Risk factors for human brucellosis in Iran:a case—control study

Masomeh Sofian a, Arezoo Aghakhani b, Ali Akbar Velayati c,Mohammad Banifazl d, Ali Eslamifar b, Amitis Ramezani b,*

International Journal of Infectious Diseases (2008) 12, 157—161

http://intl.elsevierhealth.com/journals/ijid

aArak Medical University, Arak, IranbDepartment of Clinical Research, Pasteur Institute of Iran, 13164, Pasteur Square, Tehran, IrancMasih Daneshvari Hospital, Tehran, Irand Iranian Society for the Support of Patients with Infectious Diseases, Tehran, Iran

Received 13 November 2006; received in revised form 13 March 2007; accepted 18 April 2007Corresponding Editor: Ziad Memish, Riyadh, Saudi Arabia

KEYWORDSHuman brucellosis;Risk factors;Iran

Summary

Background: Brucellosis is a zoonotic disease of worldwide distribution. Despite its control inmany countries, it remains endemic in Iran. The aim of this study was to determine the risk factorsfor brucellosis acquisition in the central province of Iran.Methods: A matched case—control study was conducted in the central part of Iran. A total of 300subjects (150 cases and 150 controls) were enrolled in the investigation. Brucellosis cases weredefined on the basis of epidemiologic, clinical, and laboratory criteria. Subjects were interviewedusing a questionnaire to obtain risk factor information. We used odds ratios and conditionallogistic regression models to explore the association between the disease and the variablesstudied.Result: Significant risk factors for infection were related to the existence of another case ofbrucellosis in the home (OR = 7.55, p = 0.0001) and consumption of unpasteurized dairy products(OR = 3.7, p = 0.014). Keeping cattle and cattle vaccination were also important risk factors.Conclusions: Pasteurization of dairy products and education regarding fresh cheese must bepursued for eradication of brucellosis. Amajor risk factor for acquiring brucellosis is the existenceof another infected family member. Therefore screening family members of an index case ofbrucellosis may lead to the detection of additional cases.# 2007 International Society for Infectious Diseases. Published by Elsevier Ltd. All rightsreserved.

* Corresponding author. Tel.: +98 21 88361922;fax: +98 21 20109550.

E-mail address: iiccom@iiccom.com (A. Ramezani).

1201-9712/$32.00 # 2007 International Society for Infectious Diseases.doi:10.1016/j.ijid.2007.04.019

Introduction

Brucellosis is an infectious zoonotic disease that is associatedwith chronic debilitating infections in humans and reproduc-tive failure in domestic animals.1 It is still an important publichealth problem throughout the world, but principally, and in

Published by Elsevier Ltd. All rights reserved.

158 M. Sofian et al.

particular, in the Mediterranean region, including Iran, Tur-key, the Arabian Peninsula, the Indian subcontinent, Mexico,and parts of Central and South America.2 It causes significanteconomic loss among those keeping domesticated animals asa source of meat and dairy products.3

Brucellosis can be transmitted to humans through contactwith animals or their products; it is an occupational hazard topersons engaged in certain professions (e.g., veterinarians,slaughterhouse workers, and farmers).4—6 Several variablesare significantly associated with the acquisition of humanbrucellosis. Both direct and indirect transmission such asconsumption of unpasteurized milk and dairy products, herd-ing, and lambing have been shown to have significant corre-lation with this disease.7

Some authors have reported exposure to aborted home-owned animals and consumption of home-made milk pro-ducts as risk factors for human brucellosis infections,4 andothers have found a statistically significant correlationbetween seropositivity for brucellosis and age, sex, andthe consumption of fresh cheese and cream made fromunboiled milk.8

Iran is an endemic area for brucellosis. Hasanjani Roushanet al. studied epidemiological features and clinical manifes-tations in 469 adult patients with brucellosis in the northernpart of Iran. They found that the consumption of fresh cheese(22.4%), animal husbandry (11.3%), working in a laboratory(8.1%), and veterinary profession (1.5%) were the main riskfactors for brucellosis infection.9

Identifying the major risk factors for brucellosis isvery important for reaching a comprehensive understand-ing of the nature of the disease and its transmission routesfor eradication of human brucellosis. The aim of thisstudy was to explore various risk factors associated withbrucellosis in the Iranian population using a case—controlstudy.

