Prevalence of Schistosoma mansoni and S. haematobium in Snail Intermediate Hosts in Africa: A Systematic Review and Meta-analysis

Background Schistosomiasis is caused by Schistosoma mansoni and S. haematobium in Africa. These schistosome parasites use freshwater snail intermediate hosts to complete their lifecycle. Varied prevalence rates of these parasites in the snail intermediate hosts were reported from several African countries, but there were no summarized data for policymakers. Therefore, this study was aimed to systematically summarize the prevalence and geographical distribution of S. mansoni and S. haematobium among freshwater snails in Africa. Methods Literature search was carried out from PubMed, Science Direct, and Scopus which reported the prevalence of S. mansoni and S. haematobium among freshwater snails in Africa. The pooled prevalence was determined using a random-effect model, while heterogeneities between studies were evaluated by I2 test. The meta-analyses were conducted using Stata software, metan command. Results A total of 273,643 snails were examined for the presence of S. mansoni and S. haematobium cercaria in the eligible studies. The pooled prevalence of schistosome cercaria among freshwater snails was 5.5% (95% CI: 4.9–6.1%). The pooled prevalence of S. mansoni and S. haematobium cercaria was 5.6% (95% CI: 4.9–6.3%) and 5.2% (95% CI: 4.6–5.7%), respectively. The highest pooled prevalence was observed from Nigeria (19.0%; 95% CI: 12.7–25.3%), while the lowest prevalence was reported from Chad (0.05%; 95% CI: 0.03–0.13). Higher prevalence of schistosome cercaria was observed from Bulinus globosus (12.3%; 95% CI: 6.2–18.3%) followed by Biomphalaria sudanica (6.7%; 95% CI: 4.5–9.0%) and Biomphalaria pfeifferi (5.1%; 95% CI: 4.1–6.2%). The pooled prevalence of schistosome cercaria obtained using PCR was 26.7% in contrast to 4.5% obtained by shedding cercariae. Conclusion This study revealed that nearly 6% of freshwater snails in Africa were infected by either S. haematobium or S. mansoni. The high prevalence of schistosomes among freshwater snails highlights the importance of appropriate snail control strategies in Africa.


Introduction
Schistosomiasis is one of the neglected tropical diseases (NTD) endemic in 78 countries and infects more than 229 million peoples in tropical and subtropical regions [1]. More than 90% of these cases are concentrated in Africa [2,3]. e burden of the disease is even more severe in sub-Saharan Africa. Poor environmental sanitation and suitability of the climate conditions for snail intermediate host contribute to the high endemicity of the region. Schistosomiasis ranks second next to malaria from parasitic infection in terms of socioeconomic and health impact in tropics [4].
Human schistosomiasis is caused by Schistosoma mansoni, S. haematobium, S. japonicum, S. intercalatum, S. mekongi, S. malayensis, and S. guineensis [5][6][7]. Among these species, S. mansoni, S. haematobium, and S. japonicum are the major causes of human schistosomiasis globally [8]. Schistosoma mansoni and S. haematobium are widely distributed and the dominant cause of human schistosomiasis in Africa [5]. e endemicity of the disease in the region is linked with the availability of freshwater snail intermediate hosts.
About 350 species of freshwater snails are known to be medically or veterinary important [9]. Among these diverse snails, Biomphalaria, Bulinus, and Oncomelania snails [4] are the dominant snail genera that are involved in the transmission of human schistosomiasis. e Biomphalaria genus consists of B. pfeifferi, B. glabrata, B. sudanica, B. straminea, B. tenagophila, B. alexandarina, and B. choanomphala [10]. Biomphalaria snails serve as the intermediate host for S. mansoni, which is responsible for intestinal and hepatic schistosomiasis. Biomphalaria pfeifferi is the most common and widely distributed snail intermediate host for S. mansoni in Africa.
Bulinus consists of 37 recognized species, which is grouped mainly into four species groups, namely, Bulinus africanus, B. forskalii, B. truncates/tropicus, and B. reticulatus [10,11]. Bulinus snails serve as intermediate hosts for S. haematobium, which is responsible for urinary schistosomiasis. Oncomelania snails consist of only a few species mainly reported from Asia. e most common snail intermediate host for S. japonicum is Oncomelania hupensis, which is found in China, the Philippines, Indonesia, and also Japan [12]. e prevalence of human schistosomiasis is varied greatly in African countries depending on the level of environmental sanitation and the suitability of the area for the snail intermediate hosts, as well as the type of snail in the area. Similarly, the prevalence of schistosomes cercaria in snail intermediate hosts is varied in different locations within the same country and also from country to country in Africa. Several epidemiological studies are available on the types and prevalence of human infecting schistosomes among snail intermediate hosts in Africa. However, up to this time, there has not been any single estimate of the prevalence of S. mansoni and S. haematobium in snail intermediate hosts in Africa that could be used by African policymakers and international organizations working on the prevention and control of schistosomiasis in the continent. erefore, this study aimed to provide summarized data on the prevalence and geographical variations of S. mansoni and S. haematobium cercaria among freshwater snails in Africa.

