Frequency of Herpes Zoster Recurrence
To the Editor: We read with great interest the study by Yawn et al,1 published in the February 2011 issue of Mayo Clinic Proceedings, which found that the frequency of herpes zoster (HZ) recurrence in a community population was higher than previously reported. The reported results are somewhat unexpected. It is generally accepted that the lifetime incidence of a second episode of HZ in immunocompetent individuals is between 1% and 5% and that the recurrence is typically many years after the first episode. A number of issues need to be clarified about the study by Yawn et al before the relevance of these findings can be determined.
First, the diagnosis of HZ in the study by Yawn et al1 was established clinically. The clinical diagnosis of HZ can be difficult and is subject to error. In a zoster prevention study,2 HZ was ruled out by laboratory testing (polymerase chain reaction or viral culture) in 24% of patients with a clinical diagnosis of HZ, suggesting that clinical diagnosis can on occasion be incorrect. No previous unreported case of HZ was revealed in closeout interviews with patients in that study.2 Similarly, a prospective study of HZ diagnoses by general practitioners in the United Kingdom found 17% of diagnoses to be incorrect.3 More specifically, of the 230 patients diagnosed clinically as having HZ, only 204 cases were confirmed by immunofluorescence and/or polymerase chain reaction. Of the 26 patients who had no evidence of HZ on laboratory tests, 10 patients had herpes simplex, and the rest had other dermatological diseases.3 In the study by Yawn et al,1 only 25% of the recurrent episodes were confirmed by laboratory analysis, but it is unclear whether the first episode was also confirmed by laboratory results.
We recently conducted a study of 173 patients (median age, 75 years) with postherpetic neuralgia. In these patients, the median duration of postherpetic neuralgia was 23 months (range, 2-207 months).4 None of the patients presented with HZ during the follow-up visit or had a history of HZ. We realize that the number of people involved is low compared with the cohort in the study by Yawn et al, but they presented with both the main risk factors that, according to that study, predict the likelihood of recurrences.
Another point in need of clarification is the choice by Yawn et al to include in the analysis the 139 individuals (8.3% of the total population sample) who were immunocompromised at the time of the index HZ episode.1 It is known that the risk of HZ and its recurrence is increased in persons with a compromised immune system. Indeed, HZ rates of 29.4 to 51.5 per 1000 person-years have been reported among adults infected with the human immunodeficiency virus,5 and high rates have also been reported in persons with systemic lupus erythematosus, rheumatoid arthritis, granulomatosis with polyangiitis (Wegener), and inflammatory bowel disease.5 For most of these conditions, data are insufficient to determine how much of the risk is attributable to the underlying disease vs its treatment. Hence, the inclusion of immunocompromised people may represent a bias.
Finally, the authors discuss the possibility of an innate (possibly genetic) predisposition for HZ, which has received attention in studies of HZ in families. We have recently examined the possibility that a family history of HZ represents a risk factor for zoster development, but our findings could not confirm the hypothesis.4
Given the ready availability and the safety profile of the HZ vaccine, we agree that it is important to identify additional categories of people who could benefit from it. However, in our opinion, well-designed prospective studies are needed to ascertain the real likelihood of recurrences in the general population.