Epidemiology of Taenia saginata taeniosis/cysticercosis: a systematic review of the distribution in southern and eastern Africa

Background The beef tapeworm, Taenia saginata, causing cysticercosis in bovines and taeniosis in humans, is thought to have a global distribution. In eastern and southern Africa, cattle production plays a crucial role in the economy, but a clear overview of the prevalence of T. saginata in the region is still lacking. This review aims to summarize existing knowledge on T. saginata taeniosis and bovine cysticercosis distribution in eastern and southern Africa. Methods A systematic review was conducted, that gathered published and grey literature, including OIE reports, concerning T. saginata taeniosis and bovine cysticercosis in eastern and southern Africa published between January 1st, 1990 and December 31st, 2017. Results A total of 1232 records were initially retrieved, with 78 full text articles retained for inclusion in the database. Unspecified taeniosis cases were reported for Angola, Ethiopia, Kenya, Madagascar, Malawi, South Africa, Tanzania, Uganda and Zambia, whereas T. saginata taeniosis cases were found for Ethiopia, Kenya, South Africa, Tanzania, Zambia and Zimbabwe. The prevalence of taeniosis ranged between 0.2–8.1% based on microscopy, and between 0.12–19.7% based on coproAg-ELISA. In Ethiopia, the percentage of tapeworm self-reporting was high (45.0–64.2%), and a substantial number of anthelmintic treatments were reported to be sold in towns. The presence of bovine cysticercosis was reported in all 27 countries/territories included in the study, except for Rwanda and Somalia, Comoros, Madagascar, Mauritius, Mayotte, Seychelles and Socotra. The prevalence of cysticercosis ranged between 0.02–26.3% based on meat inspection, and between 6.1–34.9% based on Ag-ELISA. Conclusions Although T. saginata has been reported in the majority of countries/territories of the study area, T. saginata taeniosis/cysticercosis remains a largely ignored condition, probably due to the absence of symptoms in cattle, the lack of data on its economic impact, and the fact that human taeniosis is considered a minor health problem. However, the occurrence of bovine cysticercosis is a clear sign of inadequate sanitation, insufficient meat inspection, and culinary habits that may favour transmission. Measures to reduce transmission of T. saginata are therefore warranted and the infection should be properly monitored. Electronic supplementary material The online version of this article (10.1186/s13071-018-3163-3) contains supplementary material, which is available to authorized users.


Background
The beef tapeworm, Taenia saginata, utilizes bovines as intermediate hosts and humans as final hosts. Although tapeworm infections have been reported since ancient times [1], it was not until 1782 [2] that differentiation of T. saginata from the other well-known meat-transmitted human tapeworm, Taenia solium, was established. Furthermore, it was not until 1871 that the role of cattle as intermediate hosts for the parasite was established, with "measly" beef being reported as the source of infection in patients [3].
Ingestion of raw or undercooked infected beef is indeed the mode of transmission of this zoonotic parasite to humans, in whom it develops to its adult form, a several metres long segmented worm consisting of a scolex with four suckers, neck and strobila, i.e. a chain of proglottids [4]. In contrast to T. solium, the gravid proglottids of T. saginata, which contain thousands of embryonated eggs, are mobile and can migrate from the anus independently of, as well as during, defaecation [5]. Eggs are then shed into the environment, and cattle become infected through grazing contaminated pastures, or ingesting contaminated fodder or water. After hatching, and penetration of the intestinal wall, the oncospheres reach the general circulation, distributing them throughout the body where they develop into cysticerci [4]. Common predilection sites for T. saginata cysticerci include the heart and masseter muscles [6].
In both intermediate and definitive hosts, T. saginata causes few symptoms. In humans, infection is usually characterized by anal pruritus due the active migration of T. saginata proglottids and some mild abdominal pain [7]. Nevertheless, the (potential) presence of a tapeworm in the body can cause distress [8], and some people even suffer from a pathological fear of tapeworms, often encouraged by horror stories circulating in popular media or books [9,10]. Moreover, although rare, complications due to taeniosis, such as appendicitis, have been reported [11]. In cattle, the infection is generally asymptomatic but nevertheless may incur great economic losses for the meat sector due to carcass condemnation or treatment upon detection of cysticerci during meat inspection, as well as related insurance costs [12,13].
Taenia saginata is distributed globally, with the parasite occurring in both developed and developing countries, although less frequently in countries where cultural preferences limit consumption of bovids or where adequate sanitary infrastructure reduces the likelihood of bovids ingesting human faecal matter. Thus, the prevalence of human taeniosis and bovine cysticercosis are considered particularly high in Africa, Latin America and some parts of Asia [4].
In eastern and southern Africa, the cattle population was estimated at a massive 20.6 million in 2016 [14], so the parasite is thought to be of particular relevance here. In the area, bovines are essential for the livelihoods of smallholders, serving as a source of food, draft power and manure, as well as acting as a financial buffer for challenging times. Although there are indications of the widespread presence of the parasite in at least some countries in this region (e.g. Ethiopia: [15][16][17]), an extensive overview of its distribution in this region, along with epidemiological considerations regarding its presence, is still lacking. Our aim was therefore to gather recent information on the presence of T. saginata in eastern and southern Africa.

