Pediatric Annals

The Diagnosis and Treatment of Neonatal Herpes Simplex Virus Infection

Steve Kohl, MD

Abstract

Neonatal herpes simplex virus (HSV) infection remains a challenge in the United States. Because of the success of prevention efforts regarding congenital human immuodeficiency virus (HIV) infection, it is estimated that there are now between 5 and 10 babies born with HSV infection for each baby with HIV infection per year in the United States. Neonatal HSV infection continues to be an area of medical /legal concern as well.

EPIDEMIOLOGY

Beginning in the United States in the 1960s, coincident with the so-called sexual revolution, there has been an increase in the prevalence of genital herpes infection. The latest nationwide serosurveys in the 1990s reveal a prevalence of HSV-2 infection in the sexually active age group of 20.8%, a 30% increase from that seen in the 1980s.1 Concomitant with this increase have been reports of an increase in the incidence of neonatal HSV. In King County, Washington (the county containing Seattle), the incidence of neonatal HSV infection rose from 2.6 per 100 000 live births in 1966 to more than 10 times that level (28.2 per 100000) in 1982.2 The incidence in all of California, using computerized birth records from most large hospitals in the mid 1990s, was estimated to be 11.4 per 100 000.3 The most recent data from the Seattle group, reporting on 31 810 women studied prospectively (28% of whom were HSV-2 seropositive) revealed an incidence of one case of neonatal HSV infection per 2650 live births (37.7 per 100 000).4

There are approximately 500 to 1500 infants born per year in the United States with neonatal HSV infection. Most cases of neonatal HSV infection result from maternal genital acquisition of a new HSV type (eg, type 1 or 2 if HSV seronegative, or type 2 if HSV-I positive) rather than from reactivation of old latent infection (eg, HSV-2 recurrence). The risk of new HSV infection is to a large extent dependent on the discordance of the HSV serostatus of the two parents. The highest risk is for an HSV-2 negative woman to have sexual relations with an HSV-2 positive man, a situation found 10% of the time in a recent study.5 Indeed, the risk of new HSV acquisition anytime during pregnancy in 7046 women in Seattle was 2.1%.6 It is those women, who acquire a new HSV type late in pregnancy and deliver before antibody production, whose infants are at the highest risk of neonatal HSV infection. A woman with late term acquisition of a new genital infection with HSV-I or HSV-2 has a 33% to 50% chance of infecting her neonate. Those with a recurrence of prior genital HSV infection at delivery have only a 1% to 3% chance of infecting their newborns. In addition to the difference in risk, a new infection in the mother tends to result in earlier onset of disease (first week versus second week of life) and more severe infection (usually disseminated versus local infection) in the neonate.7 One must also be aware that primary oral infection in the mother near delivery, as well as any exposed herpetic lesion on any caretaker (such as a herpetic whitlow, fever blister, or breast lesion), represents a potential risk to the neonate. Indeed the unusual habit of an Orthodox Jewish ritual circumciser with inapparent recurrent oral HSV infection who would kiss the area of incision, resulted in several cases of neonatal HSV.

Most cases of newly acquired genital HSV infection in women are asymptomatic. Thus, while a positive maternal (or male consort) history of genital HSV infection is important, its absence does not exclude neonatal HSV. Indeed, in cases of neonatal HSV, only 15%…

Neonatal herpes simplex virus (HSV) infection remains a challenge in the United States. Because of the success of prevention efforts regarding congenital human immuodeficiency virus (HIV) infection, it is estimated that there are now between 5 and 10 babies born with HSV infection for each baby with HIV infection per year in the United States. Neonatal HSV infection continues to be an area of medical /legal concern as well.

EPIDEMIOLOGY

Beginning in the United States in the 1960s, coincident with the so-called sexual revolution, there has been an increase in the prevalence of genital herpes infection. The latest nationwide serosurveys in the 1990s reveal a prevalence of HSV-2 infection in the sexually active age group of 20.8%, a 30% increase from that seen in the 1980s.1 Concomitant with this increase have been reports of an increase in the incidence of neonatal HSV. In King County, Washington (the county containing Seattle), the incidence of neonatal HSV infection rose from 2.6 per 100 000 live births in 1966 to more than 10 times that level (28.2 per 100000) in 1982.2 The incidence in all of California, using computerized birth records from most large hospitals in the mid 1990s, was estimated to be 11.4 per 100 000.3 The most recent data from the Seattle group, reporting on 31 810 women studied prospectively (28% of whom were HSV-2 seropositive) revealed an incidence of one case of neonatal HSV infection per 2650 live births (37.7 per 100 000).4

There are approximately 500 to 1500 infants born per year in the United States with neonatal HSV infection. Most cases of neonatal HSV infection result from maternal genital acquisition of a new HSV type (eg, type 1 or 2 if HSV seronegative, or type 2 if HSV-I positive) rather than from reactivation of old latent infection (eg, HSV-2 recurrence). The risk of new HSV infection is to a large extent dependent on the discordance of the HSV serostatus of the two parents. The highest risk is for an HSV-2 negative woman to have sexual relations with an HSV-2 positive man, a situation found 10% of the time in a recent study.5 Indeed, the risk of new HSV acquisition anytime during pregnancy in 7046 women in Seattle was 2.1%.6 It is those women, who acquire a new HSV type late in pregnancy and deliver before antibody production, whose infants are at the highest risk of neonatal HSV infection. A woman with late term acquisition of a new genital infection with HSV-I or HSV-2 has a 33% to 50% chance of infecting her neonate. Those with a recurrence of prior genital HSV infection at delivery have only a 1% to 3% chance of infecting their newborns. In addition to the difference in risk, a new infection in the mother tends to result in earlier onset of disease (first week versus second week of life) and more severe infection (usually disseminated versus local infection) in the neonate.7 One must also be aware that primary oral infection in the mother near delivery, as well as any exposed herpetic lesion on any caretaker (such as a herpetic whitlow, fever blister, or breast lesion), represents a potential risk to the neonate. Indeed the unusual habit of an Orthodox Jewish ritual circumciser with inapparent recurrent oral HSV infection who would kiss the area of incision, resulted in several cases of neonatal HSV.

Most cases of newly acquired genital HSV infection in women are asymptomatic. Thus, while a positive maternal (or male consort) history of genital HSV infection is important, its absence does not exclude neonatal HSV. Indeed, in cases of neonatal HSV, only 15% to 30% of mothers had a positive history, as did 10% to 15 % of their sexual contacts.

CUNICAL PRESENTATION

There are three main, roughly equally occurring, clinical categories of neonatal HSV infection. The first is localized skin, eye, and mouth (SEM) infection, in which only skin, eyes or mucous membranes are clinically involved. Second is the central nervous system (CNS) form in which the cerebrospinal fluid (CSF) is abnormal by routine analysis, or is culture or polymerase chain reaction (PCR) positive for HSV. Last is the disseminated (DISSEM) form, which is diagnosed when there are signs of involvement of lungs, liver, hematopoietic system, or sepsis, regardless of CNS involvement. These categories were proposed primarily for determination of treatment efficacy in clinical trials, but they also serve as a framework for appreciating the clinical presentations of neonatal HSV infection.

The onset of neonatal HSV infection is in the first month of life. In large series at least 9% of infants presented on the first day of life and 30% to 40% in the first week.8 From 35% to 45% of the infants are premature and 40% have a history of fetal scalp monitoring. As in group B streptococcal infection and listeria infection in the neonate, HSV infection can be thought of as having an early or a late onset. Skin, eye, and mouth and DISSEM forms of infection tend to occur in the first week of life with a mean presentation of day 5 to 6. Central nervous system infection tends to present in the second week of life, with a mean day of presentation of 9 to 12,8 although I have seen several cases as late as 6 weeks of age. Nevertheless, there is significant age overlap among these three groups of infected infants.

A number of studies continue to document that, while parents are quite good at bringing their infants to medical attention (mean time 1 to 2 days from onset of symptoms), there is often, unfortunately, a 6 to 7 day delay in initiation of antiviral therapy.8 During this delay, a significant portion (up to 70%) of infants progress from SEM infection to CNS or DISSEM disease, which are much more serious. It is hoped that with a better appreciation of risk factors and clinical presentation, this delay will be shortened and the outcome of these infants improved.

The most characteristic clinical sign are vesicles, which are thin walled, small (1 to 3mm), often grouped, and fluid filled. They begin as papules and progress over a few days to ulcers. Only one third to one half of neonates with HSV infection have vesicles on presentation. Vesicles are more common in the SEM form of illness. The lack of vesicles must not prevent consideration of the diagnosis of neonatal HSV infection, as up to one third of infected neonates never have a vesicle at any time during the illness.

Other signs of neonatal herpes are nonspecific and typical of sick neonates in general. They include signs of CNS dysfunction (seizures, lethargy, irritability, poor tone, poor feeding), sepsis (fever, hypotension, bleeding), conjunctivitis, or pneumonia. The distribution of clinical signs is to some extent dependent on the form of illness (Table 1, page 728). Thus, the presentation may be subtle, and in the absence of vesicles or keratoconjunctivitis, entirely nonspecific. Therefore, a high index of suspicion must be entertained for this uncommon illness that hides among cases of "aseptic meningitis" or "rule out sepsis" in the first month of life.

Table

TABLE 1Presenting Signs of Neonatal Herpes Simplex Virus Infection

TABLE 1

Presenting Signs of Neonatal Herpes Simplex Virus Infection

Certain clues may be helpful in early diagnosis of neonatal HSV infection. The baby's presenting part (as the head in a vertex presentation) is sometimes the earliest and most obviously infected, as evidenced by the presence of vesicles. Scalp monitoring sites are critical to examine, and a scalp site infection with a nondiagnostic Gram stain (no obvious bacteria in abundance) or one not responding to routine antibiotics is suggestive of HSV infection. Observing conjunctivitis and especially the classic keratoconjunctivitis is critical, as are mouth ulcers that are difficult to appreciate in the newborn, especially in the intubated infant. The neonate with undiagnosed lymphocytic meningitis, with or without seizures or computed tomographic scan abnormalities, should be evaluated and treated for HSV infection while other diagnostic studies are being conducted. Most of these newborns will have other etiologies established (enteroviral, listeria, candida), but waiting for the diagnosis of other causes of infection while HSV progresses is untenable. The newborn with a sepsis-like syndrome who is not responding to routine antibiotics, has negative cultures, often with signs of disseminated intravascular coagulation, thrombocytopenia and /or elevated liver function tests, should be treated empirically with acyclovir, pending evaluation for HSV. While every "rule out sepsis" evaluation need not trigger an HSV evaluation and acyclovir treatment, those newborns with atypical presentations or particular risk factors (a history of maternal or paternal HSV infection) should. Likewise, any newborn with unexplained acute hepatitis, occasionally mimicking metabolic liver disease, should be evaluated and treated for HSV infection, pending definitive diagnosis. The same is true for early onset and rapidly progressive culture-negative pneumonitis, usually beginning centrally and progressing to total lung involvement in 1 to 3 days, with onset in the first week of life. As with many other neonatal infections of maternal origin, if the infant is one of a multiple birth and one of the babies has neonatal HSV infection, all should be evaluated and treated. Table 2 (page 729) is a listing of particular signs, syndromes, or risk factors that should trigger evaluation and empiric therapy for HSV infection in the first month of life.

A number of factors are predictive of increased mortality in infants with neonatal HSV infection.8,9 Perhaps the most important, and potentially amenable to improvement by earlier recognition and therapy, is the stage of illness. Local SEM has no mortality, whereas mortality is high with CNS and DISSEM infection. Other poor prognostic signs include coma, disseminated intravascular coagulation, pneumonia, and prematurity. Morbidity is also increased in infants with CNS or DISSEM infection, those with seizures at the onset of therapy, and in SEMinfected infants with three or more cutaneous recurrences in the first 6 months of life.

Table

TABLE 2Particularly High Risk Settings for Neonatal Herpes Simplex Virus Infection

TABLE 2

Particularly High Risk Settings for Neonatal Herpes Simplex Virus Infection

DIFFERENTIAL DIAGNOSIS

The differential diagnosis of neonatal HSV infection is large and is dependent on the three main syndromes.10 Vesicular eruptions are often mistaken for pustules, thus infectious and noninfectious diseases that cause a neonatal vesicularpustular eruption are in the differential of HSV. These diseases are delineated in Tables 3 and 4. Regarding infections, the most common source of confusion clinically is Staphylococcus aureus pustular eruption, which can usually be easily excluded by a Gram stain of vesicle fluid. In HSV infection, polymorphonuclear leukocytes may be seen in the early vesicular fluid, which may mislead one to think of a bacterial infection, although bacteria are not seen. Varicella zoster virus eruption can be diagnosed by a careful history of chickenpox (not zoster) in the mother near delivery, or in a close contact, if the mother is V zoster seronegative.

Table

TABLE 3Infectious Etiologies of Vesicular-Pustular Eruptions in the Neonate10

TABLE 3

Infectious Etiologies of Vesicular-Pustular Eruptions in the Neonate10

The noninfectious etiologies of vesicular-pustular eruptions in the neonate include some common and usually easy to recognize entities like erythema toxicum, containing eosinophils on stain, or neonatal pustular melanosis. In addition, there is a long list of arcane diseases for which most pediatricians require an experienced pediatric dermatologist and careful history from the mother to make a diagnosis. I have seen several cases of incontinentia pigmenti and been impressed by the linear array of vesicles. In neonatal lupus, maternal history and blood tests have been diagnostic.

The differential diagnosis of CNS HSV infection includes many viral infections (especially enteroviral infection), the agents of the so-called TORCH syndrome (toxoplasmosis, syphilis, rubella, cytomegalovirus), and other less common causes of CSF lymphocytosis (fungi, especially candida, listeria, and, rarely, mycobacteria). The differential diagnosis of the DISSEM form includes neonatal sepsis. Finally, the differential diagnosis of HSV hepatitis includes the infectious agents of hepatitis (especially hepatitis A and B, cytomegalovirus, enterovirus, adenovirus) and metabolic illnesses causing hepatic failure.

Table

TABLE 4Noninfectious Etiologies of VesicularPustular Eruptions in the Neonate10

TABLE 4

Noninfectious Etiologies of VesicularPustular Eruptions in the Neonate10

DIAGNOSIS

Specific diagnosis of HSV infection can be made rapidly on vesicular scrapings by immunologic tests such as the enzyme-linked immunosorbant assay or fluorescent antibody assays for HSV antigen. Their specificity and sensitivity range from 75% to 85%. The old cytologic staining technique (Tzanck smear, which searches for multinucleated giant cells in the scrapings of the base of a vesicle) is neither sensitive nor specific and should not be relied on.

The gold standard for diagnosis remains viral culture. The average time for a positive culture is 2 to 3 days. If vesicles are present, culturing their fluid or swabbing their base provides the highest yield. Other culture sites include mouth (40% to 50% positive), eye (25%), and CSF (25% to 30%). Urine and stool may also be used. Laboratories can accelerate detection of HSV by combining culture and immunologic techniques (such as enzyme-linked immunosorbant assay or fluorescent antibodies), but these combination assays are not universally available. Indeed, with the reduction in the number of on-site viral laboratories because of recent changes in the medical environment, even many academic and large community medical centers must rely on outside, centralized virology laboratories, which may increase the time needed to establish a diagnosis. Furthermore, if transportation of specimens is not done carefully and expeditiously, the sensitivity of the assays can be compromised.

The most important recent advance in HSV diagnosis, especially of CNS infection, is the ability to detect minute quantities of HSV DNA rapidly by molecular amplification techniques such as PCR. A number of studies have demonstrated that in cases of CNS neonatal HSV, the PCR of CSF is positive approximately 75% of the time when performed by a qualified laboratory.11 In cases of DISSEM the CSF is almost always positive. Surprisingly, in cases of SEM, the CSF is PCR-positive about 25% of the time. This may explain some of the neurologic sequelae in babies with disease thought to be localized to SEM. In small studies, PCR of sera has been reported to be positive most of the time in CNS and DISSEM neonatal HSV as well.

Table

TABLE 5Treatment of Neonatal HSV Infection

TABLE 5

Treatment of Neonatal HSV Infection

Type-specific serologie tests able to distinguish the response to HSV-I and 2 are now commercially available (Focus technology. Cypress, CA, 1-800-445-0185). This testing has no role in the acute diagnosis of neonatal HSV infection since the titer in the neonate primarily reflects maternal antibody. Serial titers may be helpful in retrospect only. However, type-specific serology may be useful in determining if a genital HSV infection in the mother is a recurrent or primary /first episode; the latter category increases the risk to the newborn. For example, if a mother has HSV-2 isolated from a genital site, but is HSV-2 type-specific antibody seronegative, the likelihood is that the mother's genital HSV-2 infection is recent and the neonate is at a high risk for acquiring HSV-2 infection and for that infection to be severe.

The evaluation of any baby with neonatal HSV thus includes HSV cultures of any skin lesions, mouth, eye, urine, stool, and CSF. If lesions are present, enzyme-linked immunosorbant assay or fluorescent antibody rapid tests are useful, if available. Cerebrospinal fluid examination is important, with PCR analysis for HSV DNA.

Other testing includes a complete blood cell with platelet count, liver function tests, chest radiographs, and brain imaging studies with and without contrast, as clinically indicated, to determine the severity of the HSV infection.

TREATMENT

Neonatal HSV infection is a potentially lifethreatening illness. Cases may progress rapidly, and even with antiviral therapy, management of seizures, respiratory failure, shock, and disseminated intravascular coagulation is difficult. All cases, except mild forms of SEM disease, should be managed by an experienced team of intensivists, infectious disease experts, neurologists, and other specialists as needed in an intensive care setting. Table 5 (page 730) compiles the major studies and treatment outcomes beginning in the 1970s with the National Institute of Allergy and Infectious Diseases nationwide collaborative antiviral trials of therapy for babies with neonatal HSV infection.1214 Therapy first with vidarabine12 and then low dose (30 mg/kg/d)13 and more recently high dose (60 mg/kg/d) acyclovir14 have shown that once effective antiviral therapy is begun the illness rarely progresses from one stage to another (while 70% of SEM illness progresses if untreated). There has been marked improvement in the mortality associated with CNS and DISSEM disease, and morbidity from all forms of neonatal HSV infection in the past 3 decades has declined. Much of this decline is due to advances in the care of critically ill neonates as well as antiviral therapy. Even the most apparently trivial neonatal HSV infection should be treated with high dose, intravenous acyclovir, and therapy should be commenced upon serious consideration of the diagnosis not upon confirmation of the diagnosis. Acyclovir, 60 mg/kg/d intravenously administered for 14 days for SEM disease and 21 days for CNS or DISSEM neonatal HSV is the current standard.14 Ocular involvement warrants topical treatment with trifluridine, iododeoxyuridine, or vidarabine ophthalmic preparations, as well as intravenous acyclovir. The major toxicity of high dose acyclovir treatment is neutropenia (20%), requiring frequent monitoring of white blood cell counts.14 Good hydration will avoid nephrotoxicity. Several case reports have documented infection with acyclovir resistant HSV in the newborn. Thus, in neonates who fail to respond to acyclovir therapy, or who respond initially and then deteriorate, foscarnet sodium should be added. This is a moderately toxic drug (nephrotoxicity in particular) and should be used only in consultation with an experienced infectious disease expert.

It has become clear that frequent cutaneous (and perhaps silent CNS) recurrence is a poor prognostic factor for SEM disease. The use of suppressive oral acyclovir (300 mg/m2) three times per day for the first 6 months of life can decrease the risk of recurrences, but there are no data yet to show that outcome is improved.15 This strategy is not without risks, as nearly half of the infants treated experienced neutropenia and, in at least one case, an acyclovir-resistant virus was recovered. Until further data are available, chronic suppressive therapy is an investigational practice and should only be used after full explanation to parents of the possible risks and benefits.

There often is tremendous guilt, and then anger, among parents of children with neonatal HSV infection. Anticipation of these feelings and sensitive and caring support for the parents is an important part of care and may lessen the all-toocommon complication of medical-legal ramifications. Sensitive explanations of the pathophysiology, appropriate for the parents' level of education, are crucial.

PREVENTION

The ability to prevent neonatal HSV infection in women with lesions at delivery by cesarean section has not improved in many decades. With better risk-analysis data regarding the rate of transmission from women with recurrent HSV disease, the use of cesarean section in that setting has become more controversial. With the widespread availability of reliable HSV-2 type-specific serology, it is now easier to determine the category of infection in symptomatic women near delivery, to define a couple's HSV status and detect discordance so as to counsel them regarding risk of lateterm HSV infection. Recent studies showed that late-term use of oral acyclovir in pregnant women with a history of frequent genital HSV recurrences is effective in preventing HSV recurrence and thus the need for subsequent cesarean section.16 An effective HSV vaccine to prevent genital infection is years away, but mere continues to be active research in this area. It is hoped that further understanding of the immune system of neonates will allow for effective cytokine-mediated reconstitution of their ability to combat HSV infection. For now, the pediatrician's role in prevention of death and disability due to neonatal HSV infection must be to appreciate the epidemiologic risks, subtle clinical presentations, and use of effective early empiric antiviral therapy in suspect newborns.

REFERENCES

1. Fleming DT, McQuillan GM, Johnson RE, et al. Herpes simplex virus type 2 in the United States, 1976 to 1994. N Engl J Med. 1997;337:1105-1111.

2. Sullivan-Bolyai JZ, Hull HF, Wilson C, et al. Neonatal herpes simplex virus infection in King county, Washington: increasing incidence and epidemiologic correlates. JAMA. 1983;250:3059-3062.

3. Guitierrez KM, Falkovitz Halpern MS, Maldonado Y, Arvin AM. The epidemiology of neonatal herpes simplex virus infections in California from 1985 to 1995. / Inject Dis. 1999;180:199-202.

4. Brown Z, Selke S, Zeh J, et al. Transmission Rates of HSV from Mother to Infant. Paper presented at the Abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, Sept. 1999; 423.

5. Kulhanjian JA, Soroush V, Au DS, et al. Identification of women at unsuspected risk of primary infection with herpes simplex virus type 2 during pregnancy. N Engl ] Med. 1992;326:916-920.

6. Brown ZA, Selke S, Zeh J, et al. The acquisition of herpes simplex virus during pregnancy. N Engl ] Med. 1997;337:509-515.

7. Malm G, Berg U, Forsgren M. Neonatal herpes simplex: clinical findings and outcome in relation to type of maternal infection. Acta Paediatr. 1995; 84:256-260.

8. Kimberlin DW, Lin CY, Jacobs FJ, et al. Natural history of neonatal herpes simplex virus infections in the acyclovir era. Pediatrics. 2001;108:223-229.

9. Whitley RJ, Arvin A, Prober C, et al. Predictors of morbidity and mortality in neonates with herpes simplex virus infections. N Engl J Med. 1991;324:450-454.

10. Kohl S. Neonatal herpes simplex virus infection. Infections in Perinatology. 1997;24:129-150.

11. Kimberlin DW, Lakeman FD, Arvin AM, et al. Application of the polymerase chain reaction (PCR) to the diagnosis and management of herpes simplex virus disease. / Infect Dis. 1996;174:1162-1167.

12. Whitley RJ, Nahmias AJ, Soong SJ, et al. Vidarabine therapy of neonatal herpes simplex virus infection. Pediatrics. 1980;66:495-501.

13. Whitley RJ, Arvin AM, Prober C, et al. A controlled trial comparing vidarabine with acydovir in neonatal herpes simplex virus infection. N Engl J Med. 1991;324:444-449.

14. Kimberlin DW, Lin CY, Jacobs RF, et al. Safety and efficacy of high-dose intravenous acyclovir in the management of neonatal herpes simplex virus infections. Pediatrics. 2001;108:230-238.

15. Kimberlin D, Powell D, Bruger W, et al. Administration of oral acyclovir suppressive therapy after neonatal herpes simplex virus disease limited to the skin, eyes and mouth: results of a phase I/ ? trial. Pediatr Infect Dis }. 1996;15:247254.

16. Scott LL, Sanchez PJ, Jackson GL, et al. Acyclovir suppression to prevent Caesarian section after first episode genital herpes. Obstetr Gynecol. 1996;87:69-73.

TABLE 1

Presenting Signs of Neonatal Herpes Simplex Virus Infection

TABLE 2

Particularly High Risk Settings for Neonatal Herpes Simplex Virus Infection

TABLE 3

Infectious Etiologies of Vesicular-Pustular Eruptions in the Neonate10

TABLE 4

Noninfectious Etiologies of VesicularPustular Eruptions in the Neonate10

TABLE 5

Treatment of Neonatal HSV Infection

10.3928/0090-4481-20021101-09

Sign up to receive

Journal E-contents