Background and Purpose. Bell palsy is an acute facial paralysis of unknown etiology. Although recovery from Bell palsy is expected without intervention, clinical experience suggests that recovery is often incomplete. This case report describes a classification system used to guide treatment and to monitor recovery of an individual with facial paralysis. Case Description. The patient was a 71-year-old woman with complete left facial paralysis secondary to Bell palsy. Signs and symptoms were assessed using a standardized measure of facial impairment (Facial Grading System [FGS]) and questions regarding functional limitations. A treatment-based category was assigned based on signs and symptoms. Rehabilitation involved muscle re-education exercises tailored to the treatment-based category. Outcomes. In 14 physical therapy sessions over 13 months, the patient had improved facial impairments (initial FGS score=17/100, final FGS score=68/100) and no reported functional limitations. Discussion. Recovery from Bell palsy can be a complicated and lengthy process. The use of a classification system may help simplify the rehabilitation process.
Bell palsy is an acute facial paralysis of unknown etiology. Bell palsy most commonly occurs between the ages of 15 and 60 years, with 15- to 44-year-olds experiencing the highest incidence.1 In 1982, Peitersen1 outlined the natural history of Bell palsy after studying 1,011 patients for 1 year following their development of facial paralysis. Thirty-one percent of the patients had incomplete paralysis, and 69% of the patients had complete paralysis of the facial muscles. Normal facial function returned in 71% of the patients, and this recovery occurred within 3 to 8 weeks after the onset of paralysis.1 Peitersen1 reported that age has a strong influence on the recovery process. Ninety percent of the patients aged 0 to 14 years recovered completely, whereas only 37% of the patients over 60 years of age recovered completely. Peitersen concluded that the sooner some facial function returned, the more favorable the overall outcome.
Individuals with Bell palsy, in our opinion, seldom receive physical therapy. Typically, the patients are told to do nothing and that facial movement will return without intervention.2–4 Patients referred for physical therapy are often treated with electrical stimulation of the facial muscles and facial movement exercises to be completed with maximal effort.4 The outcomes of such interventions were less than optimal, with the patients often developing mass action or synkinesis (abnormal movement of the face accompanying a desired motion).5 Several studies on animal models indicate that the use of electrical stimulation is disruptive to reinnervation6–8 and thus may be contraindicated for individuals with facial nerve disorders.5
Facial neuromuscular re-education is a conservative approach to facial rehabilitation. Demonstrated outcomes of facial neuromuscular re-education include improvements in impairments associated with facial paralysis.9–12 Facial neuromuscular re-education consists of an evaluation of facial impairments and functional limitations, guided training sessions of correct movement patterns, and instruction in a specific facial movement exercise program.5,12,13
From our clinical experience in treating individuals with facial nerve disorders, we found that subgroups of patients had characteristic signs and symptoms that could be recognized prior to treatment. Based on these signs and symptoms, we found that we could identify the impairment that would respond to a certain intervention. Therefore, we developed a classification scheme based on the intervention tailored to the signs and symptoms that could also be used to guide treatment (Tab. 1).13 After the treatment-based category is identified, a physical therapy program consisting of neuromuscular re-education matched to the assigned category is then initiated.
Surface electromyography (sEMG) biofeedback or a mirror may be used as an adjunct to the retraining exercises in each of the treatment-based categories. The sEMG biofeedback is not the treatment; exercises are the treatment. The facial muscles have few, if any, muscle spindles.12,14,15 Thus, little information about muscle length and action is available to the individual. Learning facial movements is difficult without the feedback. We have found that the use of sEMG or a hand mirror is a means of providing a visual or auditory representation of facial muscle activity (sEMG) or movement (mirror). Patients are also instructed in a home facial movement exercise program, which is based on the treatment-based category (Tab. 1) and the patients' performance during the rehabilitation session. The purpose of this case report is to describe the facial rehabilitation process using facial neuromuscular re-education and a treatment-based classification system in the treatment of an individual with Bell palsy.
The patient (“MC”) was a 71-year-old woman who was diagnosed with Bell palsy of the left facial nerve and complete left facial paralysis. The initial physical therapy evaluation was conducted 2 weeks following the onset of the facial paralysis. At the time of the initial evaluation, the patient had no other active medical problems. The patient reported that her facial paralysis came on suddenly and was accompanied by pain in her left ear and a funny feeling in her tongue. The paralysis was associated with no pain or sensory deficits in the left side of the face. The patient reported no hearing loss, but she reported hearing swishing sounds in her left ear. She had a magnetic resonance imaging scan of her head, and no abnormalities were found. Electrodiagnostic testing was not performed. One week after the onset of her symptoms, she started a 7-day tapered dosage of steroid therapy.
