Research

JAMA Otolaryngology–Head & Neck Surgery | Original Investigation

Evaluation of Factors Associated With Favorable Outcomes

in Adults With Bell Palsy
Myung Chul Yoo, MD; Yunsoo Soh, MD, PhD; Jinmann Chon, MD, PhD; Jong Ha Lee, MD, PhD; Junyang Jung, MD; Sung Su Kim, MD, PhD;
Myung-Won You, MD, PhD; Jae Yong Byun, MD, PhD; Sang Hoon Kim, MD; Seung Geun Yeo, MD, PhD

IMPORTANCE Identification of the factors associated with improved facial nerve function
after treatment of Bell palsy is important to provide patients with early and effective
treatment.

OBJECTIVE To identify factors that are associated with improved treatment outcomes in
patients with Bell palsy.

DESIGN, SETTING, AND PARTICIPANTS This retrospective cohort study included 1364 patients
with Bell palsy treated at the outpatient clinic of the Department of Otolaryngology at the
Kyung Hee University Hospital, Seoul, Republic of Korea, between January 1, 2005, and
December 31, 2017. The medical records of patients admitted to this hospital for management
of acute facial palsy were reviewed by 3 otolaryngologists with more than 20 years’
experience in treating facial palsy.

MAIN OUTCOMES AND MEASURES Facial function at the initial and final visits were measured
using the House-Brackmann (H-B) grading system, which is one of several analysis tools
developed to quantify facial function and provide reproducible information. It is a widely
accepted system for grading facial function in 6 steps, from normal (H-B grade I) to total
paralysis (H-B grade VI).

RESULTS In total, 1364 patients with primary Bell palsy (718 [52.6%] women) and a mean (SD)
age of 47.7 (16.7) years were enrolled. The overall rate of favorable outcome, which was
defined as an H-B grade of I or II at the 6-month follow-up visit, was 80.6% (1099 of 1364
patients). Of 1099 patients who had a favorable outcome at 6 months, 343 (31.2%) were
younger than 40 years. Of 1364 patients, 1053 (77.2%) had moderate facial dysfunction
(H-B grade III or IV). No pathological spontaneous fibrillation activity (ie, good
electromyography [EMG] results) was detected on EMG in 937 of 1364 patients (68.7%),
492 (36.1%) had controlled hypertension, and 673 (49.3%) were treated with oral
corticosteroids alone. Multivariable analysis revealed that the following factors were
associated with favorable outcome: age younger than 40 years (odds ratio [OR], 1.56;
95% CI, 1.09-2.22), an initial H-B grade of III or IV (OR, 2.62; 95% CI, 1.93-3.57), good EMG
results after 2 weeks of treatment (OR, 3.38; 95% CI, 2.48-4.61), absence of diabetes
(OR, 1.43; 95% CI, 1.04-2.36), and control of hypertension (OR, 1.64; 95% CI, 1.16-2.33).

CONCLUSIONS AND RELEVANCE Multiple logistic regression analysis in this study suggests that
multiple clinical factors are associated with favorable outcomes in patients with Bell palsy.

Author Affiliations: Author

affiliations are listed at the end of this
article.
Corresponding Author: Seung Geun
Yeo, MD, PhD, School of Medicine,
Department of Otorhinolaryngology–
Head and Neck Surgery, Kyung Hee
University, 23 Kyung Hee Dae-ro,
Dongdaemun-gu, Seoul 02447,
JAMA Otolaryngol Head Neck Surg. doi:10.1001/jamaoto.2019.4312 Republic of Korea (yeo2park@
Published online January 23, 2020. gmail.com).

