ORIGINAL ARTICLE

Clinical and immunologic features

of recurrent herpes zoster (HZ)
Yuki Nakamura, MD,a Fumi Miyagawa, MD, PhD,a Aiko Okazaki, MD, PhD,a
Yoshinobu Okuno, MD, PhD,b Yasuko Mori, MD, PhD,c Hiroyasu Iso, MD, PhD,d
Koichi Yamanishi, MD, PhD,b,e and Hideo Asada, MD, PhD,a for the Shozu Herpes Zoster Study Group
Nara, Kagawa, Kobe, Osaka, and Ibaraki, Japan

Background: Recurrent herpes zoster (HZ) is thought to be rare, but there have been few large-scale
studies of recurrent HZ.

Objective: We conducted a large-scale prospective cohort study to characterize recurrent HZ.

Methods: We examined 12,522 participants aged 50 years or older in Shozu County and followed them up
for 3 years. We compared the incidence of HZ and postherpetic neuralgia, severity of skin lesions and acute
pain, cell-mediated immunity, and varicella-zoster virusespecific antibody titer between primary and
recurrent HZ.

Results: A total of 401 participants developed HZ: 341 with primary HZ and 60 with recurrent HZ. Skin
lesions and acute pain were significantly milder and the incidence of postherpetic neuralgia was lower in
patients aged 50 to 79 years with recurrent HZ than in those with primary HZ. Varicella-zoster virus skin test
induced a stronger reaction in patients aged 50 to 79 years with recurrent HZ than in those with primary
HZ.

Limitations: Information on previous HZ episodes was self-reported by participants, so it could not be

confirmed that they actually had a history of HZ.

Conclusion: Recurrent HZ was associated with milder clinical symptoms than primary HZ, probably
because of stronger varicella-zoster virusespecific cell-mediated immunity in the patients with recurrence.
( J Am Acad Dermatol http://dx.doi.org/10.1016/j.jaad.2016.05.037.)

Key words: cell-mediated immunity; humoral immunity; large-scale community-based prospective cohort
study; primary herpes zoster; recurrent herpes zoster; varicella-zoster virus.

V aricella-zoster virus (VZV) causes varicella in and to be limited to immunocompromised patients.

childhood as a primary infection.1,2 After Although the plausibility, actual occurrence, and
infection occurs, VZV remains latent in the frequency of recurrent HZ have been debated,4-8 there
sensory nerve ganglia throughout the lifetime of the have been few large-scale studies examining recurrent
host, with reactivation caused by impairment of HZ. Recently, we carried out a large-scale community-
immunity, aging, or stress resulting in herpes zoster based prospective cohort study on HZ.9-14 In the
(HZ).3 Recurrent HZ is generally thought to be rare current study, we analyzed data obtained from this

From the Department of Dermatology, Nara Medical University [KHC1102], and funds from the Research Foundation for
School of Medicinea; Research Foundation for Microbial Dis- Microbial Diseases of Osaka University.
eases of Osaka University, Kagawab; Division of Clinical Conflicts of interest: None declared.
Virology, Department of Microbiology and Infectious Diseases, Accepted for publication May 22, 2016.
Kobe University Graduate School of Medicinec; Public Health, Reprint requests: Hideo Asada, MD, PhD, Department of
Department of Social Medicine, Osaka University Graduate Dermatology, Nara Medical University School of Medicine, 840
School of Medicined; and National Institute of Biomedical Shijo-cho, Kashihara, Nara 634-8522, Japan. E-mail: asadah@
Innovation, Ibaraki.e naramed-u.ac.jp.
Supported by Grants-in-Aid for Scientific Research from the Published online July 26, 2016.
Japanese Ministry of Health, Labor, and Welfare for Research 0190-9622/$36.00
on Publicly Essential Drugs and Medical Devices (fiscal years Ó 2016 by the American Academy of Dermatology, Inc.
2008-2010) [KHC1102] and for Public-Private Sector Joint http://dx.doi.org/10.1016/j.jaad.2016.05.037
Research on Publicly Essential Drugs (fiscal year 2011-2012)

