Food and waterborne protozoan parasites: The African perspective

Parasitic food-borne diseases, particularly those caused by the protozoan parasites Cryptosporidium, Giardia, Cyclospora cayetanensis and Entamoeba are increasingly becoming common and have received considerable attention in the last two decades. The ability of the transmission stages of the parasites to survive in the environment for prolonged periods, globalization of the food industry and changes in eating habits have contributed to the numbers of human infections. This systematic scoping review highlights these important water- and foodborne parasites in the African context, detailing the burden in African water sources, wastewater/effluents and fresh produce. A scoping review search targeting African countries was conducted in Medline, Web of science and African journals online as well as back referencing from included studies covering the period 1990 to January 2020. Out of 1134 studies, 68 were included in the review. The articles covered 17 out of 54 African countries. There were 39/68 studies reporting on water sources while the rest reported on fresh produce. Cryptosporidium prevalence ranged from 6 to 100% in surface water, 4 to 100% in tap water and up to 100% in wastewater and sludge. In fresh produce, Cryptosporidium was reported from five countries with prevalence of 0.8–75%. Giardia was reported in 47 out of 68 articles; prevalence ranged from 2.4% in surface water; 1% to over 70% in tap water; 28–100% in wastewater and 2% - 99% in fresh produce. Prevalence of Cyclospora cayetanensis was lower. Prevalence of Entamoeba was 78% in surface water; 100% in wastewater and up to 99% in fresh produce. This study finds that Africa is no exception to the risk presented by the subject parasites from water and/or food sources. Routine screening for these parasites particularly at household level and provision of adequate and safe drinking water would help to control the parasites.


Introduction
Parasitic protozoa are ubiquitous in nature and have a (Abd El-Salam, 2012) worldwide distribution. They are responsible for epidemic and endemic human suffering in both the developed and developing countries (Cotruva et al., 2004). Some parasites are zoonotic in nature, therefore, occur in animals (Robertson and Gjerde, 2001); their occurrence on foods and water should be considered a public health concern. Food and waterborne parasites are generally under-recognised, but this scenario has changed over the years due to a number of food and water borne disease outbreaks attributed to parasites (Marshall et al., 1997). In migrant workers, refugees, immunocompromised individuals, institutionalized individuals and, possibly, children in day-care centres (Salit et al., 2009;Carrero et al., 2020).
Despite the public health importance of these parasites, which has not extensively been highlighted in Africa, little is known about the occurrence and prevalence in food and water in Africa, the common modes through which infections occur.

The scoping review question
The subject protozoan parasites can infect humans from a variety of sources which can be broadly categorized as environment, animals and human sources. Studies looking at associations between risk factors and disease often sample humans to measure the rate of disease in relation to the selected risk factors. These studies do not detail the rate of occurrence of parasites in environmental and animal sources. This information is obtained when the individual potential sources of infection are sampled. Despite the many negative health effects of the four parasites (Cryptosporidium, Giardia, Cyclospora cayetanensis and Entamoeba), very little has been done to highlight the prevalence of the parasites in water and fresh produce in Africa. Additionally, limited studies have been conducted to determine the occurrence of the parasite in fresh produce which have been reported to be common vehicles for several parasites (Robertson and Gjerde, 2001). The aim of this review is therefore to provide an overview of evidence of Cryptosporidium, Giardia, Cyclospora cayetanensis and Entamoeba contamination on fruits and vegetables that are consumed raw and various water sources (and effluents) in Africa and highlight their potential for water and foodborne disease.

Methods
The study was conducted according to the methodology by the Joana Brigs Institute (JBI) (Peters et al., 2015) and the reporting guidelines of the Preferred Reporting System for Systematic Reviews and Meta-Analysis for scoping review Protocols (PRISMA ScR) (Tricco et al., 2018).

Inclusion criteria
Primary studies meeting the following criteria were included in this ScR; (1) all primary studies on the occurrence of the four parasites Cryptosporidium, Giardia, Cyclospora cayetanensis and Entamoeba in various water bodies/sources and fresh produce particularly fruits and vegetables with samples from Africa (any of the 54 countries), (2). study design -all peer reviewed observation studies (3) studies published from1990 up to January 2020 with no language restriction.

Exclusion criteria
This review excluded (1) editorials, systematic reviews or reviews covering the subject parasites with no primary data, (2) studies on the subject parasites but with a different population such as humans, animals, other environmental samples besides water, (3) samples outside Africa or with different outcomes and studies outside the targeted time frame.

