Abstract
The aim of this study was to evaluate the factors that may cause alterations in facial nerve function during temporomandibular joint (TMJ) surgeries. Forty-six patients were included (66 joints) between the years 2000 and 2007. Study patients were those undergoing various surgical procedures for the treatment of TMJ disorders. Patients who had made an incomplete recovery from a facial nerve injury resulting from a previous operation and patients who presented with facial palsy after a previous TMJ surgery were excluded. The facial nerve function of all patients was evaluated at different time intervals using a facial nerve grading system, motor unit action potentials of the frontalis and orbicularis oculi muscles, and a facial nerve latency test. Various degrees of facial nerve affliction were initially noticed in 71% of the study cases (47 of 66 joints). Statistical analyses ( χ 2 goodness-of-fit) revealed that several factors could lead to facial nerve injury following TMJ surgery, including the design of the skin incision, prior surgeries, type of surgery, and duration of surgery. Facial nerve injury during TMJ surgery is multifactorial. Electromyographic studies are non-invasive and valuable diagnostic and prognostic tools for assessing facial nerve function.
Temporomandibular joint (TMJ) disorders represent a significant public health problem and are generally characterized by the presence of TMJ pain, tenderness, noise, and limitations of mouth opening. TMJ disorders include infection, internal derangement, degenerative joint diseases, hypomobility/ankylosis, hypermobility, tumours, and traumatic injuries. Surgical management depends on the nature of the disorder.
Regardless of the surgical procedure used for the management of TMJ disorders, there are potential complications. These include haemorrhage, infection, auriculo-temporal nerve syndrome, otological complications, trauma to the parotid gland, and facial nerve injury. Both the temporal and zygomatic branches of the facial nerve are at particular risk of injury during TMJ surgery. The facial nerve is the main anatomical structure that the surgeon should consider when performing the surgical approach to the TMJ. Skill in preserving its functional integrity is a critical factor when this type of surgery is performed.
The incidence of facial nerve injury during TMJ surgeries varies among surgeons. There are many factors that could contribute to the injury of the temporal and zygomatic branches of the facial nerve. These nerves lie in a confluence of superficial fascia, temporalis fascia, and periosteum, and may be injured by any dissection technique that attempts to violate the integrity of these regions. Excessive or heavy-handed retraction causes compression and/or stretching of nerve fibres resulting in neuropraxia; this may be the reason for a significant number of nerve injuries associated with TMJ surgery.
Facial nerve injury may also be caused by inadvertent suture ligation of facial nerve branches, particularly during wound closure. In order to prevent this undue complication, deep blind bites with the suture needle should be avoided. The use of electrocautery in deep sites that are potentially close to facial nerve branches, or within the parotid gland, should also be avoided. Furthermore, one should avoid crushing or clamping tissue indiscriminately, particularly during episodes of brisk bleeding. Excessive swelling or haematoma formation may result in transient facial nerve injury. Also, patients with previous TMJ surgery have an increased incidence of facial nerve injury.
The aim of this study was to assess the common elements encountered during TMJ surgeries that may cause facial nerve injury. The present study was conducted on a group of patients who underwent surgical procedures to treat various TMJ disorders, in order to evaluate the different factors that may contribute to facial nerve injury during TMJ surgery.
Patients and methods
In order to explore the factors that play a role in the incidence of facial nerve injury following TMJ surgery, this study recruited patients who had been diagnosed with a TMJ disorder that mandated a surgical intervention. This prospective study included 46 consecutive patients (32 females and 14 males) who underwent surgery for the treatment of TMJ disorders between the years 2000 and 2007. All study patients were recruited from the outpatient clinics of the oral and maxillofacial surgery departments of Al-Azhar University hospitals in Cairo, Egypt, between the years 2000 and 2007. The institutional ethics committee of the host institution reviewed and approved the study. After obtaining written patient consent, appropriate surgical procedures for the treatment of TMJ disorders were performed following careful clinical and radiographic evaluation to determine the nature of the disorder. The same surgical team performed all of the operations.
The reference surgery for this study was the surgical procedure undertaken for the treatment of the TMJ following the patient’s recruitment into the study. In order to be included in the study, the reference surgery needed to be either the patient’s first surgery to treat a TMJ disorder, or if she/he had undergone prior surgery, she/he was required to have no residual facial nerve dysfunction or injury following the prior procedure(s). Patients who had made an incomplete recovery from a facial nerve injury following previous TMJ or other maxillofacial surgeries and patients who presented with facial nerve palsy were excluded. Patients with a history of previous surgical treatment of the TMJ were screened preoperatively by electromyography (EMG) for motor unit action potentials (MUAPs) of the frontalis and orbicularis oculi muscles and underwent a facial nerve latency test (FNLT) to determine whether there was complete recovery of the facial nerve function. Other parameters of this study that were examined for correlation with the incidence and degree of facial nerve injury following the reference surgery were the type of surgery, surgical approach, and duration of surgery.
