Practical management of Bell's palsy QORDON •• HUeHES, MD. Cleveland, Ohio
88..', pallY II an acute unilateral weakness or paralysis of the face resulting from peripheral facial nerve dy,functlon. While there II no readily Identifiable cause, there II some recovery of function within 6 months. Thll article off.... practical guldellnea for dlagnoalng and treating Bell', pallY. These guidelines are baaed on the author's experience with 63 patlentl over the lalt 8 yeal'l. Physical examination should reveal dlffu.. nerve Involvement, normal otoscopic findings, qnd no Ikln blebs or bllste,. and parotid ma..... Other cranial nerve palsies may be present. Dlagnoltlc te,tlng shOUld InclUde basic aUdiometry If available, and computed tomography or magnetic resonance Imaging scanning If paralysis,s present. facial photography II recommended. Topognoatlc tMtIng II not helpful beCause It no longer guides the surgical approach, and prognostic t.ts are not necessary If the face Isnot paralyzed. The only sufficiently ..naltlve teat to determine the need for possible surgery " elecfroneurography. Eye care II critical In all but the mildest ca.... and aterolds may also be helpful. Surgery II rarely needed and should conaIIt only of middle fOllQ total decomPression; Ilmple decompreulon of the tympanic and mastoid aegments la seldom helpful. The natural hlltory of the dlaease and the limited role of surgery prOVide new guidelines for ,prac· tical management of 8811" pally. (OTOlARVNGOl HEAD NECK SURG 1990:102:658.)
Bell's palsy is a unilateral weakness (palsy) or paralysis of the face resulting from acute peripheral facial nerve dysfunction. There is no readily identifiable cause for BeU's palsy, and some recovery of function occurs within 6 months. The disease undoubtedly represents a spectrum of entities with varying pathogeneses. In some cases, the disease occurs as a polyneuropathy, and in some cases there are immunologic features.' Since Sir Charles Bell's report" of peripheral facial nerve dysfunction in 1821, more than 500 articles have been written on the subject of Bell's palsy. 3 During this time, basic scientific research, surgical advances, and anecdotal experience have prompted numerous and diverse recommendations for the management of Bell's palsy; however, until 1982 clinicians did not have valid, published data on the natural history of the disease. In that year, Pettersen" presented his study of 1011 untreated patients over a period of 15 years. Peitersen's report prompted some clinicians to change their management From the Department of Otolaryngology and Communicative Disorders, The Cleveland Clinic foUndation. Presented at the Annual Meeting of the American Neurotology Society, San Francisco, Calif., March 31. 1989. Submitted for pUblication March 10. 1989; revision received June 30. 1989; accepted July 6, 1989. Reprint requests: Gordon B. Hughes, A-71,OneClinicCenter, Cleveland, OH 44195. 23/1/15083
of patients with Bell's palsy. 5 The present article offers guidelines for the practical management of Bell's palsy based on Peitersen's research and the author's experience with 63 patients over the last 8 years.
MITHODS AND MATERIALS The author retrospectively reviewed charts of all patients treated for Bell's palsy from 1980 to 1988. Demographic, diagnostic, treatment, and result data were recorded on 63 patients. Thirty-six patients were women; 27 were men. The average age was 41 years (range, 12 to 76). Thirty-two palsies were on the left side; 31 were on the right side. Forty-six patients had palsy (weakness) at the time of evaluation; 17 had paralysis. Eleven patients had additional cranial nerve palsies. Forty-two patients were seen within 21 days of onset of palsy (average, 7 days; range, 1 to 21 days); 21 patients were seen at later times. Four patients had a history of previous Bell's palsy, four had diabetes, and one was pregnant. Basic audiologic studies were obtained in all but one patient, with normal hearing results in 57 of the 62 tested patients. Ipsilateral sensorineural hearing loss, presumably related to Bell's palsy, was present in five patients. CT or MRI scans of the head were obtained in all 17 patients who had paralysis and in nine other patients; radiographic studies of the nerve were normal in all 26 patients.
Volume 102 Number 6 June 1990
Practical management of Bell's palsy 619
lobi. 1. House-Brackmann facial nerve grading s1ystem6 Function Grade
I II III IV V VI
Delcrlptlon
1%)
Estimated function 1%)
8/8
100
100
7/8
76-99 51-75 26-50 1-25 0
80 60 40 20
MeQIu,."..,..