(Reprinted) E1
© 2020 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020

 Research Original Investigation
B
ell palsy is an idiopathic, acute palsy of the peripheral
facial nerve supplying all muscles of the face.1 It is a rap-
idly developing unilateral facial weakness with an un-
known cause. The condition results in a partial or complete in-
ability to spontaneously move the facial muscles on the affected
side.2 Bell palsy is clinically easy to diagnose, does not affect
a patient’s mortality and life expectancy, and usually has a good
prognosis. However, patients experience mental stress until
the condition resolves, which may take some time if sequelae
develop. The exact pathogenic mechanism is not well known,
and treatment methods for complete resolution of symp-
toms have not yet been established. Recent studies have dem-
onstrated that the major cause of Bell palsy is the reactiva-
tion of the latent herpes simplex virus type 1 or varicella zoster
virus within the geniculate ganglia.3 The pathogenesis by which
these viruses can cause neuropathy may begin with a cyto-
toxic edema induced by neuronal inflammation.4 These hy-
potheses have justified the use of corticosteroids and antivi-
ral agents, separately or together, in the treatment of Bell palsy.
The incidence of Bell palsy is 20 to 30 cases per 100 000
people per year, accounting for 60% to 75% of all unilateral fa-
cial paralyses.3,5 Men and women are equally affected and the
disease can occur at any age.6 The facial weakness usually oc-
curs within a maximum of 48 hours; it can be complete or par-
tial and is generally unilateral.7 Although most patients with
Bell palsy completely recover, up to 30% of the patients de-
velop long-term sequelae, such as a permanent facial paresis,
synkinesis, contracture, and facial asymmetry, even with ap-
propriate treatment.8,9 Therefore, the resolution of Bell palsy
and risk of paralysis are of great concern to patients.
Previous studies10-12 reported on several factors associ-
ated with prognosis based on a clinical evaluation of accom-
panying symptoms, underlying medical diseases (hyperten-
sion and diabetes), age, and the degree of degeneration of the
facial nerve as determined using an electrophysiological test.
Moreover, numerous studies13,14 have been conducted on the
epidemiology, cause, diagnosis, treatment, and prognosis of
Bell palsy with the goal of providing early and appropriate treat-
ment to achieve optimal outcomes. Current knowledge on the
prognosis of Bell palsy remains limited owing to partial re-
search of clinical factors, small study samples, and high pos-
sibility of bias in multicenter studies. In addition, no clear
guidelines have been established for treating Bell palsy with
the combination of corticosteroids and antiviral agents. Based
on this background, we analyzed variables potentially associ-
ated with the outcome of Bell palsy. This study aimed to evalu-
ate the outcomes of patients with Bell palsy and diagnosti-
cally classify specific factors that contribute to a favorable or
unfavorable outcomes.
Methods
Study Design and Participants
This retrospective cohort study included patients who were
diagnosed with acute facial palsy and were admitted to the De-
partment of Otorhinolaryngology—Head and Neck Surgery at
the Kyung Hee University Hospital, Seoul, Republic of Korea,
E2 JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020 (Reprinted)
© 2020 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020
Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy
Key Points
Question Which factors are associated with improved outcomes
in patients with Bell palsy?
Findings In this cohort study of 1364 patients diagnosed with Bell
palsy between 2005 and 2017, factors that were associated with
improved facial function included younger age, a lower degree of
initial facial nerve paralysis as measured using the
House-Brackmann grade, good electromyography result (absence
of pathological spontaneous fibrillation activity), absence of
diabetes, and control of hypertension.
Meaning The findings of this study suggest that multiple clinical
factors may be associated with a favorable outcome in patients
with Bell palsy.
from January 1, 2005, to December 31, 2017. The medical rec-
ords of 1756 patients admitted to the hospital were reviewed
by 3 of us (J.Y.B., S.H.K., and S.G.Y.) who have more than 20
years’ experience in treating facial palsy. Of 1756 patients, 392
patients were excluded based on the following exclusion cri-
teria: central nervous system disorders, polyneuropathy such
as Guillain-Barré syndrome, recurrent Bell palsy, iatrogenic fa-
cial nerve palsy, neoplasms, Ramsay Hunt syndrome, otitis me-
dia, pregnancy, herpes zoster oticus, bilateral facial palsy, un-
clear timing of onset, follow-up for less than 6 months, and
House-Brackmann (H-B) grade II at admission. A total of 1364
patients were included in this study (Figure). This retrospec-
tive study was approved by the institutional review board of
Kyung Hee University Hospital, Seoul, Republic of Korea. The
requirement for written informed consent was waived by the
Kyung Hee University Hospital owing to the retrospective na-
ture of the study.
Patients’ baseline characteristics were assessed before ini-
tiating treatment, including the findings of the otorhinolar-
yngological examination, grading of facial function, age, sex,
and previous history of facial palsy. All patients were admit-
ted to the hospital for at least 7 days. In addition, all patients
were prescribed an ophthalmic ointment to prevent eye dam-
age according to the American Academy of Otolaryngology
guideline.2 Except 61 patients who refused a corticosteroid
treatment and underwent a conservative treatment instead,
such as acupuncture or physical therapy, all other patients be-
