Clinical manifestations

Presentation and course

Herpes zoster (“shingles”) typically presents as a vesicular rash with pain and itching in a dermatomal pattern. A pre-eruptive prodrome can include headache, photophobia, and malaise, but not typically fever (78).

Herpes zoster oticus is a specific form of herpes zoster that often presents with pre-eruptive (“pre-herpetic”) pain, allodynia, burning, or itching generally localized to the ear and mastoid region (142). An erythematous maculopapular rash progresses to clusters of clear vesicles on an inflamed base; such skin lesions may variably affect the tympanic membrane, external auditory canal, auditory meatus, and less commonly the external ear, adjacent skin of the mastoid process, the mucous membranes of the soft palate (presumably via the greater superficial petrosal nerve), and the anterior two thirds of the tongue (presumably via the chorda tympani nerve) (204).

Vesicle fluid quickly changes from clear to purulent. The vesicles crust over after 3 to 5 days. Associated constitutional symptoms can include lymphadenopathy, headache, malaise, and fever. Ultimately the skin lesions heal over 2 to 4 weeks, often leaving residual scarring and pigmentary changes.

Facial palsy may precede, occur simultaneously with, or follow the rash in herpes zoster oticus (03; 125) .

In adults, the rash typically occurs slightly before or around the same time as the facial paresis, whereas in children the rash may appear several days after the onset of facial paresis (48; 93; 125). Facial palsy in herpes zoster oticus can be associated with decreased lacrimation and decreased taste or dysgeusia on the anterior two thirds of the tongue. Facial paresis is usually maximal within 1 week of onset. Increasing severity of paresis is associated with increasing age. Herpes zoster oticus is more likely than Bell palsy to be associated with a complete clinical facial paralysis (194; 02).

A small proportion of cases of facial palsy associated with varicella-zoster infection do not have an associated rash (or at least no rash in the expected location in the internal auditory canal or on the tympanic membrane), a condition referred to as “zoster sine herpete” (204; 182; 62; 64; 65) and a condition long recognized with other forms of herpes zoster (146; 52; 159; 74). Zoster sine herpete is more common in children than adults, particularly in preschool children (178).

Otologic complications of herpes zoster oticus include tinnitus, sensorineural hearing loss, hyperacusis (dysacusis), vertigo, nystagmus, and skew deviation (and associated diplopia) (24; 27; 189; 202; 113; 97; 137; 229; 02; 201; 225; 118; 135; 154; 88; 246; 142).

Abnormal hearing in patients with herpes zoster oticus is associated with otalgia, herpetic rash, and vertigo, but not with the presence or severity of facial paresis (229; 124; 207). Hearing loss is more severe in the high-frequency range (4-8 KHz) than in the low-frequency range (≤ 2 KHz) (124). Audiological findings demonstrate both sensory (ie, cochlear) and neural (ie, retro-cochlear) hearing impairment (01; 229; 246). Hyperacusis is not simply attributable to seventh nerve dysfunction as a result of loss of the stapedial reflex (160), because hyperacusis is also described in herpes zoster oticus patients with an intact stapedial reflex (36; 229); in some cases, hyperacusis may result from damage to inhibitory efferent fibers in the seventh nerve (36). Various patterns of nystagmus including direction-fixed and positional direction-changing nystagmus occur in herpes zoster oticus, likely due to inflammation of the vestibular nerve and inner ear end organs (145). Direction-fixed positional nystagmus is more common than direction-changing positional nystagmus (75% vs. 5%, respectively); direction-changing positional nystagmus may result from an alteration of the specific gravity of the lateral semicircular canal cupula or endolymph due to inflammation in the inner ear membrane (123).

Herpes zoster oticus may occur as a cranial polyneuropathy, involving especially cranial nerves VII and VIII, but also III, V, VI, IX, X, XI, and XII, and C2-C4 sensory dermatomes (46; 50; 30; 16; 159; 68; 176; 02; 15; 216; 45; 73; 211; 237; 175; 88; 114; 102; 128; 179; 186; 206; 209; 05; 205; 87; 208; 213; 242; 188; 190; 04; 07; 150; 76). Cases of sequential involvement of the trigeminal and facial nerves, with lesions on MRI involving the spinal trigeminal nucleus and tract and the facial nerve (102; 179) have raised the possibility of transaxonal spread of varicella-zoster virus between the trigeminal nerve, the facial nerve, and the spinal trigeminal nucleus and tract (102). Intersynaptic spread of reactivated varicella-zoster virus along the reflex pathways of the brainstem has been proposed to explain the development of cranial polyneuropathies in conjunction with herpes zoster oticus (05). Cranial polyneuropathy is more common in adults than children (93).

In unusual cases, atypical cranial nerves or other central nervous system structures may be involved, with manifestations including optic neuropathy, Horner syndrome, trochlear nerve palsy, cerebellar abnormalities, multifocal vasculopathy and posterior circulation strokes, hemiparesis, hemi-hypoesthesia, aseptic meningitis, encephalitis, meningo-rhombencephalitis, and myelitis (116; 122; 30; 22; 18; 155; 101; 127; 181; 56; 57; 121; 193; 150; 76). The basis for some of these symptoms is not clear, and may vary in some cases, including meningeal involvement, demyelinating events, and vasculitic involvement (181; 110). Herpes zoster oticus may follow varicella zoster viral meningitis even in young immunocompetent men (110).

Clinical manifestations of herpes zoster oticus and other forms of herpes zoster are often more severe in adults than children (100; 93; 120).

Prognosis and complications

The recurrence rate of herpes zoster is approximately 6% (240). Recurrences are significantly more likely among women, those aged at least 60 years at the index episode, those with associated pain of at least 30 days’ duration following the index episode, and immunocompromised individuals (240).