Normal Slight Moderate Moderately severe Severe Total paralysis
5/8-6/8 3/8-418 1/8-2/8
0/8
0
*A centimeter is divided into four equal parts. On the palsied side of the face, maximal voluntary lateral movement of the corner of the mouth is measured 0-4, and elevation of the eyebrow is measured Q·4. The resultant sum is 0/8·8/8.
In the past, Schirmer's tear test was recorded; 13 of these patients had undergone this test. No other topognostic (site of lesion) testing was performed, except routine stapedial reflex testing performed by the audiologist. Presently, the author does not obtain topognostic testing. In the past, prognostic testing with minimal nerve excitability and maximal nerve stimulation was obtained in all paralyzed patients. Presently, only electroneurography (ENoG) is obtained in such patients; seven of these patients underwent electroneurography. Electrical testing is not performed if the face is not paralyzed. Steroids were administered in various doses to 28 patients. The author presently gives prednisone 1 mg/kg/day for 10 to 14 days. followed by a tapering dose when patients are seen within 3 weeks after the onset of palsy, if there are no contraindications to steroid therapy. Six patients with paralysis who had poor prognostic test results underwent decompression by the author: three underwent simple transmastoid to the cochleariform process, two underwent total transmastoid to the labyrinthine segment, and one underwent totalmiddlefossa decompression. One other patienthad Previous simple decompression performed by another physician. Results were determined using the HouseBrackmann classification? (Table 1), with grade I normal function, grade VI total paralysis, and grades II through V intermediate severity. Synkinesis (mass action) and dyskinesis (involuntary twitching) were also recorded. Adequate followup was defined as the recording of a grade I result at any final office visit. or recording of a grade II through VI result if followup was at least 9 months. Thirty-seven patients had adequate follow-up: 29 had grade I; two had grade II; 2 had grade III; and four had grade IV. None had recovery worse than grade IV. Synkinesis (n = 6) and dyskinesis (n = 4) were
presentin patients with grades II-IV function. Followup was not adequate in 26 patients. Most of these 26 patientshad mild palsy, were seen for only one officevisit, had good prognoses.and returnedto their referringphysician. Presumably many of them had excellent return of function, but this was not confirmed on chart review. Of those patients who had surgery, the patient who underwent middle fossa decompression had grade IV function, and two of the patien~ who underwentsimple decompression had grade III fl\nction at 9 months after surgery or later. The remaining patient who underwent simple decompression had grade II function at 1 months, then was lost to followup. The two patients who underwent total postauricular decompression returned to their physicians and were lost to followup. DISCUlllON
In recent years, office evaluationand laboratory testing have been streamlined to provide more practical. efficientcare without sacrificing quality.Becauseof this evolution of care and the relatively small group of patients in this study. data from this retrospectivereview cannot be used statisticallyto validatethe author's opinions; however. experience does create sound judgment on which recommendations are based. This discussion covers eight general areas of Bell's palsy management: definition and diagnosis, natural history, topognostic testing, prognostic testing. eye care. medicaltreatment. surgical treatment. and reporting of results. DetlnNlon and Diagnoais
What is Bell's palsy? What diagnostic studies are recommended during the initial office visit(s)? Bell's palsy occurs as unilateral weakness or paralysis of the face resulting from acute peripheral facial nerve dysfunction.There is no readilyidentifiable cause for Bell's palsy, and some return of function occurs within 6 months."This definitionreminds the clinician that the pathogenesis is still poorly understood and the
660 HUGHES
diagnosis is made largelyby means of exclusion. There should be no historyof trauma, local infection, or central nervous system (eNS) disease. Dysacousis, dysgeusia,andfacial painare common. Othercranialnerve palsies can be present, especially of cranial nerves V, VIII, IX, and X, but thisdoes notchangethe diagnosis. By definition the peripheral facial nerve must be involved. Physical examination excludes other lesionsthat can mimic Bell's palsy: infection, tumor, trauma, and stroke. All facial nerve branches are involved diffusely, the otoscopic examination is normal, there are no skin blebs or blisters, and there are no ipsilateral parotid masses. The clinician should document these four points in the chart. Diffuse nerve involvement, mildor severe, results from