Toxoplasma gondii Infections in Animals and Humans in Southern Africa: A Systematic Review and Meta-Analysis

Background: Toxoplasma gondii is an apicomplexan parasite with zoonotic importance worldwide especially in pregnant women and immunocompromised people. This study is set to review the literature on T. gondii infections in humans and animals in southern Africa. Methods: We extracted data regarding T. gondii infections from published articles from southern Africa from 1955 to 2020 from four databases, namely Google Scholar, PubMed, EBSCO Host, and Science Direct. Forty articles from eight southern African countries were found eligible for the study. Results: This review revealed a paucity of information on T. gondii infection in southern African countries, with an overall prevalence of 17% (95% CI: 7–29%). Domestic felids had a prevalence of 29% (95% CI: 7–54%), wild felids 79% (95% CI: 60–94), canids (domestic and wild) 69% (95% CI: 38–96%), cattle 20% (95% CI: 5–39%), pigs 13% (95% CI: 1–29%), small ruminants (goats and sheep) 11% (95% CI: 0–31%), chicken and birds 22% (95% CI: 0–84%), and humans 14% (95% CI: 5–25%). Enzyme-linked immunosorbent assay (ELISA) and immunofluorescence antibody test (IFAT) constituted the most frequently used diagnostic tests for T. gondii. Conclusions: We recommend more focused studies be conducted on the epidemiology of T. gondii in the environment, food animals and human population, most especially the at-risk populations.


Introduction
Toxoplasma gondii is an apicomplexan obligate parasite that infects animals and humans worldwide [1]. The definitive hosts are felids although a recent study showed developmental success in mice subjected to certain enzymatic inhibition and diet modification [2]. The intermediate hosts include terrestrial and aquatic mammals and birds [2,3]. The pathways of T. gondii infection and transmission are multifaceted, involving the three developmental stages (tachyzoite, bradyzoite, and sporozoite) of the parasite's life cycle [2]. Intermediate hosts, including humans, can acquire infection via (i) consumption of water, vegetables, and fruits contaminated with infective oocysts; (ii) consumption of raw or undercooked meat infected with tachyzoites or bradyzoites [4]; (iii) blood transfusion; (iv) organ transplant containing cysts or tachyzoites; and (v) congenital transmission from the mother to fetus via the placenta. Feline definitive hosts acquire infections via the ingestion of sporulated oocysts or by carnivorism. However, rarely, consumption of non-pasteurized milk or milk latent infections, as antibiotics are unable to reach the bradyzoites in adequate concentrations [23,27].
Toxoplasmosis prevention is centered around avoidance of contact with sources of infection, such as cats, contaminated environment, consumption of raw or undercooked meat, personal hygiene, and regular handwashing [23]. The control of mechanical vectors of transmission, such as cockroaches, flies, or rodents in the surroundings, can also be adopted in disease control [24]. This review aims to analyze published literature on Toxoplasma infections in animals and humans in southern Africa and determine the epidemiological distribution of infection in various hosts in the region and identify gaps for future research.

Systematic Review
A total of 3197 articles were identified from the following databases: Google Scholar, PubMed, EBSCO Host, and Science Direct. After duplicates (n = 2111) were removed, title and abstracts were perused for 1086 articles. An additional eight studies were identified from other sources. Overall, 1029 articles were excluded because they were not original articles, non-relevant to research objectives to the study, or abstracts. Of the 65 reviewed full-text articles, 40 were selected for inclusion in the systematic and meta-analysis. A flow diagram illustrating this selection process is presented in Figure 1. effective but not in latent infections, as antibiotics are unable to reach the bradyzoites in adequate concentrations [23,27]. Toxoplasmosis prevention is centered around avoidance of contact with sources of infection, such as cats, contaminated environment, consumption of raw or undercooked meat, personal hygiene, and regular handwashing [23]. The control of mechanical vectors of transmission, such as cockroaches, flies, or rodents in the surroundings, can also be adopted in disease control [24]. This review aims to analyze published literature on Toxoplasma infections in animals and humans in southern Africa and determine the epidemiological distribution of infection in various hosts in the region and identify gaps for future research.

Systematic Review
A total of 3197 articles were identified from the following databases: Google Scholar, PubMed, EBSCO Host, and Science Direct. After duplicates (n = 2111) were removed, title and abstracts were perused for 1086 articles. An additional eight studies were identified from other sources. Overall, 1029 articles were excluded because they were not original articles, non-relevant to research objectives to the study, or abstracts. Of the 65 reviewed full-text articles, 40 were selected for inclusion in the systematic and meta-analysis. A flow diagram illustrating this selection process is presented in Figure 1.

