- Email: [email protected]
Effect of ibuprofen on the healing and remodeling of bone and articular cartilage in the rabbit temporomandibular joint
J Oral Maxillofac 50:843-849.
Surg
1992
Effect of Ibuprofen on the Healing and Remodeling of Bone and Articular Cartilage in the Rabbit Temporomandibular Joint GEORGE
OBEID,
DDS,*
XIEN
ZHANG,
DDS,t
AND
XING
WANG,
MD,
MSD,
PHD$
The purpose of this study was to determine the effect of ibuprofen on the healing and remodeling of bone and cartilage in the temporomandibular joint of the rabbit. Forty-two rabbits were operated on to create a groove and a hole in the articular surface of both the right and left mandibular condyles. Following surgery, the animals were divided into three groups. Group A (12 rabbits) was used as a control and the animals did not receive any medication. Group B (15 rabbits) was given a daily dose of 17 mg/kg of ibuprofen. Group C (15 rabbits) was given a daily dose of 34 mg/kg of ibuprofen. All animals were killed after 4 weeks. The 84 condyles were examined clinically and histologically. Statistical analysis showed a highly significant difference in the healing of bone and cartilage between groups A and C (P < .Ol) and a significant difference between groups A and B (P < .05). The results of this study indicate that ibuprofen has an adverse effect on the healing of bone and cartilage in the temporomandibular joint of the rabbit.
the inhibition of prostaglandins, ibuprofen has an inhibitory effect on bone healing, remodeling, resorption, and metabolism.2’-27 Our hypothesis is that the capacity of the TMJ to remodel is hampered and delayed by the use of ibuprofen as an analgesic. This study was designed to investigate the effect of ibuprofen on healing and remodeling of bone and articular cartilage of the TMJ in rabbits.
Temporomandibular joint (TMJ) dysfunction can be caused by an abnormal relationship between the disc and the condyle. The resolution of this condition is often reached through remodeling and adaptation rather than actual correction of the anomaly.‘” Pain, the most troublesome symptom of TMJ dysfunction, tends to decrease in intensity with advancing patient age. This indicates that the condition is self-limiting, with a generally favorable outcome. However, it is estimated that without treatment, the pain can last as long as 7 years before decreasing.4-8 Ibuprofen, a nonsteroidal anti-inflammatory drug (NSAID), is widely used for the treatment of joint pain and inflammation.9-20 Studies have shown that, through
Materials and Methods Forty-two adult female New Zealand white rabbits (Blue and Gray Rabbitry, Aylett, VA) were used. After general anesthesia was administered, the skin over the TMJ region was shaved and prepared with Betadine (Purdue Frederick, Norwalk, CT) solution. A 2-cm preauricular incision was made to expose the condylar process. The zygomatic arch that normally covers a large part of the condylar process in the rabbit was exposed and partially resected to facilitate the surgical access. The inferior joint compartment was opened from the lateral aspect after incising the capsule and reflecting the disc. A 1S-mm-wide and 2-mm-deep mediolateral groove was made across the top of the condyle at the junction between the anterior and middle third of the articular surface (Fig 1). Then a 1.5-mm-diameter hole was drilled from the lateral to the medial side just beneath the articular cartilage and 5 mm behind the
Received from the Department of Oral and Maxillofacial Surgery, Washington Hospital Center, Washington, DC. * Assistant Director, Residency Training Program. t Visiting Research Fellow; Professor, Beijing Medical University, School of Stomatology, Department of Oral and Maxillofacial Surgery, Beijing, China. $ Visiting Research Fellow; Associate Professor, Beijing Medical Universitv. School of Stomatology, Department of Oral and Maxillofacial Surgery, Beijing, China-. Sunnorted bv NIH Grant DE856/ 1990. .. Address correspondence and reprint requests to Dr Obeid: Department of Oral and Maxillofacial Surgery, Washington Hospital Center, 110 Irving St, NW, Washington, DC 20010. 0 1992 American Association of Oral and Maxillofacial Surgeons 0278-2391/92/5008-0010$3.00/O
843
044
FIGURE 1. Diagramshowingthe locationand sizeof the hole and groove.
groove. A IS-mm-diameter fissure bur was used for the drilling, with copious amounts of saline for cooling. The disc was replaced in its normal position and the incision was closed in layers. Identical procedures were done bilaterally in all animals. Postoperatively, 1 mL of intramuscular Combiotics (penicillin-streptomycin; Fermenta Animal Health Company, Kansas City, MO) was given daily for 7 days. None of the wounds became infected and all rabbits did well following surgery. After surgery, the 42 rabbits were divided into three groups (A, B, and C). Group A (control) consisted of 12 rabbits that did not receive any experimental medication after surgery. Group B consisted of 15 rabbits that were given oral ibuprofen (8.5 mg/kg) twice daily (Whitehall Laboratories Inc, New York, NY). Group C also consisted of 15 rabbits that were given oral ibuprofen (17 mg/kg) twice daily. The dosages of 17 mg/ kg/d and 34 mg/kg/d are equivalent to giving a 70-kg man 1,200 mg and 2,400 mg ibuprofen per day, respectively. The ibuprofen was started on the day of surgery and was continued for 28 days. All rabbits were weighed weekly and the dose was adjusted accordingly. After 4 weeks, the animals were killed by injecting a Nembutal sodium solution (Abbott Laboratories, North Chicago, IL) via an ear vein and the 84 condylar processes were resected from the mandibles for gross evaluation and histologic examination. The healing of the grooves and holes was examined from all sides and all specimens were photographed (Fig 2). Surface irregularities of grooves and lack of closure of holes were documented. The relationship between the groove and hole in each specimen was recorded to help locate the hole under the microscope. The clinical and histologic examinations were made by the three investigators. There was no attempt to blind them as to the source of the specimens. After completion of the gross observations, the specimens were put in 10% formalin solution for 72 hours
EFFECT OF 1BUPROFEN ON RABBIT TMJ
