• 
• 

Three Part Question

Is [Bell’s Palsy] best treated with [prednisolone] or [prednisolone combined with an antiviral agent]?

Clinical Scenario

A 49 year old gentleman presents with weakness of the entire right side of his face. He has no other neurological features to suggest a stroke. You diagnose idiopathic facial paralysis (Bell’s palsy). Should he be treated with prednisolone or a combination of prednisolone and an antiviral agent?

Search Strategy

Medline to week 4 December 2011 Embase, 1980–June 2011 Cinhal, 1981–June 2011 Cochrane Collaboration Database
For Medline
[Bell's palsy.mp. or exp Bell Palsy/ {limit 1 to yr="1984 -Current"} OR idiopathic facial paralysis.mp. OR idiopathic facial paralyses.mp.] AND [antiviral.mp. or exp Antiviral Agents/ OR acyclovir.mp. or exp Acyclovir/ OR aciclovir.mp. OR valacyclovir.mp. or exp Antiviral Agents/ OR valaciclovir.mp. OR famcyclovir.mp. OR famciclovir.mp. OR antivirals.mp.]
For the other databases the following terms were used: (antivirals AND Bell's AND palsy)
(prednisolone AND Bell's AND palsy)

Search Outcome

185 papers were found including four systematic reviews with meta-analyses since 2008.
Two small trials were found in abstract form by searching Ovid MEDLINE (R) In-Process & Other Non-Indexed Citations Jan 13 2012 using the search term, "Bell's Palsy".

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Numthavaj et al, 2011, UK	1805 patients with Bell’s Palsy aged 18 or over.	Meta-analysis of six Randomised Controlled Trials.	Overall recovery rate within 3 months	68.9%	Different scales used to define recovery. Different levels of facial recovery used as primary end points. Small number of patients treated with antiviral alone.
			Recovery within 3 months (five studies)	Steroids plus antiviral vs steroids alone OR 1.21 (95% CI 0.77-1.89).
			Recovery within 3 months (three high quality studies)	Steroids plus antiviral vs steroids alone OR 1.01 (95% CI 0.74-1.37).
Quant el al, 2009, US	1145 patients with Bell’s Palsy – excluding children and pregnant women.	Meta-analysis of six Randomised Controlled Trials.	Overall at least partial recovery of facial muscle function	89.7%	Different scores used to define recovery. Different levels of facial recovery used as primary end point. Varying length of follow up and time between symptoms and treatment. Dose of prednisolone used was not consistent Severity of facial muscle paralysis at presentation not consistent
			Pooled proportion patients with facial muscle recovery	Steroids 88.2% (506/754), steroids and antiviral 91.2% (521/571).
			Facial muscle recovery	Steroids plus antiviral vs steroids alone OR 1.5 (95% CI 0.83-2.69), P=0.18.
de Almeida et al, 2009, Canada	2786 patients with Bell's Palsy treated with placebo, steroids or antivirals or a combination of steroids and antivirals.	18 trials (8 compared steroids with a control, 7 compared antivirals and steroids against a steroid control and 3 evaluated steroids and antivirals in more complicated designs).	Risk of unsatisfactory recovery >/=4 months	Regression analysis for corticosteroids plus antiviral agent OR 0.54 (95% CI 0.35-0.83), P=0.004 for the combination compared with either treatment alone.	Not all studies evaluated steroids and antiviral agents against steroids. Studies rejected by other analysts were included. Outcome defined as unsatisfactory recovery, in contrast to the other meta-analyses which used full recovery at 3 months.
			Risk of unsatisfactory recovery >/=4 months	Corticosteroid plus antiviral vs corticosteroid alone RR 0.75 (95% CI 0.56-1.00), P=0.05.
Goudakos et al, 2009, Greece	738 patients with unilateral facial nerve weakness of no identifiable cause treated with either steroids or steroids with any antiviral agent.	Systematic review of five trials. Adequate data was available for meta-analysis from four trials.	Complete recovery at 3 months	Corticosteroids vs combined therapy OR 1.03 (95% CI 0.74-1.42), P=0.88. This was not materially affected by sensitivity analysis or sub group analyses.	Only 4 studies were used in the meta-analysis.
			Adverse events	Corticosteroids vs combined therapy OR 0.89 (95% CI 0.5-1.6), P=0.7.
Chen et al. 2005 China	46 patients with Bell's Palsy. 23 patients received aciclovir (0.4g TDS for 5 days) and prednisolone (50mg for 5 days) and 23 received prednisolone 50mg for 5 days. Predisolone treatment was tapered over 10 days.	Randomised Controlled Trial	Full recovery of facial nerve function at 3 weeks	12/23 (52.1%) combined treatment v 5/23 (21.7%) prednisolone only, P<0.05.	Abstract only available.
			Full recovery of facial nerve function at 4 weeks	21/23 (91.3%) combined treatment v 18/23 (78.3%) prednisolone only, P<0.05.
Shahidullah et al, 2011, Bangladesh	68 patients with Bell's Palsy. 34 were treated with famciclover and prednisolone and 34 with just prednisolone.	Randomised Controlled Trial	Full recovery of facial nerve function at 1 month.	94.1% of those on combined treatment v 61.8% of those on just prednisolone.	Abstract only available. No details of drug doses or duration of treatment.
			Full recovery of facial nerve function at 3 months	97.1% of those on combined treatment v 74.5% of those on just prednisolone.

