Weller has recounted the progression of the medical knowledge about VZV.¹ In 1875, Steiner demonstrated varicella to be an infectious disease by inoculating volunteers with vesicle fluid from an individual with varicella.² Von Bokay noted the relationship between HZ and chickenpox in 1892, in that children exposed to individuals with HZ could develop chickenpox.³ Garland suggested in 1943 that HZ might be due to reactivation of VZV acquired earlier in life.⁴ In 1952, after 11 years of laboratory research, Weller and Stoddard succeeded in isolating and propagating VZV in vitro from vesicle fluid.⁵ Restriction endonuclease patterns of isolates from a patient with varicella and subsequent zoster showed that the viral genomes are identical.⁶ Unless epidemiologically related, VZV isolates differ slightly from one another; some variation also occurs after serial passage in vitro.^6,7 The complete sequence of the VZV genome was determined in 1986,⁸ and the first genetically engineered VZV mutant was constructed in 1987.⁹ A versatile method for site-directed mutagenesis of the virus was performed using DNA cosmids to produce infectious virus in 1993.¹⁰ The varicella virus vaccine became available in the United States in 1995.

Humans are the only natural host, and the virus grows fastidiously in primary or continuous human cell lines (including human diploid fibroblasts, embryonic lung fibroblasts, foreskin fibroblasts, and primary keratinocytes, as well as melanoma and schwannoma cells) and only poorly in nonhuman lines (guinea pig embryo and monkey kidney cells).¹² The vesicle fluid from patients with both chickenpox and HZ induces similar distinct focal cytopathic changes in cultured cells.¹³ VZV is a cell-associated heat-labile virus that spreads from cell to cell by direct contact. VZV is not processed and released from cultured cells as efficiently as it is from cells in vivo, so the particle-to-infectivity ratio is quite high in cell cultures. The inability to produce high titers of cell-free virus has hampered studies of the biology and chemistry of the VZV.¹⁴

The varicella-zoster virion has a diameter of 150 to 200 nm, with a lipid envelope bearing the host cell membrane and glycoprotein spikes. The viral envelope aids in both attachment and penetration of the virus. The central core of the virus contains an icosahedral (5:3:2 symmetry) nucleocapsid composed of 162 capsomeres, within which is the viral genome. The amorphous protein-filled space between the nucleocapsid and the envelope is known as the tegument; it contains several proteins, including the immediate-early proteins and the late proteins. The specific antigenicity of the virus is conferred by the protein capsid and the glycoprotein peptomers; the host antibodies are formed against these antigens. Glycoproteins confer antigenicity through their incorporation into the viral external membrane; they are the primary contact point between the virus and the cell as infection is initiated. The glycoproteins of VZV have been examined in the greatest detail because these envelope proteins are essential to understanding the VZV-specific immunity and the production of improved vaccines.¹⁵

The virion attaches to cells by binding to heparan sulfate proteoglycans, followed by binding to a mannose-6-phosphate receptor.¹⁶ After attachment, viral glycoproteins fuse with cell membranes and the virion penetrates the cell. After penetration, the virion is uncoated and the nucleocapsid is transported to the nucleus; tegument proteins may also be transported to the nucleus, where they initiate transcription of viral genomes. Initially, the viral immediate-early genes are expressed, which activate expression of the viral early genes. The latter encode proteins important for replication of viral DNA (e.g., thymidine kinase, polymerase, major DNA-binding protein). Finally, the late viral proteins are expressed. These are structural proteins that compose the viral nucleocapsid, tegument, and envelope. Viral DNA is packaged into nucleocapsids and transported out of the nucleus, where it acquires a glycoprotein envelope (either from the nuclear membrane or cytoplasmic vacuoles), and then virions are released from the cell.¹²

The inability to produce high titers of cell-free virus has limited the ability to assign genes formally to individual kinetic classes in the replication cycle and to study viral DNA replication. We know little about the mechanism by which VZV establishes latency in the central nervous system (CNS) and the function of viral genes in this process. Even less is known about the pathways responsible for the reactivation of virus from latency.

The varicella zoster virion is among the smallest of the herpesvirus genomes. The double-stranded genome is linear in the virus particle but circularizes within the infected cell. The viral genome consists of a double-stranded linear DNA for approximately 125,000 base pairs,¹⁷ capable of encoding about 75 proteins,¹⁴ all but 5 of which have homologues with HSV-1. The genome is organized into two portions, a unique long region and a smaller unique short region. Each of these regions is surrounded by inverted repeat elements. During DNA replication, the unique short region and its repeat become inverted, producing two isomeric forms of VZV DNA.¹⁴ Each end of the viral genome contains a single unpaired nucleotide that can base pair to form a circular molecule during replication of the genome. Thus, approximately half of the molecules contain the unique short region in one orientation and half in the other. In contrast, HSV has two regions that become inverted and result in four isomeric forms. Individual virions of VZV contain one or the other isomer form, and the two forms are equally infectious. The genome contains five repeat elements that can be analyzed by restriction endonucleases, allowing a distinction between different strains of VZV.¹² The viral DNA is integrated into the host's cellular machinery; this aids in replication but also avoids immune surveillance and antiviral drug eradication.

The area controlling DNA replication in the VZV genome has been mapped to a 45-base pair palindrome in the short repeat region.¹⁸ This is homologous to corresponding regions of HSV, implying that the enzymes necessary for DNA replication in these viruses share similar properties.¹⁴ There are also several other regions with sequences similar to HSV; this has led to accurate predictions of localization of important viral genes based on the more advanced knowledge of the HSV genome.⁸ There are at least six glycoproteins in VZV, and all resemble HSV glycoproteins. These glycoproteins were originally termed gpI to gpVI but have now been reclassified to correspond to their HSV homologues: gE (gpI), gB (gpII), gH (gpIII), gI (gpIV), gC (gpV), and gL (gpVI). Because these glycoproteins are potent inducers of antibody, they may be suitable for incorporation into future subunit vaccines.¹⁴

Molecular studies, specifically restriction endonuclease analysis of viral DNA, confirmed that HZ is caused by reactivation of latent VZV initially causing varicella in the host.⁶ Different viral isolates show slight variations in the mobility of certain DNA fragments.¹⁹ Isolates from a single outbreak of varicella do not show differences in DNA mobility, implying that the genome structure is relatively stable.¹⁴ The small variability that exists between epidemiologically unrelated isolates arises most often from gains or losses in the number of small repeat elements in the DNA.¹⁴

VARICELLA

Varicella (chickenpox) is a common childhood illness, with more than 90% of the population developing clinical or serologic infection by adolescence²⁰ and almost 100% by age 60.²¹ Varicella is highly communicable, with an incubation period of approximately 14 days. The incidence of varicella in the United States approximates the annual birth rate. Varicella is spread by droplet or airborne transmission and is highly contagious. The virus infects the host through the conjunctiva and/or mucosa of the upper respiratory tract. Chickenpox can be acquired by contact with either varicella or HZ lesions or by respiratory inhalation. Patients are contagious for 2 days before onset of the rash and then until all lesions have crusted. Chickenpox generally confers lifelong protection against a subsequent attack, but second infections have been documented. Asymptomatic reinfection (documented by a rise in antibody titer) may occur after exposure to varicella.²² In immunosuppressed patients or in patients whose initial infection was mild or subclinical, symptomatic reinfection may occur.²³

The epidemiology of varicella varies remarkably between temperate and tropical climates. In tropical areas, varicella typically occurs among older persons. The distinctive age distribution of varicella seen in tropical climates is agent-specific rather than attributable to nonspecific factors, such as absence of crowding indoors in the winter.²⁴ Epidemiology studies done by Preblud²⁰ in the late 1970s and early 1980s indicated that there was significant morbidity (bacterial infection, encephalitis, pneumonia) and death caused by varicella in healthy individuals. Epidemiologic studies by the Centers for Disease Control indicate that up to 10,000 hospitalizations and 100 deaths per year occur in the United States from complications of varicella infection.

HERPES ZOSTER

Herpes zoster occurs during the lifetime of 10% to 20% of all persons.¹⁴ The most striking feature of the epidemiology of HZ is the increase in incidence observed with increasing age. The incidence is approximately 131 per 100,000 person-years for the United States white population.²⁵ The annual incidence of HZ ranges from only 0.4 to 1.6 cases per 1,000 otherwise healthy people younger than 20 years to 4.5 to 11 cases per 1,000 among those aged 80 years or older.^25,26 The lifetime incidence of HZ among African-Americans appears to be only half that reported by whites.²⁷ The great majority of dermatologic HZ cases results from reactivation of latent virus; there is an absence of any increase in HZ concurrent with epidemics of varicella.²⁸ There are reports of small epidemics of HZ, suggesting that it could be exogenously acquired or reactivated, although these reports of acquisition can usually be explained by chance.²⁹ Both silent reactivation (documented by a rise in antibody titer) and clinically apparent reinfection with a different VZV have been well documented.^22,23

Herpes zoster can occur despite substantial antibody titer, so humoral antibody is not a major determinant.³⁰ Children with immune deficiencies limited to defects in antibody synthesis do not develop severe or recurrent varicella or HZ. The known risk factors for developing HZ relate to the status of cell-mediated immunity (CMI) to VZV.^24,25 Thus, the incidence of HZ increases with age (because VZV-specific CMI and CMI in general decline with aging), with immunosuppression (such as HIV), or immunosuppressive therapy, and after primary infection in utero or early infancy, when the normal immune response is decreased. In vitro measurement of lymphocytic response to VZV antigen shows a significant decline with advanced age, with development of lymphoproliferative malignancies, and with the initiation of immunosuppressive treatment.³¹ The increased risk of HZ among persons infected with HIV requires consideration of HIV infection in any patient with HZ who is younger than 45 years or is in a recognized risk group for AIDS.³² The appearance of HZ in HIV-infected individuals or in individuals with AIDS-related complex appears to predict an increased risk for subsequent development of AIDS.³³

HZ is more likely to be severe and prolonged and to lead to dissemination in an immunosuppressed patient.³⁴ Dissemination implies a viremia. A few vesicles outside the primary dermatome are common even in normal patients and do not signify clinically important dissemination. Significant dissemination in an immunosuppressed patient is more likely to be accompanied by visceral or neurologic infection, both of which substantially increase the morbidity and mortality of HZ infection. Immunosuppressed patients may have chronic prolonged infection with sustained periods of new lesion formation, a failure of existing lesions to heal in the absence of antiviral therapy, and persistent CNS infection with progressive encephalopathy.³⁵

The frequency of dermatomes involved in HZ corresponds to the centripetal distribution of the initial varicella lesions, suggesting that the latency arises from contiguous spread of the virus during varicella from infected skin cells to sensory nerve endings with subsequent ascent to the ganglia. An alternative explanation, however, is that the ganglia are infected hematogenously during the viremic phase of varicella and that the frequency of dermatome involvement in HZ reflects the ganglia most often exposed to reactivating stimuli.¹⁴ At autopsy, attempts to recover latent VZV in culture from previously infected ganglia have been unsuccessful,⁴⁰ although the virus has been recovered when active VZV ganglion infection existed in the corresponding dermatome at the time of death.^41,42

During varicella infection, the VZV replicates efficiently initially in both neural and non-neural (satellite) ganglion cells. The total amount of latent VZV is low; therefore, conventional methods to detect latent VZV have proved limited.⁴³ Because the amount of latent VZV per cell is also very low, the question of which cell type is involved in VZV latency could not be conclusively settled by the use of traditional in situ hybridization studies. Investigators have demonstrated VZV by in situ hybridization in either neurons⁴⁴ or non-neuronal satellite cells.^45,46 By using a combination of in situ polymerase chain reaction (PCR) and in situ hybridization, latent VZV DNA has been identified in the nucleus of the neurons only⁴³ or predominantly neurons.⁴⁷

Although animal models of VZV infection are available, reactivation-associated disease does not occur in these systems. An in vitro system has been developed using human fetal ganglia and has shown that both neuronal and non-neuronal cells can support VZV growth, but that the virus can reactivate only from a combination of the two.¹¹ Originally it was thought that this ability of VZV to reactivate in satellite cells, spread to other satellite cells, and infect neuronal cells might account for the more widespread and destructive ganglionitis seen with VZV as compared with HSV.