pathologic conditions within the temporal bonewherenervecompression is mostsevere. Preservation of forehead movement indicates a pathologic condition within the brain, and involvement of onlyonebranchsuggests traumaor tumor. Onotoscopic examination, the middle ear should be normal. Middle ear disease might be coincidental, but a relationship between pathologic conditions of the middle ear and facial palsy must be assumed until proved otherwise, and the middle ear must be treated appropriately immediately. Absence of skin blebs and blisters suggests that the diagnosis is not herpes zoster oticus (Ramsay Hunt syndrome), for which antiviral therapy is most helpful. Absence of pathologic conditions of the parotid suggests that extratemporal nerve involvement is not present. The authorroutinely obtains basic audiology tests on all patients when possible, and computed tomography (CI') or magnetic resonance imaging (MRI) scans of the head on paralyzed patients. Audiologic testing detects associated sensorineural hearing loss, prompts acoustic tumorworkup if indicated, anddocuments preoperative hearing in surgical candidates. Facial palsy from acoustic neuroma is uncommon, but ipsilateral sensorineural hearing loss requires radiographic study to exclude tumor, even if palsy is mild. If hearing is normal and the face is not paralyzed, radiographic studies can be postponed indefinitely: recovery from palsy confirms the diagnosis of Bell's palsy. When the face is paralyzed, the author of this report obtains CT or MRI scans. A normal resultis expected in patients with Bell's palsy, unless coincidental pathologic conditions are present. Thus, the patient is reassured that there is nostroke,braintumor, or centralnervoussystem (CNS) disease. The clinician could wait as long as 6 months for some return of facial function before ordering x-ray films; however, undiagnosed pathologic conditions other than Bell's palsy might become worse. The di-
OtoIaryngolOgYHead and Neck Surgery
agnosis of Bell's palsy is presumptive during the acute stage and requires appropriate followup until maximal recovery has been attained.
Natural HIstory In Denmark, Peitersen" studied the natural (untreated) history of Bell's palsy in 1011 patients over 15 years. Patients were checked at short intervals until remission occurred, andthesecheckswerediscontinued only when normal function was restored or after a period of 1 year. For 85% of patients, the first signs of recovery were observedwithin3 weeksafter onset;for the other 15%, recovery occurred 3 to 6 months later. All but onepatienthad somedegreeof remission within 6 months. Graham' also reported one patient with surgically confirmed Bell's palsy who recovered after 1 year. Sequelae (synkinesis, dyskinesis) developed in all patients whoserecovery beganafter3 months. Seventyone percent of the patients in Peitersen's study had completely normal recovery (grade I equivalent), 13% had near-normal recovery (grade II equivalent), 12% had moderate recovery, and 4% had poor recovery. Thus, 84% of patients had normal or near-normal recovery without steroids or surgery. These results confirm that a majority of patients do very well with eye care alone. Most patients begin recovery within 3 weeks and virtually all begin to recover within six months. Remission beginning after 3, months usually indicates grade III or worse recovery, often with aberrant regeneration. Most important, the clinician who offers medical or surgical therapy must compare treatment morbidity with disease morbidity in lightof the generally favorable natural history of Bell's palsy. Furthermore, if treatment is to help at all, it probably should be started within 3 weeks of the onset of palsy. TopognOltlc Teltlng Whatis therole of topognostic (siteoflesion) testing? If the clinician does not believe in surgery for Bell's palsy, should topognostic tests be obtained at all? Topognostic studiesmayinclude Schirmer'stear test, stapedial reflex, and evaluation of salivation and taste on the involved side. These tests are helpful only when facial nerve dysfunction is rapid in onset because slow onset does not produce reliable test results. In surgical cases, Schirmer'stest is usedto indicate whether simple decompression to the cochleariform process is adequate (tearing present) or not (tearing absent). Schirmer'stear test is accurate in only 60% of Bell's palsy patients, however." Furthermore, using intraoperative evoked electromyography, Fisch9 " O determined that only 6% of surgically treatedcases had disease distal to the genicuJate ganglion andthatmiddlefossadecompression was
Volume102 Number6 June 1990
the surgical "gold standard." Because only middle fossa decompression should be recommended, topognostic testing is not helpful, and the author no longer obtains such tests. Likewise, clinicians who do not recommend surgery have no need for topognostic tests.
