Complications of Temporomandibular

Joint Arthroscopy: A Retrospective

Analytic Study of 670
Arthroscopic Procedures
Raúl González-García, MD, Francisco J. Rodríguez-Campo, MD,
Verónica Escorial-Hernández, MD,
Mario F. Muñoz-Guerra, MD, PhD, Jesús Sastre-Pérez, MD,
Luis Naval-Gías, MD, DMD, PhD, and
José L. Gil-Díez Usandizaga, MD, DMD

Purpose: Temporomandibular joint (TMJ) arthroscopy has been considered a safe surgical procedure
in the treatment of TMJ derangement. However, it is not exempt from complications. This study
evaluates the complications of arthroscopy in patients with internal derangement of TMJ.
Patients and Methods: Five hundred consecutive patients (670 joints) with TMJ derangement who
underwent arthroscopy between 1995 and 2004 were retrospectively analyzed. All the patients were
classified as II to V in the Wilkes classification. Lysis and lavage, electrocautery of the posterior
ligament, injection of corticoids, injection of ethanolamine, myotomy of lateral pterygoid muscle
attachments, myotomy and electrocautery, motor debridement, injection of sodium hyaluronate, and
meniscal suture
were performed in different patients.
Results: Complications were recognized during or immediately after the surgery. They were observed in
5 of 341 (1.26%) arthroscopies of the right TMJ and 4 of 329 (1.21%) arthroscopies of the left TMJ. A
1.34% complication rate was found in the whole series. No blood clots within the external auditory canal
were observed. Bleeding within the superior TMJ space was observed in 57 cases (8.5%), 36 of them in
the right TMJ and 21 in the left TMJ, but they were not considered as true complications. Lacerations of
the external auditory canal were found in 2 cases (0.3%), with no cases of perforation of the tympanic
membrane. Lesion of the auriculotemporal nerve was observed in a case. Paresia of the facial nerve was
found in 4 cases (0.6%). Alteration of visual accuracy of the ipsilateral eye was also observed in a patient
immediately after the surgery.
Conclusion: Special care must be taken to reduce complications within the upper joint space by means
of an adequate instrumentation and by paying attention to essential points of the arthroscopic technique.
© 2006 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 64:1587-1591, 2006

Temporomandibular joint (TMJ) arthroscopy has been traditionally considered a safe,

minimally invasive surgical procedure for the treatment of some internal derangement of the
TMJ. As TMJ arthroscopy is being popularized, more complications are being reported. Thus,
this technique is not exempt from complications. Several authors have reported complications
involving otologic, vascular, or neurologic injury, infection, cardiovascular alterations, and
perforation of the middle cranial fossa.1-3 Most of them took place during or immediately after
the surgical procedure. Although many cases have been reported concerning TMJ arthroscopic
complications, few large series are available.4-6 The aim of this retrospective analytic study is
to evaluate immediate or early complications of TMJ arthroscopy performed over 500
consecutive patients (670 joints) with internal derangement (stage II to V according to Wilkes7)
during 9 years.

Received from the Department of Oral and Maxillofacial–Head and

Neck Surgery, University Hospital La Princesa, Madrid, Spain.
Address correspondence and reprint requests to Dr González-
García: Department of Oral and Maxillofacial–Head and Neck Surgery,
University Hospital La Princesa, Calle Los Yébenes 35 8C,
Madrid, Spain; e-mail: raugg@mixmail.com
© 2006 American Association of Oral and Maxillofacial Surgeons
0278-2391/06/6411-0004$32.00/0 doi:10.1016/j.joms.2005.12.058 1587
1588 COMPLICATIONS OF TMJ ARTHROSCOPY

Patients and Methods

The study series consisted of 500 consecutive patients (49 male, 451 female) undergoing
arthroscopic procedures in the Department of Oral and Maxillofacial Surgery, University
Hospital La Princesa (Madrid, Spain), between 1995 and 2004. These patients had undergone
previous nonsurgical treatment, which was unsatisfactory. All patients were staged as II to V
according to Wilkes. Six hundred seventy arthroscopic procedures were performed in this
period of time (right joints, 341; left joints, 329).

All patients underwent nasoendotracheal intubation and surgery was performed under general
anesthesia. After entering the upper joint space with a 23-gauge needle, distension was
performed with salinesolution. This maneuver favored the introduction of a cannula within the
upper compartment by means of a puncture with sharp and blunt-tipped trocars. Continuous
lavage with lactated Ringer’s solution was maintained by means of an irrigation line. A 2.2-mm
Dyonics 30° arthroscope (Smith and Nephew, Melbourne, Australia) was used in all the cases.

