J Oral Maxillofac 48. 1029-1032.
Sutg 1990
A Prospective Evaluation of the Effectiveness of Temporomandibular Joint Arthroscopy DAVID H. PERROTT, DDS, MD,* AZADEH ALBORZIJ LEONARD B. KABAN, DMD, MD,* AND CLYDE A. HELMS, MD5 This is a prospective study to evaluate therapeutic arthroscopy for internal derangement of the temporomandibular joint (TMJ). Fifty-nine patients with 76 abnormal joints were evaluated preoperatively for pain, noise, maximal incisal opening (MIO), and deviation on opening. Preoperative and postoperative magnetic resonance imaging (MRI) scans were obtained for 29 joints. Patients were treated by superior joint arthroscopy, lysis of adhesions, lavage, and steroid injection, along with preoperative and postoperative splint and physiotherapy. Pain, noise, and motion were evaluated at three time periods: 1) early (10 to 30 days); 2) intermediate (1 to 6 months); and 3) late (>6 months). At early, intermediate, and late follow-up, increase in MI0 was statistically significant (P < .05). Noise did not return in the majority of patients. Disc position, evaluated by MRI, did not appear to change in 25 of 29 joints and did not correlate with clinical outcome. The results of this study indicate that TMJ arthroscopy is effective in reducing pain and increasing motion in patients with TMJ internal derangement.
studies.
Received from the University of California, San Francisco. *Assistant Professor, Department of Oral and Maxillofacial Surgery. Wenior Dental Student. SProfessor and Chairman. Department of Oral and Maxillofacial Surgery. PAssociate Professor, Department of Radiology. This work was supported in part by the UCSF Oral and Maxillofacial Research Fund and the UCSF Radiology Research and Education Fund. This paper was presented at the 71st Annual Meeting of the American Association of Oral and Maxillofacial Surgeons, San Francisco, CA, September 20 to 23, 1989. Address correspondence and reprint requests to Dr Perrott: Department of Oral and Maxillofacial Surgery, University of California, San Francisco, 513 Parnassus Ave, Rm S-738, San Francisco. CA 91943. #D 1990 American geons
Association
of Oral and Maxillofacial
1.2. IO-12
Moreover, uniformly good results have been reported in retrospective, early clinical studies of TMJ arthroscopy.2-9 However, few prospective evaluations correlating clinical and radiographic data have been published.7.9,‘3 In 1986, we began a prospective clinical evaluation of therapeutic arthroscopy for patients with pain, TMJ noise, and decreased motion. This is a report of the clinical and radiographic results in the first 59 patients (76 TMJs) treated.
Arthroscopy has recently become a common treatment modality for a variety of temporomandibular joint (TMJ) abnormalities.‘-’ Since the first description of TMJ arthroscopy in humans by Ohnishi in 1975, normal and pathologic TMJ anatomy have been documented in several animal and cadaver
Materials
and Methods
Preoperatively, patients were evaluated for age, sex, and duration of symptoms. They were classified into three groups: group 1 (organic TMJ disease or previous surgery), group 2A (myofascial pain dysfunction [MPD] syndrome with internal derangement undergoing therapeutic arthroscopy), and group 2B (MPD with internal derangement undergoing preoperative diagnostic arthroscopy). Patients were considered operative candidates when nonsurgical therapy consisting of heat, soft diet, muscle relaxants, nonsteroidal anti-inflammatory
Sur-
0278-2391/90/4810-0003$3.00/O
1029
1030
TMJ ARTHROSCOPY.