Figure 1 Map of Iran showing Arak

Patients and methods

A case—control study was undertaken to determine riskfactors for brucellosis in Iran. Data were collected duringthe year 2005. Cases consisted of patients with brucellosisattending 125 health centers from rural and urban regions ofArak City (central part of Iran, Figure 1). Their documentswere studied and then case finding was started. Clear defini-tions of case and control were established. A case wasdefined as a resident of the district of 1 year or more,who had received a clinical diagnosis of brucellosis on thebasis of the symptoms, compatible clinical findings, stan-dard tube agglutination (STA = Wright) test dilution �1:160,and in the presence of 2-mercaptoethanol (2ME) agglutina-tion�20. A control was defined as a neighbor of the case whohad been resident in the district for 1 year ormore during thestudy period and who had no history of brucellosis. Thecontrols also underwent STA and 2ME testing and did notmeet the clinical and laboratory criteria mentioned above.An infected family member was defined as a householdmember with brucellosis diagnosed by both interview andlaboratory tests.

One hundred and fifty cases (67 female and 83 male)were matched with respect to age, sex, and place ofresidence (rural or urban), with 150 controls. A physicianexamined and interviewed all the cases. All 300 subjectsanswered a questionnaire to establish risk factors for bru-cellosis. Informed consent was obtained from all studyparticipants. Complications were defined as follows: per-ipheral arthritis was diagnosed by the finding of swelling,effusion, and limitation of motion in an involved joint;epididymo-orchitis was diagnosed by the finding of swellingand tenderness of scrotal skin, testis, and epididymis withconfirmation by sonography; meningitis was diagnosed bythe presence of headache, neck stiffness, and fever and

city (central province of Iran).

Risk factors for human brucellosis in Iran 159

Table 1 Characteristics of cases and controls

Characteristics Cases (N = 150) % (n) Controls (N = 150) % (n) p Value

Sex NSMale 55.3% (83) 54% (81)Female 44.7% (67) 46% (69)

Education NSIlliterate 29.3% (44) 28% (42)No formal 8% (12) 8% (12)Elementary 37.3% (56) 34% (51)Secondary 21.3% (32) 19.3% (29)Diploma 4% (6) 8.7% (13)College 0% (0) 0.7% (1)

Type of animals owned NSCattle 16.9% (352) 7.3% (335)Sheep 78.1% (1629) 84.3% (3884)Ewe 0.2% (4) 5.3% (244)Goat 4.8% (101) 3.2% (147)

Occupation NSFarmers 33.3% (50) 21.3% (32)Butchers 0.7% (1) 0.7% (1)Animal husbandry 5.3% (8) 4% (6)Worker 58% (87) 70.7% (106)Farmers and animal husbandry 2.7% (4) 3.3% (5)

Positive family history of brucellosis 61.3% (92) 17.3% (26) <0.05Number of infected members in household 295 44 <0.05Rate of relapse in family 16% (24) 2% (3) <0.05Keeping cattle 88.7% (133) 58% (87) <0.05Cattle vaccination 76.7% (102/133) 86.2% (75/87) <0.05

NS, not significant.

confirmed if analysis of cerebrospinal fluid (CSF) showed anaseptic meningitis plus Brucella serological positivity bySTA and 2ME in CSF.

Statistical analysis

The data among study groups were compared with the Chi-square test using SPSS 11.5 program. We calculated the oddsratios and the confidence intervals for the studied variables.A conditional logistic regression model was used to explorethe association between disease and risk factors. A p value of�0.05 was defined as statistically significant.

Table 2 Risk factors for human brucellosis

Risk factors

Consumption of unpasteurized dairy productsConsumption of unpasteurized and pasteurized dairy productsSelling cattleKnowledge of the mode of brucellosis transmissionKnowledge about fresh cheese as a mode of brucellosis transmissiKnowledge about unboiled milk as a mode of brucellosis transmissKnowledge about pasteurized milk for protection against brucelloKnowledge about cattle vaccinationInfected household members

OR, odds ratio; CI, confidence interval.

Results

A total of 150 cases with a mean age 33.37 � 21.3 years(range 2—86 years) enrolled in our study. Of these, 55.3% (83)were male and 44.7% (67) were female; 85.3% (128) werehabitant in rural areas and 87.3% (131) of them kept farmanimals. Characteristics of case and control subjects areshown in Table 1.

