Introduction

Schistosoma haematobium is endemic in Delta State, Nigeria, particularly in the Ase River catchment as reported by Ito (2019); Emukah et al. (2012); Onyeneho et al. (2010); Ekwunife et al. (2009); Nwabueze and Opara (2007) where they are intermediately hosted by Bulinus truncatus, B. globosus and B. forskalii. The transmission and focal distribution of schistosomiasis is spatiotemporally restricted to water bodies inhabited by its obligate intermediate host snails and human water contact as documented by Merata et al. (2023); Nwoko et al. (2022a); Ito (2019); Ito and Egwunyenga 2015). Schistosoma haematobium is contracted by humans through exposure to polluted water. The parasite resides in freshwater snails, which discharge cercariae into the water. Upon contact with the contaminated water, the cercariae the intact skin and travel to the blood vessels surrounding the bladder and genital regions, where they develop into adult worms and lay eggs (Ito and Utebor 2023). These eggs are subsequently expelled in urine, contaminating the water anew and perpetuating the transmission cycle. Water sources such as rivers, lakes, and ponds inhabited by infected snails serve as the breeding grounds for this parasitic infection.

Ito (2019); Nduka et al. (2019); and Ito and Egwunyenga, (2015) have highlighted the endemicity of schistosomiasis in over 78 countries and considered its public health importance in the tropics and sub-tropical countries. In Sub-Sahara Africa (SSA), King and Dangerfield-Cha (2008) stated that approximately 436 million individuals are at risk of Schistosoma infections, of which 112 million are infected with S. haematobium. Similarly, Ito and Utebor (2023) and WHO (2013) documented that Nigeria accounts for the largest (29 million) number of infections with 101 million individuals at risk. While schistosomiasis is recognized as a debilitating disease, causing considerable morbidity, it is also estimated that 120 million individuals suffer severe complications of urinary schistosomiasis with an annual mortality estimate of about 200,000 worldwide (Ezeh et al. 2019; WHO 2016).

Schistosomiasis transmission is correlated with freshwater intermediate snail hosts and requires humans contact with cercaria-infested freshwater. The geospatial distribution and transmission pattern of schistosomiasis has been reported by Ezinna et al. (2023) and Amoah et al. (2017) to be determined by abiotic and biotic factors. The persistent transmission of schistosomiasis in Nigeria and other sub-Saharan Africa is influenced majorly by the proximity to water bodies, distribution of snail intermediate hosts, recreational/occupational activities, poverty, poor sanitation, lack/dilapidated sanitary infrastructure, none availability of potable water supply and climate change as opined by Alade et al. (2023); Ito et al. (2023); Ito (2019); Nduka (2019).

The impact of seasons on disease transmission cannot be over-emphasized (Nwoko et al. 2023; Ito and Egwunyenga 2023; Ito et al. 2023; Manyangadze et al. 2021; Amoah et al. 2017). Studies on the geospatial distribution of snails and infectivity are lacking in Delta State, Nigeria making the control and prevention of snail-borne disease difficult in affected communities. Thus, it is important to identify Schistosoma transmission sites and their dynamics in the Ase River catchment Delta State Nigeria for disease intervention and transmission interruptions. Specifically, this study is designed to: (i) Identify the geospatial transmission hotspots of S. haematobium in the Ase River and (ii) Understand snail transmission dynamics and cercarial infection rates. This could enhance an efficient resource allocation for targeted prevention and control interventions in specific transmission foci.

