Introduction
Schistosomiasis is a major neglected tropical disease disproportionately affecting sub-Saharan African countries, with 90.0% of the global disease burden occurring in this region (WHO, 2023). In Tanzania, the overall prevalence of schistosomiasis is 51.5% (Mazigo et al., Reference Mazigo, Nuwaha, Kinung'hi, Morona, de Moira, Wilson, Heukelbach and Dunne2012), and among school-aged children (SAC) it is reported to be 53.5% (Kavana, Reference Kavana2018), reaching up to about 80.0% in the northwestern zone around Lake Victoria (Munisi et al., Reference Munisi, Buza, Mpolya and Kinung'hi2016; Bakuza et al., Reference Bakuza, Denwood, Nkwengulila and Mable2017). However, current prevalence estimates do not include the northern region of Tanzania, including Mwanga district in the Kilimanjaro region. This district has been known to be endemic for both Schistosoma haematobium and Schistosoma mansoni (Poggensee et al., Reference Poggensee, Krantz, Nordin, Mtweve, Ahlberg, Mosha and Freudenthal2005). The population is at high risk of schistosomiasis possibly due to the presence of the intermediate snail host (Bulinus and Biomphalaria) as well as irrigation schemes, which are the conducive environment for the transmission the Schistosoma spp. The presence of the hydroelectric dam known as ‘Nyumba ya Mungu’ (Fig. 1) which ensures a constant water supply to the surrounding villages for irrigation contributes to the continues transmission of schistosomiasis in Mwanga district (Poggensee et al., Reference Poggensee, Krantz, Nordin, Mtweve, Ahlberg, Mosha and Freudenthal2005).
Map of Mwanga district Tanzania showing water sources, five primary schools and irrigation schemes
In Tanzania, including the Kilimanjaro region, mass drug administration (MDA) of praziquantel has been the major strategy to reduce the burden of schistosomiasis and has been organized annually since 2004 among SAC who are at the highest risk of infection (Mwanga et al., Reference Mwanga, Kinung'hi, Mosha, Angelo, Maganga and Campbell2020). The need to assess the success of MDA has been highlighted by the World Health Organization and tools to enhance strategic guidance for schistosomiasis control program in Tanzania have equally been reported(Clements et al., Reference Clements, Lwambo, Blair, Nyandindi, Kaatano, Kinung'hi, Webster, Fenwick and Brooker2006; WHO, 2022). The most recent data on the prevalence status of schistosomiasis among SAC in Mwanga district are from 2005, indicating a prevalence ranging from 33.5 to 70.0% (Poggensee et al., Reference Poggensee, Krantz, Nordin, Mtweve, Ahlberg, Mosha and Freudenthal2005). Conventional microscopy is the reference method to diagnose schistosomiasis in endemic settings and involves the detection of Schistosoma eggs in feces or urine, depending on the species. However, microscopy requires experienced, well-trained technical personnel, is considered a time-consuming, laborious method and has limited sensitivity in low-intensity infection settings (Hoekstra et al., Reference Hoekstra, Madinga, Lutumba, van Grootveld, Brienen, Corstjens, van Dam, Polman and van Lieshout2022c). Furthermore, the availability and access to microscopy is challenging in many rural areas in Tanzania due to a lack of trained personnel and appropriate infrastructure.