Herpes zoster oticus is more likely than Bell palsy to be associated with a complete clinical facial paralysis and a less complete clinical recovery (194; 02; 73; 241; 206; 29; 69). A shorter time between herpes zoster and the onset of neurologic symptoms appears to be a negative prognostic factor (39).

Vestibular neuritis patients demonstrate a significantly higher vestibulo-ocular reflex gain recovery compared to patients with Ramsay Hunt syndrome (156). In particular, a significantly faster reduction in the latency, velocity, and organization of the compensatory saccades was observed in vestibular neuritis than in patients with Ramsay Hunt syndrome (156).

Facial paresis in herpes zoster oticus completely resolves in about one half to two thirds of patients after incomplete loss of function, usually within about 3 months (95; 48; 176; 93; 206), whereas complete recovery is achieved in only about 10% of patients after complete loss of facial nerve function (48) and in only about 27% of patients with multiple cranial neuropathy associated with herpes zoster oticus (206).

The majority of the remaining cases are left with mild residual signs, but about 10% are left with an “unsatisfactory” outcome (95). Factors associated with less complete recovery are more severe impairment (ie, incomplete eye closure, dry eye, and poor results of electroneurography), older age, vertigo, diabetes mellitus, hypertension, and multiple cranial nerve involvement (241; 41; 206; 28; 07). Factors that predict the development of synkinesis following herpes zoster oticus include results of electroneurography, the severity of facial weakness, and the change in weakness over a 1-month period (165).

With prompt treatment with antiviral medication, the outcome of facial palsy in Ramsay Hunt syndrome approximates that of Bell palsy, which emphasizes the importance of recognizing the syndrome and treating it accordingly (119).

The recovery from hearing loss is generally good, although incomplete in some patients (229; 02); however, some authors have reported a poor prognosis for hearing loss (128). Prognostic indicators of poor hearing recovery include advanced age, retro-cochlear hearing loss, hearing loss affecting the speech frequencies (ie, approximately 250 to 8000 Hz), vertigo, and male gender (229). Elderly patients in particular can be disabled by severe and persistent imbalance (02), leading to repeated falls, falls with injury, fear of falling, and impairment in activities of daily living and psychosocial function (34).

Absence of blink reflex and low mean electroneurography value are significantly associated with incomplete recovery of facial paresis (90). An absent blink reflex is more likely to be associated with severe residual facial palsy in Ramsay Hunt syndrome than in Bell palsy (90).

Clinical vignette

Case 1. Ramsay Hunt syndrome (Hunt group 2) (17). An 11-year-old boy complained of 1 week of fever, sore throat, and otalgia. He was diagnosed with otitis media and was started on oral antibiotics. However, he developed a right peripheral facial paresis 1 week later. He was diagnosed with Bell palsy and was started on methylprednisolone. On the second day of treatment with methylprednisolone, painful vesicles developed in the right ear. Audiograms were normal. He was diagnosed with Ramsay Hunt syndrome and started on acyclovir (10 mg/kg/day). The ear lesions resolved within 3 weeks, and the facial paresis resolved within 6 weeks.

Case 2. Ramsay Hunt syndrome (Hunt group 2) (17). A 12-year-old girl complained of 3 days of sore throat, cough, left otalgia, and facial asymmetry. Exam disclosed a left peripheral facial paresis and a vesicular rash in her left ear. There were no cochleovestibular symptoms, and audiograms were normal. She was started on a 5-day course of acyclovir (800 mg/day) and a 14-day course of methylprednisolone (2 mg/kg/day for 1 week, and then 1 mg/kg/day for 1 week). The facial paresis resolved by the end of the first week. Photographs show a right peripheral facial paresis and vesicular lesion in the left auricle.

Case 3. Ramsay Hunt syndrome (Hunt group 3) (17). A 12-year-old boy presented with complaints of left otalgia, vertigo, and vomiting. He developed vesicles in the left auricle and external auditory canal and a left peripheral facial paralysis. His initial audiogram showed a mild to moderately severe left sensorineural hearing loss and normal hearing on the right. He was treated with antiviral and corticosteroid therapy, and in addition, he received an antiemetic and a sedative to control the vestibular symptoms. The pain resolved by day 5, the vesicular rash and vertigo resolved by day 10, and the facial paresis, hearing loss, and subjective tinnitus resolved within the first month. An audiogram after treatment had largely normalized; there was normal hearing to slight sensorineural hearing loss on the left and normal hearing on the right.

Case 4. Ramsay Hunt syndrome (Hunt group 3) (110). A 32-year-old man presented to the emergency room with a 1-week history of persistent headache. On the fourth day of illness, he developed an intermittent fever, and a small vesicular lesion erupted around the right ear with associated mild pain, consistent with shingles. He had a history of chickenpox as a child. At the time of admission, he was severely nauseous.

Neurologic examination revealed subtle neck stiffness. C-reactive protein was elevated at 18.4 mg/L (normal range: 0-5 mg/L), whereas the other blood tests were within the normal range, including a complete blood cell count and an erythrocyte sedimentation rate. Brain MRI was normal. CSF examination revealed a slightly increased intracranial pressure (210 mm CSF), an elevated CSF protein (255.5 mg/dL), lymphocyte-dominant pleocytosis (800/mm3, lymphocytes 90%), and a slightly decreased CSF/serum glucose ratio (47/105 = 0.45). Polymerase chain reaction testing of the CSF confirmed varicella-zoster virus positivity.