Materials and methods

Description of study area

The Ase River is one of several rivers that drain the recent delta top landforms of the western Niger Delta and it is a coalesce of many tributaries originating from the southern and western slopes of the Asaba Plateau. The tributaries include the Oboshi River, which borders the Ibusa sub-urban municipal on the west, the Atakpo Creek, which borders the settlement on the east; the Nooni River, which runs through Nsukwa and Ogwashi-Uku west; while the tributaries headwaters of the Adofi River are at Ejeme Aniogor and Agbor Aladinma (Fig. 1). These tributaries meet at the southern part of Iselegwu to form the main trunk of Ase River about 292 km in length which shares a flood plain with the River Niger and flows southwards to join the Nikeroga River in the western branch while the eastern branch of the Niger River terminates at the Nun River (Ito 2017). The Ase River further flows through Ashaka, Igbuku, Ibrede, Awah, Lagos-Iyede, Iyede-Ame, Onogboko, Itobi-Ige, Ivrogbo, Ibredeni and Ase (where the river derived its name) communities which correspond to the eleven (11) geospatial sampling sites in this study. Ase is a confluence community with surface and groundwater intrusion from the Niger River. These geospatial study locations were selected based on proximity to human dwellings and human water contact sites. The population size of Ashaka, Igbuku, Ibrede, Awah, Lagos-Iyede, Iyede-Ame, Onogboko, Itobi-Ige, Ivrogbo, Ibedeni, and Ase are 2,118, 14,500, 1518, 1149, 1331, 3046, 1763, 1104, 2217, 1238, and 3217 inhabitants respectively (NPC 2006). These communities are tropical, structurally rural and typified with lowland areas subjected to seasonal flooding interspersed with streams, lakes, canals and rivulets where faecal and other anthropogenic waste are deposited (Fig. 1). Topographically, the study area falls within 10 m at Ase study site to 22 m (at Igbuku) elevation above sea level. Rainfall is about 266.50 and 190.50-mm in the coastal and extreme northern areas respectively with precipitation peaking in July. This study area is characterized by two main seasons: dry season (November—March) and Wet season (April—October). The occupation of the inhabitants of the study area are mainly fishing, farming, snail (Pila ovata) hunting and a few artisans and civil servants. More worrisome and appalling is the reliance of the inhabitants of this catchment’s communities on perennial/intermittent streams/rivulets/rivers for hydration and domestic use.

Fig. 1
figure 1

(Source: Ito et al. 2023)

Physiographic Map of Ase River, showing the study stations in Delta State, Nigeria

Research design

The study employed a longitudinal (Nov. 2020—Oct. 2022) study design in eleven (11) geospatial coordinates (Table 1) to evaluate the spatiotemporal transmission patterns of S. haematobium by Bulinus species in the Ase River catchment using standard scooping/hand-picking techniques and cercaria shedding respectively.

Table 1 Geospatial coordinates and Elevations of selected Study Sites in the Ase River Catchment

Snail sampling techniques and identification

Using Scoop net/dip net, snail samples were collected from eleven (11) geospatial points for identification of transmission sites for a period of twenty-four (24) months. Using forceps and safety gloves, the snails were manually hand-picked and kept in a muslin bag. This was carried out at the Ase River banks and other parts of the Ase River Catchment frequently accessed by humans. The captured snails were taken to the Animal and Environmental Biology Laboratory at Delta State University Abraka where they were further identified using keys of Brown and Kristensen (1993).

Infectivity bioassay and cercaria identification

The snail species collected were first observed in beakers for cercarial shedding. To achieve this, each Bulinus snail was placed in a 20 ml beaker containing 2 ml of distilled water and exposed to light to induce cercarial shedding (Marco and Alan 2012; and Ito and Utebor 2023). After light exposure with evidences of cercariae, 1–2 drops of iodine solution were added to each beaker and left to settle for a minute to allow cercariae staining and sedimentation. The cercariae were identified morphologically using the descriptive identification key of Frandsen and Christensen (1984). Snail found shedding bifurcate cercariae of the human schistosome type was noted and recorded. Snails that tested negative for cercariae shedding during the initial light exposure were re-examined on the twelfth day for Schistosoma infectivity.

Method of data collection

The prevalence of S. haematobium infection in Bulinus snails and the cercarial output were determined and employed to establish the effect of season on transmission patterns. The Infectivity rate of Bulinus species was also calculated thus:

$$\begin{aligned}& \left({\text{a}}\right)\, \text{ Infectivity for study site}\\&\; = \frac{\text{Number of}\, Bulinus\, \text{infected in each Site }}{\text{Number of}\, Bulinus\, \text{collected from that Site}}\\&\quad \times \frac{100}{1} \end{aligned}$$
$$\begin{aligned}& \left({\text{b}}\right)\text{ Monthly Infectivity}\\&\;= \frac{\text{Number of}\, Bulinus\, \text{infected for each month}}{\text{Total Number of}\, Bulinus\, \text{collected in that month}}\\&\quad \times \frac{100}{1} \end{aligned}$$

(Okafor 1990; Rudge et al. 2013).