The House–Brackmann facial nerve injury grading system was used to grade facial nerve recovery following the reference surgery ( Table 1 ). The system involves a six-point scale, with grade I being normal function and grade VI being total paralysis.
| Grade | Description | Characteristics | |
|---|---|---|---|
| I | Normal | Normal facial function in all areas | |
| II | Mild dysfunction | Gross | Slight weakness noticeable on close inspection |
| At rest | Normal symmetry and tone motion | ||
| Motion | Forehead: moderate to good function Eye: complete closure with minimal effort |
||
| III | Moderate dysfunction | Gross | Obvious but not disfiguring difference between two sides |
| At rest | Normal symmetry and tone motion | ||
| Motion | Forehead: slight to moderate movement Eye: complete closure with effort |
||
| IV | Moderately severe dysfunction | Gross | Obvious weakness and/or disfiguring asymmetry |
| At rest | Normal symmetry and tone motion | ||
| Motion | Forehead: none Eye: incomplete closure |
||
| V | Severe dysfunction | Gross | Only barely perceptible motion |
| At rest | Asymmetry | ||
| Motion | Forehead: none Eye: incomplete closure |
||
| VI | Total paralysis | No movement |
Electrodiagnostic studies were performed using EMG or MUAP of the frontalis and orbicularis oculi muscles and FNLT. With regard to MUAP, all patients were classified into one of three groups in accordance with Danielides et al. : (1) group A were those in whom the muscle response was 51–95% of the response of the healthy muscle; (2) group B were those in whom the muscle response was 25–50% of the healthy muscle; (3) group C were those in whom the muscle response was <24% of the healthy muscle. For FNLT, all patients was classified into one of three groups in accordance with Danielides et al. : (1) group 1 were those in whom latency varied from 2.4 to 4 ms; (2) group 2 were those in whom latency varied from 4 to 6 ms; (3) group 3 were those in whom there was an absence of action potential (no response).
Postoperative evaluation of facial nerve function was performed at 24 h, 1 week, 1 month, 3 months, and 6 months. The reference values for ‘normal function’ were established for both MUAPs and FNLTs in each patient. In the case of patients undergoing unilateral joint surgery, the other ‘non-operated’ side of the face was used for reference values. In the case of patients undergoing bilateral joint surgery, preoperative MUAPs and FNLTs were collected and served as the reference values.
All statistical analyses for this study were performed using PASW Statistics version 18.0 software (SPSS Inc., Chicago, IL, USA). Bivariate tests of association and correlation between categorical variables (i.e., FNLT and MUAP outcomes, prior surgery, surgical approach, type of surgery, and facial nerve injury) were performed using Pearson correlation χ 2 goodness-of-fit tests. The χ 2 goodness-of-fit tests were performed to determine which categorical variables correlated with facial nerve injury following the reference surgery. Since the dataset contained a variety of ‘types of surgery’, an omnibus χ 2 was not valid due to the small expected frequencies for many cells. Thus subsequent χ 2 goodness-of-fit analyses tested each particular type of surgery compared to all others in order to determine if any particular type made a difference in the likelihood of facial nerve injury, using Yates’ continuity correction for instances that still had cells with small expected frequencies. Additionally, the Bonferroni corrected P -value used for these individual surgery types vs. other surgeries was employed to correct for multiple comparisons. Thus for these individual comparisons, the critical P -value was 0.008 (i.e., 0.05/6 comparisons). Bivariate tests for differences in the duration of surgery were performed using independent samples t -tests and/or one-way analysis of variance (ANOVA). Tukey’s post hoc test was used for pair-wise comparisons when the ANOVA test was significant. In order to determine if there was a difference in surgical duration for surgeries that resulted in a facial nerve injury and those that did not, an independent samples t -test was performed.
Results
The sample originally included 62 patients who provided consent to participate in this research study and who ultimately underwent surgery for the treatment of a TMJ disorder between the years 2000 and 2007. Of these 62 patients, seven did not meet the study inclusion criteria and nine were lost to follow-up. Thus 46 patients remained in the sample; 32 were females and 14 were males, and they ranged in age from 22 to 57 years (mean age 29 years). For these 46 patients, 66 TMJs were surgically treated during the study period.