Identification of studies via databases
Identification of studies via other methods
Based on animal groups, T. gondii infection in domestic felids in the region had an overall prevalence of 29% (95% CI: 7-54%) (  Forest plot of prevalence estimates of Toxoplasma gondii infections in Southern Africa. The confidence interval (CI) was 95%, and the diamond represents the pooled estimate (blue squares represent point estimation of the study weighted for population size) [3,5,19 ,28-44, 46,48-65].

Toxoplasma gondii Infections in Humans in Southern African Countries
The pooled prevalence of T. gondii infection in humans was 14% (95% CI: 5-25%), with the highest prevalence of 17% (95% CI: 4-33%) recorded in South Africa and the least prevalence of 2% (95% CI: 1-3%) from Namibia ( Figure 10). A summary of studies on Toxoplasma infections in humans in southern African countries is shown in Table 5. Out of a total of 8623 serum samples that were examined, 1342 were positive for Toxoplasma serology. Furthermore, an additional archaeological study on dead human remains was reportedly positive for T. gondii.

Toxoplasma gondii Infections in Humans in Southern African Countries
The pooled prevalence of T. gondii infection in humans was 14% (95% CI: 5-25%) with the highest prevalence of 17% (95% CI: 4-33%) recorded in South Africa and the leas prevalence of 2% (95% CI: 1-3%) from Namibia ( Figure 10). A summary of studies on Toxoplasma infections in humans in southern African countries is shown in Table 5. Out o a total of 8623 serum samples that were examined, 1342 were positive for Toxoplasma se rology. Furthermore, an additional archaeological study on dead human remains was re portedly positive for T. gondii.

Toxoplasma gondii Infections in Humans in Southern African Countries
The pooled prevalence of T. gondii infection in humans was 14% (95% CI: 5-25%), with the highest prevalence of 17% (95% CI: 4-33%) recorded in South Africa and the least prevalence of 2% (95% CI: 1-3%) from Namibia ( Figure 10). A summary of studies on Toxoplasma infections in humans in southern African countries is shown in Table 5. Out of a total of 8623 serum samples that were examined, 1342 were positive for Toxoplasma serology. Furthermore, an additional archaeological study on dead human remains was reportedly positive for T. gondii.