drugs (NSAID), biofeedback, physical therapy, and bite appliances, in a variety of combinations, failed to adequately control symptoms when used for an average period of 18 months. The patients quantified pain as none, mild, moderate, or severe on an analog scale (1 to IO). Presence of noise (clicking, popping, and grating) was documented. Maximal incisal opening (MIO) and hinge opening (HO) were measured in millimeters between the upper and lower incisors with a rigid triangulation device. The presence of deviation was recorded with the same device. All patients had a preoperative radiographic study: magnetic resonance imaging (MRI) (n = 42), tomograms (n = 19), arthrogram (n = 8), or a combination (n = 17). The MRI technique at University of California, San Francisco (UCSF) consists of direct sagittal views in closed and partial open mouth positions. No attempt was made to document disc reduction.14 Superior joint arthroscopy, lysis of adhesions, lavage, and steroid injection (100 mg of Depomedrol [Upjohn Co, Kalamazoo, MI]) were performed with either a Wolff (1.9 mm) or a Stryker (2.3 mm) TMJ arthroscope via the inferior lateral approach with two puncture sites. All patients received a flat plane stabilizing bite appliance preoperatively and were maintained postoperatively on the appliance and physical therapy. Physiotherapy consisted of manual stretching exercises, massage, heat, and ultrasonographic evaluation. Muscle relaxants and use of a stabilizing bite appliance were continued postoperatively for 3 months and then gradually discontinued. Results
There were 7 patients (male:female = 25) with 8 abnormal joints in group 1. The average age was 48 years (k9.4) and duration of symptoms 2.2 (kO.6) years. Three patients had degenerative joint disease, 1 had mixed connective tissue disease, 1 had psoriatic arthritis, and 2 patients had recurrent pain after a previous TMJ operation. There were 46 patients (male:female = 1:45) with 62 abnormal joints in group 2A. The mean age was 31.62 (28.21) years and duration of symptoms was 2.7 (? 1.2) years. No patient had previous TMJ surgery. Preoperative disc position based on radiographic findings included: 1) normal (n = 9), 2) displacement (n = 50), and 3) inconclusive (n = 3). Bony morphology of the condyle based on radiographic findings included: 1) degenerative changes (n = 16) and 2) normal (n = 46). Six patients (male:female = 1:5) with 6 abnormal joints were in group 2B. The mean age was 36.67 (t5.14) years and duration of symptoms was 4.75 (k2.1) years. These patients had a longer history of
PROSPECTIVE
STUDY
TMJ symptoms, smaller MIO, more severe degenerative joint disease, and a higher incidence of a perforated disc (n = 3) than patients in group 2A. Patients in this group underwent diagnostic arthroscopy before a planned arthrotomy procedure. Preoperatively, in the 62 joints (42 patients) in group 2A pain was rated as none (n = 3). mild (n = 4), moderate (n = 43), and severe (n = 12). There was noise in 50 joints (clicking, n = 33; popping, n = 12; grating, n = 5). Mean maximal incisal opening was 29.91 * 10.10 mm, with an HO of 20.02 ? 8.39 mm. Deviation toward the affected side on opening was noted in 34 patients. Following induction of anesthesia, the mean MI0 was 28.20 ? 6.7 mm, HO was 17.90 + 4.0 mm, and eight patients continued to deviate with opening. Intraoperative MIO, postarthroscopy, was 40.04 ? 9.95 mm and HO was 25.43 ? 6.50 mm. Findings at the time of arthroscopy included adhesions (n = 41). fibrillations (n = 42), disc perforation (n = 3), and synovitis (n = 14). The three patients who had perforation of the disk underwent open arthrotomy and were reclassified into group 2B. Pain, noise, and motion were reevaluated at three time periods (Tables 1 to 3): 1) early (10 to 30 days); 2) intermediate (1 to 6 months); and 3) late (>6 months). At early follow-up, 62 joints were evaluated. Pain was reported as none (n = 44), mild (n = 14), moderate (n = 2), and severe (n = 2). Noise was present in 15 joints, (clicking, n = 9; popping, n = 4; grating, n = 2). Mean maximal incisal opening was 35.28 + 10.72 mm, with deviation in 2 joints. At intermediate follow-up, (n = 62 joints), pain was reported as none (n = 49), mild (n = 8), moderate (n = 3), and severe (n = 2). Noise was present in 16 joints (clicking, n = 10; popping, n = 4; grating, n = 2). Mean maximal incisal opening was 37.80 -+ 9.69 mm with deviation in 3 joints. All patients maintained a bite appliance 24 hours per day for 7 days. After 7 days, they used the appliance at night as needed. Following 6 to 8 weeks, most patients did not require a bite appliance. Late follow-up was available for 26 patients (n = 34 joints) with a mean follow-up of 12.56 + 2.87 months (range 7 to 17). Pain was reported as none (n = 24), mild (n = 5), moderate (n = 3). and severe (n = 2). Noise was present in 3 joints (clicking, n = 2; popping, n = 1). Mean maximal incisal opening Table 1.
