Osteoarthrosis of the temporomandibular joint following experimental disc perforation in Macaca fascicularis

J Oral Maxillofac

Surg

46:979-990.1988

Osteoarthrosis of the Temporomandibular Joint Following Experimental Disc Perforation in Macaca fascicularis EMAD HELMY, BDS, MS,* ROBERT BAYS, DDS,t AND MOHAMED SHARAWY, BDS, PtiD* The aim of this experiment was to study the sequela of experimental temporomandibular joint (TMJ) disc perforation. Each TMJ of four Macaca fascicularis adult monkeys was surgically exposed, and a 4- to 6-mm perforation at the posterolateral portion of the avascular disc was produced by electrosurgery. Four monkeys were used as controls. The animals were killed 11 weeks (two experimental and two controls) or 12 weeks (two experimental and two controls) after disc perforation. The perforations were increased in size in five joints, and healed in one joint. In addition, two joints of one animal showed complete loss of the disc, denudation of articular surfaces, and bone-to-bone contact. In contrast to control joints, the experimental joints exhibited the following changes histopathologically: thick, highly cellular and fibrillated fibrous coverings of articular surfaces (five joints); marked hyperplasia of synovial membrane; migration of synovial cells on the surfaces of the disc and margins of perforation; multiple adhesions of disc to articular surfaces; increase in cellularity and vascularity of discs; and chondrocytic clustering in temporal fibrous covering; and osteophytes of condylar and temporal components and focal or complete denudation of articular surfaces (2 joints). Most of these changes were consistent with the diagnosis of osteoarthritis. From this study, one can conclude that disc perforation can lead to osteoarthritis.

Although perforation of the disc in the clavicular and acromioclavicular joints is ered a common normal finding,’ perforation human temporomandibular joint (TMJ) disc sidered pathologic.2*3 Such perforations are

sternoconsidof the is conusually

found at the posterolateral portion of the avascular disc4 and are frequently seen as a component of osteoarthritis.5-8 This study was carried out in an attempt to answer the following questions: (1) Would osteoarthritic changes known to accompany disc perforation in humans occur as a sequela to an experimentally produced disc perforation in primates? (2) Would surgically perforated TMJ discs undergo spontaneous regeneration or repair? and (3) If healing occurred, what elements within the TMJ would contribute to the regeneration of the injured disc?

* PhD student, Department of Oral Biology/Anatomy, Medical College of Georgia, Augusta. t Professor and Chairman, Department of Oral and Maxillofacial Surgery, Emory University, Atlanta. $ Professor and Chairman, Department of Oral Biology/ Anatomy, Medical College of Georgia, Augusta. This study was performed while Dr Helmy was a Peace Fellow in the Department of Oral Biology/Anatomy, Medical College of Georgia. Supported in part by Amideast Company grant #IO-1202-1122-64. Address correspondence and reprint requests to Dr Sharawy: Department of Oral Biology/Anatomy, Medical College of Georgia, Augusta, GA 30912. 0 1988 American Association geons 0278-2391/88/4611-0008$3.00/0

of Oral and Maxillofacial

Materials and Methods Eight adult (9 to 14 years of age) Mucacafusciculuris monkeys were used in this study. Four males constituted the control group and three males and one female were designated as the experimental group. The animals were housed one per cage and fed Wayne Monkey Diet (Continental Grain Co.,

Sur-

979

OSTEOARTHRITIS FOLLOWING DISC PERFORATION

Chicago) supplemented with fresh fruits. Water was provided ad libitum. The four experimental animals were anesthetized with intramuscular injections of Ketamine Hcl (9 mg/kg body weight). The skin at the surgical field was shaved, scrubbed and the animal was drapped exposing the surgical field. A l/100,000 epinephrine solution was injected around the joint to obtain hemostasis. The joint capsule was exposed through a transverse incision of the skin along the zygomatic arch. A horizontal incision was made through the lateral capsule to gain access to the superior joint space. A TMJ spreader was used to push the condyle down in order to expose the superior aspect of the articular disc, A 4- to 6-mm perforation was then made at the posterolatera1 aspect of the disc through the total thickness of the avascular portion using an electrosurgery loop. Extra caution was used not to touch the condylar surface. Electrosurgery was used to avoid bleeding in the joint cavities, which would complicate the interpretation of the results. After the perforation was made, the superior joint space was closed by suturing together the margins of the incised capsule using a 5-O vicryl suture. The skin wound was sutured in layers with absorbable sutures. Pressure was applied to the wound area with ice bags for approximately one half hour following surgery. The same procedure was repeated in the opposite joint of the same animal and in the other three experimental animals. Postoperative analgesics and antibiotics were administered for ten days. Postoperatively, the animals were fed their regular diet. The animals were killed by perfusion of Karnovsky’s fixative 11 weeks (two experimental and two controls) or 12 weeks (two experimental and two controls) after disc perforation. Blocks of tissue containing the TMJs were decalcified in a 50/50 formic acid and sodium citrate solution. Decalcification was verified by the loss of bone radiopacity in radiographs of the blocks. The blocks were sectioned sagittahy into three or four 3- to 5-mm thick sections using sharp razor blades. Each section was examined and photographed using a Zeiss stereomicroscope. The TMJ sections were then dehydrated in ascending grades of ethanol, cleared in xylene, infiltrated, and embedded in paraffin, The paraffin blocks were sectioned at 7 p.rn thickness and stained with hematoxlyin-eosin or alcian blue at pH 2.5 (to demonstrate acidic proteoglycans). Results STEREOSCOPIC OBSERVATION OF CONTROL JOINTS The condylar articular surface of control joint appeared smooth, glistening, and even in thickness. The temporal articular components consisted of a