Search strategy
The search was done in Medline via PubMed, Web of Science via the Web of Knowledge, Embase and African Journals online databases. A search phrase was developed based on Medline index terms and adapted to the other three databases (CM) (suppl2). The literature search was conducted on February 13, 2020. In addition, reference lists in the included articles were searched to capture possible additional publications. Majority of publications were excluded based on the title and abstract for not meeting the inclusion criteria. Independently, JS and FM read through all the retrieved articles, one-by-one and selected those that, reported studies on Cryptosporidium, Giardia, Cyclospora and Entamoeba conducted in any one of the African Countries. CM was responsible for quality assurance. In the third step, JS searched all additional relevant articles cited in the list of references of each of the initially selected articles. The selection of articles for inclusion from the search list was done based on the inclusion and exclusion criteria. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for this study is presented in Fig. 1.

Charting the results
Data extraction included basic information such as country, study site, author, publication year, study aims, sample type and sample size, diagnostic methods, number of positive samples, reported outcome measure such as prevalence or incidence, and incidental findings of other parasites. Prevalence was the major outcome measure and data has been presented according to parasite and sample type. In studies where prevalence was not given, it was estimated. Due to the heterogeneity of included studies, data was summarised and presented in tables without determining the overall average prevalence. Furthermore, incidental parasite isolations beyond the intended parasites were also reported.

Results
The results are presented in the following order; general findings (Section 3.1), prevalence and distribution of the four protozoan parasites (Cryptosporidium Section 3.2, Giardia Section 3.3, C. cayetanensis Section 3.4, Entamoeba histolytica/dispar Section 3.5) in water and sewage/effluent and, a combined prevalence and distribution of all four parasites in fresh produce (Section 3.6). Additionally, the prevalence of other parasites found in the various samples is presented (Sections 3.7 and 3.8). As a number of studies did not provide prevalence estimates, these were calculated based on the absolute numbers provided and are presented as percentages.

General findings
A total of 1134 scientific articles were retrieved from the data bases, of which 46 were included in the review. Additional publications (22) which were captured from the reference lists of the 46 included studies were also included giving a total of 68 articles. All the reviewed articles were published from 1990 to January 2020. Of these, 62 were full-length articles, three conference papers (Samie and Mashau, 2012;Petersen et al., 2014;Medani et al., 2016) and three theses (Duhain, 2011;Markos, 2013;Sampson, 2015). The publications were distributed throughout Africa covering 17 countries (Algeria, Burkina Faso, Cameroon, Cote d'Ivore, Egypt, Ethiopia, Ghana, Libya, Kenya, Nigeria, Rwanda, South Africa, Sudan, Tunisia, Uganda, Zambia and Zimbabwe) with North African countries accounting for the majority of the articles (29.4%; 20/68) while east Africa accounted for 17.6% (12/68) (Fig. 2). Egypt had the highest number of publications (11) followed by Ethiopia, Ghana, Nigeria and South Africa which had eight publications each. There were more publications from the year 2012 onwards with a peak in 2014 (Fig. 3). The studies assessed the occurrence and/or prevalence of Cryptosporidium or Giardia or Cyclospora or Entamoeba or a combination of the parasites in fresh vegetables, fruits, surface water (rivers, lakes, dams, wells, ponds), waste water (treated effluents, untreated effluents) and tap water. All publications, except one (Ssemanda et al., 2018) reported positive results for any of the parasites under review, either as a single parasite or in combination with the other parasites. A total of 51 publications reported Cryptosporidium, while 47 reported Giardia, 23 Entamoeba histolytica/dispar and 15 reported Cyclospora cayetanensis. Helminths and other protozoan parasites and various bacterial forms were also isolated in the various water sources, wastewater and sludge.
Diagnostic methods used to identify the parasites varied per study. For Cryptosporidium and/or C. cayetanensis, various studies employed different diagnostic methods including modified Ziehl Neelsen, modified Bailengar, immunofluorescence, Auramine Ophenol, immunomagnetic separation and staining and EPA 1623. For Giardia and/or Entamoeba, wet smears (after a concentration method) with or without iodine stains, and Zinc Sulfate floatation were used to identify the cysts. For water samples, they were either filtered before concentration or directly concentrated before examination. A few studies (15) employed PCR in the analysis of the water. Records after duplicates were removed (n = 820)