The monthly and seasonal Transmission Potentials (TP) as developed by Chu and Dawood (1970) were adopted as shown below:

$$\begin{aligned}& \left({\text{a}}\right)\, \text{Monthly Transmission Potential}\\&\;= \frac{\text{Monthly Number of infected snails}}{\text{The sum of infected snails in that group}}\\&\quad \times \frac{100}{1} \end{aligned}$$

(b) Seasonal T.P. = Addition of the T.P.s of the months in a season.

Reliability and validity test

Prior to snail sample collection, a pilot study to identify urinary Schistosoma hot-spot was conducted among the inhabitants of the geospatially selected sites to ascertain the prevalence of S. haematobium. The few Bulinus samples collected during the pilot study also showed evidences of cercariae shedding when expose to light. This was achieved by employing a standardized protocol to assess and observe snail-related factors, such as transmission dynamics and infection rates. This iterative approach, ensures a consistent and reliable assessment of geospatial transmission hotspots and snail transmission dynamics.

Data analysis

Statistical analysis was conducted to compare the variations in infection prevalence among different Bulinus species using chi-square and ANOVA. A significance level of p = 0.05 was set and values below this threshold were deemed significant and indicative of a potential risk factor. The obtained seasonal and communal prevalence values were subjected to rigorous examination using specialized software such as Graph-Pad Prism and PAST version 4.11. The software facilitated the application of suitable statistical tests, such as chi-square (χ2) and ANOVA, to identify noteworthy differences in the data, ensuring robust and reliable analysis.

Results

Monthly and interannual prevalence of S. haematobium in Bulinus vectors

Results of the monthly and interannual prevalence variations of S. haematobium infections in the Bulinus species are presented in Table 2. Generally, a total of 972 (41.45%) of the 2345 Bulinus snails collected were infected with S. haematobium (Table 2). The monthly prevalence of S. haematobium infections in Bulinus varied throughout the study period with the highest prevalence of 69.81 and 75.00% observed in February and March of first and second year respectively. Statistically, significant differences (P =  < 0.0001; F = 23.11; df = 11) in the monthly prevalence of S. haematobium in the intermediate host were noted in the first- and second-year with 95.68% variance. Table 2 also revealed that the second study year had the highest prevalence of 43.17% compared to the first year (39.51%). Chi-square showed that the association between the study years was strongly significant (χ2 = 23.57; df = 11; p = 0.015).

Table 2 Monthly/Interannual prevalence of S. haematobium infections in Bulinus species collected during the period of study

The Principal Component Analysis (PCA) reveals that the monthly/interannual prevalence of S. haematobium infections in Bulinus species in the Ase River catchment is primarily driven by PC1 (with an eigenvalue of 945.84), which is strongly influenced by January and February (Fig. 2). PC2 captured a smaller amount of variance and is most strongly associated with November and December with an eigenvalue of 31.01 (Fig. 2). Overall PCA reveals that the infection prevalence in the Ase River catchment is primarily characterized by strong seasonal patterns. These findings provide insights into the patterns and relationships of Bulinus infectivity between different Months.

Fig. 2
figure 2

PCA Cartesian Scatter Plot of the monthly/interannual Bulinus infectivity in the Ase River catchment

Monthly/annual Bulinus species-specific infectivity

Table 3 displays the monthly and annual species-specific infectivity in different species of Bulinus snails. B. globosus had the highest annual infection rates of 53.86 and 44.44% in the first and second year, while B. senegalensis had the lowest rates of 26.84 and 21.78% in both years respectively. In specific months, B. globosus reached a peak infection rate of 77.05% in December of the second year, while B. truncatus had its highest rate of 76.47% in February of the first year (Table 3). The prevalence of S. haematobium infection in B. truncates (P =  < 0.0001), B. globosus (P = 0.0014) and S. senegalensis (P = 0.0064) were statistically significant (P < 0.05) in the monthly prevalence with a variance of 96.14, 87.49 and 78.03% respectively but not in the annual prevalence (P > 0.05).

