Varicella Zoster antibodies among health care workers in a university hospital, Teheran, Iran.

IJOMEH 2010;23(1) 27
O R I G I N A L P A P E R S
International Journal of Occupational Medicine and Environmental Health 2010;23(1):27 – 32
DOI 10.2478/v10001-010-0011-x
VARICELLA ZOSTER ANTIBODIES
AMONG HEALTH CARE WORKERS
IN A UNIVERSITY HOSPITAL, TEHERAN, IRAN
MAHSHID TALEBI-TAHER1, MARIA NOORI2, AHMAD-REZA SHAMSHIRI3, and MITRA BARATI4
1 Iran University of Medical Sciences, Teheran, Iran
Infectious Disease Department
2 Iran University of Medical Sciences, Teheran, Iran
Internal Medicine Department
3 Teheran University of Medical Sciences, Teheran, Iran
Department of Epidemiology and Biostatistics, Faculty of Public Health and Institute of Public Health Research
4 Iran University of Medical Sciences, Teheran, Iran
Infectious Disease Department
Abstract
Objectives: This study was designed to evaluate the immune status of health care workers against varicella zoster in a uni-
versity hospital in Teheran, Iran, and to compare the history of chickenpox infection with the presence of varicella anti-
bodies in this population. Methods: Serologic testing for varicella was performed for 405 health care workers with dif-
ferent job categories and at different age. The enzyme immunoassay was used for determining IgG antibodies against
varicella zoster virus Results: A total of 405 health care workers, aged 19–50 years (median: 29 years), were examined.
Of these, 289 (71.4%) were found to be seropositive. No statistically significant differences were observed between gender, 
age, or occupation, and seropositivity (p = 0.09, 0.75, 0.54. respectively). Statistical analysis revealed that the correlation be-
tween chickenpox history and seropositivity showed a 62.3% sensitivity, 72.4% specificity, 84.9% positive predictive value, 
and 43.5% negative predictive value. Conclusions: Serologic screening of health care workers is essential to determine their
immunity to varicella, regardless of the age, occupation and history of infection. This population is recommended to be
considered a target group for future immunization programs in Iran.
Key words:
Health care workers, Varicella zoster infection, Immunization
Received: Received: August 31, 2009. Accepted: October 27, 2009.
Address reprint request to M. Talebi-Taher, Rasoul-e-akram Hospital, Sattarkhan st., Tehran, Iran (e-mail: mtalebitaher2000@yahoo.com).
INTRODUCTION
The Varicella-Zoster virus (VZV) is transmitted through
a person to person contact, inhalation of aerosols, through
contaminated fluids from vesicular skin lesions caused by 
acute chickenpox, or shingles, or infected respiratory tract
secretions. Varicella is self-limited in children but may
lead to serious complications among adolescents, adults
and immunocompromised patients [1].
Varicella is a common childhood disease. Nearly a 100% se-
ropositivity to VZV has been documented in children
between 11–13 years of age in the USA. An epidemiologi-
cal variation is being reported between the temperate and
tropical climate countries, with varicella occuring mainly
among young adults in the tropical regions [2].
The general adult seroprevalence rates amount to 81.3% in
the United Arab Emirates, 100% in Belgium, 96.6% in Is-
rael, 95% in Turkey, 95.5% in Spain, 50% in Sri Lanka,
and 87.9% in Iran [3–9].
Moreover, nosocomial transmission of VZV is also re-
ported in health care workers (HCWs) [10]. It seems that

O R I G I N A L P A P E R S M. TALEBI-TAHER ET AL.
IJOMEH 2010;23(1)28
March 2009. Serologic testing for varicella was performed
in 405 HCWs with different job categories (physicians,
nurses, nurse’s aides, medical and nursing students, and
administrative workers ) who worked at Rasoul-e-Akram
hospital, Teheran, Iran.
A checklist was completed including demographic data
and the history of varicella. Individuals with acute varicel-
la infection, those under immunosuppressive therapy, or
those having blood transfusion during the year 2008, were
excluded from the study. The local ethics committee ap-
proved the study protocol, and a signed informed consent
was obtained from each participant.