He was treated with intravenous acyclovir 10 mg/kg every 8 hours, considering the possibility of varicella-zoster virus meningitis. Antipyretics and analgesics were used to control fever, headache, and pain related to the vesicular lesions around the ear. On the third day of hospitalization, he developed a right peripheral facial palsy. On the eleventh day of hospitalization, he developed a left peripheral vestibulopathy with severe vertigo and tinnitus, with corresponding spontaneous left-beating nystagmus obeying Alexander’s law (ie, greater intensity in the quick-phase direction looking away from the responsible lesion than in the slow-phase direction looking toward the responsible lesion).

Pure tone audiometry showed a high-frequency sensorineural hearing loss in the right ear. A facial nerve conduction study showed nonspecific findings. A blink reflex test revealed absent responses of ipsilateral R1 and R2 and contralateral R2 at stimulation of the right supraorbital nerve.

After he developed a peripheral facial palsy, oral prednisolone was administered (50 mg of for 7 days followed by gradual tapering). Acyclovir was continued for a total of 14 days from the day of admission. At discharge on the 20th day of hospitalization, his headache and fever had subsided significantly, but he had persistent facial palsy, dizziness, and tinnitus.

Case 5. Ramsay Hunt syndrome (Hunt group 3) (110). A 43-year-old man presented to the emergency room with a 10-day history of persistent throbbing headache, accompanied by pain in the right ear, and right facial weakness that had developed 6 days after headache onset. The patient had received a 7-day course of antiviral and steroid therapy for suspected shingles, beginning on the third day of the headache onset.

At the time of admission, he complained of vesicles and pain around the right ear, headache, right facial palsy, hypogeusia, dizziness, hearing loss, and tinnitus. He was not febrile (37.9 °C).

Neurologic examination revealed a right peripheral facial palsy. Blood tests were normal, including CBC, ESR, and CRP. An abnormal enhancing lesion was observed in the right facial nerve on a brain MRI with contrast, suggestive of right facial neuritis.

CSF examination revealed a normal opening pressure (80 mm CSF), lymphocyte-dominant pleocytosis (53/mm3, lymphocytes 98%), slightly increased CSF protein (61.0 mg/dL), and a normal CSF/serum glucose ratio (70/117 = 0.60). A CSF PCR test confirmed varicella-zoster virus positivity.

Pure tone audiometry showed a sensorineural hearing loss with mild loss at 500 Hz and sloping mild to moderate loss above 2 KHz. A facial nerve conduction study revealed prolonged terminal latency and decreased CMAP amplitude of the right facial nerve upon its direct stimulation. A blink reflex test revealed absent responses of ipsilateral R1 and R2 and contralateral R2 at stimulation of the right supraorbital nerve.

He was treated with intravenous acyclovir (10 mg/kg every 8 hours) or oral corticosteroids for 7 days (considering the antiviral treatment from a separate facility). The accompanying pain was treated symptomatically. At discharge, the patient’s headache and dizziness had subsided, but he had persistent facial palsy, auditory symptoms, and pain around his ear.

Biological basis

Etiology and pathogenesis

Herpes zoster (“shingles”) is caused by the varicella-zoster virus, a member of the alpha subfamily of Herpesviridae. Varicella-zoster virus is an enveloped DNA virus, whose genome consists of a linear double-stranded molecule of DNA. Primary infection with varicella-zoster virus causes chickenpox, an acute, generally mild infection in children. During the primary infection, the virus establishes latency within sensory ganglia (232). Later reactivation of the virus results in herpes zoster or “shingles.”

Humoral immunity to varicella-zoster virus is not important in preventing reactivation. Antibodies have little or no role in preventing virus reactivation and development of clinical herpes zoster: (1) herpes zoster is observed in patients with positive varicella-zoster virus antibodies prior to the development of zoster; (2) varicella-zoster virus antibodies increase quickly with development of clinical herpes zoster; (3) patients with disseminated zoster have high antibody titers; (4) children with agammaglobulinemia are not at increased risk to develop herpes zoster; and (5) passive immunization with high-titered immunoglobulin neither prevents nor helps treat herpes zoster (70).

In contrast, impaired cell-mediated immunity is an important factor in reactivation of varicella-zoster virus and development of clinical herpes zoster (70; 111; 14; 89). Varicella-zoster virus-specific T cells are thought to be critical for maintaining the virus-host equilibrium and preventing herpes zoster. This is consistent with known risk factors for herpes zoster including leukemia and lymphoma, bone marrow transplant, and HIV infection. Waning of the cell-mediated immune response to varicella-zoster virus with age may help explain the marked increase in zoster with age (70).

Other factors in addition to impaired cell-mediated immunity must also be involved in the development of herpes zoster because most individuals with impaired cell-mediated immunity (and a history of chickenpox) do not develop clinical herpes zoster, and some apparently healthy children and young adults occasionally do get herpes zoster (70; 120). Indeed, most children who develop herpes zoster, even with complications such as herpes zoster oticus, are immunocompetent (83).

In 1900, Head and Campbell established that the pathologic changes of herpes zoster were typically limited to a single dorsal root ganglion or to the sensory ganglion of a cranial nerve and the corresponding nerve root (94). They corroborated earlier findings of acute inflammation, blood extravasation, and destruction of neural elements within sensory ganglia in acute cases.

Unfortunately, in Hunt’s only postmortem study, the geniculate ganglion was lost and not available for examination (103; 26). Subsequent studies by Denny-Brown and Adams did not support the concept of inflammation or degeneration of the geniculate ganglion as the basis of the clinical manifestations in herpes zoster oticus (46). Indeed, the geniculate ganglion was apparently spared, and instead, there was a necrotizing ganglionitis of the second cervical dorsal root ganglion and inflammatory changes in the facial nerve distal to the ganglion.