Search Strategies.
Relevant literature was systematically searched from online public databases (PubMed Central, Science Direct, and Scopus) using the following key-words: "Schistosomiasis" OR "S. mansoni" OR "S. haematobium" OR "parasitological study" OR "schistosome intermediate host" OR "freshwater snails" OR "malacological survey" OR "Biomphalaria snails" OR "Bulinus snails" AND "Africa". e systematic review and selection of relevant literature were conducted according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analysis) guidelines [13] (Table S1).

Inclusion and Exclusion Criteria. Literature published in
English language from 1979 to June 2020 were extracted from online public databases. Original articles reporting the prevalence of human schistosome cercariae in freshwater snails in African countries were included in the analysis. e eligibility for the inclusion of a study in our analysis had to fulfill the following criteria: (a) it was published in English, (b) the study was carried out in Africa, (c) the number of examined and infected snails with either S. mansoni or S. haematobium were clearly stated, and (d) snail species were identified at least to a genus level. Studies that reported nonhuman schistosome and other trematodes species were excluded from the analysis. Besides, review articles and meta-analysis were excluded from the analysis.

Data Extraction Protocol.
e data extraction protocol was prepared and evaluated by all authors. From each published article, we extracted the following information: author information, year of publication, study country, snail species, number of snails (collected, examined and infected), the prevalence of snail infection, the type of schistosomes reported, and methods of schistosome's detection.

Quality of Individual Study and Assessment of Bias.
e quality of studies included in the meta-analysis was assessed by using the Newcastle-Ottawa quality assessment scale (NOS) proposed by Wells et al. [14] (Text S1). e quality assessment tool consists of three parts. First, the selection of study groups graded on a scale containing five stars; mainly deals with methodological qualities of individual study. Second, comparability of groups graded on a scale containing two stars; deals with comparability of studies based on design and analysis.
ird, outcomes graded on a scale containing three stars, mainly focused on the assessment of the outcome and statistical analysis. Two authors (TH and EN) independently assessed the quality of individual study, and disagreement was solved by a discussion with the third author (AM). e overall quality of the individual study was categorized as high quality (≥8 stars), moderate quality (6-7 stars), and low quality (≤5 stars) by the total number stars obtained as described elsewhere [15].

Publication Bias across Studies.
e risks of publication bias across studies were assessed using funnel plot symmetry qualitatively. Egger's and Begg's test were used to determine the presence of publication bias across studies quantitatively.

Data Analysis.
We used a forest plot to estimate the overall pooled effect size with their 95% confidence interval (CI). e heterogeneity among studies used for this metaanalysis was evaluated using the I 2 test [16]. An I 2 -value lower than 25%, between 25% and 50%, and above 50% was regarded as low, moderate, and high heterogeneity, respectively [17]. Because of the high heterogeneity observed among the studies included in the meta-analysis, we used a random-effect model at 95% CI. To sort out the cause of heterogeneity, we used a subgroup analysis, sensitivity test, and meta-regression analysis. e data analysis was conducted using Stata software (version 14, STATA Corp College Station, TX), "metan" command.

Search Results and Eligible Studies.
A total of 2,995 relevant studies were screened from online public databases. Out of these studies, 976 articles were removed due to duplications while 1884 articles were excluded based on title and abstract screening. e remaining 135 full-text articles were assessed for eligibility. Of these, a total of 84 articles were excluded from the analysis based on specific exclusion criteria, and the remaining 51 articles were selected for this meta-analysis ( Figure 1).

Risk of Bias within Studies.
e Newcastle-Ottawa quality assessment scale indicated that there was no bias within studies. e individual study included in this review was moderate to high-quality score as indicated in Table 1.

