Screening Household Members of Acute Brucellosis Cases in Endemic Areas and Risk Factors for Brucellosis

Abstract

Early diagnosis and treatment of acute brucellosis cases were targeted by screening the household members of the index cases. We also aimed to describe the causal relations of brucellosis in an endemic region. A cross-sectional study was performed among household members (29 index cases, 113 household members). Brucellosis was diagnosed on the basis of clinical findings, serum agglutinin titer of ≥1/160 in standard tube agglutination test (STA), or a positive blood culture. Index cases were defined as patients who had been admitted to the clinic on suspicion of brucellosis and then confirmed as brucellosis cases. The people who lived in the same house as the index cases were defined as household members. The risk factors for seropositivity were studied by multivariate analysis. Independent variables of gender, consuming fresh cheese, blood groups, dealing with husbandry, and contact with the placenta of infected animals were included to the model. Backward and forward selections were performed. Nineteen out of 113 (17%) screened individuals had agglutination titers ≥1/160. The mean ages of index cases and household members were 43 years (standard deviation [SD] 18) and 29 years (SD 19), respectively. In multivariate analysis, consuming fresh cheese (odds ratio [OR]=3.1, confidence interval [CI] 1.07–9.68, p=0.049), blood group A (OR=2.6, CI 1.18–5.96, p=0.018), contact with the placenta of the infected animals (OR=3.7, CI 1.42–9.68, p=0.007), and age >30 years (OR=2.8, CI 1.25–6.51, p=0.13) were found to be associated with brucellosis. In univariate analysis, the individuals with blood group B were protected from brucella infection (p=0.013). In conclusion, screening of the people in brucellosis-endemic areas should be considered for early diagnosis and treatment. To our knowledge, blood groups were studied for the first time by this study. Higher prevalence of brucellosis among the individuals with blood group A and less prevalence among the individuals with blood group B should be considered for further studies on pathogenesis.

Introduction

Brucellosis is a disease affecting various domestic animals, including cattle, goats, swine, and dogs. Human infections derive directly or indirectly from exposure to animals or animal products (Moreno 2014). Human and animal brucellosis are responsible for enormous economic losses and considerable human morbidity in endemic areas (Sumer et al. 2003, Ducrotoy et al. 2014). Brucellosis is especially prevalent in the Mediterranean basin, including Turkey. The Turkish Ministry of Health reported 7177 cases and a morbidity rate of 10/100,000 in 2011 (Aydin 2012).

Although this disease exists worldwide, person-to-person transmission of brucellosis is extremely rare (Gotuzzo et al. 2004, Joung et al. 2010). A few reports have shown an increased incidence of brucellosis among household members of an infected person due mainly to exposure to the same source of infection. Musculoskeletal, genital, cardiac, respiratory, and nervous system involvement may be observed in brucellosis because of various clinical polymorphisms (Lubani et al. 1989, Alsubaie et al. 2005). This study aimed to diagnose and treat acute brucellosis by screening the household members of brucellosis cases. We also describe the causal relations of acute brucellosis in an endemic region.

Materials and Methods

Study population and definitions

A cross-sectional study was performed in the Infectious Diseases and Clinical Microbiology Clinic at Numune Education and Research Hospital in Ankara, which is an 1100-bed referral and tertiary-care community hospital in Turkey.

The patients who had been admitted to the clinic on suspicion of the brucellosis and were then confirmed as brucellosis were defined as the index cases. The diagnosis of acute brucellosis was on the basis of clinically compatible signs and symptoms (that usually occurried within 8 weeks) confirmed by laboratory methods (serological tests or isolation of Brucella sp. in blood culture) according to the Centers for Disease Control and Prevention (CDC) definitions (Alsubaie et al. 2005). Serologically, cases were accepted as brucellosis in which either the Brucella serum agglutination titer (SAT) was ≥1/160 in the standard tube agglutination test (STA) or there was four-fold or greater rise in Brucella SAT between acute- and convalescent-phase serum specimens obtained in a 2-week or longer period. Blood cultures were obtained from the index cases that were hospitalized and from the household members who had fever with >38°C body temperature. The household members that were living in the same house with the index cases were screened. A structured survey was administered in which demographic data, history, physical examinations, and laboratory results were recorded on individual forms. The sera of the household members were obtained, and Brucella agglutination tests were performed.

