Chickenpox represents primary infection with varicella-zoster virus (VZV), whereas herpes zoster occurs as a result of reactivation of latent VZV in the dorsal root ganglia. It has been assumed that one attack of chickenpox confers lifelong protective immunity against subsequent exogenous attacks of VZV, but not against reactivation of endogenous latent VZV. However, we have sometimes experienced clinical reinfection in immunocompetent children who had seroconverted after a live attenuated varicella vaccination, and on rare occasions, we have seen reinfection of chickenpox in immunocompetent children after natural infection. On the basis of their experience with 14 patients with recurrent chickenpox, Junker et al.1 indicated that recurrence of chickenpox may be more common than is generally accepted. There is little information concerning the relationship between recurrent chickenpox after natural infection and specific immunity. We studied VZV-specific immunity in patients with recurrent chickenpox who were seen in our hospital during a period of 4 years.
The subjects were four children younger than 15 years of age who were diagnosed as having recurrent chickenpox in the hospital of Kawasaki Medical School between June, 1991, and June, 1995. Informed consent was obtained from the parents or guardians of the children according to guidelines of the United States Department of Health and Human Services and those of Kawasaki Medical School. The primary chickenpox of the subjects had been diagnosed by different pediatricians, and the findings were documented on their medical records. We omitted subjects whose medical records of chickenpox were not confirmed. We confirmed whether the crops had become crusts in 3 to 4 days and the presence of the scars of primary chickenpox. We determined VZV-specific cellular immunity and humoral immunity using a lymphoproliferative assay and an enzyme-linked immunosorbent assay,2 respectively, in the acute phase (within 3 days after the onset) and the convalescent phase (4 weeks after the onset) of chickenpox.
Patient 1. This patient was a 14-year-old boy who had been treated for acute lymphocytic leukemia since March, 1991, with a past history of chickenpox at 3 years of age. The course of the primary chickenpox was normal; i.e. vesicles developed crusts in a few days. He was admitted to our hospital for maintenance therapy on September 30, 1991, and was isolated in a clean room until October 21, 1991, for granulocytopenia resulting from chemotherapy for acute lymphocytic leukemia. Two vesicles, each of which was 1 cm in diameter, were formed on the right chest and left thigh on October 30, 1991. We administered acyclovir intravenously. Thereafter these vesicles became crusts in 3 or 4 days. As a result, there was no herpes zoster but seven scattered vesicles were found in a few days. With treatment the chickenpox was mild. There were no patients or hospital employees with chickenpox in this ward before and after this episode.
Patient 2. This patient was a 2-year 4-month-old boy with a past history of chickenpox at 1 year 4 months of age. His first chickenpox, which was contracted from a playmate at a day-care center, followed the usual course. He had received the measles-mumps-rubella vaccine on June 22, 1992. The next day a scattered rash including more than 50 scattered red papules and vesicles was found by his mother, leading her to bring the child to our hospital. However, many of the red papules did not develop into vesicles but subsided, and most of the vesicles developed crusts in 3 days before his next visit. The morbidity of the chickenpox was mild and did not require treatment.
Patient 3. This patient was a 6-year 9-month-old boy with a past history of chickenpox at 4 years and 6 months of age. His first chickenpox, which was contracted from a playmate with chickenpox, followed the usual course. His grandfather and father are physicians. His younger sibling contracted chickenpox 2 weeks earlier. A small scattered rash including approximately 100 red papules and vesicles was found on his visit to our hospital on May 25, 1993. Many of the red papules did not develop into vesicles, and the vesicles were smaller than the usual ones of chickenpox. The morbidity of the chickenpox was mild and did not require treatment with acyclovir.
Patient 4. This patient was a 5-year 5-month-old boy with a past history of chickenpox at 2 years and 6 months of age. His mother had taken him to the home of a playmate with chickenpox to expose the child to this infection. Two weeks later approximately 10 vesicles were found and a diagnosis of chickenpox was made by a pediatrician. Thereafter he did not contract chickenpox even when he had played with playmates with chickenpox. His mother noticed that he had a rash on May 26, 1993, and we saw him the next day. We observed approximately 50 vesicles and red papules, but some of the red papules did not develop into vesicles and subsided. The morbidity of the chickenpox was mild and did not require treatment.
The VZV-specific IgG and IgM antibody and VZV-specific cellular immunity values are shown in Table 1.
Discussion. There have been a few reports regarding chickenpox reinfection after natural infection. Junker et al.1 described 14 generally healthy children who developed 2 to 5 episodes of recurrent chickenpox. Gershon et al.3 reported 8 individuals who developed clinical chickenpox after previously being determined to be VZV-seropositive. Gurevich et al.4 described 3 VZV-seropositive hospital employees who developed chickenpox. Takayama et al.5 reported 2 cases of elderly patients (83 and 90 years of age, respectively) who had recurrent mild chickenpox after exposure to grandchildren with chickenpox.