Only a very small percentage of ganglion cells (0.01% to 0.15%) exhibit VZV transcripts.⁴⁶ Several VZV genes seem to be expressed during latency in the human trigeminal ganglia,⁴⁸ although homologues of the HSV latency-associated transcripts have not been detected in VZV either during productive infection or during latency.¹² Evidence for transcriptional activity of the latent genome consists of detection of gene products and at least five RNA transcripts.^45,48 Mechanisms that limit transcription, maintain latency, or induce reactivation are unknown. The cellular factors that influence the switch of viral gene expression from latency to lytic infection are unclear but are likely to involve a dynamic interaction with the immune system that allows the expression of pivotal viral mediators of reactivation.

The immune response does not prevent reactivation totally; more likely, subclinical reactivations occur in both immunocompromised and immunocompetent patients^49–51 throughout life, and VZV-specific CMI acts to limit the spread of VZV within the ganglion, limit subsequent spread antegrade to the skin, and limit viral replication within the cutaneous lesions of HZ. Increases in VZV-specific immunity in immunocompetent patients in the absence of HZ suggest that reactivation can be limited to a subclinical event.⁵¹

The small dose of infectious virus released with HZ is immediately contained by circulating antibody or CMI and may never be evident clinically on the skin or mucous membrane. Both endogenous reactivation and exogenous re-exposure caused by either an association with a patient with varicella or with HZ^22,23,52 or by VZV vaccine are important mechanisms in maintaining enhanced T cells ready to respond to endogenous VZV reactivation.⁵³ This potential for a boost in T-cell response to VZV persists in the elderly.⁵⁴ VZV viremia in asymptomatic immunocompromised patients who subsequently have boosts in VZV immunity⁵⁰ and healthy subjects who have boosts in VZV immunity after dermatomal pain without cutaneous symptoms (zoster sine herpete),⁵⁵ and neurologic VZV-produced disease or systemic VZV-produced disease without rash are examples of contained episodes of HZ.⁵⁶ They can be confirmed by boosts in T-cell response and also by peripheral blood mononuclear cells with incomplete VZV DNA.⁵⁷ The capacity of the immune response to limit spread of the virus is a major determinant of the early and late morbidity of HZ, assuming that the acute pain of HZ and postherpetic neuralgia result from tissue destruction and neuronal changes in the ganglion.²⁸

PRIMARY INFECTION

Chickenpox infection generally follows a benign course in normal children. Complications are more common and severe in immunosuppressed patients, neonates, and adults. The major complications are bacterial infection of the skin, severe skin disease (in immunosuppressed persons), neurologic complications (encephalitis), and pneumonia.^20,59 The first viremia of varicella occurs a few days after infection, during which time viral replication has occurred within a localized area near the site of infection (usually the oral mucosa). This first viremia spreads the virus through the human host, and further cycles of replication occur. About 1 week later, a second transitory viremia occurs as circulating mononuclear cells and T lymphocytes travel through the reticular endothelial system and become infected by VZV.

The melanocytes may be an intermediary cell substrate in the viral transit process from the capillary into the basal layer of the skin. The epidermal cells are eventually involved in a process that has been called balloon degeneration because of the formation of large multinucleated giant cells. The nuclei are clustered in the center of each polykaryocyte; many of the nuclei contain eosin-staining viral inclusions. As the cells degenerate in the virally infected foci within the epidermis, an exudate develops and separates the dying cells to form a space filled with fluid. The appearance of this vesicular rash heralds the end of the incubation period.⁵⁸

Ocular complications of varicella include eyelid lesions, conjunctival vesicles, dendritic epithelial keratitis, stromal keratitis, neurotrophic keratopathy, iritis, internal ophthalmoplegia, extraocular muscle palsies, cataract, chorioretinitis, and optic neuritis.⁶⁰ The congenital varicella syndrome results most frequently from exposure to varicella during the first or second trimester and may be associated with ocular abnormalities such as chorioretinitis, optic atrophy or hypoplasia, congenital cataract, microphthalmos, and Horner's syndrome.⁶¹

RECURRENCE OF VZV

Patients with HZ have a prodrome of fever, malaise, headache, and dysesthesia for 1 to 4 days before the development of the cutaneous lesions. The exanthem consists of grouped vesicles usually involving one but occasionally up to three adjacent dermatomes. The vesicles become pustular and occasionally hemorrhagic, with evolution to crusts in 7 to 10 days. Occasionally an attack is aborted before skin lesions appear but can be confirmed serologically (zoster sine herpete).⁶² VZV reactivation in the immunosuppressed host is usually associated with a more extensive and severe local rash than in immunocompetent individuals and is often accompanied by cell-associated viremia.⁶³ The virus can be transmitted during the viremia to distant cutaneous or extracutaneous sites in scattered foci in 17% to 35% of immunocompetent HZ patients. Hematogenous spread of virus probably accounts for the appearance of a few vesicles in skin areas remote from the affected dermatome.

Disseminated HZ is defined as more than 20 vesicles outside the primary and immediately adjacent dermatome. Severely immunocompromised patients with HZ have a risk of dissemination up to 40%. In 10% of these high-risk patients, cutaneous dissemination is followed by progressive visceral involvement, particularly of the lungs, liver, and brain.^64,65 Visceral dissemination may also follow in some immunocompromised patients who have no signs of cutaneous HZ.⁶⁶ It may occur and recur frequently in otherwise clinically disease-free HIV-infected patients.⁶⁷ Chronic VZV reactivation may occur in severely immunocompromised patients, with persistent viral replication at skin sites and episodes of viremia lasting several months. Recurrent VZV lesions in patients with AIDS are characterized by epidermal hyperplasia and massive hyperkeratosis with multinucleated giant cells and necrotic acanthotic keratinocytes.⁶⁸

There are diverse neurologic complications associated with HZ, including motor neuropathies of the cranial and peripheral nervous system, encephalitis, meningoencephalitis, myelitis, and Guillain-Barré syndrome.⁶⁹ A unique cerebral vasculopathy with a mortality of approximately 25% has been recognized in which contralateral hemiparesis develops weeks to months after herpes zoster ophthalmicus (HZO).⁷⁰ Angiographic and autopsy studies confirm a cerebral vasculopathy with thrombosis and a granulomatous vasculitis involving large and small arteries, arterioles, and venules.⁷¹ Immunofluorescence and electron microscopy studies demonstrate the virus in the smooth muscle cells of the vascular media but not in the endothelium.

Involvement of the ophthalmic division of the trigeminal nerve, HZO, is disproportionately frequent and severe. It was richly described by Hutchinson in 1865.⁷² The ophthalmic division of the trigeminal nerve, by way of the frontal, nasociliary, supraorbital, and supratrochlear branches, provides sensory innervation not only to the skin of the forehead, temple, palate, and the tip of the nose, but also to the eye. The frontal branch within the ophthalmic division of the fifth nerve is most commonly involved. Rarely, all three branches of the ophthalmic division are affected simultaneously; even rarer is the simultaneous involvement of the ophthalmic, maxillary, and mandibular branches of the trigeminal nerve.

The most serious consequences are evident when the nasociliary nerve of the ophthalmic division is involved; this often signifies intraocular involvement. Subsequently, 50% of patients with HZO develop ocular complications.^73,74

There are diverse potential ocular complications of HZO because there are numerous pathogenic mechanisms of viral replication and spread and because there are varying contributions of vasculitis, neuritis, and the immunologic and host inflammatory granulomatous reactions to infection.^73,75,76 A chronic persistence of ocular symptoms is observed in approximately 29% of affected patients⁷⁷; rarely, enucleation is required (Table 1).^78,79 The inflammatory reaction can occur in any portion of the eye or adnexal tissue. The pathogenesis of the varied ocular complications of HZO is not fully known, although histopathologic studies demonstrate significant perivascular and perineural inflammation in ocular tissues.^78,79 VZV can be isolated from the eye during acute HZO but usually not in delayed or chronic ocular disease. The virus has also been recovered, however, from the delayed pseudodendrites.⁸⁰

TABLE ONE. Forms and Incidence of Herpes Zoster Keratitis

The most common infectious or immune complications of HZO include cicatricial lid retraction or loss, paralytic ptosis, conjunctivitis, scleritis, episcleritis, keratitis, iridocyclitis, secondary glaucoma, cataract, Horner's syndrome, Argyll-Robertson pupil, glaucoma, retinitis, choroiditis, optic neuritis, optic atrophy, retrobulbar neuritis, exophthalmos, extraocular muscle palsies, postherpetic neuralgia, and orbital apex and inflammatory syndromes (Table 2).^73,74,81

TABLE TWO. Incidence of Complications of Herpes Zoster Ophthalmicus

HZO IN IMMUNOCOMPROMISED PATIENTS

In immunocompromised patients, VZV infections have long been a major complication. Immunosuppressed organ transplant recipients and immune-deficient patients with cancer, leukemia, and AIDS are all at increased risk of HZ. In those with solid tumors, the incidence is 1% to 3%, in those with nonHodgkin's lymphoma 7% to 9%, and in those with Hodgkin's disease 13% to 15%, of which up to 30% disseminate.⁸² Depending on the severity of impairment of CMI, cutaneous dissemination occurs in 6% to 26% of patients within 4 to 11 days after localized HZ appears; half of these patients will have ocular, visceral, or neurologic involvement. Because reactivation of VZV in immunocompetent individuals younger than 45 is uncommon, the presence of HZ in patients younger than 45 must alert the physician to the possible coexistence of HIV infection. In a prospective study of HZ patients younger than 45, 75% had HIV risk factors or altered T-cell subsets.⁸³ The majority of HZO cases in HIV patients are characterized by severe and enduring cutaneous lesions, epithelial and stromal keratitis, and anterior uveitis.⁸⁴ In a group of 19 HIV-HZO patients with a mean age of 28 years, 89% developed punctate keratitis and anterior stromal infiltrates, 53% had iritis, and 42% contracted postherpetic neuralgia.

VZV can also involve the posterior segment of the eye in immunocompromised patients, manifesting as distinct clinical patterns of disease. It has been established as one cause of acute retinal necrosis (ARN) syndrome and the etiologic agent of progressive outer retinal necrosis (PORN) syndrome.^85–87 Retinitis develops either from reactivation of latent VZV that was previously acquired or during the course of primary infection. Visual consequences may be mild or severe, with retinal detachment a major feature. Medical and surgical management is only partially effective in preventing vision loss. The pathogenesis of the retinitis is poorly understood; the site of latency, the factors responsible for VZV reactivation, and the manner by which the virus gains access to the eye remain unclear.⁸⁸ There are disparate clinical appearances of VZV retinal infection in ARN and PORN syndromes, probably reflecting the severely altered host immune defenses in those affected with PORN syndrome.⁸⁹ ARN is usually seen in patients without AIDS.

The presence of peripheral retinal perivasculitis and sheathing was found to raise the index of suspicion greatly for coexisting HIV infection in HZO patients.⁹⁰ In general, this category of patients with combined infection requires prolonged therapy for chronic morbidity of disease.⁹¹ Prolonged therapy promotes resistant virus, with the suggestion that aggressive treatment be given early and that long-term oral suppression should be avoided.⁹²

In contradistinction to these studies was a report indicating a relatively low incidence and benign nature of the stromal keratitis, iritis, and postherpetic neuralgia in a cohort of patients with HIV infection, but an increase in chronic infectious dendritic keratitis, retinitis, and CNS disease, all probably consequences of active viral replication.⁹³ The absence of some complications may reflect the absence of an effective immune response, because many of these complications are purported to be immune in etiology. Features of HZO in HIV patients include an increasingly lower age at onset, skin eruption in multiple dermatomes, ocular disease sine herpete, the PORN syndrome, chronic infectious dendrites, and serious neurologic disease.⁹³

PRIMARY INFECTION

Despite the incubation period of 10 to 21 days after viral inoculation, serum antibodies to VZV are usually not detected until 1 to 3 days after the appearance of the varicella exanthem. Acute primary VZV infection in both healthy and immunocompromised persons is associated with the rapid induction of IgG, IgM, and IgA antibodies directed against several viral proteins; antibodies of all the immunoglobin classes appear simultaneously in most cases. The initial IgG antibodies appear to be specific for viral glycoprotein targets on both the envelope of the virus and the surface of infected cells and also against nucleocapsid protein. IgM is specific for polypeptides, and IgA is directed against three VZV-infected cell proteins.⁹⁷ IgG, IgM, and IgA antibodies appear 2 to 5 days after the onset of varicella rash and reach maximum titer during the next 2 to 3 weeks. Thereafter, IgG titers decline but persist at low levels for many years, whereas IgA and IgM antibodies usually are not detectable 12 months after the illness.^30,98,99 Subsequent household exposure to HZ or varicella results in no clinical disease or change in IgG levels, but IgA titers increased two to four times within 2 to 4 weeks before returning to baseline. This fluctuation in IgA titers indicated a subclinical reinfection. The initial IgG antibodies to VZV are predominantly in the IgG3 subclass.¹⁰⁰ Neutralizing antibodies are predominantly IgM, but IgG antibodies have also been shown to neutralize the virus.¹⁰¹

