Herpes Zoster in the First Year of Life Following Postnatal Exposure to Varicella-zoster Virus: Four Case Reports and a Review of Infantile Herpes Zoster

Background  Herpes zoster, a painful vesicular dermatomal eruption, is the result of reactivation of the varicella-zoster virus (VZV) from infected sensory ganglia. Traditionally, it is considered to be a disease of adults, in contrast to primary infection with VZV, which tends to occur mainly in children.

Observations  We report 4 cases of infantile herpes zoster in healthy immunocompetent children, all of whom were exposed to primary varicella infection within the first few months of life. A review of 62 cases from the literature reveals that postnatally acquired herpes zoster is less common than intrauterine infection (31% [n = 19] vs 69% [n = 43]) and that there is a 1.5:1 male predominance. All dermatomes are equally affected.

Conclusions  Although uncommon, herpes zoster can develop in immunocompetent children as young as a few weeks of age and should be considered in the differential diagnosis of vesicular eruptions in infants. Most frequently, it is the result of intrauterine VZV infection, but it can be secondary to postnatal exposure to VZV at an early age.

Herpes zoster may be seen in immunocompetent children and is frequent in children with acquired cellular immune deficiency from chemotherapy or human immunodeficiency virus. Herpes zoster in infancy is rare but well described following intrauterine exposure to varicella-zoster virus (VZV).^1,2 We describe 4 infants presenting with classic herpes zoster between 4 and 11 months of age following inapparent or minimally symptomatic varicella as a consequence of postnatal VZV exposure. We also review the literature on herpes zoster in infants younger than 1 year.

A 7-month-old male infant presented with a 1½-week history of a crop of vesicular lesions on an erythematous base on the right side of the back of his neck. This infant had no prior history of varicella, but at 5 months of age had household exposure to varicella in a sibling. His mother had a history of varicella many years before pregnancy. Initial physical examination of the infant showed grouped vesicles on an erythematous base scattered along the jawline and lower cheek, with a sharp demarcation at the midline on the front and back of the neck, corresponding to the right C3 dermatome (Figure 1). Some vesicles were slightly outside the primary dermatome. Findings from a Tzanck smear revealed a multinucleated giant cell. Serologic results obtained at initial presentation demonstrated a positive VZV titer. A viral culture did not grow anything. The patient was treated with 200 mg of acyclovir orally 4 times a day for 7 days, with complete resolution of zoster without sequelae.

A 7-month-old male infant presented with a 3-day history of grouped vesicles on an erythematous base over the sacrum, without other symptoms. The patient had a history of varicella at 3 weeks of age following household exposure at 1 week of age to an older sibling with varicella. At the time of the primary infection, the infant was treated with oral acyclovir. The infant's mother had a history of varicella at age 8 years. On physical examination, the patient had grouped vesicles on an erythematous base on the sacrum, left buttock, midportion of the left posterior thigh, and lateral aspect of the left foot, corresponding to the left S1 dermatome (Figure 2). No other abnormalities were noted. Findings from a Tzanck smear were positive for multinucleated giant cells. The patient was treated with 200 mg of acyclovir orally 4 times a day for 7 days. Complete resolution occurred without sequelae.

A 4-month-old infant presented to the emergency department with a 2-day history of fussiness, rhinorrhea, fever, and erythematous vesicles covering the right side of his face. The rash began with a vesicle on the right cheek and progressed to involve the eyebrow, forehead, temporal area, and right upper lip. The patient had a history of exposure to varicella during the neonatal period but never developed signs or symptoms of varicella. The infant's mother had a history of varicella many years before her pregnancy. On physical examination, the patient had a temperature of 37.8°C and an erythematous vesicular eruption with a crusty exudate in the right upper quadrant of the face over the distribution of the V1 and V2 branches of the trigeminal nerve. His conjunctivae were inflamed bilaterally, and there was a yellowish discharge from the right eye. The patient's nares and upper lip were edematous. The patient was admitted for treatment of possible bacterial superinfection complicating herpes zoster. The patient began intravenous therapy, including acyclovir and nafcillin sodium. Ophthalmologic examination showed no corneal involvement. The patient showed notable improvement in the eruption and facial edema and was discharged to home after 5 days. Viral culture of the vesicle fluid grew VZV. Serologic results obtained on admission were positive for VZV. By report of his pediatrician, the infant recovered completely without sequelae.

An 11-month-old male infant presented to the dermatologist's office with a chief complaint of a facial rash. Four days before the visit, he developed a single papule on his left cheek, which did not respond to over-the-counter 1% hydrocortisone cream. Multiple new vesicles appeared. The child had not been eating or sleeping well, but remained afebrile. History revealed that the patient's older brother had visited him in the hospital on the day he was born. The following day, the brother developed an eruption consistent with varicella infection, and he was removed from the home to live with his grandparents until his varicella resolved. Subsequently, although the patient had some febrile illness, he never had anything that appeared to be primary varicella. On physical examination, the patient had multiple grouped vesicles on erythematous plaques involving the left side of the chin, with sharp demarcation at the midline. Vesicles extended to the left side of the mouth, the lateral aspect of the left cheek in front of the left ear, and onto the temple and about 4 cm into the frontal area of the scalp. This corresponded to the left V3 dermatome (Figure 3). Findings from a Tzanck smear were negative, but direct fluorescent antibody test results for VZV and a viral culture were positive. An acute-phase VZV titer was negative. No follow-up titer was obtained. The child was treated with 200 mg of acyclovir orally 4 times a day for 1 week. His vesicles rapidly crusted over, and the erythematous plaques resolved. Two months later, he was noted to have a 4 × 5-mm hypertrophic scar on his chin. When seen 2 years later, there was a 5 × 8-mm flat white scar remaining on his chin.

Generally, primary varicella tends to occur in childhood, whereas herpes zoster is a disease of adults, with most patients being older than 45 years.^3,4 The age-adjusted incidence rates of herpes zoster are the lowest (0.45 per 1000 person-years) in the group 0 to 14 years of age and highest (4.2-4.5 per 1000 person-years) among people 75 years and older.⁵ In the pediatric population, the incidence is the lowest in the group 0 to 5 years of age (20 per 100 000 person-years) compared with adolescents (63 per 100 000 person-years).⁶ There was a male predominance (1.5:1) observed in the literature, and all 4 of our cases were boys.

Antigen-specific T cells are believed to be the principal gatekeepers of latent VZV. Conditions in which cellular responses were lost or diminished by immunosuppression pose a risk for reactivation of VZV and recurrent disease manifestation as herpes zoster.^7-9 Herpes zoster in older individuals is associated with loss of VZV-specific cellular immunity.¹⁰ Herpes zoster in individuals undergoing chemotherapy is due to suppression of cellular immunity, whereas herpes zoster in human immunodeficiency virus–infected individuals is due to viral destruction of T cells.

Acquisition of herpes zoster in healthy immunocompetent children in early childhood or during intrauterine exposure has been attributed to the immaturity of the immune system.^8,11 Terada et al¹¹ conclude that immunological status before primary infection with VZV is important and affects reactivation of VZV. They observed that, 6 to 7 weeks after primary varicella, infants had a lower response of VZV-specific cellular and humoral immunity compared with children who had infection at older ages (>1 year). In another study,¹² the peak levels of IgG antibodies after primary varicella were lower in infants compared with older children. Low response in specific VZV immunity is a valid reason to consider varicella in the first year of life as a risk factor for development of herpes zoster in otherwise healthy children.^7,8,11,13,14 Terada et al¹⁵ showed that healthy immunocompetent children who had primary VZV before 1 year of age remained positive for VZV (as determined by polymerase chain reaction) for the longest period. From these data, Terada et al¹⁵ hypothesized that a "subclinical reactivation" puts infants with a history of primary varicella at risk for herpes zoster. In 69% of infantile herpes zoster cases (ie, <12 months of age) reported in the literature, the initial event could be traced to maternal varicella during pregnancy. Dobrev¹⁶ observed that maternal varicella during the first trimester is likely to produce congenital varicella syndrome; when women have the disease later in pregnancy, the fetus can develop asymptomatic congenital infection and subsequently present clinically with herpes zoster within the first year of life. Newborns of VZV-immune mothers can also develop subclinical varicella within the first 6 months of life. In these cases, maternal VZV antibodies passively transferred to the infant may modify the disease into a subclinical form. In general, infants with primary varicella infection are at high risk for herpes zoster within the first year of life.¹⁶

