diagnosed with other nonos- teoarthritic conditions, including painful hypermobility and overloading. The sex ratio and mean age did not differ significantly be- tween the OA and the control group. Ex- clusion criteria for both groups had included the presence of TMJ growth disorders, other forms of arthritis, the presence of systemic connective tissue disorders, as well as the presence of nonarticular disorders in the oral and maxillofacial area. Additional exclusion criteria were previous TMI surgery and in- tracapsular injections of corticosteroids. For each TMJ, the duration of clinical signs and symptoms was recorded.

Arthroscopy was performed with the pa- tients under general anaesthesia via nasoen- dotracheal intubation and under optimal neuromnscular relaxation. The joint anatomy was palpated during ventral-dorsal joint movement to identify the maximum concavity of the glenoid fossa. Subsequently, a 19-gauge needle on a l ml syringe was introduced into the posterior recess of the upper joint com- partment, while the condyle was held in maxi- mum ventral position. As lavage fluid, 1 ml water for injection (B. Braun Melsungen AG, Melsungen, Germany) was used. This sterile, bidistilled water was injected and aspirated at least five times, allowing sufficient intervening time for an equilibrium to arise. Lavage samples that macroscopically demonstrated contamination with blood were excluded from this study. Ten of the 32 synovial fluid lavages of ostcoarthritic TMJs and two of the 12 syn- ovial fluid lavages of the control group were mixed immediately with a fixative containing 2% paraformaldehyde and 0.5% glutaralde- hyde in a 0.1 M phosphate buffer (pH 7.4). All samples were refrigerated until analysis. Arthroscopically, the presence of various vari- ables was recorded, including degeneration of the articular cartilage, and hypervascularity, hyperemia, and redundancy of the synovial tissues-. The synovial fluid lavage samples were analysed immediately postoperatively.

Synovlal fluid lavage sample analysis

Synovial fluid lavage samples were analysed as wet preparations, unstained and stained,

with ordinary fight, polarized light and com- pensated polarized light; A drop of lavage was placed on a precleaned slide, covered with a clean coverslip and examined with or- dinary light microscopy (LM) and polarized LM without and with a first order red com- pensator to identify MSUM and CPPD crys- tals by their morphology and sign of birefrin- genee. For BCP crystal identification, a drop of iavage was mixed on a slide with a drop of filtered (0.22/an pore size) 2% alizarin red S stain solution (pH 4.2) and subsequently examined with ordinary LM. Wright's stain was used to identify crystals and white blood cells, whilst methylene blue stain was used to stain white blood cells. Each separate identi- fication was performed at least five times by two investigators. Two samples of the unfixed and two of the fixed osteoarthritic synovial fluid lavages were centrifuged prior to analy- sis. Based on light microscopic fndings, one of the synovial fluid lavages containing aliz- arin red S positive staining was selected for subsequent electron microscopic examin- ation. The remaining synovial fluid lavage was fixed with a fixative containing 2% para- formaldehyde and 0.5% glutaraldehyde in a 0.1 M phosphate buffer (pH 7.4), and centri- fuged at 2000 rpm. After removal of the fixative, the pellet was washed with washing solution (0.1 M phosphate buffer, pH 7.4) in a water bath at 50~C. A few drops of 2% agar at 50~ were added to the pellet and shaken to suspend the pellet components into the agar, following which it was centrifuged. The newly formed pellet was cut into small blocks that were postfixed for one hour with 1% os- mium tetroxide and embedded in Epon. Sem- ithin sections of 1/an thickness were cut and studied by LM to identify the areas of inter- est. Subsequently, ultrathin sections 40-60 nm thick were cut with a diamond knife and mounted on copper grids. The sections were stained with uranyl acetate and lead citrate, and examined by electron microscopy (EM).

A correlation between the presence of syn- ovial fluid calcific material and arthro- scopically scored cartilage degeneration was analysed with Cohen's Kappa.