and then decalcified. All specimens were embedded in paraffin and cut in a way that would ensure the presence of the hole and groove in each histologic slide. Serial sections were cut at 10 pm and stained with hematoxylin and eosin. Bone healing was classified as excellent, satisfactory, poor, and no healing (Fig 3) according to the following criteria. Excellent healing. Clinical observation showed complete healing of the hole. Histologically, the hole was full of bone that was indistinguishable from the surrounding bone (Fig 3A). Satisfactory healing. Clinical observation demonstrated that the hole had healed. The histologic slide showed that most of the hole was filled with new bone, but there was still clear demarcation between the hole and surrounding bone (Fig 3B). Poor healing. A small hole or evidence of a hole could be found by gross observation. The histologic examination showed that the hole was less than half full of new bone; the remaining part was filled with connective tissue (Fig 3C). No healing. Clinical observation showed an obvious hole. Histologically there was an empty hole or a hole that was filled with connective tissue. Sometimes a limited amount of new bone could be found at the margin of the hole (Fig 3D). The healing of the grooves was classified as good, moderate, or poor (Fig 4) according to the following criteria. Good healing. The groove was tilled with new bone and cartilage. The articular cartilage was well differentiated, with a smooth and intact surface (Fig 4A). Moderate healing. The groove contained abundant, highly cellular connective tissue. A cartilage cell layer was found (Fig 4B). Poor healing. The groove was observed clinically. The histologic slide showed limited amounts of new bone and cartilage in the groove. Sometimes poorly differentiated connective tissue filled the groove (Fig 4C). x2 Analysis was used for determining the statistical significance of differences between the groups. Results HEALING OF THE HOLE
Gross observation indicated that all the holes in group A had healed. Histologic examination revealed that 62.5% (15 of 24 specimens) had excellent healing. Five of 24 (20.8%) were considered satisfactory, and 16.7% (4 of 24) showed poor healing. No specimens fell into the no-healing category (Table 1, Fig 5). In group B, gross observation detected 10.0% nonhealing holes (3 of 30). The same results were deter-
OBEID. ZHANG,
AND
845
WANG
FIGURE 2. Clinical observationof the healing of the holes and grooves. A, Condyle from group A showing excellent healing of bole and good healing of groove. B, Condyle from group B showing moderate healing of groove and satisfactory healing of bole. C, Condyle from group C showing poor healing of hole and groove. D, Condyle from group C showing no healing of hole.
mined by histologic examination. Among the remaining 27 specimens, 10 holes showed excellent healing (33.3%) 11 satisfactory healing (36.7%), and six poor healing (20%). There was a statistically significant difference between group A and B (P < .05). In group C, gross observation showed 13.3% nonhealing holes (4 of 30). The same results were determined by histologic examination. Among the remaining 26 specimens, healing in 6 holes was considered
excellent (20%) satisfactory in 12 (40%) and poor in 8 (26.7%). There was a highly significant difference between groups A and C (P < .Ol), but no statistically significant difference between groups B and C
(P> .05). HEALING
OF THE GROOVE
In group A, 62.5% of specimens (15 of 24) showed good healing of cartilage. Only 12.5% (3 of 24) had
846
EFFECT OF IBUPROFEN ON RABBIT TMJ
FIGURE 3. Histologic evaluation of holes (hematoxylin-eosin stains). A, “Excellent healing” hole from group A (original magnification X40). Arrows indicate the limit of hole. B, “Satisfactory healing” hole from group B (original magnification X40). C, “Poor healing” hole from group B (original magnification X40). D. “No healing” hole from group C (original magnification x40).
poor healing. In group B, 23.3% (7 of 30) showed good healing and 33.3% (10 of 30) showed poor healing. In group C, 23.3% (7 of 30) showed good healing and 46.7% (14 of 30) showed poor healing (Table 2, Fig 6). There was a significant difference between groups A and B (P < .05) and highly significant difference between groups A and C (P c .Ol) There was no significant difference between groups B and C (P> .05). The results, however, appeared to be better in group B. For most animals, similar clinical and histologic results for the healing of the grooves and the holes were observed on both sides of each animal. Discussion Studies have shown that ibuprofen has an inhibitory effect on bone formation, resorption, metabolism, and remodeling.2’-27 Tornkvist et al” found that ibuprofen reduced the bone mass and composition of both fractured and nonfractured rat tibia. Ibuprofen also activated calcium metabolism in the bony callus.21,22 In another study, the effect of ibuprofen and flurbiprofen on heterotopic ossification was evaluated.26
After implanting pieces of demineralized bone matrix in the abdominal muscle of growing rats, it was discovered that these drugs caused a decrease in the amount of calcium and phosphate at a constant ratio. A similar study in humans showed that ibuprofen inhibited heterotopic ossification in patients following hip replacement. 23 Rissing et al noted that ibuprofen inhibited bone resorption, minimizing radiologic changes after induced osteomyelitis in rats.28 Studies of the effect of indomethacin, flurbiprofen, and ibuprofen on periodontal disease in beagles showed that all three of the NSAIDs inhibited alveolar bone loss in the treated animals.24,29Tomkvist et a12’studied the effect of ibuprofen and indomethacin on bone metabolism and strength in surgically weakened femurs in rabbits. They found that these drugs, in pharmacologically active doses, decreased the velocity of bone formation and torsional strength. Trancik et al,25 in their quantitative histomorphometric study, also showed that indomethacin, ibuprofen, and high-dose aspirin caused a statistically significant decrease in bone ingrowth with a porous-coated chrome-cobalt implant
847
OBEID, ZHANG. AND WANG
FIGURE 4. Histologic evaluation of grooves (hematoxylin-eosin stains). A, “Good healing” groove from group A (original magnification X40). B, “Moderate healing” groove from group B (original magnification X40). C, “Poor healing” groove from group C (original magnification X40).
in rabbits. Proper et a130reported a case in which wound healing after surgery was impaired by using ibuprofen for an extended period. There is evidence that NSAIDs inhibit the synthesis of prostaglandins by blocking the cyclooxygenase pathway. ‘* Prostaglandins are important mediators of inflammation and bone metabolism and remodeling. 3’.32Some studies have shown that their level increases during fracture healing and osteomyelitis in rabbits.33 It has been found that the use of sodium salicyiate reduces the release of prostaglandins and diminishes the radiographic evidence of bone destruction and new bone formation. 34 Like the previous studies, our results showed that ibuprofen interfered with the healing and remodeling of bone and cartilage in the TMJ of rabbits, possibly as a result of inhibition of prostaglandin synthesis during the healing process. The data presented in this article indicate that there was an increased inhibitory effect on bone healing and cartilage remodeling in the animals that received the higher dose of ibuprofen. Statistical analysis showed a highly significant difference (P -C.Ol) between the control and the high-dose ibuprofen groups, which was consistent for both hole and cartilage healing and remodeling. In comparison, there was only a significant difference (P < .05) between the control and the lowdose ibuprofen groups.