Pain Preoperative
None Mild Moderate Severe
3162 4162 43162 12162
Early 44162 14162 2162 2162
Intermediate
Late
49162 8162 3162 2162
24134 s/34 3134 2134
1031
PERROTT ET AL
Table 2.
Noise Preoperative
None Clicking Popping Grating
12162 33162 12162 5162
Early
Intermediate
41162 9162 4/62 2162
46162 10162 4162 2162
Late 31134 2134 II34 o/34
was 37.93 c 10.52mm, with deviation in one patient. No new cases of moderate to severe pain or noise were recorded in the patients that reached the late follow-up period. Four patients had an unsuccessful result and subsequently underwent arthroplasty (5 of 62 joints; 8.1%). One patient had two complications: temporary facial nerve palsy (all branches) and a perforated tympanic membrane that required a corrective operation. Preoperative and postoperative MRI scans were done on 29 joints. The postoperative MRI was done I to 17 months after the procedure (mean, 8.38 + 5.09). In this study, one radiologist blinded to the patients’ clinical status evaluated the preoperative and postoperative MRI scans. Disc position was unchanged in the majority of joints in (25 of 29), improved or normal in 3 of 29, and worsened in 1 of 29 joints. There appeared to be no relationship between postoperative symptoms, overall clinical result, and disc position by MRI. Discussion
Temporomandibular joint arthroscopy is a new technique, therefore, follow-up data are needed to establish norms for outcome. Sanders reported a retrospective evaluation of 25 patients with a mean follow-up of 8.3 months.’ There was resolution of closed lock, good range of motion, and little preauricular pain. He summarized the results as good to excellent. However, all observations were subjective; there was no attempt to quantitate changes in pain, motion, or noise, and there was no MRI or radiographic correlation. Montgomery et al reported the clinical and radiographic results of arthroscopy in 52 joints.’ Patients were studied at 1, 2,4, 8, 24, and 52 weeks postoperatively. The authors found that pain, mandibular movement. function, and diet were significantly improved. The incidence of joint sounds did not significantly improve beyond the 4-week evaluation. Table 3.
MO/Deviation Preoperative
MI0 (mm) Deviation
Preoperative and postoperative MRIs were obtained in 21 patients, with disc position being unchanged in 78%. Indresano followed 50 patients (78 joints) for an average of 18.9 months following arthroscopy of the superior joint space. Outcome was judged to be successful on the basis of three criteria: 1) an MI0 increase of 50%, 2) a 70% decrease in pain (on a scale of 1 to lo), and 3) no need for a second operation during a 2-year follow-up period. Patients with closed lock had an 83% success rate, those with nonclosed lock had a 50% success rate. The overall success rate was 73%. There were four complications, and 13 patients (16%) later required an open joint procedure .s In the current study, we have attempted to quantify pain, noise, and motion preoperatively and to document the evolution of change in these parameters in early (1 to 30 days), intermediate (1 to 6 months), and late (>6 months) follow-up periods. We also have attempted to correlate clinical results with findings on preoperative and postoperative MRI scans. The results are similar to those previously reported.7.Y.‘3 At early, intermediate, and late follow-up, the increase in MI0 was statistically significant (P < .05). Range of motion improved and deviation decreased in all patients. At early and intermediate follow-up, there was a statistically significant reduction in the number of joints demonstrating noise. The presence or absence of noise did not correlate with pain relief or improvement in motion. The significance of TMJ noise remains obscure from a prognostic standpoint. Postoperative MRI data were difficult to obtain because third-party payers refused to cover the expense of a postoperative MRI scan. Naturally, patients without symptoms were reluctant to pay for the study. We were able to fund MRI scans through a grant to the UCSF Department of Radiology. Although significant clinical improvement with decreased pain and increased motion occurred in the majority of patients, this could not be correlated with radiographic changes in disc position in the 29 patients who had preoperative and postoperative MRI scans. These findings, which are consistent with previous studies.7,9*‘3 raise several questions: 1) Why is there a lack of correlation between clinical and MRI results?, 2) What is it in the arthros-