gan treatment with an oral corticosteroid or a corticosteroid
combined with an antiviral drug within 7 days from the onset
of paralysis. The corticosteroid treatment consisted of oral
prednisolone for 10 days, at a dosage of 1 mg/kg per day (maxi-
mum, 80 mg/d) for the first 4 days followed by tapering to
60 mg/d on days 5 and 6, 40 mg/d on days 7 and 8, and 20 mg/d
on days 9 and 10. The antiviral treatment consisted of oral acy-
clovir, 1000-2400 mg/d, for 5 days or famciclovir, 750 mg/d,
for 7 days. Physical therapy consisted of facial massage,
facial expression practice, and treatment with an electrical
stimulator.
Assessment of Outcome and Follow-up
The facial function was assessed using the H-B grading
system.15 The following factors were recorded: patient’s sex
jamaotolaryngology.com
 Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy
Figure. Flow Diagram of the Study Design
1756 Patients with Bell palsy
1364 Patients with Bell palsy
Age Sex Initial House-Brackmann
grade
Evaluation of prognostic factors of Bell palsy: favorable vs unfavorable recovery
and age, the degree of initial facial nerve paralysis repre-
sented by the initial H-B grade, the presence of underlying dis-
eases, such as diabetes and hypertension, electrophysiologi-
cal test results, and the degree of the final facial function. Three
otolaryngologists (J.Y.B., S.H.K., and S.G.Y.) assessed the ini-
tial severity of Bell palsy using the H-B grading system. The
degree of the initial facial palsy was classified as mild (H-B grade
II), moderate (H-B grades III-IV), or severe (H-B grade ≥V).
The degree of functional improvement was also assessed based
on the H-B grade at every follow-up visit (14 days, 1 month,
3 months, and 6 months). At the 6-month follow-up visit, we
reevaluated the H-B grade in all patients and defined favor-
able outcome as an H-B grade of I or II and an H-B grade III or
higher as an unfavorable outcome.
Electroneurography (ENoG) and electromyography (EMG)
were used for electrophysiological assessment. Electroneu-
rography was performed on day 4 or 5 after the symptom on-
set, whereas needle EMG was conducted in all patients 2 weeks
after the onset of facial palsy. Electroneurography measure-
ments were reported as the percentage maximal amplitude on
the affected side divided by the maximal amplitude on the nor-
mal side. A poor ENoG result was defined as a loss of ampli-
tude greater than 90%, whereas loss of 90% or less was clas-
sified as a good outcome.16 In addition, the presence or absence
of the blink reflex and the needle EMG findings were ana-
lyzed simultaneously to be classified as a poor or good out-
come by the physical medicine and rehabilitation physician.
The absence of pathological spontaneous fibrillation activity
was defined as a good outcome, whereas the presence of ab-
normal spontaneous activity or absence of volitional activity
was classified as a poor outcome.
Statistical Analysis
Clinical data are presented as mean (SD), with effect size
presented as the difference in rates of favorable outcome and
95% CI around the difference. To evaluate the association be-
tween baseline variables, treatment, and outcome of Bell palsy,
multivariable logistic regression analyses were performed and
associations were reported as odds ratios (ORs) with 95% CIs.
jamaotolaryngology.com (Reprinted) JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020
© 2020 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020
Original Investigation Research
392 Patients excluded
(did not meet inclusion criteria)
Electrophysiological Underlying Treatment method
test disease 673 Oral steroids only
668 Combination antiviral
23 Supportive care
Results
After excluding 392 patients who did not meet the inclusion
criteria, 1364 patients with primary Bell palsy were enrolled
(646 men [47.4%] and 718 women [52.6%]) (Table 1). The
mean (SD) patient age was 47.7 (16.7) years. Of the 1364
patients registered, 1053 (77.2%) had an initial H-B grade of
III or IV, 1279 (94.0%) had ENoG of 10% or greater, 937
(68.7%) had good EMG results, 179 (13.1%) had diabetes, 492
(36.1%) had controlled hypertension, 673 (49.3%) were
treated with corticosteroids alone, 668 (49.0%) were treated
with a combination of corticosteroid and an antiviral agent,
and 23 (1.7%) were treated with supportive care alone. The
overall rate of favorable outcome was 80.6% (1099 of 1364
patients). Among 1099 patients who had a favorable out-
come, 343 (31.2%) were younger than 40 years. In addition,
the initial H-B grade was III or IV in 885 of 1099 patients
(80.5%), 412 (37.5%) had controlled hypertension, and 571
(52.0%) were treated with corticosteroids alone.
Multivariable logistic regression analysis was performed
to identify independent factors associated with the final out-
come (Table 2). Favorable outcome (ie, H-B grade ≥III) was
associated with age younger than 40 years (OR, 1.56; 95% CI,
1.09-2.22), lower initial H-B grade (OR, 2.62; 95% CI, 1.93-
3.57), good EMG outcome after 2 weeks of treatment (OR,
3.38; 95% CI, 2.48-4.61), absence of diabetes (OR, 1.43;
95% CI, 1.04-2.36), and the control of hypertension (OR,
1.64; 95% CI, 1.16-2.33). Moreover, after controlling for
important demographic characteristics and baseline clinical
factors (age, sex, initial H-B grade, underlying disease
such as hypertension and diabetes, and electrophysio-
logical test findings), treatment with an oral corticosteroid
alone did not show any statistically significant independent
difference in the OR compared with combination antiviral
therapy. Results of the multivariable logistic regression
analysis for predicting the probability of favorable outcome
(H-B grade I-II) for Bell palsy are given in Table 2. Compared
with patients who received combination antiviral therapy,
E3
 Research Original Investigation Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy

Table 1. Demographic, Clinical, Electrophysiological, and Treatment Variables

Associated With Favorable Outcomes in Patients With Bell Palsya
All Patients, Favorable Outcome, Effect Size Difference
Variable No. (%) N = 1364 No. (%) n = 1099 (95% CI), %
Sex
Male 646 (47.4) 523 (47.6)
0.74 (−3.46 to 4.94)
Female 718 (52.6) 576 (52.4)
Age, y
<40 407 (29.8) 343 (31.2)
5.28 (0.90 to 9.66)
≥40 957 (70.2) 756 (68.8)
Initial H-B grade
III-IV 1053 (77.2) 885 (80.5)
15.24 (9.63 to 20.84)
V-VI 311 (22.8) 214 (19.5)
Electroneurography results
≥10% (Good) 1279 (94.0) 1034 (94.1)
4.37 (−4.9 to 13.65)
<10% (Poor) 85 (6.0) 65 (5.9)
Electromyography resultsb
Good 937 (68.7) 818 (74.4)
21.49 (16.51 to 26.47)
Poor 427 (31.3) 281 (25.6)
Controlled hypertension
Yes 492 (36.1) 412 (37.5)
4.96 (0.71 to 9.20)
No 872 (63.9) 687 (62.5)
Diabetes
No 1185 (86.9) 964 (87.7)
5.93 (−0.75 to 12.62) Abbreviation: H-B grade,
Yes 179 (13.1) 135 (12.3) House-Brackmann grade.
Treatment method a
Favorable outcome was defined
Oral corticosteroids alone 673 (49.3) 571 (52.0) as an H-B grade of II or lower.
b
Combination antiviral therapy 668 (49.0) 512 (46.6) 15.28 (1.48 to 34.86) Good results were defined as the
absence of pathological
Supportive care alone 23 (1.7) 16 (1.4)
spontaneous fibrillation activity.