Although there was no significant difference between the two ibuprofen groups, holes in the excellent healing category and grooves in the good healing category were less frequent in the highdose group. Similarly, the highdose group had more holes and grooves in the no healing and poor healing categories. These results, although inconclusive, seem to indicate that the inhibitory effect of ibuprofen on bone and cartilage healing in the TMJ of the rabbit is dose-related. Williams et al showed similar results in a study on the effect of different doses of ibuprofen on alveolar bone resorption in beagles. Their data showed that a tenfold reduction in dose
Table 1. The Healing of the Holes
Healing
No. in Group A (%)
Excellent Satisfactory Poor No healing
15 (62.5) 5 (20.8) 4 (16.7) 0 (0)
Total
24
No. in Group B (%)
No. in Group C (%)
10 (33.3)
6 (20.0) 12 (40.0) 8 (26.7) 4 (13.3)
1I (36.7) 6 (20.0) 3 (10.0) 30
30
Highly significant difference Between groups A and C (P < .Ol). Significant difference between groups A and B (P < .05). No significant difference between groups B and C (P > .05).
848
EFFECT OF IBUPROFEN ON RABBIT TMJ
0 EXCELLENT
SATISFACTORY
POOR
NO HEALINQ
FIGURE 5. q, group c.
Healing of the holes. R, Group A; lB, group B,
reduced the effect of ibuprofen on alveolar bone resportion. 24 Trancik et al also found that there was a dose-related response with use of indomethacin and aspirin, with higher doses having a greater inhibitory effect.25 During the experimental study we noticed that the animals in the ibuprofen groups had returned to their normal eating habits 2 to 3 days before the control group. This was probably related to the analgesic effect of the ibuprofen. However, this brief reduction in movement of the condyles in the control group did not seem to affect the healing pattern. The situation lasted only a few days, following which all animals were eating and gaining weight in a similar fashion. The data from this study suggest that ibuprofen has an adverse effect on the healing of bone and cartilage. It is conceivable that although it treated pain and inflammation in the TMJ, this drug unintentionally hampered the natural process of healing and remodeling, extending the time required for the condyle to adapt. This remodeling process could be further delayed by an increase in the dose used to treat patients who are not responding favorably. Further studies are needed to see whether the experimental findings have clinical application. Table 2.
The Healing of the Grooves No. in Group A (W)
No. in Group B (%)
No. in Group C (%)
Good Moderate Poor
I5 (62.5)
I (23.3) 13 (43.4) 10 (33.3)
7 (23.3) 9 (30.0) 14 (46.7)
Total
24
30
30
Healing
6 (25.0) 3 (12.5)
0 QOOD
HEALING
Highly significant difference between groups A and C (P < .Ol). Significant difference between groups A and B (P < .05). No signiticant difference between groups B and C (P > .05).
FIGURE 6. group c.
HEALING
MODERATE
HEALING
POOR
HEALING
Healing of the grooves. n, Group A, qI,group B; q,
References I. Morgan DH, Hall WP, Vamvas SJ, et al: Disease of the Temporomandibular Apparatus (ed 2). St Louis, Mosby, 1982 2. Blaustein DI, Scapino RP: Remodeling of the temporomandibular joint disc and posterior attachment in disc displacement specimens in relation to glycosaminoglycan content. P&t Reconstr Surg 78:756, 1986 3. Moffett BC Jr, Johnson LC, McCabe JB, et al: Articular remodeling in the adult human temporomandibular joint. Am J Anat 115:119, 1964 4. Carlsson GE: Long term effects of treatment of craniomandibular disorders. J Craniomandib Pratt 3:337, 1987 5. McNeil1 C: Non-surgical management of internal derangements, in Helms CA, Katzberg RW, Dolwick MF (eds): Internal Derangements ofthe Temporomandibular Joint. San Francisco, CA, Radiology Research and Education Foundation, 1985, p 193 6. Rasmussen DC: Description of population and Droaess of symptoms in a longitudinal study of temporomandibular arthropathy. Stand J Dent Res 89:196, 198 1 7. Von K&M, Dworkin SF, LeResche L, et ak An epidemiologic comparison of pain complaints. Pain 32: 173, 1988 8. Dworkin SF, Huggins KH, LeResche L, et al: Epidemiology of signs and temporomandibular disorders: Clinical signs in cases and controls. J Am Dent Assoc 120:273, 1990 9. DiPerri T, Auteri A, Caruso I, et al: Multicenter double-blind randomized clinical trial of imidazole salicylate versus ibuprofen in patients with osteoarthrosis. Int J Clin Pharmocol Ther Toxic01 25:479, 1987 10. Hart FD, Husk&on EC: Non-steroidal anti-inflammatory drugs. Drugs 27:232, 1984 11. Busson M: Update on ibuprofen: Review article. J Int Med Res 14:53, 1986 12. Davies EF, Avery GS: Ibuprofen: A review of its pharmacological properties and therapeutic efficacy in rheumatic disorders. Drugs 2:416, 1971 13. Ruoff G, Williams S, Cooper W, et al: Aspirin-acetaminophen vs. ibuprofen in a controlled multicenter double-blind study with patients experiencing pain associated with arthritis. Curr Ther Res 31:821, 1982 14. Cooper SA, Precheur H, Rosenheck R, et al: The analgesic efficacy of ibuprofen compared to acetaminophen with codeine. Clin Pharmacol Ther 35:232, 1984 15. Troullos ES, Hargreaves KM, Butler DP: Comparison of nonsteroidal, anti-inflammatory drugs, ibuprofen and flurbiprofen, with methylprednisolone and placebo for acute pain, swelling, and trismus. J Oral Maxillofac Surg 48:945, 1990 16. McQuary HJ, Carroll D, Watts PG, et al: Codeine 20 mg increases pain relief from ibuprofen 400 mg after third molar surgery.