Different arthroscopic procedures were performed in relation to intraoperative arthroscopic

findings and according with Wilkes staging. Lysis and lavage, electrocautery of the posterior
ligament, injection of corticoids, injection of ethanolamine, myotomy of the lateral pterygoid
muscle attachments, myotomy and electrocautery, motor debridement, injection of sodium
hyaluronate, and meniscal suture were performed in different proportions. All patients received
an intraoperative dose of amoxicillin and clavulanic acid 1 g intravenously. After this,
amoxicillin with clavulanic acid 1 g was administered each 8 hours. Moreover, a dose of
dexamethasone 4 mg was administered during and after the surgery. Patients began
physiotherapy 24 hours after the surgery. A visual analog scale (range, 0 to 10) was used for the
evaluation of pain before and after surgery. Interincisal measurement of mandibular function
and movements of laterality and protrusion were also used. Surgical procedure and evaluation of
the patients were performed by the same unique surgeon in all the cases. As part of a larger
study, patients were assessed at 1, 2, 3, 6, 9, 12, and 24 months after the surgery. All the patients
were followed up for at least 24 months. Complications during surgery and immediate
complications were considered for the analysis.

Results

Several possible complications were considered in our study, such as: 1) bleeding, 2) instrument
breakage, 3) laceration of the external auditory canal, 4) blood clots in the external auditory
canal, 5) lesion of the auriculotemporal nerve, 6) paresia of the facial nerve, 7) paralysis of the
facial nerve, 8) alteration of visual accuracy, 9) lesion of the inferior alveolar nerve, 10) cardiac
disturbances, and 11) arteriovenous fistula.

Five of 341 (1.46%) arthroscopies of the right TMJ and 4 of 329 (1.21%) arthroscopies of the
left TMJ underwent complications. A 1.34% complication rate was found for the whole series.
All these complications were recognized during or immediately after the surgery. Because
packing of the external auditory canal was performed before the intervention, no blood clots
were observed in any of the arthroscopies. Bleeding within the superior TMJ space was
observed in 57 (8.5%) arthroscopies, 36 of them in the right TMJ and 21 in the left. Bleeding
was not severe in any of the cases, and it was not considered as a real complication because no
disturbance was generated to the patient. Moreover, all cases that presented with bleeding
within the upper joint were treated by means of electrocoagulation or insufflation of a Fogarty
catheter. Arthroscopy was not finished until the bleeding was stopped in all the cases.
Lacerations of the external auditory canal were found in 2 (0.3%) cases, with no cases of
perforation of the tympanic membrane. These cases were recognized because of a sudden leak
of irrigation fluid from the external auditory canal. Patients were treated in the Department of
Otorhinolaryngology by means of irrigation of the external auditory canal and were prescribed
antibiotics and corticoid suspension ear drops. Lacerations healed uneventfully after a few
weeks.

A lesion of the auriculotemporal nerve was observed in 1 case. The patient showed paresthesia
in the maxillary region for several weeks, but recovery took place without sequelae. No other
lesions of the fifth nerve were found. Paresia of the facial nerve was found in 4 cases (0.6%),
whereas total paralysis was not observed in any of the cases. Similarly, neither lesions of the
inferior alveolar nerve nor cardiac disturbances because of trigeminocardiac reflex were
observed. Arteriovenous fistula was not found either. No instrument breakage, hearing loss,
infection, or vertigo was registered. Alteration of visual accuracy of the ipsilateral eye was also
observed in 1 patient immediately after surgery. The patient recovered uneventfully after 6
weeks. A summary of all the complications from our series is shown in Table 1. Bleeding was
the most frequent alteration and presented during different arthroscopic procedures. Patients
who experienced bleeding are shown in Table 2.

Discussion

Otologic complications resulting from TMJ arthroscopy have been reported by several
authors.1-3 These complications are caused by the close proximity of the TMJ to the external
auditory canal, tympanic membrane, and middle ear. This is the reason why blood clots and
lacerations may be found in the external auditory canal after the procedure. Moreover, serious
tympanic membrane and middle ear injuries may be found resulting from TMJ arthroscopy.
However, in concordance with the reported series,4-6,8 otologic complications rarely appear.

