Introduction

Brucellosis is recognized as an important human infection in many parts of the world especially Latin America, Southern Europe, Africa and Asia including Middle East [1]. More than 500,000 human infections are recorded every year throughout the world. This figure may be low from the actual incidence of brucellosis cases, especially in endemic areas, due to underreporting in some countries [2]. It is primarily a zoonotic disease found both in domestic and wild animals [1]. Human transmission mainly occurs through direct contact with infected animals or their products such as placenta or aborted materials, and consumption of infected animal products or undercooked meat. Also, direct contact with soil, animal feces and dust contaminated with the organism is associated with a higher risk of infection [3]. Human to human transmission is rare, but has been reported in association with blood transfusion, bone marrow transplantation, transplacental or perinatal exposure, during sexual intercourse and postnatally, through breast milk. Previously it was thought that children were uncommonly affected by this disease but quite a few reports from endemic areas changed the notion by showing a higher percentage of childhood brucellosis (10–30%) [4,5,6,7]. In children, the major source of infection is consumption of unpasteurized milk products such as soft cheese, ice cream and butter. Also, the consumption of raw milk and contact with infected animals or their excretory products can be the causes of acquiring this disease. Clinical manifestations of childhood brucellosis are varied, ranging from minimal symptoms to extreme morbidity and occasional fatality. Symptoms of childhood brucellosis vary partly because of the variable pathogenicity of different strains. Asymptomatic infections were also recorded. This hospital based study has been designed to portray a picture about the possible presence of childhood brucellosis cases in the Eastern part of India.

Material and Methods

Blood samples were collected by venipuncture from PUO (pyrexia of unknown origin) patients aged ≤18 y. In this study, PUO refers to a condition in which the patient has an elevated temperature for about 3 wk but despite investigations by a physician no explanation has been found. If a definite cause is found, it is not included under PUO. Blood was withdrawn by a single venipuncture (usually 3–5 ml but 10 ml from older children) and injected into a vacutainer tube (BD). Serum was separated by centrifugation at 2000 rpm for 15 to 20 min. The separated serum samples were then stored at −20 °C freezer before processing for further tests. The inclusion criteria of patients in this study were only children and adolescents categorized as PUO cases without any serious complications; who were not treated with any antimicrobial agent effective against brucellosis; the exclusion criteria were, patients with serious complications, or who were already treated for brucellosis. The samples were collected from different hospitals: Peerless Hospital & B.K. Roy Research Centre, a private tertiary care hospital in Kolkata, India and MGM Medical College, Kishanganj, Bihar, India. Permission for this study was taken from Institutional Ethical Committee of these hospitals. Consent for this study was taken from parents of all the children included in this study.

The duration of the study was from January, 2013 through December, 2015. A total of 236 blood samples were collected. Details of clinical findings were recorded in a typical clinical data sheet and a patient history sheet.

All the serum samples were screened by brucella specific serological tests - Standard Tube Agglutination Test (STAT; Tulip Diagnostics Pvt. Ltd., India), Rose Bengal Plate Test (RBPT; IAHVB, Bangalore, India), and ELISA (for both IgM and IgG antibodies; Immunolab GmbH, Germany).

The blood clots of serologically positive blood samples were liquefied and cultured according to Castaneda method using Castaneda’s biphasic medium (Himedia, Mumbai, India). Blood was also collected directly as described earlier, in automated blood culture bottles (BACTEC 9050, BD) and was incubated up to 3 wk for isolation of the bacterium. Subcultures in case of suspected growths were done on brucella agar plates (BD BBL ™). The bacterial growth on agar plate was tested for identification by biochemical tests - catalase using H2O2 (Merck, USA), oxidase using oxidase disc (Himedia), urease using urea agar (Himedia) and H2S production using TSI agar slant (Himedia).

The DNA was extracted from all the serologically positive serum samples using the spin column technique following the instructions provided in the QIAmp DNA Blood Mini Kit (Qiagen, Germany).

The DNA samples were tested by the brucella genus specific simplex polymerase chain reaction (PCR) using the forward and reverse bcsp gene primers B4 and B5 respectively. This primer pair was obtained from Sigma Aldrich Chemicals Pvt. Ltd., Bangalore, India. This end-point PCR was carried out in a thermal cycler (Prime, UK). The amplified product was then subjected to agarose gel electrophoresis within a gel electrophoresis trough (Peqlab, Germany). Finally the generated amplicon size of 223 bp was viewed using the gel doc apparatus (Biorad, USA). This bcsp gene encodes the 31 kDa brucella cell surface salt extractable protein (BCSP) and is considered as a marker gene present in all Brucella spp.