Prognostic Testing Prognostic testing is used to predict the treatment outcome or natural recovery from disease. What is the current role of prognostic testing? Is electroneurography the only sufficiently sensitive test to determine the need for surgery? If the clinician does not believe in surgery for Bell's palsy, should prognostic tests be obtained at all? The single most helpful prognostic sign is presence of palsy rather than paralysis. No matter how severe palsy happens to be, if it never progresses to paralysis, the patient can be reassured of an excellent eventual recovery. In such cases, electrical testing is not needed. When the face is paralyzed, the next most helpful prognostic sign is the beginning of remission within 3 weeks." These patients also can be reassured of a favorable recovery. Prognostic testing, however, is traditionally used to determine the need for possible surgery. Commonly used tests are minimal nerve excitability, maximal nerve stimulation, and electroneurography; however, only electroneurography is currently considered suffi. tl y sensinve . . to determme . the nee d f or surgery... S 9 10 cien Nerve excitability and maximal stimulation testing are of little help in managing Bell's palsy and the author no longer performs these tests. Electroneurography is ordered during the first office visit(s) in paralyzed patients seen within 3 weeks of onset of paralysis. Results indicate poor prognosis when the compound action potential (CAP) on the involved side is at least 90% reduced compared with the normal side. Electrical testing is not obtained if paralysis is not present, if there is a history of previous Bell's palsy or other neuromuscular facial disorder (comparison of each side would not be valid), or if the patient is seen more than 3 weeks after onset, when nerve injury usually has reached its maximal degree" and the opportunity to intervene has passed. Clinicians who do not believe in surgery for Bell's palsy do not need to order prognostic tests.
Eye Care Experienced clinicians agree unanimously on one point: take care of the eye. The technique of eye care will vary from one physician to the next, but usually inclUdes eyedrops during the daytime, ointment at bedtime, and a moisture chamber at night. To avoid scratching the cornea, patches and tape should not be placed
Practical management of Bell's palsy
66t
directly on the eyelid. One convenient, inexpensive moisture chamber can be created by applying a small square of clear plastic wrap or similar material to the face with gentle hairset tape. Eyeglasses should be worn when outdoors. Ophthalmologic consultation should be obtained if the patient reports eye discomfort or if the eye looks irritated despite standard care. Meetlcal Treatment
Are steroids helpful? How much should be given for how long? StankiewiczII reviewed 92 prospective and retrospective studies that recommended the use of steroids for Bell's palsy, and virtually all studies had faulty methodology and design. Stankiewicz concluded nonetheless that steroids may prevent denervation, do prevent autonomic synkinesis (crocodile tearing), may lessen synkinesis, may prevent progression of palsy to paralysis, and may shorten recovery time. The definitive, statistically valid study of steroid benefit, however, has yet to be performed. The author believes that steroids are helpful when given in adequate doses in~ the first 3 weeks of palsy. If there are no contraindications, prednisone 1 mg/kg/day for 10 to 14 days is prescribed for all patients, regardless of degree of palsy, followed by a tapering dose. Contraindications may include drug allergy, pregnancy, diabetes, hypertension, glaucoma, peptic ulcer disease, bone disease, or recent vaccination. Pending future research, the author does not offer other medications or physical therapy. Surgloal Treatment
Is surgery helpful? Should only middle fossa total decompression be performed? Should simple decompression to the cochleariform process ever be performed? Should surgeons who cannot perform middle fossa surgery offer simple decompression? When Peitersen" reported that 84% of untreated patients had good recovery, surgeons who treated Bell's palsy" began to look at their results more critically: could their patients have done well because their surgical skills minimized additional nerve injury in patients operated on? Were some surgeons creating more injury than the disease itself? Two patients in this study, for example, had mild postoperative sensorineural hearing loss. The only study in which statistical significance supports surgery is the report by Fisch": 50% of patients operated on had better final recovery than control patients when middle fossa decompression was performed urgently after electroneurography showed at least 90% reduction of compound action potential on the involved side. A prospective multiinstitutional study of Bell's