Discussion
Toxoplasma gondii is a coccidian cosmopolitan parasite of global economic and zoonotic importance. The importance of T. gondii in the meat industry and public health has been reported in a wide variety of hosts and humans, especially among immunocompromised individuals. This review revealed that there is limited information on the distribution of T. gondii in animals and humans in southern African countries. In this study, the overall pooled prevalence is estimated as 17% (95% CI: 7-29%).
The overall pooled prevalence of T. gondii infection 29% (95% CI: 7-54%) in domestic felids observed in this study is lower than the pooled seroprevalence of 51% (20-81%) reported in Africa, 52% (15-89%) in Australia [10], and 30-40% global prevalence from previous studies [66,67]. However, the pooled prevalence of T. gondii infections observed in wild felids 79% (95% CI: 60-94%) in this study is higher than the pooled prevalence reported in Africa, Asia, Europe, and South America [10], while in north African countries, no data were available on wild felids [68]. The role of felids (domestic and wild) in T. gondii epidemiology has been documented in several reports [10,69,70]. In this review, seven (7) studies were on wild felids, while five (5) studies were on domestic cats. A single infected felid is capable of shedding millions of oocysts for 10-15 days, thereby contaminating the environment and posing infection risk to various intermediate hosts [70]. Emphasis on the adequate veterinary care of animals, including frequent treatment of cats for toxoplasmosis and reduction in the population of stray cats in the environment, should be encouraged in southern African countries. Moreover, a surveillance system for Toxoplasma infection should be instituted at the wildlife-livestock interface areas in the region.
Limited studies exist on T. gondii infection in canids (domestic and wild), with an overall pooled prevalence of 69% (95% CI: 38-96%). This result is higher than the prevalence of 51.2.% reported in wild canids by Dubey et al. [71] and the global prevalence of 39.6% reported in foxes [72]. The studies in cattle were few and only done in South Africa and gave an overall pooled prevalence of 20% (95% CI: 5-39%), which is higher than the pooled prevalence of 16.3% (10.6-23.0%) from West Africa [73] and 12% (CI 8-17%) in the entire continent of Africa [1]. The estimated prevalence is, however, lower than the reported seroprevalence from Brazil and Sudan [74,75]. Studies have identified the consumption of raw or undercooked beef as a possible risk of toxoplasmosis transmission in humans [76,77].
Studies reporting the seroprevalence of T. gondii in pigs in southern Africa emanated from South Africa and Zimbabwe, with an overall pooled prevalence of 13% (95% CI: 1-29%). This is similar to the prevalence reported in pigs from Europe [80] but lower than the prevalence reported in pigs from North America, South America, Asia [82], West Africa [73], Africa [1], and globally [82]. Pigs are among the popular food animals and have been reported as a source of human toxoplasmosis through ingestion of raw or undercooked pork [83]. Toxoplasma gondii infections in pigs are either acquired prenatally via transplacental transmission or postnatally via ingestion of oocysts from a contaminated environment [1]. Hence, indoor rearing of pigs is important to reduce the exposure of pigs to T. gondii infections from the contaminated environment [1,43,84].
The overall pooled prevalence of 22% (95% CI: 0-84%) of T. gondii seroprevalence from chickens and birds in southern African countries is lower than the estimated prevalence of anti-T. gondii antibody 22% (95% CI: 0-84%) reported in chickens in West Africa [73] and 37.41% (95% CI: 29.20-46.00%) from chickens in Africa [1]. Chicken meat is a key contributor to animal protein due to affordability and availability [85]; however, it also plays a major role in human toxoplasmosis transmission when the meat is consumed raw or undercooked [1]. The free-range chickens ingest T. gondii oocysts from the contaminated environment while foraging, thus acting as zoonotic agents of human toxoplasmosis. The role of birds, especially the birds of prey, in maintaining transmission between the sylvatic cycle and domestic cycle has also been documented [86].
The pooled seroprevalence of anti-T. gondii antibody from humans came from studies that focused mainly on immunocompetent individuals, HIV+ patients, and pregnant women [8,54,57,60,62,63] as well as a few studies on blood donors and children [56,61]. Overall, the pooled prevalence of 14% (95% CI: 5-25%) of T. gondii infection in humans from southern African countries was lower than the seroprevalence reported from a metaanalysis conducted on pregnant women in African regions, American regions, eastern Mediterranean regions, Europe, the South-East Asia region, globally [87], and in some North African countries (Tunisia, Egypt, and Morocco) [68]. However, this prevalence is greater than the seroprevalence reported from Western pacific region and the World Health Organization (WHO) regions of the world, 1.1% (0.8-1.4) [87]. Humans acquire T. gondii infections either through ingestion of oocysts from the contaminated environment [88,89], via tissue bradyzoites from consumption of raw or undercooked infected meat, transplacental transmission from mother to fetus [46,90], or organ transplants or blood transfusion [11,91]. Infections in immunocompetent individuals are not associated with critical symptoms compared to the immunosuppressed, particularly AIDS patients or newborns. Congenital transmission often results in clinical manifestations, such as encephalitis, pneumonia, and ophthalmologic disorders [1,68]. The seropositivity of T. gondii prevalence in the subjects in the reviewed articles suggests an active transmission of human toxoplasmosis in the region and requires intervention to prevent infection. Control and prevention measures include environmental control of feral cats, provision of veterinary care of domestic animals, adoption of personal hygiene, creating awareness of the risk associated with consumption of raw or undercooked meat, adequate screening of blood or organ donors, and adopting a national toxoplasmosis treatment scheme for pregnant women in the region [10,92].
Diagnostic tools used in the reviewed articles varied widely and ranged from MAT, LAT, IFAT, ELISA, DT, CF, Wolstenholme's modification, and Sabin-Feldman dye test techniques to molecular approach. Studies have shown that different diagnostic techniques produce results that are heterogeneous [68]. For instance, the diagnostic performance of the MAT technique has been reported to be higher than that of ELISA [93]. In this study, the majority of articles adopted ELISA and IFAT to determine the seroprevalence of T. gondii. Although serological methods seem to lack sensitivity and specificity, they remain a standard tool for the qualitative detection of antibodies [68]. Studies that used LAT [3,56,60], histology [28,33,34], and molecular techniques [48,50,64] were few, while others used the combination of one or two of LAT, MAT, ELISA, IFAT, DT, CF, Wolstenholme's modification, and Sabin-Feldman dye test techniques [30,32,35,41,42,54,59,93,94]. A recent study comparing three serological diagnostic tools showed that ELISA and IFAT had relatively higher sensitivity and specificity than MAT [95]. Additionally, ELISA and IFAT are less laborious and time-consuming than MAT [95]. As much as molecular tools are reliable diagnostic tools, they were used in only three studies. Molecular tools are ideal for determining the distribution of T. gondii in the environment (soil and water samples), and the few studies might have been attributed to the non-availability of this diagnostic facility or the lack of competent individuals for such analysis. The adoption of molecular methods (both PCR and more discriminatory and advanced molecular tools, such as PCR-RFLP markers and DNA sequencing) will be imperative in identifying the T. gondii strains infecting various hosts.
Generally, substantial heterogeneity existed between the studies reviewed and subgroups. This may be due to a range of factors, such as people's varying hygiene practice levels, limited studies from some countries, varying diagnostic methods used, methods of rearing livestock animals, meat consumption pattern of studied individuals, or hostage.