Eligible
Records screened (n = 820) Full text articles screened for eligibility (n=69) Articles included in the review (n=68) Articles excluded: (n = 1). Did not identify parasites of interest J. Siwila, F. Mwaba, N. Chidumayo et al. Food and Waterborne Parasitology 20 (2020) e00088 The highest number of articles reporting Cryptosporidium was from North Africa followed by West Africa and Central and Southern Africa (Fig. 4). Further, North Africa accounted for the highest number of studies reporting Giardia in various water bodies while East Africa had the least. East Africa also had the least number of publications reporting C. cayetanensis and Entamoeba (Fig. 4).
Fresh produce (fruits and vegetables) that are eaten raw varies from country to country. Carrots, onions, tomatoes, green pepper, lettuce, green pepper, parsley, watercress, tomato, cucumber and cabbage were the common vegetables reported to be eaten raw in the reviewed publications. Others, mostly from west African countries included green leaf (ugwu leaf, bitter leaf, sokoyokoto, igbagba), fluted pumpkin, waterleaf, morning glory solanum and curry (Alakpa et al., 2003;Chijioke et al., 2018). Fruits screened included mango, orange, lime, banana and cherry (Hassan et al., 2013;Tefera et al., 2014;Bekele et al., 2017;Chilo et al., 2018). Nineteen publications on contamination of fruits and vegetables were retrieved. Details of the parasite prevalence and distribution in various countries are described in Section 3.6.

Prevalence and distribution of Cryptosporidium in water and sewage/effluents
For purposes of this review, water samples were categorized into surface (if obtained from river, lakes, sea, pond or well), tap (if obtained from some form of conveyance such as a tap or hydraulic pump or borehole) or waste water when specified as such. Majority of the publications were on surface water (39/68). Table 1 summarises the distribution and reported estimates of Cryptosporidium in various water sources and sewage/effluents based on the specific African countries. Thirty-three of the 39 publications (85%) on various water sources reported occurrence of Cryptosporidium. The estimates are tabulated in Table 1. The reported prevalence of Cryptosporidium ranged from 6% (Sente et al., 2016) to over 58% (Grundlingh and De Wet, 2004;Ajeagah et al., 2010;Amenu et al., 2013) in surface water while the prevalence in tap water (sachet water was considered as tap water) ranged from 4% (Ndur et al., 2015) to 100% (Fikrie et al., 2008) (Table 1). Cryptosporidium oocysts were also recovered in wastewater, sludge and sewage (Table 1) with prevalence as high as 100% (Dungeni and Momba, 2010). The regional prevalence in various water sources and effluent/sewage was reported at 1-43% in North Africa,0.9-100% in West Africa, 7-100% in East Africa and 3-100% in Central and Southern Africa.

Prevalence and distribution of Giardia in water and sewage/effluents
From all the articles captured in this review, Giardia was the second most common parasite after Cryptosporidium; with 47 out of 68 articles reporting the parasite. In publications reporting both Cryptosporidium and Giardia, Giardia appeared to have higher prevalence rates (Fig. 4). The prevalence in surface water varied from 2.4% (Khalifa et al., 2014) to 100% (Ajeagah et al., 2005(Ajeagah et al., , 2010 while that in tap water ranged from 1% (Atnafu et al., 2012) to over 70% (Fikrie et al., 2008) (Table 2).
Articles on Giardia in wastewater, sludge/sewage were from Algeria, Cote d'Ivore, South Africa and Tunisia; the parasite estimates are indicated in Table 2. Prevalence ranged from 28% (Ben Ayed et al., 2012) to 100% of the tested sites and/or samples (Dungeni and Momba, 2010;Hamaidi-Chergui et al., 2019). The regional prevalence in various water sources and effluent/sewage ranged from 2 to 100% in North Africa, 0 to 100% in Central and Southern Africa, 5 to 70% in West Africa, and 1 to 73% in East Africa.

Prevalence and distribution of Cyclospora cayetanensis in water and sewage/effluents
Like Cryptosporidium and Giardia, Cyclospora cayetanensis was isolated from various water sources, wastewater and fresh produce. In surface water, the prevalence of C. cayetanensis ranged from 0.2% (El-Shazly et al., 2007) to 22% (2/9) (Dalu et al., 2011). In tap water, the prevalence was low for all studies reporting the parasite while in wastewater, Ben Ayed et al. (2012) reported the highest prevalence of 45% (see Table 3). The same article by Ben Ayed et al. (2012)   prevalence of 58% in sludge (Table 3). The regional prevalence in various water sources and effluent/sewage was 0.2-58% in North Africa, 0-22% in Central and Southern Africa and 5-7% in West Africa.