662
OtolaryngologyHead and Neck Surgery
HUGHES
Hlatory Ezam Audlo..am Eyecare L '].\ ~ Sterolda Photo
»: ~,>\i
CT or MRJ serial ENoG Poaalble aur,ery FOUow-up photo
90% CAP reduction), or if degeneration is not severe, the patient is followed closely and retested in 1 to 2 days (ENOG
88..', pallY II an acute unilateral weakness or paralysis of the face resulting from peripheral facial nerve dy,functlon. While there II no readily Identifiable cause, there II some recovery of function within 6 months. Thll article off.... practical guldellnea for dlagnoalng and treating Bell', pallY. These guidelines are baaed on the author's experience with 63 patlentl over the lalt 8 yeal'l. Physical examination should reveal dlffu.. nerve Involvement, normal otoscopic findings, qnd no Ikln blebs or bllste,. and parotid ma..... Other cranial nerve palsies may be present. Dlagnoltlc te,tlng shOUld InclUde basic aUdiometry If available, and computed tomography or magnetic resonance Imaging scanning If paralysis,s present. facial photography II recommended. Topognoatlc tMtIng II not helpful beCause It no longer guides the surgical approach, and prognostic t.ts are not necessary If the face Isnot paralyzed. The only sufficiently ..naltlve teat to determine the need for possible surgery " elecfroneurography. Eye care II critical In all but the mildest ca.... and aterolds may also be helpful. Surgery II rarely needed and should conaIIt only of middle fOllQ total decomPression; Ilmple decompreulon of the tympanic and mastoid aegments la seldom helpful. The natural hlltory of the dlaease and the limited role of surgery prOVide new guidelines for ,prac· tical management of 8811" pally. (OTOlARVNGOl HEAD NECK SURG 1990:102:658.)
Bell's palsy is a unilateral weakness (palsy) or paralysis of the face resulting from acute peripheral facial nerve dysfunction. There is no readily identifiable cause for BeU's palsy, and some recovery of function occurs within 6 months. The disease undoubtedly represents a spectrum of entities with varying pathogeneses. In some cases, the disease occurs as a polyneuropathy, and in some cases there are immunologic features.' Since Sir Charles Bell's report" of peripheral facial nerve dysfunction in 1821, more than 500 articles have been written on the subject of Bell's palsy. 3 During this time, basic scientific research, surgical advances, and anecdotal experience have prompted numerous and diverse recommendations for the management of Bell's palsy; however, until 1982 clinicians did not have valid, published data on the natural history of the disease. In that year, Pettersen" presented his study of 1011 untreated patients over a period of 15 years. Peitersen's report prompted some clinicians to change their management From the Department of Otolaryngology and Communicative Disorders, The Cleveland Clinic foUndation. Presented at the Annual Meeting of the American Neurotology Society, San Francisco, Calif., March 31. 1989. Submitted for pUblication March 10. 1989; revision received June 30. 1989; accepted July 6, 1989. Reprint requests: Gordon B. Hughes, A-71,OneClinicCenter, Cleveland, OH 44195. 23/1/15083
of patients with Bell's palsy. 5 The present article offers guidelines for the practical management of Bell's palsy based on Peitersen's research and the author's experience with 63 patients over the last 8 years.
MITHODS AND MATERIALS The author retrospectively reviewed charts of all patients treated for Bell's palsy from 1980 to 1988. Demographic, diagnostic, treatment, and result data were recorded on 63 patients. Thirty-six patients were women; 27 were men. The average age was 41 years (range, 12 to 76). Thirty-two palsies were on the left side; 31 were on the right side. Forty-six patients had palsy (weakness) at the time of evaluation; 17 had paralysis. Eleven patients had additional cranial nerve palsies. Forty-two patients were seen within 21 days of onset of palsy (average, 7 days; range, 1 to 21 days); 21 patients were seen at later times. Four patients had a history of previous Bell's palsy, four had diabetes, and one was pregnant. Basic audiologic studies were obtained in all but one patient, with normal hearing results in 57 of the 62 tested patients. Ipsilateral sensorineural hearing loss, presumably related to Bell's palsy, was present in five patients. CT or MRI scans of the head were obtained in all 17 patients who had paralysis and in nine other patients; radiographic studies of the nerve were normal in all 26 patients.
Volume 102 Number 6 June 1990
Practical management of Bell's palsy 619
lobi. 1. House-Brackmann facial nerve grading s1ystem6 Function Grade
I II III IV V VI
Delcrlptlon
1%)
Estimated function 1%)
8/8
100
100
7/8
76-99 51-75 26-50 1-25 0
80 60 40 20
MeQIu,."..,..