The CMI response in varicella begins when macrophages produce interleukin-1 and present VZV antigen to helper T cells. The helper T cells proliferate and produce interleukin-2, which causes further stimulation of helper T cells and suppressor T cells. These T cells release interferon-γ; there is nonspecific induction of interferon-α, and both have an antiviral effect on VZV replication. Similar humoral and cell-mediated responses have been noted in individuals vaccinated with the vaccine. Memory T-cell responses and cytotoxic T-cell responses are elicited, although they are somewhat less robust in adults than in children. Antibody and CMI responses elicited by the vaccine are boosted by exposure to natural varicella.¹⁰² The infected cells are lysed by suppressor T cells along with natural killer cells, which are granular lymphocytes with no immunologic specificity. Antibody-dependent cellular cytotoxicity is also active against VZV in concert with T cells. Interferon enhances the activity of natural killer cells. The severity of infection with VZV is most strongly correlated with the depression of CMI, although antibody to many antigens occurs during the immune response to VZV. Cell-mediated responses to VZV are nonspecific in the nave host or are mediated by antigen-specific T lymphocytes that are elicited during primary exposure to the virus. Virus-specific cellular immunity is important for controlling viral replication. Viral virulence factors are likely to be important for the establishment of latent VZV infection in dorsal root ganglia, but host factors determine whether the individual with latent infection develops symptomatic VZV reactivation.⁹⁹

RECURRENCE OF VZV

The immune response to HZ is somewhat different. During the reactivation as HZ in immunocompetent patients, IgG, IgM, and IgA antibodies appear more rapidly and reach higher titers than during chickenpox, confirming an amnestic response.¹⁰³ Although a rise in IgM is usually associated with primary exposure to an infectious agent, it occurs with recurrences of VZV. Endogenous reinforcement of the immune responses also occurs as a result of transient replication of the latent virus intermittently, as evidenced by specific IgM antibody rises in asymptomatic patients.⁵² The IgG1 antibodies are predominant in HZ, compared with IgG3 in varicella.

A very robust VZV-specific T-cell proliferation coincides with the appearance of the cutaneous rash in HZ. Immunocompromised subjects have a more sluggish and less robust CMI response than healthy patients with HZ. The resolution of HZ is accompanied by local production of interferon-α. The enhanced CMI response to VZV persists for many years; this explains why second episodes are rare. When a previously infected individual is immunosuppressed, he or she still typically develops HZ rather than varicella. Persistence of immunity may be related to periodic exogenous re-exposure to the virus during epidemics of varicella, exposure to HZ, and/or subclinical reactivation of endogenous latent virus. As in varicella, the role of cellular immunity appears to be dominant in determining viral reactivations, as evidenced by a clonal expansion of T cells. Helper T cells are pivotal in VZV immunity because they carry immunologic memory, stimulate B cells to synthesize antibody, and promote the function of cytotoxic lymphocytes and natural killer cells.

Weigle and Grose⁵² have studied the temporal pattern of the antibody response to molecularly defined VZV protein during primary varicella infection, quiescence, subclinical reinfection, and overt HZ. In chickenpox, the first antibodies produced are directed against the virion envelope structural glycoprotein, gp66, gp118, gp98, and gp62, and the nucleocapsid protein, p155. One to 2 months later, antibodies to a wider variety of viral proteins appear. Because most of these antibodies become undetectable by 3 to 4 months after infection, their presence indicates recent VZV infection. Antibodies to the immunodominant gp66, gp118, and p155, however, are still present for years after infection, and therefore constitute excellent markers for previous VZV infection; they also remain unchanged during quiescence. During subclinical reinfection after exogenous exposure to VZV, there is an overall rise in glycoprotein antibody levels, with the greatest rise in gp98 and gp62, and their subsequent decrease over the next 2 years. This subclinical response is more restricted and of lesser magnitude than that noted after endogenous reactivation of herpes zoster. Immunocompromised children with varicella develop antibodies to gp66 and gp118 similar to healthy children but have lower antibody levels to gp98 and gp62 than normals.

In contrast to the varicella immune patterns, HZ is characterized by the presence of antibodies against virtually all VZV glycoproteins within a week, if not at onset. In addition to this quantitative and temporal difference, another striking difference between chickenpox and HZ is the rapid appearance of antibody to p32 in the latter but not in the former. The p32 humoral response abates over the ensuing weeks to months, but its presence in serum indicates recent HZ.

Instead of a static relationship, the situation is dynamic, with reinfection and reactivation occurring as a covert process in the absence of overt clinical signs. Although neutralizing serum antibody has been shown to prevent or attenuate varicella infection, the cellular immune response is responsible for final clearance of the virus, prevention of dissemination of disease, and perhaps maintenance of the latent state of the virus in infected ganglia.

T-lymphocyte-mediated immunity is critical in preserving the balance between the host and the virus. In contrast, the susceptibility of immunocompromised and elderly individuals to VZV reactivation does not correlate with decreasing titer of VZV IgG antibodies. Severe, prolonged suppression of cellular immunity is accompanied by a high incidence of symptomatic VZV reactivation; cellassociated VZV viremia, with life-threatening dissemination, becomes frequent. Healthy and immunocompromised individuals who develop HZ usually have a significant recovery of VZV-specific T-lymphocyte responses; the number of circulating T lymphocytes that recognize VZV antigens increases immediately as a consequence of the re-exposure to viral antigens in vivo. Enhanced CMI to VZV after HZ usually persists for a prolonged period.⁹⁹ This host immune response is not affected by acyclovir or prednisone.¹⁰⁴

VZV particles have been demonstrated in acute disease in CNS tissues, the trigeminal ganglion and its nerve axons, and the arterial walls of ocular and CNS tissues by electron microscopy and immunofluorescence.¹⁰⁷ The demonstration of afferent ganglionic fibers to intra- and extracranial blood vessels provides an explanation for the interaxonal spread of virus and the resultant granulomatous arteritis, which may be a viral or viral-induced immunopathologic event.¹⁰⁸ Granulomatous arteritis can develop during both primary infection and VZV reactivation. VZV has been cultured from the ganglia only in rare isolated cases of acute HZ.¹⁰⁹ VZV cannot usually be isolated in cultures from the explanted dorsal root ganglia, although viral RNA transcripts and DNA have been reported.¹¹⁰ Adjacent ganglia often show similar changes, and adjacent segments of the spinal cord or brain stem are involved with direct viral invasion. Even if the immune response can abort the cutaneous lesions of HZ, there may still be necrosis and an inflammatory response at the ganglion (zoster sine herpete). If the amnestic response is delayed or deficient, then both more severe local disease and cutaneous dissemination are seen.

The virus travels down the sensory nerve to the skin, producing demyelination, cellular infiltration, and subsequent scarring with fibrosis of peripheral nerves. Granulomatous inflammation is seen in the peripheral nerves. Demyelination is seen in areas of mononuclear cell infiltration and microglia proliferation. Electron microscopy and fluorescent antibody studies confirm the presence of virus and viral antigen in neuronal and satellite cells and within the corresponding sensory nerves before the development of cutaneous lesions. VZV-specific DNA has been found in circulating mononuclear cells during HZ¹¹¹ and by PCR in mononuclear cells even in the absence of recent HZ.¹¹²

The skin lesions of HZ develop because of retrograde transport of VZV from the ganglia to the skin. HZ occurs with the greatest frequency in dermatomes in which the rash of varicella is densest, presumably related to a large inoculum and a corresponding increase in infected neurons at the time of initial seeding. The lesions of HZ typically progress from discrete patches of erythema to grouped vesicles. Later, the vesicle fluid becomes cloudy from leukocytes, degenerated cells, and fibrin. These pustulate and crust in 7 to 10 days but may take a month to heal, often with anesthetic scars, changes in pigmentation, and pain. Most lesions heal without scarring, except in patients prone to keloids and in patients who develop secondary bacterial infections.

The histopathologic findings of VZV in the skin are identical with chickenpox and HZ. Vesicles involve the corium and the dermis. Replication within the skin is associated with mononuclear infiltration and multinucleated giant cells at the subepithelial and dermal tissue level. As the virus replication progresses, the epithelial cells undergo degenerative changes characterized by ballooning, with the subsequent appearance of multinucleated giant cells resulting from epithelial cell fusion and prominent eosinophilic intranuclear inclusions. The intranuclear inclusions have been labeled as Cowdry type A intranuclear inclusions ( Lipschütz bodies) indicative of HZ infection. VZV produces a spotty angiitis of sufficient intensity to cause ischemia, necrosis, and hemorrhage in the dermis; this may cause skin scars and other manifestations of infarctive necrosis of tissues.

The ocular complications of HZO are related to direct viral invasion combined with host inflammatory, immune, and vascular reactions. Rarely, enucleation is required, and the histopathologic findings of HZO have been detailed.^78,79,113 The early histopathology of HZO reveals episcleral and corneal round cell inflammation, macrophages in the corneal endothelium, and nongranulomatous (lymphocytes, plasma cells) infiltration of the iris, trabecular meshwork, ciliary body, retinal vasculature and nerves, and optic nerve.^79,114 A plasma and lymphocytic infiltration of the posterior ciliary artery and nerves in the retrobulbar space appears to be unique to HZO, probably occurring at the outset of the reactivation.⁷⁸ The early ocular inflammation is a nongranulomatous reversible reaction involving especially the superficial cornea, iris, ciliary body, choroid, and trabecular meshwork. With more severe inflammation of the ciliary body and choroid, necrosis can occur, with the onset of granulomatous infiltration of epithelioid and giant cells.⁷⁹

VZV DNA has been detected in ocular specimens up to 10 years after the clinical onset of HZO.¹¹³ However, this cannot distinguish actively replicating virus, viral latency without the presence of complete virions, or viral DNA remnants that persist after active viral infection. Wenkel and associates¹¹³ found signs of active viral replication in the corneal epithelium, the periphery of the corneal stroma, and the episclera in two patients without acute VZV infection. VZV DNA was detected in three corneas with longstanding VZV infection, suggesting a persistence of VZV in non-neuronal tissue. Secondary ischemia from occlusive vasculitis without persisting viral infection may play a major role in the histopathogenesis of the structural alterations seen in the iris and ciliary body.

Late findings include a lymphocytic infiltration of the posterior ciliary nerves and vessels, chronic inflammation and vasculitis of the iris and ciliary body with frequent necrosis, granulomatous choroiditis or giant cell arteritis, perivascular inflammation of retinal vessels, and perineuritis and perivasculitis of ocular muscles. The mechanism of smoldering VZV infection with progressive destruction of the eye is unknown, but mechanisms implicated include different immunoreactions⁷⁹ against inactive viral antigens or persisting viral replication.¹¹⁵ Alternatives include nerve damage, direct cytopathic effects of the virus, or ischemia from occlusive vasculitis.

Viral invasion and host inflammatory and immune responses are the cause of most of the corneal findings. VZV has been recovered from the HZ corneal dendrites seen in the first several days⁷⁶ and from delayed dendritic epithelial keratitis.⁸⁰ Immune reactions (lymphocytic) to viral antigen are suspected in disciform keratitis, which may represent a direct infection of the corneal endothelium.¹¹⁶ A prolonged stromal inflammation (chronic keratouveitis) seen in some patients may correlate with viral replication in the stroma from a direct cytotoxic or immunologic response.⁷⁶ Chronic HZO keratitis is sometimes characterized by a giant cell reaction centered around Descemet's membrane.⁷⁹ VZV DNA is detectable in human corneas up to 8 years after the clinical onset of HZO and may indicate VZV persistence in a latent form in corneal tissue or reactivation of the virus from an endogenous or exogenous source, causing a severe and often recurrent keratitis.¹¹⁷

Secondary ischemia from occlusive vasculitis without persisting viral infection may play a major role in the histopathogenesis of structural alteration seen in the iris and ciliary body.¹¹³ Clinically, the correlation between typical iris necrosis and occlusive vasculitis was shown using fluorescein angiography.¹¹⁸ The posterior pole showed less alteration from HZO than the anterior segment. VZV retinitis and choroiditis are rarely seen.¹¹³ The characteristic and almost pathognomonic histopathologic ocular feature in HZO is a lymphocytic perineuritis and perivasculitis of the long posterior ciliary nerves and arteries. VZV DNA has been localized within the perivascular inflammatory infiltrate and less commonly in the perineural infiltrate,¹¹³ suggesting a primary vascular event. The importance of non-neuronal cells in distinction to neurons in persistence and pathogenesis remains debated.¹¹³

Expression of VZV in blood cells is seen during varicella, thoracic HZ zoster, and postherpetic neuralgia.^119–122 Direct viral invasion of the vessel walls creates a direct cytopathic effect that may activate the immune system, leading to a granulomatous or nongranulomatous vasculitis; this may explain the neurologic complications of HZO, such as intracranial granulomatous and nongranulomatous angiitis¹¹³ and possibly the ARN syndrome.