Results

Synovial fluid analysis of the fixed and unfixed lavages of the osteoarthritic TMJs did not reveal any MSUM or CPPD crystals with ordinary or polar- ized light. Alizarin red S staining, how- ever, demonstrated in three lavages (9%) several deeply stained clumps sug- gestive of calcium-containing particles. With polarized light, these particles were brightly birefringent (Fig. 1). White blood cell staining in the unfixed lavages showed only remnants of leuko- cytes. White blood cell staining in the fixed lavages did show several intact leukocytes, although no intracyto- plasmic inclusions such as crystal-like

/ p

B

270 D i j k g r a a f et al.

Fig. 1. A) Alizarin red S positive stainjng, showing bright birdringcner (arrow) with polarized light (Alizarin red S stain x160). B) Same as in (A), under compensated polarized light.

u

A [3

Fig. 2. A) Alizarin redS negative staining, showing two highly bkefringent starehlike particles (arrowheads) with polarized light (Alizarin red S stain • 160). B) Same as in (A), under compensated polarized light.

particles were found. Erythrocytes were generally observed in every lavage. Fi- nally, in four samples, several particles were found with polarized light, show- hag a strongly birefringent beach ball/ maltese cross appearance suggestive of starch globules (Fig. 2). In the centri- fuged synovial fluid lavage samples, no identifications could be made because of the clustering of ceils and particles.

Synovial fluid analysis of the lavages of the control TMJs did not reveal any crystals, but erythroeytes and oc- casional lenkocytes were observed.

EM examination of one synovial fluid lavage of an osteoarthritic TMJ, demonstrating alizarin red S positive staining, did not convincingly show the presence of crystals.

No correlation could be found be-

tween the presence of synovial fluid cal- cific particles and arthroscopically scored cartilage degeneration.

Discussion

Synovial fluid lavage analysis of os- teoarthritic TMJs did not show any MSUM or CPPD crystals, while with alizarin red S staining the presence of possibly calcium-containing particles was identified in only three lavages. Alizarin red S staining is nonspecific for individual crystal species but is a useful screening stain for calcitic material. Staining at acidic pH (4.2) increases its specificity for hydroxyapatite but is still insufficient for definite characteriza- tion. The presence of hydroxyapatite crystals, which are BCP crystals, in syn-

ovial fluids from osteoarthritic knee joints has been related to the severity of joint damage in OA 9'24'29'33'35'46. Yet the relative incidence of crystal deposition in OA, other joint diseases, and normal control groups has not yet been estab- lished ~3. Alizarin red positive particles, however, as well as birefringent crystals, have also been found in synovial fluids obtained from normal knee joints and even at a higher frequency than found in the lavages of osteoarthritic TMJs in the present study 24. Concurrence of TMJ crystal deposition and OA ap- pears, therefore, less prominent than in other synovial joints and crystals prob- ably do not play an important role in TMJ OA.

Crystals in synovial fluid may have been derived from three possible

TMJ osteoarthritis and crystals 271

sources ~3. First, the crystals may be wear particles from the normally calci- fied cartilage zone or underlying bone that has been exposed through tissue damage in OA. Second, the crystals may have originated from abnormal cartilage mineralization and subsequent shedding. In osteoarthritic articular cartilage, active chondrocytes have been shown to form matrix vesicles that act as nucleation sites for aggregates of apatite crystals 2. In the deeper layers of TMJ osteoarthritic cartilage, matrix vesicles have been shown to contain apatite crystals 6. These were predom- inantly found in the region adjoining the calcified cartilage zone as well as in this calcified cartilage zone itself. In normal articular cartilage, mineraliza- tion is inhibited by several factors, in- cluding avascnlarity ~4. In this respect, the arthroscopic finding of angiogen- esis 19, i.e. neovascularization, of the ar- ticular eminence is quite si~ificant. Third, crystals in synovial fluid may have been derived from areas of meta- plasia in the synovial membrane or joint capsule, but there is little evidence to support this concept. A recent study, however, of so-called rice bodies, which are supposedly synovium-derived en- doarticnlar loose bodies, in synovial fluid from osteoarthritic knee joints, showed that these were composed of fibrin as well as numerous intra- and extracellular apatite and CPPD crys- tals s. They contained synovial intima cells and few inflammatory ceils, as well as collagen fibres, chondrocytes and other cartilage components. A study of loose endoarticnlar bodies in patho- logic TMJ synovial fluid, however, failed to demonstrate the presence of hydroxyapatite in these bodies, but showed that they were composed of cal- cite 47. In our opinion the presence of crystals in loose bodies in the synovial fluid does not directly imply synovial or capsular metaplasia until the origin of these bodies has been identified.