Low-cost, user-friendly rapid tests could overcome such issues, but the accuracy of available point-of-care (POC) tests to determine the prevalence of schistosomiasis in regions co-endemic for S. haematobium and S. mansoni needs to be determined. The POC test for detecting Schistosoma Circulating Cathodic Antigen (POC-CCA) in urine has been endorsed by the WHO as an alternative to conventional microscopy, in particular for the diagnosis of S. mansoni infections (WHO, 2022). It requires minimal training and has been validated in several field settings endemic for intestinal schistosomiasis (Ochodo et al., Reference Ochodo, Gopalakrishna, Spek, Reitsma, van Lieshout, Polman, Lamberton, Bossuyt and Leeflang2015; Bärenbold et al., Reference Bärenbold, Garba, Colley, Fleming, Haggag, Ramzy, Assaré, Tukahebwa, Mbonigaba, Bucumi, Kebede, Yibi, Meité, Coulibaly, N'Goran, Tchuem Tchuenté, Mwinzi, Utzinger and Vounatsou2018). Another easy-to-use rapid test is the microhaematuria dipstick test for the detection of haematuria, which has been shown to be strongly associated with urogenital schistosomiasis, although it is considered nonspecific (Emukah et al., Reference Emukah, Gutman, Eguagie, Miri, Yinkore, Okocha, Jibunor, Nebe, Nwoye and Richards2012). A quantitative Up-Converting reporter Particle Lateral Flow (UCP-LF) test detecting the genus-specific Schistosoma Circulating Anodic Antigen (CAA) is a highly accurate test to detect active infection of all Schistosoma species in urine or serum (Corstjens et al., Reference Corstjens, de Dood, Knopp, Clements, Ortu, Umulisa, Ruberanziza, Wittmann, Kariuki, LoVerde, Secor, Atkins, Kinung'hi, Binder, Campbell, Colley and van Dam2020a). Although it requires a more advanced laboratory infrastructure, it has been shown to be 100% specific and can reach a sensitivity to detect single-worm infections (Corstjens et al., Reference Corstjens, de Dood, Knopp, Clements, Ortu, Umulisa, Ruberanziza, Wittmann, Kariuki, LoVerde, Secor, Atkins, Kinung'hi, Binder, Campbell, Colley and van Dam2020; Langenberg et al., Reference Langenberg, Hoogerwerf, Koopman, Janse, Kos-van Oosterhoud, Feijt, Jochems, de Dood, van Schuijlenburg, Ozir-Fazalalikhan, Manurung, Sartono, van der Beek, Winkel, Verbeek-Menken, Stam, van Leeuwen, Meij, van Diepen, van Lieshout, van Dam, Corstjens, Hokke, Yazdanbakhsh, Visser and Roestenberg2020; Hoekstra et al., Reference Hoekstra, van Esbroeck, de Dood, Corstjens, Cnops, van Zeijl-van der Ham, Wammes, van Dam, Clerinx and van Lieshout2021)
This study aimed to determine the current prevalence of schistosomiasis in the Mwanga district Tanzania after approximately twenty years of MDA using the UCP-LF CAA test and to explore the potential of using the POC-CCA and microhaematuria dipstick as a combined POC test for diagnosing schistosomiasis in co-endemic settings in comparison to the laboratory-based UCP-LF CAA test. Furthermore, we investigated Schistosoma infection risk factors and associated parameters.
Materials and methods
Ethical considerations
Ethical approval for this study was obtained from Kilimanjaro Christian Medical University College (KCMUCo) research and ethical committee board (reference number: 2588). Administrative authorization was obtained from the district education officer and Mwanga District Medical Officer. Children were enrolled based on their availability, and those willing to participate were given a consent form to be signed by guardians and/or parents. Immediately after sample collection all children, including those who participated in our study, were provided with praziquantel under the yearly MDA program at school at the recommended dose in the presence of a local clinician.
Study area and population
The study was conducted in Mwanga district, one of the seven districts of Kilimanjaro region in Tanzania. Farming, fishing, sand collection, pebble making, soil bricking and animal keeping are the major economic activities. The study was conducted in five schools, of which two were selected based on a previous study (Poggensee et al., Reference Poggensee, Krantz, Nordin, Mtweve, Ahlberg, Mosha and Freudenthal2005). MDA of praziquantel had occurred in these schools more than 6 months prior to our study.
Sample/data collection and processing
Enrolled study participants provided consent forms from parents and were given sterile containers with unique identifiers to provide fresh urine samples. For each sample, POC-CCA and microhaematuria dipstick was done in the field for the diagnosis of S. mansoni and S. haematobium, respectively. An aliquot of 2 mL of urine was conserved (at −20°C) per participant and was shipped on ice to Leiden University Medical Center in the Netherlands for retrospective UCP-LF CAA analysis. Following urine sample collection, a face-to-face interview using a closed-ended questionnaire in English and Swahili was conducted.
Field and laboratory analysis
The POC-CCA test (batch 180817091) was obtained from Rapid Medical Diagnostics, South Africa (SA), and analysis was done according to the manufacturer's instructions. Briefly, two drops of urine were transferred into the sample window of the test cassette. The readout of the cassette was done in 20 minutes. Results were scored as negative, trace or positive. Microhaematuria dipstick (Mission Urinalysis, Lot no: URS8090018) test was performed by placing the strip on a flat surface and a drop of urine applied to the reagent pad. Readouts were done in 1 minute as either negative or positive according to the manufacturer's instructions.
