OriginAl StudieS

HerpesZosterinaPartiallyVaccinatedPediatricPopulation in Central Israel

Michal Stein, MD,* Rinat Cohen, MD, Michal Bromberg, MD, Diana Tasher, MD,* Tamar Shohat, MD, and Eli Somekh, MD*
Although there is abundant information regarding HZ in adults, including its incidence, clinical presentation and risk factors, data regarding HZ in children are limited, especially with regard to the ambulatory cases. Some of the data about the epidemiologic and clinical characterization of pediatric HZ are derived from studies done before introduction of varicella vaccine, and its incidence was found to be 42/100,000160/100,000 person-years.68 Contracting varicella in the first year of life was found, in early studies, to be a significant risk factor for HZ, and the relative risk (RR) was estimated to be between 2.8 and 20.9.6,8 However, these findings have not been reproduced. Concerns were raised that routine immunization against varicella may increase the incidence of HZ in the general population due to decreased circulation of the wild-type virus, which leads to decline in varicella-specific T cell immunity.10 Analyses performed after the introduction of varicella vaccine have shown conflicting results. Although data from the Centers for Disease Control and Prevention and from Canada suggest that vaccine-induced decline in varicella has not led to increased rate of HZ in the general population,11,12 another study demonstrated increasing rates of zoster accompanying decreasing varicella rates in Massachusetts.13 However, data regarding childhood zoster in the era of varicella vaccination are limited.1416 The objectives of this population-based study were to estimate the current rate of pediatric HZ, to describe the clinical and epidemiologic characteristics of the ambulatory cases and to identify risk factors for its appearance. The study was performed during an era of partial vaccination with VZV vaccine in Israel by using the computerized data of Maccabi Health Services in a county in Central Israel. The ability to define the pediatric population base, to determine the exact rate of varicella vaccination and the age distribution of primary varicella infection in this population prompted this study.

Background: This study was performed during an era of partial vaccination with varicella vaccine in Israel to characterize ambulatory pediatric herpes zoster (HZ) cases in a population with partial varicella vaccination coverage. Methods: Data were collected from computerized databases of a population of 114,000 children. Records of children aged 018 years, diagnosed with HZ during 2006 to 2008 were reviewed by pediatric infectious diseases experts. Telephone interviews were done with a sample of the parents to get further clinical details. Results: Of 692 medical records reviewed, 450 cases were approved for analysis, and 77 interviews were conducted. Incidence of HZ was 130 of 100,000 person life-years. Peak incidence was detected in children aged 911 years (222/100,000 person life-years). Pain and fever accompanied 52% and 13% of episodes, respectively. Higher risk for HZ was found in children who had varicella during their first year of life (relative risk and 95% confidence interval: 13.5[9.618.8]; P < 0.001), and in children who had varicella during the second year of life (relative risk = 2 [1.52.6]; P < 0.001). Vaccination was found to be protective against HZ (relative risk = 0.42 [0.330.55]; P < 0.001). Conclusions: The epidemiology of HZ seems to be changing in a population with partial varicella vaccination rate. Our results may suggest that children who contracted chicken pox in their first year of life may benefit from varicella vaccination. Key Words: children, herpes zoster, vaccination, varicella (Pediatr Infect Dis J 2012;31: 906909)

aricella zoster virus (VZV) causes 2 clinically distinct forms of disease: varicella and herpes zoster. Primary VZV infection results in the typical diffuse vesicular rash of varicella, followed by latent infection within the sensory dorsal root ganglia. Reactivation of the virus later in life, leads to the typical herpes zoster (HZ) unilateral vesicular eruption in a restricted dermatomal distribution.1 Incidence rates of HZ progressively increase with age, presumably due to the decline in VZV-specific cell-mediated immunity,2 and the cumulative lifetime incidence among the whole population is approximately 1030%.35
Accepted for publication April 26, 2012. From the *Pediatric Infectious Diseases Unit, Wolfson Medical Center, Holon; The Sackler School of Medicine, Tel Aviv University, Tel Aviv; Maccabi Health services, Hasharon County, Ramat-Hasharon; and Israel Center for Diseases Control, Tel-Hasomer, Israel. The authors have no funding or conflicts of interest to disclose. Address for Correspondence: Eli Somekh, MD, Pediatric Infectious Diseases Unit, the Edith Wolfson Medical Center, P.O. Box 5, Holon, Israel 58100. E-mail: esomeh@post.tau.ac.il. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journals Web site (www.pidj.com). Copyright 2012 by Lippincott Williams & Wilkins ISSN: 0891-3668/12/3109-0906 DOI: 10.1097/INF.0b013e31825d33f9