The physical therapy evaluation consisted of grading resting posture, voluntary movement, and the presence of synkinesis or abnormal movement, using the Facial Grading System (FGS) developed by Ross and colleagues.16 The FGS is an observer-based rating scale that is responsive to change.16 Ross et al indicated that the changes in scores on the resting symmetry component of the scale occur more slowly with rehabilitation than scores on the movement or synkinesis components of the scale. The scores of the FGS range from 0 (complete paralysis) to 100 (normal facial function).
The 3 sections to the FGS—resting posture (FGS rest), voluntary movement (FGS movement), and synkinesis (FGS synkinesis)—are scored individually, and the scores are combined for a total or composite score. The FGS rest section consists of rating 3 facial areas for symmetry: (1) palpebral fissure (normal , narrow , wide , or eyelid surgery ), (2) nasolabial fold (normal , absent , less pronounced , or more pronounced ), and (3) corner of the mouth (normal , drooped , or pulled up and out ).
The FGS rest section scores range from 0 to 4 and are weighted by a multiplier of 5 for a total FGS rest score of 0 to 20. The symmetry of 5 voluntary facial movements (brow raise, eye closure, snarl, smile, and pucker) are rated on a 5-point scale to determine the FGS movement score. The FGS movement scores range from 5 to 25 and are weighted by a multiplier of 4 for a total FGS movement score of 20 to 100. The degree of synkinesis associated with each of the voluntary movements is graded on a 4-point scale from 0 (no synkinesis, or no abnormal or pass movement patterns) to 3 (severe synkinesis, or disfiguring abnormal movement or gross mass movement of several muscles). The FGS synkinesis scores range from 0 to 15. For both the FGS rest and FGS synkinesis sections, a higher score relates to greater impairments. For the FGS movement section, a lower score relates to greater impairment. The FGS score is calculated as follows: FGS=FGS movement−FGS rest−FGS synkinesis. The reliability17 and construct validity16,17 for the use of the FGS have been demonstrated. Interrater reliability (r =.90) and intrarater reliability (r =.94) of the FGS scores were determined, using the type 2,1 intraclass correlation coefficient, for 2 physical therapists who scored videotapes of 15 individuals with facial nerve disorders.17 Construct validity was determined for the FGS by comparison with a quantitative measure of facial motion16,17 (Spearman rank-order correlation=.70–.87) and with the House-Brackmann facial grading system.16 Ross et al16 demonstrated that the FGS is sensitive to change by comparing prerehabilitation and postrehabilitation scores for 19 patients with facial nerve disorders.
We used the FGS to monitor progress and to describe the patient at different stages of recovery. The FGS scores were not used to determine the treatment-based category.
The patient's functional limitations were determined through an interview process consisting of a set of questions asked at each subsequent visit. The patient was asked questions regarding her eye and mouth function and how this function may have interfered with her daily activities.
During the initial evaluation, the patient had severe asymmetry in resting facial posture. The left side of her face was markedly drooped, and her left eye was much wider than her right eye (FGS rest score=15). Voluntary movement, as compared with movement of the uninvolved side, was trace to minimal. She initiated slight movement with severe asymmetry throughout all regions of the face (FGS movement score=32). As is typical in this stage of recovery when movement is minimal, the patient had no signs of synkinesis or abnormal movement patterns (FGS synkinesis score=0). Her composite FGS score on the initial evaluation was 17/100 (ie, 32−15−0=17).
MC was retired and lived alone. She reported little difficulty in eating, drinking, speaking, and closing her eye; however, she relied on compensatory techniques such as drinking from the uninvolved side of her mouth, lifting her cheek with her hand while speaking, and manually closing her eye. Because of her poor corneal protection, she had to stop her regular swimming exercise program, and she appeared motivated to improve her facial function so that she could return to swimming.
Overview of Intervention
To assist the patient in her goal of improved facial functioning, she was treated with facial neuromuscular retraining (NMR) techniques, using a hand-held mirror or sEMG biofeedback.5,11–13 Treatment planning was based on the evaluation findings and on treatment-based categories. Treatment sessions were one on one with a physical therapist for approximately 1 hour (see Tab. 2 and the “Service Delivery” section for details). A typical physical therapy session consisted of a brief re-evaluation, training with sEMG or a mirror, and instruction in an exercise program to be completed at home.