Quality control was done as a part of DBT (Department of Biotechnology) network project on brucellosis (Government of India) in eight different units throughout India, which includes authors’ Unit BE-8. A thorough training of all the investigators and fellows, quality check of all the recommended kits, and random inter lab check of the samples were done to achieve a uniform high standard in all the units.

Results

During the period of study, altogether 236 blood samples of child PUO cases were screened for the detection of brucella infection. Out of 236 blood samples, 29 (12.3%) were denoted as brucella positive samples based on any one of the serological (STAT, RBPT, IgM ELISA, IgG ELISA) and molecular (PCR) tests. The total number of child PUO cases screened year wise during the study period and also the positive brucella cases were noted. Based on serological and molecular screening it was noted that brucella positive cases were high during 2013 and again in 2015 but somewhat low incidence was observed in 2014. The total number of child PUO cases screened each month during the study tenure and the month wise positive brucella cases were also noted. Based on this analysis it was found that the maximum brucella positive patients were identified during the months, June to September, which is the monsoon season of the year.

Results of serological tests (Table 1) revealed that 10.6% serum samples were STAT positive, 7.2% were RBPT positive, 7.2% were IgM ELISA positive and 0.85% was IgG ELISA positive. None of the cultures showed brucella positivity. PCR test was positive in 1.3% cases (Table 1). Figure 1 indicates the three brucella positive cases obtained by PCR during the course of the study.

Table 1 Results of serological tests and molecular test among brucella positive cases
Full size table
Fig. 1
figure 1

Gel picture represents brucella positive cases during 2013–2015. Lane 1: 100 bp molecular size marker, Lane 2: Positive control for bcsp gene (Brucella abortus; S99), Lane 3: Negative control, Lanes 5, 7, 9: Samples positive for bcsp gene. The amplicon size for bcsp gene is 223 bp

Full size image

It was expected that Brucella positive results may be more in male children as their number was more (58.5%) in this study, but it was found that Brucella positive results were more in female child patients (14.3%) (Table 2). Chi –squared test was done in between positive results of male and females. The calculated P values (significance level) for differences of positive results in between male and females in relation to STAT, RBPT, IgM ELISA, IgG ELISA and PCR tests were 0.9419, 0.3488, 0.5566, 0.9600, and 0.0628 respectively. The results indicate that although there was practically no difference in STAT, RBPT, IgM ELISA, IgG ELISA test positive results, there was some significance of increased positive PCR tests in females; which also indicates there is a possibility of increased brucellosis in female child.

Table 2 Gender distribution among PUO (aged ≤18 y) and brucella positive cases
Full size table

It was found that during the study period majority (88.6%) of the child PUO cases came from Kolkata and its adjoining districts and 11.4% were from adjoining states and Bangladesh.

Another interesting observation was that during the tenure of this study three female child patients were identified as brucella positive based on all the serological and molecular findings obtained from the laboratory except IgG ELISA. Among these three child patients, one resides in Kolkata city and the other two are in one of its adjoining district. It was also recorded that only three child patients below 5 y of age were found affected by this disease in this study. All of them were 2-y-old; one was male and the other two were females. They showed brucella positivity in all serological tests (STAT, RBPT and IgM ELISA) except IgG ELISA test.

Discussion

Brucellosis has an important public health implication and cause of economic loss in many countries, particularly in the countries of Mediterranean region and other developing countries [8]. Worldwide, reported incidence of human brucellosis in endemic disease areas varies widely, from <0.01 to >200 per 100,000 population [9]. Clinical picture of this disease is variable as it may manifest as a systemic disease. After infection, the microorganism, which exhibits multisystem involvement, proliferates in the regional lymph nodes, passes into blood stream and usually affects several tissues and multiple organs, primarily the reticuloendothelial system [10]. It is considered as an important cause of pyrexia of unknown origin (PUO) and has shown considerable mortality in endemic countries. The incidence of childhood brucellosis is usually underreported and is found to be ranging between 0.8 and 1.6% [11]. However, the rate of childhood brucellosis in endemic regions has been reported to vary from 11% [12] to 56% [13]. Brucellosis, as an infectious disease in children has been accounted rarely in India [14]. It is very uncommon under 5 y of age. The present study was conducted with child and adolescent patients aged ≤18 y and in this study three child patients aged less than 5 y were found affected by this disease. The pediatric patients included in this study presented symptoms like persistent fever, joint pain, low backache, fatigue and night sweat but fever was the most common complaint which is in accordance with previous reports [15, 16]. These symptoms were also mainly observed in the Brucella positive cases. The period of illness prior to testing and diagnosis was less than 1 mo in most of the cases, which shows consistency with previous studies [11].