849
MARJORIE JEFFCOAT
17.
18. 19.
20. 2 1.
22.
23.
24.
25.
A repeat-dosing comparison of ibuprofen and an ibuprofencodeine combination. Pain 34:7, 1989 Beaver WT. Forbes JA, Barkasz BA, et al: An evaluation of ibuprofen and acetaminophen in postoperative oral surgery pain. Clin Pharmacol Ther 41:180, 1987 Abramowicz M: Ibuprofen vs acetaminophen in children. 3 I : 109, 1989 Mehlisch DR, Sollecito WA, Helfrick JE, et al: Multicenter chnical trial of ibuprofen and acetaminophen in the treatment of postoperative dental pain. J Am Dent Assoc I2 1:257, 1990 Cooper SA: Five studies on ibuprofen for postsurgical dental pain. Am J Med 77:70. 1984 Tomkvist H, Lindholm TS, Netz P. et al: Effect of ibuprofen and indomethacin on bone metabolism reflected in bone strength. Clin Orthop 187:255, 1984 Lindholm TS, Tomkvist H: Ibuprofen effect on bone formation and calcification exerted by the anti-inflammatory drug ibuprofen. Stand J Rheumatol 10:38, 1981 Elmstedt E, Lindholm TS, Nilsson OS, et al: Effect of ibuprofen on heterotopic ossification after hip replacement. Acta Orthop Stand 5625, 1985 Williams RC, Jeffcoat MK. Howell TH, et al: Ibuprofen: An inhibitor of alveolar bone resorption in beagles. J Periodont Res 23:225, 1988 Trancik T, Mills W, Vinson N: The effect of indomethacin. as-
J Oral Maxillofac 50:849-850,
pirin and ibuprofen on one ingrowth into a porous-coated implant. Clin Orthop 249: 113. 1989 26. Tomkvist H, Nilsson OS, Bauer FCH, et al: Experimentally induced heterotopic ossification in rats influenced by anti-inflammatory drugs. Stand J Rheumatol 12:177, 1983 27. Tomkvist H, Lindholm TS: Effect of ibuprofen on mass and composition of fracture callus and bone. Stand J Rheumatol 9:167, 1980 28. Rissing JP, Buxton IB: Effect of ibuprofen on gross pathology, bacterial count, and levels of prostaglandin E2 in experimental staphylococcal osteomyelitis. J Infect Dis 154:627, 1986 29. Williams RC, Jeffcoat MK, Howell TH. et al: Indomethacin or flurbiprofen treatment of periodontitis in beagles: Comparison of effect on bone loss. J Periodont Res 22:403, 1987 30. Proper SA, Fenske NA, Burnett SM, et al: Compromised wound repair caused by perioperative use of ibuprofen. J Am Acad Dermatol 18:1173, 1988 3 1. Kaley G, Weiner R: Prostaglandin E 1: A potential mediator of inflammatory response. Am NY Acad Sci 1803338, 1971 32. Dietric JW, Raisz LG: Prostaglandin in calcium and bone metabolism. Clin Orthop 111:228, 1975 33. Dekel S, Lenthall G, Francis MJO: Release of pro$taglandins from bone and muscle after tibia1 fracture. J Bone Joint Surg 63:185. 1981 34. Dekel S, Francis MJO: The treatment of osteomyelitis of the tibia with sodium salicylate. J Bone Joint Surg 63:178, 1981
Surg
1992
Discussion Effect of Ibuprofen on the Healing and Remodeling of Bone and Articular Cartilage in the Rabbit Temporomandibular Joint Marjorie Jefioat,
DMD
University of Alabama School of Dentistry, Birmingham
Dr Obeid and colleagues are to be congratulated on their excellent article, which provides evidence that nonsteroidal anti-inflammatory drugs (NSAIDs) may adversely influence healing in a rabbit temporomandibular joint model. This article addresses a topical and clinically important controversy in oral and maxillofacial surgery, orthopedics, and rheumatology. Many studies, including the present report, point to an inhibitory effect of NSAIDs on healing, whereas others have reported either no effect or enhanced bone formation. This is no small point, because NSAIDs are among the most frequently prescribed medications in the United States today. Patients may be taking these medications over a long or short period either for their analgesic or anti-inflammatory properties. At first glance, the hypothesis underlying the effect(s) of NSAIDs on bone may appear deceptively straightforward. The NSAID used in Dr Obeid’s study, ibuprofen, like all drugs of its class, is a cyclooxygenase inhibitor. Thus, it exerts its anti-inflammatory effects by inhibiting the conversion of arachadonic acid to the prostagIandin series of metabolites. Prostaglandins themselves have been implicated in both bone resorption and formation.‘,’ Prostaglandin levels often are elevated in a wide range of inflamed tissues, and prostagIandin E2 (PGE,) has been shown to cause bone resorption in vitro.
Following this line of reasoning, it is plausible to hypothesize that inhibiting conversion of arachadonic acid to prostaglandin(s) could result in a bone sparing effect. Paradoxically, appropriate pharmacological levels of PGEs in vivo have been shown to enhance bone formation. Thus, these data may be used to support the hypothesis that NSAID administration could inhibit bone formation. The reasons for the apparently contradictory res#ultsusing in vivo models for study ofthe effects of NSAIDs on cartilage and bone fall into at least three basic categories, including 1) the model or disease entity studied, 2) the dosage administered, and 3) the specific NSAID under study. Each of these categories will be briefly addressed. Data supporting a negative role for NSAIDs in healing are numerous, and include the present study of artificially induced temporomandibular joint lesions, fracture healing without rigid fixation,3,4 and models of orthopedic implants.5,6 On the other hand, no effect on fracture healing has been observed in a fracture model with rigid fixation.* Positive effects on bone have been described in an immobilized rat model for osteoporosis,’ in periodontitis7** and preliminary reports have shown increased bone density in and around titanium implants in patients receiving influence of NSAIDS.~,” To further complicate the picture, when considering the literature on NSAIDs one must also include potential effects on cartilage because temporomandibular joint disorders may also involve this tissue.“J2 In different experimental models, NSAIDs have exhibited either cartilage sparing or inhibitory effects on proteoglycan synthesis. Furthermore, it has been suggested that the inhibition of proteoglycan synthesis is greater in the diseased and inflamed cartilage of osteoarthritis
Surg
1992
Effect of Ibuprofen on the Healing and Remodeling of Bone and Articular Cartilage in the Rabbit Temporomandibular Joint GEORGE
OBEID,
DDS,*
XIEN
ZHANG,
DDS,t
AND
WANG,
MD,
MSD,
PHD$
The purpose of this study was to determine the effect of ibuprofen on the healing and remodeling of bone and cartilage in the temporomandibular joint of the rabbit. Forty-two rabbits were operated on to create a groove and a hole in the articular surface of both the right and left mandibular condyles. Following surgery, the animals were divided into three groups. Group A (12 rabbits) was used as a control and the animals did not receive any medication. Group B (15 rabbits) was given a daily dose of 17 mg/kg of ibuprofen. Group C (15 rabbits) was given a daily dose of 34 mg/kg of ibuprofen. All animals were killed after 4 weeks. The 84 condyles were examined clinically and histologically. Statistical analysis showed a highly significant difference in the healing of bone and cartilage between groups A and C (P < .Ol) and a significant difference between groups A and B (P < .05). The results of this study indicate that ibuprofen has an adverse effect on the healing of bone and cartilage in the temporomandibular joint of the rabbit.