29.91 2 10.10 34162
Early 35.28 + 10.72 2/62
Intermediate 37.80 2 9.69 3162
Late 37.93 * 10.52 I/34
1032
TMJ ARTHROSCOPY,
copy procedure that produces clinical improvement?, and 3) What is the significance of disc position? There are at least three potential reasons for the lack of correlation between clinical and MRI results. One, the MRI scan may be inaccurate in locating disc position. This is unlikely because of well documented cadaver and human intraoperative findings. Two, abnormal disc position is not important as a cause of TMJ symptoms; patients may function on the posterior attachments or without a disc and remain asymptomatic. Three, TMJ arthroscopy releases adhesions in the superior joint space, but not in the inferior joint space. Function returns, but with the disc tightly adhered to the condyle. Return to adequate motion, however, may allow for resolution of symptoms. There were several limitations of this prospective study. It did not have a control group, which may weaken the conclusions derived from the data. However, all patients had failed to achieve resolution of their symptoms after long periods of nonsurgical treatment. They were referred to our group for an operative procedure and would not have been receptive to randomization to another course of nonsurgical treatment. Another factor that may have potentially obscured the results was the use of an intra-articular injection of steroids. The fact that patients received a steroid injection at the time of arthroscopy might explain our early results, but would not account for the long-term resolution of pain and increased motion. Conclusion
STUDY
patient can be predicted during the early to intermediate follow-up periods, that is, failure usually appeared in the first 3 months and only rarely afterward. Acknowledgment We would like to acknowledge Anna Gerrero, DDS, and Stephen Supancic, DDS. for their assistance in gathering data for this study.
References 1.Ohnishi M: Clinical application of arthroscopy in the tem2.
3.
4. 5. 6.
7.
8. 9.
10.
1I.
Results of this study indicate that TMJ arthroscopy is effective in reducing pain and increasing motion in patients with internal derangement. However, arthroscopy as done in this study appears to have little or no affect on the position of the disc. At early, intermediate, and late follow-up, pain was reduced and increases in MI0 were statistically significant. Noise did not return in the majority of patients. It appears that the long-term condition of the
PROSPECTIVE
12. 13.
14.
poromandibular joint diseases. Bull Tokyo Med Dent Univ 27: 141, 1980 Heffez L, Blaustein D: Diagnostic arthroscopy of the temporomandibular joint. Oral Surg Oral Med Oral Path 64:653, 1987 Sanders B: Arthroscopic surgery of the temporomandibular joint: Treatment of internal derangement with persistent closed lock. Oral Surg Oral Med Oral Path 62:361, 1986 Murakami K, Ito K: Arthroscopy of the temporomandibular joint third report; clinical experiences. Arthroscopy 9:49, 1984 Tarro A: Arthroscopic diagnosis and surgery of the temporomandibular joint. J Oral Maxillofac Surg 46:282, 1984 Israel H, Roser SM: Patient response to temporomandibular joint arthroscopy: Preliminary findings in 24 patients. J Oral Maxillofac Surg 47:570, 1984 Montgomery MT, Van Sachels JE, Thrash WJ, et al: Signs and symptoms, patient satisfaction and meniscal position following TMJ surgery. AAOMFS 70th Annual Meeting & Scientific Sessions, September 29 to October 3, 1988 (abstr) Indresano AT: Arthroscopic surgery of the temporomandibular joint: Report of 64 patients with long-term follow-up. J Oral Maxillofac Surg 47:439, 1989 Mosseo JJ, Sartores D, Glass R, et al: The effect of arthroscopic surgical lysis and lavage of the superior joint space on TMJ disc position and mobility. J Oral Maxillofac Surg 471676, 1989 Hilsabeck RB, Laskin DM: Arthroscopy of the temporomandibular joint of the rabbit. J Oral Surg 36:938, 1978 Williams RA, Laskin DM: Arthroscopic examination of experimentally induced pathologic conditions of the rabbit temporomandibular joint. J Oral Surg 38:65, 1980 Holmlund A, Hellsing G: Arthroscopy of the temporomandibular joint. An autopsy study. Int J Oral Surg 14:169, 1985 McCain JP, Humberto DLR, LeBlanc W: Correlation of clinical, radiographic and arthroscopic findings in the internal derangements of the TMJ. J Oral Maxillofac Surg 47:913, 1989 Helms CA, Doyle GW, Orwig D, et al: Staging of derangements of TMJ with MRI: Preliminarv observations. J Craniomand Disorders 3:93, 1989 .