The diagnosis of chickenpox is based on the characteristic rash and the course of the disorder. Pediatricians usually do not examine the specific antibody except for suspicious cases with an abnormal course. Our cases were not determined to be VZV-seropositive after primary chickenpox. To determine whether the diagnosis of primary chickenpox was correct, their past history of chickenpox was confirmed by examination of their medical records with their guardians' cooperation. All of the first incidences of chickenpox in these patients followed the usual course of chickenpox with the exception of Patient 4, in whom it was mild. We also confirmed the presence of chickenpox scars in three of the four patients. In addition Patients 2, 3 and 4 apparently had been exposed to VZV from playmates with chickenpox before the primary chickenpox. The second form of chickenpox was confirmed by alteration of VZV-specific IgG and/or IgM antibodies.
Patient 1 is thought to have been reactivation of latent endogenous VZV because the patient had been isolated in a clean room until 9 days before the onset. The specific IgG antibody was detected in the acute phase of Patient 1's illness. It is considered that the specific IgG antibody cannot protect against reactivation of VZV, because specific cellular immunity is low. Specific IgG antibody is detected in most patients with herpes zoster.6 The specific cellular immunity of Patient 1 was higher than that of the other patients, but the stimulation index (SI) was low in comparison with those of VZV-seropositive children.2 We have sometimes experienced dissemination of vesicles like those of chickenpox after herpes zoster in immunocompromised patients. Although this patient did not have typical zoster clinically, we believe the findings were consistent with atypical herpes zoster.
Patients 2, 3 and 4, on the other hand, are immunocompetent children, who are thought to have been exposed to exogenous VZV. These patients did not have specific IgG antibody at the onset of recurrent chickenpox. The specific IgM antibody was detected in Patients 2 and 3. The presence of VZV-specific IgM antibody does not necessarily indicate primary infection. Cradock-Watson et al.7 and we,8 respectively, reported that 78 and 50% of patients with herpes zoster had the VZV-specific IgM antibody. If we define positive as SI ≥ 2.0 in specific cellular immunity, only Patient 4 was positive at the onset. Even 4 weeks after the onset, the SI of these cases was 2.0 to 3.2. The specific cellular and humoral immunity responses of these cases were lower than those of immunocompetent children after primary chickenpox.2 Junker et al.1 reported that sequential follow-up of their eight patients revealed one who became seronegative and two who lost cellular immunity and that some of them developed more than three episodes of recurrent chickenpox. We infer that these immunocompetent cases may lose specific immunity with waning of the disease, because the specific immunity responses were low after recurrent chickenpox in spite of the booster effect after reexposure.
Acknowledgments. This study was supported in part by Grant 6-503 from Kawasaki Medical School and Grant 076-709-21 from the Ministry of Education, Science, Sports and Culture in Japan.
Kihei Terada, M.D.; Shoji Kawano, M.D.; Yoshina Shimada, M.D.; Yasuhiro Yagi, M.D.; Naoki Kataoka, M.D.
Department of Pediatrics
Kawasaki Medical School
Kurashiki City, Japan
1. Junker AK, Angus E, Thomas EE. Recurrent varicella-zoster virus
infections in apparently immunocompetent children. Pediatr Infect Dis J 1991;10:569-75.
2. Terada K, Kawano S, Yoshihiro K, Morita T. Varicella-zoster virus
(VZV) reactivation is related to the low response of VZV-specific immunity after chickenpox
in infancy. J Infect Dis 1994;169:650-2.
3. Gershon AA, Steinberg SP, Gelb L, the National Institute of Allergy and Infectious Diseases Collaborative Varicella Vaccine Study Group. Clinical reinfection
with varicella-zoster virus
. J Infect Dis 1984;149:137-42.
4. Gurevich I, Jensen L, Kalter R, Cunha BA. Chickenpox
in apparently “immune” hospital workers. Infect Control 1990;11:510-11.
5. Takayama N, Takayama M, Negishi M. Clinical varicella-zoster virus reinfection
observed in two advanced-age persons. Kansenshogaku Zasshi 1992;66:1373-7.
6. Arvin AM, Pollard RB, Rasmussen LE, Merigan TC. Cellular and humoral immunity in the pathogenesis of recurrent herpes viral infections in patients with lymphoma. J Clin Invest 1980;65:869-78.
7. Cradock-Watson JE, Ridehalgh MKS. Specific immunoglobulin responses after varicella and herpes zoster. J Hyg [Camb] 1979;82:319-36.
8. Terada K, Kawano S, Yoshihiro K, Morita T. Natural killer cell activity in herpes zoster in children without underlying disease. Scand J Infect Dis 1993;25:521-4.