shallow glenoid fossa, a flat articular eminence with a long preglenoid articular plane, and a welldeveloped postglenoid spine (Fig 1). The articular surface of the temporal articular surface appeared smooth, glistening, and even in thickness (Fig 2). The articular disc, similar to a human TMJ disc, consisted of an anterior band, intermediate thin zone, posterior band, and bilaminar zone (Figs 1 and 2). The anterior band extended anteriorly and superiorly to insert on the ascending slope of the articular eminence and inferiorly to insert on the anterior inferior aspect of the condyle (Fig 1). The retrodiscal tissue of the bilaminar zone split into a superior stratum which inserted into the petrotympanic fissure and inferior stratum which inserted into the posterior inferior surface of the condyle. The discal attachments divided the joint space into anterior superior, anterior inferior, posterior superior and posterior inferior recesses. In a sagittal view, the articular capsule is more distinct posteriorly (Fig 2). The superior head of the lateral pterygoid ran parallel to the anterior extension of the anterior band of the disc and a few of its fibers (- 15%) inserted into the lower fibers of the anterior band. The rest of the fibers of the superior head and all the fibers of the inferior head inserted into the anteromedial aspect of the head of the condyle (Figs 1 and 3). The synovial membrane appeared as a glistening membrane with fingerlike projections that were more prominent at the posterior superior joint recess (Fig 2). HISTOLOGIC OBSERVATION OF CONTROL JOINTS The superior and inferior joint spaces were free of tissue

debris. The surfaces of the condyle, disc, artitular eminence, and glenoid fossa were smooth (Fig 3). Disc. The disc consisted of avascular, dense collagen bundles that ran parallel at the superior and inferior surfaces and were slightly interwoven in the middle portion of the intermediate zone. Interwoven or interlacing bundles were mostly found at the anterior and posterior bands (Fig 4). The discal cells were of two varieties, fibroblasts and chondrocytelike cells. The latter appeared rounded, contained round nuclei, and were surrounded with lacuna and alcian blue-positive territorial matrix (Fig 5). The retrodiscal tissue consisted of highly vascular loosely organized collagen bundles and a large number of elastic fibers that were more abundant at the superior stratum but were also found in the inferior stratum. The superior stratum inserted into the petrotympanic fissure and postglenoid tubercle while the inferior stratum inserted into the posterior inferior aspect of the condyle (Fig 3).

HELMY, BAYS, AND SHARAWY

FIGURES l-15. Figs l-5 and 7-11 are saggital sections of control joints; Fig 6 is a coronal section. All histologic sections were stained with hematoxylin and eosin unless otherwise specified. FIGURE 1. Intact disc, consisting of anterior band (l), intermediate thin zone (2), posterior band (3), and bilaminar zone (4). The superior head of lateral pterygoid (5) inserts into the anterior band of the disc and to the condyle along with the inferior head (6). FIGURE 2. Note the smoothness of the articular surfaces and the presence of synovial tissue (Sv) in the posterior superior recess. The posterior capsule (C) is seen. FIGURE 3. Histological section showing the anterior band (I), intermediate zone (2), posterior band (3), and bilaminar zone (4). The superior head of the lateral pterygoid (5) inserts mostly into the condyle (asterisk). (Original magnification x 18.) FIGURE 4. High magnification showing a portion of the posterior band of the disc. Note the interwoven collagen bundles and the presence of chondrocytelike cells (arrowheads). (Original magnification x 116.) FIGURE 5. Alcian blue-stained section of the disc showing alcian blue-positive territorial matrices around the chondrocytelike cells (arrowheads). (Original magnification x290.) FIGURE 6. Lateral superior recess showing synovial membrane (arrows) lining the articular capsule (C) and extending onto the surfaces of articular eminence (AE) and disc (D). (Original magnification x29.) FIGURE 7. Alcian blue stained-section of synovium, showing intimal synovial cells (arrows) and vascular subintimal connective tissue. (Original magnification x290.) FIGURE 8. Portion of the superior aspect of the condyle (shown sideways) showing the smooth surface of the avascular fibroelastic (F) covering with an underlying secondary cartilage (C). (Original magnification X116.) FIGURE 9. Alcian blue-stained

982

OSTEOARTHRITIS

FOLLOWING

DISC PERFORATION

section of condyle showing alcian blue-positive territorial matrices of the chondrocytes in the hypertrophied cell zone (arrows). (Original magnification x 181.) FIGURE 10. Portion of the articular eminence. Note the smooth surface and the compactness of the normal fibrous covering and the presence of chondroid bone (Ch) and absence of organized secondary cartilage. (Original magnification X73.) FIGURE 11. The cells found in the avascular fibrous covering of the articular eminence were chondrocytelike cells (arrowheads) and fibroblasts. Note the presence of chondroid bone (Ch). (Original magnification x 181.) Figs 12-15 are all stereoscopic photographs of experimental joints that were cut sag&ally into lateral, middle, and medial sections. FIGURE 12. Middle section showing enlarged discal perforation with irregular margins (arrows). Note the thickened condylar (C) fibrous covering located under the perforation (asterisk). FIGURE 13. Loss of the disc and denudation of both condylar (C) and temporal bone fibrous coverings were noted in two joints of one animal. FIGURE 14. Low magnification of the joint shown in Fig 13. The areas adjacent to denuded articular bones showed a “reparative” fibrous thickening (arrows). FIGURE 15. A joint section medial to disc perforation. Note that this portion of the disc (D) appeared normal except for marked thickening of the condylar and temporal fibrous coverings (arrows).

Synovial membrane. Synovial membranes lined the walls of all the joint recesses and extended for a short distance onto the temporal, condylar, and discal surfaces (Fig 6). The synovial membrane consisted of a cellular layer facing the joint space known as intima, and a subjacent vascular connective tissue or subintimal layer (Fig 7). The cells of the synovial intima appeared as either elongated, tibroblastlike or polygonal, macrophagelike. Gaps between the intimal ceils exposed the subintimal connective tissue to the joint space (Fig 7). The subintimal tissue of the synovial membrane in the anterior recesses was mostly fibroadipose in nature while that of the posterior recesses was mostly of the fibroareolar type. In coronal sections the synovial membrane of the medial superior recess was fibroadipose, while that of the medial inferior recess was fibroareolar in nature. The synovial membrane of the lateral superior recess was a dense fibrous type, while the lateral inferior recess was fibroareolar. In both the lateral and medial recesses the synovial membranes were seen to extend onto the disc for a short distance (Fig 6). The connective tissue of the synovial villi contained delicate elastic fibers in the resorcin fuchsin stained sections (not included). A small amount of alcian blue-positive material was also noted between the cells of the intima (Fig 7). Condyle. The condylar articular surface was covered with avascular fibroelastic connective tissue. The collagen bundles ran mostly in an anteroposterior direction. Few fibroblastlike cells, with occasional chondrocytelike cells, were seen within the fibrous covering. At the interface between the fibroelastic tissue and the underlying secondary cartilage a distinct two- to three-cell-thick layer was found. This layer corresponded to the prechondrocytic resting cell layer or undifferentiated mesenchymal cell layer (Fig 8). Below the reserve cell zone the cartilage could be indistinctly divided into a proliferative zone, a zone of the hypertrophied chondrocytes, and a zone of calcified cartilage (Figs 8 and 9). The latter contained either osteoid material or bone. Alcian blue stained the chondrocytic