Search Strategy
A systematic literature search was conducted in the following databases: Google Scholar, PubMed, EBSCO Host, and Science Direct using the following terms and Boolean operators (AND, OR): Toxoplasma AND Toxoplasmosis in southern Africa, Toxoplasma in cats AND southern Africa, Toxoplasma in livestock (sheep, goats, cattle) AND southern Africa, Toxoplasma in wildlife AND Southern Africa, Toxoplasma in felids, Toxoplasma in fowls AND Southern Africa, and Toxoplasma in humans AND southern Africa. The titles and abstracts of the search results were perused for the retrieval of relevant articles. References from selected articles were further used as a guide to other literature. The literature search was concluded in June 2021. Full-text articles were retrieved and managed in Endnote reference manager, version X7 (Clarivate Analytics, Philadelphia, PA, USA). This systematic review was performed following the PRISMA protocol (Reporting Items for Systematic Reviews and Meta-Analyses).

• Inclusion and exclusion criteria
An article was included in this study if it was published between 1955 and 2020 in a peer-reviewed journal and reported on (1) prevalence of T. gondii in cats and/or other animals and (2) Toxoplasma seroprevalence in humans in southern Africa. Dead links, duplicates, and grey pieces of literature were excluded during the literature review. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) used in this review is shown in (Figure 1).

•
Data extraction and quality assessment From each selected article, data on the study period, country of study, type of hosts, sample size, number of infected subjects/hosts, prevalence (%), and the diagnostic method(s) used were retrieved. Quality assessment of the identified articles was done as described by Munn et al. [96]. Quality assessment of each article was based on the following information: (1) relevance of research objective(s) to Toxoplasma, (2) prevalence of Toxoplasma as the main objective of the study, (3) study design was appropriately defined (case reports, cross-sectional), (4) samples randomly selected, (5) study subjects categorized by age/sex were relevant, (6) use of valid diagnostic methods in the study, (7) reliability of diagnostic methods, (8) representativeness of target sample to the general population, (9) description of the prevalence of Toxoplasma infection in the study community/animals, and (10) geographical location of Toxoplasma infection defined. The index score for each article was calculated by dividing the quality assessment of the study by ten. Detailed information about reasons for inclusion/exclusion and quality assessment is shown in Supplementary File S1.

Data Analysis
The extracted data from the search were entered in Microsoft Excel for analysis. The MetaXL (www.epigear.com accessed on 15 October 2021) was used to carry out a meta-analysis. An Inverse Heterogeneity (IVhet) model was used to compute the prevalence estimates with their 95% confidence intervals (CIs). The inverse variance statistic (I 2 index) was used to quantify heterogeneity, and we tested for its significance using Cochran's Q test. The I 2 index was interpreted as no, low, moderate, or high heterogeneity if the value was 0%, ≤25%, 50%, or ≥75%, respectively. Forest plots were generated to show the prevalence of Toxoplasma among the study subjects. Furthermore, subgroup analysis was carried out to assess the mean pooled prevalence estimates according to host types and regions within southern Africa. The risk of publication bias was assessed using the Luis Furuya-Kanamori (LFK) index and funnel plot [97]. The symmetry of the Doi plots was determined using the LFK index and a value within the range of ±1 was considered as symmetrical and classified as the absence of publication bias, while an LFK value within the range of ±2 was considered as minor asymmetry with slight publication bias, and an LFK value outside the range of ±2 was considered as major asymmetry and high publication bias [97].

Conclusions and Recommendation
This study showed that there are limited studies on T. gondii in humans and animals in southern Africa. Considering the limited information on the prevalence of T. gondii in southern African countries, more studies targeting the epidemiology of this parasite in the environment (soil and water), vegetable, food animals, wild animals, and humans (children, pregnant women, immunocompromised, and healthy people) must be conducted to better understand the transmission dynamics in the region. Additionally, there is a need to establish a surveillance system at the wild animals-livestock interface for monitoring transmission between livestock, wildlife, and humans. Furthermore, emphasis should be focused on health education and the preventive measures of toxoplasmosis, which include adequate cooking of meat, washing of fruits and vegetables before eating, and provision of potable water.