Section for Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand
Division of Neurology, Department of Medicine, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand
Centre for Clinical Epidemiology and Biostatistics, The University of Newcastle, Hunter Medical Research Institute, and Department of General Medicine, John Hunter Hospital, Newcastle, NSW, Australia
Abstract
Background
Previous metaanalyses of treatments for Bell's palsy are still inconclusive due to different comparators, insufficient data, and lack of power. We therefore conducted a network metaanalysis combining direct and indirect comparisons for assessing efficacy of steroids and antiviral treatment (AVT) at 3 and 6 months.
Methods
We searched Medline and EMBASE until September 2010 using PubMed and Elsviere search engines. A network metaanalysis was performed to assess disease recovery using a mixed effects hierarchical model. Goodness of fit of the model was assessed, and the pooled odds ratio (OR) and 95% confidence interval (CI) were estimated.
Results
Six studies (total n = 1805)were eligible and contributed to the network metaanalysis. The pooled ORs for resolution at 3 months were 1.24 (95% CI: 0.79  1.94) for Acyclovir plus Prednisolone and 1.02 (95% CI: 0.73  1.42) for Valacyclovir plus Prednisolone, versus Prednisolone alone. Either Acyclovir or Valacyclovir singly had significantly lower efficacy than Prednisolone alone, i.e., ORs were 0·44 (95% CI: 0·28  0·68) and 0·60 (95% CI: 0·42  0·87), respectively. Neither of the antiviral agents was significantly different compared with placebo, with a pooled OR of 1·25 (95% CI: 0·78  1·98) for Acyclovir and 0·91 (95% CI: 0·63  1·31) for Valacyclovir. Overall, Prednisolonebased treatment increased the chance of recovery 2fold (95% CI: 1·55  2·42) compared to nonPrednisolonebased treatment. To gain 1 extra recovery, 6 and 26 patients need to be treated with Acyclovir and prednisolone compared to placebo and prednisolone alone, respectively.
Conclusions
Our evidence suggests that the current practice of treating Bell's palsy with AVT plus corticosteroid may lead to slightly higher recovery rates compared to treating with prednisone alone but this does not quite reach statistical significance; prednisone remains the best evidencebased treatment.
Background
Bell's palsy is a condition characterized by an acute onset of facial nerve palsy with no known cause. The incidence is about 20/year/100,000 population
Although the actual cause of Bell's palsy is unknown, the widely accepted mechanism is inflammation of the facial nerve during its course through the bony labyrinthine part of the facial canal, which leads to compression and demyelination of the axons, and disruption of blood supply to the nerve itself
Treatment of Bell's palsy varies, and no clear consensus exists. Most physicians prescribe corticosteroids as a primary treatment due to its potential to reduce swelling and inflammation. The addition of antiviral treatment (AVT) such as Acyclovir or Valacyclovir is aimed at eradication of HSV infection. Acyclovir, a nucleoside analog, inhibits HSV replication through inhibition of viral DNA polymerase. It is absorbed slowly from the gastrointestinal tract, necessitating the use of a fivetimes daily regimen. Valacyclovir, a Valine derivative of Acyclovir, is claimed to lead to higher drug levels through conversion to Acyclovir by intestinal and hepatic esterase enzymes, leading to less intensive regimens. Its distribution, cellular kinetics, mechanism, and excretion are otherwise identical to Acyclovir
Efficacy of AVT in Bell's palsy is still not established, and the question exists whether adding AVT to another treatment such as corticosteroid can lead to better and faster recovery compared with corticosteroids alone or without treatment. The original Cochrane systematic review of this topic
Methods
Search strategy
One author (NP) located studies in MEDLINE (from 1966 to August 2010) and EMBASE (from 1950 to September 2010) using PubMed and Ovid search engines. Search terms used were as follows: (Bell's palsy or idiopathic facial palsy) and (antiviral agents or acyclovir or valacyclovir), limited to randomized controlled trials. Search strategies for both databases are described in the additional file
Appendix.
Click here for file
Selection of study and inclusion/exclusion criteria
Abstracts and/or full papers of identified studies were reviewed by one author (NP) and checked by another author (TA). Studies were included if they were RCTs, and studied subjects aged 18 years or older with sufficient data. NonEnglish papers were excluded from the review. Where eligible papers had insufficient information, corresponding authors were contacted by email for additional information. The reference lists of the retrieved papers were also reviewed to identify relevant publications. Where there were multiple publications from the same study group, the most complete and recent results were used.
Data extraction & Quality assessment (QA)
Data extraction was independently performed in duplicate by PN and AT using a standardized data extraction form, which included study design, sample size, patient characteristics (i.e., age, gender), type of intervention and comparator, outcomes, and followup time. Any disagreement was resolved by discussion.
Quality of studies was also independently assessed by PN and AT based on a modified Jadad score which takes into account randomization technique, allocation concealment, blinding, intention to treat, and patient attrition
Outcome
Complete recovery was defined as a score ≤2 on the HouseBrackman Facial Recovery scale
Statistical analysis
For direct metaanalysis, the odds ratio (OR) and variance of each study were estimated and pooled. Heterogeneity of ORs was assessed using Cochran's Q test and I^{2}. If heterogeneity was present, ORs were pooled using the random effects model, i.e. DerSimonian and Laird method. Metaregression was applied to assess whether age, gender, and QA were sources of heterogeneity, if these data were available. Contour enhanced funnel plots were used to detect publication bias due to small study effects
For network metaanalysis, treatment groups were considered in a mixed effects hierarchical model with logit link function using the xtmelogit command in STATA
Results
A flow diagram of study selection is shown in Figure
Diagram of selection of studies
Diagram of selection of studies.
Baseline characteristics of included studies
Author (Year)
N
Mean age
% Male Participants
Mean disease severity score
Type of treatment
Outcomes
FU time(month)
QA score
Intervention (dosage)
Control
Adour et al. (1996)
99
43
50
3 (FPRP)
Acyclovir 2,000 mg/day × 10 days Prednisolone 30 mg/kg/day × 5 days, 10 mg/day × next 5 days
Prednisolone with the same dosage
FPRI ≥ 8
4
7
De Diego et al. (1998)
101