The family members who had a Brucella antibody titer of ≥1/160 and symptoms compatible with brucellosis (fever, night sweats, undue fatigue, anorexia, weight loss, headache, and arthralgia) were defined as symptomatic seropositive members, whereas individuals without any clinical symptoms suggestive of brucellosis were defined as asymptomatic seropositive members.

Microbiological studies

Serum samples were examined by serial tube dilution with STA and the Coombs test for incomplete Brucella antibodies. A serum agglutinin titer of ≥1/160 in the Coombs test was considered positive, similar to STA. Samples of blood were cultured using an automated system (Organon Teknika BacT/Alert Biomerieux, France).

Statistical analysis

Data were analyzed by Stata Statistical software, version 11.0 (Stata Corporation, TX). A multivariate model analysis was performed to detect the predictors of Brucella SAT of >1/160. Blood groups (A, B, and O), contact with placenta, consumption of fresh cheese, and age were included in the model for logistic regression. Forward and backward selections were performed, and no discrepancy was noted. The statistical significance was set as p<0.05

Results

After admission of index cases to the clinic, household members of these cases were screened. A total of 142 subjects were included in the study. Among these, the total number of index cases was 29 (20%). A total of 113 household members who were related to index cases were contacted, interviewed, and then tested by using SAT. Nineteen out of 113 (17%) household members were found to be seropositive (Fig. 1). Sixty-five (46%) of the total 142 cases were male and 77 (54%) were female. Additionally, there was a similar distribution among sexes in index cases and household members (Table 1). There was no significant difference in sex ratios between seropositive and seronegative groups; in all 48 subjects of seropositive group, 23 (48%) cases were females and 25 (52%) cases were males (p=0.281).

FIG. 1.

Study population distribution according to tube agglutination, blood culture, and symptoms. STA, standard tube agglutination test.

Table 1.

Demographic Characteristics of the Subjects

	Index cases n=29 (%)	Screened household members n=113 (%)	p
Female gender	14 (48)	63 (56)	0.471
Mean age (years)	43 (16–79)	29 (1–77)	<0.0005
Wright agglutination ≥160	29 (100)	19 (17)	<0.001

The mean age of cases was 32 (1–79) years. Children <12 months of age were accounted as 1 year old. The mean ages of index cases and household members were 43 years (standard deviation [SD] 18) and 29 years (SD 19), respectively. As shown in Table 1, there was a statistically significant difference between mean ages of index cases and household members. This might be due to the fact that our department accepts only adult patients. However, there were children in household members.

SAT of all cases were between negative to 1/5120. Nevertheless, SAT combinations with Coombs agglutination were a maximum of 1/10,240. Although 10 of 29 index cases (34%) had high-degree SAT (1/1280–1/5120), 10 of 19 seropositive household members (52%) had high-degree SAT.

Symptoms of brucellosis are variable. For that reason, the cases with any symptoms that may be compatible with brucellosis (such as fever, sweats, anorexia, weight loss, bone and joint symptoms [including arthralgia, back pain, and joint swelling], neuropsychiatric symptoms [headache, depression, fatigue, and changes in mental status], gastrointestinal, or dermatologic symptoms) were assessed symptomatic cases. The relation between any symptom and seropositivity was evaluated. Forty-three (58%) of the all symptomatic cases (n=74) were seropositive, 31 (42%) cases were seronegative, and five (7%) of asymptomatic cases (n=68) were seropositive (p<0.001). Nevertheless, among 48 seropositive cases, Brucella SAT of symptomatic seropositive cases were not significantly higher than titers of asymptomatic seropositive cases (p=0.550). The mean values of SAT were 1/1240 and 1/1699, respectively.