Acute varicella has been acquired postnatally less frequently (31% [19/62] of the reported cases) than as an intrauterine infection. There are some instances, however, in which the initial episode of varicella has not been documented. Although there are 43 documented cases of intrauterine exposure to varicella with subsequent herpes zoster in the first year of life, there are only 9 cases with known postnatal exposure and 10 in which the time of exposure was not well documented. Baba et al^8,17 suggest that in infancy the presence of maternal antibody modifies primary infection and that subclinical primary infection may predispose to herpes zoster. Although the mothers of the 4 infants in our series all had varicella in childhood, we hypothesize that their antibodies were not protective enough to have prevented primary infection with VZV in their infants.

The diagnosis of herpes zoster can be made by a Tzanck smear of scrapings from the floor of the vesicles,¹⁸ direct fluorescent antibody tests on similar smears,¹⁹ presence of high or rising titers exposed to VZV,¹⁸ and, definitively, by culture findings of the VZV virus.¹⁸ A high index of suspicion should be aroused when vesicular lesions are noted to be in a dermatomal distribution. The most common differential diagnosis is impetigo, and bacterial culture can usually distinguish the 2 conditions, unless there is concurrent bacterial infection of the herpetic lesions. Viral and bacterial cultures are often both performed.

In general, children tolerate herpes zoster much better than adults, with the disease usually being mild and lasting 1 to 3 weeks.^7,9 Clinical features include pruritus and pain, with rare postherpetic neuralgia.^4,7,20 Systemic reactions include fever, headache, and lymphadenopathy. Secondary bacterial infections and ophthalmic herpes zoster have been reported.⁴ The most frequently involved dermatomes are cranial, cervical, and thoracic dermatomes.^7,9,21 Among the cases reviewed in this article, the thoracic dermatome was affected in most patients. Involvement of more than 1 dermatome can occur.

Summaries of the reported cases of herpes zoster in the first year of life are presented in Table 1 and Table 2. There were 43 cases (69%) of prenatal acquisition (Table 1) and 19 cases (31%) of postnatal acquisition (Table 2). Of the cases in which patient sex was reported, 29 (60%) occurred in boys and 19 (40%) in girls. In 14 cases (23%), the reports did not specify sex. Of the 51 cases in which location was specified, 14 (27%) were in cranial nerve, 10 (20%) in cervical, 17 (33%) in thoracic, and 10 (20%) in lumbosacral dermatomes. In many cases, more than 1 dermatome was affected.

Oral acyclovir is recommended by the manufacturer to be given at an oral dosage of 20 mg/kg of body weight per dose 4 times a day to children older than 2 years, and it remains the first-line therapy for VZV in children. Oral (40-60 mg/kg per day) or intravenous (30 mg/kg per day) treatment has been used for 5 to 8 days or for 2 days after new lesions stop developing.³⁸ Other authors suggest 100 mg/kg per day for 7 days⁴⁷ or 50 mg/kg per day for 5 to 7 days.^27,31 In acute herpes zoster, it decreases the time of new vesicle formation and the number of days to crusting. Analgesics and appropriate skin care provide relief and minimize the risk of secondary infection.⁹

Although herpes zoster is not common in children younger than 1 year, a review of the literature reveals that it occurs. The most frequent cause in immunocompetent patients is intrauterine exposure to VZV. We have described 4 cases of herpes zoster following postnatally acquired primary varicella. Siblings infected with varicella are the most usual contact source. We hypothesize that this condition is rarely recognized because of the mild clinical manifestations in this age group and the expectation that maternal antibodies will be protective.¹⁶ It is likely that the vesicular lesions of herpes zoster in this age group are misdiagnosed as impetigo or other cutaneous disorders. With a high degree of suspicion, the dermatomal distribution of a vesicular eruption in infancy should point the clinician toward a correct diagnosis of herpes zoster. Why some children are not protected from infantile VZV infection by maternal antibodies remains a question for further investigation.

Correspondence: Anne W. Lucky, MD, Division of Pediatric Dermatology, Cincinnati Children's Hospital Medical Center, 3333 Burnet Ave, Cincinnati, OH 45229.

Accepted for publication April 12, 2004.

We thank Miriam Hakim, MD, for assistance in the preparation of the manuscript.