Search strategy
A systematic review of published literature was conducted to collect data on the occurrence, prevalence, and geographical distribution of bovine cysticercosis and human taeniosis in eastern and southern Africa, published between January 1st, 1990 and December 31st, 2017. For the purpose of this study, eastern and southern Africa was defined as the area covered by the following countries/territories: Angola, Botswana, Burundi, Comoros, Djibouti, Eritrea, Ethiopia, Kenya, Lesotho, Madagascar, Malawi, Mauritius, Mayotte (French), Mozambique, Namibia, Réunion (French), Rwanda, Seychelles, Socotra (Yemini), Somalia (including the autonomous regions Puntland and Somaliland), South Africa, Swaziland, Tanzania (including the semi-autonomous region of Zanzibar), Uganda, Zambia and Zimbabwe. The PRISMA guidelines were followed whilst conducting the review [18] (Additional file 1). The search protocol can be found in Additional file 2.
The international bibliographic databases PubMed (http://www.ncbi.nlm.nih.gov/pubmed) and Web of Science (http://ipscience.thomsonreuters.com/product/webof-science/) were searched using the following search phrase: (cysticerc* OR cisticerc* OR "C. Furthermore, a range of databases for grey literature and MSc/PhD thesis documents were searched using keywords from the above search phrase (the full list of databases is presented in Additional file 3). Data on bovine cysticercosis from the different scientific databases were complemented with data from OIE databases "Handistatius" (1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004) and "WAHIS" (2005) [19,20]. Finally, reference lists of reviews on the topic were screened and additional relevant records were added to the database.

Selection criteria
Upon compilation of search results from the different databases, duplicate records were removed. Thereafter, titles and abstracts were screened for relevance, applying the following exclusion criteria: (i) studies concerning a parasite other than T. saginata; (ii) studies conducted outside the study area; (iii) studies published outside the study period; (iv) studies reporting results outside the scope of the review question (e.g. review, experiment, intervention trial); and (v) duplicated data. After the screening process, full text articles were evaluated using the same criteria listed above (Additional file 4).

Data extraction and generation
Data from included records were extracted. In reports where the numerator and denominator of the study sample were available, prevalence data were calculated, if not already provided. When not presented in the manuscript, the 95% exact confidence intervals (CI) were calculated, using the "binom.test" function ("stats" package) in R 3.5.1 [21].

Search results
A total of 1228 records were obtained from the database search, and four additional records were added through screening of the reference lists of relevant reviews (Additional file 4). After removal of duplicate records (n = 71), 1161 records were screened based on title and, thereafter, abstract. During title screening, 987 records were excluded, and a further 85 records were removed upon abstract screening; three of these were remaining duplicate records, whereas the other removed records focussed on a different parasite (n = 32) or study area (n = 18), were published outside the study period (n = 1), or had a different scope (e.g. laboratory experiments, review) (n = 31). Thus, 89 full text articles (n = 89) fulfilled the eligibility criteria for evaluation, but three of these were unavailable. During the evaluation of the remaining 86 records, eight were excluded due to having a different scope.