The Schistosoma genus-specific UCP-LF CAA test was employed to detect CAA in urine samples and to confirm active infection with Schistosoma spp (Corstjens et al., Reference Corstjens, de Dood, Knopp, Clements, Ortu, Umulisa, Ruberanziza, Wittmann, Kariuki, LoVerde, Secor, Atkins, Kinung'hi, Binder, Campbell, Colley and van Dam2020a). All urine samples were subjected to the UCAAhT417 wet format of the test. Briefly, 500 μL of each urine sample was mixed with 100 μL of 12% trichloroacetic acid, then incubated at room temperature for 5 minutes and centrifuged. The clear supernatant was then concentrated to 20 μL using a 0.5 mL centrifugal device (Amicon Ultra-0.5, Millipore, Merck Chemicals B.V., Amsterdam, The Netherlands). The resulting concentrate was then applied to the lateral flow test strip. To quantify CAA concentrations and to validate the assay cutoff (0.6 pg mL−1), reference standards with known CAA-levels were included. A CAA concentration above 0.6 pg mL−1 was considered positive. Infection intensity was categorized as low positive (>0.6–10 pg mL−1), moderate positive (>10–100 pg mL−1) and high positive (>100 pg mL−1).
Statistical analysis
The agreement between the combination of POC-CCA an microhaematuria dipstick (combined POC test) and UCP-LF CAA was performed using Kappa (K) statistics. For POC-CCA, trace results were considered negative. Furthermore, the association between socio-demographic characteristics and risk factors associated with Schistosoma infection, based on the UCP-LF CAA test, was performed using chi-square statistics, binary and multiple logistic regression analyses. Statistical analysis was performed using IBM Statistical Package for Social Sciences version 29 (SPSS Inc., Chicago, United States of America). For generation of plots, GraphPad Prism version 9.3.1 for Windows (GraphPad Software, San Diego, California USA, www.graphpad.com) was used.
Results
Socio-demographic characteristics
A total of 576 children provided informed consent and subsequently provided a urine sample and were therefore included in the final analysis. In Table 1 socio-demographic characteristics of the study population are given. The children's age ranged from 5 to 16 years, with a mean age of 9.8 years (s.d. 2.4) and 50.7% were females. Furthermore, the majority of the children were in class range 1 to 3 (50.3%). Farming was the most common father's profession (46.7%) followed by fishing (23.4%). The most common mother's profession was farming (44.8%), followed by small businesses (34.2%).
The prevalence of schistosomiasis across all five schools based on UCP-LF CAA, POC-CCA, microhaematuria dipstick and a combination of the POC-CCA and microhaematuria dipstick
a Combination of POC-CCA and/or microheamaturia dipstick positive outcome.
Prevalence and intensity of Schistosoma infection
In total 117 (20.3%) of children were found to be CAA positive (Table 1). The highest proportion of positives was observed among children attending Kivulini primary school (33.3%), followed by Mkombozi primary school (29.2%). The lowest proportion was 12.5% and was found among school children at Kagongo primary school. The majority of moderate to high intensity infections based on CAA-levels was observed in the schools in Kivulini and Mkombozi (Fig. 2a). Furthermore, SAC within the age category 5–10 years were found to be more often CAA positive than those aged 11–16 years (Fig. 2b). The outcome of the point-of-care tests are also summarized in Table 1. Based on POC-CCA and microhaematuria dipstick tests the prevalence of schistosomiasis were, 54.5% and 25.7% respectively. Assuming that the combination of two tests will give more clearer prevalence, combining POC-CCA and microhaematuria dipstick, the prevalence was 65.3%.
Prevalence and intensity of Schistosoma infection based on UCP-LF CAA amongst (A), five selected schools (B) the age categories.
The agreement between the combination of POC-CCA and microhaematuria dipstick and the reference test UCP-LF CAA
The combination of POC-CCA and microhaematuria dipstick showed no agreement with the UCP-LF CAA test (Table 2). Furthermore the P value indicates that this lack of agreement is not statistically significant, suggesting that the disagreement between tests is likely due to chance at significance level of 0.05. Analysis of the individual POC-CCA and microhaematuria dipstick tests also showed no agreement with the UCP-LF CAA test.
The level of agreement between point-of-care circulating cathodic antigen (POC-CCA) test, microhaematuria dipstick and upconverting particle lateral flow circulating anodic antigen (UCP-LF CAA) urine test by Cohen's kappa coefficient in 576 school-aged children from Mwanga Tanzania
Risk factors associated with schistosomiasis
Using multivariate logistic regression analysis (adjusted odd ratio), the potential risk factors associated with Schistosoma infection, based on the presence of CAA, showed that children in class level 1–3 had two times higher odds of having schistosomiasis than children in higher classes. Children involved in farming and swimming activities had respectively 5.6 and 3.6 odds of being infected than those who did not farm nor swim. Furthermore, children attending Kileo, Kivulini, Mkombozi and Mnoa primary schools had 2.2, 2.6, 2.4 and 2.7 times higher odds of CAA positive results respectively, when compared to those attending Kagongo primary school. More details can be found in Supplementary Table 1.