Prevalence and distribution of Entamoeba histolytica/dispar in water and sewage/effluents
Most studies did not apply molecular methods to distinguish E. histolytica from E. dispar and results were therefore reported as positive for E. histolytica/dispar. Similar to the other parasites discussed above, Entamoeba histolytica/dispar was reported in various water bodies and sources (Table 4) and from all the four categorized regions (North, West, East and Central and Southern Africa) with prevalence ranging from 8 to 36% in West Africa, 0.3 to 100% in North Africa, and 0 to 78% in Southern Africa. In surface water, the prevalence ranged from 1% (El-Shazly et al., 2007) to 78% (Mtapuri-Zinyowera et al., 2014). No publication reported  3.6. Prevalence and distribution of Cryptosporidium, Giardia, Cyclospora cayetanensis and Entamoeba in fresh produce Nineteen publications out of the 68 publications reported contamination of fruits and vegetables with the subject parasites. The prevalence of Cryptosporidium on fresh produce (fruits and vegetables) ranged from 0.8% (Tchounga et al., 2017) to over 80% (Saaed and Ongerth, 2019). Most publications that reported the parasites in vegetables/fruits were from West Africa (8/ 12), this review did not capture any article from the Central and Southern African region ( Table 5). The prevalence in East, West and North Africa were reported at 5-13%, 0.8-75% and 29-81% respectively.
Publications reporting Giardia cysts from fruits and vegetables in various countries are indicated in Table 5. Giardia cysts were identified in over 50% of the studies (73.7%; 14/19). Prevalence rates ranged from 2% (3/150) (Amaechi et al., 2016) to over 99% (159/160) (Chijioke et al., 2018). The regional prevalence in North, West and East Africa ranged from 3 to 84%, 2 to 99% and 8 to 28% respectively. Similar studies from central and southern Africa appear to be lacking as no articles on the subject matter were captured in the search.

Prevalence and distribution of other protozoan parasites in water, sewage/effluents & fresh produce
The subject protozoan parasites in this review were Cryptosporidium, Giardia, Cyclospora cayetanensis and Entamoeba. The reviewed articles also reported other protozoa as tabulated in Table 6 with varying prevalence rates. In surface water, the following protozoa were reported: Toxoplasma gondii, Sarcocystis spp., Isospora belli, Balantidium coli, Blastocystis hominis, Acanthamoeba, Entamoeba coli, Trichomonas vaginalis and Chilomastix mesnilli; in wastewater. Isospora belli and Eimeria spp. were reported while in tap water, only Sarcocystis was reported (Table 6). Balantidium coli, T. gondii and I. belli were also reported in vegetables (Table 6).