the inhibition of prostaglandins, ibuprofen has an inhibitory effect on bone healing, remodeling, resorption, and metabolism.2’-27 Our hypothesis is that the capacity of the TMJ to remodel is hampered and delayed by the use of ibuprofen as an analgesic. This study was designed to investigate the effect of ibuprofen on healing and remodeling of bone and articular cartilage of the TMJ in rabbits.
Temporomandibular joint (TMJ) dysfunction can be caused by an abnormal relationship between the disc and the condyle. The resolution of this condition is often reached through remodeling and adaptation rather than actual correction of the anomaly.‘” Pain, the most troublesome symptom of TMJ dysfunction, tends to decrease in intensity with advancing patient age. This indicates that the condition is self-limiting, with a generally favorable outcome. However, it is estimated that without treatment, the pain can last as long as 7 years before decreasing.4-8 Ibuprofen, a nonsteroidal anti-inflammatory drug (NSAID), is widely used for the treatment of joint pain and inflammation.9-20 Studies have shown that, through
Materials and Methods Forty-two adult female New Zealand white rabbits (Blue and Gray Rabbitry, Aylett, VA) were used. After general anesthesia was administered, the skin over the TMJ region was shaved and prepared with Betadine (Purdue Frederick, Norwalk, CT) solution. A 2-cm preauricular incision was made to expose the condylar process. The zygomatic arch that normally covers a large part of the condylar process in the rabbit was exposed and partially resected to facilitate the surgical access. The inferior joint compartment was opened from the lateral aspect after incising the capsule and reflecting the disc. A 1S-mm-wide and 2-mm-deep mediolateral groove was made across the top of the condyle at the junction between the anterior and middle third of the articular surface (Fig 1). Then a 1.5-mm-diameter hole was drilled from the lateral to the medial side just beneath the articular cartilage and 5 mm behind the
Received from the Department of Oral and Maxillofacial Surgery, Washington Hospital Center, Washington, DC. * Assistant Director, Residency Training Program. t Visiting Research Fellow; Professor, Beijing Medical University, School of Stomatology, Department of Oral and Maxillofacial Surgery, Beijing, China. $ Visiting Research Fellow; Associate Professor, Beijing Medical Universitv. School of Stomatology, Department of Oral and Maxillofacial Surgery, Beijing, China-. Sunnorted bv NIH Grant DE856/ 1990. .. Address correspondence and reprint requests to Dr Obeid: Department of Oral and Maxillofacial Surgery, Washington Hospital Center, 110 Irving St, NW, Washington, DC 20010. 0 1992 American Association of Oral and Maxillofacial Surgeons 0278-2391/92/5008-0010$3.00/O
843
044
FIGURE 1. Diagramshowingthe locationand sizeof the hole and groove.
groove. A IS-mm-diameter fissure bur was used for the drilling, with copious amounts of saline for cooling. The disc was replaced in its normal position and the incision was closed in layers. Identical procedures were done bilaterally in all animals. Postoperatively, 1 mL of intramuscular Combiotics (penicillin-streptomycin; Fermenta Animal Health Company, Kansas City, MO) was given daily for 7 days. None of the wounds became infected and all rabbits did well following surgery. After surgery, the 42 rabbits were divided into three groups (A, B, and C). Group A (control) consisted of 12 rabbits that did not receive any experimental medication after surgery. Group B consisted of 15 rabbits that were given oral ibuprofen (8.5 mg/kg) twice daily (Whitehall Laboratories Inc, New York, NY). Group C also consisted of 15 rabbits that were given oral ibuprofen (17 mg/kg) twice daily. The dosages of 17 mg/ kg/d and 34 mg/kg/d are equivalent to giving a 70-kg man 1,200 mg and 2,400 mg ibuprofen per day, respectively. The ibuprofen was started on the day of surgery and was continued for 28 days. All rabbits were weighed weekly and the dose was adjusted accordingly. After 4 weeks, the animals were killed by injecting a Nembutal sodium solution (Abbott Laboratories, North Chicago, IL) via an ear vein and the 84 condylar processes were resected from the mandibles for gross evaluation and histologic examination. The healing of the grooves and holes was examined from all sides and all specimens were photographed (Fig 2). Surface irregularities of grooves and lack of closure of holes were documented. The relationship between the groove and hole in each specimen was recorded to help locate the hole under the microscope. The clinical and histologic examinations were made by the three investigators. There was no attempt to blind them as to the source of the specimens. After completion of the gross observations, the specimens were put in 10% formalin solution for 72 hours
EFFECT OF 1BUPROFEN ON RABBIT TMJ
and then decalcified. All specimens were embedded in paraffin and cut in a way that would ensure the presence of the hole and groove in each histologic slide. Serial sections were cut at 10 pm and stained with hematoxylin and eosin. Bone healing was classified as excellent, satisfactory, poor, and no healing (Fig 3) according to the following criteria. Excellent healing. Clinical observation showed complete healing of the hole. Histologically, the hole was full of bone that was indistinguishable from the surrounding bone (Fig 3A). Satisfactory healing. Clinical observation demonstrated that the hole had healed. The histologic slide showed that most of the hole was filled with new bone, but there was still clear demarcation between the hole and surrounding bone (Fig 3B). Poor healing. A small hole or evidence of a hole could be found by gross observation. The histologic examination showed that the hole was less than half full of new bone; the remaining part was filled with connective tissue (Fig 3C). No healing. Clinical observation showed an obvious hole. Histologically there was an empty hole or a hole that was filled with connective tissue. Sometimes a limited amount of new bone could be found at the margin of the hole (Fig 3D). The healing of the grooves was classified as good, moderate, or poor (Fig 4) according to the following criteria. Good healing. The groove was tilled with new bone and cartilage. The articular cartilage was well differentiated, with a smooth and intact surface (Fig 4A). Moderate healing. The groove contained abundant, highly cellular connective tissue. A cartilage cell layer was found (Fig 4B). Poor healing. The groove was observed clinically. The histologic slide showed limited amounts of new bone and cartilage in the groove. Sometimes poorly differentiated connective tissue filled the groove (Fig 4C). x2 Analysis was used for determining the statistical significance of differences between the groups. Results HEALING OF THE HOLE