Comment(s)

The 2 meta analyses by Quant et al. and Numthavaj et al. had 5 RCTs in common. It is unclear why they did not agree on the 6th. Both demonstrated significant heterogeneity. Both found pooled ORs in favour of the combination therapy, but acknowledged that this failed to meet the accepted criteria for statistical significance. Quant et al. showed that the lower quality studies provided the "pull" in favour of combination therapy. Both meta-analyses hinted that patients with more severe facial palsy may benefit from combination therapy.

Goudakos et al. only analysed five trials, while de Almeida et al. included many trials excluded by the other analysts. The two small trials not included in the meta-analyses both found in favour of combination therapy. However, the largest and highest quality randomised controlled trials have all shown no benefit from combination therapy.

Clinical Bottom Line

The use of antivirals in the treatment of Bell's Palsy is not supported. The use of Prednisolone alone is recommended.

Level of Evidence

Level 1 - Recent well-done systematic review was considered or a study of high quality is available.

References

1. Numthavaj P, Thakkinstian A, Dejthevaporn C, et aL. Corticosteroids and Antiviral Therapy for Bell’s Palsy : A Network Meta-Analysis BMC Neurology 2011;11:1
2. Quant EC, Jeste SS, Muni RH, et al. The Benefit of Steroids Versus Steroids and Antivirals for Treatment of Bell’s Palsy : A Meta-analysis BMJ 2009;339:b3354.
3. de Almeida JR, Al Khabouri M, Guyatt GH, et al. Combined corticosteroid and antiviral treatment for Bell Palsy JAMA 2009;302(9):985-993.
4. Goudakos JK, Markou KD. Corticosterids vs corticosteroids plus antiviral agents in the treatment of Bell Palsy Arch Otolaryngol Head Neck Surg. 2009;135(6):558-564.
5. Chen WL, Yang ZH, Huang ZQ. Outcome of treatment 46 patients with Bell's palsy with aciclovir and prednisone. [Chinese] Shanghai Kou Qiang Yi Xue/Shanghai Journal of Stomatology. 2005;14(6):590-2.
6. Shahidullah M, Haque A, Islam MR, et al. Comparative Study between Combination of Famciclovir and Prednisolone with Prednisolone Alone in Acute Bell's Palsy. Mymensingh Medical Journal: MMJ. 2011;20(4):605-13.