Discussion
This review has brought out important findings indicating the widespread contamination of African water bodies, tap water and fresh produce with protozoan parasites Cryptosporidium, Giardia, Cyclospora and Entamoeba. The review has also revealed the presence of the protozoa and other parasites in sewage and treated effluents. A significant number of articles have been published across the African continent on Cryptosporidium and Giardia in water, fruits and vegetables especially after the year 2010 (Fig. 3). There is, however, lack of uniformity in the design, determination of sample size and reporting of findings making it difficult to compare the various studies.
The review has further revealed increasing research on the waterborne protozoan parasites in various water sources and fresh produce over time. Publications were distributed throughout Africa with North African countries accounting for the majority of the articles while East Africa had the least number (Fig. 2). However, despite the clustering in the northern region, some countries produced more publications than others, for example, Egypt had 11 publications while there was only one from Algeria. The status was similar for east Africa where there is limited information in Uganda and Kenya compared to Ethiopia. There is need for more countrywide research in the various parts of the continent to appreciate the burden of the subject parasites in Africa.
Studies in surface water revealed widespread contamination with Cryptosporidium and Giardia, information which corroborates with studies in developed countries (Bouzid et al., 2008;Karanis et al., 2007). For Cryptosporidium, prevalence of over 50% (Grundlingh and De Wet, 2004;Amenu et al., 2013) were reported in most studies while Giardia prevalence was as high as 100% in some of the water samples tested (Ajeagah et al., 2005(Ajeagah et al., , 2010. Despite the variations in sample size, it is still clear that surface African surface waters are contaminated with the two important waterborne parasites. Tap water was equally contaminated, including packaged water (in sachets) (Ndur et al., 2015). Even though no major waterborne outbreak of diarrhoeal disease linked to protozoa contamination of water has been reported in Africa, the two parasites have previously been associated with waterborne outbreaks in the USA and other parts of the world (MacKenzie et al., 1994;Karanis et al., 2007). The high burden of Cryptosporidium and Giardia demonstrated in the reviewed studies in surface water, untreated and treated effluents provide evidence necessary to carry out causation studies in Africa. It is possible that contamination of water bodies (some of which are sources of water for many households) and tap water by these parasites could be responsible for some of the diarrhoea disease in affected countries. Future human studies reporting Cryptosporidium and/or Giardia infections should endeavor to establish sources of infections as this will assist in coming up with targeted intervention measures. The review also highlighted a high level of parasite contamination in wastewater (treated and untreated) and sewage/sludge. Wastewater as well as treated effluents were variably contaminated with the parasites under study. Wastewater (treated and untreated) has been reported to be increasingly being used in agriculture with the majority in untreated form in developing countries due to scarcity of water (Scott et al., 2010;Dickin et al., 2016). The levels of contamination exhibited in this review therefore raises concern. In places where wastewater may be used for vegetable cultivation, such contamination as seen in the reviewed publications poses a public health risk. It is important that such acts, where practiced, are discouraged and safe water should be provided to communities to safeguard their health. In the current review, 13 out of the 19 studies and 14 of the 19 studies on parasite contamination in vegetables and/or fruits that are normally eaten raw reported presence of Cryptosporidium and Giardia parasites, respectively. Cyclospora cayetanensis and E. histolytica were also detected (Table 5). With the known clinical consequences of human infection with these parasites especially in immunocompromised individuals (Supplementary Table 1) and the zoonotic potential of the parasites, it is important that people are sensitized about the risks of consuming raw or undercooked foods. Health education could therefore assist in preventing infections. Fresh produce can be contaminated with enteric pathogens throughout the process of planting to consumption. It is therefore, important that strict hygiene measures are advocated for in the production process as well as at household level to prevent human infections. Several publications also reported presence of other pathogens including helminths, other protozoa and bacteria in water, sewage and effluents and fruits and vegetables (Tables 6  and 7). It is evident that there is widespread helminths contamination especially in surface water and vegetables.
With the current evidence, it is important that research linking diarrhoea disease with contamination of water bodies/sources or fresh produce with protozoan parasites be carried out in Africa. This will provide evidence-based data on the various possible causes and sources of diarrhoea disease especially in children and immunocompromised individuals in disadvantaged communities. Furthermore, only 17 out of the 54 African countries have conducted studies on these parasites; more studies are necessary in other countries to provide conclusive evidence on the exact burden of the four protozoan parasites in water sources, sewage/effluents and fresh produce.
The limitation of this review is the possibility of failing to capture publications that are not indexed in the searched data bases. Further, only four databases were searched. However, we believe most were obtained from the reference lists of the reviewed publications and these sufficiently give a picture of the status of the target parasites in Africa. Secondly, some studies did not strictly follow the STROBE guidelines making it impossible to extract uniform data from the articles. Another limitation is that most studies did not perform molecular tests to distinguish E. histolytica and E dispar. The prevalence of E. histolytica may therefore be overestimated.

Conclusions and future perspectives
In many African communities, particularly those in rural areas, individuals have limited access to adequate and safe household water. This coupled with inadequate or lack of water treatment, poor hygiene practices, and lack of awareness and education programmes, significantly contributes to predisposition of many such communities to parasitic infections. With population growth  Hassan et al., 2013 in the midst of inadequate infrastructure, sanitary facilities and lack of a systematic way of determining the prevalence Cryptosporidium, Giardia, Cyclospora and Entamoeba histolytica/dispar from food and water sources, parasitism in humans in the African setting will continue to be a challenge. The findings in this review echo this. Despite the importance and popularity of swimming pools particularly in hotter months of year, reports of protozoan parasites in these waters poses a public health risk to users.  Amaechi et al., 2016;Tchounga et al., 2017;Chijioke et al., 2018 Further, swimming in contaminated water bodies, a common practice in rural settings, adds to the public health risk to users of these facilities. Periodic screening for these parasites in treated and untreated raw water particularly at household level and provision of adequate safe drinking water is advocated for. This should be supported by education and awareness programmes on the importance of using clean and safe water at household level including washing of fresh produce. Active and passive surveillance should also be conducted, and efforts should be made to minimise dissemination of oocysts and cysts in the farming environment and via human waste management. To achieve meaningful outcomes, it is also important to note that diseases that may result from infection with the discussed parasites cannot be managed with one method alone; they require an integrated and one health approach.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.