Surface EMG biofeedback was used initially to measure muscle activity associated with voluntary facial movements. Surface EMG biofeedback devices can be used to record and display small changes in muscle activity that cannot be seen in a mirror. MC found this information helpful when she started regain movement. As she was able to move more, she used the surface EMG biofeed-back less and a mirror more.
When MC developed abnormal movement patterns or synkinesis, the surface EMG biofeedback again played an important role in the physical therapy session. She developed an abnormal movement pattern such that when she snarled, her left eye would close while her right eye stayed open. MC was unaware that this was happening. Surface EMG biofeedback during exercise sessions helped to make her more aware of the abnormal movement. We placed the recording electrodes over the levator labii muscle (snarl muscle) and the inferior oculi muscle (eye closure). MC would practice snarling and raising the line on the sEMG biofeedback screen associated with the levator labii muscle activity while keeping the activity from the inferior oculi muscle to a minimum. The sEMG biofeedback would often record activity in the inferior oculi muscle prior to any visible eye closure, providing MC with the information necessary to correct her movements.
Based on the initial signs and symptoms (severe resting asymmetry, minimal voluntary movement, absent synkinesis, and impaired function), MC was considered to be in an initiation treatment category (Tab. 1). Exercises typical for the initiation category include active assisted range of motion exercises, during which the patient used the fingers of her hand to position a part of her face at a position in the range of motion for a specific movement and tried to hold the position using the targeted facial muscle while removing the passive assist. Often, patients find that it is easier to hold a position with a muscle than it is to move to the desired position (eg, having a patient passively raise the involved eyebrow with a hand, then slowly removing the passive assist as the patient tries to activate the frontalis muscle and maintain the brow raise). MC used these techniques, as part of her home exercise program, for the following facial expressions: smile, pucker, brow raise, and frown.
Because MC could not voluntarily close her eye and had signs and symptoms of corneal irritation typical of patients in the initiation category, exercises focusing on closing the eye. Squinting or raising the lower eyelid was also included in the home facial exercise program. An exercise that appears to allow the patient control over the Bell reflex18 (eye rolling backward) is helpful to achieve a more complete eye closure. The patient is instructed to focus both eyes on an object positioned 30.5 cm (12 in) down and in front of the patient and then to attempt to close both eyes. The eyes are to remain focused on this point until they are closed. Focusing the eyes downward helps to initiate the lowering of the upper eyelid. Maintaining the focused position until the eyes are closed prevents the Bell reflex, which can trick the patient into thinking that the eye is closed.
Muscle fatigue is often a concern when a patient is learning to initiate facial movements. To help avoid fatigue, MC was instructed to do 5 to 10 repetitions of the facial exercises (smile, pucker, brow raise, frown, and eye closure) 3 times a day. The number of exercises was kept to a minimum (3–5 exercises) because, in our experience, patients are more likely to adhere to a regimen consisting of a few exercises than to a regimen consisting of many exercises. MC often reported doing more exercises than were given to her because she wanted to expedite her recovery.
A re-evaluation done 6 weeks and 3 physical therapy sessions later (1 visit every other week) revealed that the patient's resting posture was unchanged, as measured by the FGS (FGS rest score=15). Her face was less drooped but still not symmetrical. Voluntary movement had increased to minimal to moderate movement. She initiated movement with mid-excursion and moderate asymmetry for all facial movements (FGS movement score=56), and there was no evidence of synkinesis (FGS synkinesis score=0). The composite or total FGS score was 41/100. MC reported less difficulty with eating and drinking than at the initiation of treatment, but she had continued difficulty protecting the cornea of her eye. She was able to close her eye completely, but only with conscious effort. She was still unable to return to swimming.
Based on the increased voluntary movement and absent synkinesis, the patient was considered to be in the facilitation category of treatment. The patient was instructed in active and resisted facial movement exercises typical for patients with some movement, no abnormal movement, and no difficulty with activities of daily living. She was instructed to do symmetrical active facial movements without allowing the voluntary movement of the uninvolved side of the face to distort the movement of the involved side of the face. Maintaining symmetry is an important part of facial movement exercises. When the uninvolved facial muscles overpower the involved facial muscles, the facial posture tends to shift to the uninvolved side. When the facial posture shifts, the involved muscles are placed at a less-than-optimal length for functioning (stretched). By maintaining symmetry and a more optimal length of the involved facial muscles during voluntary facial movements, we believe that the involved muscles have a better chance of functioning. In our opinion, small symmetrical facial movements also make it easier to detect small amounts of facial motion that may not be apparent if the resting facial posture is shifted due to overpowering of the uninvolved facial muscles.