Prevalence of S. mansoni and S. haematobium Cercaria
Among Freshwater Snails. A total of 273, 643 snails from Biomphalaria and Bulinus genera were examined for the presence of S. mansoni and S. haematobium cercaria in the 51 eligible studies, respectively ( Table 2). Out of these snails, 8,682 of them were infected by either S. mansoni or S. haematobium. e prevalence of schistosome cercaria in the individual study ranged from 0.05% to 58.03% with substantial heterogeneity across studies within and across countries. e pooled prevalence of schistosome cercaria among freshwater snails was 5.5% (95% CI: 4.9-6.1%, I 2 � 99.4%, and p < 0.001) ( Figure 2).
Twelve snail species from Biomphalaria and Bulinus snails were reported in the eligible articles used for this metaanalysis (  Table 2). e studies included in this meta-analysis used shedding of cercariae and PCR-based detection of schistosomes from snail tissue. e pooled prevalence of schistosomes obtained by shedding of cercariae was 4.5% (95% CI: 3.9-5.1%) in contrast to 26.7% (95% CI: 10.5-43.0%) obtained by PCR techniques ( Figure 6).

Publication Bias across Studies.
e funnel plot symmetry demonstrates the presence of publication bias among studies included in this meta-analysis ( Figure 7). Similarly, Egger's test results (p � 0.02) and Begg's test (p � 407) confirm the presence of publication bias among studies.

Metaregression Analysis and Sensitivity Test.
ere were clear heterogeneities across studies included in this metaanalysis. We performed a meta-regression analysis to identify the sources of heterogeneity across studies. e metaregression analysis showed that methods of schistosome detection from snails (regression coefficient: 2.55, 95% CI, 1.17-5.54, p � 0.02) might be the source of heterogeneity. e country of study (regression coefficient: 0.99, 95% CI, 0.89-1.11, p � 0.97), years of publication (regression coefficient: 1.41, 95% CI, 0.91-2.18, p � 0.12), snail genus (regression coefficient: 1.19, 95% CI, 0.56-2.55, p � 0.65), and snail species (regression coefficient: 0.92, 95% CI, 0.8 p � 0.65 p � 0.14) did not contribute for the heterogeneity. Besides, a sensitivity analysis was performed by recalculating the pooled prevalence by sequentially removing one-by-one to assess the effect of individual studies to overall effect. e pooled prevalence remained stable, and the result was not driven by individual studies included in the meta-analysis.