Acyclovir 2,400 mg/day ×10 days
Prednisolone 1 mg/kg/day × 10 days, taper over next 6 days
HB grade ≤ II, FPRI score≥ 8
3
3
Hato et al. (2007)
221
50
53
15 (Yanagihara)
Valacyclovir 1,000 mg/day × 5 days Prednisolone 60 mg/day for 5 days, taper with Mecobalamin 1·5 mg/day × 6 months
Placebo
Yanagihara score > 36 points, no facial contracture or synkinesis
6
9
Sullivan et al. (2007)
551
44
51
3.6 (HB)
Acyclovir 2000 mg/day × 10 days Prednisolone 50 mg/day × 10 days
Placebo
HB grade I
3, 9
12
Yeo et al. (2008)
91
41
45
3.7 (HB)
Acyclovir 2,400 mg/day × 5 days Prednisolone 1 mg/kg/day × 4 days with maximum of 80 mg/day, 60 mg/day × 56 day, 40 mg/day × 78 day, 20 mg/day × 910 day
Placebo
HB grade ≤ II
2, 6
2
Engstrom et al. (2008)
829
40
59
4 by HB
Valacyclovir 3,000 mg/day × 7 days Prednisolone 60 mg/day × 5 days, 10 mg/day until 10 days
Placebo
Sunnybrook 100/100, HB ≤ II
1, 2, 3, 6, 12
12
Direct comparisons
Among 6 studies, 3 studies
Describe numbers of recovery according to treatment groups for each included study
Author
Treatment groups
Recovery at 3 months
Total N
Recovery after 3 months
Total N
Adour et al.
Acy+Pred
34
38
49
53
Pred
20
29
35
46
De Diego
Acy
42
54


Pred
44
47


Engstorm et al.
Val+Pred
134
206
149
206
Val
113
207
120
207
Pred
137
210
150
210
Plac
111
206
127
206
Hato et al.
Val+Pred
94
114
110
114
Pred
80
107
96
107
Sullivan et al.
Acy+Pred
99
124
115
124
Acy
76
123
96
123
Pred
109
127
122
127
Plac
79
122
104
122
Yeo et al.
Acy + Pred
36
44
41
44
Pred
35
47
40
47
Combining 5 studies
Contour enhanced funnel plot: AT plus Prednisolone versus Prednisolone
Contour enhanced funnel plot: AT plus Prednisolone versus Prednisolone.
Three studies
Network metaanalysis
All six studies
All treatment comparisons and results of the network metaanalysis for 3 month outcomes are displayed in Figure
Comparisons of recovery rates between treatment groups: A network metaanalysis
Comparisons of recovery rates between treatment groups: A network metaanalysis. a) at 3 months.
b) at > 3 months
Treatment effects for all comparisons are shown in Figure
Comparisons of treatment effects on disease recovery at 3 months between direct and network metaanalyses
Time at assessment
Intervention Group
Reference Group
Direct metaanalysis
Network metaanalysis
Discrepancy
n
OR
95% CI
SE
n
OR
95% CI
SE
Z
P value
3 month
Acy + Pred
Pred
409
1·39
0·523·75
0·32
773
1·24
0·791·94
0·28
0.27
0.788
Val + Pred
Pred
637
1·17
0·751·81
0·22
887
1·02
0·731·42
0·17
0·49
0·621
Acy
Pred
351
0·26
0·150·46
0·28
744
0·44
0·280·68
0·10
1·77
0·077
Val
Pred
417
0·64
0·430·95
0·21
774
0·60
0·420·87
0·11
0·27
0·785
Plac
Pred
665
0·50
0·360·70
0·17
895
0·54
0·400·74
0·09
0·40
0·689
Acy+Pred
Acy
247
2·45
1·33  4·53
0·29
383
2·84
1·68  4·78
0·75
0·83
0·20
Val+Pred
Acy