territorial matrix mostly in the zone of hypertrophy and zone of calcified cartilage (Fig 9). Temporal articular surface. The fibrous covering of the glenoid fossa was thin, dense, and avascular and contained very few cells. The fibrous covering of the descending slope of the articular eminence was much thicker than that of the glenoid fossa but also consisted of dense fibrous tissue with few fibroblasts (Fig 10). Chondrocytelike cells also were occasionally seen (Fig II). Chondroid bone, characterized by the presence of chondrocytes surrounded with eosinophilic matrix, was consistently seen beneath the fibrous covering of the descending slope of the articular eminence (Figs 10 and 11). The territorial matrix of chondrocytelike cells stained positive with alcian blue stain (Fig 11). Under the chondroid bone, the temporal bone was of the mature lamellar type (Fig IO). STEREOSCOPICOBSERVATION OF EXPERIMENTAL JOINTS

Before the joints were sectioned into lateral, middle, and medial sections they were observed intact with the dissecting stereoscope. All eight experimental joints exhibited thickening of the articular capsule laterally and anteriorly. After the first lateral section was made, the joint could be observed under the stereoscope. In five of the eight discs that were perforated 11 or 12 weeks earlier, large perforations could be recognized (Fig 12). However, in both joints of one monkey the discs were almost completely degenerated on both lateral and medial sections, and bone-to-bone contact was observed (Figs 13 and 14). The disc perforation in one joint of another monkey that was killed 12 weeks postoperatively had apparently healed (Fig 41). In five joints the perforations increased in size both medially and anteroposteriorly. Adhesion of the perforated discs to the underlying condyle was seen in the lateral section of the five joints in which perforation could be identified (Fig 16). Complete adhesion of the anterior lip of the perforation to the condyle was seen in three perforated discs, while

HELMY, BAYS, AND SHARAWY

983

FIGURES 16-30. Figs 16 and 17 are stereoscopic photographs, of experimental joints that were cut sagittally into lateral, middle, and medial sections. FIGURE 16. Note the fusion of the anterior lip of perforated disc (arrow) to a steplike depression (asterisk) of the condyle (C). FIGURE 17. In this severely osteoarthtitic joint, the large glistening globular masses in the joint spaces (arrows) were found histologically to be hypertrophied synovium. Figs 18-30 are histologic sagittal sections of experimental joints, stained with hematoxylin and eosin unless otherwise specified. FIGURE 18. In contrast to control (Fig IO), note the increased thickness of the fibrous covering of the articular eminence. Also note the high cellularity and the presence of bone microcysts (arrows). Asterisk indicates the superior joint space. (Original magnification x46.) FIGURE 19. Fibrous covering of glenoid fossa in a midsection showing abnormally numerous fibroblasts and chondrocytelike cells. Note the surface next to the joint space (asterisk) is covered by two to three cell layers (arrows). (Original magnification x46.) FIGURE 20. The cells covering the surface of the Bbro cartilage (arrows) could be traced to synovial membrane (Sv). The latter also covered the perforated disc (D and arrows). Note the abnormal high vascuhuity of the fibrous tissue (arrowheads). (Original magnification X18.) FIGURE 21. Section of temporal fibrous covering medial to the perforation showing chondrocytic clusters (arrows), or “brood capsules”. (Original magnification X 181.) FIGURE 22. The territorial matrix of the chondrocytes were alcian blue-positive (arrowheads). Note the loss of metachromasia in areas that appeared as chondroid bone (Ch). (Original magnification x93.) FIGURE 23. Temporal articular surface close to the articular eminence showing doubling of the tide mark

984

OSTEOARTHRITIS

FOLLOWING

DISC PERFORATION

(arrows) next to cartilage (Cr). (Original magnification x 181.) FIGURE 24. In contrast to Fig 23, areas of the glenoid fossa showed osteoclastic bone resorption (arrows). Note the presence of chondroid bone (Ch). (Original magnification x73.) FIGURE 25. Areas of alcian blue-positive cartilage was sometimes seen interposed between temporal fibrous covering (F) and bone (B). The alcian bluepositive matrix of many cells was lost in focal areas (arrowheads). Note the presence of a bone microcyst with cartilagenous remnants within the cavity (Cv). (Original magnification x73.) FIGURE 26. Middle section of temporal surface opposite to the disc perforation. Note the marked fibrillation of the fibrous covering. Asterisk indicates joint space. (Original magnification x 116.) FIGURE 27. Middle section opposite disc perforation showing marked flaking (arrows) of the fibrous covering of articular eminence, characteristic of osteoarthrosis. (Original magnification x46.) FIGURE 28. Fibrous covering of articular eminence occasionally showed amianthoid or asbestoid degeneration. Asterisks indicate superior joint space. (Original magnification x464.) FIGURE 29. Fibroosseous osteophyte found on the posterior slope of the articular eminence (asterisk). (Original magnification x29.) FIGURE 30. An example of fibrous osteophyte found on the articular eminence. (Original magnification x 18.)