those who received only supportive care had lower odds of
having a favorable outcome (OR, 0.47; 95% CI, 0.17-1.26).
However, the CI was wide, indicating low precision of this
estimate, and included the null value; therefore, no defini-
tive conclusion could be made.
Discussion
Several factors were evaluated, including age, sex, initial H-B
grade, underlying disease, such as hypertension and diabe-
tes, electrophysiological test findings, and treatment method.
Several previous studies have reported on factors associated
with better outcomes in patients with Bell palsy. Peitersen8 re-
ported that patient age at the time of complete or incomplete
paralysis was associated with treatment outcome for Bell palsy.
Children younger than 14 years had a favorable outcome, and
older patients had the worst outcome. However, other stud-
ies have found that age was not associated with treatment out-
come of Bell palsy. Takemoto et al17 reported little correlation
between age and treatment outcome. Mantsopoulos et al18 also
showed that age was not a significant prognostic factor for a
favorable outcome in Bell palsy. The association between age
and outcome of Bell palsy may seem to be controversial; age
less than 40 years was associated with favorable outcomes in
this study.
The present study used the H-B grading system, which is
most frequently used to assess the degree of facial function
in Bell palsy. In other studies,19-21 H-B grade I was set as the
threshold for a favorable outcome according to stricter crite-
ria; in the present study, an H-B grade of II or lower is be-
lieved to indicate favorable outcomes in the context of nor-
mal function in daily life.19,22 Fujiwara et al23 reported that the
facial grading score at 1 week after treatment was associated
with an unfavorable outcome of Bell palsy at 6 months. They
examined clinical variables associated with long-term out-
comes in patients with idiopathic facial nerve paralysis, in-
cluding the Yanagihara facial grading system. Mantsopoulos
et al18 reported that the most important factor in the long-
term outcome (2-6 years) after the idiopathic nerve paralysis
was the initial severity of facial weakness. In our study, the re-
sults were consistent with those of previous reports; patients
with Bell palsy had favorable outcomes when the initial H-B
grade was IV or lower.
Generally, uncontrolled hypertension is associated with
facial palsy. The association between severe hypertension
and peripheral facial palsy has been described primarily in
children. Arterial hypertension is diagnosed after a substan-
tial delay. Adequate antihypertensive treatment is associated
with a favorable outcome, and prognosis is good in patients
with hypertension controlled by medication.24 Hypertension
may increase the risk of Bell palsy among patients aged older

E4 JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020 (Reprinted) jamaotolaryngology.com

© 2020 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020

 Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy Original Investigation Research

diabetes is known to affect the facial nerve fibers, which could

Table 2. Multivariable Logistic Regression Analysis
for Favorable Outcome for Bell Palsya cause considerable delays in the blink reflex on both sides of
the face.32,33 This finding suggests that diabetes may have nega-
Variable OR (95% CI)
tive consequences on outcomes in patients’ peripheral facial
Gender
paralysis. The present study demonstrated that the OR for fa-
Male 1.04 (0.78-1.38)
vorable outcomes was higher in patients without diabetes than
Female 1 [Reference]
in those with diabetes. Moreover, we found that the absence
Age, y
of diabetes was statistically significantly associated with fa-
<40 1.56 (1.09-2.22)
vorable outcomes.
≥40 1 [Reference]
Electrophysiological tests, such as ENoG and EMG, were
Initial H-B grade
originally introduced by Esslen and Fisch.34,35 Both tests have
III-IV 2.62 (1.93-3.57)
been used to assess the severity of facial nerve injury and are
V-VI 1 [Reference]
electrophysiological measures that indirectly quantify the fa-
Electroneurography results
cial nerve function by recording the motor unit action poten-
≥10% (Good) 1.29 (0.73-2.28)
tials and/or compound muscle action potentials.36 By com-
<10% (Poor) 1 [Reference]
paring the peak-to-peak amplitude of the compound muscle
Electromyography resultsb
action potentials from the affected side with the response am-
Good 3.38 (2.48-4.61)
plitude from the nonaffected side, an estimate of the amount
Poor 1 [Reference]
of the degenerated nerve could be presented. Electroneurog-
Controlled hypertension
raphy is useful for investigating the risk of poor outcomes in
Yes 1.64 (1.16-2.33)
the early stages of acute facial palsy. In patients with Bell palsy,
No 1 [Reference]
ENoG combined with a conduction block and assessment of
Diabetes
axonal degeneration can be used to evaluate the degenerated
No 1.43 (1.04-2.36)
portion of the axons.35 It is considered the most valuable test
Yes 1 [Reference]
for investigating the risk of unfavorable outcomes in patients
Treatment method
with Bell palsy.17,18,37 Patients with more than 90% degenera-
Oral corticosteroids only 1.13 (0.82-1.54)
tion on the affected side within 14 days after onset of facial pa-
Combination antiviral therapy 1 [Reference]
ralysis are considered to have poor outcome.10 The treatment
Only supportive care 0.47 (0.17-1.26)
effects were assessed separately in patients with good (ENoG
Abbreviations: H-B, House-Brackmann; OR, odds ratio. results ≥10%) vs poor (ENoG results <10%) outcomes.38,39
a
Favorable outcome was defined as an H-B grade of II or lower. In the present study, the ENoG was performed at 4 to 5 days
b
Good electromyography results were defined as the absence of pathologic after symptom onset, and the needle EMG was performed
spontaneous fibrillation activity.
2 weeks later. Abnormal ENoG responses, such as wallerian
degeneration, usually occur within 72 hours after nerve
than 40 years.25 Controlled hypertension was a major factor injury; ENoG should never be performed before that time.
for favorable outcomes in patients with Bell palsy.26 It has Tojima et al37 found that ENoG results could reveal the extent
long been reported that hemorrhages into the facial canal are of wallerian degeneration 7 days after symptom onset in pa-
responsible for facial paralysis with severe hypertension.27,28 tients with Bell palsy. In the present study, ENoG was per-
Previous studies have shown that the incidence of Bell palsy formed at 4 to 5 days after the onset of facial paralysis. Tests
is higher in patients with uncontrolled hypertension owing performed this soon after onset did not yield clinically mean-
to poor compliance with medication. In 3 adult case studies ingful results.
with known chronic hypertension, facial palsy occurred dur- The results of EMG obtained 2 weeks after onset of Bell
ing the exacerbation of the hypertension owing to nonadher- palsy have been found to be a reliable factor associated with
ence of medication.29-31 Facial palsy immediately resolved the outcome. In studies showing the use of EMG in treating fa-
after better blood pressure control, suggesting an association cial nerve injuries,40,41 the detection of abnormal spontane-
between the controlled hypertension and resolution of facial ous activity on needle EMG is known to be a factor associated
palsy. Lee et al26 demonstrated that the initial severity of the with unfavorable outcomes. Abnormal spontaneous activity,
facial paralysis and controlled hypertension were factors including positive sharp waves and fibrillation potentials,
associated with favorable outcome. In the present study, are well accepted as signs of axonal degeneration.42 In the
appropriate control of blood pressure owing to adequate present study, the results of EMG performed 2 weeks after the
antihypertensive treatment was associated with a good out- onset of Bell palsy were considered important for assessing out-
come after the development of facial palsy. comes, and this finding is consistent with those of other
Diabetes has been reported to contribute to a poor out- studies.36,43
come in patients with Bell palsy. Takemoto et al17 found that To our knowledge, no definitive treatment guidelines
diabetes was significantly associated with unfavorable out- have been established for Bell palsy owing to the high spon-
come. Moreover, Peitersen8 showed that vascular insuffi- taneous recovery rate (up to 85.2%) and unknown cause.13
ciency and diabetic polyneuropathy were associated with The American Academy of Otolaryngology treatment guide-
poorer outcomes. Long-term hyperglycemia in patients with lines for Bell palsy suggest that treatment with oral cortico-