Blood cultures were obtained in 31 cases, and Brucella bacteria were isolated from 18 cases (58%). The positivity rate for Brucella bacteria among index cases and household members was 64% and 44%, respectively. All of these culture-positive 18 cases were also seropositive, but there was no statistical correlation between blood culture positivity and SAT degree (p=0.622). (The mean values of SAT of culture positive and negative cases were 1/1573 and 1/1261, respectively).

In addition, as shown in Table 2, on evaluation of 120 cases that were older than 10 years old, 34 (72%) of the total 47 seropositive cases dealt with husbandry and 43 (91%) cases consumed fresh cheese. Animal husbandry and consuming fresh cheese were statistically significant predisposing factors for Brucella infection (p=0.001 and p=0.001). Moreover, when the occupational risk factors were evaluated, coming into contact with the placenta of infected animals, cleaning the barn, milking, and having a history of abortion among animals were found to be associated with brucellosis (Table 2).

Table 2.

The Risks for Brucella Infection

	SAT≥160 n=47 (%)	SAT<160 n=73 (%)	p
Female gender	23 (49)	47 (64)	0.094
Mean age (years)	39	35	0.16
Husbandry	34 (72)	30 (41)	0.001
Consuming fresh cheese	43 (91)	48 (66)	0.001
Occupational risks (n=120)
Contact with placenta	18 (38)	11 (15)	0.004
Abortion among the animals	31 (66)	27 (37)	0.002
Cleaning the barn	34 (72)	28 (38)	<0.001
Milking	24 (51)	22 (30)	0.021
Blood groups
A	23 (49)	18 (26)	0.01
O	10 (21)	24 (34)	0.129
B	4 (9)	19 (27)	0.013

The analysis was performed among the subjects who are older than 10 years of age.

SAT, serum agglutination titers.

On the other hand, the different blood groups were evaluated with univariate analysis, and whereas blood group A was associated with brucellosis (p=0.01), the individuals with blood group B were protected from Brucella infection (p=0.013) (Table 2).

In multivariate analysis, consuming fresh cheese (odds ratio [OR]=3.1, confidence interval [CI] 1.07–9.68, p=0.049), blood group A (OR=2.6, CI 1.18–5.96, p=0.018), contact with the placenta of the infected animals (OR=3.7, CI 1.42–9.68, p=0.007), and age>30 (OR=2.8, 1.25–6.51, p=0.13) were found to be associated with brucellosis (Table 3).

Table 3.

The Multivariate Analysis for Brucella Serum Agglutination Titers≥1/160

	Odds ratio	Confidence interval (95%)	p
Blood group A	2.6	1.18–5.96	0.018
Contact with placenta of the animal	3.7	1.42–9.68	0.007
Age (years) >30	2.8	1.25–6.51	0.013
Consuming fresh cheese	3.1	1.07–9.68	0.049

Discussion

Turkey is a Mediterranean country in which brucellosis is endemic, especially in the mid-Anatolia region (Sumer et al. 2003). In Turkey, seropositivity of brucellosis among farmers, rural areas, and dairy farms was found to be 6.2%, 7.2%, and 5.7%, respectively. Each year many people are infected with brucellosis, which is an important public health problem (Cetinkaya et al. 2005, Tabak et al. 2008). Therefore, the population at risk should be identified, and screening programs should be established to detect unrecognized cases, especially in families living in rural areas. Through this study of prospective screening, we could identify other cases among all family members of index cases, making these cases available for early treatment.

In this study, it was found that the seroprevalence of brucellosis in household members of acute brucellosis cases was 17% (19/113). Similar to our study, 18% seroprevalence was demonstrated in a study organized in a different region of Turkey (Tabak et al. 2008). The prevalence of brucellosis among household members were reported as 19% in Saudi Arabia (Alsubaie et al. 2005), 50% in Peru (Gotuzzo et al. 1989), and 9% in Israel (Abramson et al. 1991).