PATIENTS AND METHODS

This retrospective study was done with collaboration of Maccabi Health Services, which is the second largest health maintenance organization in Israel that uses a computerized database for recording all medical information of its patients. We retrieved the details of children who were 018 years of age, residing in Hasharon County in Central Israel, who were insured by Maccabi Health Services and were diagnosed with HZ by using International Classification of Diseases codes (053 group) from January 2006 until December 2008. The records were evaluated by 2 pediatric infectious disease experts to confirm the diagnoses. We excluded cases where there was no satisfactory description of the rash, when the diagnosis was not conclusive or the clinical description was not compatible with HZ. From the computerized files, we retrieved clinical information about HZ episodes and epidemiologic information regarding

906

| www.pidj.com

The Pediatric Infectious Disease Journal Volume31,Number9,September2012

The Pediatric Infectious Disease Journal Volume31,Number9,September2012

Herpes Zoster in Israeli Children

age, gender and underlying diseases. Medications consumption was evaluated according to actual purchase recordings from the computerized data of Maccabi Health Services. Data regarding primary varicella infection and vaccination against varicella were also recorded. The vaccine used was varicella Oka/GSK vaccine,Varilrix (GlaxoSmithKline Biologicals, Brentford, United Kingdom). Telephone interviews with a sample of patients parents who had HZ, were conducted by research investigators of the Israel Center for Disease Control, to get more clinical details. Parents were inquired about accompanying fever and pain, symptoms duration and exposure to chicken pox, before the appearance of HZ.

Incidence
Maccabi Health Services provides medical care to about 114,000 children residing in Hasharon County. Based on these data, we calculated a mean incidence of 130 of 100,000 person-years. Peak incidence was noted in children 911 years of age222 of 100,000 person-years (Fig., Supplemental Digital Content 1, http:// links.lww.com/INF/B247).

Varicella Vaccination
Varicella vaccine had been recommended by the Israeli Society of Pediatrics since 2000 but was not funded by the Israeli Ministry of Health until 2008 when routine varicella vaccination was begun. Fifty-five children with HZ were vaccinated against varicella, whereas 31,920 children residing in the county were vaccinated for varicella as compared with 395 cases of zoster among 82,080 children who presumably had natural varicella. Therefore, the protective effect of varicella vaccine for the emergence of HZ during childhood was found to be 0.36, (95% confidence interval: 0.270.48; P < 0.001).

Statistical Analysis
Data were recorded in an EXCEL spreadsheet, and analyzed by standard statistical tests (student t test, Pearson correlation and 2 tests).

Calculation of the Relative Risk for Zoster in Vaccinated Children

We estimated the population at risk for HZ by vaccination and disease history status by comparing the incidence of HZ among children younger than 18 years of age residing in the county who were vaccinated with the incidence among those with a presumed history of varicella disease (patients were considered to have a history of varicella disease if they had no documented history of vaccination). Age distribution of contracting varicella in children with HZ was compared with that of the other children in the county to determine the RR for HZ in children who had varicella during their first 2 years of life. The nominator was the rate of contracting varicella at a certain age among the 147 patients with zoster in whom the exact age of contracting varicella was known. The denominator was the rate of contracting varicella at the same age among children residing in Hasharon County who contracted varicella.

Location
Detailed clinical information regarding the location of the rash was available in 442 files. The detailed distribution of the rash is shown in Table 1. We found that vaccinated children were more likely to have the rash in the head, neck and lower extremities and less likely to have the rash in the lower trunk compared with nonvaccinated children.