jamaotolaryngology.com (Reprinted) JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020 E5

© 2020 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020

 Research Original Investigation
steroids within 72 hours of symptom onset is highly likely to
be effective in patients with new-onset Bell palsy with or
without the use of concurrent antiviral therapy.2 Although
there is a consensus that early use of prednisolone is effec-
tive, prescription of antiviral agents remains controversial.
The treatment effect of prednisolone suggests that inflam-
mation by neural edema of the facial nerve is part of the
pathogenesis in Bell palsy.44
The use of additional antiviral treatment is based on the
hypothesis that a simple herpes virus infection could cause
inflammation of the facial nerve. Antiviral agents cannot
destroy viruses that have already replicated; these drugs
inhibit viral replication by interfering with the viral DNA
polymerase. Numerous studies that have compared gluco-
corticoid treatment with placebo in patients with Bell palsy
have demonstrated considerable improvement in symptoms
with the use of glucocorticoids. Sullivan and colleagues13
performed a large, randomized, controlled, and double-blind
study with predefined and specified outcome measures and
a follow-up at 9 months. They noted that an antiviral agent
was not beneficial in improving outcomes of facial paralysis.
Engström et al20 aimed to compare the efficacy of predniso-
lone and valacyclovir for resolution of facial paralysis. They
concluded that treatment with prednisolone alone reduced
the time to resolution of symptoms, whereas valacyclovir
did not, indicating that prednisolone alone is sufficient to
treat patients with Bell palsy. In contrast, Hato et al 45
reported that combination antiviral therapy was more effec-
tive in treating Bell palsy, excluding zoster sine herpete, than
the conventional prednisolone therapy and described the
importance of an early treatment with valacyclovir and
prednisolone. This result is in accordance with those of a
previous study by Adour and colleagues46 who showed that
treatment with acyclovir-prednisone is superior to that with
prednisone alone in treating patients with Bell palsy.
In discussing the treatment effects of antiviral agents, sev-
eral factors should be considered. In the study by Hato et al,45
the mean Yanagihara score was approximately 15, which cor-
responds to H-B grades IV and V. The study also examined pa-
tients with more severe facial paralysis compared with of the
patients in the study by Sullivan et al,13 who had a mean H-B
grade of 3.6 of 6. The Yanagihara facial nerve grading system47
was developed as a representative regional scale in Japan and
was standardized to facial function classes in Japan and sev-
eral other countries. The Yanagihara system measures 10 in-
dividual aspects of various facial functions. The score for each
feature can be 0 (complete palsy), 2 (partial palsy), or 4 (al-
most normal), with a maximum total score of 40. The total
score provides information on the degree of facial nerve dys-
function. The patients in the study by Sullivan et al13 seemed
to have milder facial palsy than those in the study by Hato
et al.45 The mean H-B grade for the patients in our study was
similar to the mean H-B grade in the study by Sullivan et al13
(3.59 vs 3.6) In addition, Engström et al20 reported that only
30% of patients (245 of 829) had severe palsy (H-B grade ≥V),
and the H-B grade in patients treated with prednisolone plus
placebo and prednisolone plus valacyclovir were 3.7 and 3.8,
respectively. In the present study, the proportion of patients
E6 JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020 (Reprinted)
© 2020 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020
Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy
with severe palsy (H-B grade ≥V) was 22.8%, suggesting that
the different outcome might be attributed to the large propor-
tion of patients with mild to moderate Bell palsy. de Ru et al48
also reported that a large number of patients with mild paraly-
sis recovered spontaneously, and the positive treatment ef-
fect of the antiviral drugs may have been diluted. If the study
included only those patients with severe Bell palsy, the com-
bination of corticosteroid and antiviral agents might have been
different. Some studies suggested that the combination of cor-
ticosteroid and antiviral agents is more effective than cortico-
steroids alone, particularly in patients with severe Bell palsy.
Lee et al49 showed that combined treatment with a cortico-
steroid and an antiviral agent (prednisolone plus famciclovir)
resulted in better outcomes than those of corticosteroid treat-
ment alone for treating severe Bell palsy (H-B grade ≥V). In the
present study, multivariable logistic regression analyses re-
vealed that treatment with corticosteroids alone was associ-
ated with a slightly higher odds (1.13) of favorable outcome than
the combination antiviral therapy (1.13 [95% CI, 0.82-1.54] vs
1.00 [reference]), and the upper bound of the CI (1.54) sug-
gested that the true difference may be clinically meaningful.
Because the lower bound of the CI was 0.82, combination an-
tiviral therapy could also be more beneficial than corticoste-
roids alone. Thus, these results indicate that a prospective, ran-
domized, controlled, double-blind study with adequate sample
size to detect clinically meaningful differences between treat-
ments should be conducted for patients with severe Bell palsy.
The present study investigated the factors associated
with favorable outcomes in patients with Bell palsy. Many
studies evaluated the association between several factors
and outcomes of Bell palsy. Prior studies that included larger
numbers of patients than those included in the present
analysis were meta-analyses. Therefore, intervariability
in the assessment process may be lower in our monocentric
study. To our knowledge, few studies with more than 1300
patients have been conducted on the outcomes of Bell palsy.
Strengths and Limitations
A strength of our study was its presentation of factors, base-
line characteristics, and outcomes for a large number of pa-
tients assessed for more than 10 years. This study also has some
limitations. First, we performed the ENoG at 4 to 5 days after
symptom onset. The timing of the ENoG testing, one of the im-
portant factors associated with outcomes of Bell palsy, was not
properly implemented. Second, we could not conduct a double-
blind, placebo-controlled, randomized study to evaluate the
effectiveness of the treatment method. Because of its retro-