Vasculitis, therefore, can be a significant component of many of the ocular complications. VZV may lead to ischemic necrosis of the iris (with subsequent iris atrophy),¹¹⁸ of the retina (with herpes zoster retinitis),¹²³ of the choroid,⁷⁸ or of the ciliary body (with phthisis). The optic nerve may be damaged by viral-induced vasculitis or chronic inflammatory response.⁷⁹ Viral invasion of the retina has been confirmed in several patients with ARN.^82,123 The extraocular muscles, orbital myositis, and orbital edema seen in HZO may be a result of perineuritis and perivasculitis associated with generalized orbital inflammation.^75,124

Therefore, although viral replication with persistence in tissue is responsible for some of the acute and recalcitrant features of HZO, clearly there are other inflammatory, perivascular, and host immune responses that are not well defined at present.

TABLE THREE. Diagnostic Tests for Varicella-Zoster Virus (VZV)

  Vesicular Scrapings
  Viral isolation on tissue culture
  Morphologic tests

  Tzanck staining techniques (e.g., Giemsa, Wright, Papanicolaou, light and electron microscopy)

  Direct immunofluorescence or immunoperoxidase stains
  Radioimmunoassay
  ELISA
  Agar gel immunodiffusion
  Immunoelectrophoresis

  In situ hybridization
  Polymerase chain reaction

VIRAL DETECTION

The Tzanck smear is a simple procedure for the rapid presumptive diagnosis of VZV.¹²⁷ It can be performed by scraping the base of the vesicle and staining with hematoxylin and eosin, Giemsa, Wright, toluidine blue, or Papanicolaou stain. The presence of multinucleated giant cells, homogenization of nuclear chromatin, and acidophilic intranuclear bodies with faceted contours rapidly identifies herpesvirus but cannot distinguish VZV.

Vesicle fluid can be collected in a syringe and used for viral isolation or electron microscopic study. VZV is more difficult to recover than HSV because of temperature lability, the variability of viral infectivity in vesicular lesions, and the strong cell-associated nature of the virus. Amplification methods allow identification of viral antigen at an earlier culture stage. Restriction enzyme digestion has been used to differentiate strains. Viral culture is less sensitive than immunofluorescence staining or PCR because of the strong cell-associated nature of VZV.

Electron microscopy can confirm a herpes-like particle, but none of these cytologic techniques alone can distinguish HSV from VZV. Smears from the base of a vesicle can be used for immunologic staining to detect VZV antigens with monoclonal antibodies. PCR is an exquisite technique as a research tool but is moving rapidly into the clinical arena.

The most sensitive host system for isolation of VZV from clinical material is the diploid human cell lines or primary human cell cultures. Inoculation should be performed as soon as possible on monolayer cultures.¹²⁸ Viral growth is usually not evident for 3 to 5 days after inoculation. The cytopathic effect is characterized by acidophilic intranuclear inclusion bodies and multinucleated giant syncytial epithelial cells, similar to those seen with the Tzanck technique. Specific identification of VZV is then confirmed with monoclonal antibody using direct or indirect immunofluorescence.

Antigen detection by direct immunofluorescence is the method of choice for laboratory identification. Virus isolation is second best in sensitivity, but IgG and IgM are not sensitive enough, as least as tested in HIV-infected patients.¹²⁹

SEROLOGIC TESTS FOR ANTIBODY DETECTION

Normally, approximately 90% of the adult population have detectable titers of VZV antibodies, but only 5% have titers of 1:640 or higher in the absence of disease. A high titer suggests VZV as an etiology, whereas an absent titer usually can eliminate VZV as a diagnosis.^30,98,130 Although IgM indicates recent infection, not all patients demonstrate more than a mild rise in titer, whether stimulated by exogenous exposure to new virus or endogenous reactivation of latent VZV.^22,23,130 Moreover, the IgM titer may not change in immunosuppressed patients at the time of either varicella or HZ because of immune failure. High IgG titers then may be a more reliable indicator of the cause of illness in the presence or absence of IgM. Nevertheless, clinically useful information from serologic tests can be helpful, especially paired acute and convalescent sera. For the serodiagnosis of infection, acute and convalescent sera are best analyzed with the same test run: a fourfold antibody titer rise is considered clinically significant. In varicella, the antibody titer reaches a peak at 2 to 3 weeks, but the response is more rapid with HZ. Although the presence of VZV-specific IgM is indicative of a recent active infection, it can be induced by exogenous response to VZV, by asymptomatic “controlled conversions” of the latent VZV, and by symptomatic varicella or HZ. The measurement of VZV-specific IgG, IgM, and IgA by radioimmunoprecipitation and gel fractionation has yielded a specific pattern with certain nonglycosylated polypeptide complexes in HZ reactivation.¹³¹

Many different assays have been used to measure anti-VZV antibodies as an indicator of prior or recurrent infection, to predict susceptibility to disease, and to evaluate immune responses to vaccination. These assays include anticomplement immunofluorescence, complement fixation, immune adherence hemagglutination, passive hemagglutination, radioimmunoassay, immunoblot, latex agglutination, fluorescent-antibody-to-membrane-antigen, immunofluorescence, neutralization, and enzyme-linked immunosorbent assay (ELISA).¹³² Objectivity, sensitivity, specificity, precision, ruggedness, and throughput are considerations for the specific applications of the test (e.g., rapid result, maximum sensitivity, ability to measure antibody responses after immunization or disease). Antibody responses to immunization with live attenuated varicella vaccine are typically lower than those achieved after natural VZV infection and therefore require increased test sensitivity above that suitable for monitoring titers after varicella.¹³² The ideal assay format would provide sensitivity sufficient to detect antibody responses after either disease or vaccination, would be able to distinguish nonspecific antibody, would use well-characterized antigens, would be commercially available, and would be adaptable to screening large numbers of sera in a short period of time.

Fluorescent-antibody-to-membrane-antigen and enhanced neutralization assays that measure antibodies to viral glycoprotein have served as reference standards for the sensitive and specific measurement of antibodies contributing to protection from varicella. However, they are relatively cumbersome for routine use and are not amenable to the timely testing of large numbers of sera.

For measuring antibody responses after vaccination, assays including complement fixation, ELISA, fluorescent-antibody-to-membrane-antigen, immune adherence hemagglutination, immunoblot, immunofluorescence, latex agglutination, enhanced neutralization, passive hemagglutination, and radioimmunoassay have been used. ELISA formats and agglutination are amenable to testing large numbers of sera and do not require significant specialized equipment or facilities. ELISAs have been most widely accepted and applied for routine use.¹³² The ELISA directed at glycoprotein seems ideal.¹³²

PCR has been used to detect VZV DNA in ocular tissues,^114,133 although it is difficult to determine if one is detecting inactive viral particles, active disease, or recrudescent disease. Use of this technique has detected VZV DNA in varicella keratitis,¹³⁴ in the trigeminal ganglia, in delayed pseudodendrites, and in the aqueous of patients with the ARN syndrome or zoster sine herpete.^80,135–137

In VZV infection, as well as other viral diseases, the cellular immunity is a more important diagnostic assay than humoral immunity. Measurement of in vitro proliferative responses of peripheral blood T lymphocytes to VZV antigen and the VZV skin test have been used to distinguish the immune status.^22,138 Patients with HZ respond more dramatically with sensitization of lymphocytes than with changes in antibody level. Correspondingly, patients with diminished VZV CMI have an increased susceptibility to HZ. It remains a cumbersome research tool.

Most attempts to produce varicella in animals have led to the production of antibody but few clinical signs of infection or pathologic changes. Seroconversion has been demonstrated in the rat, rabbit, and guinea pig. The simian varicella virus produces varicella-like disease but does not infect humans. VZV resembles closely this simian varicella virus, including a limited DNA homology distributed throughout the genome.¹³⁹ Simian varicella virus uniformly causes a severe disease with a high fatality rate in the natural hosts, but reactivation as HZ has not been demonstrated. Reactivation of simian varicella appears as a whole-body rash, suggesting that recurrence occurs from multiple ganglia, or more likely that hematogenous spread of virus is important in the pathogenesis. The simian virus can become latent in multiple ganglia without the development of clinical varicella, paralleling the known phenomenon of subclinical varicella infection in humans.¹⁴⁰

The guinea pig model with VZV inoculated on the cornea produced superficial corneal disease in all inoculated animals that resolved within 9 days. VZV was recovered by cocultivation from the conjunctiva and from trigeminal ganglia, superior cervical ganglia, midbrain, and cerebellum up to 20 days after inoculation.¹⁴¹ CNS and pulmonary infection was also evident. Electron microscopy confirmed early (14 to 21 days) latent infection in the trigeminal ganglia.

A rabbit model has also been successful in producing infection of the cornea and the trigeminal ganglia. Persistence or latency has been more difficult to establish, although this may be possible in the rat.^99,142

Acyclovir is converted to acyclovir-monophosphate by the VZV thymidine kinase and is subsequently converted to the active triphosphate form by cellular kinases within infected cells. Phosphorylation is catalyzed by cellular enzymes, resulting in high acyclovir triphosphate concentration. Acyclovir directly inhibits the viral DNA polymerase by competing with deoxyguanosine triphosphate as a substrate for viral DNA polymerase and is incorporated into viral DNA, resulting in chain termination because of the lack of a hydroxyl group. Studies of high-dose oral acyclovir have shown a modest benefit for localized HZ in the normal host, yielding more rapid resolution of viral shedding, acute neuralgia, and new lesion formation.

The literature is not completely consistent in HZO. Several studies have supported oral acyclovir in the treatment of HZO by lessening some of the ocular signs and symptoms,^144–147 but other studies have failed to find the benefit of this expensive therapy.¹⁴⁸ The results of famciclovir, valacyclovir, and other drugs specifically in HZO are not currently published. Acyclovir use in HZO leads to a reduced incidence of pseudodendritic keratopathy, episcleritis, and iritis and also a reduced incidence, although perhaps not severity, of stromal keratitis.^77,144,146 There is little if any effect on corneal hypesthesia, neurotrophic keratitis, or postherpetic neuralgia.¹⁴⁴ Other studies of patients receiving high doses of oral acyclovir showed a lower incidence of keratitis and uveitis in the acyclovir group compared with the untreated patients.^145–147 A large, prospective, double-blind controlled study confirmed that late ocular inflammatory complications occurred in approximately 29% of treated patients versus 50% to 71% of untreated patients.¹⁴⁷ In contrast, one retrospective, paired controlled study failed to confirm a beneficial effect of acyclovir on ocular complications.¹⁴⁸

Use of this agent may not be cost-effective in developing countries¹⁴⁹ or even perhaps in developed countries.¹⁴⁸ The cost of acyclovir is enormous, and there is some disagreement about the cost/benefit analysis in normal hosts.¹⁴⁸ Oral acyclovir combined with prednisone for 21 days improves the quality of life in the immediate acute period associated with the use of steroids, although there does not seem to be an effect of the steroids on postherpetic neuralgia.¹⁵⁰ These authors believed that 10 days of acyclovir might be sufficient but suggested 21 days of steroids (tapered); therefore, they recommended that antivirals be administered during the time steroids are given. Steroids are usually avoided in patients with osteoporosis, diabetes, hypertension, or immunosuppression.

HZO in immunocompromised patients may be associated with dissemination, chronicity, and severe complications and is best treated initially with intravenous acyclovir. The bioavailability of oral acyclovir is erratic, especially in AIDS patients, who may suffer from enteropathies. Also, patients with sight-threatening conditions such as ARN syndrome should receive intravenous acyclovir. Acyclovir-resistant strains tend to develop in patients with AIDS and these patients are best treated with foscarnet.