1
2 Nakamura et al J AM ACAD DERMATOL

cohort study over the entire follow-up period of 3 years from the Research Foundation for Microbial
to explore the characteristics of recurrent HZ in Diseases of Osaka University, Japan). The VZV skin
comparison with primary HZ. We found that recurrent test antigen was licensed in Japan in 1990.15,16 In
HZ was characterized by less severe skin lesions, less brief, to perform this test, culture fluid of MRC-5 cells
severe pain, and a lower risk of postherpetic neuralgia infected with VZV (Oka parental strain) was har-
(PHN) than primary HZ in the age group 50 to 79 years. vested and centrifuged. Then the supernatant was
collected and concentrated by ultrafiltration for
METHODS storage as a bulk prepara-
Study design tion, mainly consisting of
The study design was re- CAPSULE SUMMARY
VZV glycoproteins (III and
ported previously.9 Briefly, IV). The VZV glycoprotein
dRecurrent herpes zoster (HZ) has not
12,522 Japanese persons content was evaluated by
been well characterized.
aged 50 years or older in enzyme-linked immunosor-
Shozu County were enrolled dSymptoms of recurrent HZ were bent assay (ELISA). We
in this study on October 1, significantly milder than those of primary injected 100 L of VZV skin
2008. Among them, 5683 HZ. test antigen intradermally
participants underwent a d
Relatively strong cell-mediated immunity into the forearm of each
VZV skin test at registration. against varicella-zoster virus might participant at registration.
Participants were followed contribute to less severe symptoms in Erythema and edema were
up for 3 years by telephone patients with recurrent HZ. evaluated 48 hours after the
interview every 4 weeks. If injection, and the longest
they developed possible HZ, diameter was measured as
the participants attended hospitals or clinics in Shozu the test result. The extent of edema was also assessed
County registered with this study where their symp- by palpation with the index finger.17
toms, severity of pain, and humoral immunity were
evaluated. Photographs of skin lesions were also Assessment of humoral immunity for VZV
taken for later assessment, and samples of vesicles Blood samples were obtained from participants
and crusts were collected for detection of VZV by who developed HZ during the follow-up period to
polymerase chain reaction (PCR). All of the partici- assess humoral immunity. Serologic tests for
pants who developed possible HZ were examined VZV antibodies, such as the neutralization test,
by PCR and serologic testing. Cases were confirmed immunoadherence hemagglutination test, and
by the clinical evaluation committee of Nara Medical glycoprotein ELISA, were performed as described
University School of Medicine in Japan, consisting of previously.17-22 The glycoprotein ELISA was per-
3 dermatologists with expertise in HZ, and the final formed by using microtiter plates coated with
diagnosis was based on symptoms together with the purified VZV glycoproteins to determine the titer of
results of PCR and serologic tests. IgG antibodies against viral glycoproteins. The
Statistical analyses were performed on the partic- neutralization test for VZV was done by the plaque
ipants who developed HZ during the 3-year reduction technique, with the neutralizing antibody
follow-up period, after excluding 6.4% who titer corresponding to a 50% reduction in plaque
withdrew from the study, were lost to follow-up, or count, whereas the immunoadherence hemaggluti-
died during the study. This study was approved by nation test for VZV antibodies was based on fixation
research ethics committee in Nara Medical University of complement by antigen-antibody complexes.
School of Medicine.
Evaluation of HZ skin lesions
Diagnosis of recurrent HZ Initial evaluation of subjects with possible HZ was
A diagnosis of recurrent HZ was made in subjects done by physicians in Shozu County using a standard
who met the following criteria: (1) a previous survey form that we developed. The clinical diag-
episode of HZ before this study, and (2) diagnosis nosis was confirmed by the dermatologists using
of HZ during the study period based on symptoms photographs of skin lesions. The following variables
combined with the results of PCR and serologic tests. were also assessed: the presence of underlying
diseases, immunosuppressant/antineoplastic ther-
VZV skin test apy, date of onset of the rash, distribution of the
To evaluate sensitivity to VZV and the level of rash, properties of the rash (erythema; number of
cell-mediated immunity (CMI), we used the VZV vesicles, pustules, erosions, and crusts; and ulcera-
antigen Biken (a commercially available reagent tion and fusion of vesicles), date of onset of pain, and
J AM ACAD DERMATOL Nakamura et al 3