Normal Slight Moderate Moderately severe Severe Total paralysis
5/8-6/8 3/8-418 1/8-2/8
0/8
0
*A centimeter is divided into four equal parts. On the palsied side of the face, maximal voluntary lateral movement of the corner of the mouth is measured 0-4, and elevation of the eyebrow is measured Q·4. The resultant sum is 0/8·8/8.
In the past, Schirmer's tear test was recorded; 13 of these patients had undergone this test. No other topognostic (site of lesion) testing was performed, except routine stapedial reflex testing performed by the audiologist. Presently, the author does not obtain topognostic testing. In the past, prognostic testing with minimal nerve excitability and maximal nerve stimulation was obtained in all paralyzed patients. Presently, only electroneurography (ENoG) is obtained in such patients; seven of these patients underwent electroneurography. Electrical testing is not performed if the face is not paralyzed. Steroids were administered in various doses to 28 patients. The author presently gives prednisone 1 mg/kg/day for 10 to 14 days. followed by a tapering dose when patients are seen within 3 weeks after the onset of palsy, if there are no contraindications to steroid therapy. Six patients with paralysis who had poor prognostic test results underwent decompression by the author: three underwent simple transmastoid to the cochleariform process, two underwent total transmastoid to the labyrinthine segment, and one underwent totalmiddlefossa decompression. One other patienthad Previous simple decompression performed by another physician. Results were determined using the HouseBrackmann classification? (Table 1), with grade I normal function, grade VI total paralysis, and grades II through V intermediate severity. Synkinesis (mass action) and dyskinesis (involuntary twitching) were also recorded. Adequate followup was defined as the recording of a grade I result at any final office visit. or recording of a grade II through VI result if followup was at least 9 months. Thirty-seven patients had adequate follow-up: 29 had grade I; two had grade II; 2 had grade III; and four had grade IV. None had recovery worse than grade IV. Synkinesis (n = 6) and dyskinesis (n = 4) were
presentin patients with grades II-IV function. Followup was not adequate in 26 patients. Most of these 26 patientshad mild palsy, were seen for only one officevisit, had good prognoses.and returnedto their referringphysician. Presumably many of them had excellent return of function, but this was not confirmed on chart review. Of those patients who had surgery, the patient who underwent middle fossa decompression had grade IV function, and two of the patien~ who underwentsimple decompression had grade III fl\nction at 9 months after surgery or later. The remaining patient who underwent simple decompression had grade II function at 1 months, then was lost to followup. The two patients who underwent total postauricular decompression returned to their physicians and were lost to followup. DISCUlllON
In recent years, office evaluationand laboratory testing have been streamlined to provide more practical. efficientcare without sacrificing quality.Becauseof this evolution of care and the relatively small group of patients in this study. data from this retrospectivereview cannot be used statisticallyto validatethe author's opinions; however. experience does create sound judgment on which recommendations are based. This discussion covers eight general areas of Bell's palsy management: definition and diagnosis, natural history, topognostic testing, prognostic testing. eye care. medicaltreatment. surgical treatment. and reporting of results. DetlnNlon and Diagnoais
What is Bell's palsy? What diagnostic studies are recommended during the initial office visit(s)? Bell's palsy occurs as unilateral weakness or paralysis of the face resulting from acute peripheral facial nerve dysfunction.There is no readilyidentifiable cause for Bell's palsy, and some return of function occurs within 6 months."This definitionreminds the clinician that the pathogenesis is still poorly understood and the
660 HUGHES
diagnosis is made largelyby means of exclusion. There should be no historyof trauma, local infection, or central nervous system (eNS) disease. Dysacousis, dysgeusia,andfacial painare common. Othercranialnerve palsies can be present, especially of cranial nerves V, VIII, IX, and X, but thisdoes notchangethe diagnosis. By definition the peripheral facial nerve must be involved. Physical examination excludes other lesionsthat can mimic Bell's palsy: infection, tumor, trauma, and stroke. All facial nerve branches are involved diffusely, the otoscopic examination is normal, there are no skin blebs or blisters, and there are no ipsilateral parotid masses. The clinician should document these four points in the chart. Diffuse nerve involvement, mildor severe, results from