CHAPTER 88: HERPES SIMPLEX VIRUS INFECTIONS

Herpes simplex viruses (HSV-1, HSV-2; Herpesvirus hominis) produce a variety of infections involving mucocutaneous surfaces, the central nervous system (CNS), and—on occasion—visceral organs. Prompt recognition and treatment reduce the morbidity and mortality rates associated with HSV infections.

The genome of HSV is a linear, double-strand DNA molecule (molecular weight, ~100 × 10⁶) that encodes >90 transcription units with 84 identified proteins. The genomic structures of the two HSV subtypes are similar. The overall genomic sequence homology between HSV-1 and HSV-2 is ~50%, whereas the proteome homology is >80%. The homologous sequences are distributed over the entire genome map, and most of the polypeptides specified by one viral type are antigenically related to polypeptides of the other viral type. Many type-specific regions unique to HSV-1 and HSV-2 proteins do exist, however, and a number of them appear to be important in host immunity. These type-specific regions have been used to develop serologic assays that distinguish between the two viral subtypes. Either restriction endonuclease analysis or sequencing of viral DNA can be used to distinguish between the two subtypes and among strains of each subtype. The variability of nucleotide sequences from clinical strains of HSV-1 and HSV-2 is such that HSV isolates obtained from two individuals can be differentiated by restriction enzyme patterns or genomic sequences. Moreover, epidemiologically related sources, such as sexual partners, mother-infant pairs, or persons involved in a common-source outbreak, can be inferred from such patterns.

The viral genome is packaged in a regular icosahedral protein shell (capsid) composed of 162 capsomeres (see Fig. 86-1). The outer covering of the virus is a lipid-containing membrane (envelope) acquired as the DNA-containing capsid buds through the inner nuclear membrane of the host cell. Between the capsid and lipid bilayer of the envelope is the tegument. Viral replication has both nuclear and cytoplasmic phases. Initial attachment to the cell membrane involves interactions of viral glycoproteins C and B with several cellular heparan sulfate–like surface receptors. Subsequently, viral glycoprotein D binds to cellular co-receptors that belong to the tumor necrosis factor receptor family of proteins, the immunoglobulin superfamily (nectin family), or both. The ubiquity of these receptors contributes to the wide host range of herpesviruses. HSV replication is highly regulated. After fusion and entry, the nucleocapsid enters the cytoplasm and several viral proteins are released from the virion. Some of these viral proteins shut off host protein synthesis (by increasing cellular RNA degradation), whereas others “turn on” the transcription of early genes of HSV replication. These early gene products, designated α genes, are required for synthesis of the subsequent polypeptide group, the β polypeptides, many of which are regulatory proteins and enzymes required for DNA replication. Most current antiviral drugs interfere with β proteins, such as viral DNA polymerase. The third (γ) class of HSV genes requires viral DNA replication for expression and encodes most structural proteins specified by the virus.

After viral genome replication and structural protein synthesis, nucleocapsids are assembled in the cell’s nucleus. Envelopment occurs as the nucleocapsids bud through the inner nuclear membrane into the perinuclear space. In some cells, viral replication in the nucleus forms two types of inclusion bodies: type A basophilic Feulgen-positive bodies that contain viral DNA and eosinophilic inclusion bodies that are devoid of viral nucleic acid or protein and represent a “scar” of viral infection. Enveloped virions are then transported via the endoplasmic reticulum and the Golgi apparatus to the cell surface.

Viral genomes are maintained by some neuronal cells in a repressed state called latency. Latency, which is associated with transcription of only a limited number of virus-encoded RNAs, accounts for the presence of viral DNA and RNA in neural tissue at times when infectious virus cannot be isolated. Maintenance and growth of neural cells from latently infected ganglia in tissue culture result in production of infectious virions (explantation) and in subsequent permissive infection of susceptible cells (co-cultivation). Activation of the viral genome may then occur, resulting in reactivation—the normal pattern of regulated viral gene expression and replication and HSV release. The release of virions from the neuron follows a complex process of anterograde transport down the length of neuronal axons. In experimental animals, ultraviolet light, systemic and local immunosuppression, and trauma to the skin or ganglia are associated with reactivation.

Three noncoding RNA latency-associated transcripts (LATs) are found in the nuclei of latently infected neurons. Deletion mutants of the LAT region exhibit reduced efficiency in their later reactivation. Substitution of HSV-1 LATs for HSV-2 LATs induces an HSV-1 reactivation pattern. These data indicate that LATs apparently maintain—rather than establish—latency. HSV-1 LATs promote the survival of acutely infected neurons, perhaps by inhibiting apoptotic pathways. LAT transcript abundance and low genome-copy number correlate with subnuclear positioning of HSV genomes around the centromere. Indeed, chromatization of HSV DNA appears to play a vital role in silencing expression of lytic replication genes. Highly expressed during latency, LAT-derived micro-RNA appears to silence expression of the key neurovirulence factor infected-cell protein 34.5 (ICP34.5) and to bind in an antisense configuration to the immediate-early protein ICP0 messenger RNA to prevent expression, which is vital to HSV reactivation. Although certain viral transcripts are known to be necessary for reactivation from latency, the molecular mechanisms of HSV latency are not fully understood, and strategies to interrupt or maintain latency in neurons are in developmental stages.

While latency is the predominant state of virus on a per-neuron basis, the high frequency of oral and genital tract reactivation for HSV-1 and HSV-2 suggests that the viruses are rarely quiescent within the entire biomass of ganglionic tissue. Recent data indicate that HSV-2 antigen is often shed: most persons infected with HSV-2 have frequent subclinical bursts of reactivation lasting 2–4 h, and the host mucosal immune system can contain viral reactivation in the mucosa before the development of clinical reactivation. Supporting this clinical observation, recent work using microdissection plus real-time polymerase chain reaction (PCR) of individual neurons from cadaveric trigeminal ganglia explants revealed that many more neurons (2–10%) harbor HSV than would be predicted by in situ hybridization studies for LATs. Viral copy number is highly variable between neurons, with extremely high levels in certain neurons, and HSV DNA copy numbers are similar in LAT-positive and LAT-negative neurons; these findings add to the uncertainty about the role that LATs play in preventing reactivation.