RESULTS
Abbreviations used:
Characteristics of the study population
CMI: cell-mediated immunity Among 12,522 participants (aged $50 years), 401
ELISA: enzyme-linked immunosorbent assay
HZ: herpes zoster developed HZ: 341 with primary HZ and 60 with
PCR: polymerase chain reaction recurrent HZ (Table I). The 341 patients with primary
PHN: postherpetic neuralgia HZ were 121 men and 220 women (median age
VZV: varicella-zoster virus
69.4 years), comprising 59 patients aged 50 to
59 years, 92 patients aged 60 to 69 years, 113 patients
aged 70 to 79 years, and 77 patients aged 80 years
other associated symptoms (eg, fever, headache, and older. Malignancies were reported in 19 patients
generalized HZ, multidermatomal HZ, eye compli- and 20 patients were on immunosuppressive therapy
cations, motor paralysis, Ramsay Hunt syndrome). (Table I). The patients with 60 recurrent HZ were 18
We developed a clinical severity score for HZ skin men and 42 women (median age 70.1 years),
lesions in which scores from 0 up to 3 points were comprising 9 patients aged 50 to 59 years, 16 patients
assigned for each of the following criteria: the aged 60 to 69 years, 21 patients aged 70 to 79 years,
percentage of erythema in the lesional dermatome and 14 patients aged 80 years and older.
(score: 0-3); the number of vesicles, pustules, Malignancies were reported in 2 patients and 3
erosions, or crusts (score: 0-3); the presence or patients were on immunosuppressive therapy
absence of fusion of vesicles (score: 0 or 1); the (Table I). The characteristics of all subjects with
presence or absence of ulcer formation (score: 0 or primary or recurrent HZ are summarized in Table I.
1); the number of lesional dermatomes (score: 0 or There were no significant differences of age or most
1); and the presence or absence of generalized other characteristics between the subjects with
eruptions (score: 0 or 1). The severity score for HZ primary or recurrent HZ. The incidence of primary
skin lesions was calculated as the sum of each of HZ was significantly higher in women than in men
these scores at the peak of disease activity with a (OR, 1.542; 95% CI, 1.234e1.927; P \.001), whereas
maximum possible score of 10, as described previ- there was no significant gender difference in the
ously.12,13 Please see the Supplementary Data, avail- incidence of recurrent HZ (Table I).
able at http://www.jaad.org.
Less severe skin lesions in recurrent HZ
Evaluation of zoster-associated pain Using the highest skin lesion score obtained
The severity of pain was evaluated by using a face during the course of HZ as the severity score for
scale from 0 (no pain) to 5 (pain disturbing sleep)12 each subject, the average scores in each age group
on days 0, 1, 2, 3, 4, 5, 6, 13, 20, 27, 34, 41, 48, 55, 85, were compared between primary and recurrent HZ
115, 145, and 175 after the initial medical consulta- (Fig 1, A). The severity score was lower for recurrent
tion. Some patients with early resolution of pain HZ than primary HZ in the patients aged 70 to
were only followed up until day 7 after the pain 79 years (P \.001), but not in the other age groups
disappeared. The severity score for zoster-associated (Fig 1, A). When the 3 younger age groups (50-59,
pain was calculated as the area under the curve of 60-69, and 70-79 years [50-79 years]) were combined
pain versus time during the 12-week period after the for comparison, patients with recurrent HZ exhibited
onset of HZ rash. We identified patients with PHN by less severe skin lesions than those with primary HZ
conducting telephone interviews once every 4 weeks (P \ .001) (Fig 1, B). However, no significant
for each study participant who developed HZ. A difference was observed between primary and
diagnosis of PHN was made in patients with either: recurrent HZ in patients aged 80 years and older
(1) zoster-associated pain for at least 3 months after (Fig 1, B).
the onset, or (2) recurrence of zoster-associated pain
after 3 months or longer.
Less severe zoster-associated pain in recurrent
HZ
Statistical analysis Next we evaluated zoster-associated pain in the
Analysis of covariance was used to compare the acute and subacute stages of the disease. Similar to
severity of skin lesions, and zoster-associated pain the findings for skin lesions (Fig 1), acute and
between subjects with primary or recurrent HZ. subacute zoster-associated pain was significantly
Logistic regression analysis was used to compare less severe in recurrent HZ than primary HZ in
the risk of PHN and the VZV skin test reaction patients aged 70 to 79 years (Fig 2, A) (P \ .001)
between primary and recurrent HZ. and those aged 50 to 79 years (Fig 2, B) (P \.001).
 4 Nakamura et al J AM ACAD DERMATOL