Table 2 displays the frequency distributions of the various categorical measures taken in this study. Surgical procedures consisted of TMJ reconstruction (12%, n = 8), gap arthroplasty (26%, n = 17), eminectomy (42%, n = 28), high condylar shave (9%, n = 6), meniscectomy (3%, n = 2), and disc plication and discoplasty (8%, n = 5). The surgical approaches used were either via modified endaural incision ( n = 33) or modified pre-auricular incision ( n = 33). Twelve of the patients, accounting for 20 joints (30%), had undergone prior surgical interventions, but exhibited no residual facial nerve dysfunction prior to the reference surgery used in this study. A preoperative EMG study for MUAPs of the frontalis and orbicularis oculi muscles and FNLT confirmed that there was complete recovery of the facial nerve function of this particular group of patients before the reference surgery. The duration of surgery ranged from 70 to 200 min, with a mean of 128.6 min (standard deviation 43.1 min).
| Variable | Number | % |
|---|---|---|
| Surgical approach | ||
| Modified endaural | 33 | 50.0 |
| Modified pre-auricular | 33 | 50.0 |
| Surgery type | ||
| Eminectomy | 28 | 42.4 |
| Gap arthroplasty | 17 | 25.8 |
| TMJ reconstruction | 8 | 12.1 |
| High condylar shave | 6 | 9.1 |
| Disc plication and discoplasty | 5 | 7.6 |
| Meniscectomy | 2 | 3.0 |
| Facial nerve injury | ||
| Yes | 47 | 71.2 |
| No | 19 | 28.8 |
| Prior TMJ surgery | ||
| Yes | 20 | 30.3 |
| No | 46 | 69.7 |
| FNLT at 1 week a | ||
| Group 1 | 34 | 51.5 |
| Group 2 | 32 | 48.5 |
| FNLT at 1 month a | ||
| Group 1 | 40 | 60.6 |
| Group 2 | 26 | 39.4 |
| FNLT at 3 months a | ||
| Group 1 | 53 | 80.3 |
| Group 2 | 13 | 19.7 |
| FNLT at 6 months a | ||
| Group 1 | 66 | 100.0 |
| Group 2 | 0 | 0.0 |
| MUAP at 1 week b | ||
| Group A | 28 | 42.4 |
| Group B | 38 | 57.6 |
| MUAP at 1 month b | ||
| Group A | 37 | 56.1 |
| Group B | 29 | 43.9 |
| MUAP at 3 months b | ||
| Group A | 60 | 90.9 |
| Group B | 6 | 9.1 |
| MUAP at 6 months b | ||
| Group A | 66 | 100.0 |
| Group B | 0 | 0.0 |
a Group 1 FNLT represents latency varying from 2.4 to 4 ms; group 2 FNLT represents latency varying from 4 to 6 ms.
b Group A MUAP represents a muscle response that is 51–95% of the response of the healthy muscle; group B MUAP represents a muscle response that is 25–50% of the healthy muscle.
Following the reference surgery, the incidence of initial facial nerve injury was found to be 71% ( n = 47). Table 3 displays the outcomes of the χ 2 analyses that tested each particular type of surgery compared to all others in order to determine if any particular type made a difference in the likelihood of facial nerve injury. These analyses revealed that the incidence of facial nerve injury was higher for joints that underwent a modified pre-auricular approach (87.9%) than joints that underwent a modified endaural approach (54.5%; P = 0.003). Joints with prior surgeries also had a higher incidence of facial nerve injuries (100%) than joints with no prior surgeries (58.7%; P = 0.001). Among the various types of surgery, joints that underwent gap arthroplasty, TMJ reconstruction, and meniscectomy had a higher incidence of facial nerve injury (100%) than other procedures (61.2%; P = 0.006).
| Facial nerve injury | No facial nerve injury | P -value | |||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Surgical approach | |||||
| Modified endaural | 18 | 54.5 | 15 | 45.5 | 0.003 * |
| Modified pre-auricular | 29 | 87.9 | 4 | 12.1 | |
| Type of surgery (omnibus) | |||||
| Eminectomy | 17 | 60.7 | 11 | 39.3 | NA |
| Gap arthroplasty | 17 | 100.0 | 0 | 0.0 | |
| TMJ reconstruction | 8 | 100.0 | 0 | 0.0 | |
| High condylar shave | 2 | 33.3 | 4 | 66.7 | |
| Discoplasty | 1 | 20.0 | 4 | 80.0 | |
| Meniscectomy | 2 | 100.0 | 0 | 0.0 | |
| Type of surgery | |||||
| Eminectomy | 17 | 60.7 | 11 | 39.3 | 0.106 |
| Other | 30 | 78.9 | 8 | 21.1 | |
| Type of surgery | |||||
| Gap arthroplasty | 17 | 100.0 | 0 | 0.0 | 0.006 ** |
| Other | 30 | 61.2 | 19 | 38.8 | |
| Type of surgery | |||||
| TMJ reconstruction | 8 | 100.0 | 0 | 0.0 | 0.133 |
| Other | 39 | 67.2 | 19 | 32.8 | |
| Type of surgery | |||||
| High condylar shave | 2 | 33.3 | 4 | 66.7 | 0.094 |
| Other | 45 | 75.0 | 15 | 25.0 | |
| Type of surgery | |||||
| Discoplasty | 1 | 20.0 | 4 | 80.0 | 0.034 |
| Other | 46 | 75.4 | 15 | 24.6 | |
| Type of surgery | |||||
| Meniscectomy | 2 | 100.0 | 0 | 0.0 | 0.904 |
| Other | 45 | 70.3 | 19 | 29.7 | |
| Previous surgery | |||||
| Yes | 20 | 100.0 | 0 | 0.0 | 0.001 * |
| No | 27 | 58.7 | 19 | 41.3 | |
* Statistically significant at α < 0.05.
