Seroprevalence and Molecular Diagnosis of Brucellosis Among Patients at Selected Hospitals of Abbottabad, Pakistan

Background: Brucellosis is a neglected disease of ruminants with zoonotic potential. It causes severe health problems in humans and economic loss. Only a limited number of studies have been conducted in Pakistan to determine the prevalence of human brucellosis and related risk factors. The objectives of the current cross-sectional study were to determine prevalence of Brucella infection in sera collected from patients at three hospitals of Abbottabad. Risk factors were investigated. Methods: A total of 500 blood samples were collected. A questionnaire was lled in for each patient to obtain information on age, gender, living area, brucellosis associated symptoms, associated risk factors, pregnancy and abortion history. Serum agglutination test (SAT) was performed to detect antibodies against brucellae. A genus specic BCSP-31 gene RT-PCR was used to detect Brucella DNA in the sample. Statistical analysis was done to determine odd ratios, risk ratios, 95% condence intervals and p values. Results: A total of 13.6% (n=68) patients were found to be SAT positive. DNA was found in 11.4% (n=57) samples. The prevalence of brucellosis was reported to be higher in women (14.6%, n=44) than in man (12.1%, n=24). The age group of 25-50 years was found to be at higher risk for brucellosis (14.5%, n=50). “Animal contact” was reported as the main risk factor followed by “consumption of raw animal products”. About 9.9% pregnant women (n=13) were found brucellosis positive. Of these 23.8% (n=5) had an abortion history. Conclusions: The present study reports a striking prevalence of brucellosis among patients including at three hospitals Abbottabad. These ndings must foster strategies for controlling in an dairy The present study reported the prevalence of brucellosis among patients including pregnant women from hospitals of Abbottabad, Pakistan. The study showed that the population of Abbottabad is at higher risk of acquiring brucellosis because most people although living in urban areas have close contact with animals and consume raw products of animal’s e.g. unpasteurized milk. The results of this study can be used to develop strategies for controlling human brucellosis in rural settings of Pakistan, to raise awareness about brucellosis in livestock professionals, consumers and physicians and to develop control programs by the authority in charge.