There was no significant difference in age, sex, maritalstatus, area of residence (rural and urban), educationallevel, consumption of uncooked meat, knowledge aboutprevention routes of brucellosis, and occupation between

Adjusted OR 95% CI p Value

3.7 1.64—8.3 0.0142.9 6.4—14 0.0071.9 1.14—3.14 0.00141.14 0.41—3.15 0.8

on 0.44 0.23—0.85 0.01ion 0.48 0.2—1.18 0.11sis 0.94 0.49—1.83 0.87

1.63 0.87—3.03 0.137.55 3.91—14.61 0.0001

160 M. Sofian et al.

Table 3 The most common complaints and complications ofpatients with brucellosis

Percentage

Clinical complaintsFever 78.7%Myalgia 66%Chills 64%Anorexia 52%Headache 50.7%Malaise 46%Low back pain 44.7%Fatigue 43%Sweating 42.7%Weight loss 34%Arthralgia 34%Paresthesia 18.7%Dyspnea 18%Palpitations 16.7%Nausea 15.3%Rash 12%Dysuria 11.3%

Clinical complicationsArthritis 14.6%Meningitis 0.6%

case and control subjects. We could find no correlationbetween acquisition of brucellosis and abortion or infertilityin home-owned animals.

Significant risk factors for infection were related to theexistence of another case of brucellosis in the home(OR = 7.55, 95% CI = 3.91—14.61, p = 0.0001) and consump-tion of unpasteurized dairy products (OR = 3.7, 95%CI = 1.64—8.3, p = 0.014) (Table 2).

Knowledge about the mode of brucellosis transmission byfresh cheese appeared to be protective against diseasetransmission (OR = 0.44, 95% CI = 0.23—0.85, p = 0.01).

Significant differences were found between case andcontrol subjects with regard to positive family history ofbrucellosis (61.3% vs. 17.3%, p < 0.05), number of infectedmembers in the household (295 persons vs. 44 persons,p < 0.05), rate of relapse in the family (16% vs. 2%,p < 0.05), keeping cattle (88.7% vs. 58%, p < 0.05), andcattle vaccination (76.7% vs. 86.2%, p < 0.05).

When specific animal species were considered, the great-est risk was associated with products derived from sheep asopposed to goats and cattle.

The most common symptoms recorded were fever(78.7%), myalgia (66%), and chills (64%). Of our patients,17.8% showed disease complications including: arthritis(14.6%), meningitis (0.6%), and epididymo-orchitis (2.6%).Complaints and complications of patients with brucellosis areshown in Table 3.

Epididymo-orchitis 2.6%

Discussion

Brucellosis occurs naturally in domestic animals and is trans-mitted to human beings by direct and indirect routes such asconsumption of unpasteurized milk and dairy products, closecontact with infected livestock, their tissues or secretions,herding, lambing, and others.7,10 Several studies have beenperformed in the Kingdom of Saudi Arabia (KSA),11—13 Jor-dan,14 Gaza Strip,15 Lebanon,16 Israel,17 and in Yemen18 todetermine risk factors for brucellosis.

Some authors have emphasized that direct contact withinfected animals and their products is more important thaningestion of contaminated animal products for brucellosistransmission.19—21 Exposure to aborted home-owned animalshas been identified as a probable source of human brucellosisinfections in some studies.4 Alballa et al. reported that directcontact with domestic animals and consumption of rawproducts of animal origin were the main risk factors.10

In a case—control study in Saudi Arabia the greatest riskfactor was indirect contact with animals (the consumption ofunpasteurized dairy products), as opposed to direct contactwith animals. When direct contact with animals was consid-ered, the study found a very high-risk association with assist-ing in animal parturition but no significant risk associatedwith other direct (unspecified) animal contact.11

Other scholars believe that public health programs shouldfocus on educating the population about the risks of con-suming dairy products made from unpasteurized milk, such assoft cheeses.22

Brucellosis is endemic in certain parts of Iran. The pre-valence of brucellosis in Iran has been reported as 0.5% to10.9% in different provinces. In a study in Tehran province theincidence of brucellosis was 17.5/100 000 and the main

source of infection was fresh cheese and unboiled milk.Another cross-sectional study in Mazandaran provinceshowed that the main risk factors for brucellosis acquisitionwere age, sex, and consumption of milk and cheese (datapublished in Iranian journals).

