Definition, prevalence and aetiology
Herpes zoster oticus is a viral infection of the inner, middle and external ear characterised by herpetic blisters (small vesicles) of the skin of the external canal, pinna and/or the oral mucosa and severe otalgia (ear pain). These symptoms can be accompanied by one or more of the following: vestibulocochlear dysfunction (e.g. vertigo, hearing loss, hyperacusis, tinnitus), taste impairment, dry mouth and eyes.

When herpes zoster oticus is associated with acute peripheral facial paralysis the condition is known as Ramsay Hunt syndrome type 2 (Ramsay Hunt 1907). Although the terms 'herpes zoster oticus' and 'Ramsay Hunt syndrome' are frequently used interchangeably in the literature, this review will focus only on Ramsay Hunt syndrome and will include trials with participants suffering from herpes zoster oticus with accompanying facial weakness.

The incidence of Ramsay Hunt syndrome is about five cases per 100,000 of the US population annually, with a significantly increased incidence in patients older than 60 years (Murakami 1997). The condition is considered to be both less frequent and less severe in children (Hato 2000; Sweeney 2001), but reports of children are limited in the literature and the clinical features and prognosis of paediatric Ramsay Hunt syndrome remain unclear. Vaccination against chickenpox (both in childhood and in later life) may prevent or reduce its occurrence (Guess 1985; Hato 2000; Oxman 1995).

Ramsay Hunt syndrome is caused by the reactivation of latent varicella‐zoster virus in the geniculate ganglion (the sensory ganglion of the facial nerve) which affects the seventh and eighth cranial nerves. This is due to the close proximity of the facial nerve (cranial nerve VII) and the vestibulocochlear nerve (cranial nerve VIII) (Sweeney 2001). The subsequent dysfunction of cranial nerves VII and VIII is thought to be caused primarily by varicella‐zoster virus neuritis and to a lesser extent by inflammatory oedema. Maximal facial paralysis is generally seen within one week. Although less common than Bell's palsy (i.e. idiopathic facial palsy), Ramsay Hunt syndrome generally causes more severe dysfunction and has a poorer prognosis for facial nerve recovery (Uri 2003).

History, examination and diagnosis
Early diagnosis of Ramsay Hunt syndrome is extremely important, as antiviral treatment within the first 72 hours of the onset of symptoms is generally considered to be crucial to the resolution of some (if not all) symptoms (Dworkin 2007).

Diagnosis is primarily clinical and is based on the presentation of a combination of severe ear pain, small vesicles on the pinna or oral mucosa and facial palsy. When patients present with acute facial weakness a high index of suspicion is required in order to recognise the condition, deliver prompt treatment and obtain the best possible outcome. Although the appearance of vesicles usually precedes or presents simultaneously with facial paralysis, it may be delayed until after facial weakness is clinically evident, thus making the condition initially indistinguishable from Bell's palsy. Symptoms and signs of vestibulocochlear dysfunction are not always present, but can include hearing loss, tinnitus, vertigo, nystagmus, nausea and vomiting.

The clinical picture remains the cornerstone of diagnosis, as serum titres of varicella‐zoster virus antibodies are only useful when comparing the acute and convalescent stages of the condition (Dickins 1988). The use of a polymerase chain reaction (PCR) to detect varicella‐zoster virus from skin of the affected area of the ear may help to distinguish between patients with Bell's palsy and patients with Ramsay Hunt syndrome in the very early stages (Murakami 1998).

If diagnosis is made and antiviral treatment commenced within 72 hours of the onset of symptom(s), the facial palsy recovery rate has been reported in non‐randomised studies to be as high as 75% (Murakami 1997; Dickins 1988). However, even with early treatment, the absence of demonstrable lack of nerve excitability, complete paralysis and age over 50 years at presentation are said to be significant factors for increased risk of poor prognosis (Sweeney 2001). By contrast, the prognosis for associated sensorineural hearing loss associated with Ramsay Hunt syndrome is excellent, and only around 5% of patients will have residual hearing impairment (Kaberos 2002; Murakami 1997). Other potentially chronic sequelae are post‐herpetic neuralgia, tinnitus and vestibular dysfunction (Dworkin 2007).

Management options
The standard first‐line treatment for herpes zoster infections at sites in the body other than the ear is the anti‐viral agent acyclovir, which is given either intravenously for immunocompromised patients or orally in otherwise healthy patients. Other antiviral agents that may be prescribed include valacyclovir, famcyclovir or brivudin (Dworkin 2007).

Antiviral therapy is only effective against virus replication in herpes zoster ‐ that is, it can prevent further proliferation and spread of the varicella‐zoster virus ‐ and cannot eradicate herpes viruses. Oral antiviral treatment has been shown by two meta‐analyses (of four and five randomised controlled trials respectively) to lessen the severity of herpes zoster virus infection in immunocompetent adults by reducing the duration of viral shedding and new lesion formation, thereby accelerating rash healing and reducing the duration of pain (Wood 1996; Jackson 1997).

It is generally held that the use of oral antivirals is limited to a period of seven to ten days and needs to be started within 72 hours of the onset of a rash. They are generally very well tolerated if administered at standard doses (acyclovir: 800 mg five times/day for 7 to 10 days; famcyclovir: 500 mg three times/day for seven days; valacyclovir: 1000 mg three times/day for seven days) and providing patients are kept well hydrated (Goodman 2006; Lampee 1986; Kinishi 2001; Dworkin 2007). The most common side effects are nausea (and occasionally vomiting) and headache, which occur in 10% to 20% of cases. Other possible side effects include renal impairment, diarrhoea, dizziness, fatigue, skin rash, anorexia, leg pain, medication taste, sore throat and hair loss from prolonged use (Bean 1982; Dickins 1988). Any allergic reaction to antiviral medications is a contraindication to their use (Peterslund 1981).

Oral brivudin appears to have a higher potency against varicella‐zoster virus than the oral regimes of acyclovir, valacyclovir and famcyclovir. It also has a simpler regime (once a day) and no nephrotoxic effects (Dworkin 2007; Gross 2003). However, antivirals cannot prevent either the acute or chronic pain experienced by about 20% of patients over 50 years of age with herpes zoster and therefore supplementary therapies, most frequently adjuvant corticosteroids, are often prescribed for herpes zoster infections, including Ramsay Hunt syndrome. The rationale for their use is to reduce nerve inflammation and associated pain.

Although antivirals have become standard therapy for Ramsay Hunt syndrome in Europe and the US in recent years, and a number of primary studies assessing their effectiveness have been published, we found no systematic review of randomised studies addressing this question. Therefore, this systematic review aims to evaluate whether antiviral therapy does significantly improve outcomes in patients with Ramsey Hunt syndrome.