Exposure to HSV at mucosal surfaces or abraded skin sites permits entry of the virus into cells of the epidermis and dermis and initiation of viral replication therein. HSV infections are usually acquired subclinically. Whether clinical or subclinical, HSV acquisition is associated with sufficient viral replication to permit infection of either sensory or autonomic nerve endings. On entry into the neuronal cell, the virus—or, more likely, the nucleocapsid—is transported intra-axonally to the nerve cell bodies in ganglia. In humans, the transit interval of spread to the ganglia after virus inoculation into peripheral tissue is unknown. During the initial phase of infection, viral replication occurs in ganglia and contiguous neural tissue. Virus then spreads to other mucocutaneous surfaces through centrifugal migration of infectious virions via peripheral sensory nerves. This mode of spread helps explain the large surface area involved, the high frequency of new lesions distant from the initial crop of vesicles that is characteristic in patients with primary genital or oral-labial HSV infection, and the ability to recover virus from neural tissue distant from neurons innervating the inoculation site. Contiguous spread of locally inoculated virus also may take place and allow further mucosal extension of disease. Recent studies have demonstrated HSV viremia—another mechanism for extension of infection throughout the body—in ~30–40% of persons with primary HSV-2 infection. Latent infection with both viral subtypes in both sensory and autonomic ganglia has been demonstrated. For HSV-1 infection, trigeminal ganglia are most commonly infected, although extension to the inferior and superior cervical ganglia also occurs. With genital infection, sacral nerve root ganglia (S2–S5) are most commonly affected.

After resolution of primary disease, infectious HSV can no longer be cultured from the ganglia; however, latent infection, as defined by the presence of viral DNA, persists in 2–11% of ganglionic cells in the anatomic region of the initial infection. The mechanism of reactivation from latency is unknown. Increasingly, studies indicate that host T cell responses at the ganglionic and peripheral mucosal levels influence the frequency and severity of HSV reactivation. HSV-specific T cells have been recovered from peripheral-nerve root ganglia. Many of these resident CD8+ T cells are juxtaposed with latently HSV-1-infected neurons in the trigeminal ganglia and can block reactivation with both interferon (IFN) γ release and granzyme B degradation of the immediate-early protein ICP4. In addition, there appears to be a latent viral load in the ganglia that correlates positively with the number of neurons infected and the rate of reactivation but inversely with the number of CD8+ cells present. It is not known whether reactivating stimuli transiently suppress these immune cells, independently upregulate transcription of lytic genes, or both. Moreover, host containment in the mucosa has been demonstrated. Once virus reaches the dermal-epidermal junction, there are three possible outcomes: rapid host containment of infection near the site of reactivation; spread of virus into the epidermis, with a micro-ulceration associated with low-titer subclinical shedding; and subsequent rapid (within hours) containment of virus with widespread replication and necrosis of epithelial cells and subsequent clinical recurrence (the latter defined clinically by a skin blister and ulceration). Histologically, herpetic lesions involve a thin-walled vesicle or ulceration in the basal region, multinucleated cells that may include intranuclear inclusions, necrosis, and an acute inflammatory infection. Re-epithelialization occurs once viral replication is restricted, almost always in the absence of a scar.

Analysis of the DNA from sequential isolates of HSV or from isolates from multiple infected ganglia in any one individual has revealed similar, if not identical, restriction endonuclease or DNA sequence patterns in most persons. As more sensitive genomic technologies are developed, evidence of multiple strains of the same subtype is increasingly being reported. For example, infection of individual neurons with multiple strains of drug-susceptible and drug-resistant virus in severely immunosuppressed patients indicates that ganglia can be reseeded during chronic infection. Because exposure to mucosal shedding is relatively common during a person’s lifetime, current data suggest that exogenous infection with different strains of the same subtype, while possible, is uncommon.

Host responses influence the acquisition of HSV disease, the severity of infection, resistance to the development of latency, the maintenance of latency, and the frequency of recurrences. Both antibody-mediated and cell-mediated reactions are clinically important. Immunocompromised patients with defects in cell-mediated immunity experience more severe and more extensive HSV infections than those with deficits in humoral immunity, such as agammaglobulinemia. Experimental ablation of lymphocytes indicates that T cells play a major role in preventing lethal disseminated disease, although antibodies help reduce titers of virus in neural tissue. Some clinical manifestations of HSV appear to be related to the host immune response (e.g., stromal opacities associated with recurrent herpetic keratitis). The surface viral glycoproteins have been shown to be targets of antibodies that mediate neutralization and immune-mediated cytolysis (antibody-dependent cell-mediated cytotoxicity). Monoclonal antibodies specific for each of the known viral glycoproteins have, in experimental infections, conferred protection against subsequent neurologic disease or ganglionic latency. In humans, however, subunit glycoprotein vaccines have been largely ineffective in reducing acquisition of infection. Multiple cell populations, including natural killer cells, macrophages, and a variety of T lymphocytes, play a role in host defenses against HSV infections, as do lymphokines generated by T lymphocytes. In animals, passive transfer of primed lymphocytes confers protection from subsequent HSV challenge. Maximal protection usually requires the activation of multiple T cell subpopulations, including cytotoxic T cells and T cells responsible for delayed hypersensitivity. The latter may confer protection by the antigen-stimulated release of lymphokines (e.g., IFNs), which in turn have a direct antiviral effect and both activate and enhance a variety of specific and nonspecific effector cells. The HSV virion contains a variety of genes that are directed at the inhibition of host responses. These include gene no. 12 (US-12), which can bind to the cellular transporter-activating protein TAP-1 and reduce the ability of this protein to bind HSV peptides to human leukocyte antigen (HLA) class I, thereby reducing recognition of viral proteins by cytotoxic T cells of the host. This effect can be overcome by the addition of IFN-γ, but this reversal requires 24–48 h; thus, the virus has time to replicate and invade other host cells. Entry of infectious HSV-1 and HSV-2 inhibits several signaling pathways of both CD4+ and CD8+ T cells, leading to their functional impairment in killing and influencing the spectrum of their cytokine secretion.