497
2·32
1·37  3·93
0·62


Val
Acy




384
1·37
0·79  2·37
0·38


Plac
Acy
245
1·14
0·65  1·98
0·27
505
1·25
0·78  1·98
0·29
0·23
0·41
Acy+Pred
Val




413
2·06
1·20  3·56
0·57
Val+Pred
Val
413
1·55
1·02  2·35
0·20
527
1·69
1·16  2·48
0·33
0·22
0·411
Plac
Val
413
0·97
0·65  1·45
0·20
535
0·91
0·63  1·31
0·17
0·24
0·404
Acy+Pred
Plac
246
2·16
1·17  4·00
0·29
534
2·27
1·31  2·66
0·55
0·08
0·468
Val+Pred
Plac
412
1·59
1·05  2·42
0·20
648
1·87
1·31  2·66
0·34
0·41
0·340
Acy+Pred
Val+Pred




526
1·22
0·72  2·06
0·32


6 month
Acy + Pred
Pred
441
1·63
0·47  5·75
0·83
758
1·74
0·933·24
0·56
0·07
0·474
Val + Pred
Pred
637
1·58
0·55  4·48
0·40
857
1·15
0·781·69
0·23
0·67
0·246
Acy
Pred
250
0·15
0·04  0·41
0·25
660
0·33
0·180·61
0·10
2·93
0·002
Val
Pred
417
0·55
0·36  0·85
0·21
744
0·55
0·370·81
0·11
0·00
0·500
Plac
Pred
665
0·44
0·17  1·14
0·34
865
0·62
0·430·88
0·11
0·96
0·169
Acy+Pred
Acy
247
3·6
1·54  9·07
0·41
344
5·21
2·48  10·94
1·97
0·18
0·427
Val+Pred
Acy




443
3·45
1·78  6·68
1·16


Val
Acy




330
1·64
0·85  3·18
0·55


Plac
Acy
245
1·63
0. 80  3·34
0·34
451
1·84
1·02  3·33
0·56
0·18
0·427
Acy+Pred
Val




428
3·17
1·57  6·40
1·14
Val+Pred
Val
413
1·89
1·23  2·92
0·21
527
2·10
1·40  3·15
0·44
0·22
0·414
Plac
Val
413
1·16
0·77  1·76
0·20
535
1·12
0·76  1·65
0·22
0·12
0·453
Acy+Pred
Plac
246
2·21
0·89  5·82
0·43
549
2·83
1·46  5·47
0·95
0·24
0·406
Val+Pred
Plac
412
1·63
1·05  2·51
0·21
648
1·86
1·26  2·77
0·38
0·30
0·381
Acy+Pred
Val+Pred