Sutg 1990
A Prospective Evaluation of the Effectiveness of Temporomandibular Joint Arthroscopy DAVID H. PERROTT, DDS, MD,* AZADEH ALBORZIJ LEONARD B. KABAN, DMD, MD,* AND CLYDE A. HELMS, MD5 This is a prospective study to evaluate therapeutic arthroscopy for internal derangement of the temporomandibular joint (TMJ). Fifty-nine patients with 76 abnormal joints were evaluated preoperatively for pain, noise, maximal incisal opening (MIO), and deviation on opening. Preoperative and postoperative magnetic resonance imaging (MRI) scans were obtained for 29 joints. Patients were treated by superior joint arthroscopy, lysis of adhesions, lavage, and steroid injection, along with preoperative and postoperative splint and physiotherapy. Pain, noise, and motion were evaluated at three time periods: 1) early (10 to 30 days); 2) intermediate (1 to 6 months); and 3) late (>6 months). At early, intermediate, and late follow-up, increase in MI0 was statistically significant (P < .05). Noise did not return in the majority of patients. Disc position, evaluated by MRI, did not appear to change in 25 of 29 joints and did not correlate with clinical outcome. The results of this study indicate that TMJ arthroscopy is effective in reducing pain and increasing motion in patients with TMJ internal derangement.
studies.
Received from the University of California, San Francisco. *Assistant Professor, Department of Oral and Maxillofacial Surgery. Wenior Dental Student. SProfessor and Chairman. Department of Oral and Maxillofacial Surgery. PAssociate Professor, Department of Radiology. This work was supported in part by the UCSF Oral and Maxillofacial Research Fund and the UCSF Radiology Research and Education Fund. This paper was presented at the 71st Annual Meeting of the American Association of Oral and Maxillofacial Surgeons, San Francisco, CA, September 20 to 23, 1989. Address correspondence and reprint requests to Dr Perrott: Department of Oral and Maxillofacial Surgery, University of California, San Francisco, 513 Parnassus Ave, Rm S-738, San Francisco. CA 91943. #D 1990 American geons
Association
of Oral and Maxillofacial
1.2. IO-12
Moreover, uniformly good results have been reported in retrospective, early clinical studies of TMJ arthroscopy.2-9 However, few prospective evaluations correlating clinical and radiographic data have been published.7.9,‘3 In 1986, we began a prospective clinical evaluation of therapeutic arthroscopy for patients with pain, TMJ noise, and decreased motion. This is a report of the clinical and radiographic results in the first 59 patients (76 TMJs) treated.
Arthroscopy has recently become a common treatment modality for a variety of temporomandibular joint (TMJ) abnormalities.‘-’ Since the first description of TMJ arthroscopy in humans by Ohnishi in 1975, normal and pathologic TMJ anatomy have been documented in several animal and cadaver
Materials
and Methods
Preoperatively, patients were evaluated for age, sex, and duration of symptoms. They were classified into three groups: group 1 (organic TMJ disease or previous surgery), group 2A (myofascial pain dysfunction [MPD] syndrome with internal derangement undergoing therapeutic arthroscopy), and group 2B (MPD with internal derangement undergoing preoperative diagnostic arthroscopy). Patients were considered operative candidates when nonsurgical therapy consisting of heat, soft diet, muscle relaxants, nonsteroidal anti-inflammatory
Sur-
0278-2391/90/4810-0003$3.00/O
1029
1030
TMJ ARTHROSCOPY.