Telephone Interviews
Telephone interviews were performed with 77 families chosen randomly. Forty parents (54.8%) noted that the zoster episode was accompanied by pain. The pain preceded the rash in 35%, and continued after the resolution of the rash in 7.5% of the 40 cases of HZ accompanied by pain. There were no cases of postherpetic neuralgia. Only 13 parents reported use of analgesics (paracetamol, ibuprofen or dipyrone) for pain relief. Ten parents (15.6%) had reported accompanying fever. The median durations of the rash, pain and fever were 8, 7.6 and 4.3 days, respectively. Only 5.6% of responders noted exposure to patients with chicken pox before zoster appearance. The exposure was either at school or in households.

Ethics Approval
The study was approved by the Ethical Committees of Maccabi Health Services, and the Wolfson Medical Center.

RESULTS
A total of 739 children were diagnosed with HZ during 2006 to 2008. We were able to gain access to 692 computerized files. After review of the records by pediatric infectious disease experts, 450 cases (215 male and 235 female) were approved and used for further analysis. Median age was 9.46 years. There was no seasonality in the distribution of cases. About half of the cases (221 cases) were diagnosed by pediatricians, 23.3% were diagnosed by dermatologists and 15.5% by general practitioners. The number of office visits for zoster episodes ranged from 1 to 7; median visit number was 1.

Acyclovir Treatment
One hundred ninety-eight (44%) children received systemic treatment with acyclovir, 130 (28.8%) were treated only topically and 52 children (13.6%) received both systemic and topical treatment.

Underlying Illnesses
Three patients had some degree of immunodeficiency. One child received methotrexate for juvenile rheumatoid arthritis, one had chronic renal failure and one patient had insulin-dependent diabetes mellitus.

TABLE 1. Location of Herpes Zoster Rash in Vaccinated and Unvaccinated Children

Location of the Rash* Head and neck Upper extremity Upper trunk Lower trunk Lower extremity
*

Total Group % (n = 442) 9.3 (n = 41) 9.9 (n = 44) 46.3 (n = 205) 21.7 (n = 96) 12.6 (n = 56)

Vaccinated Children % (n = 52) 19.2 (n = 10) 15.4 (n = 8) 36.5 (n = 19) 7.7 (n = 4) 21.1 (n = 11)

Nonvaccinated Children % (n = 390) 7.94 (n = 31) 9.2 (n = 36) 47.6 (n = 186) 23.6 (n = 92) 11.5 (n = 45)

P 0.017 NS NS 0.01 0.009

Some children had rash in more than one location, for example, involvement of both the neck, and the upper extremity. For the difference between vaccinated and unvaccinated children. NS indicates not significant.

2012 Lippincott Williams & Wilkins

www.pidj.com |

907

Stein et al

The Pediatric Infectious Disease Journal Volume 31, Number 9, September 2012

Age of Contracting Varicella

Records regarding primary infection with varicella were available for 147 of the 450 children with HZ. The age of contracting varicella in this group was compared with that of the other children in the county to determine the RR for HZ according to the age of contracting varicella (Figs., Supplemental Digital Content 2 and 3, http://links.lww.com/INF/B248 and http://links.lww.com/INF/B249). A significantly increased risk for HZ was found in children who contracted varicella during the first year (RR:13.5 [9.618.8]; P < 0.001) and also in the second year of life (RR: 2 [1.52.6]; P < 0.001). In addition, there was a significant correlation between the age of contracting chicken pox and the time interval to zoster breakout (RR = 0.16; P < 0.05). Specifically, the time period between varicella infection to HZ was shorter in children who had chicken pox before 2 years of age compared with children who had chicken pox at an older age (5.07 3.36 versus 6.08 3.5 years, respectively; P = 0.039).

Varicella in Vaccinated Children

Nine of the 55 children with zoster who were vaccinated for varicella had a history of varicella (Table, Supplemental Digital Content 4, http://links.lww.com/INF/B250). Five of them contracted varicella after vaccination (breakthrough varicella), whereas 4 children had varicella during the first year of life, before vaccination. None of these patients had the zoster rash over the upper extremities, and only 1 child had the zoster rash over the lower extremities.