Table 3 Monthly/annual prevalence of S. haematobium infections in Bulinus specific-species collected during the period of study

Geospatial prevalence of S. haematobium in Bulinus species in ase river catchment

The geospatial prevalence of S. haematobium infections varied among the three snail species (B. truncatus, B. globosus and B. senegalensis) across different sites during the study period (Table 4). The results also indicated variations in infection rates within and between years for each snail species. Overall, the infectivity increased slightly from the first year (36.59%) to the second year (40.51%) for B. truncates (Table 4). The highest infection rate for B. truncatus was observed at the Iyede-Ame study site with a value of 63.33 and 59.22% for the first and second years respectively. Similarly, B. globosus showed a significant increase in infection rates from the first year to the second year, with the highest rates observed at the Onogboko and Lagos-Iyede study sites in both years. On the other hand, B. senegalensis exhibited the lowest infection rates among the three snail species, with a decrease in infection rates from first-year to second year. However, the study sites' prevalence of S. haematobium infection in B. truncates (P = 0.0012), B. globosus (P = 0.0009) and S. senegalensis (P = 0.0172) were statistically significant (P < 0.05) with 89.26, 87.55 and 78.89% variance respectively.

Table 4 Geospatial prevalence of S. haematobium infections in species-Specific Snail Vector During the Period of Study

Species-specific transmission potential (T.P) of S. haematobium in Ase River

Table 5 shows the T.P of S. haematobium by different Bulinus species across months and years. B. truncatus consistently had the highest T.P, followed by B. globosus and B. senegalensis. In first-year, B. truncatus had the highest T.P in January (23.66%) and December (16.03%), while B. globosus surpassed B. senegalensis in January (22.35%) and May (17.50%). In the second year, B. senegalensis showed higher T.P in January (38.64%) and February (20.45%), while B. globosus increased in May (10.14%). Statistical analysis confirmed significant differences (P < 0.05) in T.P among the species. However, yearly transmission potentials did not differ significantly.

Table 5 Monthly/Interannual Species-Specific Transmission Potential of S. haematobium by Bulinus species in the Study Area

Seasonal species-specific T.P of S. haematobium vectors in Ase River catchment

The results in Table 6 revealed distinct seasonal variations in the T.P of S. haematobium by different Bulinus species. During the dry season in November, B. truncatus exhibited a T.P of 7.44%, followed by B. globosus (12.73%) and B. senegalensis (5.66%). However, during the wet season, the T.P increased for all three species (Table 6). Specifically, the T.P of B. senegalensis (12.90%) higher than B. truncatus (12.20%) and B. globosus (11.66%) respectively. Monthly analysis of transmission potential for Bulinus species revealed distinct trends. In the dry season, B. truncatus, B. globosus and B. senegalensis had their highest T.P in January (Table 6). During the wet season, B. truncatus and B. globosus peaked in July, while B. senegalensis reached its peak T.P in May. Significant differences were observed between seasons (P < 0.05), but not across the years for all three species of Bulinus.

Table 6 Seasonal species-specific T.P of S. haematobium vectors in Ase River catchment

Seasonal T.P of S. haematobium infections by Bulinus species in Ase River

The results presented in Table 7 revealed the T.P of S. haematobium infections in Bulinus species from the Ase River Catchment. In the dry season, transmission potentials varied across the months, with November having the highest monthly transmission potential in first-year (8.97%) and subsequent months showing a decreasing trend. Second-year had lower T.P overall. The wet season also showed variations, with June and July having the highest monthly T.P in the first year (9.66 and 13.10% respectively), while the second year had higher T.P overall. The cumulative seasonal T.P for the two years was 55.04% in the dry season and 44.96% in the wet season. Generally, the seasonal T.P of the two years was 55.04 and 44.96% in dry and wet seasons respectively.