the other two discs showed adhesion of the posterior lip of the perforation to the condyle. Thickening and irregularities of the condylar artitular covering and the temporal articular surfaces were seen in all five joints with perforated discs (Figs 13, 14, 16, and 17). This was clearly seen in the lateral two thirds of each joint. In the two joints in which the discs were almost completely degenerated, the coverings of the articular surfaces were denuded (one joint) or much thinner than that in normal joints (one joint) (Figs 13 and 14). In many areas the articular covering of the condyle was thickened, particularly adjacent to the discal perforation, and seemed to abut into a reverse configuration on the opposing temporal surface (Figs 14 and 17). Compared with controls, the condyle was deformed in seven of the eight experimental joints (Figs 13-17). The temporal articular surfaces in the same seven joints were also altered in their normal morphology and became highly irregular, with multiple fibrous adhesions to the disc and condyle. In the one joint in which the disc was apparently healed, there was minimal morphological deformation in the shape of either the condyle or the temporal bone (Fig 41). Synovial proliferation was seen in the form of globular masses filling mostly the anterior and/or posterior recesses of five joints (Fig 17). The remaining three experimental joints also showed synovial proliferations, but to a lesser extent. The posterior superior attachment of the disc seemed to have a more anteriorly located insertion at the glenoid fossa than that of the control joints. These changes were seen exclusively in those cases that showed adhesion of the posterior lip of the perforation to the glenoid fossa. The posterior inferior recesses of five joints with perforated discs were more shallow than those of the control joints. The sections of TMJs medial to the perforations appeared to have near normal morphology except for roughness of the anterior superior surface of the disc in four of the seven joints (Fig 15). In addition, there was a slight fibrillation and thickening of the

condylar articular covering and a slight irregularity along the temporal articular surfaces (Fig 15). HISTOPATHOLOGICAL

OBSERVATIONS

The experimental joints were divided for description purposes into three groups: (1) joints with perforated discs (n = 5); (2) joints in which the discs were completely degenerated (n = 2); and (3) joints with healed disc perforation (n = 1). Group 1 Changes in the temporal surface. Three of the five joints showed marked thickening of the articular coverings (Figs 18-20). In contrast to control joints, there was also a marked increase in the number of cells, particularly along the descending slope of the articular eminence (Figs 18 and 19). In sagittal sections of the joints, lateral to the perforations, the thickened fibrous covering of the glenoid fossa was highly cellular and was covered by cells similar in morphology to intimal synovial cells (Fig 20). These cells could be traced to the synovial membrane lining the posterior superior recess. The numerous cells within the fibrous tissue appeared to be undifferentiated mesenchymal cells (Fig 19). They were either rounded or stellate in shape, with large basophilic nuclei. It was not uncommon to find areas of the thickened fibrous covering that were abnormally increased in vascularity (Fig 20). The vessels were mostly capillaries and venules and were either distributed throughout the fibrous tissue (Fig 20) or restricted to an area between the avascular fibrous covering and the underlying bone (Fig 24). The bone close to the vascular fibrous tissue often appeared irregular and broken in its continuity (Fig 24). In sections medial to the perforation, the fibrous covering of the articular eminence as far as the preglenoid plane appeared smooth and contained large areas of hyalinelike cartilage (Figs 21 and 22). In the alcian blue-stained sections the matrix was stained deep blue but there were some areas of reduced metachromasia that appeared to be

HELMY, BAYS, AND SHARAWY

chondroid bone (Figs 22 and 24). Areas of new bone formation were found next to the cartilage as indicated by duplication of the tide mark (the interface between the calcified and noncalcified cartilage) (Fig 23). In some areas of the temporal covering, chondrocytes were grouped into clusters, sometimes referred to in the literature as “brood capsules”’ (Fig 21). One chondrocytic cluster was surrounded by alcian blue ground substance and hyalinized collagen (Figs 21 and 22). However, the phenomenon of chondrocytic clustering also was observed in association with fibrillated fibrous covering. Fibrillation was more commonly observed in the temporal covering opposite the enlarged disc perforation (Figs 26 and 27). In the fibrillated areas, looser packing and occasional fragmentation of collagen bundles were seen in the superficial layers. Some of the flakes or strands of the articular surfaces were projecting from the surface into the joint space (Fig 27). Close observation of many of these projections revealed their nature as avascular, acellular, scarlike collagenous tissue, very similar to what was described in arthritic joints as asbestoid or amianthoid degeneration’ (Fig 28). It was remarkable that throughout these degenerative areas in the fibrous coverings of the temporal articular surfaces, little or no inflammatory cells were noted. Areas of bone resorption were frequently seen at the posterior slope of the articular eminence and at the depth of the glenoid fossa in the middle sections of the affected joints. These areas were characterized by the presence of osteoclasts in Howship’s lacunae, multiple bone reversal lines, and granulation tissues (Fig 24). In some sections, the granulation tissue communicated directly with the joint space. In places where cartilage was present close to areas of bone resorption, chondroclastic activity, manifested by focal dissolution of the alcian blue positive territorial matrix and its replacement with a loosely arrayed fibrous tissue, was frequently observed (Fig 25). A similar type of cartilage resorption has been described in the literature as Weichselbaum’s lacunae resorption.’ Bone microcysts were frequently observed. The cysts appeared empty, contained loose connective tissue, or contained remnants of cartilage (Fig 25). In two joints of two animals in this group, focal areas of bone were completely denuded of fibrous covering and the bone appeared sclerotic with empty osteocytic

985

lacunae (Fig 31). Fitting of the denuded bone corresponded with ruptured osteocytic lacunae that opened into the joint cavity. The areas of denuded bone were seen mostly at the posterior slope of the articular eminence and in proximity to areas of bone which were covered by a thickened highly cellular and highly vascular fibrous covering as previously described (Fig 31).Areas of osteophytes were observed in the temporal articular surfaces of two joints. The osteophytes were either flbroosseous (Fig 29) or fibrous in nature (Fig 30). In all five experimental joints of this group, the superior stratum of the retrodiscal tissue appeared to have more anteriorly placed bony attachments than the control joints (Fig 33). The underlying bone in the area of attachment frequently showed lines indicative of extensive remodelling. Changes in the condyle. The histopathologic study confirmed the stereoscopic observation of a marked deformation of the condyle of the affected joints (Fig 33). Fibrous adhesions of the anterior lip of the perforated disc to the condyle were seen clearly in three joints, while the posterior lip of the perforation was adherent to the condyle in only two joints. Significant condylar bone resorption was seen beneath these adhesions, resulting in a marked craterlike depression under the perforation (Fig 33). In the controls, the posterior inferior slope of the condyle lacked secondary cartilage. In contrast, cartilage was consistently present at this site in the experimental condyles (Fig 34). Along the condyles there was also a marked thickening of the secondary cartilage (Figs 34 and 35). Unlike the temporal articular surfaces, the condyles of the five joints in group 1 showed little or no flaking or fibrillation. At the anterior slopes of the condyles a thickened, highly cellular fibrous covering was noted (Fig 35). The cells appeared undifferentiated and similar in morphology to the reserve cell zone of the condylar secondary cartilage. Chondrocytic clustering, frequently observed in the degenerating temporal tibrous covering, was only occasionally noted in the condylar covering. The area of the condyle under the perforation frequently showed vascular granulation tissue adjacent to areas of bone resorption. In several instances, focal areas of denuded bone could be seen. Condylar osteophytes were seen in the five joints. Some of these osteophytes took the shape of a raised, pedunculated projection consist-