Subsequent pathologic studies have demonstrated variable perineural, intraneural, and perivascular lymphocytic infiltration of the facial nerve, and in some cases the chorda tympani, auditory, and vestibular nerves, the modiolus and organ or Corti in the cochlea, and the skin of the external auditory meatus (86; 24; 244). Although the geniculate ganglion may show scattered lymphocytic infiltration, most of the neurons in the ganglion are well preserved in herpes zoster oticus (86; 24; 244). None of the existing pathologic studies support Hunt’s concept of extensive damage to the geniculate ganglion as the basis for herpes zoster oticus (46; 86; 24; 244). Grossly pronounced swelling of the facial nerve may be evident at surgery for as long as 4 months following herpes zoster oticus (99).

Studies have demonstrated varicella-zoster virus DNA in the geniculate ganglia of patients with herpes zoster oticus (66; 228; 67; 227; 215). Varicella-zoster virus DNA is present not only in neurons but also in perineuronal satellite cells and non-neuronal cells of latently infected ganglia (73). Varicella-zoster virus DNA may also be present in the auditory and vestibular primary afferent ganglia, and in the facial nerve sheath, CSF, middle ear mucosa, and vesicles on the auricles or oral cavity (228; 227; 169; 180). Different patterns of development of eighth cranial nerve dysfunction can be caused by progression of neuritis or labyrinthitis after varicella-zoster virus reactivation (180). Latent varicella-zoster virus is not integrated into the human chromosome; instead, it may exist in a circular or end-to-end arrangement (37).

Epidemiology

Varicella (chickenpox) is usually an illness of childhood that results from primary infection with varicella-zoster virus and is characterized by a febrile, pruritic, maculopapular vesicular rash. The vesicles evolve into noninfectious dried crusts over 5 to 6 days. Varicella is highly infectious, with secondary infection rates in susceptible household contacts of two thirds or greater. Transmission occurs from person-to-person by direct contact or by airborne spread of respiratory secretions. The incubation period is usually about 2 weeks but ranges from 10 to 21 days. Humans are the only source of infection. The vast majority of patients (approximately 90%) with chickenpox are less than 15 years old, and there is a marked winter-spring seasonal pattern (149).

Following chickenpox, varicella-zoster virus remains latent in human nerve tissue and reactivates in about 15% of infected people over their lifetimes, resulting in herpes zoster (“shingles”). Approximately two thirds of patients with herpes zoster (including herpes zoster oticus) are 65 years of age or older and there is no seasonal pattern (149). Both the incidence and severity of herpes zoster increase sharply with age (85; 54). An initial episode of herpes zoster does not apparently protect against or significantly increase the risk of a subsequent episode (100).

The absolute risk of Ramsay Hunt syndrome within 3 months of zoster diagnosis was 0.01% (95% CI 0.0- 0.01) (58).

Herpes zoster oticus is rare in children under 6 years of age (85; 23; 129; 93; 132) but is an important and often unrecognized cause of acute peripheral facial paralysis in children between 6 and 15 years of age (63; 17). Risk factors for zoster and herpes zoster oticus in children include chickenpox during infancy, childhood leukemia, and childhood HIV infection (85; 93; 171; 163).

Risk factors for zoster and herpes zoster oticus in adults include increasing age, bone marrow and organ transplants, lymphoma, HIV infection, treatment of AIDS with protease inhibitors, treatment of advanced non-small cell lung cancer with immune checkpoint inhibitor, and systemic lupus erythematosus (191; 151; 200; 172; 233; 06; 161; 53; 157; 149; 171; 54; 163; 78; 197). There is no significantly increased risk of a subsequent diagnosis of malignancy following the onset of herpes zoster and, therefore, there is little evidence to support an aggressive search for malignancy in patients diagnosed with herpes zoster (192; 59; 85; 78), although some data suggest that the sites of primary tumor or local radiotherapy may predispose to development of zoster in that specific location (196). Although herpes zoster is associated with HIV infection, herpes zoster is apparently unrelated to CD4 cell count, period from seroconversion, viral load, or rate of disease progression (including progression to AIDS) (77; 06; 161; 157; 163).

The live-attenuated varicella vaccine (Varivax) was licensed by the United States Food and Drug Administration in 1995. Varicella-zoster vaccination can prevent or reduce the occurrence of chickenpox, herpes zoster, and associated complications, including herpes zoster oticus, among those immunized (09; 184; 10; 32; 11; 12; 93; 203; 79; 223; 243). As a result of introduction of this vaccine, there has been a dramatic decline in the incidence of varicella in surveillance areas with moderate vaccine coverage (203; 79; 243). However, although the incidence of varicella decreased dramatically as varicella vaccine coverage in children increased, the incidence of herpes zoster increased (243). Although several potential explanations are possible, it seems likely that the observed increase in herpes zoster infection is a result of a loss of “immunologic boosting” (including loss of exposure to wild-type varicella-zoster virus) for those previously infected with varicella-zoster (79). This loss of immunologic boosting would allow a waning of cell-mediated immunity and would lead to a relative increase in herpes zoster incidence in the population.

Prevention

The live-attenuated varicella vaccine (Varivax) was licensed by the United States Food and Drug Administration in 1995. Varicella-zoster vaccination can prevent or reduce the occurrence of chickenpox, herpes zoster, and associated complications, including herpes zoster oticus, among those immunized (09; 184; 10; 32; 11; 12; 93; 203; 79; 223; 243).