Increasing evidence suggests that HSV-specific CD8+ T cell responses are critical for clearance of virus from lesions. Immunosuppressed patients with frequent and prolonged HSV lesions have fewer functional CD8+ T cells directed at HSV. HSV-specific CD8+ T cells have been shown to persist in the genital skin at the dermal–epidermal junction contiguous to neuronal endings. Even during clinical quiescence, these CD8+ T cells make both antiviral and cytotoxic proteins indicative of immune surveillance. These resident memory CD8+ T cells appear to be “first responders” capable of controlling viral reactivation at the site of viral release into the dermis. This rapid “on and off” interplay between the virus and host helps explain the variability in clinical disease severity between episodes in any single individual. Differences of 30–60 min in host responses can result in 100- to 1000-fold differences in viral levels and can determine whether an episode of disease is subclinical or clinical.

There is a strong association between the magnitude of the CD8+ T lymphocyte response and the clearance of virus from genital lesions. The location, effectiveness, and longevity of the T lymphocytes (and perhaps of other immune effector cells) may be important in the expression of disease and the likelihood of transmission over time.

Seroepidemiologic studies have documented HSV infections worldwide. The past 15 years have shown that the prevalence of HSV-2 is even higher in the developing than in the developed world. In sub-Saharan Africa, HSV-2 seroprevalence among pregnant women may approach 60%, and annual acquisition rates among teenage girls may verge on 20%. The global incidence has been estimated at ~23.6 million infections per year. As in the developed world, the rate of HSV-2 coital acquisition as well as the serologic prevalence is higher among women than among men. Most of this HSV-2 acquisition is preceded by acquisition of HSV-1; the frequency of genital HSV-1 in the developing world is low at present.

Infection with HSV-1 is acquired more frequently and earlier than infection with HSV-2. More than 90% of adults have antibodies to HSV-1 by the fifth decade of life. In populations of low socioeconomic status, most persons acquire HSV-1 infection before the third decade of life. Antibodies to HSV-2 are not detected routinely until puberty. Antibody prevalence rates correlate with past sexual activity and vary greatly among different population groups. There is evidence that the prevalence of HSV-2 has decreased slightly over the past decade in the United States. Serosurveys indicate that 15–20% of the U.S. population has antibodies to HSV-2. In most routine obstetric and family planning clinics, 25% of women have HSV-2 antibodies, although only 10% of those who are seropositive for HSV-2 report a history of genital lesions. As many as 50% of heterosexual adults attending sexually transmitted disease clinics have antibodies to HSV-2.

Many studies continue to show that both incident and—more important—prevalent HSV-2 infection enhances the acquisition rate of HIV-1. More specifically, HSV-2 infection is associated with a two- to fourfold increase in HIV-1 acquisition. This association has been amply demonstrated in heterosexual men and women in both the developed and the developing worlds. Epidemiologically, regions of the world with high HSV-2 prevalence and selected populations within such regions have a higher population-based incidence of HIV-1. One study indicated that approximately one-quarter of HIV infections in the high-prevalence city of Kisumu, Kenya, were directly attributable to HSV-2.

In addition, HSV-2 facilitates the spread of HIV into low-risk populations on a per-coital basis, and prevalent HSV-2 appears to increase the risk of HIV infection by seven- to ninefold. Mathematical models suggest that ~33–50% of HIV-1 infections may be attributable to HSV-2 both in men who have sex with men (MSM) and in sub-Saharan Africa. In addition, HSV-2 is more frequently reactivated in and transmitted by persons co-infected with HIV-1 as opposed to persons not co-infected. Thus, most areas of the world with a high HIV-1 prevalence also have a high HSV-2 prevalence. A wide variety of serologic surveys have indicated a similar or even higher seroprevalence of HSV-2 in most parts of Central America, South America, and Africa. In Africa, HSV-2 seroprevalence has ranged from 40% to 70% in obstetric and other sexually experienced populations. Antibody prevalence rates average ~5–10% higher among women than among men.

Several studies suggest that many cases of “asymptomatic” genital HSV-2 infection are, in fact, simply unrecognized or confined to anatomic regions of the genital tract that are not easily visualized. Asymptomatic seropositive persons shed virus on mucosal surfaces almost as frequently as do those with symptomatic disease. This large reservoir of unidentified carriers of HSV-2 and the frequent asymptomatic reactivation of the virus from the genital tract have fostered the continued spread of genital herpes throughout the world. HSV-2 infection is an independent risk factor for the acquisition and transmission of infection with HIV-1. Among co-infected persons, HIV-1 virions can be shed from herpetic lesions of the genital region. This shedding may facilitate the spread of HIV through sexual contact. HSV-2 reactivation is associated with a localized persistent inflammatory response consisting of high concentrations of CCR5-enriched CD4+ T cells as well as inflammatory dendritic cells in the submucosa of the genital skin. These cells can support HIV infection and replication and hence are likely to account for the almost threefold increase in HIV acquisition among persons with genital herpes. Unfortunately, antiviral therapy does not reduce this subclinical postreactivation inflammation, probably because of the inability of current antiviral agents to prevent the release of small amounts of HSV antigen into the genital mucosa.