FIGURES 31-44. Histologic sections of experimental joints, stained with hematoxylin and eosin unless otherwise specified. FIGURE 3 1. Focal area of denudation of bone of the articular eminence into the joint superior space (asterisk). Note the areas of bone resorption under thickened highly vascular fibrous tissue (arrows). (Original magnification x29.) FIGURE 32. Severe osteoarthritis following disc perforation. Note the loss of the disc, complete denudation of the bone of both condyle (C) and articular eminence (AE). Note the presence of ebumation (opening of osteocytic lacunae to joint space; arrows). (Original magnification x 18.) FIGURE 33. Low magnification showing deformed condyle and adhesion of fibrillated disc (D) to condyle (C) (asterisk) and to temporal fibrous covering (two asterisks). (Original magnification x 15.) FIGURE 34. Some areas of the deformed condyles showed thickenning of secondary cartilage

986

OSTEOARTHRITIS

FOLLOWING

DISC PERFORATION

(Cr) and the appearance of cartilage abnormally in the posterior inferior aspect of the condyle (arrow). Note the hypertrophied synovial villi (Sv). (Original magnification x 18.) FIGURE 35. Fibrillated fibrous covering of the condyle (F) overlying cartilage (Cr). (Original magnification x23.) FIGURE 36. An osteophyte (0) of the posterior aspect of the condyle causing indentation of the overlying disc. (Original magnification x37.) FIGURE 37. Perforated disc became highly cellular and vascular. Note the cells of the surface of the disc and the free margin of the perforation. (Original magnification x29.) FIGURE 38. High magnification showing the surface cells and underlying highly cellular disc. (Original magnification x290.) FIGURE 39. Portion of the perforated disc stained with methyi pyronine which stained the surface cells red (arrows). These cells could be traced back to synovium. (Original magnification x290.) FIGURE 40. Anterior margin of perforated disc. Note the high cellularity of discal tissue and the presence of dystrophic calcification (asterisks). (Original magnification x73.) FIGURE 41. Low magnification showing a healed discal perforation (arrows). (Original magnification x 15.) FIGURE 42. High magnification showing synovial cells covering the disc (arrows) and regenerating fibrous tissue attempting to fill the gap. (Original magnification x29.) FIGURE 43. Higher magnification showing highly cellular regenerating fibrous tissue shown in Fig 42. (Original magnification x 116.) FIGURE 44. Low magnification showing the marked hypertrophy of synovial membrane (Sv) of the fibroadipose type following discal perforation. (Original magnification ~46.)

HELMY,

BAYS,

AND SHARAWY

987

ing of a core of bone covered with vascularized Iibrous tissue (Fig 36). The osteophytes appeared to have compressed the adjacent areas of the injured disc and created a depression (Fig 36). Changes membrane.

in the perforated

discs

and synovial

In contrast to the avascular control discs, the perforated discs were highly cellular and highly vascular, particularly close to the free margins of the perforation (Fig 37). The superior and inferior surfaces of the perforated discs were consistently covered with two to three layers of cells, which continued and ensheathed the margins of the perforations (Figs 37-39). These cells could be traced back to the synovial membrane of the joint recesses. Methyl pyronine (stains cells rich in rough endoplasmic reticulum [RER] red-stained sections) showed that the majority of cells were rich in RER and therefore possibly belonged to the B-type cells of the synovial membrane (Fig 39). In four joints, metaplastic bonelike tissue was seen in the margins of the perforation (Fig 40). The disc medial to the perforation was, in contrast to control discs, highly cellular and highly vascular. The cells within the perforated discs were mainly tibroblasts and undifferential mesenchymal cells (Figs 38 and 40) surrounded by thin, young collagen fibers (Fig 38). Group 2

In the two joints (one animal) in which stereoscopic examination revealed absence of disks, the disks were degenerated in lateral, middle, and medial sagittal sections. The temporal articular surface was completely denuded along the anterior posterior length of the joint except at the most posterior aspect of the articular fossa (nonarticular region), where thickened fibrous tissue was observed. The bone appeared sclerosed, with many empty lacunae. Minute cracks, cysts, and pitting of the surface were seen along the denuded bone (Fig 32). The condylar articular surface was also denuded of its fibrous covering (Fig 32). Areas of bone resorption and granulation tissue rich in osteoclasts were observed in both condyles (Fig 32). Group 3

Histologic examination of lateral, middle, and medial sections of the one joint in which the disc perforation healed revealed that the perforation was accidentally made in the retrodiscal tissue, which was further posterior than the perforations made in discs of the joints in groups 1 and 2. The degenerative changes observed in groups 1 and 2 were absent, and the articular surfaces appeared almost normal (Fig 41). Thickening of the fibrous covering and the appearance of cartilage at the posterior in-