The analysis of our series showing an otologic complication rate of 0.3% referred to the whole
series, in accordance with the 0% to 1% reported in the literature. A recent report by Tsuyama et
al9 reported 24 (8%) of 301 cases with otologic complications derived from TMJ arthroscopy.
By far, blood clots and lacerations of the external auditory canal were the most frequent in their
series (5.3%). The first may be explained by the absence of a cotton pellet packing the ear
before the surgery. Because we have had no blood clots, this procedure constitutes a reliable
method to avoid the occupation of the external auditory canal. In relation to lacerations of the
canal, Tsuyama et al9 reported a 2.3% complication rate, in contrast to the 0.3% of our series.
The introduction of the insufflation needle and the arthroscope should be directed forward,
rather than at a 90° angle. As reported by McCain,10 the insertion of the operative cannula
should be performed under direct arthroscopic visualization. However, because anatomic
variations may be found in some cases, damage to the external auditory canal is possible. Both
of our cases with laceration of the external auditory canal required otologic treatment and
recovered uneventfully.

Although we found neither hearing loss nor vertigo or ear fullness, this complication has been
reported by several authors.9,11 It has been suggested that these complications take a similar
pathway, via the ligaments within Huguire’s canal or the foramen of Huschke.12,13

In relation to neurovascular injuries, no direct injury of the superficial temporal vessels or the
facial nerve have been reported when a postero-lateral approach is used.14 Our data do not
agree with these authors because we observed damage to the superficial temporal vessels
despite the use of this approach. Moreover, although we used this approach with the canthus-
tragus line as a guide, 1 patient (0.15%) with a lesion of the auriculotemporal nerve and 4
patients (0.6%) with paresia of the seventh nerve were observed. However, no cases of paralysis
of the facial nerve were found. The auriculotemporal nerve was damaged in a medial position
because of an excessive introduction of the cannula medial to the upper joint. Thus, according to
these data, arthroscopy is a relatively safe, but not exempt of neurovascular complications,
procedure.
The reported incidence of neurologic complications has been as high as 3.9%.4,6 In our series,
0.75% of the cases showed neurologic lesion after the arthroscopy, which is lower than that
reported in the literature. All cases recovered in a few days. In the routine practice, a careful
checking for continuous flow of the irrigation fluid may reduce the extravasation rate, and
therefore the risk of neurapraxia and paralysis. In fact, the majority of cases with neurapraxia
and paralysis are caused by the compression of the nerves by serum extravasation. If an
important extravasation occurs, we should explore the pharynx before removal of the intubation
tube. This maneuver will show a possible parapharyngeal edema that could compromise the
airway. We have found this situation in some cases, but no special treatment was necessary.
After a few minutes, parapharyngeal edema disappeared and removal of the intubation tube was
possible in all the cases. In summary, by using the adequate puncture techniques, these
complications may be reduced in frequency, but not completely avoided.

Cardiac depression has been referred to in the literature as the result of the stimulation of the
trigeminal nerve.15,16 In fact, subsequent cardiac dysfunction is caused by the stimulation of
the vagus nerve via the central nucleus of the vagus nerve, which receives the afferent fibers of
the fifth nerve. This condition is similar to the more frequently reported oculocardiac
reflex.17,18 Although extremely rare, this complication may compromise vital parameters. A
total interruption of the arthroscopic maneuvers and the rapid intervention of the
anesthesiologist are mandatory to re-establish patient parameters.

Bleeding was observed in 8.5% of our cases, which was higher than that reported by other
authors.8,9 By far, it was the most frequent alteration in our series. Although bleeding was not
severe in any case, some problems with visibility within the upper joint space were observed
during the surgery. Irrigation at a higher flow was necessary to maintain an adequate visibility,
and bleeding stopped after electrocoagulation or by means of the insufflation of a Fogarty
catheter. In our series, a higher proportion of cases were from patients classified as stage II or
III according to Wilkes. However, bleeding was minimal in all the cases and it was not
considered a true complication. Within the vascular complications, arteriovenous fistula has also
been referred to as a possible complication of the TMJ arthroscopy, but it usually appears after a
long follow-up period.19

One of our cases showed an alteration of visual accuracy. No clear explanation has been
obtained to clarify this aspect because no direct pathway has been described to explain this
complication. Moreover, we have not found works referring to ocular complications derived
from arthroscopic procedures. The patient experienced this alteration for 6 weeks; after this
he/she experienced complete recovery.