PHN less likely in recurrent HZ

3.77 6 0.05 5.37 6 0.19 5.68 6 0.20

3.85 6 0.11 5.28 6 0.40 5.70 6 0.45

(log2) 6SE

n = 59

n = 10
There was a significantly lower incidence of PHN

IAHA,
Acute phase VZV antibody titer
in recurrent HZ than primary HZ in patients aged 50
to 79 years (OR, 0.199; 95% CI, 0.047-0.852; P = .03),
but not in those aged 80 years or older (Fig 3).
(log2) 6SE

n = 60

n = 10
NT,

Stronger VZV skin test reaction with more

edema in recurrent HZ
From among 5683 participants who underwent a
Patients with Glycoprotein-

(log10) 6SE

VZV skin test at registration, we extracted the data of

n = 60

n = 10
ELISA,

participants who developed HZ. There was no

significant difference of the erythema response in
the skin test between subjects with primary and
recurrent HZ (Table I). The diameter of erythema
($10 mm)

n = 148y

21.88%y
n = 32y
9.46%y
edema

tended to be larger in patients aged 50 to 79 years

Patient background factors

with recurrent HZ (mean 6 SE: 11.44 6 2.01 mm)

VZV skin test

than in those with primary HZ (mean 6 SE:

8.69 6 0.71 mm), but the difference was not
10.11 6 1.83
8.03 6 0.66

HZ, Herpes zoster; F, female; IAHA, immunoadherence hemagglutination; M, male; NT, neutralization test; VZV, varicella-zoster virus.
Diameter of
erythema,
mm, 6SE

n = 139

n = 31

significant. On the other hand, the percentage of

patients with edema ($10 mm) in the skin test was
significantly higher in recurrent HZ than primary HZ
when patients aged 50 to 79 years were compared
immunosuppressive

immunosuppressive

(OR, 3.05; 95% CI, 1.076-8.627; P = .036) (Fig 4).

The percentage of patients with edema in the skin test was significantly higher in recurrent HZ than primary HZ.
M (n = 121); Malignancy (n = 19);
Immunosuppression

Malignancy (n = 2);
therapy (n = 20)

therapy (n = 3)

Similar VZV-specific humoral immunity in

primary and recurrent HZ
We compared serum levels of VZV antibodies
between primary and recurrent HZ. Participants who
Table I. Characteristics of patients with primary and recurrent herpes zoster

developed HZ during this study were examined for

antibodies at the initial medical consultation. To
F (n = 220)

F (n = 42)
M (n = 18);

minimize the effect of VZV reactivation on antibody

Sex

titers, we only included patients with HZ who

presented within 2 days after the onset of rash. The
*The incidence of primary HZ was significantly higher in women than in men.

serum VZV antibody titers determined by glycopro-

Median

tein ELISA, neutralization, and immunoadherence

age, y
69.4

70.1

hemagglutination test results did not differ between

the primary and recurrent HZ groups (Table I),
1.10% (M 0.84%;

1.01% (M 0.91%;

suggesting that the risk of developing recurrent HZ

HZ incidence/y

F 1.33%)*

might not depend on the humoral immune status.

F 1.07%)

DISCUSSION
Among the 12,522 participants investigated in this
study, 401 participants developed HZ, including 341
subjects years with HZ
No. of Person- patients
No. of

60
341

with primary HZ and 60 with recurrent HZ. The

annual incidence of primary HZ was 1.10%, whereas
that of recurrent HZ was 1.01%.
10,530 31,015

5917

We showed that patients with recurrent HZ aged

50 to 79 years developed less severe skin lesions, had
1992

less severe acute pain, and had a lower risk of PHN

than patients with primary HZ of the same age group,
whereas these differences were not noted in patients
Recurrent HZ

aged 80 years and older. These findings may be

Primary HZ

related to stronger CMI in patients with recurrent HZ

Group

than in patients with primary HZ, because the VZV

skin test induced more edema and tended to induce
y
J AM ACAD DERMATOL Nakamura et al 5

Fig 1. Severity of skin lesions between primary and recurrent herpes zoster (HZ ) in each age
group (A) and in 2 age groups (50-79 vs $80 years) (B). Bars indicate the mean severity scores
adjusted for sex (error bars, SEM).