ferior aspect of the condyle was observed under the healed disc (Fig 42). Although the margins of the perforation contained acellular, hyalinized, necrotic, and calcified collagen, the continuity of the disc was restored by a highly cellular fibrous tissue (Figs 41-43). At the site of healing, intense synovial membrane hyperplasia was also observed (Fig 44). The synovial villi appeared highly vascular (Fig 44). Synovial cells also were observed on the disc surface and lining the margins of the old perforation (Fig 42). Discussion In this study, bilateral discal perforation in monkeys produced changes consistent with osteoarthritis in seven of eight joints studied. The changes we observed were similar to the osteoarthritic changes described by other investigators, both in experimental models’0-13 and in human studies.‘*2,7,14 Two joints of one animal showed what can be classified as severe degenerative joint disease 12 weeks following discal perforations. These changes were characterized by almost complete loss of the disc, denudation of articular surface, and eburnation of the exposed bone. Sprinz12 also reported complete disc degeneration in a rabbit in response to excision of half of the disc. Dubecq” stated that in one of his animals the trauma to the disc led to its complete desintegration. We produced a 4- to 6-mm perforation in eight joints; in five joints the perforations were greatly enlarged 11 to 12 weeks later. Such degeneration of discal tissues undoubtedly contributed to the reported osteoarthritic changes. Yaillen et all6 observed degenerative joint changes following unilateral meniscectomy in Macaca fascicularis. Stevenson et ali7 and Silbermann and Livine’* observed that severe degenerative changes occur in animal TMJs in the absence of the disc. They also reported the occurrence of fibrous ankylosis as a sequela to disc disintegration. Moffet et all9 mentioned that after thinning of the disc and perforation in the human TMJ , degenerative changes appeared. Y aillen et ali6 observed a similar sequence of events in three of their experimental monkeys. The protective nature of the disc appear to be true for other stress-bearing joints. Lufti” observed that knee artitular cartilages in monkeys were more liable to undergo arthritic changes when they were deprived of meniscal protection. Appel14 reported that degenerative arthritis was more common in persons with meniscectomized knees. In the one joint in which the perforation was made close to the bilaminar zone and healed, no degenerative changes were observed. Also, when

988

bilateral discal perforations in Macaca fascicularis monkeys were repaired unilaterally using a synovial membrane flap procedure, the degenerative changes noted in the contralateral unrepaired perforation were not found in the repaired joints.*’ These results point out to the importance of an intact disc for the maintenance of the health of the articulating surfaces. Few or no inflammatory changes were observed 11 or 12 weeks after discal perforation. It is possible that inflammation occurred following perforation and contributed to the degenerative changes, and then subsided in the absence of infection. Other contributing factors may also have led to the observed osteoarthritis. Bruce and Walmsley** suggested that the primary function of the joint meniscus is to distribute synovial fluid over the articular surfaces. Goose and Appleton23 also proposed that the joint disc reduces the space between the articular surfaces so that a fluid film lubrication can take place. The presence of disc perforation may have interfered with the even distribution of synovial fluid to the articulating surfaces, which may have compromised the lubrication and nourishment role of the synovial fluid and increased the biomechanical friction between the bony components of the joints. In as much as we did not immobilize the joints following surgery, their mobility may have accelerated and exaggerated the degenerative response. The temporal articular surfaces were more affected by the osteoarthritic changes, particularly the posterior slope of the articular eminence, which could be because of excessive loading and pressure at this site following the perforation. Similar results were reported by Lambert et al in human arthritic TMJs.* Some of the changes noted after discal perforation can easily be classified as regenerative in nature (eg, thickening of fibrous coverings, high cellularity and vascular&y of the discs, appearance of chondrocytic clustering, and synovial hyperplasia). Changes such as fibrillation and flaking of fibrous coverings, denudation of bone, eburnation, resorption of bone, and appearance of condylar osteophytes are considered degenerative in nature. Both reparative and degenerative changes are known to occur simultaneously in osteoarthritis of other joints.8’24 Moffet et all9 and Carlsson and Oberg25 stated that if demands placed on the TMJ tissues exceed the remodelling capacity of the joint to respond to stimuli, degenerative changes may develop. Our experiment proved that discal perforation is a strong stimulus to both remodelling and degenerative responses of the monkey joint. Thickening of the fibrous coverings of the condyle and most of the temporal articular surface (particularly the posterior slope of the articular em-

OSTEOARTHRITIS

FOLLOWING

DISC PERFORATION

inence) was a consistent feature in five of the eight experimental joints. The mechanism by which these changes occurred cannot be determined from our study. However, one may speculate that in the case of the condyle, the undifferentiated mesenchymal cell layer consistently seen in the control specimens may have undergone proliferation under the injurious stress brought about by discal perforation. These cells could differentiate either into tibroblasts, known to make collagen and ground substance, or cartilage cells that would eventually lead to increase in the thickness of the secondary cartilage and bone resulting in deviation in form of the condyles. Hansson and Nordstrom26 reported a negative correlation between the thickness of the undifferentiated mesenchymal cells in human condyles and the thickness of condylar cartilage. They suggested that the undifferentiated mesenchyma1 cells contribute to the remodelling of the tissue, resulting in deviations in form as an early stage of osteoarthritis. Lubsen et al,27 in their study of human mandibular condyles, have supported the concepts raised by Hansson and Nordstrum.26 However, the same hypothesis could not be applied to the increased fibrous covering and the area of new cartilage observed at the temporal surfaces because an undifferentiated mesenchymal cell layer was not found in these areas. Tolle?* proposed that new fibrous surfaces can grow in from the edges of the synovial membrane to cover the denuded articular bone. We observed synovial intimal cells extending over the temporal articular surfaces in all of the experimental joints, but whether these cells contributed to the thickening of the fibrous covering cannot be determined from our study. The bone marrow was proposed as an alternate source for the fibrous tissue by Feroze.24 This hypothesis could not be confirmed or denied by our morphologic observations. Regardless of the source of these cells that abnormally populated the fibrous coverings, they undoubtedly had the ability to differentiate into cartilage and chondroid bone forming cells. We also observed the phenomenon of chondrocytic clustering, which is indicative of the proliferative ability of these chondrocytelike cells, as was commonly found in joints affected with osteoarthritis. ~4 The clustering was found more in the stress bearing areas than in the non-stress-bearing areas. 9 Anders et ali3 found tritiated thymidinelabelled cluster cells in experimentally produced osteoarthritis in the knee joints of rabbits. In our study the territorial matrix of these clusters of cells was alcian blue-positive, similar to cartilage-ground substance. Changes in the magnitude of physical forces and their directions (eg, from tensile to compressive) have been shown to cause a change of fibrous tissue to fibrocartilage.29 One may speculate