When some active movements are difficult to perform, such as lowering the bottom lip, functional activities, such as saying specific sounds, are used for exercise. The activity of lowering the bottom lip is an important component of saying words that begin with the letter “F.” MC reported practicing a word list to be easier than doing lip movement exercises, presumably because of her greater familiarity with the word task than with isolated oral movements.
Resistive facial exercises may be appropriate if the patient has no signs of synkinesis. Manual resistance is applied in the opposite direction of the desired movement. Resistance should be applied to only isolated facial movements, without causing mass action or synkinesis. Care must be taken not to overstrengthen the uninvolved facial muscles, which would cause an even greater imbalance. An example of resistive facial exercises would be for the patient to provide resistance to the upper lip with a finger while attempting to pucker.
Facial muscle fatigue is no longer a primary concern when the patient is in the facilitation category. The patient is instructed to do a large number of repetitions (10–20) of active or resistive exercises 1 to 2 times a day. Again, the number of exercises is limited to 3 to 5 to keep the patient focused on the area needing the most work and to improve adherence to the exercise program. A typical exercise program for MC at this time would be 10 to 20 repetitions of 3 to 5 exercises to be completed 1 to 2 times a day.
Seven months after the initiation of therapy and 11 physical therapy sessions, MC's resting posture had changed from a drooping brow, lower eyelid, cheek, and mouth corner to a raised lower eyelid and a retracted cheek and mouth corner. The FGS rest score remained 15 but now represented the narrowing of her left eye as compared with her right eye, and the retraction of the left cheek and mouth corner. Voluntary movement had improved throughout the left side of the patient's face and was almost symmetrical with that of the uninvolved side (FGS movement score=84). At this point, MC had started to develop mild abnormal movement patterns or synkinesis with brow raise and snarl motions (FGS synkinesis score=2). When she would raise her eyebrows or snarl, her left eye would close slightly. Her FGS score was 67/100.
The patient's facial functioning had continued to improve. She had no problems with eating or performing oral hygiene (brushing her teeth). She reported only slight difficulty drinking from a glass without compensation techniques and only occasional problems with eye closure and protection. She was still unable to resume swimming because she could not adequately protect her cornea.
Based on the appearance of inappropriate muscle activity and the presence of abnormal movement, the patient was now considered to be in the movement control category of treatment, with the facilitation category a secondary classification. Exercises focused on controlling the abnormal or synkinetic movement, such as raising the brow while keeping the eye open and controlling the ocular synkinesis. Movement control facial exercises emphasize moving only as much as the patient can without triggering the abnormal facial movement. The range of the movement is increased as long as the abnormal movement is controlled. The patient is told to concentrate on the quality of the exercise and not the quantity of the exercises completed. It is better for a patient to do 5 repetitions of an exercise correctly than it is to do 20 repetitions incorrectly. MC was instructed to do as many repetitions of the control exercises that she could do correctly and to perform these exercises several times a day.
Because facial muscle tightness often accompanies synkinesis, it is important to teach the patient facial muscle stretching exercises. The patient was instructed in a stretching exercise that consisted of placing her right thumb inside her mouth, grasping the left cheek, and pulling the cheek down and across her face, thus applying a stretch to the cheek musculature. The stretch was held for 20 seconds. The patient was instructed to stretch her cheek 2 to 3 times, twice a day, to prevent shortening of muscle tissue. She was instructed to stretch her cheek more often if she experienced cheek muscle tightness throughout the day.
Strengthening exercises for specific movements were continued as long as they did not cause synkinesis. MC was instructed to continue with 2 to 3 of the strengthening exercises (10–20 repetitions), 1 to 2 times a day, as explained in the “Facilitation” section.
The patient's last physical therapy visit was 13 months after the initiation of therapy. She continued to demonstrate asymmetry in resting posture, which consisted of a narrow eye and a tight cheek (FGS rest score=15). Voluntary movement had improved slightly to almost complete to complete movement between the sides (FGS movement score=88), and the abnormal movement or synkinesis had increased slightly to minimal with all movements (FGS synkinesis score=5). Her FGS score was 68/100. The biggest change appeared to be in function. The patient reported no difficulties with eating, drinking, speaking, or protecting the cornea of her eye. She had even resumed swimming. Patient satisfaction was high by patient report.