Human taeniosis occurrence
A total of 48 records reported the presence of human taeniosis cases (excluding those with confirmed T. solium taeniosis). Unspecified taeniosis cases were reported from Angola, Ethiopia, Kenya, Madagascar, Malawi, South Africa, Tanzania, Uganda and Zambia, whereas known T. saginata taeniosis cases were reported from Ethiopia, Kenya, South Africa, Tanzania, Zambia and Zimbabwe (Fig. 1). Microscopy results were included in 32 reports, most of which presented data from Ethiopia (18) ( Table 1). Taeniosis  Confirmed by PCR Abbreviations: CI, confidence interval; na, not available; HAART, highly active antiretroviral therapy prevalence based on microscopy alone ranged between 0.2-8.1% (villagers in Kenya [22] and Ethiopia [23], respectively), and one study reported the absence of taeniosis (in geophagous pregnant women in Kenya [24]). Four records presented data from coproAg-ELISA studies conducted in Kenya and/or Zambia, with a prevalence ranging between 0.12-19.7% (villagers in Zambia [25] and Kenya [26], respectively) ( Table 2), two of which involved confirmed T. saginata cases. Overall, common study groups were school-children, patients suffering from other diseases [e.g. HIV infection, sleeping sickness and active pulmonary tuberculosis (TB)], as well as occupational groups (e.g. tobacco farm workers, food handlers). Furthermore, eight studies reported taeniosis prevalence in communities based on selfreporting by questionnaire respondents (prevalence range: 45.5-64.2%) ( Table 3), and five records presented data on anthelmintic sales in towns (Table 4), both in Ethiopia. Another two records contained data on household latrine sampling, thus presenting prevalence at the household level (Malawi: 40.4% [27]; South Africa: 18.0% [28]). Finally, one report discussed a case of intestinal obstruction due to impaction of a T. saginata tapeworm in Zimbabwe, requiring enterotomy with bolus removal as well as appendectomy [29].

Bovine cysticercosis
Based on the retrieved data sources (both OIE databases and manuscripts/reports), the presence of bovine cysticercosis was reported in all of the 27 countries/territories studied, except for Comoros, Madagascar, Mauritius, Mayotte and Seychelles. In addition, no information was available for Rwanda, Somalia, Mayotte and Socotra (Fig. 2). Data from the two OIE data sources indicating the occurrence and/or number of cases are presented in Table 5. Apart from the OIE data sources, a total of 39 records were found to document results on bovine cysticercosis in the study region. Meat inspection results were included in 35 records (Table 6), with prevalence estimates ranging between 0.02-26.3%, while two records reported the absence of positive animals (Tanzania: 2011 [30], Zambia: 2001 [31]). Seven records provided serological data, mostly based on Ag-ELISA results (prevalence range: 6.1-53.5%), while one presented Ab-ELISA data (prevalence: 10.0%) [32] and another IHAT results (prevalence: 25.7%) [33] (Table 7). One study estimated the town level costs due to condemnation caused by bovine cysticercosis [Mekelle, abattoir level: 31,952 ETB/6 months (991 EUR, according to July 2018 exchange rates; 1 ETB = 0.0310 EUR) [34]], and another five studies provided data on total economic losses due to condemnation for a wide variety of conditions [17,30,[35][36][37]. Overall, the

Taeniosis and bovine cysticercosis occurrence
The co-occurrence of both bovine cysticercosis and taeniosis during the study period was reported in Angola, Ethiopia, Malawi, South Africa, Tanzania, Uganda, Zambia and Zimbabwe, but this was not the case for the other countries/territories studied. The occurrence of bovine cysticercosis or taeniosis was reported for all the countries/territories studied, except for Somalia, Rwanda and the Comoros, Mauritius, Mayotte, Seychelles, Mayotte and Socotra islands.