Blood samples were collected from each individual
and the separated serum was stored at –20°C prior to
testing.  EIA  for  varicella-specific  IgG  was  performed 
using  commercial  virus-specific  IgG  EIA  kits  (vari-
cella IgG EIA well, RADIM, Italy; sensitivity 100%,
specificity  88%).  Optical  density  values  were  indexed 
according to the manufacturer’s instructions. Sera were
classified  as  negative  if  the OD was  less  than  0.20  and 
as positive if it was higher than 0.70; sera with OD be-
tween 0.20 and 0.70 were considered equivocal. Equivo-
cal values were considered negative.
Statistical analysis was performed using the SPSS 13 soft-
ware. After conducting descriptive statistical analyses, we
used the chi square test to study the differences in the
proportions of the categorical variables between the study
groups. The P value of less than 0.05 was considered sta-
tistically significant. We compared the history of varicella 
infection in HCWs with the results of testing for antibody
against varicella zoster antigen to assess the reliability
of the history of varicella in this population. Sensitivity,
specificity,  positive  predictive  value  (PPV)  and  negative 
predictive value (NPV) were calculated. In this study, sen-
sitivity refers to the percentage of persons with a history
of varicella among seropositive HCWs,  specificity  to  the 
percentage of persons without a history of varicella among
seronegative HCWs, PPV to the percentage of seroposi-
tives among HCWs with a positive history, and NPV to the
percentage of seronegatives among HCWs with a negative
history of varicella.
age  alone  is  not  sufficient  to  guarantee  immunity.  The 
community at risk must have a clear history of previous
varicella infections or a serologic testing. As a small pro-
portion of the population with a history of the disease
might be susceptible, some experts recommend the se-
rologic testing in all HCWs regardless of the disease his-
tory [11].
In a study performed in two neonatal units in Brazil, 150
of 215 (70%) workers had a history of varicella infec-
tion and all of them had VZ antibodies, which yielded
a 100% sensitivity of the varicella history. Of the remain-
ing 65 work ers who did not remember having had varicella
in the past, 60 (92%) proved to have serologic evidence
of varicella infection, and 5 individuals were considered
non-immune to varicella. The Brazilian authors have con-
cluded that the positive history of varicella infection is
a reliable evidence of immunity against the disease [12].
In a Japanese study, 97.2% of HCWs in a Tokyo hospi-
tal were immune against varicella but the authors sug-
gested that aggressive screening and vaccination of sus-
ceptible HCWs were essential because most of the Japa-
nese HCWs seemed to have little knowledge about their
history of VZV infection [13].
Varicella history was not a reliable predictor of the pres-
ence or absence of varicella antibodies in HCWs in Sau-
di Arabia. Almuneef and colleagues showed that 84%
of HCWs in a hospital were seropositive, and the positive
predictive value (PPV) of the history of chickenpox for
the seropositivity was 89%. The negative predictive value
(NPV) was 22% [14].
As no report on varicella zoster virus immunity in HCWs
in Iran is available, this study has been designed to investi-
gate the seropositivity of VZV among HCWs in a 625-bed
university hospital. In this study, we aimed at determining
the relationship between immunity and the history of va-
ricella infection.
METHODS
This cross-sectional, prevalence study was designed to
investigate the seropositivity of VZV among HCWs
in a 625-bed university hospital between February and

VARICELLA ZOSTER IMMUNITY AMONG HEALTH CARE WORKERS O R I G I N A L P A P E R S
IJOMEH 2010;23(1) 29
DISCUSSION
Nowadays, the nosocomial transmission of varicella is well
recognized. Although all susceptible hospitalized adults
are at risk, certain patients, such as immunocompromised
persons and pregnant women are at increased risk for
a severe complicated infection [9].
Iran is a country in the tropical area. Here, the seropos-
itivity rate against varicella zoster antigen was reported
to be about 35% to 60% in children below 10 years to
nearly 90% in individuals at the age of 30 years [9,15].