Infections of the TMJ have been also reported ranging from 0% to 1%.4-6,8 No postoperative
infections were observed in our study. This is in accordance with results from Tsuyama et al.9
In fact, this complication seems extremely uncommon, although joint infection, otitis media,
and infratemporal joint infection have been reported.20,21

Arthroscopy can be used to reduce pain and dysfunction almost immediately. The vast majority
of complications appear during or immediately after the arthroscopy and most of them recover
uneventfully. Special care must be taken to reduce bleeding within the upper joint space by
means of adequate instrumentation and by paying attention to essential points of the surgical
technique. Because of the low rate of complications from 670 TMJ arthroscopies, we strongly
recommend its use in cases in which nonsurgical treatment has failed to solve a patient’s
disabilities.We have increased our skills in arthroscopic surgery for the last 9 years. Although
we now use more complex arthroscopic procedures, a parallel increase in the complication rate
has not been observed. This condition may be explained by the augmentation of the surgeon’s
experience. Further studies are necessary to analyze whether or not early complications may
influence the general outcome of TMJ derangement after a long postoperative follow-up
period.
References
1. Van Sieckels JE, Nishioka GJ, Hegewald MD, et al: Middle ear
injuring resulting from temporomandibular joint arthroscopy.
J Oral Maxillofac Surg 45:962, 1987
2. McCain JP: Complication of TMJ arthroscopy. J Oral Maxillofac
Surg 46:256, 1988
3. Applebaum EL, Kumar A: Otologic complication following temporomandibular
joint arthroscopy. Ann Otol Rhinol Laryngol
97:665, 1988
4. Carls FR, Engelke W, Locker MC, et al: Complications following
arthroscopy of the temporomandibular joint: Analysis covering
a 10-year period (451 arthroscopies). J Craniomaxillofac Surg
24:12, 1996
5. White RD: Retrospective analysis of 100 consecutive surgical
arthroscopies of the temporomandibular joint. J Oral Maxillofac
Surg 47:1014, 1989
6. McCain JP, Sanders B, Koslin MG, et al: Temporomandibular
joint arthroscopy. A 6-year multicentric retrospective study of
4,831 joints. J Oral Maxillofac Surg 50:926, 1992
7. Wilkes CG: Arthrography of the temporomandibular joint in
patients with the TMJ pain-dysfunction syndrome. Minn Med
61:645, 1978
8. Indresano AT: Arthroscopic surgery of the temporomandibular
joint: Report of 64 patients with long-term follow-up. J Oral
Maxillofac Surg 47:439, 1989
9. Tsuyama M, Kondoh T, Seto K, et al: Complications of temporomandibular
joint arthroscopy: A retrospective analysis of
301 lysis and lavage procedures performed using the triangulation
technique. J Oral Maxillofac Surg 58:500, 2000
10. McCain JP: Principles and Practice of Temporomandibular
Joint Arthroscopy. St Louis, MO, Mosby, 1996, p 128
11. Sugisaki M, Ikai A, Tanabe H: Dangerous angles and depth for
cranial fossa injury during arthroscopy of the temporomandibular
joint. J Oral Maxillofac Surg 53:803, 1995
12. Loughner BA, Larkin LH, Mahan PE: Discomalleolar and anterior
malleolar ligaments: Possible causes of middle ear damage
during temporomandibular joint surgery. Oral Surg Oral Med
Oral Pathol Endod 68:14, 1989
13. Herzog S, Fiese R: Persistent foramen of Huscke: Possible risk
factor for otologic complication after arthroscopy of the temporomandibular
joint. Oral Surg Oral Med Oral Pathol Endod
68:267, 1989
14. Westesson PL, Erickson L, Liedberg L: The risk of damage to
facial nerve, superficial temporal vessels, discs, and articular
surfaces during arthroscopic examination of the temporomandibular
joint. Oral Surg 62:124, 1986
15. Roberts RS, Best JA, Shapiro RD: Trigeminocardiac reflex during
temporomandibular joint arthroscopy: Report of a case.
J Oral Maxillofac Surg 57:854, 1999
16. Momota Y, Kotani J, Ueda Y, et al: Cardiac asistolia during
arthroscopic surgery of the temporomandibular joint: A case
report. J Oral Maxillofac Surg 57:188, 1999
17. Mikakhur RK, Sheperd WF, Jones CJ: Ventilation and the oculocardiac
reflex. Anesthesia 41:825, 1986
18. Gold RS, Pollard Z, Buchwald IP: Asystole due to the oculocardiac
reflex during strabismus surgery: A report of two cases.
Ann Ophtalmol 20:473, 1988
19. Martin-Granizo R, Caniego JL, de Pedro M, et al: Arteriovenous
fistula after temporomandibular joint arthroscopy successfully
treated with embolization. Int J Oral Maxillofac Surg 33:301,
2004
20. Tarro AW: Arthroscopic treatment of anterior disc displacement:
A preliminary report. J Oral Maxillofac Surg 47:353, 1989
21. McCain JP, Zabiegalski NA, Levine RL: Joint infection as a
complication of temporomandibular joint arthroscopy: A case
report. J Oral Maxillofac Surg 51:1389, 1993