DISCUSSION
HZ, a well-characterized entity in adults, has not been extensively investigated in healthy pediatric populations, and a large amount of the data about its incidence and associated risk factors are derived from studies performed several decades ago.68 In addition, there are only a few reports with regard to pediatric HZ in the era of varicella vaccination.1416 Our study was conducted in a population that was partially vaccinated for chicken pox, because this vaccine was recommended by the Israeli Society of Pediatrics but was not funded by the Israeli Ministry of Health. The average coverage in the pediatric population of Hasharon County with varicella vaccine during the study period (20062008) was 28%. As reported previously, HZ in the immune-competent child is a mild and self-limited illness that generally does not result in postherpetic neuralgia.7,8 The incidence found in the present study was higher than reported historically7: 130 versus 42 cases per 100,000 person-years. This rate may either reflect a real increase in childhood HZ rate, or may reflect a recording bias, as our study was based on analysis of computerized data where there was better access to medical records. The higher rate found in our study is probably not related to the introduction of varicella vaccine, because a similarly high rate was already found in a study done before varicella vaccination era in Iceland.8 A recent study from California16 also demonstrated rates that were close to our finding. However, as opposed to previous series showing a steady rise in the rate of childhood zoster with increasing age,6 our data demonstrated a significant peak in the incidence of HZ in children 916 years of age, with the highest incidence in children 911 years of age. This finding, which was consistent throughout each of the 3 study years, could be due to the effect of partial varicella vaccination; children 916 years of age were more likely to be infected with wild-type virus, but subsequently had limited re-exposure to chicken pox because of the introduction of varicella vaccination. On the other hand, older children had presumably more re-exposure since varicella vaccine was introduced in Israel in the year 2000. Evidence

supporting this assumption can be found in the recent populationbased active surveillance of Civen et al among residents of Atelope Valley, CA.15 This survey demonstrated an age-specific effect of the implementation of varicella vaccination on the incidence of HZ in children. Although the incidence of HZ among children < 10 years of age declined by 55%, the incidence among children 1019 years of age increased by 63%. Civen et als and our results are in accord with theoretical models that predicted the epidemiologic effect of varicella vaccination on the incidence of HZ.17,18 This situation is apparently only temporary, because universal vaccination (that has already been used in Israel since 2008) is eventually supposed to result in a lower incidence of HZ. The RR for HZ in vaccinated children was 0.36 (0.270.48) compared with unvaccinated children. In addition, only 1 of the 9 patients who were vaccinated and also contracted varicella before or after vaccination had the zoster rash over the extremities, suggesting that zoster in at least part of these cases might have been caused by wild-type varicella virus rather than by vaccine virus. Our study, comparing the rates of HZ in vaccinated and unvaccinated children in a population with partial varicella vaccination rate, has suggested that there is a protective effect of varicella vaccination on the incidence of HZ. This effect has been demonstrated previously in clinical studies19 and also in surveillance studies.1416 These results are reassuring with regard to the long-term effect of routine varicella vaccination on the incidence of HZ. The age for contracting varicella has been found in earlier studies to be a major factor for the early appearance of HZ,7,10 but this finding has not been demonstrated subsequently. We found that children who were younger than 1 year of age while contracting varicella had higher incidence of HZ (RR: 13.5 [9.618.8]). Moreover, our study is the first to suggest that even children who had chicken pox during their second year of life had a substantial rise in HZ incidence (RR: 2 [1.52.6]). In contrast to previous reports,8 there was a significant correlation between the age of contracting varicella and the time period from primary varicella infection to the appearance of HZ, and the time period between varicella infection and HZ was shorter in children who had chicken pox before 2 years of age. Current guidelines do not recommend administering varicella vaccine to children who are infected before 1 year of age. Our finding may suggest that the cellular immune response to varicella in this group of children is less robust and they may benefit from varicella vaccination.

Limitations
The main limitations of our study are the retrospective nature of data collection and the absence of virologic diagnosis. However, HZ has a distinct clinical presentation and data were meticulously reviewed by infectious diseases experts minimize misdiagnosis of HZ. In addition, there are no reference data regarding the incidence of pediatric HZ in Israel before the introduction of varicella vaccine. We were able to retrieve the age of contracting varicella in only 147 patients. This could be either due to the fact that some children had varicella without reporting it to their pediatricians. Another explanation is that many children, especially older children, had their varicella infection before their pediatricians were enrolled in the computerized recording system; therefore, the office visit for varicella infection was not recorded in the computerized system. REFERENCES
1. Gnann JW Jr, Whitley RJ. Clinical practice. Herpes zoster. N Engl J Med. 2002;347:340346. 2. Weinberg A, Lazar AA, Zerbe GO, et al. Influence of age and nature of primary infection on varicella-zoster virus-specific cell-mediated immune responses. J Infect Dis. 2010;201:10241030.