pathologic conditions within the temporal bonewherenervecompression is mostsevere. Preservation of forehead movement indicates a pathologic condition within the brain, and involvement of onlyonebranchsuggests traumaor tumor. Onotoscopic examination, the middle ear should be normal. Middle ear disease might be coincidental, but a relationship between pathologic conditions of the middle ear and facial palsy must be assumed until proved otherwise, and the middle ear must be treated appropriately immediately. Absence of skin blebs and blisters suggests that the diagnosis is not herpes zoster oticus (Ramsay Hunt syndrome), for which antiviral therapy is most helpful. Absence of pathologic conditions of the parotid suggests that extratemporal nerve involvement is not present. The authorroutinely obtains basic audiology tests on all patients when possible, and computed tomography (CI') or magnetic resonance imaging (MRI) scans of the head on paralyzed patients. Audiologic testing detects associated sensorineural hearing loss, prompts acoustic tumorworkup if indicated, anddocuments preoperative hearing in surgical candidates. Facial palsy from acoustic neuroma is uncommon, but ipsilateral sensorineural hearing loss requires radiographic study to exclude tumor, even if palsy is mild. If hearing is normal and the face is not paralyzed, radiographic studies can be postponed indefinitely: recovery from palsy confirms the diagnosis of Bell's palsy. When the face is paralyzed, the author of this report obtains CT or MRI scans. A normal resultis expected in patients with Bell's palsy, unless coincidental pathologic conditions are present. Thus, the patient is reassured that there is nostroke,braintumor, or centralnervoussystem (CNS) disease. The clinician could wait as long as 6 months for some return of facial function before ordering x-ray films; however, undiagnosed pathologic conditions other than Bell's palsy might become worse. The di-
OtoIaryngolOgYHead and Neck Surgery
agnosis of Bell's palsy is presumptive during the acute stage and requires appropriate followup until maximal recovery has been attained.
Natural HIstory In Denmark, Peitersen" studied the natural (untreated) history of Bell's palsy in 1011 patients over 15 years. Patients were checked at short intervals until remission occurred, andthesecheckswerediscontinued only when normal function was restored or after a period of 1 year. For 85% of patients, the first signs of recovery were observedwithin3 weeksafter onset;for the other 15%, recovery occurred 3 to 6 months later. All but onepatienthad somedegreeof remission within 6 months. Graham' also reported one patient with surgically confirmed Bell's palsy who recovered after 1 year. Sequelae (synkinesis, dyskinesis) developed in all patients whoserecovery beganafter3 months. Seventyone percent of the patients in Peitersen's study had completely normal recovery (grade I equivalent), 13% had near-normal recovery (grade II equivalent), 12% had moderate recovery, and 4% had poor recovery. Thus, 84% of patients had normal or near-normal recovery without steroids or surgery. These results confirm that a majority of patients do very well with eye care alone. Most patients begin recovery within 3 weeks and virtually all begin to recover within six months. Remission beginning after 3, months usually indicates grade III or worse recovery, often with aberrant regeneration. Most important, the clinician who offers medical or surgical therapy must compare treatment morbidity with disease morbidity in lightof the generally favorable natural history of Bell's palsy. Furthermore, if treatment is to help at all, it probably should be started within 3 weeks of the onset of palsy. TopognOltlc Teltlng Whatis therole of topognostic (siteoflesion) testing? If the clinician does not believe in surgery for Bell's palsy, should topognostic tests be obtained at all? Topognostic studiesmayinclude Schirmer'stear test, stapedial reflex, and evaluation of salivation and taste on the involved side. These tests are helpful only when facial nerve dysfunction is rapid in onset because slow onset does not produce reliable test results. In surgical cases, Schirmer'stest is usedto indicate whether simple decompression to the cochleariform process is adequate (tearing present) or not (tearing absent). Schirmer'stear test is accurate in only 60% of Bell's palsy patients, however." Furthermore, using intraoperative evoked electromyography, Fisch9 " O determined that only 6% of surgically treatedcases had disease distal to the genicuJate ganglion andthatmiddlefossadecompression was
Volume102 Number6 June 1990
the surgical "gold standard." Because only middle fossa decompression should be recommended, topognostic testing is not helpful, and the author no longer obtains such tests. Likewise, clinicians who do not recommend surgery have no need for topognostic tests.