Table 7 Monthly and Seasonal T.P of S. haematobium infections in Bulinus species Collected in Ase River Catchment, Delta State, Nigeria

Vector-specific geospatial T.P of S. haematobium in Ase river catchment

Table 8 provide insights into the T.P of S. haematobium infections among different snail species across various locations and years. Ashaka consistently exhibited moderate T.P values for all three vector species. Igbuku had higher T.P values for B. truncatus in the first and second year but decreased in the second year. Conversely, Igbuku had low T.P values for B. globosus and B. senegalensis, except for a higher T.P value of B. globosus in First-year. Ibrede exhibited higher T.P values for B. truncatus and B. senegalensis in the second year. Awah had minimal T.P values for B. truncatus and B. senegalensis in the first year, indicating low T.P. The in-land communities (Lagos-Iyede, Iyede-Ame, Onogboko and Itobi-Ige) consistently maintain a very high T.P of S. haematobium infection. Specifically, Lagos-Iyede showed high T.P values for all three vector species, ranging from 16.43 to 20.45% in both years. Iyede-Ame exhibited the highest T.P values among all locations, particularly for B. truncatus and B. senegalensis in Second-year (29.46 and 34.09%, respectively). Onogboko displayed a mixed pattern, with B. truncatus having the highest T.P value of 23.66% in first-year, while B. senegalensis had its peak T.P value of 34.09% in in the second year (Table 8). Generally, Iyede-Ame had the highest geospatial T.P of 21.71% cercaria while Awah recorded 0.41%. The study sites exhibited significant differences (P < 0.05). However, no significant difference (P > 0.05) was noted in the yearly vector species T.P.

Table 8 Vector-Specific Geospatial Transmission of Potentials of S. haematobium infections in Ase River Catchment, Delta State, Nigeria

Discussion

Understanding the schistosomiasis transmission dynamics is crucial for effective control. Ito and Egwunyenga (2015) and Kristensen et al. (2013) opined that this knowledge helps in developing integrated control strategies for managing snail populations and reducing the spread of snail-borne diseases. This study provides important insights into the monthly and annual transmission dynamics of S. haematobium infections in the Bulinus species. The overall infectivity of 41.45% in the Bulinus species indicates a significant concern of schistosomiasis in the study area. These findings corroborate those of Nwoko et al. (2023), Pam et al. (2021), N’Guessan et al. (2017) and Mohammed (2017) in South Africa, Mauritania and Sudan respectively. Interannually, the first- and second-year recorded a prevalence of 39.51 and 43.17% respectively. The observed variations in prevalence indicate the presence of climatic and seasonal fluctuations in disease transmission (Ito and Egwunyenga 2023). The differences in prevalence between first-year and second-year could also be attributed to various factors such as climatic conditions, snail population dynamics and water quality reported by Nwoko et al. (2023) and Mareta et al. (2023). Additionally, the variations in prevalence rates between the first- and second year suggest potential temporal changes in disease transmission dynamics, which should be considered when designing long-term control strategies.

The higher prevalence from January to March suggests a potential peak in transmission, while the lower prevalence in September and October indicates reduced transmission during that period. This observation is contrary to the assertions of Amoah et al., (2017) who documented higher snail infectivity in September and October. The variations in snail infectivity across different months may be attributed to several factors, including climatic/environmental conditions, water quality and snail population dynamics as opined by Abe et al. (2017), Ito et al. (2023), and Ezinna et al. (2023). The higher prevalence rates observed during certain months such as February, highlight the need for targeted interventions during specific periods of the year.

