Better specimen collection urged for herpes zoster detection in vaccinated children

**Chun C. Pediatr Infect Dis J. 2011;doi:10.1097/INF.0b013e3182137e35.**

The varicella vaccine has significantly lowered the rate of varicella disease, but it has had one unfortunate adverse effect, the introduction of herpes zoster from vaccine-type virus.

A brief report published by Colleen Chun, MD, and colleagues of Kaiser Permanente Northwest in Portland, Ore., noted six confirmed cases of herpes zoster in vaccinated children, two of which they attributed to the vaccine.

Chun and colleagues recruited vaccinated patients and collected specimens in their hospital’s pediatric population from March 2000 to March 2001 and April 2002 to September 2003.

The researchers said they could confirm the diagnosis using varicella polymerase chain reaction (PCR) in about half of the skin specimens collected from children with suspected herpes zoster, and about one-third of these were attributed to vaccine-type virus.

These rates were somewhat lower than data reported by the Vaccine Adverse Event Reporting System (VAERS), which reported that “61% of skin specimens from patients with clinically diagnosed herpes zoster were PCR-positive and that 67% of confirmed cases were due to vaccine-type varicella,” they said.

The difference may have been caused by a higher percentage of “scab-type specimens,” in the VAERS data, according to the researchers.

“Although our study size was small, the methodology may be more accurate, as it involved prospective, population-based, comprehensive electronic surveillance with review of the electronic medical records of all identified subjects,” they said. “This contrasts with the limitations of passive reporting bias with VAERS, including lack of consistent diagnostic criteria prompting a report, limited clinical data, lack of verification for each identified case, and underreporting of events occurring much later after vaccination.”

The researchers said serum varicella zoster immunoglobulin M is nondiagnostic for herpes zoster among vaccinated children, and they found similar results using IgG. Further study to optimize specimen collection for PCR confirmation in patients with herpes zoster who do not have scabs is encouraged.

Disclosures: The Kaiser Permanente Northwest and CDC institutional review boards approved the study.

In the Chun study the authors state: "The varicella vaccine has significantly lowered the rate of varicella disease, but it has had one unfortunate adverse effect, the introduction of herpes zoster from vaccine-type virus," I would point out that the incidence of zoster is lower after vaccination than after natural infection, so Oka zoster is not exactly an "adverse event" of vaccination. Better a few cases of zoster due to Oka than many more due to wild-type VZV. For example, the incidence reported by Civen et al (Pediatr infect Dis 2009) in young children who were vaccinated was 0.33/1,000 person-years of observation from 2000-2006, down from 0.75/1,000 person-years of observation in 2000.

With regard to the low rate of zoster shown as due to Oka by Chun and colleagues, this study involves very small numbers of patients. PCR data exists for only six patients; 33% Oka for zoster cases in vaccinees may seem lower than other reports involving many more patients (Chavez 67% J Infect Dis 2008; and Galea 60% J Infect Dis 2008) but is probably related to the very low numbers of patients. In addition, 3/4 children with zoster due to wild type virus gave a history of having had varicella before vaccination. If these three children are eliminated, 66% had zoster due to Oka, but this is only considering three children. What is particularly interesting in all the published studies, however, is the large percentage of vaccinees who develop zoster caused by the wild-type virus. Most of them must have been infected with wild-type VZV, despite vaccination, presumably without symptoms of varicella.

In some hands, PCR works best when using vesicular fluid or scabs, and swabs are generally less satisfactory for diagnosis (Vazquez et al, New Engl J Med 2001; JAMA 2004). Another paper , however, indicates that swabs are very useful diagnostically, and that oral fluids, and even urine may yield a positive PCR for VZV in varicella (Leung et al, Clin Infect Dis, 2010). The methodology in the Chen paper (population-based, etc) trumps that of VAERS, but the specimen collection may not. There is an area of general agreement though; serology is not now the way to go for diagnosis of varicella or zoster.

– Anne A. Gershon, MD
Director, Division of Pediatric Infectious Diseases, Columbia University Medical Center

Disclosure: Dr. Gerson reports no relevant financial disclosures

Follow the PediatricSuperSite.com on Twitter.