Discussion
The present study aimed at describing the epidemiology of T. saginata taeniosis/cysticercosis in eastern and southern Africa (1990-2017). Based on our findings, both human taeniosis and bovine cysticercosis were widespread in the 27 countries/territories studied, except for Somalia, Rwanda and six island states/territories, indicating that T. saginata is present in most countries of the study area. However, lack of diagnosis and reporting, particularly in rural areas, mean that the data accrued are likely to underestimate occurrence. The absence of data for some countries does not exclude the possibility that this parasite is present there as well. For example, given that of the three countries bordering Rwanda that are included in this review (Burundi, Tanzania and Uganda) all report the presence of this parasite, it seems unlikely that Rwanda is free from T. saginata. On the other hand, one potential hypothesis for the lack of reported T. saginata in Rwanda is the remarkably higher rate of access to improved sanitation services, at 60.8% in comparison to neighbouring Burundi at 35.5% [38]. The Rwandan civil war, during 1990-1994, culminating in the genocide of 1994, may have impacted reporting during that period, but does not explain the more recent lack of reporting. For Somalia, the ongoing civil war might explain the lack of reported data for the country, whereas for the six island states and territories, governmental or scientific interest in reporting cases may be lacking. Cases of taeniosis were reported for Angola, Ethiopia, Kenya, Madagascar, Malawi, South Africa, Tanzania,  Uganda, Zambia and Zimbabwe, yet the majority of reports on human taeniosis cases did not, unfortunately, provide species determination. Thus, cases of T. saginata taeniosis were not differentiated from infections caused by other Taenia spp. The pork tapeworm, T. solium, for instance, is presumed to also be widely distributed throughout eastern and southern Africa [39] and therefore we cannot conclude that all reported, unspecified taeniosis cases are due to T. saginata. Taenia solium is known to be the causative agent of the severe condition neurocysticercosis, associated with epilepsy, severe headaches, cognitive deficits [40] and a major cause of deaths among the food-borne diseases [41]. The presence of a single T. solium tapeworm carrier poses a major risk for his/her surroundings, as humans acquire neurocysticercosis through the ingestion of T. solium eggs transmitted through poor hygiene practices resulting in faecal-oral transmission [42]. Although tapeworm infections usually have an asymptomatic course [7] apart from some sporadic complications (e.g. intestinal obstruction in the Zimbabwean patient [29]), it is thus paramount to register cases as well as to differentiate case species, to allow precise prevalence estimates, and to guide appropriate control measures. Species determination, however, is hampered by the fact that Taenia spp. eggs cannot be differentiated upon coprological examination. Expelled proglottids of T. solium and T. saginata can be distinguished on the basis of the number of uterine branches, but such material is not always available. Moreover, more advanced diagnostic tools (e.g. copro-PCR) to differentiate species are often lacking in resource poor settings [43], and even in developed countries are not often performed due to lack of awareness about neurocysticercosis [12].
In certain countries in the study area, specific culinary habits put the consumers at great risk of contracting T. saginata taeniosis. For instance, in Ethiopia, "kitfo" is a very popular beef dish, in which the meat is usually consumed raw or lightly cooked, while "tibs" is another dish often containing undercooked beef. Furthermore, "kurt"   [91] refers to the habit of eating cubes of raw beef, finished off with local spices. Unsurprisingly, a high proportion of the Ethiopian population reports having had a tapeworm, and sales of taeniicidal drugs in Ethiopia are high [15,16,[44][45][46][47][48][49]. Access to adequate clean water and sanitation services (WASH) is notoriously poor across the whole of sub-Saharan Africa, including the region of interest to this paper. There are large between and within-country disparities, but overall sub-Saharan Africa lags far behind the goals set out by the international community in both the millennium development and the sustainable development goals with only 25.7% (23.1-28.6%) of the population having access to improved sanitation [38]. This lack of WASH capacity is strongly reflected by the presence of parasites such as T. saginata which requires ingestion of eggs passed in faecal material for the propagation of its life-cycle.
In eastern and southern Africa the cattle population is large, and bovine products, including meat, are an  important protein source for humans, as well as a source of draft power and form of investment. Beef cattle are typically kept in an extensive manner; animals are basically free-ranging. The presence of human T. saginata carriers shedding eggs into the environment puts these cattle at risk of bovine cysticercosis, and this presumably occurs widely in the study area. In developed countries, the condition is known to cause economic losses due to freezing or condemnation of the carcass as well as related insurance costs (e.g. Belgium: 3,408,455 EUR/year [13]). Studies investigating the magnitude of this economic loss in the study area are, however, limited, with data available from only one abattoir in Ethiopia [34]. Furthermore, reporting of bovine cysticercosis to OIE appeared to be inconsistent, with large variations in number of cases reported even within the same country, and gaps in the annual reporting (e.g. no data available after 2005).

Conclusions
Taenia saginata taeniosis/cysticercosis is a widespread, yet largely ignored, condition in southern and eastern Africa. This is probably due to the lack of symptoms in cattle, the lack of good data on its economic impact, and because human taeniosis is considered a minor health problem. Nevertheless, the presence of bovine cysticercosis is a clear sign of inadequate sanitation, insufficient meat inspection, and culinary habits that may favour transmission. Measures to reduce transmission of T. saginata are therefore warranted, and the infection should be properly monitored, in both humans and cattle. It should also be noted that as cattle are an important source of human protein and livelihoods in the area, ensuring optimal health and productivity of cattle is of indirect importance to human health and welfare as well as any direct impact. Species identification in tapeworm carriers is paramount to gain detailed insights in the distribution of the different Taenia spp. in the area, as well to avoid the development of the severe condition neurocysticercosis within communities due to ingestion of eggs shed by a T. solium tapeworm carrier. We conclude that in order to ensure both the safety of beef consumed in the southern and eastern Africa, and to improve the underlying sanitary conditions perpetuating the parasitic life-cycle, concerted, co-ordinated efforts must be made by integrating public, animal and environmental health in a One Health approach.