Routine (pre-exposure) varicella-zoster vaccination is recommended for susceptible children aged 12 months to 13 years and is also desirable for susceptible adolescents and adults (10). Pre-exposure varicella-zoster vaccination is also particularly recommended for all adolescents and adults who do not have a reliable clinical history of chickenpox, or serologic evidence of varicella-zoster virus infection, and who are at high risk of exposure or transmission (224). This includes healthcare workers (particularly those who have close contact with patients at high risk of serious complications), residents and staff in institutional settings, family contacts of immunocompromised people, and women of childbearing age who are not pregnant and not planning to become pregnant in the next 4 weeks (10). Susceptible patients 13 years of age or older should receive two doses of vaccine, given at least 4 weeks apart.

Varicella vaccine is also recommended for susceptible people following exposure to varicella because if administered within 72 hours, and possibly up to 120 hours, it can prevent or attenuate clinical disease. Post-exposure vaccination should be considered particularly in health care settings and households. Additional considerations for prevention in at-risk people exposed to varicella-zoster virus include passive immunization with varicella-zoster immune globulin and in some cases pharmacologic prophylaxis with acyclovir (09; 10; 11; 12). Varicella-zoster immune globulin is a hyperimmune globulin prepared from healthy volunteer blood donors who are found to have high antibody titers to varicella-zoster virus.

Varicella-zoster vaccine is contraindicated in patients with an allergy to any of the vaccine components (including neomycin or gelatin), during pregnancy, in adults (and some children) with HIV infection, or in children or adults with severe intercurrent illness, leukemia, lymphoma, metastatic cancer, or immunosuppression due to corticosteroids, other immunosuppressive medications, or radiation (10). Varicella-zoster virus vaccine should not be given within 5 months following administration of blood, plasma, immunoglobulins, or varicella-zoster immune globulin. Salicylates should be avoided for 6 weeks following varicella-zoster virus vaccination because of the association between aspirin use and Reye syndrome following varicella.

Varicella-zoster virus infection can cause serious infection in adult women (particularly pneumonia), is teratogenic, and can produce a life-threatening illness in newborns (32). Therefore, women of childbearing age should be questioned about a history of chickenpox or varicella-zoster virus vaccination before conception and also during their first prenatal visit (32). If there is a question about their immunity to varicella-zoster virus, they should be offered serologic testing and vaccination if indicated based on the testing (10; 32). Although a test-then-vaccinate strategy is more cost-effective, testing is not required, and the vaccine is well tolerated in people who are already immune. Vaccine virus transmission to a susceptible contact can occur only if the vaccine has a rash; this risk is low in healthy individuals.

Varicella-zoster virus infection can also be associated with serious complications in immunocompromised individuals, older adults, and young infants. Passive immunization with varicella-zoster immune globulin early in the incubation period can prevent or attenuate varicella-zoster virus infections in susceptible contacts at high risk of infection (eg, young infants, immunocompromised individuals, and older adults with no history or questionable history of chickenpox) (177; 11).

Varicella-zoster virus vaccine is much safer than getting chickenpox. Side effects of varicella-zoster virus vaccine are generally mild and include soreness or swelling at the injection site, fever, or mild rash up to 1 month after vaccination. About 3% of adults develop a localized varicella rash at the injection site (median of two lesions), and about 6% develop a mild generalized rash (median of five lesions, not all at the injection site). The risk of any rash is much lower (less than 1%) after the second dose of vaccine. The rash may be atypical, with macules or papules rather than vesicles. Less common problems include febrile seizures and rarely pneumonia. Healthcare workers who develop a rash should be evaluated at their employer’s occupational health service and furloughed until the rash resolves (typically in 1 to 2 days) to prevent iatrogenic transmission.

Varicella-zoster immune globulin is indicated following exposure to varicella or herpes zoster for people at high risk of developing severe disease and those who are not eligible for vaccination: (1) susceptible pregnant women; (2) newborns whose mothers have chickenpox (or rarely herpes zoster) 5 days before through 2 days after delivery; (3) hospitalized premature infants more than 28 weeks’ gestation, whose mother had no history of varicella; (4) hospitalized premature infants less than 28 weeks or less than 1000 g (regardless of the mother’s immune status or maternal history); (5) children with leukemia or lymphoma who have not been vaccinated; (6) people with HIV/AIDS or other cell-mediated immunodeficiency, including people receiving high-dose systemic steroids or other drugs that suppress the immune system. Varicella-zoster immune globulin is most effective in preventing varicella infection if given within 96 hours of varicella-zoster virus exposure.

In particular, if exposure occurs during pregnancy, the mother can be reassured as to their protection if they have a history of chickenpox. Otherwise (ie, if there is no prior history of chickenpox or the history is uncertain), varicella-zoster immune globulin should be administered if within 10 days of exposure and preferably within 96 hours to try to prevent maternal infection (32; 96; 164). The live-virus vaccine should not be administered during pregnancy, although cases of inadvertent administration continue to occur instead of the intended varicella-zoster immune globulin (235). Oral acyclovir should be considered for susceptible pregnant women who did not receive varicella-zoster immune globulin or who have risk factors for severe disease, and intravenous acyclovir should be administered to pregnant women who develop complicated varicella at any stage of pregnancy (96). Infants whose mothers develop varicella-zoster virus infection peripartum should be treated with varicella-zoster immune globulin (ie, if the onset was from 7 days prior up to 28 days after delivery) and acyclovir (if the onset was from 4 days before to 2 days after delivery) (96; 164).