Valacyclovir, the L-valine ester prodrug of acyclovir, is cleaved to acyclovir after absorption. Valacyclovir has enhanced the plasma concentration of acyclovir, using a stereospecific transporter to enhance gastrointestinal absorption. Enhanced benefit was established for accelerated healing and resolution of zoster-associated pain. Valacyclovir has efficacy and safety similar to acyclovir while permitting more convenient dosing and 7 days of treatment in immunocompetent patients.¹⁵¹ Cases of thrombocytopenic purpura/hemolytic uremic syndrome have been reported in patients with advanced HIV disease. There are not enough data specifically available concerning valacyclovir in HZO, but theoretically it should be more effective than acyclovir.

Famciclovir, a member of the guanine nucleoside family, is a prodrug of penciclovir. Famciclovir is converted into penciclovir and acts similarly to acyclovir once in the cell but has a prolonged intracellular half-life. Penciclovir does not have significant oral bioavailability, but famciclovir has excellent bioavailability. Famciclovir is administered three times a day and is at least equivalent to acyclovir. Both famciclovir and valacyclovir offer ease of administration compared with acyclovir therapy. Acyclovir and famciclovir may accelerate resolution of postherpetic neuralgia, especially among person 50 years or older.^152,153 There are no specific published studies in HZO.

In contrast to acyclovir, valacyclovir, and famciclovir, which are guanosine analogues, sorivudine (BV-araU) is a synthetic pyrimidine nucleoside analogue. It is the most potent of all drugs tested in vitro, with a 1,000-fold increased efficacy against VZV.¹⁵⁴ Sorivudine inhibits VZV replication at a concentration of only 0.0003 μg/ml.¹² It inhibits viral DNA polymerase, as do the other antivirals discussed above, but does not result in DNA chain termination. It has been licensed for use in Japan since 1994 and is still undergoing additional trials with early promising results.^155,156 Because of potential severe toxicity related to an interaction between sorivudine and 5-fluorouracil (can eventuate in neutropenia), the pharmaceutical company has discontinued clinical development.¹⁵⁶ It may be licensed for once-daily therapy in HIV-infected individuals.¹¹

Cidofovir, a phosphorylated nucleoside analogue, may have some use in the future for therapy of acyclovir-resistant VZV because it bypasses the viral thymidine kinase.

Brivudin (bromovinyldeoxyuridine) is a nucleoside analogue licensed in parts of Europe that has shown efficacy in the oral treatment of HZ in immunocompromised patients, including those with ophthalmic involvement.¹⁵⁷ It is not available in the United States.

The majority of acyclovir-resistant mutants of VZV have been found in AIDS patients, related to diminished viral-encoded thymidine kinase function. In vitro resistance to one thymidine kinase-dependent antiviral agent may not necessarily preclude the use of another thymidine kinasedependent drug. DNA polymerase inhibitors such as foscarnet or vidarabine are usually indicated, although this does not guarantee a clinical response.⁸⁶ Foscarnet does not require phosphorylation by the VZV thymidine kinase; thus, viruses with mutations in the thymidine kinase gene are still susceptible to inhibition by foscarnet. Foscarnet inhibits DNA replication by directly inhibiting the viral DNA polymerase; thus, mutations in the DNA polymerase can result in resistance to foscarnet.

Although a substantial increase in the incidence of HZ has been noted over the past 40 years, there has not been a change in the risk of specific or overall complications in a population-based sample of 859 patients compared to earlier studies.¹⁵⁸ A corollary of this (not mentioned by the authors), however, is that the introduction of antivirals may not have influenced the long-term course of the disease. Among 859 patients, the 60-day risk of postherpetic neuralgia was 7.9%, skin superinfections 2.3%, ocular complications 1.6%, motor neuropathy 0.9%, meningitis 0.5%, and herpes zoster oticus 0.2%; this was the same incidence reported years ago. The risk of sequelae increases with age and appears to be more common among those with comorbid conditions. Antiviral therapy alone may never be successful in eliminating the clinicopathophysiologic effects of HZO, especially if the focus of infection and response is in the posterior nerves and blood vessels behind the eye, as histopathologic studies have shown. Evidence that systemic or topical antiviral drugs prevent ophthalmic complications is not absolute.¹⁵⁹

McGill and Chapman¹⁶⁰ have reported several studies documenting that 5% acyclovir ophthalmic ointment is highly effective in resolving HZO epithelial keratitis and in preventing recurrent disease, in comparison with control patients receiving topical corticosteroids. Conversely, Marsh¹⁶¹ noted a more rapid trend toward resolution of inflammation in HZ patients treated with a topical acyclovir-steroid combination compared with acyclovir alone. Other reports failed to show any effect of topical acyclovir on eye complications.¹⁶² Another report found a poor clinical effect from topical acyclovir compared with oral acyclovir.¹⁶³ Topical ophthalmic acyclovir is not available in the United States.

Although systemic corticosteroids have been touted in the treatment of HZ to prevent postherpetic neuralgia, more recent analysis refutes this claim.^164,165 Systemic steroids are indicated, however, to improve the quality of life in the early period after HZ. Topical or systemic steroids are indicated for some of the immunologic and vascular epiphenomena that accompany VZV infection, especially HZO. Topical steroids can be beneficial in the therapy of scleritis, episcleritis, uveitis, and corneal stromal inflammation. Systemic steroids can be beneficial in the therapy of retinal vasculitis, optic neuropathy, and anterior segment ischemia. Although they do not influence viral expression, they quiet the reaction against antigen, which is a prominent part of the ocular and CNS destruction. Steroids must be administered with prudence, tapered slowly to avoid recrudescence, and withdrawn when no longer indicated or if contraindicated. Patients must be monitored for specific side effects.

The pathophysiology of PHN remains obscure and probably is multideterminate. There may be multiple changes in the afferent pathways of both the peripheral nervous system and the CNS. The CNS plays a key role in PHN, because sectioning of peripheral nerves does not relieve the pain. In many patients, stimulus-evoked pain occurs in large areas of skin surrounding the territory of the infected nerve, suggesting abnormal sensory processing in the CNS.^169,170 There is convincing evidence that PHN is a disease of the CNS, with peripheral C-fiber degeneration resulting in an abnormality in how the CNS processes pain signals. There is abnormal persistence of nociceptor-evoked central hyperexcitability, with heightened severity of response to mild stimuli. The injury caused by HZ infection may result in a loss of normal descending inhibitory inputs and a hyperexcitable state of ascending fibers.

The acute pain of shingles is accompanied by several phenomena that might also set the stage for the chronic pain of PHN; for example, the initiation of nociceptor-evoked central hyperexcitability and the production of axonal damage, which may lead to ectopically discharging nociceptor sprouts. This has therapeutic implications. Autopsy studies of individuals with PHN demonstrate histopathologic changes not just in the peripheral nerves but also in the dorsal horn and spinal cord; these CNS changes are not found in patients who recover without PHN.¹⁷¹ Therefore, the morphologic changes in HZ may be reversible, whereas those in PHN are not.¹⁷² VZV-specific proteins have been found in the mononuclear cells of HZ patients with PHN months or years after the rash, suggesting the continued persistence, reactivation, and expression of VZV.¹²² The persistence of VZV DNA in blood mononuclear cells as detected by PCR has been correlated with PHN.¹⁷³

Increased age and prodromal symptoms are associated with a higher prevalence of PHN.¹⁷⁴ Age-related alterations in the availability of specific neurotransmitters and growth factors have yet to be completely elucidated, but they may contribute to PHN as well. Although HZO is thought to have a higher prevalence of PHN compared with HZ elsewhere, this has not been confirmed by studies.^25,174 Ocular complications also were not found to correlate with PHN.¹⁷⁴

There have been many attempts to prevent PHN. If begun within 72 hours after the appearance of the rash, famciclovir, valacyclovir, or acyclovir will reduce the acute pain in immunocompetent patients with HZ, thus providing relief in the greatest number of patients, irrespective of the lack of significant effect on PHN.¹⁷⁵ The antiviral agents limit viral replication, accelerate healing, and reduce early pain, but none consistently reduce the likelihood of PHN at 6 months after onset of rash.^{151,152,174,175} The effect of oral acyclovir on PHN in patients with HZO remains unresolved: some studies show no effect, others show an effect over a limited early period but not at 6 months, and still others show significant reductions in both incidence and severity.^{144,146–148,176} In a meta-analysis of antiviral therapy trials, the prevalence of PHN as denoted by proportions of patients with pain at 3 and 6 months was substantially lower in acyclovir-treated patients than in those randomized to placebo.¹⁵³ Famciclovir may shorten the duration of PHN; the effect of valacyclovir is unknown in PHN. Valacyclovir and famciclovir are given less frequently and are superior in shortening the duration of acute zoster-associated pain.^151,152,177 Others have reported that acyclovir and famciclovir are cost-effective in reducing the duration of PHN.^178,179

Corticosteroids have been studied for three decades in HZ.^{146,150,174,180,181} Corticosteroids ameliorate the pain and debility of acute HZ but do not prevent PHN.^150,175 Although corticosteroids do not alter the course of PHN, the demonstration that they improve the quality of life in the early stages after HZ may justify their administration in combination with an antiviral drug in high-risk patients 50 years of age or older with moderate to severe pain in whom corticosteroids are not contraindicated.¹⁸² However, not all agree.

Aggressive pain control at the onset of HZO, especially in elderly patients prone to PHN, may reduce the magnitude of the initiation phase of nociceptor-evoked central hyperexcitability and lessen the probability that some subsequent factor will be able to maintain abnormal central processing. Treatment recommendations must be individualized.¹⁷⁵ An emerging literature suggests that antagonists to N-methyl-D-aspartate receptors may lessen neuropathic pain by altering abnormal CNS processing. In effect, the CNS learns a pain response to non-noxious stimuli, such as stroking or touch. The available antagonists of these receptors, such as ketamine and dextromethorphan, reduce windup and neuropathic pain but have many adverse effects.¹⁸³ The best available treatment for PHN consists of adrenergically active antidepressants, such as amitriptyline, nortriptyline, or desipramine; the sooner they are given, the better.¹⁷² If pain is incompletely relieved by these medications, oral opioids on a regular basis may be given with little risk of psychological and physical dependency in this particular population.

Topical agents may be of modest benefit. Several studies have evaluated topical capsaicin cream, with mixed results.^184–188 It depletes substance P and other neuropeptides from central and peripheral endings of afferent unmyelinated nerve fibers. At present, the efficacy of topical capsaicin is unproven or modest at best.¹⁷⁵ Clinical trials are difficult to mask because of the burning induced by capsaicin.

Nonpharmacologic, noninvasive, and nontraditional therapies (e.g., transcutaneous electrical nerve stimulation, hypnosis, biofeedback, and other cognitive and behavioral techniques) complement the traditional medical treatment of PHN, with only anecdotal evidence of effect.¹⁸⁹ A pain management clinic can be helpful. A summary report suggests a poor effect of analgesics, anticonvulsants, and sympathetic blockade.¹⁹⁰

Current therapeutic strategies for acute HZO include an antiviral (to lessen acute pain and reduce some ocular complications), systemic steroids (to improve the quality of life in the immediate period), an antidepressant (to modify the CNS effect), and good pain control (a narcotic such as morphine if needed). The pre-emptive use of tricyclic antidepressants such as amitriptyline in all patients older than 60 may avoid PHN.¹⁹¹ In shingles, it seems prudent to treat the pain aggressively, especially in the elderly patient prone to PHN; good pain control during shingles may reduce the magnitude of the initiation phase of nociceptor-evoked central hyperexcitability and lessen the probability that some subsequent factors will be able to maintain abnormal central processing. In the presence of PHN, consideration should be given to a two-pronged strategy aimed at both peripheral and central pain mechanisms. For example, one could combine topical application of a local anesthetic with a tricyclic antidepressant. DMDA receptor antagonists such as dextrorphan, dextromethorphan, and memantine may prove to be particularly effective in suppressing nociceptor-evoked central hyperexcitability.^192,193

In Japan, vaccine trials progressed from healthy children to immunocompromised children, whereas in the United States they progressed from immunocompromised children to healthy children. Gershon and Steinberg¹⁹⁵ conducted the clinical trials in immunocompromised children and adults in the United States with vaccine manufactured both in Europe and in the United States; they confirmed the studies in Japan that the vaccine was effective in preventing severe disease in appropriate target groups. Clinical trials in healthy children also proved the vaccine was relatively safe and very effective in preventing severe disease. Most vaccinees who were not completely protected had mild rashes and sometimes mild fever on exposure to infected individuals (modified varicella-like syndrome). These individuals usually did not mount a robust immune response to the vaccination. In March 1995, the U.S. Food & Drug Administration licensed a live attenuated varicella vaccine for immunization of healthy infants, children, adolescents, and adults in the United States who have not had chickenpox and are not pregnant.