spective design, the analysis of the optimal method for treat-
ing patients with Bell palsy was inferior to that of high-
quality randomized clinical trials. Furthermore, although the
initial severity of Bell palsy was assessed by 3 otolaryngolo-
gists, the evaluators were not blinded, which may have intro-
duced selection bias. Third, we used the H-B grading system
to assess the severity of palsy, but, we did not evaluate synki-
nesis, a risk factor for poor outcome and a major complica-
tions of Bell palsy. To overcome these limitations, further re-
search is needed to validate our results on the effectiveness
of the treatments for idiopathic peripheral facial palsy.
jamaotolaryngology.com
 Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy
Conclusions
The rate of favorable outcome (H-B grade ≤II) in patients
with Bell palsy was 80.6%. Multiple clinical factors were
associated with favorable outcomes in patients with Bell
ARTICLE INFORMATION
Accepted for Publication: November 29, 2019.
Published Online: January 23, 2020.
doi:10.1001/jamaoto.2019.4312
Author Affiliations: Department of Physical
Medicine & Rehabilitation, School of Medicine,
Kyung Hee University, Seoul, Republic of Korea
(Yoo, Soh, Chon, Lee); School of Medicine,
Department of Anatomy and Neurobiology, Kyung
Hee University, Seoul, Republic of Korea (Jung);
Medical Research Center for Bioreaction to
Reactive Oxygen Species and Biomedical Science
Institute, School of Medicine, Graduate School,
Kyung Hee University, Seoul, Republic of Korea
(S. S. Kim); School of Medicine, Department of
Radiology, Kyung Hee University, Seoul, Republic of
Korea (You); School of Medicine, Department of
Otorhinolaryngology–Head and Neck Surgery,
Kyung Hee University, Seoul, Republic of Korea
(Byun, S. H. Kim, Yeo).
Author Contributions: Drs Yoo and Yeo had full
access to all of the data in the study and take
responsibility for the integrity of the data and the
accuracy of the data analysis.
Concept and design: Yoo, Soh, Chon, Lee, Jung,
S. H. Kim, Yeo.
Acquisition, analysis, or interpretation of data: Yoo,
Soh, S. S. Kim, You, Byun, Yeo.
Drafting of the manuscript: Yoo, Soh, Chon, You,
Yeo.
Critical revision of the manuscript for important
intellectual content: Yoo, Lee, Jung, S. S. Kim, Byun,
S. H. Kim, Yeo.
Statistical analysis: Yoo, Soh, Jung, You, Byun.
Obtained funding: Jung, S. S. Kim, Yeo.
Administrative, technical, or material support: Yoo,
Soh, Byun, S. H. Kim, Yeo.
Supervision: Soh, Chon, Lee, S. S. Kim, S. H. Kim,
Yeo.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by
grant KHU-20191237 from Kyung Hee University in
2019 (Dr Yeo).
Role of the Funder/Sponsor: The funder had no
role in the design and conduct of the study;
collection, management, analysis, and
interpretation of the data; preparation, review, or
approval of the manuscript; and decision to submit
the manuscript for publication.
REFERENCES
1. Tiemstra JD, Khatkhate N. Bell’s palsy: diagnosis
and management. Am Fam Physician. 2007;76(7):
997-1002.
2. Baugh RF, Basura GJ, Ishii LE, et al. Clinical
practice guideline: Bell’s palsy. Otolaryngol Head
Neck Surg. 2013;149(3)(suppl):S1-S27. doi:10.1177/
0194599813505967
3. Murakami S, Mizobuchi M, Nakashiro Y, Doi T,
Hato N, Yanagihara N. Bell palsy and herpes simplex
virus: identification of viral DNA in endoneurial fluid
jamaotolaryngology.com
© 2020 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020
palsy, including younger age, lower initial H-B grade, good
EMG result, absence of diabetes, and control of hyperten-
sion. Therefore, the present study may help identify patients
likely to have favorable outcomes based on specific factors,
which may aid clinicians in selecting patients for therapeutic
clinical trials.
and muscle. Ann Intern Med. 1996;124(1, pt 1):27-30.
doi:10.7326/0003-4819-124-1_Part_1-199601010-
00005
4. Lagalla G, Logullo F, Di Bella P, Provinciali L,
Ceravolo MG. Influence of early high-dose steroid
treatment on Bell’s palsy evolution. Neurol Sci.
2002;23(3):107-112. doi:10.1007/s100720200035
5. Adour KK, Byl FM, Hilsinger RL Jr, Kahn ZM,
Sheldon MI. The true nature of Bell’s palsy: analysis
of 1,000 consecutive patients. Laryngoscope. 1978;
88(5):787-801. doi:10.1002/lary.1978.88.5.787
6. Katusic SK, Beard CM, Wiederholt WC,
Bergstralh EJ, Kurland LT. Incidence, clinical
features, and prognosis in Bell’s palsy, Rochester,
Minnesota, 1968-1982. Ann Neurol. 1986;20(5):
622-627. doi:10.1002/ana.410200511
7. Teixeira LJ, Valbuza JS, Prado GF. Physical
therapy for Bell’s palsy (idiopathic facial paralysis).
Cochrane Database Syst Rev. 2011;(12):CD006283.
doi:10.1002/14651858.CD006283.pub3
8. Peitersen E. Bell’s palsy: the spontaneous course
of 2,500 peripheral facial nerve palsies of different
etiologies. Acta Otolaryngol Suppl. 2002;(549):4-30.
doi:10.1080/000164802760370736
9. Morgenlander JC, Massey EW. Bell’s palsy.
Ensuring the best possible outcome. Postgrad Med.
1990;88(5):157-161, 164. doi:10.1080/00325481.
1990.11716398
10. Lee DH. Clinical efficacy of electroneurography
in acute facial paralysis. J Audiol Otol. 2016;20(1):8-12.
doi:10.7874/jao.2016.20.1.8
11. Kudoh A, Ebina M, Kudo H, Matsuki A. Delayed
recovery of patients with Bell’s palsy complicated
by non-insulin-dependent diabetes mellitus and
hypertension. Eur Arch Otorhinolaryngol. 1998;255
(3):166-167. doi:10.1007/s004050050036
12. Akcan FA, Dundar Y, Uluat A, Korkmaz H,
Ozdek A. Clinical prognostic factors in patients with
idiopathic peripheral facial nerve paralysis (Bell’s
palsy). Eur Res J. 2017;3(2):170-174. doi:10.18621/
eurj.293246
13. Sullivan FM, Swan IR, Donnan PT, et al. Early
treatment with prednisolone or acyclovir in Bell’s
palsy. N Engl J Med. 2007;357(16):1598-1607.
doi:10.1056/NEJMoa072006
14. Monini S, Lazzarino AI, Iacolucci C, Buffoni A,
Barbara M. Epidemiology of Bell’s palsy in an Italian
Health District: incidence and case-control study.
Acta Otorhinolaryngol Ital. 2010;30(4):198-204.
15. House JW, Brackmann DE. Facial nerve grading
system. Otolaryngol Head Neck Surg. 1985;93(2):
146-147. doi:10.1177/019459988509300202
16. Smith IM, Maynard C, Mountain RE,
Barr-Hamilton R, Armstrong M, Murray JA.
TheprognosticvalueoffacialelectroneurographyinBell’s
palsy. Clin Otolaryngol Allied Sci. 1994;19(3):201-203.
doi:10.1111/j.1365-2273.1994.tb01215.x
17. Takemoto N, Horii A, Sakata Y, Inohara H.
Prognostic factors of peripheral facial palsy:
(Reprinted) JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020
Original Investigation Research
multivariate analysis followed by receiver operating