Discussion
Using a highly accurate diagnostic approach (UCP-LF CAA), this study indicated that after nearly two decades of MDA schistosomiasis remains highly prevalent (20%) among school-aged children in Mwanga district, Tanzania. Although POC-CCA and microhaematuria dipstick test showed an even higher prevalence than the UCP-LF CAA test, no agreement was found between these tests and the UCP-LF CAA results nor any association was observed between these tests and known risk factors for schistosomiasis, highlighting the limitation of these currently available rapid diagnostics tests (POC-CCA and microhaematuria) in accurately determining the true prevalence in this specific setting that is known to be co-endemic for S. mansoni and S. haematobium.
Different prevalence's have been observed throughout Tanzania (Mnkugwe et al., Reference Mnkugwe, Minzi, Kinung'hi, Kamuhabwa and Aklillu2020; Mazigo et al., Reference Mazigo, Uisso, Kazyoba, Nshala and Mwingira2021; Kajembe et al., Reference Kajembe, Gasarasi, Tarimo, Lushina and Sylvester2022; Maganga et al., Reference Maganga, Campbell, Angelo, Mosha, Mwanga and Kinung'hi2023; Ogweno et al., Reference Ogweno, Mushi, Silvestri, Bonaventura, Justine, Noah, Yoram, Mohamed and Tarimo2023). As commonly known as well as shown in the current study, measurement of prevalence highly depends on the diagnostic method used. Since we have used a highly accurate diagnostic method in our study, i.e. the UCP-LF CAA test, it is difficult to directly compare our results to previously published results based on microscopy and/or POC-CCA as these methods have limited sensitivity/specificity. Our data confirm regional variation in the burden of schistosomiasis in Mwanga district, which would, extrapolated to Tanzania as a whole, argue for a more focally oriented schistosomiasis control approach.
A significant difference in infection rates among different age groups was identified. Younger children (5–10 years) exhibited a higher prevalence of schistosomiasis than the older age group (11–16 years), indicating early exposure to the infection (Ogweno et al., Reference Ogweno, Mushi, Silvestri, Bonaventura, Justine, Noah, Yoram, Mohamed and Tarimo2023). The possible reason for older children having low prevalence is through acquired immunity due repeated infections as indicated by other previous studies for example possible presence of IgE antibodies (Woolhouse, Reference Woolhouse1998; Oettle and Wilson, Reference Oettle and Wilson2017). A significant association was found between schistosomiasis and children who swim in water bodies, which may be attributed to playful behavioural activities common among children (Munisi et al., Reference Munisi, Buza, Mpolya, Angelo and Kinung'hi2017). The children involved in swimming activities had 3.6 times more risk of being infected, in line with recent systematic review demonstrated by Reitzug and colleagues (Reitzug et al., Reference Reitzug, Ledien and Chami2023). Children's involvement in farming was found to be associated with an increased risk of being infected with schistosomiasis. Finally, children attending Kileo, Kivulini, Mkombozi and Mnoa primary schools were identified to have higher rates of Schistosoma infection compared to those attending Kagongo primary school. These findings are likely due to the proximity of irrigation schemes/rivers to these schools, which are perceived as safer water sources for young children and so most likely to visit compared to larger water bodies like the dams located closer to Kagongo, where parents have concerns about the risk of children drowning and so cautioned not to go there for water.