A correlation between the presence of calcific material in the TMJ synovial fluid lavages and arthroscopically scored cartilage degeneration could not be demonstrated in this study. This was probably caused by the limited number of lavages demonstrating these crystals. One of the two problems with this study was the dilution of TMJ synovial fluid through the lavage. The average volume of synovial fluid of pathologic TMJs is 37/zl 1. Because this volume is too small for fluid to be aspirated, lavage of the

joint was the only way to obtain the TMJ synovial fluid. In this study, 1 ml water for injection was used as lavage fluid rather than saline to prevent de novo sodium chloride crystal precipita- tion in the slides. Inherent to the use of lavage fluid is the considerable dilution of synovial fluid, which may result in crystal dissolution. It has also yet to be determined what concentration of crys- tals in TMJ synovial fluid would be considered clinically significant since these crystals have also been found in the synovial fluid of other normal joints. In addition to the dilution prob- lem, an underdiagnosis may have oc- curred due to the low threshold of re- liable crystal identification by LM 31,56. Until now, two methods have been pro- posed to determine the dilution factor in TMJ synovial fluid lavages m, but both have limitations 2~. In future TMJ synovial fluid analysis for crystals, this dilution problem should be solved and the concentration of crystals that has a clinical relevance should be established. For this reason, the use of additional analytical techniques may be necessary, e.g. nuclide binding 34 or hypochlorite solution 45. Based on the findings in this study, we believe that crystals play only a limited role in the osteoarthritic pro- ceSS.

The second problem with this study was the lapse of time between the mo- ment of lavage and synovial fluid analy- sis (2-4 hours). It has been shown that leukocyte counts drop after one hour, whereas after six hours they may have been reduced by as much as 50% 53 . This may explain why white blood cell stain- ing in the unfixed lavages in this study showed only remnants of leukocytes. However, this could also have been caused by the use of pure water rather than saline as lavage fluid and the consequent difference in osmotic pres- sure. Overall, only few leukocytes were found. White blood cell counts of less than 2000/mm 3 are generally considered noninflammatory 2s. A white blood cell count was not performed in this study because of the dilution problem men- tioned above, and because the leuko- cytes observed in the lavages may have been derived from blood that had con- taminated the lavage due to the ma- nlpnlation of the needle. The presence of erythrocytes, as well as the presence of leukocytes in the control lavages may support this possibility. The absence of increased numbers of leukocytes in the synovial fluid of osteoarthritic TMJs is

in concurrence with the relative absence of inflammatory cells in the synovial membrane of these TMJs ~.

An unexpected finding in the syn- ovial fluid lavages was the presence of starchlike globules, which probably de- rived from the use of rubber gloves dur- ing the arthroscopy. Although starch contamination might have occurred after the lavage, we support the recom- mendation by BRONS~IN 7 regarding glove washing in order to prevent the occurrence of iatrogenically induced starch synovitis.

Acknowledgments. The authors wish to thank Dr G. Zardeneta, Department of Oral and Maxillofaeial Surgery, University of Texas Health Science Center at San Antonio, Tex- as, USA, for his editorial input.

References

I. AGHABEIGI B, t'IENDn~SON B, HOPPER C, I-IAnRIS M. Temporomandibular joint synovial fluid analysis. Br J Oral Maxil- lofac Surg 1993: 31: 15-20.

2. ALI SY. Mineral-containing matrix ves- icles in human osteoarthrotic cartilage. In: NugI G, ed.: The etiopathogenesis of osteoarthrosis. London: Pitman Medical, 1980: 105-16.

3. ALSTERGR~ P, AP~IX3n~ A, APPEL- GREN B, KOPP S, LUNDI~ERG T, THEO- OORSSON E. Determination of temporo- mandibular joint fluid concentrations using vitamin BI2 as an internal stan- dard. Eur J Oral ScJ 1995: 103: 214-8.