The present investigation determined risk factors forhuman brucellosis in the central part of Iran using a case—control study. According to our survey, the significant riskfactors for infection are related to the existence of anothercase of brucellosis in the household and the consumption ofunpasteurized dairy products. There was no statisticallysignificant correlation between acquisition of brucellosisand abortion or infertility in home-owned animals, but keep-ing cattle and cattle vaccination were significant risk factorsbetween case and control subjects. Our results on the con-sumption of unpasteurized dairy products are in concordancewith the studies published by Cooper,11 Husseini et al.,7

Cetinkaya et al.,8 Kozukeev et al.,4 and Issa and Jamal,14

but are in contrast to the report of Serra Alvarez and GodoyGarcia.23 Our data regarding aborted home-owned animalsdid not concord with the study of Kozukeev et al.4

Occupation and educational levels are also risk factors forbrucellosis acquisition.4 A hospital-based case—control studyin Yemen showed significant risk factors for infection relatedto occupation and drinking fresh milk and laban. Socio-economic and educational factors were also independentrisk factors.24 Araj and Azzam from Lebanon concluded thatexposure to brucellosis is high among persons in high-riskoccupations.16 Our data showed that occupation and educa-tional levels were not important risk factors and were incontrast with the above findings. One possible explanationcould be due to our case and control selection. Case and

Risk factors for human brucellosis in Iran 161

control groups were neighbors and presumably had similaroccupations and educational levels, especially in rural areas.Therefore we should not expect to find a significant differ-ence between study groups.

A recent study from KSA25 showed that 63.5% of 137detected cases in Tabuk district came from rural areas, whichis comparable to the result of a study in Palestine7 that showed61.1% of the cases being village inhabitants. Our studydetected higher cases (85.3%) from rural areas than theabove-mentioned studies, but we found similarities in unpas-teurizedmilk consumption and animal keeping and ownership.

Familiar antecedents and household members have asignificant relation with the seropositivity for brucello-sis.7,22,26,27 We observed significant differences betweencase and control subjects with regard to a positive familyhistory of brucellosis, number of infected members in thehousehold, and rate of relapse in the family. Therefore,active serological screening of the other household membersof brucellosis cases may detect additional unrecognizedcases.

In conclusion the findings of this study suggest that humanbrucellosis is still an important public health problem in Iran.However, most of the risk factors for brucellosis acquisitionare modifiable. Pasteurization of milk and dairy products andeducation regarding eating habits must be pursued for era-dication of human brucellosis, especially in rural areas.Knowledge about mode of brucellosis transmission by freshcheese appeared to be protective against disease transmis-sion.

Since brucellosis is a zoonotic disease and the vast major-ity of cases are infected through direct or indirect contactwith animals, a campaign to control the infection by animalvaccination would be an important step. In this study wefound that the greatest risk factor for acquiring brucellosis isthe existence of another infected family member, and itshould be emphasized that the exposure of family membersto the same epidemiological factors leads to more than onemember of the family being infected. Thus the basic principalshould be to screen family members once a case is diagnosed.

Acknowledgments

The authors acknowledge Arak Medical Sciences Universityfor financial support of this study.

Conflict of interest: No conflict of interest to declare.

References

1. Corbell JM. Brucellosis: an overview. Emerg Infect Dis 1997;2:213—21.

2. Geyik MF, Gur A, Nas K, Cevik R, Sarac J, Dikici B, et al.Musculoskeletal involvement in brucellosis in different agegroups: a study of 195 cases. Swiss Med Wkly 2002;132:98—105.

3. Young EJ. Brucella species. In: Mandell GL, Bennett JE, Dolin R,editors. Mandell, Douglas and Bennett’s principles and practiceof infectious diseases. 4th Ed. New York: Churchill Livingstone;1995. p. 2053—7.

4. Kozukeev TB, Ajeilat S, Maes E, Favorov M, Centers for DiseaseControl, Prevention (CDC). Risk factors for brucellosis–—Leylek andKadamjay districts, Batken Oblast, Kyrgyzstan, January—Novem-ber, 2003. MMWR Morb Mortal Wkly Rep 2006;55(Suppl 1):31—4.

5. Mendez Martınez C, Paez Jimenez A, Cortes-Blanco M, SalmoralChamizo E, Mohedano Mohedano E, Plata C, et al. Brucellosisoutbreak due to unpasteurized raw goat cheese in Andalucia(Spain), January—March 2002. Euro Surveill 2003;8:164—8.