Despite of providing crucial updates on the prevalence of schistosomiasis among school children in the area, the study has several limitations. Firstly, the laboratory UCP-LF CAA test detects all Schistosoma species, but it does not provide any species-specific information (Corstjens et al., Reference Corstjens, de Dood, Knopp, Clements, Ortu, Umulisa, Ruberanziza, Wittmann, Kariuki, LoVerde, Secor, Atkins, Kinung'hi, Binder, Campbell, Colley and van Dam2020). In control settings, where species information would be relevant, other measures can provide this, e.g. determining the presence of specific snail species, or performing egg microscopy or species-specific PCR. For treatment, species information as such is not needed, and CAA has been demonstrated to be an excellent marker for monitoring treatment efficacy (Sousa et al., Reference Sousa, van Dam, Pinheiro, de Dood, Peralta, Peralta, Daher, Corstjens and Bezerra2019; Hoekstra et al., Reference Hoekstra, van Esbroeck, de Dood, Corstjens, Cnops, van Zeijl-van der Ham, Wammes, van Dam, Clerinx and van Lieshout2021, Reference Hoekstra, Casacuberta-Partal, van Lieshout, Corstjens, Tsonaka, Assaré, Silué, N'Goran, N'Gbesso, Brienen, Roestenberg, Knopp, Utzinger, Coulibaly and van Dam2022a, Reference Hoekstra, Chernet, de Dood, Brienen, Corstjens, Labhardt, Nickel, Wammes, van Dam, Neumayr and van Lieshout2022b). Although POC-CCA and microhaematuria rapid tests are user-friendly, kappa statistics revealed a poor agreement between these tests and the UCP-LF CAA. It was expected that the combined positivity rates of POC-CCA and microhaematuria dipstick test would reflect the UCP-LF CAA results, however this was not the case (Table 1). The poor agreement may be due to production batch differences, sexually transmitted infections (STIs), low-intensity infections, and subjectivity to test readouts, which might also affect results (Assaré et al., Reference Assaré, Tra-Bi, Coulibaly, Corstjens, Ouattara, Hürlimann, van Dam, Utzinger and N'Goran2021). Furthermore, more accurate result with both tests might have been possible if the test was scored in a more quantitative manner. For example, following the recently described G-score scoring method for POC-CCA, an inclusion of control samples as a way to standardize the readout and to determine the cut-off for positivity. The microhaematuria test can be scored semi-quantitatively based on colour intensity linked to the level of red blood cells. However, registering of more quantitative results was not foreseen in this study.
In conclusion, this study demonstrates a moderate prevalence of schistosomiasis in Mwanga district Tanzania, implicating that the ~20 years annual MDA of praziquantel in this region may have had an effect on reducing the schistosomiasis burden, but transmission is still ongoing. To improve the efficacy of MDA strategies, diagnosis at acute stages of the disease in combination with treatment could be extended not only to higher risk groups but also to all persons above 2 years of age as recommended by WHO (Faust et al., Reference Faust, Osakunor, Downs, Kayuni, Stothard, Lamberton, Reinhard-Rupp and Rollinson2020; WHO, 2022). Apart from that, integrated approaches including improved access to WASH infrastructure, political willingness, and production of reliable data are important for controlling schistosomiasis in Tanzania (Abe et al., Reference Abe, Tambo, Xue, Xu, Ekpo, Rollinson, Yang, Li and Zhou2020). The presence of persistent hotspots in countries like Tanzania where provision of MDA program failed to provide long terms solution in some villages shows the need for such an integrated approach (Kittur et al., Reference Kittur, King, Campbell, Kinung'hi, Mwinzi, Karanja, N'Goran, Phillips, Gazzinelli-Guimaraes, Olsen, Magnussen, Secor, Montgomery, Utzinger, Walker, Binder and Colley2019). Furthermore, a combination of the POC-CCA and microhaematuria dipstick did not prove to be useful as a screening tool for schistosomiasis in this S. haematobium and S. mansoni co-endemic setting. However, efforts are ongoing to make CAA detection generally available, with a recent initiative focusing on the development of a more easy-to-use, accurate, affordable and visually scored CAA-RDT (FIND, 2020; GHIT, 2020). The CAA-RDT could be of great potential in resource-poor endemic settings and assist in the development of targeted control measures and interventions to effectively combat schistosomiasis.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182024001045.
Data availability statement
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary and that the raw data is available upon request to the corresponding author.
Acknowledgements
We gratefully acknowledge Claudia de Dood (LUMC) for her assistance in performing the CAA analysis, Dieuwke Kornelis (LUMC) for technical assistance with the POC-CCA, and George A. Masisila for assisting in data entry and POC-CCA processing.
Authors’ contributions
Conceptualization of the study: J.J.P, s.e.M, G.J.v.D, M.Y. Data collection, analysis interpretation of study findings: J.J.P, B.M, P.T.H, L.v.L, E.M, N.M, S.H, P.L.A.M.C, s.e.M, G.J.v.D, Editing: J.J.P, B.M, P.T.H, L.v.L, P.L.A.M.C, G.J.v.D; Original draft: J.J.P,B.M
Financial support
This study was funded by LUMC Global PhD fellowship fund which supported Jeremia J.Pyuza, LUMC-SF-GLOBAL and partly by NWO-WOTRO Science for Global Development program, Grant Number W 07.30318.009 (INSPiRED – Inclusive diagnostics for Poverty Related parasitic Diseases in Nigeria and Gabon) which supported Brice Meulah. This work is part of the EDCTP2 program supported by the European Union.
Competing interests
None.
Ethical standards
This study received ethical clearance from the Research and Ethics Committee of Kilimanjaro Christian Medical University College (KCMUCo), with reference number 2588.
Open access