In this study, although the majority of all cases recorded were female, the ratio of female in the seropositive group was slightly lower than males (23/48 [48%], 25/48 [52%] [p=0.281]), which is similar to other Mediterranean and regional countries like Greece and Iran (Hasanjani Roushan et al. 2004, Avdikou et al. 2005), which are also endemic countries. We detected a higher rate of seropositivity in middle and older age groups than in younger age groups. A study from Israel reported that 77% of the seropositive persons were younger than 17 years old (Abramson et al. 1991). This might be due to the fact that the frequency of contact with animals varies according to different groups daily routines.

It was reported that symptomatic family members tended to have higher Brucella antibody titers compared to asymptomatic group (Gotuzzo et al 1989, Alsubaie et al. 2005). However, we did not detect higher serum Brucella SAT values among symptomatic seropositive cases than asymptomatic ones.

Symptomatic cases with Brucella infection may have multiorgan involvement, most often in the musculoskeletal system. Consistent with various studies, fever, night sweats, malaise, and arthralgia were the most frequently observed symptoms in our study (Tasbakan et al. 2003, Hasanjani Roushan et al. 2004, Aminzadeh et al. 2010).

In our study, animal husbandry was the statistically significant predisposing factor for brucellosis, which was compatible with other studies reported from Yemen, Eritrea, and Spain (Castel Monsalve et al. 1996, Al-Shamahy et al. 2000, Omer et al. 2002). Exposure to infected animals poses a considerable risk for human brucellosis infections. Farmers live in close proximity to their livestock that may be kept in the lower floor of their home, and they usually contact with them without protective measures, such as the use of gloves and a mask. In our study, the infected patients tended to participate more in calf deliveries and had contact with cow's blood and placenta, compared with the healthy subjects. In addition, daily activities like cleaning the barn and milking were detected as being statistically significant predisposing factors for brucellosis.

There are various studies that have shown the relationship between human blood groups and different infectious diseases (Berger et al. 1989, Garratty et al. 2005). Blood group antigens may interact with many infectious agents or affect the host immune response indirectly (Rios et al. 2000). In our study, blood groups were also evaluated as risk factors, and we detected a higher rate of Brucella seropositivity among indivuals with blood group A, whereas there was a lower rate among the individuals with blood group B. These observations should be supported by molecular studies.

The limitation of the study was as follows. Our study was performed among the household members of the index patients; however, screening in a large population would provide a better description of the similar risk factors.

Conclusions

Brucellosis is still a public health problem in Turkey. Health education programs should be prepared in cooperation with veterinarians. In addition, people should be educated about the habit of consumption of pasteurized milk and milk products instead of unpasteurized products and using preventive measures during contact with animals. Screening and treatment of symptomatic and asymptomatic family members of people from these areas provide a significant decrease in morbidity and mortality rates of brucellosis. A higher prevalence of brucellosis among the individuals with blood group A and lower prevalence among the individuals with blood group B should be considered for further studies regarding the pathogenesis of brucellosis.

Footnotes

Acknowledgments

The authors are grateful to the colleagues and the clinics personnel in the Department.

Author Disclosure Statement

No competing financial interests exist.

References

Abramson

, Rosenvasser

ZVI

, Block

, Dagan

. Detection and treatment of brucellosis by screening a population at risk. Pediatr Infect Dis J, 1991; 10:434–438.

Al-Shamahy

, Whitty

CJM

, Wright

. Risk factors for human brucellosis in Yemen: A case control study. Epidemiol Infect, 2000; 125:309–313.

Alsubaie

, Almuneef

, Alshaalan

, Balkhy

, et al. Acute brucellosis in Saudi families: Relationship between Brucella serology and clinical symptoms. Int J Infect Dis, 2005; 9:218–224.