drugs (NSAID), biofeedback, physical therapy, and bite appliances, in a variety of combinations, failed to adequately control symptoms when used for an average period of 18 months. The patients quantified pain as none, mild, moderate, or severe on an analog scale (1 to IO). Presence of noise (clicking, popping, and grating) was documented. Maximal incisal opening (MIO) and hinge opening (HO) were measured in millimeters between the upper and lower incisors with a rigid triangulation device. The presence of deviation was recorded with the same device. All patients had a preoperative radiographic study: magnetic resonance imaging (MRI) (n = 42), tomograms (n = 19), arthrogram (n = 8), or a combination (n = 17). The MRI technique at University of California, San Francisco (UCSF) consists of direct sagittal views in closed and partial open mouth positions. No attempt was made to document disc reduction.14 Superior joint arthroscopy, lysis of adhesions, lavage, and steroid injection (100 mg of Depomedrol [Upjohn Co, Kalamazoo, MI]) were performed with either a Wolff (1.9 mm) or a Stryker (2.3 mm) TMJ arthroscope via the inferior lateral approach with two puncture sites. All patients received a flat plane stabilizing bite appliance preoperatively and were maintained postoperatively on the appliance and physical therapy. Physiotherapy consisted of manual stretching exercises, massage, heat, and ultrasonographic evaluation. Muscle relaxants and use of a stabilizing bite appliance were continued postoperatively for 3 months and then gradually discontinued. Results
There were 7 patients (male:female = 25) with 8 abnormal joints in group 1. The average age was 48 years (k9.4) and duration of symptoms 2.2 (kO.6) years. Three patients had degenerative joint disease, 1 had mixed connective tissue disease, 1 had psoriatic arthritis, and 2 patients had recurrent pain after a previous TMJ operation. There were 46 patients (male:female = 1:45) with 62 abnormal joints in group 2A. The mean age was 31.62 (28.21) years and duration of symptoms was 2.7 (? 1.2) years. No patient had previous TMJ surgery. Preoperative disc position based on radiographic findings included: 1) normal (n = 9), 2) displacement (n = 50), and 3) inconclusive (n = 3). Bony morphology of the condyle based on radiographic findings included: 1) degenerative changes (n = 16) and 2) normal (n = 46). Six patients (male:female = 1:5) with 6 abnormal joints were in group 2B. The mean age was 36.67 (t5.14) years and duration of symptoms was 4.75 (k2.1) years. These patients had a longer history of
PROSPECTIVE
STUDY
TMJ symptoms, smaller MIO, more severe degenerative joint disease, and a higher incidence of a perforated disc (n = 3) than patients in group 2A. Patients in this group underwent diagnostic arthroscopy before a planned arthrotomy procedure. Preoperatively, in the 62 joints (42 patients) in group 2A pain was rated as none (n = 3). mild (n = 4), moderate (n = 43), and severe (n = 12). There was noise in 50 joints (clicking, n = 33; popping, n = 12; grating, n = 5). Mean maximal incisal opening was 29.91 * 10.10 mm, with an HO of 20.02 ? 8.39 mm. Deviation toward the affected side on opening was noted in 34 patients. Following induction of anesthesia, the mean MI0 was 28.20 ? 6.7 mm, HO was 17.90 + 4.0 mm, and eight patients continued to deviate with opening. Intraoperative MIO, postarthroscopy, was 40.04 ? 9.95 mm and HO was 25.43 ? 6.50 mm. Findings at the time of arthroscopy included adhesions (n = 41). fibrillations (n = 42), disc perforation (n = 3), and synovitis (n = 14). The three patients who had perforation of the disk underwent open arthrotomy and were reclassified into group 2B. Pain, noise, and motion were reevaluated at three time periods (Tables 1 to 3): 1) early (10 to 30 days); 2) intermediate (1 to 6 months); and 3) late (>6 months). At early follow-up, 62 joints were evaluated. Pain was reported as none (n = 44), mild (n = 14), moderate (n = 2), and severe (n = 2). Noise was present in 15 joints, (clicking, n = 9; popping, n = 4; grating, n = 2). Mean maximal incisal opening was 35.28 + 10.72 mm, with deviation in 2 joints. At intermediate follow-up, (n = 62 joints), pain was reported as none (n = 49), mild (n = 8), moderate (n = 3), and severe (n = 2). Noise was present in 16 joints (clicking, n = 10; popping, n = 4; grating, n = 2). Mean maximal incisal opening was 37.80 -+ 9.69 mm with deviation in 3 joints. All patients maintained a bite appliance 24 hours per day for 7 days. After 7 days, they used the appliance at night as needed. Following 6 to 8 weeks, most patients did not require a bite appliance. Late follow-up was available for 26 patients (n = 34 joints) with a mean follow-up of 12.56 + 2.87 months (range 7 to 17). Pain was reported as none (n = 24), mild (n = 5), moderate (n = 3). and severe (n = 2). Noise was present in 3 joints (clicking, n = 2; popping, n = 1). Mean maximal incisal opening Table 1.
