NATURAL VARICELLA-ZOSTER VIRUS REACTIVATION SHORTLY AFTER VARICELLA IMMUNIZATION IN A CHILD

Abstract

Zoster in childhood is an unusual occurrence and probably even less common after varicella-zoster virus (VZV) immunization.1-5 We report a normal child with no history of varicella who developed uncomplicated zoster caused by natural VZV 12 days after VZV immunization. The rates of zoster in children and the possible role of varicella immunization in the stimulation of natural VZV occurrence are reviewed. Case report. HV is a 6-year-old boy with no personal or family history of varicella. It is not known whether his mother had varicella during her pregnancy with him. The child had had no serious prior infections, HIV risk factors or other evidence of immunodeficiency. Because of the negative chickenpox history, he received one dose of the varicella vaccine (Varivax; Merck and Co., West Point, PA) in the right arm. He experienced intermittent fevers for 8 days, as high as 40.0°C. with no other physical findings. No other laboratory evaluations were done during this time. Twelve days after the immunization he had onset of a pruritic rash beginning on his back and extending to the posterior portion of his left arm. When seen 3 days later, 15 days after the immunization, he was afebrile, complained of nausea and vomiting for 3 days and had a vesicular rash in a left second thoracic dermatome pattern on his back and extending to the posterior aspect of his left arm. Viral cultures were obtained as were specimens for VZV PCR; prochlorperazine was prescribed for nausea and ibuprofen for mild pain. Nine days later there was complete crusting of all lesions, resolution of pain and no further complaints. The viral cultures were negative for herpes simplex virus and VZV. VZV sample preparation and PCR amplification were performed as previously described.6, 7 Briefly 2 sets of primer pairs were used to amplify a 222-base pair (bp) fragment in gene 54 and a 350-bp fragment in gene 38 of the VZV genome. The 222-bp product contains an asymmetrically placed BglI restriction site present in the Oka vaccine strain and in ∼15 to 20% of American wild-type strains. The 350-bp product contains an asymmetrically placed PstI restriction site present in all known wild-type strains but is absent in the vaccine strain. Ethidium bromide-stained amplified product (8 μl) was then visualized under ultraviolet light after electrophoresis in a 1.8% agarose gel. If the expected bands were seen, then 60 μl of remaining amplified product were divided into two aliquots which were digested with the restriction endonucleases, BglI and PstI. The products of digestion were then separated by electrophoresis in a 4% agarose gel. The pattern of the bands produced by the two digestions were compared with those of wild-type and vaccine strains. VZV-negative controls are included in every assay. If no VZV DNA was amplified, the presence of amplifiable DNA was assessed by subjecting these specimens to a second amplification using a primer pair that amplifies a 268-bp fragment of the human beta-globin gene. Analysis of the patient's sample demonstrated a pattern consistent with wild-type varicella-zoster virus. Discussion. Zoster in a normal child is an unusual occurrence. In several population-based studies the rate of zoster in children ages 0 to 19 years was from 0.63 to 1.6/1000 person years.1-3 The rate in children age 6 years ranged from 0.3 to 1.5/1000 person years.1-3 It is generally agreed that varicella occurring in the first year of life is associated with a higher risk of zoster,1, 4 although this finding is not uniform.3 It is not clear when the child in the present case was first infected with VZV because the disease was either asymptomatic or so mild as to be inapparent to the child and his family. There are few data regarding the risk of zoster after VZV immunization. The risk of zoster in 9454 healthy children after immunization was 0.18/1000 person year (8 cases in 44 994 person years followed). The one positive culture recovered was of natural VZV. The time from immunization ranged from 1.2 to 5.8 years.5, 8 Review of the world's literature on PCR or culture-confirmed zoster in VZV-immunized healthy9-11 and leukemic12 children and the 1 adult case13 revealed 12 published cases. Six were wild type. None of the 12 cases occurred before 32 days postimmunization. In immunocompromised children, the risk of zoster after immunization is considerably lower than that after natural varicella infection.14-17 Indeed receipt of a booster dose of vaccine or a household exposure to varicella further decreased the risk of zoster in immunized leukemic children.12 A population-based study has provided evidence that a high intensity of varicella transmission suppresses viral reactivation.18 The risk of zoster has been associated with lower cell-mediated immunity to VZV in infants and in children with leukemia after immunization.16, 19 VZV immunization has been shown to increase cell-mediated immunity in older individuals.20 A large long term placebo-controlled VZV immunization trial is ongoing to determine whether VZV immunization can decrease the rate of zoster in the elderly, a group with particularly high rates of zoster (M Oxman, personal communication, 1998). It is against this background that we report a child with the onset of zoster caused by natural VZV shortly after receiving VZV immunization. The probability of a normal 6-year-old having zoster in a given 2-week period are 1 to 5/100 000 person years.1-3 It would be even lower for a child with a negative history for chickenpox, as in the present case. While this case may be an extraordinary coincidence, it mandates a careful review of all cases of zoster occurring after VZV immunization and their time from immunization that have been reported in postmarketing surveys. A statistical analysis by investigators at Merck using the adverse vaccine reports database is ongoing to determine whether cases of zoster caused by natural VZV are closely related to VZV immunization. Induction of natural zoster by immunization could result from cytokine stimulation of viral emergence from latency or from transient viral-induced immunosuppression. Zoster induced by VZV immunization could have special implications for the use of immunization to prevent zoster in the elderly, a population with almost uniform VZV latent virus and higher risks for zoster. Steve Kohl, M.D. Jonathan Rapp, M.D. Phillip La Russa, M.D. Anne A. Gershon, M.D. Sharon P. Steinberg, B.S. Departments of Pediatrics and Family and Community Medicine; University of California, San Francisco; San Francisco General Hospital; San Francisco, CA (SK, JR) Department of Pediatrics; Columbia University College of Physicians and Surgeons; New York, NY (PLR, AAG, SPS)