541
1·51
0·75  3·05
0·54


These effects were largely consistent when recovery was judged at later time points (Figure
Comparison of direct and network metaanalysis
Results of direct and network analyses are compared in the last column of table
Finally, estimates of NNT and NNH for recovery at 3 and 6 months are given in table
Estimated numbers needed to treat and numbers needed to harm of treatments
Intervention Group
Reference Group
NNT/NNH (3 months)
NNT/NNH (6 months)
Point Estimate
95% CI
Point Estimate
95% CI
Acy + Pred
Pred
NNT 26
NNH 21
NNT 10
NNT 15
NNH 93
NNT 9
Val + Pred
Pred
NNT 271
NNH 16
NNT 17
NNT 52
NNH 26
NNT 16
Acy
Pred
NNH 6
NNH 3
NNH 13
NNH 5
NNH 3
NNH 12
Val
Pred
NNH 9
NNH 5
NNH 37
NNH 10
NNH 5
NNH 31
Plac
Pred
NNH 8
NNH 5
NNH 17
NNH 13
NNH 6
NNH 52
Acy+Pred
Acy
NNT 8
NNT 10
NNT 4
NNT 6
NNT 8
NNT 5
Val+Pred
Acy
NNT 3
NNT 15
NNT 5
NNT 7
NNT 12
NNT 6
Val
Acy
NNT 15
NNH 18
NNT 6
NNT 14
NNH 34
NNT 7
Plac
Acy
NNT 21
NNH 17
NNT 8
NNT 11
NNT 296
NNT 7
Acy+Pred
Val
NNT 6
NNT 22
NNT 4
NNT 4
NNT 10
NNT 3
Val+Pred
Val
NNT 8
NNT 27
NNT 5
NNT 6
NNT 13
NNT 4
Plac
Val
NNH 43
NNH 9
NNT 15
NNT 37
NNH 15
NNT 9
Acy+Pred
Plac
NNT 6
NNT 16
NNT 5
NNT 6
NNT 14
NNT 4
Val+Pred
Plac
NNT 7
NNT 16
NNT 5
NNT 9
NNT 22
NNT 6
Acy+Pred
Val+Pred
NNT 26
NNH 14
NNT 8
NNT 18
NNH 21
NNT 8
Discussion
Our network metaanalysis indicates that treatment with AVT alone or placebo is significantly inferior to Prednisolone alone; the effect of AVT alone and placebo are similar to each other. Current practice of adding AVT (either Acyclovir or Valacyclovir) in the regimen with Prednisolone may increase disease recovery rates compared with Prednisolone alone, but at this point this difference is not statistically significant. Prednisolone remains the strongest evidencebased treatment, whether compared to placebo or AVT monotherapy.
The possible explanations for the lack of any incremental effect of AVT when added to corticosteroids might include:
• corticosteroids reduces the inflammatory process in Bell's palsy and this facilitates remyelination of facial nerve.
• Bell's palsy is a postinfectious immune mediated facial neuropathy rather than direct viral infection
• There may be a small incremental increase in efficacy but there is not sufficient power, even with all the trials to date, to demonstrate this. Large RCTs are needed to specifically compare corticosteroid and corticosteroid plus AVT.
This metaanalysis demonstrates well the advantage of the network approach. Assessing the efficacy of treatments for Bell's palsy based on results of individual RCTs and direct metaanalysis is difficult due to the fact that there are various treatment regimens, and too few studies performing the same treatment comparisons for pooling. For instance, 6 treatment regimens are possible in clinical practice (i.e., Acyclovir, Valacyclovir, Prednisolone, Placebo, combination of Acyclovir + Prednisolone, and Valacyclovir + Prednisolone) resulting in 15 possible treatmentpair comparisons. Previous reviews have had problems with this multiplicity of comparisons:
• a previous systematic review of AVT versus corticosteroid
• one systematic review
• Another review
The small numbers in these previous metaanalyses clearly led to lack of power. Two more complete reviews
We have applied a mixed model for network metaanalysis
Quality of included studies varied; quality assessment scores ranged from 212. Metaregression of direct metaanalysis indicated that this might be a source of heterogeneity in pooling effects of AVT. However, we could not adjust for the effects of quality assessment score and other covariables in the mixed effect model since this requires individual patient data. An individual patient data metaanalysis could be attempted, although individual level raw data are often difficult and more time consuming to access. However, with this method, covariables in both study (e.g., quality assessment) and individual levels (e.g., age, disease severity) can be assessed using a multilevel analysis approach.
Conclusion
Our evidence suggests that the current practice of treating Bell's palsy with AVT plus corticosteroid may lead to slightly higher recovery rates at 3 and 6 months compared to treating with corticosteroid alone, although at this point the sum of the data do not show that this is a significant difference; prednisone remains the best evidencebased treatment. Treating with AVT alone is significantly worse than treating with corticosteroid alone and is no better than placebo.
List of abbreviations
AVT: Antiviral treatment; OR: Odds ratio; CI: Confidence interval; HSV: Herpes Simplex virus; NNT/NNH: Number needed to treat/harm
Compteting interests
The authors declare that they have no completing interests
Authors' contributions
AT had full accessed to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: PN, AT, CD. Acquisition of data: PN, AT. Analysis and interpretation of data: PN, AT, JA. Drafting of the manuscript: PN, AT. Critical revision of the manuscript for important intellectual content: JA, CD. Final approval of the version to be published: all authors read and approved the final manuscript
Study supervision: AT, JA
Prepublication history
The prepublication history for this paper can be accessed here: