Serological Survey of Tularemia Among Butchers and Slaughterhouse Workers in Iran

Trans R Soc Trop Med Hyg 2014; 108: 516–518 doi:10.1093/trstmh/tru094 Advance Access publication 18 June 2014

Serological survey of tularemia among butchers and slaughterhouse workers in Iran

Saber Esmaeilia,b, Behzad Esfandiaria,b, Max Maurinc,d, Mohammad Mehdi Gouyae,

SHORT COMMUNICATION Mohammad Reza Shirzadie, Fahimeh Bagheri Amiria,f and Ehsan Mostafavia,b,*

aDepartment of Epidemiology, Pasteur Institute of Iran, Tehran, Iran; bResearch Centre for Emerging and Reemerging Infectious Diseases (Akanlu), Pasteur Institute of Iran, Kabudar-Ahang, Hamadan, Iran; cCNR Francisella, Laboratoire de Bacte´riologie, De´partement des Agents Infectieux, Institut de Biologie et de Pathologie, CHU de Grenoble, Universite´ Joseph Fourier, Grenoble, France; dLAPM, CNRS UMR e f EAFC, Grenoble, France; Center of Disease Control (CDC), Ministry of Health, Tehran, Iran; Faculty of Veterinary Medicine, University of Downloaded from Tehran, Tehran, Iran

*Corresponding author: Present address: Department of Epidemiology, Pasteur Institute of Iran, 69 Pasteur Ave., Tehran 1316943551, Iran. Tel/Fax: +98 21 66496448; E-mail: [email protected]

Received 4 January 2014; revised 29 March 2014; accepted 25 April 2014 http://trstmh.oxfordjournals.org/

Background: Tularemia is a zoonotic disease caused by the Gram-negative bacterium Francisella tularensis. Human infections often occur through manipulation of infected animals or animal carcasses. Methods: In this study, we determined the tularemia seroprevalence in butchers and slaughterhouse workers in 10 counties of Sistan and Baluchestan Province in Iran. Results: A mean seroprevalence of 6.5% for IgG antibodies against F. tularensis was seen. The highest seropositivity rates were observed in the counties of Zabol and Nikhshahr. There was no difference in the seroprevalence

rates between butchers and slaughterhouse workers (p¼0.25). at Queen's University on September 19, 2014 Conclusion: These data suggest that tularemia is endemic in Sistan and Baluchestan Province in Iran.

Keywords: Francisella tularensis, Iran, Seroprevalence, Sistan and Baluchestan Province, Tularemia, Zoonosis

Introduction antibodies were also detected in an Afghan hedgehog in the same year.7 The ﬁrst human case of tularemia in Iran was Francisella tularensis is the etiological agent of tularemia. There reported in the city of Marivan, southwest Kurdistan (western are two subspecies of F. tularensis, namely F. tularensis subspecies Iran) in 1980.8In a recent study (2011–2012) among different tularensis (type A strain) commonly found in North America and groups in western Iran, the rate of tularemia seroprevalence F. tularensis subspecies holarctica (type B strain) found in the was 14.4%.9 In this study, 16% of the butchers were seropositive 1 whole northern hemisphere. Type B strains are responsible for for tularemia. The aim of the study was to determine the sero- almost all tularemia cases in Europe and Asia, and are usually asso- prevalence of F. tularensis in southeastern Iran and to establish ciated with less severe symptoms and lower mortality rates as com- a baseline for surveillance of the disease in the country. Because 2,3 pared to type A strains. F. tularensis has a large animal reservoir, anti-F. tularensis antibodies were previously reported in domestic which includes many terrestrial mammals (especially rodents and animals, sera from butchers and slaughterhouse workers were 2 lagomorphs) and arthropods (especially ticks). Human infections studied. These occupational groups may develop tularemia usually occur through direct contact with infected animals or ani- after contact with blood or tissues from animals infected with mal carcasses. Alternatively, infection may occur via tick (or other F. tularensis or healthy carriers of this bacterium, or via the bite arthropod) bites or after exposure to a contaminated environ- of infected ticks during slaughtering operations.10 ment.3,4 The clinical manifestations of tularemia may vary from an asymptomatic infection to a severe and possibly fatal disease. Six clinical forms are classically recognized: ulceroglandular, glandu- Materials and methods lar, oculoglandular, oropharyngeal, pneumonic and typhoidal.1,5 In Iran, antibodies against F. tularensis were detected in The study area domestic animals (cattle and sheep) in the northwest, and in a This cross-sectional study was carried out in Sistan and porcupine in the southeast (Zabol county) in 1973.6 F. tularensis Baluchistan Province, in southeastern Iran, in 2011. This province

516 Transactions of the Royal Society of Tropical Medicine and Hygiene

is bordered by the Oman Sea to the south, Afghanistan and Statistical analysis Pakistan to the east, the South Khorasan province to the north, Data were analyzed with SPSS statistical software, 16th version and the provinces of Kerman and Hormozgan to the west (SPSSInc,Chicago,IL,USA).Logisticregressionandx2 tests (Supplementary Figure 1). It has a dry, arid to semiarid climate. were used to compare the variables; p-values less than 0.05 were considered statistically signiﬁcant. ArcGIS software version 9.3 (ESRI, Redlands, CA, USA) was used for mapping the results. Sampling In this study, 184 serum samples were collected from 120 butchers and 64 slaughterhouse workers from 10 counties of Sistan and Ethical considerations Baluchistan Province (including Zahak and Zabol in the north, The study was approved by the ethics committee of Pasteur Iranshahr,ZahedanandKhashinthecenterandChabahar, Institute of Iran. Sarbaz, Saravan and Konarak in the south of the province; Table 1). All ofﬁcial slaughterhouses in this province were recruited

and participants were selected randomly from among the Results Downloaded from employees. All participants were male. The median (interquartile In total, 12 samples (6.5%, 95% CI 3.58–10.82%) had positive range) age and length of employment were 34 (25–45) years and IgG titers and 19 samples (10.3%) had borderline IgG titers 8 (3–15) years, respectively. The inclusion criteria were being over against F. tularensis antigen. Only two (6%) of the 31 sera positive 18 and working as a butcher or slaughterhouse worker for a min- or borderline for F. tularensis antibodies also displayed antibodies imum of 6 months. A 10 mL blood sample was collected from against Brucella abortus antigen. The highest F. tularensis sero- http://trstmh.oxfordjournals.org/ each participant after obtaining informed consent and demo- prevalence was observed in Zabol (10.5%) and Nikshahr (10.0%), graphic characteristics. Sera were kept at 2208C and transferred which are northern and southern counties, respectively. However, to the Department of Epidemiology of the Pasteur Institute of the mean seropositivity rates progressively declined from the nor- Iran (Tehran). thern (9.3%) to the central (6.3%) to southern (2.2%) regions of Sistan and Baluchistan Province, although these differences were not statistically signiﬁcant (p¼0.37). We found no statistic- Serological tests ally signiﬁcant difference (p¼0.25) between the F. tularensis sero- Collected sera were tested for the presence of IgG antibodies prevalence observed in butchers (5.0%) and slaughterhouse against F. tularensis, using a commercial ELISA kit (Virion/Serion, workers (9.4%). The F. tularensis seroprevalence rates were not

Wu¨rzburg, Germany) according to the manufacturer’s instructions correlated to age (p¼0.94) and length of employment (p¼0.19) at Queen's University on September 19, 2014 (positive cut-off titer .15 U/mL, borderline cut-off titers 10–15 U/mL). of participants. Serum samples with positive or borderline results for F. tularensis were further tested against Brucella antigen because of the pos- sibility of serological cross-reactions between both antigens. We Discussion used a locally prepared antigen and a standard tube agglutination Our study shows the presence of IgG antibodies against F. tularen- test elaborated by the Pasteur Institute of Iran. sis in 6.5% of the 184 healthy butchers and slaughterhouse workers investigated in Sistan and Baluchestan province, in south-eastern Iran. The ELISA test we used for screening of anti-F. tularensis IgG antibodies is considered to have a specificity Table 1. Seroprevalence of IgG antibodies against Francisella rate higher than 95%, close to the 98.1% specificity reported for 11 tularensis among butchers and slaughterhouse workers in different the western blot technique. False positive results for tularemia counties and regions of Sistan and Baluchestan Province in serology have been mainly observed in brucellosis patients, southeastern Iran owing to serological cross reactions between F. tularensis and Brucella spp.antigens.12 However, in this study, only 6.5% of serum samples positive for anti-F. tularensis antibodies also dis- Region No. tested County No. tested played anti-Brucella sp. antibodies. Thus, it is highly probable (seropositivity %) (seropositivity %) that positive F. tularensis ELISA tests found in butchers and slaughterhouse workers represent true tularemia infections. North 43 (9.3) Zabol 38 (10.5) Because the seropositive participants were asymptomatic, posi- Zahak 5 (0) tive antibody titers probably indicated past tularemia infections. Central 96 (6.3) Zahedan 70 (8.8) It is also possible that they were recent but asymptomatic Iranshahr 14 (0) because not all infections with Francisella result in symptoms. Khash 12 (8.3) Moreover, we did not find an increase in tularemia seroprevalence South 45 (2.2) Chabahar 10 (0) in older patients as reported in other studies,4 which suggests that Sarbaz 5 (0) tularemia has recently emerged or re-emerged in the studied Saravan 10 (0) province. Altogether, our data suggest that tularemia is currently Nikshahr 10 (10.0) endemic in Sistan and Baluchistan Province, which warrants fur- Konarak 10 (0) ther investigations in the general population and in domestic ani- Total 184 (6.5) mal herds to which the two occupational groups studied are exposed.

517 S. Esmaeili et al.

In recent years, there have been no official reports of human Authors’ contributions: MMG and EM conceived the study; MMG, MRS and tularemia cases in Sistan and Baluchestan Province. However, the EM designed the study protocol; SE, BE, MRS and EM carried out the clinical clinical manifestations of tularemia in humans are vary and often assessment; MM, FBA and EM analysed and interpreted the data. SE and unspecific.3 Infections caused by type B F. tularensis,foundin EM drafted the manuscript; all authors critically revised the manuscript Europe and Asia, are usually of mild to moderate severity, with mor- for intellectual content. All authors read and approved the final tality rates lower than 1%.3 The more frequent clinical forms of manuscript. SE, MM and EM are guarantors of the paper. tularemia may be easily confounded with other diseases, such as tuberculosis for chronic lymphadenopathies or Streptococcus Acknowledgments: We would like to express our gratitude to the staff of Zahedan and Zabol University of Medical Sciences for their help in pyogenes infection for pharyngitis. Also, significant antibody titers sampling and Ms Manijeh Yousefi Behzadi, who assisted us in data entry. are usually detected 2 to 3 weeks only after the onset of symptoms, and may persist for months. Thus, serological tests are often nega- Funding: We appreciate the financial support of the Center for Disease tive at the time patients seek medical attention, and a positive test 13 Control and Prevention of the Iranian Ministry of Health. This work was does not necessarily reflect acute and/or recent infection. Thus, approved by the Pasteur Institute of Iran [Grants no, 508 and 582]. tularemia may be an underdiagnosed disease in Iran. Downloaded from In the 1970s, antibodies against F. tularensis were found in wild Competing interests: None declared. animals in Zabol county in the northern region of Sistan and Baluchistan province,6,7 and it was subsequently expected that Ethical approval: The study was approved by the ethics committee of the seroprevalence of tularemia would be higher in this region. Pasteur Institute of Iran. In the present study, the seroprevalence of tularemia in the nor-