6. Chomel BB, DeBess EE, Mangiamele DM, Reilly KF, Farver TB, SunRK, et al. Changing trends in the epidemiology of human bru-cellosis in California from 1973 to 1992: a shift toward foodbornetransmission. J Infect Dis 1994;170:1216—23.

7. Husseini AS, Ramlawi AM. Brucellosis in theWest Bank, Palestine.Saudi Med J 2004;25:1640—3.

8. Cetinkaya Z, Aktepe OC, Ciftci IH, Demirel R. Seroprevalence ofhuman brucellosis in a rural area of Western Anatolia, Turkey. JHealth Popul Nutr 2005;23:137—41.

9. Hasanjani Roushan MR, Mohrez M, Smailnejad Gangi SM, Solei-mani Amiri J, Hajiahmadi M. Epidemiological features and clin-ical manifestations in 469 adult patients with brucellosis inBabol, Northern Iran. Epidemiol Infect 2004;132:1109—14.

10. Alballa SR. Epidemiology of human brucellosis in southern SaudiArabia. J Trop Med Hyg 1995;98:185—9.

11. Cooper CW. Risk factors in transmission of brucellosis fromanimals to humans in Saudi Arabia. Trans R Soc Trop Med Hyg1992;86:206—9.

12. Kiel F, Khan M. Brucellosis among hospital employees in SaudiArabia. Infect Control Hosp Epidemiol 1993;14:268—72.

13. Opawoye A. Brucellosis in children of Faifa-Gizan, Southwest ofSaudi Arab: a review of 32 cases. Ann Saudi Med 1994;14:5.

14. Issa H, Jamal M. Brucellosis in children in South Jordan. EastMediterr Health J 1999;5:895—902.

15. Awad R. Human brucellosis in the Gaza Strip, Palestine. EastMediterr Health J 1998;4:225—33.

16. Araj G, Azzam R. Seroprevalence of Brucella antibodies amongpersons in high-risk occupation in Lebanon. Epidemiol Infect1996;117:281—8.

17. Mishal J, Ben-Israel N, Levin Y, Sherf S, Jafari J, Embon E, et al.Brucellosis outbreak: analysis of risk factors and serologicscreening. Int J Mol Med 1999;4:655—8.

18. Al-Shamaly H. Seropositivity for brucellosis in a sample of ani-mals in the Republic of Yemen. East Mediterr Health J 1999;5:1035—41.

19. Abo-Shehada MN, Odeh JS, Abu-Essud M, Abuharfeil N. Seropre-valence of brucellosis among high-risk people in northern Jor-dan. Int J Epidemiol 1996;25:450—4.

20. De Massis F, Di Girolamo A, Petrini A, Pizzigallo E, Giovannini A.Correlation between animal and human brucellosis in Italy duringthe period 1997—2002. Clin Microbiol Infect 2005;11:632—6.

21. Perez-Rendon Gonzalez J, Almenara Barrios J, Rodriguez MartinA. [The epidemiological characteristics of brucellosis in theprimary health care district of Sierra de Cadiz] [Article inSpanish]. Aten Primaria 1997;19:290—5.

22. Karabay O, Serin E, Tamer A, Gokdogan F, Alpteker H, Ozcan A,et al. Hepatitis B carriage and Brucella seroprevalence in urbanand rural areas of Bolu province of Turkey: a prospective epi-demiologic study. Turk J Gastroenterol 2004;15:11—3.

23. Serra Alvarez J, Godoy Garcia P. [Incidence, etiology and epi-demiology of brucellosis in a rural area of the province of Lleida][Article in Spanish]. Rev Esp Salud Publica 2000;74:45—53.

24. Al-Shamahy HA, Wright SG. A study of 235 cases of humanbrucellosis in Sana’a, Republic of Yemen. East Mediterr HealthJ 2001;7:238—46.

25. Elbeltagy K. An epidemiological profile of brucellosis in TabukProvince, Saudi Arabia. East Mediterr Health J 2001;7:790—8.

26. Almuneef MA, Memish ZA, Balkhy HH, Alotaibi B, Algoda S, AbbasM, et al. Importance of screening household members of acutebrucellosis cases in endemic areas. Epidemiol Infect 2004;132:533—40.

27. Memish Z. Brucellosis control in Saudi Arabia. Prospects andchallenges. J Chemother 2001;13(Suppl 1):11—7.