Prognostic Testing Prognostic testing is used to predict the treatment outcome or natural recovery from disease. What is the current role of prognostic testing? Is electroneurography the only sufficiently sensitive test to determine the need for surgery? If the clinician does not believe in surgery for Bell's palsy, should prognostic tests be obtained at all? The single most helpful prognostic sign is presence of palsy rather than paralysis. No matter how severe palsy happens to be, if it never progresses to paralysis, the patient can be reassured of an excellent eventual recovery. In such cases, electrical testing is not needed. When the face is paralyzed, the next most helpful prognostic sign is the beginning of remission within 3 weeks." These patients also can be reassured of a favorable recovery. Prognostic testing, however, is traditionally used to determine the need for possible surgery. Commonly used tests are minimal nerve excitability, maximal nerve stimulation, and electroneurography; however, only electroneurography is currently considered suffi. tl y sensinve . . to determme . the nee d f or surgery... S 9 10 cien Nerve excitability and maximal stimulation testing are of little help in managing Bell's palsy and the author no longer performs these tests. Electroneurography is ordered during the first office visit(s) in paralyzed patients seen within 3 weeks of onset of paralysis. Results indicate poor prognosis when the compound action potential (CAP) on the involved side is at least 90% reduced compared with the normal side. Electrical testing is not obtained if paralysis is not present, if there is a history of previous Bell's palsy or other neuromuscular facial disorder (comparison of each side would not be valid), or if the patient is seen more than 3 weeks after onset, when nerve injury usually has reached its maximal degree" and the opportunity to intervene has passed. Clinicians who do not believe in surgery for Bell's palsy do not need to order prognostic tests.
Eye Care Experienced clinicians agree unanimously on one point: take care of the eye. The technique of eye care will vary from one physician to the next, but usually inclUdes eyedrops during the daytime, ointment at bedtime, and a moisture chamber at night. To avoid scratching the cornea, patches and tape should not be placed
Practical management of Bell's palsy
66t
directly on the eyelid. One convenient, inexpensive moisture chamber can be created by applying a small square of clear plastic wrap or similar material to the face with gentle hairset tape. Eyeglasses should be worn when outdoors. Ophthalmologic consultation should be obtained if the patient reports eye discomfort or if the eye looks irritated despite standard care. Meetlcal Treatment
Are steroids helpful? How much should be given for how long? StankiewiczII reviewed 92 prospective and retrospective studies that recommended the use of steroids for Bell's palsy, and virtually all studies had faulty methodology and design. Stankiewicz concluded nonetheless that steroids may prevent denervation, do prevent autonomic synkinesis (crocodile tearing), may lessen synkinesis, may prevent progression of palsy to paralysis, and may shorten recovery time. The definitive, statistically valid study of steroid benefit, however, has yet to be performed. The author believes that steroids are helpful when given in adequate doses in~ the first 3 weeks of palsy. If there are no contraindications, prednisone 1 mg/kg/day for 10 to 14 days is prescribed for all patients, regardless of degree of palsy, followed by a tapering dose. Contraindications may include drug allergy, pregnancy, diabetes, hypertension, glaucoma, peptic ulcer disease, bone disease, or recent vaccination. Pending future research, the author does not offer other medications or physical therapy. Surgloal Treatment
Is surgery helpful? Should only middle fossa total decompression be performed? Should simple decompression to the cochleariform process ever be performed? Should surgeons who cannot perform middle fossa surgery offer simple decompression? When Peitersen" reported that 84% of untreated patients had good recovery, surgeons who treated Bell's palsy" began to look at their results more critically: could their patients have done well because their surgical skills minimized additional nerve injury in patients operated on? Were some surgeons creating more injury than the disease itself? Two patients in this study, for example, had mild postoperative sensorineural hearing loss. The only study in which statistical significance supports surgery is the report by Fisch": 50% of patients operated on had better final recovery than control patients when middle fossa decompression was performed urgently after electroneurography showed at least 90% reduction of compound action potential on the involved side. A prospective multiinstitutional study of Bell's
662
OtolaryngologyHead and Neck Surgery
HUGHES
Hlatory Ezam Audlo..am Eyecare L '].\ ~ Sterolda Photo
»: ~,>\i
CT or MRJ serial ENoG Poaalble aur,ery FOUow-up photo
90% CAP reduction), or if degeneration is not severe, the patient is followed closely and retested in 1 to 2 days (ENOG
