Anthrax outbreaks in the humans - livestock and wildlife interface areas of Northern Tanzania: a retrospective record review 2006–2016

Anthrax is a zoonotic infectious disease caused by Bacillus anthracis, a spore-forming, Gram-positive bacterium [1]. The disease occurs in humans, and wild and domestic mammalian species, in particular, herbivores [2]. Anthrax cases in humans are classified into three forms according to the clinical features and transmission routes: the cutaneous form, accounting for about 95% of all reported human cases worldwide, the gastrointestinal form, and the pulmonary form [3]. There is no evidence of person-to-person transmission of B. anthracis [4], and humans normally acquire the disease from direct contact with anthrax-infected animals or anthrax-contaminated animal products [4, 5]. Both domestic and wild animals serve as potential sources of infections in humans [6, 7]. The clinical presentation of this disease in susceptible herbivores is usually characterised by septicaemia and sudden death with/without bleeding from natural orifices and subcutaneous haemorrhages. Other symptoms in livestock and some wild herbivores are fever, dyspnoea, agitation, convulsions followed by sudden death. In pigs, carnivores, and primates the main symptoms are local oedema and swelling of the face and neck. Failure of the blood to clot, the absence of rigor mortis and the presence of splenomegaly are the most significant necropsy findings [6].

Worldwide, anthrax occurs at a low incidence in developed countries but remains endemic in African and Asian regions [6]. The African experience also illustrates the classic One Health aspects of anthrax where humans, livestock, wildlife and environment are important part of the epidemiological pattern. Anthrax outbreaks in Tanzania have been reported in humans, livestock, and wildlife over several years, and areas mostly affected are those in the livestock-wildlife interface [7]. Anthrax outbreaks in hippos were reported in Ugalla Game Reserve in 2000 and 2001 as well as in Mtera dam in 2003 [4]. Several large outbreaks (> 500 deaths) have also been reported in cattle, goats, and sheep in the eastern part of the Serengeti National Park [7]. In September 2016, a total of 153 hippopotamus died in Kilombero River due to anthrax outbreak, and in early October 2016, an anthrax outbreak in livestock occurred in Ngorongoro district, where ten humans were infected, two of them died [8]. Anthrax outbreaks have been reported in the Serengeti ecosystem for many years, mostly with sporadic outbreaks in several endemic hotspot areas affecting humans, livestock and wildlife animals [9].

Anthrax incidence in a given locality is related to temperature, rains or drought, soil, vegetation, host condition and population density [10, 11]. The local weather condition of an area may directly or indirectly influence possibilities for animals to come into contact with B. anthracis spores. This may include grazing closer to the soil in dry periods when grasses are short or sparse, and movement of herds to protected areas for wildlife conservation when water becomes scarce. The general state of health of the hosts may also affect their level of resistance to infection [6].

The health personnel in the Ministries responsible for health of human, livestock and wildlife (Epidemiology Unit) and the Ministry of Regional Administration and Local Government are responsible for responding to disease outbreaks as soon as they get outbreak notification from lower levels. During the disease outbreak response, their role is to identify and characterize the outbreak etiologic agents, monitoring the progress of the outbreak and putting the effectiveness of control and preventive strategies in place [12]. From the national level, information of public health emergence or disease outbreak is required to be communicated to WHO within 24 h [13, 14]. For the human surveillance system, communication during surveillance, reporting, and the response is by telephones (mobile and landlines), internet, fax, radio (national and local stations), television, letters, technical Meetings (National task force) and workshops [15]. Laboratory diagnostic reports of anthrax from Tanzania Veterinary Laboratory Agency (TVLA) and Tanzania Wildlife Research Institute (TAWIRI) Serengeti are regularly shared by the Ministries responsible for livestock and wildlife respectively. These reports are crucial for setting up control measures of anthrax outbreaks in livestock and wildlife.

Regular analysis of diagnostic and surveillance data from livestock and wildlife are essential for efficient management of anthrax outbreaks in animals and protecting human population [16].

However, despite the frequent occurrence of anthrax outbreaks in Tanzania, there is no comprehensive analysis of data indicating the magnitude and spatial distribution of the disease from the human, livestock and wildlife health sectors. It is therefore important to coalesce and summarize the available information to assess more comprehensive epidemiological patterns of anthrax in Tanzania.

We conducted a retrospective review of reported anthrax outbreak records from the human, livestock, and wildlife surveillance systems of Tanzania from January 2013 to December 2016. This was followed by a more thorough examination of data from Arusha and Kilimanjaro regions for 2006 to 2016. The specific objectives were (i) to identify the districts assumed to be an anthrax hotspots in Arusha and Kilimanjaro regions, (ii) to evaluate the efficiency of the anthrax reporting and response system and diagnostic capacity at national, regional and district levels, (iii) to describe the epidemiology of anthrax in the hotspot areas and (iv) to identify potential areas for further observational studies to better understand the complex ecology of anthrax.