Fig 2. Severity of acute and subacute zoster-associated pain between primary and recurrent
herpes zoster (HZ ) in each age group (A) and in 2 age groups (50-79 vs $80 years) (B). Bars
indicate the mean severity scores adjusted for sex (error bars, SEM).

a larger area of erythema in the patients with infection boosts immunity against VZV. We previ-
recurrent HZ aged 50 to 79 years than in those with ously reported that CMI for VZV was inversely
primary HZ. The VZV skin test used in the current associated with the severity of skin lesions and
study does not cause the typical induration that is HZ-associated acute/subacute pain.12 These findings
commonly induced by the tuberculin skin test, but suggest that the stronger VZV-specific CMI observed
instead induces edema accompanied by erythema. in patients with recurrent HZ in the current study
Okuno et al11 reported that the diameters of edema may reduce the severity of skin lesions and acute
and erythema induced by the VZV skin test were pain in this group. We have also previously reported
inversely correlated with the incidence of HZ. In that patients with HZ with a strong VZV skin test
contrast, there was no significant difference of response had a significantly lower risk of PHN than
VZV-specific humoral immunity between primary patients with a weak skin test response.13 That report
HZ and recurrent HZ. These results suggest that the is consistent with our current finding that recurrent
clinical differences between primary HZ and HZ is associated with a lower risk of PHN.
recurrent HZ may be more dependent on CMI than In patients aged 80 years and older with primary
humoral immunity. or recurrent HZ, there were no significant differences
VZV-specific CMI may be stronger in patients with in the severity of skin lesions and acute pain, the
recurrent HZ because reactivation of primary VZV incidence of PHN, and CMI for VZV. There are 2
6 Nakamura et al J AM ACAD DERMATOL

weaker booster response to VZV and thereby be

susceptible to recurrent HZ despite a recent episode
of primary HZ. Second, in patients with earlier onset
of primary HZ, immunity to VZV might be expected
to decrease with aging and recurrent HZ could occur
after a long interval.
It has been reported that the risk of recurrent HZ
ranges from 1% to 6%, with most studies showing
that the annual incidence of recurrent HZ is nearly
1%.23 We obtained a similar finding, with the annual
incidence of recurrent HZ being 1.01%.
In this study, there was no significant difference in
the incidence of recurrent HZ between men and
women or between immunocompetent and immu-
nocompromised patients. In contrast to our findings,
previous studies conducted in clinics or hospitals
have shown that the incidence of recurrent HZ and
primary HZ are higher in women than in men.24 This
Fig 3. Incidence of postherpetic neuralgia (PHN ) discrepancy is most likely a result of the differences
between primary and recurrent herpes zoster (HZ ) by in study design. We performed a community-based
logistic regression analysis in 2 age groups (50-79 cohort study that included patients who do not
vs $80 years). Bars indicate the mean incidence of PHN usually seek medical attention unless specifically
adjusted for sex (error bars, SEM). asked to do so. In addition, we found that recurrent
HZ-associated pain tended to be less severe in men
than in women, suggesting that men may be less
likely to seek medical attention (data not shown). For
the above 2 reasons, the incidence of HZ in men
might be underreported by studies conducted in
clinics or hospitals.
Because the risk of HZ increases as the population
ages and the proportion of elderly persons in the
population is continuously increasing in Japan, the
incidence of recurrent and primary HZ are expected
to rise substantially in the future. Our results
showed that recurrent HZ could occur even if
patients had stronger anti-VZV CMI. Therefore,
identification of factors associated with recurrent
HZ is required to allow the implementation of
preventive measures.

REFERENCES
1. Grose C. Variation on a theme by Fenner: pathogenesis of
chickenpox. Pediatrics. 1981;68:735-737.
2. Ku CC, Besser J, Abendroth A, et al. Varicella-zoster virus
pathogenesis and immunobiology: new concepts emerging
Fig 4. Percentage of patients with edema ($10 mm) in the from investigations with the SCIDhu mouse model. J Virol.
varicella-zoster virus (VZV) skin test among patients with 2005;79:2651-2658.
primary and recurrent herpes zoster (HZ ) in 2 age groups 3. Mims C, White D. Viral pathogenesis and immunology. Oxford
(United Kingdom): Blackwell Scientific Publications; 1984.
(50-79 vs $80 years). Bars indicate the mean positive rate
4. Chien AJ, Olerud JE. Why do so many clinicians believe that
of edema ($10 mm) adjusted for sex (error bars, SEM).
recurrent zoster is common? Dermatol Online J. 2007;13(2):
517-524.
possible explanations for these findings. First, 5. Heskel NS, Hanifin JM. ‘‘Recurrent herpes zoster’’: an unproved
entity? J Am Acad Dermatol. 1984;10:486-490.
although we have no precise data about the interval
6. Horiuchi Y. Recurrent herpes zoster. J Dermatol. 1998;25:
period between primary and recurrent HZ in 347-348.
individual patients, it is possible that the patients 7. Burkhart CN. Recurrent herpes zoster revisited. Int J Dermatol.
with later onset of primary HZ might develop a 2002;41:528.
J AM ACAD DERMATOL Nakamura et al 7