In our study, 405 HCWs were tested to determine the
immunity against VZV, and 71.4% of them were found
RESULTS
According to the demographic data and serologic test re-
sults of the HCWs (Table 1), 405 HCWs were tested for anti
varicella IgG. Their age ranged from 19 to 50 years (me-
dian 29 years). Of these workers, 289 (71.4%) were positive
but we did not find any significant correlation of the gender, 
age, or occupation and seropositivity (p = 0.09, 0.75, 0.54,
respectively). There was no significant difference between 
the presence of VZV and age (p = 0.75).
Table 2 presents the results regarding the sensitivity, speci-
ficity, positive predictive value and negative predictive val-
ue of the history of varicella for the seropositivity.
Table 1. Demographic data and serologic results for 405 health care workers studied
Charateristics
Total
n (%)
Serologic results
positive
n (%)
negative
n (%)
p
Gender 0.09
female 235 (58.0) 160 (68.1)1 75 (31.9)
male 170 (42.0) 129 (75.9) 1 41 (24.1)
Age (year) 0.75
≤ 25 136 (33.5) 95 (69.8) 41 (30.2)
26–30 103 (25.4) 74 (71.8) 29 (28.2)
31–35 1 58 (14.3) 40 (68.9) 18 (31.1)
≥ 36 108 (26.8) 80 (74.0) 28 (26.0)
Occupation 0.54
nurse 1 78 (19.2) 55 (70.5) 23 (29.5)
nurse’s aide 106 (26.1) 78 (73.6) 28 (26.4)
physician 1 42 (10.3) 35 (83.3) 1 7 (16.7)
medical student 83 (20.4) 59 (71.0) 24 (28.9)
nursing student 33 (8.1) 22 (66.7) 11 (33.3)
administrative worker 63 (15.5) 41 (65.1) 22 (34.9)
Table 2. Varicella antibodies in two groups with and without a history of varicella, and sensitivity analysis
Hx.1 of varicella
Ab titer
total (n)
positive
n (%)
negative
n (%)
PPV NPV sensitivity specificity
Positive Hx. 212 180 (84.9) 32 (15.1) 84.9 43.5 62.3 72.4
Negative Hx.2 193 109 (56.5) 84 (43.5)
1 History.
2 Including uncertain Hx.

O R I G I N A L P A P E R S M. TALEBI-TAHER ET AL.
IJOMEH 2010;23(1)30
varicella antibody status in HCWs. Indeed, a positive
history of chickenpox better correlated with the pres-
ence of antibody than did a negative serologic test re-
sult for varicella [14,23,25]. In the populations of Brazil
and Turkey, and the child population of Switzerland,
the positive predictive value of the clinical history of
varicella ranged from 95% to 100% [16,31–33]. In the
applicants for nurse training in Scotland, the positive
predictive value of the history of varicella for seroposi-
tivity was 98%. The negative predictive value was 14%.
A positive history of chickenpox showed a sensitivity
of 84% and specificity of 60%. The authors concluded 
that the absence of the chickenpox infection in the pa-
tient’s past medical history is an unreliable predictor
of susceptibility to infection in HCWs [34].
The low PPV in our study might be due to the lack of
data regarding the past history of viral infections like
varicella, and the subclinical diseases. The higher NPV
in our study (43.5%), which is not consistent with oth-
er reports, may reflect a lower general seroprevalence 
in the Iranian HCWs [14,34].
In conclusion, we suggest that, before any decision is
made to start mass vaccination against VZV in Iran,
it is necessary to conduct further large-scale studies
on seroepidemiology of varicella in different popula-
tions (including immunocompromised patients, wom-
en of childbearing age, HCWs, children from different
age groups,  etc.). The findings of our  study,  showing 
a weak correlation between the clinical history and
seropositivity, indicate that the serologic screening of
all HCWs should be recommended. The limitations
of our study include a small population size and the
failure to determine which HCWs come from which
province of Iran.
ACKNOWLEDGEMENTS
The authors thank Dr. Leila Zahedi-Shoolami, Dr. Tahereh
Mousavi, and Ms. Yeganeh Shoaee for their assistance. This
study was supported by the Deputy Research, Iran University of
Medical Sciences, Teheran, Iran (Grant No. 499).
to  be  seropositive.  In  fact,  a  significant  proportion 
of HCWs are susceptible to varicella. This is important
in view of their close contact with pregnant women and
high risk patients whereby the virus can be easily trans-
mitted.
The results of a recent study regarding seropositiv-
ity  differ  from  the  findings  of  other  studies.  In  Tur-
key, 98% of 363 HCWs had antibodies against varicel-
la [16]. The immunity rates for varicella among medi-
cal students in Germany and Switzerland were 96.9%
and 97%, respectively, [17,18], while in our study,
only 71% of medical students were immune. In Israel,
the seroprevalence of VZ among the HCWs in three
studies was 94.8%, 94.4% and 98.5% [19–21]. In Japan
and Brazil, the seropositivity for VZ among HCWs
ranged from 97.2% to 99.1% [13,22,23]. The seroposi-
tivity rate in the population of 1106 HCWs was 91%
in North Italy [24] and 93.1% in Canada [25]. In Saudi
Arabia, the seropositivity amounted to 83%, which
was less than in the other countries studied [26]. Alto-
gether, these data indicate that there is a difference in
the immunity against VZV among HCWs in the coun-
tries of the temperate and tropical climatic zones.
In our study, we found a susceptibility rate of 31.9%
among 235 women of childbearing age (20–40 years).
This rate is higher than in Finland (seronegativi-
ty 3.8%) [27], Ireland (seronegativity 11.3%) [28], Italy
(seronegativity 20%) [29], and lower than in Sri Lanka
(seronegativity 56.2%) [30]. Therefore, the chance of
acquiring and transmitting varicella to, or from, the
patients among our female HCWs is much higher than
in  other  countries.  Thus,  the  findings  of  the  present 
study indicate the importance of the screening test and
vaccination for susceptible HCWs in order to reduce
the rate of varicella in this high risk group.
On the other hand, we compared the history of varicel-
la in HCWs with the results of testing for VZ antibod-
ies to assess the reliability of the history of varicella in
this group. These data are fundamental to analyze the
cost effectiveness of different vaccination strategies in
Iran. Like in other studies, our data showed that the
history of varicella was not a reliable predictor of the

VARICELLA ZOSTER IMMUNITY AMONG HEALTH CARE WORKERS O R I G I N A L P A P E R S
IJOMEH 2010;23(1) 31
11. Tennenberg AM, Brassard JE, Lieu J, Drusin LM. Varicella
vaccination for healthcare workers at a university hospital: an
analysis of costs and benefits. Infect Control Hosp Epidemi-
ol 1997;18:405–11.
12. Santos AM, Ono E, Weckx LY, Coutinho AP, de Moraes-
Pinto MI. Varicella zoster antibodies in healthcare workers
from two neonatal units in Sao Paulo, Brazil — assessment of
a staff varicella policy. J Hosp Infect 2004;56(3):228–31.
13. Hatakeyama S, Moriya K, Itoyama S, Nukui Y, Uchida M,
Shintani Y, et al. Prevalence of measles, rubella, mumps and
varicella antibodies among healthcare workers in Japan. Infect
Control Hosp Epidemiol 2004;25(7):591–4.
14. Almuneef M, Memish ZA, Abbas ME, Balkhy HH. Screening
healthcare workers for varicella-zoster virus: Can we trust the
history? Infect Control Hosp Epidemiol 2004;25(7):595–8.
15. Motamedifar M, Handjani F, Hadi N, Shahkarami MK,
Mehrabani D. Seroprevalence of Varicella-Zoster Virus in chil-
dren from Shiraz, Iran. Iran J Immunol 2006;3:43–6.