Gross observation indicated that all the holes in group A had healed. Histologic examination revealed that 62.5% (15 of 24 specimens) had excellent healing. Five of 24 (20.8%) were considered satisfactory, and 16.7% (4 of 24) showed poor healing. No specimens fell into the no-healing category (Table 1, Fig 5). In group B, gross observation detected 10.0% nonhealing holes (3 of 30). The same results were deter-
OBEID. ZHANG,
AND
845
WANG
FIGURE 2. Clinical observationof the healing of the holes and grooves. A, Condyle from group A showing excellent healing of bole and good healing of groove. B, Condyle from group B showing moderate healing of groove and satisfactory healing of bole. C, Condyle from group C showing poor healing of hole and groove. D, Condyle from group C showing no healing of hole.
mined by histologic examination. Among the remaining 27 specimens, 10 holes showed excellent healing (33.3%) 11 satisfactory healing (36.7%), and six poor healing (20%). There was a statistically significant difference between group A and B (P < .05). In group C, gross observation showed 13.3% nonhealing holes (4 of 30). The same results were determined by histologic examination. Among the remaining 26 specimens, healing in 6 holes was considered
excellent (20%) satisfactory in 12 (40%) and poor in 8 (26.7%). There was a highly significant difference between groups A and C (P < .Ol), but no statistically significant difference between groups B and C
(P> .05). HEALING
OF THE GROOVE
In group A, 62.5% of specimens (15 of 24) showed good healing of cartilage. Only 12.5% (3 of 24) had
846
EFFECT OF IBUPROFEN ON RABBIT TMJ
FIGURE 3. Histologic evaluation of holes (hematoxylin-eosin stains). A, “Excellent healing” hole from group A (original magnification X40). Arrows indicate the limit of hole. B, “Satisfactory healing” hole from group B (original magnification X40). C, “Poor healing” hole from group B (original magnification X40). D. “No healing” hole from group C (original magnification x40).
poor healing. In group B, 23.3% (7 of 30) showed good healing and 33.3% (10 of 30) showed poor healing. In group C, 23.3% (7 of 30) showed good healing and 46.7% (14 of 30) showed poor healing (Table 2, Fig 6). There was a significant difference between groups A and B (P < .05) and highly significant difference between groups A and C (P c .Ol) There was no significant difference between groups B and C (P> .05). The results, however, appeared to be better in group B. For most animals, similar clinical and histologic results for the healing of the grooves and the holes were observed on both sides of each animal. Discussion Studies have shown that ibuprofen has an inhibitory effect on bone formation, resorption, metabolism, and remodeling.2’-27 Tornkvist et al” found that ibuprofen reduced the bone mass and composition of both fractured and nonfractured rat tibia. Ibuprofen also activated calcium metabolism in the bony callus.21,22 In another study, the effect of ibuprofen and flurbiprofen on heterotopic ossification was evaluated.26
After implanting pieces of demineralized bone matrix in the abdominal muscle of growing rats, it was discovered that these drugs caused a decrease in the amount of calcium and phosphate at a constant ratio. A similar study in humans showed that ibuprofen inhibited heterotopic ossification in patients following hip replacement. 23 Rissing et al noted that ibuprofen inhibited bone resorption, minimizing radiologic changes after induced osteomyelitis in rats.28 Studies of the effect of indomethacin, flurbiprofen, and ibuprofen on periodontal disease in beagles showed that all three of the NSAIDs inhibited alveolar bone loss in the treated animals.24,29Tomkvist et a12’studied the effect of ibuprofen and indomethacin on bone metabolism and strength in surgically weakened femurs in rabbits. They found that these drugs, in pharmacologically active doses, decreased the velocity of bone formation and torsional strength. Trancik et al,25 in their quantitative histomorphometric study, also showed that indomethacin, ibuprofen, and high-dose aspirin caused a statistically significant decrease in bone ingrowth with a porous-coated chrome-cobalt implant
847
OBEID, ZHANG. AND WANG
FIGURE 4. Histologic evaluation of grooves (hematoxylin-eosin stains). A, “Good healing” groove from group A (original magnification X40). B, “Moderate healing” groove from group B (original magnification X40). C, “Poor healing” groove from group C (original magnification X40).
in rabbits. Proper et a130reported a case in which wound healing after surgery was impaired by using ibuprofen for an extended period. There is evidence that NSAIDs inhibit the synthesis of prostaglandins by blocking the cyclooxygenase pathway. ‘* Prostaglandins are important mediators of inflammation and bone metabolism and remodeling. 3’.32Some studies have shown that their level increases during fracture healing and osteomyelitis in rabbits.33 It has been found that the use of sodium salicyiate reduces the release of prostaglandins and diminishes the radiographic evidence of bone destruction and new bone formation. 34 Like the previous studies, our results showed that ibuprofen interfered with the healing and remodeling of bone and cartilage in the TMJ of rabbits, possibly as a result of inhibition of prostaglandin synthesis during the healing process. The data presented in this article indicate that there was an increased inhibitory effect on bone healing and cartilage remodeling in the animals that received the higher dose of ibuprofen. Statistical analysis showed a highly significant difference (P -C.Ol) between the control and the high-dose ibuprofen groups, which was consistent for both hole and cartilage healing and remodeling. In comparison, there was only a significant difference (P < .05) between the control and the lowdose ibuprofen groups.
Although there was no significant difference between the two ibuprofen groups, holes in the excellent healing category and grooves in the good healing category were less frequent in the highdose group. Similarly, the highdose group had more holes and grooves in the no healing and poor healing categories. These results, although inconclusive, seem to indicate that the inhibitory effect of ibuprofen on bone and cartilage healing in the TMJ of the rabbit is dose-related. Williams et al showed similar results in a study on the effect of different doses of ibuprofen on alveolar bone resorption in beagles. Their data showed that a tenfold reduction in dose
Table 1. The Healing of the Holes
Healing
No. in Group A (%)
Excellent Satisfactory Poor No healing
15 (62.5) 5 (20.8) 4 (16.7) 0 (0)
Total
24
No. in Group B (%)
No. in Group C (%)
10 (33.3)
6 (20.0) 12 (40.0) 8 (26.7) 4 (13.3)
1I (36.7) 6 (20.0) 3 (10.0) 30
30
Highly significant difference Between groups A and C (P < .Ol). Significant difference between groups A and B (P < .05). No significant difference between groups B and C (P > .05).
848
EFFECT OF IBUPROFEN ON RABBIT TMJ
0 EXCELLENT
SATISFACTORY
POOR
NO HEALINQ
FIGURE 5. q, group c.
Healing of the holes. R, Group A; lB, group B,
reduced the effect of ibuprofen on alveolar bone resportion. 24 Trancik et al also found that there was a dose-related response with use of indomethacin and aspirin, with higher doses having a greater inhibitory effect.25 During the experimental study we noticed that the animals in the ibuprofen groups had returned to their normal eating habits 2 to 3 days before the control group. This was probably related to the analgesic effect of the ibuprofen. However, this brief reduction in movement of the condyles in the control group did not seem to affect the healing pattern. The situation lasted only a few days, following which all animals were eating and gaining weight in a similar fashion. The data from this study suggest that ibuprofen has an adverse effect on the healing of bone and cartilage. It is conceivable that although it treated pain and inflammation in the TMJ, this drug unintentionally hampered the natural process of healing and remodeling, extending the time required for the condyle to adapt. This remodeling process could be further delayed by an increase in the dose used to treat patients who are not responding favorably. Further studies are needed to see whether the experimental findings have clinical application. Table 2.
The Healing of the Grooves No. in Group A (W)
No. in Group B (%)
No. in Group C (%)
Good Moderate Poor
I5 (62.5)
I (23.3) 13 (43.4) 10 (33.3)
7 (23.3) 9 (30.0) 14 (46.7)
Total
24
30
30
Healing
6 (25.0) 3 (12.5)
0 QOOD
HEALING
Highly significant difference between groups A and C (P < .Ol). Significant difference between groups A and B (P < .05). No signiticant difference between groups B and C (P > .05).
FIGURE 6. group c.
HEALING
MODERATE
HEALING
POOR
HEALING
Healing of the grooves. n, Group A, qI,group B; q,
References I. Morgan DH, Hall WP, Vamvas SJ, et al: Disease of the Temporomandibular Apparatus (ed 2). St Louis, Mosby, 1982 2. Blaustein DI, Scapino RP: Remodeling of the temporomandibular joint disc and posterior attachment in disc displacement specimens in relation to glycosaminoglycan content. P&t Reconstr Surg 78:756, 1986 3. Moffett BC Jr, Johnson LC, McCabe JB, et al: Articular remodeling in the adult human temporomandibular joint. Am J Anat 115:119, 1964 4. Carlsson GE: Long term effects of treatment of craniomandibular disorders. J Craniomandib Pratt 3:337, 1987 5. McNeil1 C: Non-surgical management of internal derangements, in Helms CA, Katzberg RW, Dolwick MF (eds): Internal Derangements ofthe Temporomandibular Joint. San Francisco, CA, Radiology Research and Education Foundation, 1985, p 193 6. Rasmussen DC: Description of population and Droaess of symptoms in a longitudinal study of temporomandibular arthropathy. Stand J Dent Res 89:196, 198 1 7. Von K&M, Dworkin SF, LeResche L, et ak An epidemiologic comparison of pain complaints. Pain 32: 173, 1988 8. Dworkin SF, Huggins KH, LeResche L, et al: Epidemiology of signs and temporomandibular disorders: Clinical signs in cases and controls. J Am Dent Assoc 120:273, 1990 9. DiPerri T, Auteri A, Caruso I, et al: Multicenter double-blind randomized clinical trial of imidazole salicylate versus ibuprofen in patients with osteoarthrosis. Int J Clin Pharmocol Ther Toxic01 25:479, 1987 10. Hart FD, Husk&on EC: Non-steroidal anti-inflammatory drugs. Drugs 27:232, 1984 11. Busson M: Update on ibuprofen: Review article. J Int Med Res 14:53, 1986 12. Davies EF, Avery GS: Ibuprofen: A review of its pharmacological properties and therapeutic efficacy in rheumatic disorders. Drugs 2:416, 1971 13. Ruoff G, Williams S, Cooper W, et al: Aspirin-acetaminophen vs. ibuprofen in a controlled multicenter double-blind study with patients experiencing pain associated with arthritis. Curr Ther Res 31:821, 1982 14. Cooper SA, Precheur H, Rosenheck R, et al: The analgesic efficacy of ibuprofen compared to acetaminophen with codeine. Clin Pharmacol Ther 35:232, 1984 15. Troullos ES, Hargreaves KM, Butler DP: Comparison of nonsteroidal, anti-inflammatory drugs, ibuprofen and flurbiprofen, with methylprednisolone and placebo for acute pain, swelling, and trismus. J Oral Maxillofac Surg 48:945, 1990 16. McQuary HJ, Carroll D, Watts PG, et al: Codeine 20 mg increases pain relief from ibuprofen 400 mg after third molar surgery.
849
MARJORIE JEFFCOAT
17.
18. 19.