Background
Brucellosis is a disease caused by bacteria of species of the genus Brucella with a high zoonotic potential [1]. In developing countries, the disease is of great importance for public and veterinary health [2] by affecting both, human and animal health [3]. Endemic areas include the Mediterranean region, the Middle East, the Arab peninsula, Africa, Latin America and Asia [4]. Four species of Brucella (B. abortus, B. melitensis, B. suis, B. canis) are known to cause disease also in humans regularly. Other Brucella species i.e. B. inopinata, B. cetaceae and B. microti cause disease in animals but rarely in humans [5]. The number of new Brucella infections in humans exceeds 500,000 cases per year worldwide [4].
Bacteria of the genus Brucella are Gram negative coccobacilli without capsule and slow growing organisms [6]. They are aerobic and some are capnophilic. Brucella are facultative intracellular pathogens and have the ability to survive and propagate inside epithelial cells, macrophages, dendritic cells and placental trophoblasts [7]. Their main virulence factor and antigen which is important for survival in the host is liopolysaccharide (LPS) which is involved in the prevention of apoptosis of infected cells [8]. Cytoplasmic antigens, structural proteins of the outer membrane such as OMPs and periplasmic proteins are antigens identi ed by the immune system of the host [9].
Brucellosis is a zoonotic disease. It is transmitted from animals to humans through ingestion of contaminated raw meat and unpasteurized dairy products like soft cheese and raw milk usually. It is also an occupational hazard for people with close contact to infected animals. They get infected via skin abrasion and cuts or inoculation. Inhalation of contaminated aerosols during bacterial culture has caused disease in laboratory staff. Brucellae can actively penetrate conjunctiva of the eyes. While transmission by person to person is rare [10]. Thus, brucellosis in humans mostly coincides with endemic infection of livestock [4].
Some Brucella species are highly infectious and are able to cause human brucellosis with an infection dose of less than 10 organisms [11]. The incubation period of brucellosis varies and the average incubation period is 2 to 3 weeks in man [12]. Brucellosis in humans is commonly known as "Malta fever" or "Undulant fever" and has been found to be one of the main causes of fever of unknown origin in endemic areas [13]. In humans, brucellosis causes undulant fever (39-40 °C), sweating, headache, fatigue, malaise, anorexia, weakness, arthralgia, back and abdominal pain. Symptoms may be present for some weeks or may get chronic or latent for months to years. In case of absence of proper treatment, severe complications may be seen [14]. The infection can affect any system or organ due to dissemination of bacteria via the blood [11]. Osteoarticular, hematologic, gastrointestinal, genitourinary, respiratory, cardiovascular and neurologic disorders can be provoked. Osteoarticular disorders are most common. The highest mortality is caused by endocarditis [9]. During the rst trimester, spontaneous abortion and in utero death of the fetus in pregnant women have been documented [15]. In animals, the disease causes orchitis, infertility, abortion and reduced production of milk [3].
Due to the non-speci c symptoms of the disease, brucellosis is misdiagnosed with other diseases such as typhoid fever, tuberculosis, malaria, leishmaniasis, malignancy and rheumatic fever [9]. Final diagnosis, therefore, must rely on laboratory techniques including bacterial isolation by culture, phenotypic or genotypic con rmation, and testing for Brucella antibodies by serological methods. Various serological methods are available but there is a lack of a perfect method. The speci cities of serological tests are low with sensitivities ranging from 65-95% [16]. Nowadays, molecular techniques are used to detect DNA circulating in the blood stream to amend serology. These techniques have high speci cities and are rapid [17].
Treatment of human brucellosis needs combination of antibiotics. Doxycycline with gentamycin or rifampicin is most commonly used for treatment [18]. Currently, doxycycline in combination with streptomycin is considered the best choice of therapy mainly for localized brucellosis and acute cases having less relapses and side effects [19].
There is no effective vaccine which has been approved to prevent brucellosis in humans. Therefore, control of human brucellosis depends on its prevention in animals to reduce exposure [20]. In endemic areas, food hygiene, i.e. pasteurizing dairy products is an important measure of safety. The infection risk is also reduced through introduction of other biosafety and biosecurity measures such as personal hygiene, adopting safe practices of working or environment protection [21]. Water sources can be prevented from contamination through burying and proper handling of abortion material [22]. Vaccination in animals can reduce nancial losses but is not an appropriate measure for eradication. In nal steps of eradication, test and slaughter policy has to be applied [23].
According to OIE (O ce for International des Épizooties, Paris), brucellosis is a neglected zoonotic disease which has a negative impact on human health and production of animals [19]. In Pakistan, multiple zoonotic diseases (i.e. toxoplasmosis, brucellosis etc.) are prevalent in the human population [24,25]. Because brucellosis has a considerable impact on animal production and human health, the socioeconomic life situation of people of rural regions who mainly depend on cultivation of land and livestock rearing is substantially impaired [26]. Few studies have been conducted in Khyber Pakhtunkhwa province of Pakistan to investigate the prevalence of and risk factors for Brucella infection in patients of hospitals of Abbottabad.