This study has revealed that there is a close temporal correlation between the occurrence of anthrax outbreaks in animals (livestock and wildlife) and humans in the Arusha/ Kilimanjaro ecosystems. It might be attributed to the ongoing interactions between humans and animals such as types of husbandry, humans looking for food (meat and milk) and other livelihood issues like the use of animal skin as bedding materials, a common practice in the pastoral community.

We also, found more cases occurring in the dry season starting from September through November which might be a facilitating factor for anthrax transmission in animals and then into humans. The observed seasonal occurrences of anthrax show that climate-related factors (precipitation and ambient temperature) play a crucial role in triggering outbreaks, although there are variations between locations and therefore contributing factors are debated [9]. Some African countries, like this study have been reporting anthrax outbreaks at the end of dry seasons which indicates that over grazing, nutritional stress and congregation of animals at watering points might propagate the disease transmission [18, 19]. Also, animals tend to assemble themselves in certain places when there is pasture shortage, increasing chances of occurrences of anthrax [20].

Furthermore, water bodies may collect and accumulate spores in “storage areas” [21]. As water storage points are the last locations to hold water during dry seasons, these are the dangerous areas where animals tend to acquire the infection through drinking spore-contaminated water [22]. Our data demonstrated a seasonal pattern of anthrax outbreaks in northern Tanzania with peaks of outbreaks in humans, livestock and wildlife during March (start of long rain season) and September through November (end of dry season) each year in the last decade. This shows a high potential for anthrax infection in the human-livestock and wildlife interface areas of northern Tanzania [12], representing critical information to decision makers that they will need to set up preventive measures.

These measures could include strategic vaccination of livestock against anthrax, distribution of human antibiotic prophylactics to hotspot areas, and health education to high risk communities a few months before the expected time of anthrax outbreaks. Nevertheless, effective anthrax control depends on ensuring that the disease is controlled in livestock through routine targeted vaccination which may automatically control the problem in humans [20]. Restriction of free movement of healthy livestock during outbreak periods and safe disposal of dead animals (in a pit of six feet deep added with 10% formalin poured on top of the carcasses). This should be followed by soil decontamination on the area where the carcass is laying and removal of bloody soil, so it can help to prevent further occurrences of the disease [23].

We further found that men were at higher risk of acquiring anthrax (65.2%) than women, which might be due to slaughtering and handling the meat from the carcass of dead or slaughtered sick animals without inspection by a designated livestock officer. This is more likely to be the route of exposure for many of the cutaneous anthrax cases we found in this review [24]. More often they also eat meat while grazing their animals in the wilderness, only bringing home any remaining meat and offal for the wives and children. Moreover, men are the decision makers of the family who also dictates whether women and children should go to the hospital when they fall sick. This might impact on the health seeking behaviour of women creating a false representation in hospital registers regardless of the true disease status. Effective clinical management of zoonotic diseases depends on various factors including health seeking behaviour of individuals [16].

Pastoralists handle sick animals before dying and dress the carcasses after death [25]. Also, extensive handling of meat at different stages of preparations with direct skin contact with anthrax infected materials. It is a risk for contracting the infection and entry of B. anthracis into the human skin abrasions and can cause the cutaneous form of anthrax which we mostly found during this review. The cutaneous forms of anthrax are easily diagnosed clinically in health facilities and in laboratories by performing a Giemsa or Methylene blue staining on the discharges from lesions to detect the presence of B. anthracis. Other forms of anthrax like gastrointestinal and pulmonary are not as easily diagnosed at most of the existing health facilities in the anthrax hotspot districts in Tanzania.

High numbers of recorded human anthrax cases (Table 1) may partly be due to many patients that report at the specified health facility and good systems of recording patients in the health management information system (HMIS) outpatient & in-patient department booklets. In a few instances, some health facilities had no HMIS booklets which may account for no or a low number of reviewed human anthrax cases. For example, the Magaiduru dispensary in Ngorongoro district had no HMIS booklets to keep records of human anthrax cases regardless of some verbal information on the presence of human anthrax suspected cases in the village they serve. The same applies for IDSR reporting forms that in past years anthrax was not one of the IDSR priority diseases. Therefore some facilities did not bother to report the disease until 2013 when a revised National IDSR Guidelines included anthrax as one of the immediately reportable diseases. However, in Hai district they historically improvised a slot on the reporting forms for capturing anthrax cases in the infectious diseases weekly ending (IDWE) reporting forms which accounts for a high number of anthrax cases in Hai district compared to other hotspot districts of Kilimanjaro region. Overall, Arusha region has reported more anthrax human cases in a time series of the last ten years and Ngorongoro district having more anthrax human cases compared to other districts. This might be contributed by the pastoral communities living in close proximity with wildlife conservation areas and facilitating disease transmission between livestock and wildlife animals and then to humans.