HSV infections occur throughout the year. Transmission can result from contact with persons who have active ulcerative lesions or with persons who have no clinical manifestations of infection but who are shedding HSV from mucocutaneous surfaces. HSV reactivation on genital skin and mucosal surfaces is common. The frequency of sampling influences the frequency of detection. Recent studies indicate that most HSV-1 and HSV-2 episodes last <4–6 h; thus, replication of virus and clearance by the host are rapid. Even with once-daily sampling, HSV DNA can be detected on 20–30% of days by PCR. Corresponding figures for HSV-1 in oral secretions are similar. Rates of shedding are highest during the initial years after acquisition, with viral shedding occurring on as many as 30–50% of days during this period. Immunosuppressed patients shed HSV from mucosal sites at an even higher frequency (20–80% of days). These high rates of mucocutaneous reactivation suggest that exposure to HSV from sexual or other close contact (kissing, sharing of glasses or silverware) is common and help explain the continuing spread and high seroprevalence of HSV infections worldwide. Reactivation rates vary widely among individuals. Among HIV-positive patients, a low CD4+ T cell count and a high HIV-1 load are associated with increased rates of HSV reactivation. Daily antiviral chemotherapy for HSV-2 infection can reduce shedding rates but does not eliminate shedding, as measured by PCR or culture.

HSV has been isolated from nearly all visceral and mucocutaneous sites. The clinical manifestations and course of HSV infection depend on the anatomic site involved, the age and immune status of the host, and the antigenic type of the virus. Primary HSV infections (i.e., first infections with either HSV-1 or HSV-2 in which the host lacks HSV antibodies in acute-phase serum) are frequently accompanied by systemic signs and symptoms. Compared with recurrent episodes, primary infections, which involve both mucosal and extramucosal sites, are characterized by a longer duration of symptoms and virus isolation from lesions. The incubation period ranges from 1 to 26 days (median, 6–8 days). Both viral subtypes can cause genital and oral-facial infections, and the infections caused by the two subtypes are clinically indistinguishable. However, the frequency of reactivation of infection is influenced by anatomic site and virus type. Genital HSV-2 infection is twice as likely to reactivate and recurs 8–10 times more frequently than genital HSV-1 infection. Conversely, oral-labial HSV-1 infection recurs more frequently than oral-labial HSV-2 infection. Asymptomatic shedding rates follow the same pattern.

Oral-facial infections

Gingivostomatitis and pharyngitis are the most common clinical manifestations of first-episode HSV-1 infection, whereas recurrent herpes labialis is the most common clinical manifestation of reactivation HSV-1 infection. HSV pharyngitis and gingivostomatitis usually result from primary infection and are most common among children and young adults. Clinical symptoms and signs, which include fever, malaise, myalgias, inability to eat, irritability, and cervical adenopathy, may last 3–14 days. Lesions may involve the hard and soft palate, gingiva, tongue, lip, and facial area. HSV-1 or HSV-2 infection of the pharynx usually results in exudative or ulcerative lesions of the posterior pharynx and/or tonsillar pillars. Lesions of the tongue, buccal mucosa, or gingiva may occur later in the course in one-third of cases. Fever lasting 2–7 days and cervical adenopathy are common. It can be difficult to differentiate HSV pharyngitis clinically from bacterial pharyngitis, Mycoplasma pneumoniae infections, and pharyngeal ulcerations of noninfectious etiologies (e.g., Stevens-Johnson syndrome). No substantial evidence suggests that reactivation of oral-labial HSV infection is associated with symptomatic recurrent pharyngitis.

Reactivation of HSV from the trigeminal ganglia may be associated with asymptomatic virus excretion in the saliva, development of intraoral mucosal ulcerations, or herpetic ulcerations on the vermilion border of the lip or external facial skin. About 50–70% of seropositive patients undergoing trigeminal nerve-root decompression and 10–15% of those undergoing dental extraction develop oral-labial HSV infection a median of 3 days after these procedures. Clinical differentiation of intraoral mucosal ulcerations due to HSV from aphthous, traumatic, or drug-induced ulcerations is difficult.

In immunosuppressed patients, HSV infection may extend into mucosal and deep cutaneous layers. Friability, necrosis, bleeding, severe pain, and inability to eat or drink may result. The lesions of HSV mucositis are clinically similar to mucosal lesions caused by cytotoxic drug therapy, trauma, or fungal or bacterial infections. Persistent ulcerative HSV infections are among the most common infections in patients with AIDS. HSV and Candida infections often occur concurrently. Systemic antiviral therapy speeds the rate of healing and relieves the pain of mucosal HSV infections in immunosuppressed patients. The frequency of HSV reactivation during the early phases of transplantation or induction chemotherapy is high (50–90%), and prophylactic systemic antiviral agents such as IV acyclovir and penciclovir or the oral congeners of these drugs are used to reduce reactivation rates. Patients with atopic eczema may also develop severe oral-facial HSV infections (eczema herpeticum), which may rapidly involve extensive areas of skin and occasionally disseminate to visceral organs. Extensive eczema herpeticum has resolved promptly with the administration of IV acyclovir. Erythema multiforme may also be associated with HSV infections (see Fig. 14-25); some evidence suggests that HSV infection is the precipitating event in ~75% of cases of cutaneous erythema multiforme. HSV antigen has been demonstrated both in circulatory immune complexes and in skin lesion biopsy samples from these cases. Patients with severe HSV-associated erythema multiforme are candidates for chronic suppressive oral antiviral therapy.

HSV-1 and varicella-zoster virus (VZV) have been implicated in the etiology of Bell’s palsy (flaccid paralysis of the mandibular portion of the facial nerve). Some but not all trials have documented quicker resolution of facial paralysis with the prompt initiation of antiviral therapy, with or without glucocorticoids. However, other trials have shown little benefit. Thus there is no consensus on the relative value of antiviral drugs alone, glucocorticoids alone, and the two modalities combined for the treatment of Bell’s palsy.