Study area
The study was carried out at Ayub Medical Hospital, Jinnah Medical Hospital and DHQ of Abbottabad city. Abbottabad city is the capital of the district Abbottabad located in the Hazara region, province Khyber Pakhtunkhwa. It has a total area of about 1,967 km 2 or 759 square miles. According to the census of 2017, the total population is 1,332,912 and the density is 680 inhabitants per km 2 [27]. The total number of households of the district reporting livestock such as cattle, buffaloes, sheep, goats and camels etc. is 1,263,547 and the total number of reported animals of the KPK province is 5,967,886 according to data of 2006 [28]. Most of the rural population has to do livestock farming as there is little land available for agriculture in this district. Thus, a higher risk of acquiring brucellosis due to close contact to livestock can be supposed [29].
Collection of patients detail data by questionnaire A questionnaire was lled in personally for each patient. Questions on age, gender, dwelling area, animal ownership or presence of animals in household, contact with animals, processing or handling raw animal products or meat, consumption of raw animal products, access of livestock to the household's source of drinking water, abortion in animals or contact with aborted animals, presence or previous history of symptoms such as fever, night sweats, head ache, arthralgia, generalized ache, nausea, anorexia and fatigue and presence of such symptoms or brucellosis in any other house-hold member had to be answered. Women were asked to report on previous pregnancies and abortion history.

Blood collection
Page 6/18 500 blood samples from persons who visited the Outdoor Patient departments (OPD) of the hospitals and agreed to take part in this study were collected from April, 2019 to August, 2019. About 4 ml blood was collected aseptically from the brachial vein with disposable and sterile syringes. Blood was immediately injected and transferred into serum separating gel-tubes and tubes were labeled immediately. The serum was obtained by centrifugation at 6,000 rpm for ve minutes. Each serum sample was divided into two parts for serum agglutination testing and DNA extraction to perform RT-PCR, respectively.

Serology
The Brucella abortus antigen of the Febrile Antigen Kit (Plasmatec, Lab21 Healthcare Ltd, Bridport, Dorset, United Kingdom) was used for serum agglutination slide test as per manufacturer instructions. Brie y, 80 µl, 40 µl, 20 µl, 10 µl and 5 µl of undiluted serum was added onto a row of 3 cm diameter circles of a reaction slide. Then a drop of the undiluted suspension of antigen was added to each serum sample by using the dropper provided with the kit. The content was mixed using a stirring stick. The slide was shaken gently for one minute and then observed for any agglutination. A test was positive when agglutination was observed at 1:160. Genus speci c primers and probes targeting the BCSP-31 gene were used according to [30]. The BCSP-31 gene codes for a 31KDa immunogenic protein of the membrane and is conserved among all Brucella species and biovars. The sequences of primers and uorescent tagged probe are given in the Table 1.

Results
Out of 500 samples, 68 samples were found to be SAT positive. While 57 samples were con rmed positive by RT-PCR.
The associations of demographic factors with seropositivity for Brucella antibodies are given in Table 2.
The study showed that the prevalence of brucellosis was higher in the age group 25-50 years (n = 50). The prevalence of brucellosis was 12.1% (n = 24) in males and 14.6% (n = 44) in females but this nding was not signi cant (p = 0.493). The prevalence of disease was reported to be 31.6% (n = 49) in participants of rural areas and 5.5% (n = 19) of urban area which was signi cant (p = < 0.0001).    Table 4.