Because there is little or no viremia in vaccinees, in contrast to the high rate of viremia for several days before the appearance of rash in natural varicella, there is less possibility for the vaccine virus to become latent in ganglia, and the possibility of less HZ in the future. When live attenuated varicella vaccine is injected into an immunocompetent human host, the virus can replicate at a local site without apparent systemic spread. If a large dose of vaccine is given, or if the host is immunocompromised, the virus may be able to spread beyond the injection site to distant internal organs for further cycles of replication, followed by a viremia and rash. In the latter circumstance, virus can be carried to the mucosal region of the head and neck, whence transmission can occur.

There is concern that the widespread use of vaccine in these healthy populations will shift the prevalence of disease to older individuals, in whom varicella is more serious than in children, or that it will alter the epidemiology of HZ. Mathematical models and scientific assumptions suggest that the societal benefit will be maximized if there is widespread use of the vaccine in all susceptible individuals. Mathematical modeling suggests that routine immunization of preschool children would greatly reduce the number of primary varicella cases, whereas the shift in age distribution of cases would not result in increased overall morbidity as measured by the number of hospitalizations.¹⁹⁶ The role of exposure to varicella in the maintenance of immunity among immunologically normal persons may play a role; responsiveness to boosting by wild-type VZV infections was especially important in reducing the number of older cases. The direct and indirect effects of vaccinees and vaccination programs interact. Models, however, cannot replace biologic understanding; uncertainties still exist in predicting the long-term results of vaccination. Vaccine efficacy studies in the field should be designed to obtain better estimates of residual susceptibility, residual infectiousness, duration of protection, and the effects of boosting by reinfection with wild-type virus.

Patients have developed HZ after the VZV vaccine, although the frequency of reinfection after vaccination appears to be less than that after natural infection.^102,197 The incidence of HZ has not increased in children vaccinees and may have decreased. Both the vaccine and wild-type strains have been recovered from these patients. The incidence of HZ in vaccinees is more common in children who develop rash,¹⁹⁴ suggesting that the peripheral sensory nerve route from skin vesicles is a major route of migration of VZV to sensory ganglia. In leukemic children, the incidence of HZ, as determined by life-table analysis, was 3% in vaccinees versus 15% in controls.^198–200 This may represent the attenuated state of the virus or the lower rate of latency after vaccination compared with natural infection, because the skin is not infected after most vaccinations, and access of the virus to sensory nerves may be less. The rate of HZ in vaccinated healthy children and adults also appears to be several times lower than that resulting from natural infection, although data are still limited and follow-up is still short. The HZ in vaccinees has generally been mild; some cases have resulted from reactivation of vaccine virus, others from superinfection of wild-type strain and subsequent reactivation. HZO might be specifically less likely to occur because the inoculation is performed on an extremity. There has been no evidence to date to indicate waning immunity after vaccination. Experience with virus reactivation after adult vaccination has been limited by the size of the vaccinee pool.

Theoretically, a vaccine could be used both to provide immunity to varicella and to boost the immunity of patients with previous varicella to prevent HZ.²⁰¹ Pilot studies have been done in the Veterans Administration hospital system with live attenuated vaccine in patients 60 or older to determine if varicella vaccine will decrease the incidence of HZ, and to evaluate its effect on the cellular immune response to VZV. The expected result is that it may protect against reactivation of HZ.²⁰² Consideration is being given to booster shots in adolescence or in adulthood with either attenuated or glycoprotein subunit vaccines to prevent or modify HZ.²⁰³

1. Weller TH: Varicella and herpes zoster: A perspective and overview. J Infect Dis 166:S1, 1992

2. Steiner P: Zur Inokulation der Varicellen. Wein Med Wochenschr 25:306, 1875

3. von Bokay J: Das Auftreten der Schafblattern unter besonderen Umstanden. Unger Arch Med 1:159, 1892

4. Garland J: Varicella following exposure to herpes zoster. N Engl J Med 228:336, 1943

5. Weller TH, Stoddard MB: Intranuclear inclusion bodies in cultures of human tissue inoculated with varicella vesicle fluid. J Immunol 68:311, 1952

6. Straus SE, Reinhold W, Smith HA et al: Endonuclease analysis of viral DNA from varicella and subsequent zoster infections in the same patient. N Engl J Med 311:1362, 1984

7. Zweerink HJ, Morton DH, Stanton LW, Neff BJ: Restriction endonuclease analysis of the DNA from VZV: Stability of the DNA after passage in vitro. J Gen Virol 55:207, 1981

8. Davison AJ, Scott JE: The complete DNA sequence of VZV. J Gen Virol 67:1759, 1986

9. Lowe RS, Keller PM, Keech BJ et al: VZV as a live vector for the expression of foreign genes. Proc Natl Acad Sci USA 84:3896, 1987

10. Cohen JI, Seidel KE: Generation of VZV and viral mutants from cosmid DNAs: VZV thymidylate synthetase is not essential for replication in vitro. Proc Natl Acad Sci USA 90:7376, 1993

11. Harper DR, Gilbert RL, Jeffries DJ: Molecular biology of VZV. A review prepared for the UK Advisory Group on Chickenpox. J Infect 36:1, 1998

12. Cohen JI: VZV. The virus. Infect Dis Clin North Am 10: 457, 1996

13. Weller TH: Serial propagation in vitro of agents producing inclusion bodies derived from varicella and herpes zoster. Proc Soc Exp Biol Med 83:340, 1953

14. Straus SE, Ostrove JM, Inchauspe G et al: NIH conference: VZV infections. Biology, natural history, treatment, and prevention. Ann Intern Med 108:221, 1988

15. Grose C: Glycoproteins encoded by VZV: Biosynthesis, phosphorylation, and intracellular trafficking. Ann Rev Microbiol 44:59, 1990

16. Zhu Z, Gershon MD, Ambron R et al: Infection of cells by VZV: Inhibition of viral entry by mannose 6-phosphate and heparin. Proc Natl Acad Sci USA 92:3546, 1995

17. Dumas AM, Geelen JL, Mares W, Van der Noordaa J: Infectivity and molecular weight of VZV DNA. J Gen Virol 47:233, 1980

18. Stow ND, Davison AJ: Identification of a VZV origin of DNA replication and its activation by herpes simplex virus type 1 gene products. J Gen Virol 67:1613, 1986

19. Straus SE, Hay J, Smith H, Owens J: Genome differences among VZV isolates. J Gen Virol 64:1031, 1983

20. Preblud SR, Orenstein WA, Bart KJ: Varicella: Clinical manifestations, epidemiology and health impact in children. Pediatr Infect Dis 3:505, 1984

21. Gershon AA, Steinberg SP: Antibody responses to VZV and the role of antibody in host defense. Am J Med Sci 282:12, 1981

22. Arvin AM, Koropchak CM, Wittek AE: Immunologic evidence of reinfection with VZV. J Infect Dis 148:200, 1983

23. Gershon AA, Steinberg SP, Gelb L: Clinical reinfection with VZV. J Infect Dis 149:137, 1984

24. Wharton M: The epidemiology of VZV infections. Infect Dis Clin North Am 10:571, 1996

25. Ragozzino MW, Melton LJ, Kurland LT et al: Population-based study of herpes zoster and its sequelae. Medicine 61:310, 1982

26. Donahue JG, Choo PW, Manson JE, Platt R: The incidence of herpes zoster. Arch Intern Med 155:1605, 1995

27. Schmader K, George LK, Burchett BM et al: Racial differences in the occurrence of herpes zoster. J Infect Dis 171:701, 1995

28. Levin MJ, Hayward AR: The varicella vaccine. Prevention of herpes zoster. Infect Dis Clin North Am 10:657, 1996

29. Palmer SR, Caul EO, Donald DE et al: An outbreak of shingles? Lancet 2:1108, 1985

30. Brunell PA, Gershon AA, Uduman SA, Steinberg S: Varicella-zoster immunoglobulins during varicella, latency, and zoster. J Infect Dis 132:49, 1975

31. Arvin AM, Pollard RB, Rasmussen LE, Merigan TC: Cellular and humoral immunity in the pathogenesis of recurrent herpes viral infections in patients with lymphoma. J Clin Invest 65:869, 1980

32. Friedman-Kien AE, Lafleur FL, Gendler E et al: Herpes zoster: A possible early clinical sign for development of acquired immunodeficiency syndrome in high-risk individuals. J Am Acad Dermatol 14:1023, 1986

33. Melbye M, Grossman RJ, Goedert JJ et al: Risk of AIDS after herpes zoster. Lancet 1:728, 1987

34. Dolin R, Reichman RC, Mazur MH, Whitley RJ: NIH Conference: HZV infections in immunosuppressed patients. Ann Intern Med 89:375, 1978

35. Gallagher JG, Merigan TC: Prolonged herpes-zoster infection associated with immunosuppressive therapy. Ann Intern Med 91:842, 1979

36. Furuta Y, Takasu T, Fukuda S et al: Detection of VZV DNA in human geniculate ganglia by polymerase chain reaction. J Infect Dis 166:1157, 1992

37. Liedtke W, Opalka B, Zimmermann CW, Lignitz E: Age distribution of latent herpes simplex virus 1 and VZV genome in human nervous tissue. J Neurol Sci 116:6, 1993

38. Mahalingam R, Wellish MC, Dueland AN et al: Localization of herpes simplex virus and varicella zoster virus DNA in human ganglia. Ann Neurol 31:444, 1992

39. Mahalingam R, Wellish M, Wolf W et al: Latent varicella-zoster viral DNA in human trigeminal and thoracic ganglia. N Engl J Med 323:627, 1990

40. Plotkin SA, Stein S, Snyder M, Immesoete P: Attempts to recover varicella virus from ganglia. Ann Neurol 2:249, 1977

41. Esiri MM, Tomlinson AH: Herpes zoster. Demonstration of virus in trigeminal nerve and ganglion by immunofluorescence and electron microscopy. J Neurol Sci 15:35, 1972

42. Bastian FO, Rabson AS, Yee CL, Tralka TS: Herpesvirus varicellae isolated from human dorsal root ganglia. Arch Pathol 97:331, 1974

43. Dueland AN: Latency and reactivation of varicella zoster virus infections. Scand J Infect Dis 100(supp):46, 1996

44. Gilden DH, Rozenman Y, Murray R et al: Detection of VZV nucleic acid in neurons of normal human thoracic ganglia. Ann Neurol 22:377, 1987

45. Meier JL, Holman RP, Croen KD et al: VZV transcription in human trigeminal ganglia. Virology 193:193, 1993

46. Croen KD, Ostrove JM, Dragovic LJ, Straus SE: Patterns of gene expression and sites of latency in human nerve ganglia are different for varicella-zoster and herpes simplex viruses. Proc Natl Acad Sci USA 85:9773, 1988

47. Kennedy PG, Grinfeld E, Gow JW: Latent VZV is located predominantly in neurons in human trigeminal ganglia. Proc Natl Acad Sci USA 95:4658, 1998

48. Cohrs RJ, Barbour MB, Mahalingam R et al: VZV transcription during latency in human ganglia: Prevalence of VZV gene 21 transcripts in latently infected human ganglia. J Virol 69:2674, 1995

49. Gilden DH, Dueland AN, Devlin ME et al: VZV reactivation without rash. J Infect Dis 166:S30, 1992

50. Wilson A, Sharp M, Koropchak CM et al: Subclinical VZV viremia, herpes zoster, and T lymphocyte immunity to varicella-zoster viral antigens after bone marrow transplantation. J Infect Dis 165:119, 1992

51. Ljungman P, Lonnqvist B, Gahrton G et al: Clinical and subclinical reactivations of VZV in immunocompromised patients. J Infect Dis 153:840, 1986

52. Weigle KA, Grose C: Molecular dissection of the humoral immune response to individual varicella-zoster viral proteins during chickenpox, quiescence, reinfection, and reactivation. J Infect Dis 149:741, 1984

53. Arvin AM: Cell-mediated immunity to VZV. J Infect Dis 166:S35, 1992

54. Hayward A, Levin M, Wolf W et al: VZV-specific immunity after herpes zoster. J Infect Dis 163:873, 1991

55. Gilden DH, Wright RR, Schneck SA et al: Zoster sine herpete, a clinical variant. Ann Neurol 35:530, 1994

56. Luby JP, Ramirez-Ronda C, Rinner S et al: A longitudinal study of VZV infections in renal transplant recipients. J Infect Dis 135:659, 1977