8. Bansal R. Recurrent herpes zoster. Int J Dermatol. 2001;40: 17. Takahashi M, Okada S, Miyagawa H, et al. Enhancement of
542-543. immunity against VZV by giving live varicella vaccine to the
9. Takao Y, Miyazaki Y, Onishi F, et al. The Shozu herpes zoster elderly assessed by VZV skin test and IAHA, gpELISA antibody
(SHEZ) study: rationale, design, and description of a assay. Vaccine. 2003;21:3845-3853.
prospective cohort study. J Epidemiol. 2012;22:167-174. 18. Sadaoka K, Okamoto S, Gomi Y, et al. Measurement of
10. Tang H, Moriishi E, Okamoto S, et al. A community-based varicella-zoster virus (VZV)-specific cell-mediated immunity:
survey of varicella-zoster virus-specific immune responses in comparison between VZV skin test and interferon-gamma
the elderly. J Clin Virol. 2012;55:46-50. enzyme-linked immunospot assay. J Infect Dis. 2008;198:
11. Okuno Y, Takao Y, Miyazaki Y, et al. Assessment of skin test 1327-1333.
with varicella-zoster virus antigen for predicting the risk of 19. Provost PJ, Krah DL, Kuter BJ, et al. Antibody assays suitable
herpes zoster. Epidemiol Infect. 2013;141:706-713. for assessing immune responses to live varicella vaccine.
12. Asada H, Nagayama K, Okazaki A, et al. An inverse correlation Vaccine. 1991;9:111-116.
of VZV skin-test reaction, but not antibody, with severity of 20. Gershon AA, Kalter ZG, Steinberg S, et al. Detection of
herpes zoster skin symptoms and zoster-associated pain. antibody to varicella-zoster virus by immune adherence
J Dermatol Sci. 2013;69:243-249. hemagglutination. Proc Soc Exp Biol Med. 1976;151:762-765.
13. Imoto K, Okazaki A, Onishi F, et al. VZV skin-test reaction, but 21. Kalter ZG, Steinberg S, Gershon AA. Immune adherence hem-
not antibody, is an important predictive factor for agglutination: further observations on demonstration of anti-
postherpetic neuralgia. J Dermatol Sci. 2013;79:235-240. body to varicella-zoster virus. J Infect Dis. 1977;135:1010-1013.
14. Takao Y, Miyazaki Y, Okeda M, et al. Incidences of herpes 22. Yamada A, Ogino S, Asano Y, et al. Comparison of 4 serological
zoster and postherpetic neuralgia in Japanese adults aged testsecomplement fixation, neutralization, fluorescent
50 years and older from a community-based prospective antibody to membrane antigen and immune adherence
cohort study: the SHEZ study. J Epidemiol. 2015;25(10): hemagglutinationefor assay of antibody to varicella-zoster
617-625. (V-Z) virus. Biken J. 1979;22:55-60.
15. Kamiya H, Ihara T, Hattori A, et al. Diagnostic skin test 23. Kawai K, Gebremeskel BG, Acosta CJ. Systematic review of
reactions with varicella virus antigen and clinical application incidence and complications of herpes zoster: towards a
of the test. J Infect Dis. 1977;136(6):784-788. global perspective. BMJ Open. 2014;4:e004833.
16. Asano Y, Shiraki K, Takahashi M, et al. Soluble skin test antigen 24. Yawn BP, Wollan PC, Kurland MJ, et al. Herpes zoster
of varicella-zoster virus prepared from the fluid of infected recurrences more frequent than previously reported. Mayo
cultures. J Infect Dis. 1981;143:687-692. Clin Proc. 2011;86:88-93.
7.e1 Nakamura et al J AM ACAD DERMATOL

SUPPLEMENTARY DATA the table, on the basis of the distribution and

The severity score for herpes zoster skin lesion properties of rash at the peak of disease activity.
was calculated as the sum of each score defined in

Criteria for determining severity score of HZ skin lesions

Percentage
of erythema in Number of vesicles,
the lesional pustules,erosions, Fusion of Ulcer Number of lesional Generalized
Score dermatome or crusts vesicles formation dermatomes eruption
0 0 0 Absent Absent Single Absent
1 \20 \10 Present Present More than one Present
2 20-50 10-50
3 $ 50 $ 50