20. 2 1.
22.
23.
24.
25.
A repeat-dosing comparison of ibuprofen and an ibuprofencodeine combination. Pain 34:7, 1989 Beaver WT. Forbes JA, Barkasz BA, et al: An evaluation of ibuprofen and acetaminophen in postoperative oral surgery pain. Clin Pharmacol Ther 41:180, 1987 Abramowicz M: Ibuprofen vs acetaminophen in children. 3 I : 109, 1989 Mehlisch DR, Sollecito WA, Helfrick JE, et al: Multicenter chnical trial of ibuprofen and acetaminophen in the treatment of postoperative dental pain. J Am Dent Assoc I2 1:257, 1990 Cooper SA: Five studies on ibuprofen for postsurgical dental pain. Am J Med 77:70. 1984 Tomkvist H, Lindholm TS, Netz P. et al: Effect of ibuprofen and indomethacin on bone metabolism reflected in bone strength. Clin Orthop 187:255, 1984 Lindholm TS, Tomkvist H: Ibuprofen effect on bone formation and calcification exerted by the anti-inflammatory drug ibuprofen. Stand J Rheumatol 10:38, 1981 Elmstedt E, Lindholm TS, Nilsson OS, et al: Effect of ibuprofen on heterotopic ossification after hip replacement. Acta Orthop Stand 5625, 1985 Williams RC, Jeffcoat MK. Howell TH, et al: Ibuprofen: An inhibitor of alveolar bone resorption in beagles. J Periodont Res 23:225, 1988 Trancik T, Mills W, Vinson N: The effect of indomethacin. as-
J Oral Maxillofac 50:849-850,
pirin and ibuprofen on one ingrowth into a porous-coated implant. Clin Orthop 249: 113. 1989 26. Tomkvist H, Nilsson OS, Bauer FCH, et al: Experimentally induced heterotopic ossification in rats influenced by anti-inflammatory drugs. Stand J Rheumatol 12:177, 1983 27. Tomkvist H, Lindholm TS: Effect of ibuprofen on mass and composition of fracture callus and bone. Stand J Rheumatol 9:167, 1980 28. Rissing JP, Buxton IB: Effect of ibuprofen on gross pathology, bacterial count, and levels of prostaglandin E2 in experimental staphylococcal osteomyelitis. J Infect Dis 154:627, 1986 29. Williams RC, Jeffcoat MK, Howell TH. et al: Indomethacin or flurbiprofen treatment of periodontitis in beagles: Comparison of effect on bone loss. J Periodont Res 22:403, 1987 30. Proper SA, Fenske NA, Burnett SM, et al: Compromised wound repair caused by perioperative use of ibuprofen. J Am Acad Dermatol 18:1173, 1988 3 1. Kaley G, Weiner R: Prostaglandin E 1: A potential mediator of inflammatory response. Am NY Acad Sci 1803338, 1971 32. Dietric JW, Raisz LG: Prostaglandin in calcium and bone metabolism. Clin Orthop 111:228, 1975 33. Dekel S, Lenthall G, Francis MJO: Release of pro$taglandins from bone and muscle after tibia1 fracture. J Bone Joint Surg 63:185. 1981 34. Dekel S, Francis MJO: The treatment of osteomyelitis of the tibia with sodium salicylate. J Bone Joint Surg 63:178, 1981
Surg
1992
Discussion Effect of Ibuprofen on the Healing and Remodeling of Bone and Articular Cartilage in the Rabbit Temporomandibular Joint Marjorie Jefioat,
DMD
University of Alabama School of Dentistry, Birmingham
Dr Obeid and colleagues are to be congratulated on their excellent article, which provides evidence that nonsteroidal anti-inflammatory drugs (NSAIDs) may adversely influence healing in a rabbit temporomandibular joint model. This article addresses a topical and clinically important controversy in oral and maxillofacial surgery, orthopedics, and rheumatology. Many studies, including the present report, point to an inhibitory effect of NSAIDs on healing, whereas others have reported either no effect or enhanced bone formation. This is no small point, because NSAIDs are among the most frequently prescribed medications in the United States today. Patients may be taking these medications over a long or short period either for their analgesic or anti-inflammatory properties. At first glance, the hypothesis underlying the effect(s) of NSAIDs on bone may appear deceptively straightforward. The NSAID used in Dr Obeid’s study, ibuprofen, like all drugs of its class, is a cyclooxygenase inhibitor. Thus, it exerts its anti-inflammatory effects by inhibiting the conversion of arachadonic acid to the prostagIandin series of metabolites. Prostaglandins themselves have been implicated in both bone resorption and formation.‘,’ Prostaglandin levels often are elevated in a wide range of inflamed tissues, and prostagIandin E2 (PGE,) has been shown to cause bone resorption in vitro.
Following this line of reasoning, it is plausible to hypothesize that inhibiting conversion of arachadonic acid to prostaglandin(s) could result in a bone sparing effect. Paradoxically, appropriate pharmacological levels of PGEs in vivo have been shown to enhance bone formation. Thus, these data may be used to support the hypothesis that NSAID administration could inhibit bone formation. The reasons for the apparently contradictory res#ultsusing in vivo models for study ofthe effects of NSAIDs on cartilage and bone fall into at least three basic categories, including 1) the model or disease entity studied, 2) the dosage administered, and 3) the specific NSAID under study. Each of these categories will be briefly addressed. Data supporting a negative role for NSAIDs in healing are numerous, and include the present study of artificially induced temporomandibular joint lesions, fracture healing without rigid fixation,3,4 and models of orthopedic implants.5,6 On the other hand, no effect on fracture healing has been observed in a fracture model with rigid fixation.* Positive effects on bone have been described in an immobilized rat model for osteoporosis,’ in periodontitis7** and preliminary reports have shown increased bone density in and around titanium implants in patients receiving influence of NSAIDS.~,” To further complicate the picture, when considering the literature on NSAIDs one must also include potential effects on cartilage because temporomandibular joint disorders may also involve this tissue.“J2 In different experimental models, NSAIDs have exhibited either cartilage sparing or inhibitory effects on proteoglycan synthesis. Furthermore, it has been suggested that the inhibition of proteoglycan synthesis is greater in the diseased and inflamed cartilage of osteoarthritis