** Statistically significant at α < 0.008 (Bonferroni corrected for multiple comparisons among ‘type of surgery’ comparisons).
While the likelihood of an injury was higher when using a modified pre-auricular approach ( Table 3 ), there was also interest in whether or not the different approaches resulted in different grades of facial nerve function at different points in time. χ 2 tests were used to determine if the House–Brackmann facial nerve grade differed between the surgical approaches at 1 week, 1 month, 3 months, and 6 months. Table 4 shows the results of these comparisons for the different time periods. The surgical approaches differed in grade of injury/function at 1 week ( P = 0.005), but not at 1 month ( P = 0.721) or 3 months ( P = 0.080), while at 6 months all joints in both groups were grade I functioning. At 1 week, joints that underwent the modified endaural approach were more likely to be back to normal functioning, while those that underwent the modified pre-auricular approach were more likely to have grade II, III, and IV injuries. There were no differences between the surgical approaches for the grade of injury at 1 month or 3 months, but there was a general trend over time for both groups to regain normal functioning.
| Time | Grade a | Modified endaural | Modified pre-auricular | P -value | ||
|---|---|---|---|---|---|---|
| n | % | n | % | |||
| 1 Week | I | 15 | 45.5 | 4 | 12.1 | 0.005 * |
| II | 1 | 3.0 | 8 | 24.2 | ||
| III | 7 | 21.2 | 6 | 18.2 | ||
| IV | 10 | 30.3 | 15 | 45.5 | ||
| 1 Month | I | 16 | 48.5 | 12 | 36.4 | 0.721 |
| II | 1 | 3.0 | 2 | 6.1 | ||
| III | 7 | 21.2 | 7 | 21.2 | ||
| IV | 9 | 27.3 | 12 | 36.4 | ||
| 3 Months | I | 24 | 72.7 | 14 | 42.4 | 0.080 |
| II | 2 | 6.1 | 7 | 21.2 | ||
| III | 5 | 15.2 | 8 | 24.2 | ||
| IV | 2 | 6.1 | 4 | 12.1 | ||
| 6 Months | I | 33 | 100.0 | 33 | 100.0 | – |
| II | 0 | 0.0 | 0 | 0.0 | ||
| III | 0 | 0.0 | 0 | 0.0 | ||
| IV | 0 | 0.0 | 0 | 0.0 | ||
| 1 Week | FNLT 1 | 19 | 57.6 | 15 | 45.5 | 0.325 |
| FNLT 2 | 14 | 42.4 | 18 | 54.5 | ||
| 1 Month | FNLT 1 | 22 | 66.7 | 18 | 54.5 | 0.314 |
| FNLT 2 | 11 | 33.3 | 15 | 45.5 | ||
| 3 Months | FNLT 1 | 28 | 84.8 | 25 | 75.8 | 0.353 |
| FNLT 2 | 5 | 15.2 | 8 | 24.2 | ||
| 6 Months | FNLT 1 | 33 | 100.0 | 33 | 100.0 | – |
| FNLT 2 | 0 | 0.0 | 0 | 0.0 | ||
| 1 Week | MUAP A | 16 | 48.5 | 12 | 36.4 | 0.319 |
| MUAP B | 17 | 51.5 | 21 | 63.6 | ||
| 1 Month | MUAP A | 21 | 63.6 | 16 | 48.5 | 0.215 |
| MUAP B | 12 | 36.4 | 17 | 51.5 | ||
| 3 Months b | MUAP A | 33 | 100.0 | 27 | 81.8 | 0.032 * |
| MUAP B | 0 | 0.0 | 6 | 18.2 | ||
| 6 Months | MUAP A | 33 | 100.0 | 33 | 100.0 | – |
| MUAP B | 0 | 0.0 | 0 | 0.0 | ||
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