Discussion
Schistosomiasis is the second leading cause of infectious diseases next to malaria in Africa. Despite intensive efforts to tackle schistosomiasis, the prevalence is still unacceptably high in many African countries. e control and prevention strategies mainly rely on treatment of infected cases and mass drug administration of school-aged children. In many African countries, snail control strategies are not routinely implemented and sometimes considered as old-fashion approaches [70] despite their vital contributions to the elimination of schistosomes witnessed from Japan, Iran, and Tunisia [71][72][73]. In addition, enormous progresses have been observed in the elimination program from Morocco, Oman, Lebanon, and Caribbean Islands [73,74]. e intensity and prevalence of schistosomes' infection among freshwater snails are scarce from many African countries. Summarized information about the prevalence of schistosomes among snail intermediate hosts is important for policymakers to give better attention to snail control strategies in Africa. e overall pooled prevalence of schistosome cercaria was nearly 6% among freshwater snails in Africa. is finding is slightly lower than 9% reported from freshwater snails in Brazil [75]. In contrast, a lower prevalence of infected snails was observed from Indonesia [76] and Brazil [77]. ese differences might be associated with prevalence and intensity of schistosome infection in the community, the level of environmental sanitation, suitability of the climate for the snails, level of existing snail control strategies, level of human exposure to open surface water, methods of schistosome detection, and seasons of snail collection and examination.
e highest pooled prevalence of schistosome cercaria among freshwater snails was observed in Nigeria followed by Ethiopia. In contrast, low prevalences of schistosomes were observed from Benin, Burundi, and Chad. e high prevalence of schistosomes among snail species in Nigeria and Ethiopia might be associated with the high prevalence of schistosomes in the community. e prevalence of schistosomiasis could reach as high as 90% in Ethiopia [78] and 94% in Nigeria [79,80]. In addition, the difference in the level of environmental sanitation and the suitability of the area for the intermediate host, as well as the types of snail species in the area, may contribute for the difference in infection of snails across countries. Moreover, larger numbers of studies were reported from these two countries. Out of 51 studies, 17 (23.9%) studies included in this review were from the two countries.
Several species of freshwater snails that potentially serve as intermediate hosts for S. mansoni and S. haematobium have been recently reviewed [81]. Biomphalaria and Bulinus snails are the common and widely distributed intermediate hosts for schistosomes in Africa. e twelve snail species observed in this review belonged to either Biomphalaria or Bulinus genus.
Biomphalaria snails are well-known intermediate hosts of S. mansoni in Africa. is review showed that 5.6% of Biomphalaria snails were infected by S. mansoni in Africa. e highest pooled prevalence of S. mansoni was observed from Tanzania followed by Ethiopia and Nigeria, while low pooled prevalence was reported from Benin, Burundi, and Chad. ese reported differences might be associated with       Biomphalaria pfeifferi were the most common snails infected by schistosome parasite. About 40% of the eligible studies included in this meta-analysis reported B. pfeifferi. e role of B. pfeifferi as an intermediate host of S. mansoni varied from country to country. Biomphalaria pfeifferi is the sole intermediate host for S. mansoni in Côte d'Ivoire [83] and Senegal [84] and the dominant intermediate host in many other African countries [85][86][87][88].
Biomphalaria sudanica was the second common Biomphalaria snails that serve as an intermediate host for S. mansoni as observed in this study. Biomphalaria sudanica is an intermediate host for S. mansoni in Kenya [47] and Tanzania [19,89]. Biomphalaria sudanica is also reported from Ethiopia but limited to around Lake Ziway [90] and Tikur Wuha [87].
Biomphalaria alexandrina was the third common intermediate host for S. mansoni observed in this study. However, its contribution as an intermediate host for S. mansoni is restricted in geographical distribution, mainly reported from Egypt [91][92][93]. Biomphalaria choanomphala was another intermediate host for Schistosoma mansoni reported from Uganda [64] and Tanzania [61]. is snail species is widely distributed around Lake Victoria, which is divided among three countries (Kenya, Tanzania, and Uganda). Biomphalaria choanomphala is also reported from Kenya [94], but the infection intensity was not determined.  [42] and Côte d'Ivoire [23,24].
Detection of schistosome infection is determined by shedding of cercariae and/or PCR based approaches from snail tissue. ere was a significant difference in the pooled  [97]. is difference is associated with the sensitivity of PCR to detect schistosome infection from snail tissue. Cercarial shedding is suffered by several limitations such as low parasite burden; snails may not shed cercariae during the prepatent period; time-consuming, and labour-intensive [98]. PCR-based detection of schistosome infection from snails is generally rapid, efficient, sensitive, and cost-effective for large-scale detection [99,100]. Despite the ongoing schistosomiasis control strategies in many African countries, the pooled prevalence of schistosomes among freshwater snails had increased over time from 1.3% (before 2000) to 8.3% between 2016 and 2020. is might be attributed to the large number of epidemiological     countries (varied from 1 study to 11 studies in a country). ird, the studies included in this review were published in English, and we did not include studies published in other languages such as French due to language barriers and translation-related challenges. Fourth, most of the studies included in this review used cercarial shedding rather than     PCR for the detection of schistosome infection. Cercarial shedding is less sensitive for the detection of schistosomes due to its inherent limitation that may result in low prevalence of schistosomes among infected snails in Africa. e pooled prevalence of schistosome cercaria among freshwater snails of Africa observed in this review might be below the actual infection intensity. Fifth, this review showed that there was high heterogeneity across   studies included in this review. ese might be associated with study design, seasons of snail collection, method of detection, and variation of endemicity of schistosomes across countries.

Conclusions
is review showed that nearly 6% of freshwater snails in Africa were infected by either S. haematobium or S. mansoni. e pooled prevalences of schistosome cercaria among freshwater snails have increased in the recent years in many African countries. e higher and increased trends in the prevalence of schistosomes among freshwater snails highlight the need for appropriate snail control strategies in the region. Policymakers should give better attention about integration of snail control strategies to the ongoing treatment-based prevention of schistosomiasis in Africa.

CI:
Confidence interval I 2 : Inverse variance index NOS: Newcastle-Ottawa quality assessment scale PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-analysis PCR: Polymerase chain reaction NTD: Neglected tropical disease.

Data Availability
All the datasets are included in the manuscript.

Conflicts of Interest
e authors declare that there are no conflicts of interest.

Authors' Contributions
TH was involved in the design, conducted the review, data analysis, and interpretation of the findings, and drafted the manuscript. EN and AM were involved in drafting the manuscript, reviewing, and editing the manuscript.