Genital infections

First-episode primary genital herpes is characterized by fever, headache, malaise, and myalgias. Pain, itching, dysuria, vaginal and urethral discharge, and tender inguinal lymphadenopathy are the predominant local symptoms. Widely spaced bilateral lesions of the external genitalia are characteristic (Fig. 88-1). Lesions may be present in varying stages, including vesicles, pustules, or painful erythematous ulcers. The cervix and urethra are involved in >80% of women with first-episode infections. First episodes of genital herpes in patients who have had prior HSV-1 infection are associated with systemic symptoms in a few patients and with faster healing than primary genital herpes. Subclinical DNAemia has been found in ~30% of cases of true primary genital herpes. The clinical courses of acute first-episode genital herpes are similar for HSV-1 and HSV-2 infection. However, the recurrence rates of genital disease differ with the viral subtype: the 12-month recurrence rates among patients with first-episode HSV-2 and HSV-1 infections are ~90% and ~55%, respectively (median number of recurrences, 4 and <1, respectively). Recurrence rates for genital HSV-2 infections vary greatly among individuals and over time within the same individual. HSV has been isolated from the urethra and urine of men and women without external genital lesions. A clear mucoid discharge and dysuria are characteristics of symptomatic HSV urethritis. HSV has been isolated from the urethra of 5% of women with the dysuria-frequency syndrome. Occasionally, HSV genital tract disease is manifested by endometritis and salpingitis in women and by prostatitis in men. About 15% of cases of HSV-2 acquisition are associated with nonlesional clinical syndromes, such as aseptic meningitis, cervicitis, or urethritis. A more complete discussion of the differential diagnosis of genital herpes is presented in Chap. 35.

Both HSV-1 and HSV-2 can cause symptomatic or asymptomatic rectal and perianal infections. HSV proctitis is usually associated with rectal intercourse. However, subclinical perianal shedding of HSV is detected in women and men who report no rectal intercourse. This phenomenon is due to the establishment of latency in the sacral dermatome from prior genital tract infection, with subsequent reactivation in epithelial cells in the perianal region. Such reactivations are often subclinical. Symptoms of HSV proctitis include anorectal pain, anorectal discharge, tenesmus, and constipation. Sigmoidoscopy reveals ulcerative lesions of the distal 10 cm of the rectal mucosa. Rectal biopsies show mucosal ulceration, necrosis, polymorphonuclear and lymphocytic infiltration of the lamina propria, and (in occasional cases) multinucleated intranuclear inclusion-bearing cells. Perianal herpetic lesions are also found in immunosuppressed patients receiving cytotoxic therapy. Extensive perianal herpetic lesions and/or HSV proctitis is common among patients with HIV infection.

Herpetic whitlow

Herpetic whitlow—HSV infection of the finger—may occur as a complication of primary oral or genital herpes by inoculation of virus through a break in the epidermal surface or by direct introduction of virus into the hand through occupational or some other type of exposure. Clinical signs and symptoms include abrupt-onset edema, erythema, and localized tenderness of the infected finger. Vesicular or pustular lesions of the fingertip that are indistinguishable from lesions of pyogenic bacterial infection are seen. Fever, lymphadenitis, and epitrochlear and axillary lymphadenopathy are common. The infection may recur. Prompt diagnosis (to avoid unnecessary and potentially exacerbating surgical therapy and/or transmission) is essential. Antiviral chemotherapy is usually recommended (see below).

Herpes gladiatorum

HSV may infect almost any area of skin. Mucocutaneous HSV infections of the thorax, ears, face, and hands have been described among wrestlers. Transmission of these infections is facilitated by trauma to the skin sustained during wrestling. Several recent outbreaks have illustrated the importance of prompt diagnosis and therapy to contain the spread of this infection.

Eye infections

HSV infection of the eye is the most common cause of corneal blindness in the United States. HSV keratitis presents as an acute onset of pain, blurred vision, chemosis, conjunctivitis, and characteristic dendritic lesions of the cornea. Use of topical glucocorticoids may exacerbate symptoms and lead to involvement of deep structures of the eye. Debridement, topical antiviral treatment, and/or IFN therapy hastens healing. However, recurrences are common, and the deeper structures of the eye may sustain immunopathologic injury. Stromal keratitis due to HSV appears to be related to T cell–dependent destruction of deep corneal tissue. An HSV-1 epitope that is autoreactive with T cell–targeting corneal antigens has been postulated to be a factor in this infection. Chorioretinitis, usually a manifestation of disseminated HSV infection, may occur in neonates or in patients with HIV infection. HSV and VZV can cause acute necrotizing retinitis as an uncommon but severe manifestation.

Central and peripheral nervous system infections

HSV accounts for 10–20% of all cases of sporadic viral encephalitis in the United States. The estimated incidence is ~2.3 cases per 1 million persons per year. Cases are distributed throughout the year, and the age distribution appears to be biphasic, with peaks at 5–30 and >50 years of age. HSV-1 causes >95% of cases.

The pathogenesis of HSV encephalitis varies. In children and young adults, primary HSV infection may result in encephalitis; presumably, exogenously acquired virus enters the CNS by neurotropic spread from the periphery via the olfactory bulb. However, most adults with HSV encephalitis have clinical or serologic evidence of mucocutaneous HSV-1 infection before the onset of CNS symptoms. In ~25% of the cases examined, the HSV-1 strains from the oropharynx and brain tissue of the same patient differ; thus some cases may result from reinfection with another strain of HSV-1 that reaches the CNS. Two theories have been proposed to explain the development of actively replicating HSV in localized areas of the CNS in persons whose ganglionic and CNS isolates are similar. Reactivation of latent HSV-1 infection in trigeminal or autonomic nerve roots may be associated with extension of virus into the CNS via nerves innervating the middle cranial fossa. HSV DNA has been demonstrated by DNA hybridization in brain tissue obtained at autopsy—even from healthy adults. Thus, reactivation of long-standing latent CNS infection may be another mechanism for the development of HSV encephalitis.