thern region was higher than in the central and the southern http://trstmh.oxfordjournals.org/ regions of this province. This suggests that tularemia endemic foci may predominate in the northern part of Sistan and Baluchistan Province, and thus the presence of unrecognized References tularemia cases should be carefully investigated in this area. 1 Hepburn MJ, Simpson AJH. Tularemia: current diagnosis and treatment However, a seroprevalence of 10% was also found in the southern options. Expert Rev Anti-Infe 2008;6:231–40. county of Nikhshahr, suggesting an extended tularemia endemic area in south-eastern Iran. 2 Petersen JM, Schriefer ME. Tularemia: emergence/re-emergence. Vet The risk of tularemia in butchers and slaughterhouse workers Res 2005;36:455–67. could have been better evaluated if we had included a control 3 Sjo¨stedt A. Tularemia: history, epidemiology, pathogen physiology, and group. It is recommended that for similar studies a group from clinical manifestations. Ann NY Acad Sci 2007;1105:1–29. at Queen's University on September 19, 2014 the general population be recruited as a control group. Another 4 Clark DV, Ismailov A, Seyidova E et al. Seroprevalence of tularemia in limitation of this study was not having access to the medical rural Azerbaijan. Vector-Borne Zoonot 2012;12:558–63. records of the study participants to tentatively correlate their 5 Weber IB, Turabelidze G, Patrick S et al. Clinical recognition and past or present clinical manifestations with positivity of tularemia management of tularemia in Missouri: a retrospective chart review serology. of 121 cases. Clin Infect Dis 2012;55:1283–90. 6 Arata A, Chamsa M, Farhang-Azad A et al. First detection of tularaemia Conclusions in domestic and wild mammals in Iran. Bull World Health Organ 1973;49:597–603. Our results emphasize the need to improve the knowledge of 7 Farhang-Azad A, Mescerjakova I, Neronov V. Afghan hedgehog, a new the clinical manifestations and diagnosis of tularemia among reservoir of tularemia. Bull Soc Pathol Exot 1973;66:266–9. health-care workers, to better clarify the real impact of the 8 Karimi Y, Salarkia F, Ghasemi MA. Tularemia: ﬁrst human case in Iran. disease in the studied occupational groups, but also in other J Med Coun Iran 1981;8:134–41. high-risk groups (such as veterinarians and hunters) and in the general population. Further work is also needed to clarify the 9 Esmaeili S, Gooya MM, Shirzadi MR et al. Seroepidemiological survey of tularemia among different groups in western Iran. Int J Infect Dis current reservoirs and modes of transmission of F. tularensis.In 2014; 18:27–31. the meantime, because our data indicate that butchers and slaughterhouse workers may be exposed to F. tularensis 10 Coleman EA. Tularemia: a practical guide to management in the event infection during their work, it is recommended that these of a tick bite or terrorist attack. AJN 2002;102:65–6. employees be tested for tularemia at the time of recruitment 11 Schmitt P, Splettstosser W, Porsch-Ozcurumez M et al. A novel and then on a regular basis during their active period. They screening ELISA and a conﬁrmatory Western blot useful for should also use personal protective equipment to prevent such diagnosis and epidemiological studies of tularemia. Epidemiol Infect contaminations. 2005;133:759–66. 12 Ta¨ rnvik A, Chu MC. New approaches to diagnosis and therapy of tularemia. Ann NY Acad Sci 2007;1105:378–404. Supplementary data 13 Jenzora A, Jansen A, Ranisch H et al. Seroprevalence study of Supplementary data are available at Transactions Online Francisella tularensis among hunters in Germany. Fems Immunol (http://trstmh.oxfordjournals.org/). Med Mic 2008;53:183–9.

518