Secondary prevention of herpes zoster is now possible for older adults previously infected with chickenpox. In 2005, Oxman and colleagues reported the results of the Shingles Prevention Study, a large, randomized, double-blind, placebo-controlled trial of a high-potency, live-attenuated varicella-zoster vaccine (185; 14; 71). More than 38,000 adults over 60 years of age were enrolled. The varicella-zoster vaccine markedly reduced the morbidity from both herpes zoster and postherpetic neuralgia among older adults. After a median of more than 3 years of surveillance, the use of the varicella-zoster vaccine reduced the incidence of herpes zoster by 51%, the burden of illness due to herpes zoster by 61%, and the incidence of postherpetic neuralgia by 67%. Local reactions at the vaccination site were generally mild, and the vaccine had low rates of serious adverse events, hospitalization, and death. The U.S. Food and Drug Administration licensed this live-attenuated zoster vaccine (Zostavax) in 2006, for use in people aged 60 years and older (U.S. Food and Drug Administration 2006), and in March 2011 the FDA expanded approval of a live-attenuated zoster vaccine (Zostavax) to adults aged 50 to 59 years (31). The live zoster vaccine has a minimum potency at least 14 times greater than the minimum potency of the vaccine licensed to prevent varicella, because a much higher potency is needed to produce a significant increase in cell-mediated immunity to varicella-zoster virus in older adults (185). The vaccine is administered as a single subcutaneous injection, preferably in the upper arm.

In 2012, Schmader and colleagues reported a randomized, double-blind, placebo-controlled trial of the live, attenuated zoster vaccine (Zostavax) among 22,439 subjects aged 50 to 59 years (199). The study was conducted at 105 sites in North America and Europe. In subjects aged 50 to 59 years, the live, attenuated zoster vaccine (Zostavax) significantly reduced the incidence of herpes zoster, with vaccine efficacy at preventing herpes zoster estimated at 70%. The vaccine was well tolerated, with adverse effects primarily limited to injection site reactions and headaches. On March 24, 2011, on the basis of prepublication results of this study, the U.S. Food and Drug Administration expanded the licensing of this live, attenuated zoster vaccine (Zostavax) for use in people aged 50 years and older (U.S. Food and Drug Administration 2017).

Consequently, the presently licensed live, attenuated zoster vaccine (Zostavax) reduces the incidence of herpes zoster by 70% among those aged 50 to 59 years, by 64% among those aged 60 to 69 years, and by 38% among those aged 70 years and older (185; 199; 38). The vaccine reduces the incidence of postherpetic neuralgia by 66% among those aged 60 to 69 years, and by 67% among those aged 70 years and older (185; 38).

The long-term persistence of efficacy for the live, attenuated zoster vaccine (Zostavax) was reported in 2015, and this study showed that efficacy declines significantly over time (166). Among 6867 participants who were enrolled from the original Shingles Prevention Study into the Long-Term Persistence Substudy, estimated vaccine efficacy decreased from 61% to 37% for the herpes zoster burden of illness, from 67% to 35% for incidence of postherpetic neuralgia, and from 51% to 21% for incidence of herpes zoster. Efficacy declined for all three outcome measures from 7 through 11 years after vaccination. The vaccine significantly reduced the burden of illness caused by herpes zoster for 10 years after vaccination, but significantly reduced the incidence of herpes zoster for only 8 years after vaccination. The safety and efficacy of administering a booster dose of the live, attenuated zoster vaccine 10 or more years after the initial dose is not yet clear (38).

Given the limited efficacy and persistence of the live, attenuated zoster vaccine (Zostavax), vaccine development has continued. In 2015, Lal and colleagues reported a randomized, double-blind, placebo-controlled, phase 3 trial of an adjuvanted herpes zoster subunit vaccine in older adults (aged 50 years and older) that had been conducted in 18 countries (38; 141). The recombinant vaccine contains a surface varicella-zoster virus glycoprotein E (gE) antigen obtained from cultured cells and a liposomal adjuvant (AS01B) to enhance the immune response (43; 13; 21). Participants received two intramuscular doses of the vaccine or placebo over an interval of 2 months. During a mean follow-up of 32.2 years, overall vaccine efficacy against herpes zoster was 97%, and this estimate was remarkably uniform at between 97% to 98% for those aged 50 to 59 years, 60 to 69 years, and 70 years and older. Injection site and systemic reactions were more common in the vaccine group, but the proportions who had serious adverse events, possible immune-mediated disease (of concern given the adjuvants), or who died were similar in the vaccine and control groups.

In 2016, Cunningham and colleagues reported a second trial of the subunit vaccine (43; 174). This trial, conducted concurrently with the previously reported trial at the same sites, evaluated the safety and efficacy of the subunit vaccine in adults aged 70 years and older. During a mean follow-up of 3.7 years, vaccine efficacy against herpes zoster was 90%, and was similar in those aged 70 to 79 years and in those aged 8 years and older. Injection site and systemic reactions were more common in the vaccine group, but the proportions who had serious adverse events, possible immune-mediated disease (of concern given the adjuvants), or who died were again similar in the vaccine and control groups.

In pooled analyses of the two trials of the subunit vaccine for those aged 70 years and older, vaccine efficacy against herpes zoster was 91% and against postherpetic neuralgia was 89% (43). The subunit vaccine appears to be more effective than the live, attenuated vaccine, especially in those over 70 years of age (185; 38; 141; 43; 174). It may also prove to be useful in those with impaired cell-mediated immunity, for whom the live, attenuated vaccine is contraindicated (38): such individuals include those with hematopoietic-cell transplantation, those with HIV infection and low CD4+ cell counts, or those receiving potent immunosuppressive medications. Whether the subunit vaccine will elicit a sufficiently protective immune response in those with impaired cell-mediated immunity remains to be proven (38).