16. Celikbas A, Ergonul O, Aksaray S, Tuygun N, Esener H,
Tanir G, et al. Measles, rubella, mumps, and varicella seroprev-
alence among health care workers in Turkey: Is prevaccination
screening cost-effective? Am J Infect Control 2006;34(9):583–7.
17. Wicker S, Rabenau HF, Gottschalk R, Doerr HW, All-
winn R. Seroprevalence of vaccine preventable and blood-
transmissible viral infections (measles, mumps, rubella, polio,
HBV, HCV and HIV) in medical students. Med Microbiol Im-
munol 2007;196(3):145–50.
18. Baer G, Bonhoeffer J, Schaad UB, Heininger U. Seropreva-
lence and immunization history of selected vaccine preventable
diseases in medical students. Vaccine 2005;23(16):2016–20.
19. Chodick G, Ashkenazi S, Livni G, Lerman Y. Increased
susceptibility to varicella-zoster virus among Israeli physi-
cians and nurses born in the Middle-East region. J Occup
Health 2006;48(4):246–52.
20. Lerman Y, Chodick G, Tepper S, Livni G, Ashkenazi S.
Seroepidemiology of varicella-zoster virus antibodies among
health-care workers and day-care centre workers. Epidemiol
Infect 2004;132(6):1135–8.
21. Chazan B, Colodner R, Teitler N, Chen Y, Raz R. Varicella
zoster virus in health care workers in northern Israel: Seroprev-
alence and predictive value of history of varicella infection. Am
J Infect Control 2008;36(6):436–8.
REFERENCES
1. Centers for disease control and prevention. Prevention of
varicella: Recommendation of the Advisory Committee on
Immunization Practices (ACIP). MMWR 2007;56(RR-4):
1–40.
2. Merrett P, Schwartzman K, Rivest P, Greenaway C. Strate-
gies to prevent varicella among newly arrived adult immi-
grants and refugees: A cost-effectiveness analysis. Clin Infect
Dis 2007;44:1040–8.
3. Uduman SA, Tahira AM, AL-Wash R, Usmani MA, Ben-
er A. Varicella susceptibility among children and healthy adults
in the United Arab Emirates. East Mediterr Health J 2001;7
(4–5):604–8.
4. Thiry N, Beutels P, Shkedy Z, Vranckx R, Vandermeulen C,
Wielen MV, et al. The seroepidemiology of primary varicel-
la-zoster virus infection in Flanders (Belgium). Eur J Pe-
diatr 2002;161(11):588–93.
5. Cohen DI, Davidovici BB, Smetana Z, Balicer RD, Kle-
ment E, Mendelson E, et al. Seroepidemiology of varicella
zoster in Israel prior to large-scale use of varicella vaccines. In-
fection 2006;34(4):208–13.
6. Alp H, Altinkaynak S , Ertekin V, Kilicaslan B , Giirak-
sin A. Seroepidemiology of varicella-zoster virus infection in
a cosmopolitan city (Erzarum) in the eastern Turkey. Health
Policy 2005;72(1):119–24.
7. Salleras L, Dominguez A, Plans P, Costa J, Cardenosa N,
Torner N, et al. Seroprevalence of varicella zoster virus infec-
tion in child and adult population of Catalonia (Spain). Med
Microbiol Immunol 2008;197(3):329–33.
8. Kurukulasooriya GM, Thevanesam V, Agampodi SB, Abey-
koon AM, Amarasinghe SP, Goonasekara KP. Susceptibilty
of new entrant university students in Sri Lanka to varicella
zoster infection. Asia Pac J Public Health 2009 May 1 [cited
March 26, 2010]. Available from http://www.ncbi.nlm.nih.
gov/pubmed/19411280.
9.  Sharifi Z, Emadi Ghanjin S. The seroepidemiology of varicella
zoster virus (vzv) in different age groups in Tehran, Iran. Iran
J Allergy Asthma Immunol 2005;4(2):95–8.
10. Weber DJ, Rutala WA, Hamilton H. Prevention and control
of varicella-zoster infections in health care facilities. Infect
Control Hosp Epidemiol 1996;17:694–705.