4. BAm.HAUS S, M~s K, VOOL TH. Infra- temporaler Gichttophus - eine serene Differentialdiagnose zur primaren Parot- iserkrankung. Laryngol Rhinol Otol 1989: 68: 638--41.

5. BEST JA, SHAPIRO RD, KALMAR J, WEST- F.SSON P-L. Hydroxyapatite deposition disease of the temporomandibular joint in a patient with renal failure. J Oral MaxiUofax Surg 1997: 55: 1316-22.

6. DB Bo1,rr LGM, LW.M RSB, Boegr~G G. Ultrastructure of the articular cartilage of the mandibular condyle: ageing and degeneration. Oral Surg 1985: 60: 631- 41.

7. BnoNsrEIN SL. Starch synovitis of the TMJ treated by arthroscopic debride- ment: a case report. J Craniomandib Dis- ord Facial Oral Pain 1988: 2: 9-12.

8. BucKI B, LANSAMAN J, JANSON X, et al. Rice bodies containing calcium micro- crystals in osteoarthritic synovial fluids. A report of four cases studied by trans- mission electron microscopy. Rev Rhum 1994: 61: 359-64.

9. C~,ROLL GJ, STU.~T P.A, ARMSTRONG JA, Bn~mArm PD, L ~ G BA. Hydroxy- apatite crystalsare a frequent finding in osteoarthritic synovial fluid, but are not related to increased concentrations of

2 7 2 Dijkgraaf et al.

keratan sulfate or interleukin lfl. J Rheu- matol 1991: 18: 861~5.

10. CrIEtn~G HS, RYAN LM. Role of crystal deposition in matrix degradation. In: WOESSNER JF, HOWELL DS, eds.: Joint cartilage degradation. Basic and clinical aspects. New York: Marcel Dekker, 1993: 209-23.

11. C-MtrONG R, PIPER MA. Bilateral pseudogout of the temporomandibuiar joint: Report of a case and review of literature. J Oral Maxillofac Surg 1995: 53: 691-4.

12. CO~mELLES R, GASQtmT F, PAGES M. Cal- cinosis of the temporomandibular joint: apropos of a case and critical review of the literature. Rev Stomatol Chir Mar, il- lofac 1992: 93:183-4 [in French].

13. DmPPE P, WATT I. Crystal deposition in osteoarthritis: an opportunistic event? Clin Rheum Dis 1985: 11: 367-92.

14. DIEPPE R CALVERT E Crystals and joint disease. London: Chapman & Hall, 1983.

15. D1EPPE P, CAMPION G, DOHERTY M. Mixed crystal deposition. Rheum Dis Clin North Am 1988: 14: 415-26.

16. DtrKGRAXF LC, DE Borer LGM, LmM RSB. C~cium pyrophosphate dihydrate crystal deposition disease of the tem- poromandibular joint: report of a case. J Oral MaxiUofae Surg 1992: 50: 1003- 9.

17. D u K t ~ LC, LmM RSB, DE BONT LGM, BOERn~ G. Calcium pyrophos- phate dihydrate crystal deposition dis- ease: a review of the literature and a tight and electron microscopic study of a case of the temporomandibttlar joint with nu- merous intracellular crystals in the chon- droeytes. Osteoarthritis Cart 1995: 3: 35- 45.

18. DIJKGRmW LC, Lm~t RSB, DE BOlqT LGM. Ultrastructural characteristics of osteoarthritic temporomandibular joints. J Oral Maxillofac Surg 1997: 55: 1269- 79.

19. DUKGRAAF LC, SPIJKI~VET FKL, DE Bol, rr LGM. Arthroseopic findings in os- teoarthritic temporomandibular joints. J Oral Maxillofac Surg (submitted).

20. DIJK~RA~ LC, DE Borer LGM, BOERING G, Lima RSB. The structure, biochemis- try, and metabolism of osteoarthritic car- tilage: a review of the literature. J Oral Maxillofac Surg 1995: 53: 1182-92.