The adjuvanted, recombinant varicella-zoster virus subunit vaccine (Shingrix) was approved by the U.S. Food and Drug Administration on October 20, 2017 (84). The vaccine is administered in two intramuscular doses of 0.5 ml given 2 to 6 months apart. Based on immunogenicity data, the protective effect of recombinant varicella-zoster virus vaccine (Shingrix) is expected to persist for at least 9 years after vaccination (13). The efficacy of recombinant varicella-zoster virus vaccine (Shingrix) in persons who receive only one dose is not known. The safety and efficacy of administering the subunit zoster vaccine among those who had previously received the live, attenuated vaccine have not been tested (38), nor has there been a head-to-head trial of the live-attenuated varicella-zoster virus vaccine (Zostavax) and the recombinant varicella-zoster virus vaccine (Shingrix). The Advisory Committee on Immunization Practices now recommends recombinant varicella-zoster virus vaccine (Shingrix) over live-attenuated varicella-zoster virus vaccine (Zostavax) for herpes zoster prevention and, furthermore, recommends that healthy adults aged 50 years or older, including those who have already received live-attenuated varicella-zoster virus vaccine (Zostavax), be vaccinated with two doses of recombinant varicella-zoster virus vaccine (Shingrix) (2 to 6 months apart).

Differential diagnosis

Acute facial palsy without the pathognomonic vesicles is generally diagnosed as Bell palsy (158; 93; 211). In children, particularly, the appearance of vesicles is often delayed (93), leading to misdiagnosis of Bell palsy in patients with herpes zoster oticus. In adults, herpetic cranial polyneuritis may occasionally mimic brainstem infarction (20). In an elderly man, herpes zoster oticus was originally misdiagnosed as malignant otitis externa (42). Several clinical features associated with clinical peripheral facial paresis should increase diagnostic suspicion of herpes zoster oticus: complete clinical facial paralysis; severe otalgia; sensorineural hearing loss; or a reddish ear (even without a herpetic rash) (02).

Diagnostic workup

Herpes zoster oticus can usually be diagnosed clinically. The situation is more complicated in the absence of a rash, or in the presence of other neurologic manifestations (eg, hemiparesis, cranial polyneuropathy, etc.). Associated facial swelling in some cases can suggest bacterial odontogenic infection and facial cellulitis (115).

In high-risk settings (eg, hospitals, nursing homes, prisons), rapid case identification is important to prevent susceptible people at high risk (eg, immunocompromised or pregnant individuals) from developing serious complications of varicella-zoster virus infection.

Laboratory testing for varicella-zoster virus in herpes zoster oticus is not routinely required but can be helpful in selected cases to confirm the diagnosis (eg, zoster sine herpete) (72; 73; 211). Laboratory diagnosis or varicella-zoster virus infection can include (1) positive serologic test for varicella-zoster immunoglobulin M (IgM) antibody (214); (2) isolation of varicella-zoster virus from vesicles (or possibly blood or CSF in immunocompromised patients), although varicella-zoster virus is labile and difficult to recover from swabs of cutaneous lesions (78; 40); (3) demonstration of varicella-zoster virus antigen by direct fluorescent antibody testing in material swabbed from the base of a freshly unroofed fluid-filled vesicle or in lesion crusts; (4) detection of varicella-zoster virus DNA by polymerase chain reaction tests on clinical specimens from cutaneous lesions (168), tear fluid or saliva (98; 140), blood monocytes (214; 73), or CSF (72; 73; 40); or (5) a significant rise in varicella immunoglobulin G (IgG) antibody level in serum (214). Because viral proteins persist after the virus is no longer replicating, direct fluorescent antibody tests may be positive when viral cultures are negative. Virus isolation should be attempted in cases with severe disease, especially in immunocompromised cases. Virus can usually be cultured from zoster lesions for 7 days or longer, and results may be available within 2 to 3 days.

CSF, though rarely examined in patients with herpes zoster oticus, may be normal (244), or may show lymphocytosis and occasionally a modest elevation of protein concentration (46; 24; 133). Almost two thirds of patients with herpes zoster oticus have a modest CSF pleocytosis (mean 84 cells/mm3; range 7 to 250 cells/mm3) (133). Varicella-zoster virus IgM antibody and DNA (ie, by polymerase chain reaction techniques) can be demonstrated in CSF (72; 73; 81), which can be diagnostically helpful in some cases of zoster sine herpete. Variations in commercial, real-time polymerase chain reaction tests can impact the sensitivity of diagnostic testing (238).

Although not as sensitive as MRI, computed tomography may show abnormalities including abnormal enhancement (76).

The membranous labyrinth normally appears on T1-weighted MRI images as a nonenhancing intermediate-signal-intensity structure surrounded by the low-signal intensity of the petrous temporal bone (51). In approximately 50% to 70% of cases with facial nerve palsy due to the varicella-zoster virus, variable degrees of enhancement of the seventh and eighth cranial nerves, the labyrinth, the geniculate ganglion, and the internal and external auditory canals enhance on postcontrast MRI using gadolinium (139; 183; 134; 147; 195; 51; 212; 234; 117; 136; 25; 19; 198; 210; 102; 179; 144).

In rare cases, there may also be enhancement of the pontine facial nerve nucleus (198), as well as the internal auditory canal dura and sometimes the parotid gland (33).

Patients with herpes zoster oticus typically have high signals in the inner ear on precontrast 3D-FLAIR MRI, a finding that is rare in patients diagnosed with Bell palsy (peripheral facial paresis without vesicular rash or audiovestibular symptoms) (173; 35; 138). Precontrast hyperintensity on 3D-FLAIR MRI is associated with the presence of vertigo and severity of hearing impairment but not with clinical outcome (35).

Audio-vestibular deficits are well-correlated with increased signal intensity of the inner ear end organs on 3D-FLAIR images 4 hours post-contrast (144).