Based on these signs and symptoms, we still considered the patient to be in the movement control treatment category, with relaxation the secondary treatment category. Because minimal changes were noticed in voluntary movement in the previous 7 months, strengthening was no longer, in our opinion, a reasonable goal. We instructed the patient in a final program to help maintain her facial function and to prevent any inappropriate muscle activity or synkinesis. The program consisted of isolated facial movements, stretching, facial massage, and relaxation exercises19 typical for patients in the movement control and relaxation treatment categories. Jacobsen's relaxation exercises19 and the same technique of progressively contracting and relaxing of muscles was applied to specific facial muscles. For example, MC was told to wrinkle her nose and to raise her upper lip as much as she could, holding the contraction for 3 to 5 seconds, and then to “let go,” releasing the muscle contraction completely. MC was instructed to continue with this program one time a day, gradually weaning herself from the exercise program. She was told to continue with the facial muscle stretches at least one time a day or more as she felt she needed it to prevent further facial muscle tightness.
The patient was treated over a 13-month period and seen for only 14 physical therapy sessions. Initially, the treatment sessions were more frequent (2–4 times per month) because of the need for instruction and for the patient to become familiar with the exercise process. As the patient became more aware of her facial movements, she was treated less frequently (once every 3 months). Table 2 shows the physical therapy schedule.
Impairment and Functional Limitation
The patient demonstrated improvements as facial impairments and functional limitations became less severe (Tab. 3). In our opinion, moderate improvements were made in symmetry of the face at rest, even though these improvements were not evident in the FGS rest scores. The FGS grades resting posture as being either symmetrical or asymmetrical and does not account for levels of severity. The most noticeable changes were the improvement of her voluntary movement (FGS movement), which occurred in the first 7 months of treatment, and the development of synkinesis (FGS synkinesis) in the seventh month.
The patient's functional activities improved so that after 13 months she had no difficulty eating, drinking, speaking, or protecting the cornea of her eye. She no longer had to rely on compensatory techniques to complete her activities of daily living. She had even returned to swimming between the 7th and 13th months of treatment. The patient was highly satisfied with her outcome.
In our experience, individuals with Bell palsy are seldom referred for physical therapy at the onset of the disorder. Often, they are told to wait and that this condition will get better on its own. Complete recovery does not always occur, especially in high-risk populations such as people who are elderly or who have delayed recovery.1
Physical therapists rarely continue to treat patients for 13 months. We believed, however, that this treatment duration was necessary to achieve the outcomes for this patient. For the first 7 months, the patient had facial weakness and was treated with strengthening exercises. At the 7-month visit, she had facial muscle overactivity and synkinesis. At this point, the treatment plan was adjusted to fit the changes in her facial impairments. If the physical therapy had been terminated prior to this 7-month mark, her problems of facial muscle tightness and synkinesis would not have been addressed. Instructing the patient in a maintenance program at the last physical therapy session may help to prevent an increase in facial muscle tightness and synkinesis over time. Although 13 months may seem like a long time to treat a patient, the total number of physical therapy visits was only 14 visits.
Physical therapy for patients with facial paralysis traditionally has consisted of generic facial exercises or electrical stimulation.4 Facial neuromuscular re-education techniques (ie, the use of facial exercises to address a patient's impairments and functional limitations) are different from the traditional intervention for facial paralysis. In our approach, the exercise program changes over time as the patient's impairments change with recovery. The facial neuromuscular re-education exercise program emphasizes accuracy of facial movement patterns and isolated muscle control, and it excludes exercises that promote mass contraction of muscles related to more than one facial expression. In our approach, the number of exercise repetitions and the frequency of the exercise program depend on the treatment-based categories, which are based on the patient's impairments (Tab. 1).
Continued research is needed to determine the best treatment for individuals with facial neuromuscular disorders. A first step could be to validate the treatment-based classification system based on the physical signs and symptoms of individuals with facial neuromuscular disorders. If the classification system is validated, the effectiveness of physical therapy intervention with a “tailored” treatment approach for each of the treatment categories can be determined.
- Received April 15, 1998.
- Accepted September 3, 1998.
- Physical Therapy