Discussion
Brucellosis is a zoonotic disease of worldwide distribution. It negatively impacts human health, and animal production and economy by signi cant loss [12]. Brucellosis may become chronic causing severe osteoarticular, cardiovascular, neurological and genitourinary complications including epididymo-orchitis and abortion in pregnant women if left untreated [31,32]. Main reasons for human disease are animal brucellosis in bovines and small ruminants, several risky behaviors such as consumption of unpasteurized and contaminated dairy products and not wearing protective clothing when handling potentially infectious animals and their products [33].
In slaughterhouse workers, villagers and farm workers of rural Pakistan, who are considered to be at special risk to get infected with brucellosis, a seroprevalence of 14% was found [34]. These authors found a prevalence of 6.9% in a group at risk from the Potohar Plateau in 2013 [26]. In comparable settings in India [35] and in Bangladesh [36] prevalence's were reported to be 7.32% and 4.4%, respectively. It is also well known that a signi cant number of brucellosis cases can be found in patients with fever of unknown origin in endemic countries. Involving 7567 patients with suspect, yearly prevalence's from 10.4 to 15.7% were found at a Saudi hospital between 2014 and 2018 (37). At Ayub Teaching Hospital, Abbottabad 70 patients from different districts presenting with nonspeci c symptoms (fever, body aches, myalgias, arthralgia, headache, backache, malaise and insomnia) were recently screened with SAT (cutoff titer > 80) for Brucella antibodies and amazingly 49 were found positive (38). In contrast to that extraordinary high prevalence these authors documented a SAT seroprevalence of 13.6% (n = 68) in persons seeking advice at three different outdoor hospitals from Abbottabad. This prevalence shows the high burden of disease in the rural population although it could not be gured out how many admissions were brucellosis related indeed. The strongly diverging results of the two preliminary studies from Abbottabad show the need for future research to provide sustainable data for public health use. It has to be stressed that data from different studies cannot be compared without caution as various not standardized or harmonized tests are still used and those tests are not used on a routine basis at the labs either. Hence, these data show that physicians at hospitals in endemic areas should be aware of brucellosis in their day to day work.
The prevalence of brucellosis was highest in the age group 25-50 years. This nding can be explained by the fact that participants of this middle-aged group were mainly veterinarians, butchers and milking personnel who were in close contact to animals. However, seropositive cases of brucellosis were reported in participants of all age groups. As already stressed different studies cannot not be compared due to the variety of techniques used. So, the highest prevalence was found in the age group of 20-30 years (26.92%) while low prevalence was recorded in older age > 40 years (7.80%) for study participants from the Punjab, Pakistan [39 Ali 2018]. An Indian study found persons of age group 26-35 years affected most (10.8%) [40]. A study conducted in Southern Saudi Arabia among febrile patients determined the highest seroprevalence in patients 21 and 40 years of age (35.8-45.3%), while low prevalence was recorded in young children and older people (3 and 15%), respectively [37]. More positive cases were found in the age group of 41-80 years in Bangladesh which contrasts with our study [36]. Hence, an interpretation of the data points to the fact, that in rural populations the presence of antibodies comes along with contact to brucellae or their LPS at work and so it is very likely to nd more positives aged from 15 to 55 years. Neither the fact that anti-Brucella antibodies are detected in study participants nor that patients at hospitals are involved allow the nal statement that an active infection caused those due to the shortcomings of brucellosis serology. The interpretation of these data has to be done in the light of the epidemiological context. Trends, however, are obvious and can be used to guide countermeasures.
Brucella infection was found more often in female than in male patients. Similarly, a higher prevalence of brucellosis in female patients (37%) was also recorded from Peshawar [41]. The explanation is that animal husbandry is done mainly by women in Pakistan. Therefore, they are in direct contact to animals during their daily activities and also help during parturition without using precautionary measures. In contrast, all seropositive patients were reported to be man in a study conducted among persons of Ludhiana, India where only few women were involved in activities that exposed them to animals and other potential risk-factors [40]. Similarly, a study conducted in high-risk group persons from Bangladesh also found a higher prevalence in man (5.6%) than in women (0.8%). The main reason for this nding was, that mainly butchers, milkers, livestock farmers and veterinary practitioners were tested. These occupations are traditionally in the hands of men there and expose them to a high risk of infection [36]. This is also true for Egyptian setting where more man is involved in management of livestock [42].