Aminzadeh

, Farrokhi

, Zanjani

, Baiati

, et al. Screening of household members and contacts of patients with acute brucellosis to detect unrecognized cases. J Infect Dis Immun, 2010; 2:42–43.

Avdikou

, Maipa

, Alamanos

. Epidemiology of human brucellosis in a defined area of northwestern Greece. Epidemiol Infect, 2005; 133:905–910.

Aydin

. Epidemiology of Brucella infections in Turkey. In: Fourth Symposium on Zoonotic Diseases in Turkey, Ankara, 2012:57–63.

Berger

, Young

, Edberg

. Relationship between infectious diseases and human blood type. Eur J Clin Microbiol Infect Dis, 1989; 8:681–689.

Castel Monsalve

, Rullan

, Peiro Callizo

, Nieto-Sandoval Alcolea

. Epidemic outbreak of 81 cases of brucellosis following the consumption of fresh cheese without pasteurization. Rev Esp Salud Publica, 1996; 70:303–311.

Cetinkaya

, Aktepe

, Ciftci

, Demirel

. Seroprevalence of human brucellosis in a rural area of Western Anatolia, Turkey. J Health Popul Nutr, 2005; 23:137–141.

10.

Ducrotoy

, Bertu

, Ocholi

, Gusi

, et al. Brucellosis as an emerging threat in developing economies: Lessons from Nigeria. PLoS Negl Trop Dis, 2014; 8:e3008.

11.

Garratty

. Relationship of blood groups to disease: Do blood group antigens have a biological role?. Rev Med Inst Mex Seguro Soc, 2005; 43:113–121.

12.

Gotuzzo

, Carrillo

, Seas

, Guerra

, et al. Epidemiological and clinical features of brucellosis in 39 family groups. Enferm Infecc Microbiol Clin., 1989; 7:519–524.

13.

Gotuzzo

, Carrilla

. Brucella In: Gorbach

, Bartlett

, Blacklow

, eds. Infectious Diseases, 3 ^th ed. Philadelphia: Lippincott Williams Wilkins, 2004:1717–1724.

14.

Hasanjani Roushan

, Mohrez

, Smailnejad Gangi

, Soleimani Amiri

, et al. Epidemiological features and clinical manifestations in 469 adult patients with brucellosis in Babol, Northern Iran. Epidemiol Infect., 2004; 132:1109–1114.

15.

Joung

. Brucella Species. In: Mandell

, Bennett

, Mandell

, eds. Douglas and Bennett's Principles and Practice of Infectious Diseases, 7 ^th ed. Philadelphia: Churchill Livingstone, 2010:2921–2925.

16.

Lubani

, Dudin

, Araj

, Manandhar

, et al. Neurobrucellosis in children. Pediatr Infect Dis J, 1989; 8:79–82.

17.

Moreno

. Retrospective and prospective perspectives on zoonotic brucellosis. Front Microbiol, 2014; 5:213.

18.

Omer

, Assefaw

, Skjerve

, Tekleghiorghis

, et al. Prevalence of antibodies to Brucella spp. and risk factors related to high-risk occupational groups in Eritrea. Epidemiol Infect, 2002; 129:85–91.

19.

Rios

, Bianco

. The role of blood group antigens in infectious diseases. Semin Hematol, 2000; 37:177–185.

20.

Sumer

, Sumer

, Alim

, Nur

, et al. Seroprevalence of Brucella in an elderly population in Mid-Anatolia, Turkey. J Health Popul Nutr, 2003; 21:158–161.

21.

Tabak

, Hakko

, Mete

, Ozaras

, et al. Is family screening necessary in brucellosis?. Infection, 2008; 36:575–577.

22.

Tasbakan

, Yamazhan

, Gokengin

, Arda

, et al. Brucellosis: A retrospective evaluation. Trop Doct, 2003; 33:151–153.