Recent studies have identified genetic polymorphisms in two separate genes among families with a high frequency of HSV encephalitis. Peripheral-blood mononuclear cells from these patients (predominantly children) appear to secrete reduced levels of IFN in response to HSV. These observations suggest that some cases of sporadic HSV encephalitis may be related to host genetic determinants.

The clinical hallmark of HSV encephalitis has been the acute onset of fever and focal neurologic symptoms and signs, especially in the temporal lobe (Fig. 88-2). Clinical differentiation of HSV encephalitis from other viral encephalitides, focal infections, or noninfectious processes is difficult. Elevated cerebrospinal fluid (CSF) protein levels, leukocytosis (predominantly lymphocytes), and red blood cell counts due to hemorrhagic necrosis are common. While brain biopsy has been the gold standard for defining HSV encephalitis, a highly sensitive and specific PCR for detection of HSV DNA in CSF has largely replaced biopsy for defining CNS infection. Although titers of antibody to HSV in CSF and serum increase in most cases of HSV encephalitis, they rarely do so earlier than 10 days into the illness and therefore, although useful in retrospect, generally are not helpful in establishing an early clinical diagnosis. In rare cases, demonstration of HSV antigen, HSV DNA, or HSV replication in brain tissue obtained by biopsy is highly sensitive; examination of such tissue also provides the opportunity to identify alternative, potentially treatable causes of encephalitis. Antiviral chemotherapy with acyclovir reduces the rate of death from HSV encephalitis. Most authorities recommend the administration of IV acyclovir to patients with presumed HSV encephalitis until the diagnosis is confirmed or an alternative diagnosis is made. All confirmed cases should be treated with IV acyclovir (30 mg/kg per day in three divided doses for 14–21 days). After the completion of therapy, the clinical recurrence of encephalitis requiring more treatment has been reported. For this reason, some authorities prefer to treat initially for 21 days, and many continue therapy until HSV DNA has been eliminated from the CSF. Even with therapy, neurologic sequelae are common, especially among persons >50 years of age.

HSV DNA has been detected in CSF from 3–15% of persons presenting to the hospital with aseptic meningitis. HSV meningitis, which is usually seen in association with primary genital HSV infection, is an acute, self-limited disease manifested by headache, fever, and mild photophobia and lasting 2–7 days. Lymphocytic pleocytosis in the CSF is characteristic. Neurologic sequelae of HSV meningitis are rare. HSV is the most commonly identified cause of recurrent lymphocytic meningitis (Mollaret’s meningitis). Demonstration of HSV antibodies in CSF or persistence of HSV DNA in CSF can establish the diagnosis. For persons with frequent recurrences of HSV meningitis, daily anti-viral therapy has reduced the occurrence of such episodes.

Autonomic nervous system dysfunction, especially of the sacral region, has been reported in association with both HSV and VZV infections. Numbness, tingling of the buttocks or perineal areas, urinary retention, constipation, CSF pleocytosis, and (in males) impotence may occur. Symptoms appear to resolve slowly over days or weeks. Occasionally, hypoesthesia and/or weakness of the lower extremities persists for many months. Transitory hypoesthesia of the area of skin innervated by the trigeminal nerve and vestibular system dysfunction (as measured by electronystagmography) are the predominant signs of disease. Whether antiviral chemotherapy can abort these signs or reduce their frequency and severity is not yet known. Rarely, transverse myelitis, manifested by a rapidly progressive symmetric paralysis of the lower extremities or Guillain-Barré syndrome, follows HSV infection. Similarly, peripheral nervous system involvement (Bell’s palsy) or cranial polyneuritis may be related to reactivation of HSV-1 infection.

Visceral infections

HSV infection of visceral organs usually results from viremia, and multiple-organ involvement is common. Occasionally, however, the clinical manifestations of HSV infection involve only the esophagus, lung, or liver. HSV esophagitis may result from direct extension of oral-pharyngeal HSV infection into the esophagus or may occur de novo by reactivation and spread of HSV to the esophageal mucosa via the vagus nerve. The predominant symptoms of HSV esophagitis are odynophagia, dysphagia, substernal pain, and weight loss. Multiple oval ulcerations appear on an erythematous base with or without a patchy white pseudomembrane. The distal esophagus is most commonly involved. With extensive disease, diffuse friability may spread to the entire esophagus. Neither endoscopic nor barium examination can reliably differentiate HSV esophagitis from Candida esophagitis or from esophageal ulcerations due to thermal injury, radiation, or corrosives. Endoscopically obtained secretions for cytologic examination and culture or DNA detection by PCR provide the most useful material for diagnosis. Systemic antiviral chemotherapy usually reduces the severity and duration of symptoms and heals esophageal ulcerations.

HSV pneumonitis is uncommon except in severely immunosuppressed patients and may result from extension of herpetic tracheobronchitis into lung parenchyma. Focal necrotizing pneumonitis usually ensues. Hematogenous dissemination of virus from sites of oral or genital mucocutaneous disease may also occur, producing bilateral interstitial pneumonitis. Bacterial, fungal, and parasitic pathogens are commonly present in HSV pneumonitis. The mortality rate from untreated HSV pneumonia in immunosuppressed patients is high (>80%). HSV has also been isolated from the lower respiratory tract of persons with acute respiratory distress syndrome and prolonged intubation. Most authorities believe that the presence of HSV in tracheal aspirates in such settings is due to reactivation of HSV in the tracheal region and localized tracheitis in persons with long-term intubation. Such patients should be evaluated for extension of HSV infection into the lung parenchyma. Controlled trials assessing the role of antiviral agents used against HSV in morbidity and mortality associated with acute respiratory distress syndrome have not been conducted. The role of lower respiratory tract HSV infection in overall rates of morbidity and mortality associated with these conditions is unclear. HSV is an uncommon cause of hepatitis in immunocompetent patients. HSV infection of the liver is associated with fever, abrupt elevations of bilirubin and serum aminotransferase levels, and leukopenia (<4000 white blood cells/μL). Disseminated intravascular coagulation may also develop.