This study found a higher prevalence in persons from the rural area demonstrating a higher risk of acquiring brucellosis than for people of the urban area. This nding is similar to that of a study conducted in Peshawar among hospital patients [41]. Persons from rural areas are often involved in birthing and herding of livestock and they are more dependent on livestock production putting them at the risk of infection [39]. A cross sectional study was conducted on rural population of the Punjab in India and seropositivity was also linked to a history assisting with abortions and calving [43].
Nicoletti stated that each case of brucellosis in humans is related to an animal source and its presence in animals causes a major risk of Brucella infection for humans [44]. Thus, animal contacts, processing or handling raw animal products or meat and consumption of raw foods are the main risk factors to be considered [45,46]. Additionally, the risk factor "access of livestock to source of drinking water" was considered to take into account epidemiological circumstances. Indeed, this study showed the highest seroprevalence in the group of persons who had direct contact with livestock (30.9%) or raised livestock at home (31.9%). As expected, processing or handling raw animal foods proved to be an important risk factor (21.3%). 28 patients recorded consumption of undercooked meat or unpasteurized milk. 13 (39.4%) participants had direct contact with materials of aborted animals. Aborted foeti are usually left for decomposition by scavengers instead of proper disposing. This procedure increases the infection risk because large numbers of organisms are excreted with the uterine uid, placenta and fetus at the calving/lambing time [47].
Brucellosis in pregnant women bares the risk of abortion and may also cause repeatedly abortions after becoming chronic. Thus, participants of this study were asked about the course of previous pregnancies.
Indeed, 23.8% (n = 5) seropositive women reported on abortion. A recent study in Pakistan involving 429 pregnant women mostly from rural areas reported 5.8% seroprevalence and 14.6% of these had abortion history [24]. Due to the low number of cases in both studies, we only can recommend further studies to evaluate these ndings Brucellosis can also pose a serious risk to newborns. Mortality in newborns was reported as well as transmission of brucellosis to a neonate via the congenital route or via breastmilk [48][49][50][51][52]. Future studies in Pakistan should also consider this neglected aspect of brucellosis in childhood. 47 reference deleted The most common clinical sign observed among SAT seropositive patients was fever i.e. 94.1% (n = 64), followed by arthralgia with 55.8% (n = 38). Similarly, most often fever (77.71%) and arthritis (83.43%) were reported in 175 brucellosis positive hospital patients in Bikaner, India [53]. 30.8% (n = 21) participants of this study reported persons with similar symptoms of brucellosis in their households.
Transmission from person to person is rare, so these household members were most probably infected by the same animals or foods. Brucellosis in endemic countries is a family problem. Physicians should be aware of that fact and include all family members in their investigations. As brucellosis can be attracted again and again from the same source it is of imminent importance to identify and eliminate this source as well. Patients must be made aware of the epidemiology of brucellosis and local veterinary o cers need to be involved nally.
RT-PCR was performed for con rmation of brucellosis in patient samples. RT-PCR is a rapid, reliable, highly sensitive and speci c method for molecular diagnosis. Genus speci c primers and probes targeting the conserved BCSP-31 gene were used. 11.4% samples were RT-PCR positive. The study showed that RT-PCR is a rapid method to con rm brucellosis within 2-3 hours as compared to conventional methods which take several days to weeks and also poses a high risk of infection to laboratory personnel. Cases of brucellosis might have been lost by serological tests because they were in early stage of infection and antibodies were absent still.

Conclusions
The present study reported the prevalence of brucellosis among patients including pregnant women from hospitals of Abbottabad, Pakistan. The study showed that the population of Abbottabad is at higher risk of acquiring brucellosis because most people although living in urban areas have close contact with animals and consume raw products of animal's e.g. unpasteurized milk. The results of this study can be used to develop strategies for controlling human brucellosis in rural settings of Pakistan, to raise awareness about brucellosis in livestock professionals, consumers and physicians and to develop control programs by the authority in charge.

Consent for publication
Not applicable.

Availability of supporting data
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This research was funded by German Federal Foreign O ce, funded project "Building a network of laboratories in Pakistan to enhance biosafety and biosecurity in Pakistan".
Authors' contributions LH, SA, MAS, FM, HEA and HN conceptualized the study and did the manuscript write-up. LH, ST, US, AAS, SAA and GM analyzed the data. LH, SA, MAS, US and AUK wrote the article. All authors read and approved the nal manuscript.