21. DUK~RAAF LC, Wlmss JB, DE BONT LGM. Endothelial cell stimulating angiogenesis factor in synovial fluid of osteoarthritic temporomandibular joints. Pitfalls in temporomandibuiar joint syn- ovial fluid analysis. In: DUKGgtO, F LC: Temporomandibular joint osteoarthritis and crystal deposition diseases. Structure and pathogenesis. Thesis. University of Groningen, 1997.

22. DUKGRAAF LC, LIEM RSB, DE BONT LGM. Synovial membrane involvement in osteoarthritic temporomandibular joints. A light microscopic study. Oral Surg 1997: 83: 37346.

23. FAgs I. Ein Gichttophus im Kiefergelenk mad auf der Tuba Eustachii. Laryngol Rhinol Otol 1983: 62: 574-7.

24. FAWTrmOP F, HoRrou J, SWAN A, Htrr- TON C, Dol-maTY M, DmPPE P. A com- parison of normal and pathological syn- ovial fluid. Br J Rheumatol 1985: 24: 61- 9.

25. FIFE RS, RACI-IOW JVg, RYAN LM. Syn- ovial fluid and plasma levels of cartilage matrix glycoprotein in arthritis. Calcif Tissue Int 1994: 55: 100-2.

26. FLUUR E, HAVERLING M, MOLIN C. Gout in the temporomandibuiar joint. Report on a case. J Otorhinolaryngol Relat Spec 1974: 36: 16-20.

27. G)~RTNER F, PAST A. Arthritis urica und die Mitbeteiligung der K.iefergelenke. ZWR 1984: 93: 812-4.

28. GATTER RA. A practical handbook of joint fluid analysis. Philadelphia: Lea & Febiger, 1984.

29. GIalLISCO PA, SCHUUACI-IER HR, HOL- LANDER JL, SOPER KA. Synovial fluid crystals in osteoarthritis. Arthritis Rheum 1985: 28: 511-5.

30. Good AE, UPTON LG. Acute temporo- mandibular arthritis in a patient with bruxism and calcium pyrophosphate de- position disease. Arthritis Rheum 1982: 25: 353-5.

31. GOSDON C, SwAN A, DmPPE P. Detection of crystals in synovial fluids by light microscopy: sensitivity and reliability. Ann Rheum Dis 1989: 411: 737-42.

32. GROSS BD, W I L ~ RB, DICoSIMO CJ, WmLL~tS SV. Gout and pseudogout of the temporomandibular joint. Oral Surg 1987: 63: 551-4.

33. HALVERSON PB, McCARTY DJ. Patterns of radiographic abnormalities associated with basic calcium phosphate and cal- cium pyrophosphate dihydrate crystal de- position in the knee. Ann Rheum Dis 1986: 45: 603-5.

34. HALVERSON PB, McCARTY DJ. Identili- cation of hydroxyapatite crystals in syn- ovial fluid. Arthritis Rheum 1979: 22: 389-95.

35. HAMILTON E, PATTRICK M, HORNBY J, DERRICK G, DOrI~TY M. Synovial fluid calcium pyrophosphate dihydrate crystals and alizarin red positivity: analysis of 3000 samples. Br J Rheumatol 1990: 29: 101-4.

36. HUTTON CW, DOm~TY M, DmPPE PA. Acute pseudogout of the temporoman- dibular joint: a report of three cases and review of the literature. Br J Rheumatol 1987: 26: 51-2.

37. ISHIDA 17, D O W N HD, BULLOUGH PG. Tophaceous pseudogout (tumoral cal- cium pyrophosphate dihydrate crystal de- position disease). Human Pathol 1995: 26: 587-93.

38. KAMATANI Y, TAGAWA T, HIRANO Y, NO- MURA J, MURATA M. Destructive calcium pyrophosphate dihydrate temporo-man- dibular arthropathy (pseudogout). Int J Oral Maxillofac Surg 1987: 16: 749-52.

39. K L ~ HZ, EWaANK RL. Gout of the teraporomandibular joint. Report of three cases. Oral Surg 1969: 27: 281-2.

40. LAMBERT RGW, BECKER EJ, PIUTZKER KPH. Case report 597. Skeletal Radiol 1990: 19: 139-41.

41. LEO~GHAM J, REt3AN M, JONES A, DonrmTV M. Factors affecting radio- graphic progression of knee osteoar- thritis. Ann Rheum Dis 1995: 54: 53- 8.