HELMY, BAYS, AND SHARAWY

that the presence of disc perforation created abnormal functional forces that caused certain areas of the temporal fibrous covering to change to fibrous cartilage. If these forces exceeded the physiological tolerance of the tissue, they would cause fibrillation or flaking of the fibrous tissue or complete denudation in extreme cases. All of these conditions were observed in the experimental joints. In addition to excessive mechanical forces, the change in the nature or amount of proteoglycans may be a contributing factor. The major sequela of proteoglycan destruction is decreased resistance to compressive loading generated during joint function.30 Also, depletion of glycosaminoglycans causes the tissue to be deformed and subjects the collagen mesh to excessive amounts of strain.3’ This can progress to complete denudation of the articular soft tissue. Following denudation, the bone surface becomes sclerosed, osteocytes die, and the lacunae open to the bone surface. This phenomenon is known as eburnation. Eburnation was noted in our study and has also been noted in other studies of osteoarthritis.24 Although we used alcian blue stain at pH 2.5, which stains acidic proteoglycans and shows loss of metachromasia, the qualitative nature of our study did not lead to the conclusion that proteoglycans were reduced in the experimental tissues. Further studies using quantitative and more precise techniques (such as lectin binding, monoclonal antibody binding, and biochemical determination) are needed. We have also observed vascularization of the otherwise avascular disc and fibrous coverings. The vascularity may have increased the water content of the fibrocartilage. Excessive percolation of fluid may enhance the deficiency of proteoglycans, which could lead to the weakening of collagen and its physical stripping by mechanical forces, and henceforth to the appearance of flaking and fibrillation of the articular fibrous coverings of the experimental joints. Histologic evidence of bone destruction was observed in some areas of the condyles and the temporal surfaces. Both osteoclastic bone resorption and chondrocytic lacunar resorption were observed. The stimulus for such resorption cannot be deduced with certainty, but could be attributed to excessive biomechanical forces caused by bone-to-bone contact following disc perforation. In most experimental joints the perforations were increased in diameter, and in some cases the discs were partially or completely degenerated. The tissue damage caused by overloading may lead to a release of lysosomal hydrolytic enzymes which can cause further degeneration.32‘35 Bone microcysts were a prominent feature of the experimental joints. Similar changes were reported

989 in osteoarthritis of other joints and during bone remodelling.36-40 The mechanism of formation of these cysts is not known. Numerous rounded cells were found on the surface of the perforated discs and ensheathing the free margins of the perforations. These cells could be traced back to the synovial intimal cells. Small blood vessels characteristic of subintimal connective tissue also were frequently seen. It is highly likely that the subintimal vessels were the source of the newly developed vessels in the perforated disc. Carlsson et a14’ reported similar findings and regarded the vascularization of injured discs as a regenerative response. Synovial hypertrophy was noted in all recesses of the experimental joints. We regard the synovial response as part of the osteoarthritic changes that occur following disc perforation. Other factors that traumatize the TMJ components, such as extraction of posterior teeth, were found to cause osteoarthritic changes, including synovial hypertrophy.42 Ramfjord and Enlow also reported synovial hypertrophy in the anterior inferior recesses of monkey TMJs after chronic anterior displacement of their mandibles. The hypertrophy of synovial villi, the migration of the intimal cells, and the vascularization of injured discs could all be regarded as regenerative responses in an attempt to repair the perforated discs. The cellularity of the perforated discs, particularly at the margins of the perforation, was markedly increased. The cells appeared as typical tibroblasts. A similar response was reported to occur in the injured knee menisci.4*45 The origin of these cells could be the proliferating synovial cells. Veth and DenHeeten46 observed that invasion of synovial cells was important in repair of injured menisci; repair occurred only in wedge-shaped meniscal lesions with a base adjacent to the synovial membrane. It is interesting to note that in the one healed perforation in our study, we unintentionally created the perforation close to the bilaminar zone in which vascularized synovial membrane is normally found. Conclusions From the results of our study, the following conclusions were reached: 1. Macroscopic and microscopic study of the Macaca fascicularis monkey TMJ revealed a joint that is markedly similar to the human TMJ. 2. Surgical creation of bilateral disc perforation in Macaca fascicularis monkeys resulted in pathologic alterations in the majority of the experimental joints (87.5%), which was consistent with a diagnosis of secondary osteoar-

990

thritis or degenerative joint disease. Therefore, we propose this procedure as a nonhuman primate experimental model for secondary osteoarthrites of the TMJ. 3. Perforated discs may have the capacity for repair when the perforation is close to the retrodiscal tissue or synovial membrane. 4. The synovial membrane appeared to be the tissue that showed most of the regenerative response to injury. 5. After discal injury the synovial cells behave like undifferentiaed mesenchymal cells, becoming highly proliferative and attempting to repair the disc perforation. References 1. Silberberg M, Frank EL, Jarrett S, et al: Aging and osteoarthritis of the human stemoclavicular joint. Am J Path01 35:85-l, 1959 2. Westesson PL, Rohlin M: Internal derangement related to osteoarthritis in temporomandibular joint autopsy specimens. Oral Surg 57:17, 1984 3. Westesson PL, Bronstein SL, Leidberg J: Internal derangement of the TMJ: Morphologic description with correlation to joint function. Oral Surg 59:323, 1985 4. Summa R: The importance of the interarticular fibrocartilage of the temporomandibular articulation. Dent Cosmos 70: 512, 1918 5. Cary B, Simonds CB, Morgan DH: Treatment of fracture and ankylosis, in, Morgan DH, House LR, Hall WP, et al (eds): Diseases of the Temporomandibular Apparatus, a Multidisciplinary Approach (ed 2). St. Louis, Mosby, 1982 6. Bean WR, Omrell KA, Oberg T: Comparisons between radiographic observations and macroscopic tissue changes in TMJs. Dentomaxillop Radio1 690, 1977 7. Carlsson GE, Signard K, Oberg T: Arthritis and allied diseases of the TMJ, in, Zarb GA, Carlsson GE: TMJ Function and Dysfunction. St. Louis, Mosby, 1979, p 284 8. DeBont LGM, Boering G, Liem RSB: Osteoarthritis and internal derangement of the TMJ: A light microscopic study. J Oral Maxillofac Surg 44:634, 1986 9. Sokoioff L: Osteoarthritis, in Ackerman LV, Spjut HJ, Abel1 MR (eds): Bones and Joints. Baltimore, Williams & Wilkins, 1976, p 110 10. Nuller W: Experimentalle untersuchungen uber druckusuren am Gelenkknorpel and ihre Beziehingen zur Arthritis deformans. Bnms Bietre Klur Chir 131642, 1924 11. Sprinz R: T.M.J. menisectomy in rabbit. J Anat 88514, 1954 12. Sprinz R: Further observation on the effect of surgery on the meniscus of the mandibular joints in rabbits. Arch Oral Biol 5:195, 1961 13. Anders H, Lindberg S, Telhaf H: Experimental osteoarthritis in rabbit. Acta Orthop Stand 41:522, 1970 14. Appel H: Late results after menisectomy in the knee joint. Acta Orthoo Stand lSu~ol1 133:6. 1970 15. Dobecq XJ: Recherchks morphologiques, physiologues et cliniques sur le menisque mandibulaire: Fuxation habituelle et craquements temporomaxillaires. J Med Bordeaux 114:125, 1937 16. Yaillen DM, Shapiro PA, Luschei ES: Temporomandibular joint menisectomy effects on joint structure and ma;+icatory function in Macaca fascicularis. J Maxillofac Surg 7:264, 1979 17. Stevenson TR, Evaskus DS, Laskin DM: Role of the meniscus in the TMJ ankylosis. A histological study. J Dent Res 58:269, 1979 (special issue A)