Other reported complications of HSV infection include monarticular arthritis, adrenal necrosis, idiopathic thrombocytopenia, and glomerulonephritis. Disseminated HSV infection in immunocompetent patients is rare. In immunocompromised patients, burn patients, or malnourished individuals, HSV occasionally disseminates to other visceral organs, such as the adrenal glands, pancreas, small and large intestines, and bone marrow. Rarely, primary HSV infection in pregnancy disseminates and may be associated with the death of both mother and fetus. This uncommon event is usually related to the acquisition of primary infection in the third trimester. Disseminated HSV infection is best detected by the presence of HSV DNA in plasma or blood.

Neonatal HSV infections

Of all HSV-infected populations, neonates (infants younger than 6 weeks) have the highest frequency of visceral and/or CNS infection. Without therapy, the overall rate of death from neonatal herpes is 65%; <10% of neonates with CNS infection develop normally. Although skin lesions are the most commonly recognized features of disease, many infants do not develop lesions at all or do so only well into the course of disease. Neonatal infection is usually acquired perinatally from contact with infected genital secretions at delivery. Congenitally infected infants have been reported. Of neonatal HSV infections, 30–50% are due to HSV-1 and 50–70% to HSV-2. The risk of developing neonatal HSV infection is 10 times higher for an infant born to a mother who has recently acquired HSV than for other infants. Neonatal HSV-1 infections may also be acquired through postnatal contact with immediate family members who have symptomatic or asymptomatic oral-labial HSV-1 infection or through nosocomial transmission within the hospital. All neonates with presumed herpes should be treated with IV acyclovir and then placed on maintenance oral antiviral therapy for the first 6–12 months of life. Antiviral chemotherapy with high-dose IV acyclovir (60 mg/kg per day) has reduced the mortality rate from neonatal herpes to ~15%. However, rates of morbidity, especially among infants with HSV-2 infection involving the CNS, are still very high.

HSV in pregnancy

In the United States, 22% of all pregnant women and 55% of non-Hispanic black pregnant women are seropositive for HSV-2. However, the risk of mother-to-child transmission of HSV in the perinatal period is highest when the infection is acquired near the time of labor—that is, in previously HSV-seronegative women. The clinical manifestations of recurrent genital herpes—including the frequency of subclinical versus clinical infection, duration of lesions, pain, and constitutional symptoms—are similar in pregnant and nonpregnant women. Recurrences increase in frequency over the course of pregnancy. However, when women are seropositive for HSV-2 at the outset of pregnancy, no effect on neonatal outcomes (including birth weight and gestational age) is seen. First-episode infections in pregnancy have more severe consequences for mother and infant. Maternal visceral dissemination during the third trimester occasionally occurs, as does premature birth or intrauterine growth retardation. The acquisition of primary disease in pregnancy, whether related to HSV-1 or HSV-2, carries the risk of transplacental transmission of virus to the neonate and can result in spontaneous abortion, although this outcome is relatively uncommon. For newly acquired genital HSV infection during pregnancy, most authorities recommend treatment with acyclovir (400 mg three times daily) or valacyclovir (500–1000 mg twice daily) for 7–10 days. However, the impact of this intervention on transmission is unknown. The high HSV-2 prevalence rate in pregnancy and the low incidence of neonatal disease (1 case per 6000–20,000 live births) indicate that only a few infants are at risk of acquiring HSV. Therefore, cesarean section is not warranted for all women with recurrent genital disease. Because intrapartum transmission of infection accounts for the majority of cases, abdominal delivery need be considered only for women who are shedding HSV at delivery. Several studies have shown no correlation between recurrence of viral shedding before delivery and viral shedding at term. Hence, weekly virologic monitoring and amniocentesis are not recommended.

The frequency of transmission from mother to infant is markedly higher among women who acquire HSV near term (30–50%) than among those in whom HSV-2 infection is reactivated at delivery (<1%). Although maternal antibody to HSV-2 is protective, antibody to HSV-1 offers little or no protection against neonatal HSV-2 infection. Primary genital infection with HSV-1 leads to a particularly high risk of transmission during pregnancy and accounts for an increasing proportion of neonatal HSV cases. Moreover, during reactivation, HSV-1 appears more transmissible to the neonate than HSV-2. Only 2% of women who are seropositive for HSV-2 have HSV-2 isolated from cervical secretions at delivery, and only 1% of infants exposed in this manner develop infection, presumably because of the protective effects of maternally transferred antibodies and perhaps lower viral titers during reactivation. Despite the low frequency of transmission of HSV in this setting, 30–50% of infants with neonatal HSV are born to mothers with established genital herpes.

Isolation of HSV by cervicovaginal swab at the time of delivery is the greatest risk factor for intrapartum HSV transmission (relative risk = 346); however, culture-negative, PCR-positive cases of intrapartum transmission are well described. New acquisition of HSV (odds ratio [OR] = 49), isolation of HSV-1 versus HSV-2 (OR = 35), cervical versus vulvar HSV detection (OR = 15), use of fetal scalp electrodes (OR = 3.5), and young age confer further risk of transmission, whereas abdominal delivery is protective (OR = 0.14). Physical examination poorly predicts the absence of shedding, and PCR far exceeds culture in terms of sensitivity and speed. Therefore, PCR detection at the onset of labor should be used to aid clinical decision-making for women with HSV-2 antibody. Because cesarean section appears to be an effective means of reducing maternal-fetal transmission, patients with recurrent genital herpes should be encouraged to come to the hospital early at the time of delivery for careful examination of the external genitalia and cervix as well as collection of a swab sample for viral isolation. Women who have no evidence of lesions can have a vaginal delivery. The presence of active lesions on the cervix or external genitalia is an indication for cesarean delivery.