On a bithermal caloric test, the lateral semicircular canal showed high signal intensity on 3D-FLAIR images 4 hours post-contrast in 75% of patients with abnormal canal paresis (144). Although the cochlea showed high signal intensity on 3D-FLAIR images 4 hours post-contrast in 75% of patients with hearing loss, the vestibulo-cochlear nerve showed enhancement in post-contrast T1-weighted images in only 33% of patients with hearing loss (144).

MRI enhancement of inner-ear structures and cranial nerves following varicella-zoster virus infection typically lasts less than 6 weeks but may apparently persist for at least 6 years in rare patients (51; 245). Such protracted resolution of MRI enhancement warrants consideration of other causes of labyrinthitis or intracanalicular acoustic neuroma (51; 245).

Some studies have found no clear prognostic indicators based on MRI images (134); others have suggested that enhancement limited to the geniculate ganglion and to the labyrinthine segment of the facial nerve indicates a good prognosis, whereas widespread enhancement of the facial nerve is often associated with a poor prognosis for facial nerve recovery (19).

Simultaneous postcontrast enhancement of the labyrinth and intratemporal cranial nerves can occur with viral or bacterial labyrinthitis, including herpes zoster oticus and Bell palsy (139; 134; 195; 212; 117; 210; 33). Abnormal contrast enhancement of inner ear soft-tissue structures may also occur with facial nerve or cochlear tumors, with posttraumatic facial paresis, following temporal bone surgery, and with rare arteriovenous anomalies of the area, although the intensity and distribution of enhancement can often differentiate such lesions (147; 195; 234). MRI findings in herpes zoster oticus can mimic intracanalicular acoustic neuroma (08; 91; 147b; 51; 80).

Management

The treatment of herpes zoster oticus remains somewhat controversial, particularly in terms of the role of corticosteroids in combination with antivirals (41; 47; 60; 61).

Antivirals. Acyclovir facilitates healing of varicella-zoster virus infections, has a low rate of toxicity, and may help prevent postherpetic neuralgia (70). In particular, immunocompromised patients with varicella-zoster virus infection should be promptly treated with high-dose acyclovir (1500 mg/m2 per day for children and 30 mg/kg per day for adults in three divided doses) (70). Acyclovir is also indicated in rare patients who develop associated central nervous system findings (101).

Early treatment with acyclovir (eg, 800 mg five times daily in adults) (and possibly steroids such as prednisone 60 mg daily for 3 to 5 days) may improve the prognosis of facial palsy and also possibly hearing loss associated with herpes zoster oticus, but adequate randomized, controlled trial data are lacking (49; 112; 218; 02; 167; 64; 131; 130; 211; 82; 217; 221; 162; 60).

Treatment is typically given for 1 to 2 weeks. Longer treatment does not apparently improve the outcome (236).

An alternative to acyclovir is famciclovir (500 mg three times daily in adults), although the latter is more toxic. In a retrospective study, recovery rates were higher following treatment with famciclovir plus steroids than with acyclovir plus steroids, especially in patients without hypertension and diabetes mellitus (126).

Corticosteroids. The addition of steroids is controversial, and randomized trials have suggested that corticosteroid therapy does not apparently improve the outcome (236), although restrospective observational data suggest a benefit of high-dose corticosteroid therapy in conjunction with antiviral therapy (61). Unfortunately, there have been few randomized controlled trials to assess the utility of corticosteroids as an adjuvant to antiviral therapy in herpes zoster oticus (222).

In a systematic review involving collectively 350 patients, the authors concluded that intratympanic corticosteroids have the potential to reduce nonrecovery in patients with Ramsay Hunt syndrome (60). However, the methods of intratympanic corticosteroid treatment varied among included studies. The timing of intratympanic corticosteroid treatment varied from daily for 10 days to once or twice a week. Patients treated with a combination of systemic corticosteroids and intratympanic corticosteroids were likely to achieve recovery of normal facial function. A subgroup analysis revealed that daily intratympanic corticosteroids reduced nonrecovery whereas non-daily intratympanic corticosteroids did not. Further well-designed randomized controlled trials are needed to confirm the effectiveness of intratympanic corticosteroids and establish an optimal treatment regimen.

Eye protection. With severe peripheral facial paresis, a major consideration is ensuring protection of the ipsilateral eye, especially when the eyelids do not close fully volitionally or during sleep. Such patients have poor reflex eye closure and may have inadequate lacrimation and are, therefore, susceptible to corneal abrasions and infections. Ophthalmic lubricant should be placed in the affected eye (a viscous product is preferred especially at night). The eyelid can be held closed with tape or an inflexible convex eye shield can be placed over the eye for protection. A hydrophilic soft contact lens has also been used to bandage the cornea in such cases to prevent injury (239). It is potentially dangerous to use flexible eye patches or to attempt to keep the lid closed by applying a dressing held with pressure and tape over the eye; in such cases, the eye often opens under cover and is abraded by the patch or dressing.

Vertigo. Vestibular sedatives may help alleviate acute vertigo, especially during the first few days, when the eighth nerve is involved. Anecdotal reports are also available of vestibular rehabilitation training for peripheral vestibular dysfunction following herpes zoster oticus (75).

Special considerations

Pregnancy

Varicella infection in women during pregnancy can result in maternal varicella pneumonia, intrauterine death, or severe congenital disease (congenital or fetal varicella syndrome). Congenital varicella syndrome is characterized by limb hypoplasia, skin abnormalities in dermatomal distribution, ocular abnormalities, encephalitis, microcephaly, cerebral malformations, epilepsy, mental retardation, and low birth weight.

Congenital varicella syndrome occurs in approximately 0.4% to 2% of infants born to women infected with varicella-zoster virus during the first two trimesters of pregnancy (55; 187; 92). Subclinical maternal infection can be cause congenital varicella syndrome (170).