OSTEOARTHRITIS

FOLLOWING

DISC PERFORATION

18. Silbermann M, Livine E: Age related degenerative changes in the mouse mandible. J Anat 129(3):507, 1979 19. Moffet BC, Johnson LC, McCabe IB, et al: Articular remodelling in adult human T.M.J. Am J Anat 115:119, 1964 20. Lufti AM: Morphologic changes in the articular cartilage after menisectomy: An experimental study in the monkey. J Bone Joint Surg 57B:525. 1975 21. Sharawy M, Bays R, Helmy E: Repair of the TMJ disc perforation using synovial membrane flap. A histopathological and ultrastructural study. J Dent Res 65:851, 1986 22. Bruce J, Walmsley R: Replacement of semilunar cartilage of the knee after operating excision. Br J Surg 25: 17, 1937 23. Goose DH. Annleton J: Human Dentofacial Growth. Elmsford, NY, Pergamon, 1982, p 86 24. Feroze NG: Fine structure of synovial joints. A text and atlas of the ultrastructure of normal and pathological artitular tissues. Boston, Butterworths, 1983 25. Carlsson GE, Oberg T: Remodelling of the TMJ. Oral Sci Rev 6:53, 1974 26. Hansson T, Nordstrom B: Thickness of the soft tissue layers and the articular disc in the TMJ with deviation in form. Acta Odontol Stand 35:281, 1977 27. Lubsen CC, Hansson TL. Nordstrom BB, et al: Histomorphometric analysis of cartilage and subchondral bone in mandibular condyles of young human adults at autopsy. Arch Oral Biol 30(2):129, 1985 28. Toller PA: The synovial apparatus and TMJ function. Br Dent J 111:355, 1961 29. Scapinelli R, Little K: Observation on the mechanically induced differentiation of cartilage from fibrous connective tissue. J Path01 101:85, 1970 _ 30. Kemnson GE. Miur H. Swanson SAV. et al: Correlations between stiffness and chemical constituents of cartilage on the femoral head. Biochem Biophys Acta 215:70, 1970 31. Freeman MAR. Kemnson GE: Load carriage. in. Adult Articular Cartilage. Oxford, Alden Press, 1573. p 1973 32. Ali SY, Evans I: Enzymatic degradation of cartilage in osteoarthritis. Fed Proc 32: 1491, 1973 33. Pugh JW, Radin EL, Rose RM: Qualitative studies of human subchondral cancellous bone. Its relationship to the state of its overlying cartilage. J Bone Joint Surg 56:313, 1974 34. Radin EL, Paul IW, Rose RM: Role of mechanical factors in pathogenesis of primary osteoarthritic. Lancet 1:519,1972 35. Radin EL, Parker GH, Pugh JW, et al: Response ofjoints to impact loading. III. Relationship between trabecular microfractures and cartilage degeneration. J Biochem 6:51, 1973 36. Crane AR, Scamo JJ: Synovial cysts (ganglion) of bone. J Bone Joint Surg 49A:355, 1967 37. Gorlin RJ, Goldman HJ: Thoma’s Oral Pathology, vol 2 (ed 6). St Louis, Mosby. 1970, p 590 38. Harrison MHM, Schaiowicz F, Trueta J: Osteoarthritis of the hip. A study of the nature and evolution of the disease. J Bone Joint Sum 35B:598. 1953 39. Resenick D, Nlwafama G, Coutts RD: Subchondral cysts (geodes) in arthritis disorders. Pathologic and radiographic appearance of the hip joint. Am J Roentgen01 128~799, 1977 40. Johnson LC: Kinetics of osteoarthritis. Lab Invest 8:1223, 1959 41. Carlsson GE. Oberg T, Bergman F, et al: Morphological changes in the mandibular ioint disc in TMJ nain dvsfunction syndrome. Acta Odontol Stand 25: 163; 1967 _ 42. Cinasoui G: Histopathology of the TMJ following bilateral extraction of molars in rats. Oral Surg 16:613, 1963 43. Ramfjord SP, Enlow RD: Anterior displacement of the mandible in adult Rhesus monkeys: Long term observation. J Prosthet Dent 26(5):517, 1971 44. Bruce J, Walmsley R: Replacement of semilunar cartilage of the knee after operating excisions. Br J Surg 25:17, 1937 45. Cox JS, Corden LD: The degenerative effects of medial meniscus tears in dogs’ knees. Clin Orthop 125:236, 1977 46. Veth RPH, Den Heeten GJ: Repair of the meniscus. Clin Orthop Rel Res 175:258, 1983