STAT with titres greater than 1:160 suggests active infection. STAT measures the total amount of agglutinating antibodies i.e., both IgM and IgG. Thus, STAT is known to be the best diagnostic modality available. RBPT is a rapid serological screening test. It is suitable for large scale screening. In this study, several serological parameters were checked such as STAT, RBPT and ELISA. Results indicated that the majority of the positive child patients showed STAT positivity followed by RBPT and IgM ELISA.

In brucellosis endemic region, the interpretation of Brucella serological tests may be difficult due to persistent presence of Brucella specific antibodies in general population. Thus over diagnosis and unnecessary treatment against brucellosis will continue until and unless each country establishes a normal range of the titre of serological tests for the population of that country. Although rapid and low cost STAT and RBPT tests are preferred tests for brucellosis as per WHO guidelines, each positive result should be correlated with clinical symptoms and history of exposure to animals or animal products if present. This is due to the fact that the predictive value of a positive test declines as prevalence of the disease decreases in a country.

Again, serological tests vary in their ability to give a true positive result. False reaction may occur due to group cross reaction with other bacteria including Yersinia enterocolitica 0:9 (common major O-polysaccharide antigen), E. coli O116: H21, O157: H7, Francisella tularensis, Salmonella serotypes of Kauffman-White group N, Pseudomonas maltophilia, Vibrio cholerae and by non specific inhibitors mimicking antibodies. Specificity, sensitivity and predictive values of all serological tests should be considered during interpretation of the results. Usually specificity and sensitivity are inversely related.

The culture of brucella is difficult and usually takes a lot of time. Moreover, the culture yield remains very low. The bacteremia condition prevails mostly in the first 2 wk of illness, when PUO diagnosis is not done. Thus as per WHO guidelines, serological tests are normally preferred for diagnosis of brucellosis in the laboratory. In this study authors could not retrieve the organism from any blood culture samples probably due to these facts. In brucellosis endemic regions, clinical symptoms associated with positive serology but without isolation of the organism are considered confirmed human cases [17].

The molecular tests like PCR are being employed for the diagnosis of brucellosis both at the genus and species levels. The technique has advantages because it is easy to perform, require a short span of time and is non-hazardous as it does not carry the risk of laboratory acquired infection. So there is a probability that this diagnostic test will be the future diagnostic modality of choice [18].

In this study ELISA (both IgM and IgG) positive results were considered true positive results, for their better sensitivity and specificity [19, 20], recent WHO guidelines also suggested that ELISA can be considered as a confirmatory test for brucellosis [21] and when compared to individual test, sensitivity was found as 82.6%, 73.1%, 90.5%, 52.8%, 54.3% with STAT, RBPT, IgM ELISA, IgG ELISA, and PCR respectively. Specificity was minimum with STAT which was 50.0%. Specificity of other tests was 100.0%. Positive predictive values of STAT, RBPT - the two commonly used tests, were 65.5% and 73.1% respectively, while negative predictive values of these two tests were 71.4% and 41.7%, respectively. Thus STAT may be used for screening Brucella infection, then the positive results may be confirmed by ELISA test results and for final diagnosis, a positive result should be correlated with clinical findings thoroughly before beginning the treatment.

During the tenure of the study, it was noticed that the incidence of positive cases of child brucellosis was significant in 2013 and 2015, with monsoon being the most favorable season for acquiring this infectious disease. This is because in this season Brucella infection is more in animals from where the disease is transmitted to human beings. Similarly, the disease was found slightly more in female children particularly in relation to PCR tests. Although it is very difficult to explain why female children are more affected in brucellosis than male children, there are several studies done previously in rats, where it was found that testosterone in male rats gives resistance to Brucella infection [22], while oestradiol in female rats decreases resistance to Brucella infection [23]. Thus similar factors may be the reason for the differences in brucella positivity among both the sexes in this study. To prevent the disease in children, drinking of unpasteurized milk, eating cheese or ice cream should be avoided and children should not come in contact with sick or dead animal. Education of general public is needed as an adjunct and not as an alternative strategy to control human brucellosis [24] as till now no vaccine is available for the prevention of brucellosis in humans.

Conclusions

It can be summarized that the present study was designed to evaluate the presence of childhood brucellosis presenting as PUO cases in Eastern zone of India. It is visualized that childhood brucellosis is prevalent in India and in PUO cases, brucellosis should also be looked for even in children below 5 y of age.