57. Gilden DH, Hayward AR, Krupp J et al: VZV infection of human mononuclear cells. Virus Res 7:117, 1987

58. Grose C: Pathogenesis of infection with varicella vaccine. Infect Dis Clin North Am 10:489, 1996

59. Fleisher G, Henry W, McSorley M et al: Life-threatening complications of varicella. Am J Dis Child 135:896, 1981

60. Uchida Y, Kaneko M, Hayashi K: Varicella dendritic keratitis. Am J Ophthalmol 89:259, 1980

61. Lambert SR, Taylor D, Kriss A et al: Ocular manifestations of the congenital varicella syndrome. Arch Ophthalmol 107:52, 1989

62. Lewis G: Zoster sine herpete. Br Med J 2:418, 1958

63. Balfour HH Jr: Varicella zoster virus infections in immunocompromised hosts. A review of the natural history and management. Am J Med 85:68, 1988

64. Rockley PF, Tyring SK: Pathophysiology and clinical manifestations of varicella zoster virus infections. Int J Dermatol 33:227, 1994

65. Schmidbauer M, Budka H, Pilz P et al: Presence, distribution and spread of productive varicella zoster virus infection in nervous tissues. Brain 115:383, 1992

66. Locksley RM, Flournoy N, Sullivan KM, Meyers JD: Infection with VZV after marrow transplantation. J Infect Dis 152:1172, 1985

67. Litoff D, Catalano RA: Herpes zoster optic neuritis in human immunodeficiency virus infection. Arch Ophthalmol 108:782, 1990

68. Alessi E, Cusini M, Zerboni R et al: Unusual varicella zoster virus infection in patients with the acquired immunodeficiency syndrome [letter]. Arch Dermatol 124:1011, 1988

69. Reichman RC: Neurologic complications of varicella-zoster infections. Ann Intern Med 89:379, 1978

70. Hilt DC, Buchholz D, Krumholz A et al: Herpes zoster ophthalmicus and delayed contralateral hemiparesis caused by cerebral angiitis: Diagnosis and management approaches. Ann Neurol 14:543, 1983

71. MacKenzie RA, Forbes GS, Karnes WE: Angiographic findings in herpes zoster arteritis. Ann Neurol 10:458, 1981

72. Hutchinson J: A clinical report on herpes zoster frontalis ophthalmicus (shingles affecting the forehead and nose). Roy London Ophthalm Hosp Rep 5:191, 1865

73. Cobo M, Foulks GN, Liesegang T et al: Observations on the natural history of herpes zoster ophthalmicus. Curr Eye Res 6:195, 1987

74. Womack LW, Liesegang TJ: Complications of herpes zoster ophthalmicus. Arch Ophthalmol 101:42, 1983

75. Liesegang TJ: The VZV: systemic and ocular features. J Am Acad Dermatol 11:165, 1984

76. Liesegang TJ: Corneal complications from herpes zoster ophthalmicus. Ophthalmology 92:316, 1985

77. Harding SP, Lipton JR, Wells JC: Natural history of herpes zoster ophthalmicus: Predictors of postherpetic neuralgia and ocular involvement. Br J Ophthalmol 71:353, 1987

78. Naumann G, Gass JD, Font RL: Histopathology of herpes zoster ophthalmicus. Am J Ophthalmol 65:533, 1968

79. Hedges TR, Albert DM: The progression of the ocular abnormalities of herpes zoster. Histopathologic observations of nine cases. Ophthalmology 89:165, 1982

80. Pavan-Langston D, Yamamoto S, Dunkel EC: Delayed herpes zoster pseudodendrites. Polymerase chain reaction detection of viral DNA and a role for antiviral therapy. Arch Ophthalmol 113:1381, 1995

81. Karbassi M, Raizman MB, Schuman JS: Herpes zoster ophthalmicus. Surv Ophthalmol 36:395, 1992

82. Hirsch M: Herpes group in the immunocompromised host. In Rubin R, Young L (eds): Clinical Approach to Infection in the Compromised Host, p 347. New York, Plenum, 1988

83. Sandor EV, Millman A, Croxson TS, Mildvan D: Herpes zoster ophthalmicus in patients at risk for the acquired immune deficiency syndrome (AIDS). Am J Ophthalmol 101:153, 1986

84. Cole EL, Meisler DM, Calabrese LH et al: Herpes zoster ophthalmicus and acquired immune deficiency syndrome. Arch Ophthalmol 102:1027, 1984

85. Culbertson WW, Blumenkranz MS, Pepose JS et al: Varicella zoster virus is a cause of the acute retinal necrosis syndrome. Ophthalmology 93:559, 1986

86. Engstrom RE Jr, Holland GN, Margolis TP et al: The progressive outer retinal necrosis syndrome. A variant of necrotizing herpetic retinopathy in patients with AIDS. Ophthalmology 101:1488, 1994

87. Garweg J, Bohnke M: VZV is strongly associated with atypical necrotizing herpetic retinopathies. Clin Infect Dis 24:603, 1997

88. Culbertson WW, Dix RD: VZV diseases: Posterior segment of the eye. In Pepose JS, Holland GN, Wilhelmus KR (eds): Ocular Infection and Immunity, p 1131. St. Louis, Mosby-Year Book, 1995

89. Pepose JS: The potential impact of the varicella vaccine and new antivirals on ocular disease related to VZV. Am J Ophthalmol 123:243, 1997

90. Kestelyn P, Stevens AM, Bakkers E et al: Severe herpes zoster ophthalmicus in young African adults: A marker for HTLV-III seropositivity. Br J Ophthalmol 71:806, 1987

91. Schuman JS, Orellana J, Friedman AH, Teich SA: Acquired immunodeficiency syndrome (AIDS). Surv Ophthalmol 31:384, 1987

92. Jacobson MA, Berger TG, Fikrig S et al: Acyclovir-resistant varicella zoster virus infection after chronic oral acyclovir therapy in patients with the acquired immunodeficiency syndrome (AIDS). Ann Intern Med 112:187, 1990

93. Margolis TP, Milner MS, Shama A et al: Herpes zoster ophthalmicus in patients with human immunodeficiency virus infection. Am J Ophthalmol 125:285, 1998

94. Hope-Simpson R: The nature of herpes zoster: A long-term study and a new hypothesis. Proc R Soc Med 58:9, 1965

95. Miller AE: Selective decline in cellular immune response to varicella-zoster in the elderly. Neurology 30:582, 1980

96. Burke BL, Steele RW, Beard OW et al: Immune responses to varicella-zoster in the aged. Arch Intern Med 142:291, 1982

97. Arvin AM: Immune responses to VZV. Infect Dis Clin North Am 10:529, 1996

98. Cradock-Watson JE, Ridehalgh MK, Bourne MS: Specific immunoglobulin responses after varicella and herpes zoster. J Hyg (Lond) 82:319, 1979

99. Arvin AM: VZV. Clin Microbiol Rev 9:361, 1996

100. Sundqvist VA, Linde A, Wahren B: Virus-specific immunoglobulin G subclasses in herpes simplex and VZV infections. J Clin Microbiol 20:94, 1984

101. Schmidt NJ, Lennette EH: Neutralizing antibody responses to VZV. Infect Immun 12:606, 1975

102. White CJ: VZV vaccine. Clin Infect Dis 24:753, 1997

103. Miller LH, Brunell PA: Zoster, reinfection or activation of latent virus? Observations on the antibody response. Am J Med 49:480, 1970

104. Bates CJ, Kudesia G, McKendrick MW: Effect of acyclovir and prednisolone on the serological response in herpes zoster. J Med Virol 50:244, 1996

105. Ghatak NR, Zimmerman HM: Spinal ganglion in herpes zoster. A light and electron microscopic study. Arch Pathol 95:411, 1973

106. Doyle PW, Gibson G, Dolman CL: Herpes zoster ophthalmicus with contralateral hemiplegia: Identification of cause. Ann Neurol 14:84, 1983

107. Nagashima K, Nakazawa M, Endo H: Pathology of the human spinal ganglia in VZV infection. Acta Neuropathol (Berl) 33:105, 1975

108. Mayberg MR, Zervas NT, Moskowitz MA: Trigeminal projections to supratentorial pial and dural blood vessels in cats demonstrated by horseradish peroxidase histochemistry. J Comp Neurol 223:46, 1984

109. Shibuta H, Ishikawa T, Hondo R et al: Varicella virus isolation from spinal ganglion. Arch Gesamte Virusforsch 45:382, 1974

110. Hyman RW, Ecker JR, Tenser RB: VZV RNA in human trigeminal ganglia. Lancet 2:814, 1983

111. Gilden DH, Devlin M, Wellish M et al: Persistence of VZV DNA in blood mononuclear cells of patients with varicella or zoster. Virus Genes 2:299, 1989

112. Devlin ME, Gilden DH, Mahalingam R et al: Peripheral blood mononuclear cells of the elderly contain VZV DNA. J Infect Dis 165:619, 1992

113. Wenkel H, Rummelt V, Fleckenstein B, Naumann GO: Detection of varicella zoster virus DNA and viral antigen in human eyes after herpes zoster ophthalmicus. Ophthalmology 105:1323, 1998

114. Liesegang TJ: The use of polymerase chain reaction techniques to detect VZV in corneal transplant tissue [editorial]. Arch Ophthalmol 115:664, 1997

115. Zaal MJ, Maudgal PC, Rietveld E, Suir EP: Chronic ocular zoster. Curr Eye Res 10:125, 1991

116. Maudgal PC, Missotten L, De Clercq E, Descamps J: VZV in the human corneal endothelium: A case report. Bull Soc Belge Ophthalmol 190:71, 1980

117. Wenkel H, Rummelt C, Rummelt V et al: Detection of varicella zoster virus DNA and viral antigen in human cornea after herpes zoster ophthalmicus. Cornea 12:131, 1993

118. Marsh RJ, Easty DL, Jones BR: Iritis and iris atrophy in herpes zoster ophthalmicus. Am J Ophthalmol 78:255, 1974

119. Koropchak CM, Solem SM, Diaz PS, Arvin AM: Investigation of VZV infection of lymphocytes by in situ hybridization. J Virol 63:2392, 1989

120. Ozaki T, Ichikawa T, Matsui Y et al: Lymphocyte-associated viremia in varicella. J Med Virol 19:249, 1986

121. Vonsover A, Leventon-Kriss S, Langer A et al: Detection of VZV in lymphocytes by DNA hybridization. J Med Virol 21:57, 1987

122. Vafai A, Wellish M, Gilden DH: Expression of VZV in blood mononuclear cells of patients with postherpetic neuralgia. Proc Natl Acad Sci USA 85:2767, 1988

123. Schwartz JN, Cashwell F, Hawkins HK, Klintworth GK: Necrotizing retinopathy with herpes zoster ophthalmicus: A light and electron microscopical study. Arch Pathol Lab Med 100:386, 1976

124. Chang-Godinich A, Lee AG, Brazis PW et al: Complete ophthalmoplegia after zoster ophthalmicus. J Neuroophthalmol 17:262, 1997

125. Gershon AA, LaRussa P, Steinberg SP: VZV. In Murray PR, Baron EJ, Pfaller MA et al (eds): Manual of Clinical Microbiology, 6th ed, p 895. Washington DC, ASM Press, 1995

126. Liesegang TJ: Diagnosis and therapy of herpes zoster ophthalmicus. Ophthalmology 98:1216, 1991

127. Solomon AR, Rasmussen JE, Weiss JS: A comparison of the Tzanck smear and viral isolation in varicella and herpes zoster. Arch Dermatol 122:282, 1986

128. Schmidt NJ: VZV. In Lennette EN, Ballows A, Hausler WJ, Shadomy HJ (eds): Manual of Clinical Microbiology, 4th ed. Washington DC, American Society for Microbiology, 1985

129. Dahl H, Marcoccia J, Linde A: Antigen detection: The method of choice in comparison with virus isolation and serology for laboratory diagnosis of herpes zoster in human immunodeficiency virus-infected patients. J Clin Microbiol 35:347, 1997

130. Gershon AA, Steinberg SP, Borkowsky W et al: IgM to VZV: Demonstration in patients with and without clinical zoster. Pediatr Infect Dis 1:164, 1982

131. Harper DR, Grose C: IgM and IgG responses to VZV p32/p36 complex after chickenpox and zoster, congenital and subclinical infections, and vaccination. J Infect Dis 159: 444, 1989