Pain Preoperative
None Mild Moderate Severe
3162 4162 43162 12162
Early 44162 14162 2162 2162
Intermediate
Late
49162 8162 3162 2162
24134 s/34 3134 2134
1031
PERROTT ET AL
Table 2.
Noise Preoperative
None Clicking Popping Grating
12162 33162 12162 5162
Early
Intermediate
41162 9162 4/62 2162
46162 10162 4162 2162
Late 31134 2134 II34 o/34
was 37.93 c 10.52mm, with deviation in one patient. No new cases of moderate to severe pain or noise were recorded in the patients that reached the late follow-up period. Four patients had an unsuccessful result and subsequently underwent arthroplasty (5 of 62 joints; 8.1%). One patient had two complications: temporary facial nerve palsy (all branches) and a perforated tympanic membrane that required a corrective operation. Preoperative and postoperative MRI scans were done on 29 joints. The postoperative MRI was done I to 17 months after the procedure (mean, 8.38 + 5.09). In this study, one radiologist blinded to the patients’ clinical status evaluated the preoperative and postoperative MRI scans. Disc position was unchanged in the majority of joints in (25 of 29), improved or normal in 3 of 29, and worsened in 1 of 29 joints. There appeared to be no relationship between postoperative symptoms, overall clinical result, and disc position by MRI. Discussion
Temporomandibular joint arthroscopy is a new technique, therefore, follow-up data are needed to establish norms for outcome. Sanders reported a retrospective evaluation of 25 patients with a mean follow-up of 8.3 months.’ There was resolution of closed lock, good range of motion, and little preauricular pain. He summarized the results as good to excellent. However, all observations were subjective; there was no attempt to quantitate changes in pain, motion, or noise, and there was no MRI or radiographic correlation. Montgomery et al reported the clinical and radiographic results of arthroscopy in 52 joints.’ Patients were studied at 1, 2,4, 8, 24, and 52 weeks postoperatively. The authors found that pain, mandibular movement. function, and diet were significantly improved. The incidence of joint sounds did not significantly improve beyond the 4-week evaluation. Table 3.
MO/Deviation Preoperative
MI0 (mm) Deviation
Preoperative and postoperative MRIs were obtained in 21 patients, with disc position being unchanged in 78%. Indresano followed 50 patients (78 joints) for an average of 18.9 months following arthroscopy of the superior joint space. Outcome was judged to be successful on the basis of three criteria: 1) an MI0 increase of 50%, 2) a 70% decrease in pain (on a scale of 1 to lo), and 3) no need for a second operation during a 2-year follow-up period. Patients with closed lock had an 83% success rate, those with nonclosed lock had a 50% success rate. The overall success rate was 73%. There were four complications, and 13 patients (16%) later required an open joint procedure .s In the current study, we have attempted to quantify pain, noise, and motion preoperatively and to document the evolution of change in these parameters in early (1 to 30 days), intermediate (1 to 6 months), and late (>6 months) follow-up periods. We also have attempted to correlate clinical results with findings on preoperative and postoperative MRI scans. The results are similar to those previously reported.7.Y.‘3 At early, intermediate, and late follow-up, the increase in MI0 was statistically significant (P < .05). Range of motion improved and deviation decreased in all patients. At early and intermediate follow-up, there was a statistically significant reduction in the number of joints demonstrating noise. The presence or absence of noise did not correlate with pain relief or improvement in motion. The significance of TMJ noise remains obscure from a prognostic standpoint. Postoperative MRI data were difficult to obtain because third-party payers refused to cover the expense of a postoperative MRI scan. Naturally, patients without symptoms were reluctant to pay for the study. We were able to fund MRI scans through a grant to the UCSF Department of Radiology. Although significant clinical improvement with decreased pain and increased motion occurred in the majority of patients, this could not be correlated with radiographic changes in disc position in the 29 patients who had preoperative and postoperative MRI scans. These findings, which are consistent with previous studies.7,9*‘3 raise several questions: 1) Why is there a lack of correlation between clinical and MRI results?, 2) What is it in the arthros-