Three Bulinus species (B. truncatusB. senegalensis and B. senegalensis) known to be intermediate hosts for S. haematobium were collected from the Ase River catchment, Delta State, Nigeria. These snail assemblages are similar to a preliminary report by Fryer and Probert (1988) and Abe et al. (2017). The Bulinus-species-specific infectivity by S. haematobium fluctuated throughout the study period. For example, B. globosus had the highest infection rate in December of Second-year, with 77.05% of the examined snails infected. Contrarily, B. truncatus had its highest infection rate in February of the first year, with 76.47% of the snails infected. These variations in monthly prevalence may be attributed to various factors, such as environmental conditions and the presence of suitable habitats for snails and parasites (Mereta et al. 2019; Nwoko et al. 2023). Overall study revealed varying levels of S. haematobium infection rates among the three snail species and across different stations. These findings suggest that different snail species may have different transmission dynamics and should be targeted for specific control interventions.

This study highlights the temporal variation in transmission potential among the Bulinus species. The highest transmission potentials generally occurred during the dry seasons, which is consistent with the cercariae infectivity period of S. haematobium and its intermediate hosts. Contrarily, in the Ase River catchment, the dry season witness a low relative Bulinus abundance compared to the wet season as documented by Ito et al. (2023). B. truncatus consistently showed the highest transmission potential, suggesting its importance in maintaining the parasite's life cycle. B. globosus and B. senegalensis also played significant roles, albeit with lower transmission potentials. The observed variations in the transmission potential of S. haematobium by different Bulinus species can be attributed to several factors such as change in temperature and seasons (dry and wet seasons). Furthermore, the seasonal variations in transmission potential suggest the existence of specific environmental conditions or behavioral patterns that contribute to increased human exposure during certain months (Ito and Egwunyenga 2023; Ito et al. 2023; Manyangadze et al. 2021).

Previous studies by Posa and Sodhi (2006), Mereta et al. (2023), Ito et al. (2023) have established the impact of environmental factors such as vegetation, water quality and human water contact pattern on snail abundance and infectivity. The differences in transmission potential between years indicate the influence of environmental factors and population dynamics of the snail intermediate hosts (Nwoko et al. 2022a). The findings from this study highlight the dynamic nature of the species-specific transmission potential of S. haematobium by different Bulinus species in the study area. The variations observed in transmission potential between different months and years suggest the influence of seasonal factors, environmental conditions, increased snail activity, potential interactions between the parasite and intermediate host species and higher parasite burdens in humans (Ito 2019; Nwoko et al. 2022b).

The results demonstrate a clear distinction in transmission potentials between the dry and wet seasons. The dry season exhibited higher transmission potentials in the initial months, gradually decreasing towards the end. This may be attributed to factors such as temperature, rainfall patterns and water availability. The dry season in the study locations of Ase River is characterized by an increase in temperature. Elevated temperatures increase snail metabolic rate, fecundity and feeding frequency, reducing the duration of the development periods and increasing the number of generations per year as reported by Ito et al. (2023); De La Rocque et al. (2008) and Kristensen et al. (2001).

To control S. haematobium in the Ase River Catchment, targeted interventions are crucial. These include chemical control of snails through molluscicidal treatments and habitat modifications; health education programs can raise awareness and promote safe water practices; Mass Drug Administration (MDA) with praziquantel, especially for school-age children, is vital; improved sanitation and biological control methods can reduce transmission; seasonal interventions, community engagement, and robust monitoring systems are key for sustainable success. Adopting these measures, tailored to the area, will effectively combat schistosomiasis in the Ase River Catchment.

Conclusion

The results showed 41.45% infectivity of S. haematobium in snail vectors and are of a public health concern as water contact activities in the Ase River catchment can lead to an exponential prevalence of urogenital schistosomiasis. This study also revealed the geospatial hotspots and the dynamic nature of the species-specific transmission potential of S. haematobium by B. truncatus, B. globosus and B. senegalensis. The high transmission potential by B. truncatus and B. senegalensis in the dry season indicate their significance as potential contributors to the overall transmission of S. haematobium in the Ase River catchment. The variations observed in transmission potential between different months and years suggest the influence of seasonal factors, environmental conditions and potential interactions between the parasite and intermediate host species. These results demonstrate a clear distinction in transmission potentials between the dry and wet seasons which calls for the need for continuous monitoring and targeted interventions to control schistosomiasis transmission in the study area.