If first-episode exposure has occurred (e.g., if HSV serologies show that the mother is seronegative or if the mother is HSV-1-seropositive and the isolate at delivery is found to be HSV-2), many authorities would initiate antiviral therapy for the infant with IV acyclovir. At a minimum, samples for viral cultures and PCR should be obtained from the throat, nasopharynx, eyes, and rectum of these infants immediately and at 5- to 10-day intervals. Lethargy, skin lesions, or fever should be evaluated promptly. All infants from whom HSV is isolated 24 h after delivery should be treated with IV acyclovir at recommended doses.

Both clinical and laboratory criteria are useful for diagnosing HSV infections. A clinical diagnosis can be made accurately when characteristic multiple vesicular lesions on an erythematous base are present. However, herpetic ulcerations may resemble skin ulcerations of other etiologies. Mucosal HSV infections may also present as urethritis or pharyngitis without cutaneous lesions. Thus, laboratory studies to confirm the diagnosis and to guide therapy are recommended. While staining of scrapings from the base of the lesions with Wright’s, Giemsa’s (Tzanck preparation), or Papanicolaou’s stain to detect giant cells or intranuclear inclusions of Herpesvirus infection is a well-described procedure, few clinicians are skilled in this technique, the sensitivity of staining is low (<30% for mucosal swabs), and these cytologic methods do not differentiate between HSV and VZV infections.

HSV infection is best confirmed in the laboratory by detection of virus, viral antigen, or viral DNA in scrapings from lesions. HSV DNA detection by PCR is the most sensitive laboratory technique for detecting mucosal or visceral HSV infections and should be used when available. HSV causes a discernible cytopathic effect in a variety of cell culture systems, and this effect can be identified within 48–96 h after inoculation. Spin-amplified culture with subsequent staining for HSV antigen has shortened the time needed to identify HSV to <24 h. The sensitivity of all detection methods depends on the stage of the lesions (with higher sensitivity for vesicular than for ulcerative lesions), on whether the patient has a first or a recurrent episode of the disease (with higher sensitivity in first than in recurrent episodes), and on whether the sample is from an immunosuppressed or an immunocompetent patient (with more antigen or DNA in immunosuppressed patients). Laboratory confirmation permits subtyping of the virus; information on subtype may be useful epidemiologically and may help to predict the frequency of reactivation after first-episode oral-labial or genital HSV infection.

Serologic assays with whole-virus antigen preparations, such as complement fixation, neutralization, indirect immunofluorescence, passive hemagglutination, radioimmunoassay, and enzyme-linked immunosorbent assay, are useful for differentiating uninfected (seronegative) persons from those with past HSV-1 or HSV-2 infection, but they do not reliably distinguish between the two viral subtypes. Serologic assays that identify antibodies to type-specific surface proteins (epitopes) of the two viral subtypes have been developed and can distinguish reliably between the human antibody responses to HSV-1 and HSV-2. The most commonly used assays are those that measure antibodies to glycoprotein G of HSV-1 (gG1) and HSV-2 (gG2). A western blot assay that can detect several HSV type-specific proteins can also be used.

Acute- and convalescent-phase serum samples can be useful in demonstrating seroconversion during primary HSV-1 or HSV-2 infection. However, few available tests report titers, and increases in index values do not reflect first episodes in all patients. Serologic assays based on type-specific proteins should be used to identify asymptomatic carriers of HSV-1 or HSV-2. No reliable IgM method for defining acute HSV infection is available.

Several studies have shown that persons with previously unrecognized HSV-2 infection can be taught to identify symptomatic reactivations. Individuals seropositive for HSV-2 should be told about the high frequency of subclinical reactivation on mucosal surfaces that are not visible to the eye (e.g., cervix, urethra, perianal skin) or in microscopic ulcerations that may not be clinically symptomatic. Transmission of infection during such episodes is well established. HSV-2-seropositive persons should be educated about the high likelihood of subclinical shedding and the role condoms (male or female) may play in reducing transmission. Antiviral therapy with valacyclovir (500 mg once daily) has been shown to reduce the transmission of HSV-2 between sexual partners.

The success of efforts to control HSV disease on a population basis through suppressive antiviral chemotherapy and/or educational programs will be limited. Barrier forms of contraception (especially condoms) decrease the likelihood of transmission of HSV infection, particularly during periods of asymptomatic viral excretion. When lesions are present, HSV infection may be transmitted by skin-to-skin contact despite the use of a condom. Nevertheless, the available data suggest that consistent condom use is an effective means of reducing the risk of genital HSV-2 transmission. Chronic daily antiviral therapy with valacyclovir can also be partially effective in reducing acquisition of HSV-2, especially among susceptible women. There are no comparative efficacy studies of valacyclovir versus condom use. Most authorities suggest both approaches. The need for a vaccine to prevent acquisition of HSV infection is great, especially in light of the role HSV-2 plays in enhancing the acquisition and transmission of HIV-1.

A substantial portion of neonatal HSV cases could be prevented by reducing the acquisition of HSV by women in the third trimester of pregnancy. Neonatal HSV infection can result from either the acquisition of maternal infection near term or the reactivation of infection at delivery in the already-infected mother. Thus strategies for reducing neonatal HSV are complex. Some authorities have recommended that antiviral therapy with acyclovir or valacyclovir be given to HSV-2-infected women in late pregnancy as a means of reducing reactivation of HSV-2 at term. Data are not available to support the efficacy of this approach. Moreover, the high treatment-to-prevention ratio makes this a dubious public health approach, even though it can reduce the frequency of HSV-associated cesarean delivery.