Anthrax outbreaks cause substantial economic losses through livestock and wildlife losses, the cost of laboratory reagents and carcass disposal (burial or incineration). Therefore investment in the control of this disease is inevitable [26]. Response to these outbreaks requires joint collaborative efforts of the Ministries responsible for human health, livestock, and wildlife services. However, one of the biggest challenges in the control of zoonotic diseases is the current lack of joint approaches for responding to disease outbreaks. Therefore, there is a need for the creation of joint response action plans with combined technologies and infrastructures from both public health and veterinary professionals including sociologists, and ecologists to initiate approaches to contain zoonotic diseases [27]. Worldwide, a One Health approach is a call to action for the establishment of closer professional interactions, collaborations, capacity building and research opportunities across the science professionals and related disciplines to improve the health status of humans, livestock, wildlife, and the environment [28]. In Tanzania, one of the challenges for initiating the joint surveillance system under a One Health approach would be lack of compatible surveillance systems between the ministries responsible for human health, livestock and wildlife. Nonetheless, the right opportunity is the existence of a strong IDSR system within human health sector which can be improved and expanded to cover a harmonized list of priority zoonotic diseases  in a ‘One Health’ approach.

In countries like Kenya the five diseases identified as top priority zoonotic diseases are anthrax, trypanosomiasis/HAT, rabies, brucellosis and Rift Valley Fever (RVF) in descending order [29]. This highlights the importance of prioritizing zoonotic diseases in Tanzania, as well as presenting opportunities to focus on diseases with the greatest local public health burden and not focus only on diseases that have greater global attention [29]. Most often, authorities start looking for the disease in livestock and take appropriate actions only after they report human cases and deaths in that particular area. When disease surveillance and control take this approach, humans essentially serve as a sentinel species (human illness and death) act as proxy indicators of disease prevention and control in livestock [12].

We also observed that there is poor anthrax diagnostic capacity, not only in the hotspot districts in northern Tanzania, but the entire country. The only routine diagnostic techniques performed at TVLA in Arusha and TAWIRI in Serengeti National Park are either Giemsa or Methyline blue staining of fixed blood smear from either humans or animals (livestock and wildlife). They also test swabs collected from skin lesions of human anthrax suspected cases by the same technique at some health facilities of the hotspot districts in Northern Tanzania. The Ministry of Livestock and Fisheries headquarters has a laboratory (Biosafety Level 2 Laboratory) with a capacity for diagnostic polymerase chain reaction (PCR) for anthrax, but is located in Dar es Salaam about 700 Km away from the hotspot areas. This hampers the diagnosis of other forms like pulmonary and gastrointestinal anthrax, which are currently managed clinically at respective health facilities and may be confused with so many other diseases with a potential of causing pneumonia and/or bloody diarrhoea within the hotspot areas. There is a concern for biosafety in clinical laboratories; the requirements vary in different countries. We consider Biosafety Cabinet level 2 appropriate for clinical laboratory analysis, while biosafety level 3 is more suitable for research related studies involving spore suspensions in liquids formulation or large-scale cultures [30]. We would therefore recommend an animal US snap test for anthrax, which needs a specimen from a dead animal (tissue or blood) and obtain results within a short time. Similar tests can be pursued for humans after having good response in the animal sector in terms of supportive political will, user acceptance and accessibility of the relevant technology and gadgets.

In most instances, many anthrax outbreaks are mismanaged, particularly in rural areas, where it is unlikely to have been adequately diagnosed, reported on time and forwarded to the central levels for rapid response. Under-reporting of the IDSR priority diseases (including anthrax) through the electronic surveillance system was revealed in Arusha and Kilimanjaro regions. Therefore, anthrax cases detailed in this study include only those that were recorded, communicated and reviewed in the surveillance systems of Tanzania. It only provides an estimate of the magnitude of the disease which could be a significant under - estimate of the disease burden. Furthermore, the collected anthrax data would assist in future ecological niche modelling in order to map for areas where anthrax outbreaks are more likely to be occurring. This may be a tool for optimization of control measures and improving epidemiologic knowledge of this disease in Tanzania. The causality of various potential risk factors for anthrax transmission in the affected communities of northern Tanzania could also be tested with appropriately designed future observational studies.

The findings of this study are critical for consideration by respective authorities for setting up prevention and control measures of anthrax outbreaks in the human, livestock and wildlife sectors within Tanzania. There is a gross under-reporting of anthrax cases in existing human and animal surveillance systems, which can be an obstacle for estimating the real burden of anthrax in the hotspot districts. We also noticed that people living in the marginalised communities like the Maasai remain at high risk of contracting anthrax infection given their ties to cultural practices of handling and consuming dead animals and their products. Moreover, repeated occurrences of anthrax in livestock, wildlife, and humans suggest for strengthening links and promoting inter–disciplinary and multi-sectoral collaboration to enhance the improved prevention and control measures for anthrax outbreaks in a One Health approach.