132. Krah DL: Assays for antibodies to VZV. Infect Dis Clin North Am 10:507, 1996

133. Mietz H, Eis-Hubinger AM, Sundmacher R, Font RL: Detection of VZV DNA in keratectomy specimens by use of the polymerase chain reaction. Arch Ophthalmol 115: 590, 1997

134. Power WJ, Hogan RN, Hu S, Foster CS: Primary varicella-zoster keratitis: Diagnosis by polymerase chain reaction. Am J Ophthalmol 123:252, 1997

135. Yamamoto S, Tada R, Shimomura Y et al: Detecting VZV DNA in iridocyclitis using polymerase chain reaction: A case of zoster sine herpete [case report]. Arch Ophthalmol 113:1358, 1995

136. Stavrou P, Mitchell SM, Fox JD et al: Detection of VZV DNA in ocular samples from patients with uveitis but no cutaneous eruption. Eye 8:684, 1994

137. Nishi M, Hanashiro R, Mori S et al: Polymerase chain reaction for the detection of the varicella-zoster genome in ocular samples from patients with acute retinal necrosis. Am J Ophthalmol 114:603, 1992

138. Tanaka Y, Harino S, Danjo S et al: Skin test with VZV antigen for ophthalmic herpes zoster. Am J Ophthalmol 98:7, 1984

139. Gray WL, Oakes JE: Simian varicella virus DNA shares homology with human VZV DNA. Virology 136:241, 1984

140. Vafai A, Murray RS, Wellish M et al: Expression of VZV and herpes simplex virus in normal human trigeminal ganglia. Proc Natl Acad Sci USA 85:2362, 1988

141. Pavan-Langston D, Dunkel EC: Ocular VZV infection in the guinea pig. A new in vivo model. Arch Ophthalmol 107:1068, 1989

142. Myers MG, Connelly BL: Animal models of varicella. J Infect Dis 166:S48, 1992

143. Marsh RJ: Herpes zoster ophthalmicus. J R Soc Med 90: 670, 1997

144. Cobo LM, Foulks GN, Liesegang T et al: Oral acyclovir in the treatment of acute herpes zoster ophthalmicus. Ophthalmology 93:763, 1986

145. Herbort CP, Buechi ER, Piguet B et al: High-dose oral acyclovir in acute herpes zoster ophthalmicus: The end of the corticosteroid era. Curr Eye Res 10:171, 1991

146. Harding SP, Porter SM: Oral acyclovir in herpes zoster ophthalmicus. Curr Eye Res 10:177, 1991

147. Hoang-Xuan T, Buchi ER, Herbort CP et al: Oral acyclovir for herpes zoster ophthalmicus. Ophthalmology 99:1062, 1992

148. Aylward GW, Claoue CM, Marsh RJ, Yasseem N: Influence of oral acyclovir on ocular complications of herpes zoster ophthalmicus. Eye 8:70, 1994

149. Kubeyinje EP: Cost-benefit of oral acyclovir in the treatment of herpes zoster. Int J Dermatol 36:457, 1997

150. Whitley RJ, Weiss H, Gnann JW Jr et al: Acyclovir with and without prednisone for the treatment of herpes zoster. A randomized, placebo-controlled trial. The National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group. Ann Intern Med 125:376, 1996

151. Beutner KR, Friedman DJ, Forszpaniak C et al: Valacyclovir compared with acyclovir for improved therapy for herpes zoster in immunocompetent adults. Antimicrob Agents Chemother 39:1546, 1995

152. Tyring S, Barbarash RA, Nahlik JE et al: Famciclovir for the treatment of acute herpes zoster: Effects on acute disease and postherpetic neuralgia. A randomized, double-blind, placebo-controlled trial. Collaborative Famciclovir Herpes Zoster Study Group. Ann Intern Med 123:89, 1995

153. Wood MJ, Kay R, Dworkin RH et al: Oral acyclovir therapy accelerates pain resolution in patients with herpes zoster: A meta-analysis of placebo-controlled trials. Clin Infect Dis 22:341, 1996

154. Whitley RJ: Sorivudine: A promising drug for the treatment of VZV infection. Neurology 45:S73, 1995

155. Bodsworth NJ, Boag F, Burdge D et al: Evaluation of sorivudine (BV-araU) versus acyclovir in the treatment of acute localized herpes zoster in human immunodeficiency virus-infected adults. The Multinational Sorivudine Study Group. J Infect Dis 176:103, 1997

156. Gnann JW, Crumpacker CS, Lalezari JP et al: Sorivudine versus acyclovir for treatment of dermatomal herpes zoster in human immunodeficiency virus-infected patients: Results from a randomized, controlled clinical trial. Antimicrob Agents Chemother 42:1139, 1998

157. Wutzler P, De Clercq E, Wutke K, Farber I: Oral brivudin vs. intravenous acyclovir in the treatment of herpes zoster in immunocompromised patients: A randomized double-blind trial. J Med Virol 46:252, 1995

158. Galil K, Choo PW, Donahue JG, Platt R: The sequelae of herpes zoster. Arch Intern Med 157:1209, 1997

159. Guidelines for the management of shingles. Report of a working group of the British Society for the Study of Infection (BSSI). J Infect 30:193, 1995

160. McGill J, Chapman C: A comparison of topical acyclovir with steroids in the treatment of herpes zoster keratouveitis. Br J Ophthalmol 67:746, 1983

161. Marsh RJ: Ophthalmic herpes zoster. Br J Hosp Med 15:609, 1976

162. Marsh RJ, Cooper M: Double-masked trial of topical acyclovir and steroids in the treatment of herpes zoster ocular inflammation. Br J Ophthalmol 75:542, 1991

163. Neoh C, Harding SP, Saunders D et al: Comparison of topical and oral acyclovir in early herpes zoster ophthalmicus. Eye 8:688, 1994

164. Post BT, Philbrick JT: Do corticosteroids prevent postherpetic neuralgia? A review of the evidence. J Am Acad Dermatol 18:605, 1988

165. Esmann V, Geil JP, Kroon S et al: Prednisolone does not prevent post-herpetic neuralgia. Lancet 2:126, 1987

166. deMoragas J, Kierland R: The outcome of patients with herpes zoster. Arch Dermatol 75:193, 1957

167. Watson PN, Evans RJ: Postherpetic neuralgia. A review. Arch Neurol 43:836, 1986

168. Nurmikko T: Clinical features and pathophysiologic mechanisms of postherpetic neuralgia. Neurology 45:S54, 1995

169. Rowbotham MC, Fields HL: Post-herpetic neuralgia: The relation of pain complaint, sensory disturbance, and skin temperature. Pain 39:129, 1989

170. Watson CPN, Deck JH: The neuropathology of herpes zoster with particular reference to postherpetic neuralgia and its pathogenesis. In Watson CPN (ed): Herpes Zoster and Postherpetic Neuralgia. Pain Research and Clinical Management, p 139. Amsterdam, Elsevier, 1993

171. Watson CP, Deck JH, Morshead C et al: Post-herpetic neuralgia: Further post-mortem studies of cases with and without pain. Pain 44:105, 1991

172. Bowsher D: Pathophysiology of postherpetic neuralgia: Towards a rational treatment. Neurology 45:S56, 1995

173. Mahalingam R, Wellish M, Brucklier J, Gilden DH: Persistence of VZV DNA in elderly patients with postherpetic neuralgia. J Neurovirol 1:130, 1995

174. Choo PW, Galil K, Donahue JG et al: Risk factors for postherpetic neuralgia. Arch Intern Med 157:1217, 1997

175. Kost RG, Straus SE: Postherpetic neuralgia—pathogenesis, treatment, and prevention. N Engl J Med 335:32, 1996

176. McGill JI, White JE: Acyclovir and post-herpetic neuralgia and ocular involvement. Br Med J 309:1124, 1994

177. deGreef H: Famciclovir, a new oral antiherpes drug: Results of the first controlled clinical study demonstrating its efficacy and safety in the treatment of uncomplicated herpes zoster in immunocompetent patients. Int J Antimicrob Agents 4:241, 1995

178. Huse DM, Schainbaum S, Kirsch AJ, Tyring S: Economic evaluation of famciclovir in reducing the duration of postherpetic neuralgia. Am J Health Syst Pharm 54:1180, 1997

179. Jackson JL, Gibbons R, Meyer G, Inouye L: The effect of treating herpes zoster with oral acyclovir in preventing postherpetic neuralgia. A meta-analysis. Arch Intern Med 157:909, 1997

180. Clemmensen OJ, Andersen KE: ACTH versus prednisone and placebo in herpes zoster treatment. Clin Exp Dermatol 9:557, 1984

181. Wood MJ, Johnson RW, McKendrick MW et al: A randomized trial of acyclovir for 7 days or 21 days with and without prednisolone for treatment of acute herpes zoster. N Engl J Med 330:896, 1994

182. Whitley RJ, Weiss H, Gnann J: The efficacy of steroid and acyclovir therapy of herpes zoster in the elderly [abstract]. J Invest Med 43:252A, 1995

183. Eide PK, Jorum E, Stubhaug A et al: Relief of post-herpetic neuralgia with the N-methyl-D-aspartic acid receptor antagonist ketamine: A double-blind, cross-over comparison with morphine and placebo. Pain 58:347, 1994

184. Bernstein JE, Bickers DR, Dahl MV, Roshal JY: Treatment of chronic postherpetic neuralgia with topical capsaicin. A preliminary study. J Am Acad Dermatol 17:93, 1987

185. Bernstein JE, Korman NJ, Bickers DR et al: Topical capsaicin treatment of chronic postherpetic neuralgia. J Am Acad Dermatol 21:265, 1989

186. Gamse R, Petsche U, Lembeck F, Jancso G: Capsaicin applied to peripheral nerve inhibits axoplasmic transport of substance P and somatostatin. Brain Res 239:447, 1982

187. Jessell TM, Iversen LL, Cuello AC: Capsaicin-induced depletion of substance P from primary sensory neurones. Brain Res 152:183, 1978

188. Watson CP, Evans RJ, Watt VR: Post-herpetic neuralgia and topical capsaicin. Pain 33:333, 1988

189. Crue BL Jr: Neuritis, neuropathy, and neuralgia. In Tollison CD (ed): Handbook of Pain Management, 2d ed, p 377. Baltimore, Williams & Wilkins, 1994

190. Bowsher D: Postherpetic neuralgia and its treatment: A retrospective survey of 191 patients. J Pain Symptom Manage 12:290, 1996

191. Bowsher D: The effects of pre-emptive treatment of postherpetic neuralgia with amitriptyline: A randomized, double-blind, placebo-controlled trial. J Pain Symptom Manage 13:327, 1997

192. Bennett GJ: Hypotheses on the pathogenesis of herpes zoster-associated pain. Ann Neurol 35:S38, 1994

193. Hanania MM, Brietstein D: Postherpetic neuralgia: A review. Cancer Invest 15:165, 1997

194. Takahashi M: The varicella vaccine.Vaccine development. Infect Dis Clin North Am 10:469, 1996

195. Gershon AA, Steinberg SP: Live attenuated varicella vaccine: Protection in healthy adults compared with leukemic children. National Institute of Allergy and Infectious Diseases Varicella Vaccine Collaborative Study Group. J Infect Dis 161:661, 1990

196. Halloran ME: Epidemiologic effects of varicella vaccination. Infect Dis Clin North Am 10:631, 1996

197. Lawrence R, Gershon AA, Holzman R, Steinberg SP: The risk of zoster after varicella vaccination in children with leukemia. N Engl J Med 318:543, 1988

198. Gershon AA, Steinberg SP, Gelb L et al: Live attenuated varicella vaccine. Efficacy for children with leukemia in remission. JAMA 252:355, 1984

199. Hardy I, Gershon AA, Steinberg SP, LaRussa P: The incidence of zoster after immunization with live attenuated varicella vaccine. A study in children with leukemia. Varicella Vaccine Collaborative Study Group. N Engl J Med 325:1545, 1991

200. Brunell PA, Taylor-Wiedeman J, Geiser CF et al: Risk of herpes zoster in children with leukemia: Varicella vaccine compared with history of chickenpox. Pediatrics 77:53, 1986

201. Berger R, Luescher D, Just M: Enhancement of varicella-zoster-specific immune responses in the elderly by boosting with varicella vaccine. J Infect Dis 149:647, 1984

202. Oxman MN: Immunization to reduce the frequency and severity of herpes zoster and its complications. Neurology 45:S41, 1995

203. Vafai A: Boosting immune response with a candidate VZV glycoprotein subunit vaccine. Vaccine 13:1336, 1995