42. MAGNO WB, LEE SH, SCnMmT J. Chon- drocaleinosis of the temporomandibular joint: an external ear canal pseudotumor. Oral Surg 1992: 73: 262-5.

43. McCAaTY D J, K o o P ~ WJ, eds. Ar- thritis and allied conditions. 12th ed. Philadelphia: Lea & Febiger, 1993: 1773- 872.

44. MoGI G, KUCA M, KAWAUCHI H. Chon- drocalcinosis of the temporomandibular joint. Calcium pyr0phosphate dihydrate deposition disease. Arch Otolaryngol Head Neck Surg 1987: 113: 1117-9.

45. MORADt-BmHENDI N, "ILmNER IG. Devel- opment of a new technique for the ex- traction of crystals from synovial fluids. J Mater SCI Mater Med 1995: 6: 51-5.

46. PATTRICK M, ~ T O r ~ E, WILSON R, AUSTIN S, DOI-IERTY M. Association of radiographic changes of osteoarthritis, symptoms and syaovial fluid particles in 300 knees. Ann Rheum Dis 1993: 52: 97- 103.

47. PiAcm, rrnm C, MARCI-IETn C, CALLEGARI A, et al. Endoarticular loose bodies and calcifications of the disk of the temporo- mandibuiar joint: morphological features and chemical composition. Scanning Microsc 1995: 9: 789--95.

48. PoPOV K, DnvnTROV D. A case of gout with manifestations in the temporoman- dibular joint. Stomatologiia-Soliia 1975: 57:119-24 [in Bulgarian].

49. PmTZKER KPH, PmLLIPS H, LUK SC, KOVEN IH, KISS A, HOUPT JB. Pseudotu- mor of temporomandibular joint: de- structive calcium pyrophosphate dihy- drate arthropathy. J Rheumatol 1976: 3: 70-81.

50. PYlqN BR, WEIh'BERG S, IRISH J. Calcium pyrophosphate dihydrate deposition dis- ease of the temporomandibular joint. A case report and review of the literature. Oral Surg 1995: 79: 278-84.

51. SAILER HE MAKEK MS. Arthropathie durch Pyrophosphat-Ablagerung als tu- morvort~usehende Erkrankung des Kie- fergelenks. Sehweiz Mschr Zahnmed 1986: 96: 594-603.

52. ScHtrs~r HR. Synodal inflam- mation, crystals, and osteoarthritis. J Rheumatol 1995:22 (Suppl 43): 101-3.

53. ScmnemCHER HR, REGINATO kJ. Atlas of synovial fluid analysis and crystal identi- fication. Philadelphia: Lea & Febiger, 1991.

54. SCULLY RE, MARK EJ, MCNEELY WE EBELING SO. Case reports of the Massa- chusetts General Hospital. Weekly

clinicopathological exercises. Case 29 1996. N Eng J Med 1996: 335: 876--81.

55. StmA Z, SzAno G. Gouty tophus of the temporomandibular joint. Fogorv Sz 1989: 82:33-6 [in Hungarian].

56. SWAN A, CHAPMAN B, HEAP P, SEWARD H, DmPPE P. Submicroscopic crystals in osteoarthritic synovial fluids. Ann Rheum Dis 1994: 53: 467-70.

57. DE Vos RAI, BRANTS J, KUSL~ GJ, BECKER AE. Calcium pyrophosphate di- hydrate arthropathy of the temporo- mandibular joint. Oral Surg 1981: 51: 497-502.

58. ZEMPLENYI J, CALCATERRA TC. Chondro- calcinosis of the temporomandibular joint. A parotid pseudotumor. Arch Oto- laryngol 1985: 111: 403-5.

T M J osteoarthritis and crystals

Address: Leonore C. Dijkgraaf, DDS, PhD The University of Texas Health Science

Center at San Antonio Department of Oral and

Maxillofacial Surgery 7703 Floyd Curl Drive San Antonio, TX 78284 USA

273