908

| www.pidj.com

2012 Lippincott Williams & Wilkins

The Pediatric Infectious Disease Journal Volume31,Number9,September2012

Herpes Zoster in Israeli Children

3. Straus SE, Ostrove JM, Inchausp G, et al. NIH conference. Varicella-zoster virus infections. Biology, natural history, treatment, and prevention. Ann Intern Med. 1988;108:221237. 4. Brisson M, Edmunds WJ, Law B, et al. Epidemiology of varicella zoster virus infection in Canada and the United Kingdom. Epidemiol Infect. 2001;127:305314. 5. Yawn BP, Saddier P, Wollan PC, et al. A population-based study of the incidence and complication rates of herpes zoster before zoster vaccine introduction. Mayo Clin Proc. 2007;82:13411349. 6. Hope-Simpson RE. The nature of herpes zoster: a long-term study and a new hypothesis. Proc R Soc Med. 1965;58:920. 7. Guess HA, Broughton DD, Melton LJ 3rd, et al. Epidemiology of herpes zoster in children and adolescents: a population-based study. Pediatrics. 1985;76:512517. 8. Petursson G, Helgason S, Gudmundsson S, et al. Herpes zoster in children and adolescents. Pediatr Infect Dis J. 1998;17:905908. 9. Whitley RJ. Changing dynamics of varicella-zoster virus infections in the 21st century: the impact of vaccination. J Infect Dis. 2005;191: 19992001. 10. Baba K, Yabuuchi H, Takahashi M, et al. Increased incidence of herpes zoster in normal children infected with varicella zoster virus during infancy: community-based follow-up study. J Pediatr. 1986;108:372377. 11. Jumaan AO, Yu O, Jackson LA, et al. Incidence of herpes zoster, before and after varicella-vaccination-associated decreases in the incidence of varicella, 1992-2002. J Infect Dis. 2005;191:20022007.

12. Tanuseputro P, Zagorski B, Chan KJ, et al. Population-based incidence of herpes zoster after introduction of a publicly funded varicella vaccination program. Vaccine. 2011;29:85808584. 13. Yih WK, Brooks DR, Lett SM, et al. The incidence of varicella and herpes zoster in Massachusetts as measured by the Behavioral Risk Factor Surveillance System (BRFSS) during a period of increasing varicella vaccine coverage, 19982003. BMC Public Health. 2005;5:68. 14. Tseng HF, Smith N, Marcy SM, et al. Risk factors of herpes zoster among children immunized with varicella vaccine: results from a nested case-control study. Pediatr Infect Dis J. 2010;29:205208. 15. Civen R, Chaves SS, Jumaan A, et al. The incidence and clinical characteristics of herpes zoster among children and adolescents after implementation of varicella vaccination. Pediatr Infect Dis J. 2009;28:954959. 16. Weinmann S, Chun C, Riedlinge RK. Herpes zoster in children in the varicella vaccine era. Paper presented at: 48th Annual Meeting of the Infectious Diseases Society of America (IDSA); October 2124, 2010; Vancouver, Canada. 17. Brisson M, Gay NJ, Edmunds WJ, et al. Exposure to varicella boosts immunity to herpes-zoster: implications for mass vaccination against chickenpox. Vaccine. 2002;20:25002507. 18. Garnett GP, Grenfell BT. The epidemiology of varicella-zoster virus infections: the influence of varicella on the prevalence of herpes zoster. Epidemiol Infect. 1992;108:513528. 19. Hardy I, Gershon AA, Steinberg SP, et al. The incidence of zoster after immunization with live attenuated varicella vaccine. A study in children with leukemia. Varicella Vaccine Collaborative Study Group. N Engl J Med. 1991;325:15451550.

2012 Lippincott Williams & Wilkins

www.pidj.com |

909