29.91 2 10.10 34162
Early 35.28 + 10.72 2/62
Intermediate 37.80 2 9.69 3162
Late 37.93 * 10.52 I/34
1032
TMJ ARTHROSCOPY,
copy procedure that produces clinical improvement?, and 3) What is the significance of disc position? There are at least three potential reasons for the lack of correlation between clinical and MRI results. One, the MRI scan may be inaccurate in locating disc position. This is unlikely because of well documented cadaver and human intraoperative findings. Two, abnormal disc position is not important as a cause of TMJ symptoms; patients may function on the posterior attachments or without a disc and remain asymptomatic. Three, TMJ arthroscopy releases adhesions in the superior joint space, but not in the inferior joint space. Function returns, but with the disc tightly adhered to the condyle. Return to adequate motion, however, may allow for resolution of symptoms. There were several limitations of this prospective study. It did not have a control group, which may weaken the conclusions derived from the data. However, all patients had failed to achieve resolution of their symptoms after long periods of nonsurgical treatment. They were referred to our group for an operative procedure and would not have been receptive to randomization to another course of nonsurgical treatment. Another factor that may have potentially obscured the results was the use of an intra-articular injection of steroids. The fact that patients received a steroid injection at the time of arthroscopy might explain our early results, but would not account for the long-term resolution of pain and increased motion. Conclusion
STUDY
patient can be predicted during the early to intermediate follow-up periods, that is, failure usually appeared in the first 3 months and only rarely afterward. Acknowledgment We would like to acknowledge Anna Gerrero, DDS, and Stephen Supancic, DDS. for their assistance in gathering data for this study.
References 1.Ohnishi M: Clinical application of arthroscopy in the tem2.
3.
4. 5. 6.
7.
8. 9.
10.
1I.
Results of this study indicate that TMJ arthroscopy is effective in reducing pain and increasing motion in patients with internal derangement. However, arthroscopy as done in this study appears to have little or no affect on the position of the disc. At early, intermediate, and late follow-up, pain was reduced and increases in MI0 were statistically significant. Noise did not return in the majority of patients. It appears that the long-term condition of the
PROSPECTIVE
12. 13.
14.
poromandibular joint diseases. Bull Tokyo Med Dent Univ 27: 141, 1980 Heffez L, Blaustein D: Diagnostic arthroscopy of the temporomandibular joint. Oral Surg Oral Med Oral Path 64:653, 1987 Sanders B: Arthroscopic surgery of the temporomandibular joint: Treatment of internal derangement with persistent closed lock. Oral Surg Oral Med Oral Path 62:361, 1986 Murakami K, Ito K: Arthroscopy of the temporomandibular joint third report; clinical experiences. Arthroscopy 9:49, 1984 Tarro A: Arthroscopic diagnosis and surgery of the temporomandibular joint. J Oral Maxillofac Surg 46:282, 1984 Israel H, Roser SM: Patient response to temporomandibular joint arthroscopy: Preliminary findings in 24 patients. J Oral Maxillofac Surg 47:570, 1984 Montgomery MT, Van Sachels JE, Thrash WJ, et al: Signs and symptoms, patient satisfaction and meniscal position following TMJ surgery. AAOMFS 70th Annual Meeting & Scientific Sessions, September 29 to October 3, 1988 (abstr) Indresano AT: Arthroscopic surgery of the temporomandibular joint: Report of 64 patients with long-term follow-up. J Oral Maxillofac Surg 47:439, 1989 Mosseo JJ, Sartores D, Glass R, et al: The effect of arthroscopic surgical lysis and lavage of the superior joint space on TMJ disc position and mobility. J Oral Maxillofac Surg 471676, 1989 Hilsabeck RB, Laskin DM: Arthroscopy of the temporomandibular joint of the rabbit. J Oral Surg 36:938, 1978 Williams RA, Laskin DM: Arthroscopic examination of experimentally induced pathologic conditions of the rabbit temporomandibular joint. J Oral Surg 38:65, 1980 Holmlund A, Hellsing G: Arthroscopy of the temporomandibular joint. An autopsy study. Int J Oral Surg 14:169, 1985 McCain JP, Humberto DLR, LeBlanc W: Correlation of clinical, radiographic and arthroscopic findings in the internal derangements of the TMJ. J Oral Maxillofac Surg 47:913, 1989 Helms CA, Doyle GW, Orwig D, et al: Staging of derangements of TMJ with MRI: Preliminarv observations. J Craniomand Disorders 3:93, 1989 .