characteristic and Kaplan-Meier analyses. Otol
Neurotol. 2011;32(6):1031-1036. doi:10.1097/MAO.
0b013e31822558de
18. Mantsopoulos K, Psillas G, Psychogios G,
Brase C, Iro H, Constantinidis J. Predicting the
long-term outcome after idiopathic facial nerve
paralysis. Otol Neurotol. 2011;32(5):848-851.
doi:10.1097/MAO.0b013e31821da2c6
19. Berg T, Marsk E, Engström M, Hultcrantz M,
Hadziosmanovic N, Jonsson L. The effect of study
design and analysis methods on recovery rates in
Bell’s palsy. Laryngoscope. 2009;119(10):2046-2050.
doi:10.1002/lary.20626
20. Engström M, Berg T, Stjernquist-Desatnik A,
et al. Prednisolone and valaciclovir in Bell’s palsy:
a randomised, double-blind, placebo-controlled,
multicentre trial. Lancet Neurol. 2008;7(11):
993-1000. doi:10.1016/S1474-4422(08)70221-7
21. Prakash KM, Raymond AA. The use of nerve
conduction studies in determining the short-term
outcome of Bell’s palsy. Med J Malaysia. 2003;58
(1):69-78.
22. Yeo SW, Lee DH, Jun BC, Chang KH, Park YS.
Analysis of prognostic factors in Bell’s palsy and
Ramsay Hunt syndrome. Auris Nasus Larynx. 2007;
34(2):159-164. doi:10.1016/j.anl.2006.09.005
23. Fujiwara T, Hato N, Gyo K, Yanagihara N.
Prognostic factors of Bell’s palsy: prospective
patient collected observational study. Eur Arch
Otorhinolaryngol. 2014;271(7):1891-1895. doi:10.1007/
s00405-013-2676-9
24. Jörg R, Milani GP, Simonetti GD, Bianchetti MG,
Simonetti BG. Peripheral facial nerve palsy in severe
systemic hypertension: a systematic review. Am J
Hypertens. 2013;26(3):351-356. doi:10.1093/ajh/
hps045
25. Savadi-Oskouei D, Abedi A, Sadeghi-Bazargani
H. Independent role of hypertension in Bell’s palsy:
a case-control study. Eur Neurol. 2008;60(5):
253-257. doi:10.1159/000151701
26. Lee HY, Byun JY, Park MS, Yeo SG. Effect of
aging on the prognosis of Bell’s palsy. Otol Neurotol.
2013;34(4):766-770. doi:10.1097/MAO.
0b013e3182829636
27. Moxon W. Apoplexy into canal of Fallopius in a
case of Bright’s disease, causing facial paralysis.
Trans Pathol Soc London. 1869;20:420-422.
28. Matsumoto Y, Pulec JL, Patterson MJ,
Yanagihara N. Facial nerve biopsy for etiologic
clarification of Bell’s palsy. Ann Otol Rhinol Laryngol
Suppl. 1988;137(6)(suppl 3):22-27. doi:10.1177/
00034894880976S307
29. Agarwal R, Manandhar L, Saluja P, Grandhi B.
Pontine stroke presenting as isolated facial nerve
palsy mimicking Bell’s palsy: a case report. J Med
Case Rep. 2011;5(5):287. doi:10.1186/1752-1947-
5-287
30. Ellis SL, Carter BL, Leehey MA, Conry CM.
Bell’s palsy in an older patient with uncontrolled
E7
 Research Original Investigation
hypertension due to medication nonadherence.
Ann Pharmacother. 1999;33(12):1269-1273. doi:10.
1345/aph.19129
31. Lavin PJ, Weissman BM. ‘Bell’s palsy’ in
accelerated hypertension. Postgrad Med.
1985;77(8):165-168, 168. doi:10.1080/00325481.
1985.11699035
32. Kiziltan ME, Akalin MA, Şahin R, Uluduz D.
Peripheral neuropathy in patients with diabetes
mellitus presenting as Bell’s palsy. Neurosci Lett.
2007;427(3):138-141. doi:10.1016/j.neulet.2007.
09.029
33. Kiziltan ME, Uluduz D, Yaman M, Uzun N.
Electrophysiological findings of acute peripheral
facial palsy in diabetic and non-diabetic patients.
Neurosci Lett. 2007;418(3):222-226. doi:10.1016/j.
neulet.2007.03.028
34. Esslen E. The acute facial palsies: investigations
on the localization and pathogenesis of
meato-labyrinthine facial palsies. Schriftenr Neurol.
1977;18:1-164.
35. Fisch U. Maximal nerve excitability testing vs
electroneuronography. Arch Otolaryngol. 1980;106
(6):352-357. doi:10.1001/archotol.1980.
00790300040008
36. Mannarelli G, Griffin GR, Kileny P, Edwards B.
Electrophysiological measures in facial paresis and
paralysis. Oper Tech Otolaryngol–Head Neck Surg.
2012;23(4):236-247. doi:10.1016/j.otot.2012.08.003
E8 JAMA Otolaryngology–Head & Neck Surgery Published online January 23, 2020 (Reprinted)
© 2020 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ by a East Carolina University User on 01/25/2020
Evaluation of Factors Associated With Favorable Outcomes in Adults With Bell Palsy
37. Tojima H, Aoyagi M, Inamura H, Koike Y. Clinical
advantages of electroneurography in patients with
Bell’s palsy within two weeks after onset. Acta
Otolaryngol Suppl. 1994;511:147-149. doi:10.3109/
00016489409128320
38. Fisch U. Surgery for Bell’s palsy. Arch Otolaryngol.
1981;107(1):1-11. doi:10.1001/archotol.1981.
00790370003001
39. May M, Klein SR, Taylor FH. Idiopathic (Bell’s)
facial palsy: natural history defies steroid or surgical
treatment. Laryngoscope. 1985;95(4):406-409.
doi:10.1288/00005537-198504000-00007
40. Sittel C, Stennert E. Prognostic value of
electromyography in acute peripheral facial nerve
palsy. Otol Neurotol. 2001;22(1):100-104. doi:10.
1097/00129492-200101000-00019
41. Grosheva M, Wittekindt C, Guntinas-Lichius O.
Prognostic value of electroneurography and
electromyography in facial palsy. Laryngoscope.
2008;118(3):394-397. doi:10.1097/MLG.
0b013e31815d8e68
42. Stålberg E. Invited review: electrodiagnostic
assessment and monitoring of motor unit changes
in disease. Muscle Nerve. 1991;14(4):293-303.
doi:10.1002/mus.880140402
43. Yamout BI, Zaytoun G, Nuweihed I. The role of
facial nerve conduction studies and electromyography
in predicting the outcome of Bell’s palsy. Eur J Neurol.
1997;4(4):348-351. doi:10.1111/j.1468-1331.1997.
tb00360.x
44. Furuta Y, Fukuda S, Chida E, et al. Reactivation
of herpes simplex virus type 1 in patients with Bell’s
palsy. J Med Virol. 1998;54(3):162-166. doi:10.1002/
(SICI)1096-9071(199803)54:3<162::AID-JMV3>3.
0.CO;2-3
45. Hato N, Yamada H, Kohno H, et al. Valacyclovir
and prednisolone treatment for Bell’s palsy:
a multicenter, randomized, placebo-controlled
study. Otol Neurotol. 2007;28(3):408-413.
doi:10.1097/01.mao.0000265190.29969.12
46. Adour KK, Ruboyianes JM, Von Doersten PG,
et al. Bell’s palsy treatment with acyclovir and
prednisone compared with prednisone alone:
a double-blind, randomized, controlled trial. Ann
Otol Rhinol Laryngol. 1996;105(5):371-378.
doi:10.1177/000348949610500508
47. Hato N, Fujiwara T, Gyo K, Yanagihara N.
Yanagihara facial nerve grading system as a
prognostic tool in Bell’s palsy. Otol Neurotol. 2014;
35(9):1669-1672. doi:10.1097/MAO.
0000000000000468
48. de Ru JA, van Benthem PP, Janssen LM.
Is antiviral medication for severe Bell’s palsy still
useful? Lancet Neurol. 2009;8(6):509. doi:10.1016/
S1474-4422(09)70114-0
49. Lee HY, Byun JY, Park MS, Yeo SG.
Steroid-antiviral treatment improves the recovery
rate in patients with severe Bell’s palsy. Am J Med.
2013;126(4):336-341. doi:10.1016/j.amjmed.2012.
08.020
jamaotolaryngology.com