Polyparasitism in School-Aged Children Living in the Schistosomiasis Endemic Focus of Muyuka-Cameroon after >8 Years of Sustained Control Measures: A Cross Sectional Study


 Background: Despite the ubiquity of polyparasitism, its health impacts have been inadequately studied. The aim of this study was to determine the prevalence and determinants of polyparasitism with Schistosoma haematobium, Plasmodium and soil-transmitted helminths (STH) following sustained control measures, as well as evaluate the outcomes and the clinical correlates of infection in school-aged children (SAC) living in the schistosomiasis endemic focus of Muyuka-Cameroon.Methods: Microhaematuria in urine sample was detected using reagent strip and S. haematobium ova by filtration/microscopy methods. Plasmodium in blood sample was detected using Giemsa stained blood films and complete blood count was obtained using an auto-haematology analyser. STH in stool sample was detected by the Kato-Katz method. Categorical and continuous variables were compared as required, Kappa value estimated and the adjusted odds ratio (AOR) in the multivariate analysis was used to evaluate association of the risk factors with infection.Results: Out of the 638 SAC examined, single infection was prevalent in 33.4% while polyparasitism was 19.9%. Prevalence of S. haematobium + Plasmodium was 7.8%; S. haematobium + STH was 0.8%; Plasmodium + STH was 0.8% while, S. haematobium + Plasmodium + STH was 0.9%. Higher preponderance of S. haematobium + Plasmodium infection occurred in females, those from Likoko, did not use potable water, practiced bathing in stream and carried out open defecation than their equivalents. However, being female (AOR = 2.38, P = 0.009) was the only significant risk factor identified. Anaemia was a common morbidity (74.3%) with a slight agreement with microscopy in predicting S. haematobium and Plasmodium infections. The sensitivity and specificity of haematuria (13.0%) in predicting S. haematobium infection was 46.5% and 100% with a moderate agreement with microscopy. Co-infection with S haematobium and malaria parasite was significantly associated with 3-fold odds of history of fever in the last three days.Conclusions: Polyparasitism is a public health problem in the Ikata-Likoko area in Muyuka with females most at risk. Anaemia prevalence is exacerbated in co- and triple-infections and together with a history of fever are of value in predicting polyparasitism.


Background
Polyparasitism is a common condition in human populations in which a person experiences disease from two or more concomitant, chronic infections with helminthic and/or protozoan parasites [1,2]. In addition to malaria, schistosomiasis and soil-transmitted helminths (STH) constitute a major public health problem in many parts of sub-Saharan Africa. A total of 229.2 million people in 2018 needed preventive chemotherapy (PC) for schistosomiasis globally, of which 124.4 million were school-aged children [3]. More than 1.5 billion people are infected with STH worldwide and over 568 million school-aged children (SAC) live in areas of intensive transmission and need treatment and preventive interventions [4]. On the other hand, out of the 228 million malaria cases reported in 2018, 213 million (93%) occurred in the World Health Organisation (WHO) African Region [5].
As helminth and Plasmodium infections overlap geographically in developing countries, it is therefore a probable phenomenon for polyparasitism to occur causing high morbidity however, these infections are rarely studied together. Neglected tropical diseases (NTD) such as schistosomiasis, a water-borne parasitic disease remains a focal disease while infection with STHs such as Ascaris lumbricoides, Trichuris trichiura and hookworms are ubiquitous in developing regions of Africa, Asia and the Americas [6]. In the ecological settings of Mount Cameroon, the transmission of Schistosoma haematobium, Plasmodium spp., and STH is common and concurrent urogenital schistosomiasis, malaria, and or ascariasis have been reported [7,8,9]. Universal factors attributed to the co-occurrence of these infections include poor sanitation, inadequate toilet facilities, lack of potable water, and ineffective public health enlightenment programme and services [8,10].
Despite the ubiquity of polyparasitism, its health impacts have been inadequately studied. The effects of polyparasitism are often clinically inapparent however, concomitant infection of two parasites may modulate the effects of each other within their host [11]. While it is unclear if co-infections with helminths such as schistosomes can modulate susceptibility to malaria in humans [12], it is di cult to attribute morbidity-related outcomes in situations where multiple causative pathogens co-exist within the same person. Sparse evidence on the effect of polyparasitism in SAC is suggestive of the occurrence of anaemia, malnutrition, impaired cognitive development, splenomegaly, fatigue and multiplicative impact on organ pathology [2,13,14,15,16]. These subtle morbidities like malnutrition as well as its common presentation stunting, anaemia, and leucocytosis are correlates of both helminths and protozoan infections [8,17].
The control of schistosomiasis and STH in endemic regions in Africa rely on regular mass drug administration (MDA) and monitoring of infections in SAC. In 2015, 53.2 million and 417 million SAC received PC for schistosomiasis and STH respectively [18]. With an estimated two million people with schistosomiasis in Cameroon and an additional ve million living in high transmission areas, annual MDA of praziquantel to SAC in endemic areas is the country's main control strategy against the disease [19,20]. Among the integrated malaria control strategies in different epidemiological settings in Cameroon, vector control intervention through distribution of long-lasting insecticidal nets (LLINs) has been scaled-up [21]. While the prevalence of these parasites have experienced a reduction due to mapping of schistosomiasis endemic areas and sustained MDA campaigns with praziquantel, PC with albendazole for STH and the scale-up of treated bed nets across the country between 2000 and 2015 for control of malaria [20,22,23], these have not been su cient in interrupting the transmission cycle of the parasites. Hence there is the need for regular monitoring studies in endemic areas to tailor strategies to ensure site-speci c interruption of the disease transmission.
Due to the commonality of Plasmodium and helminth co-infections [24], improved understanding of polyparasitism is of concern in resource-limited settings in developing countries where diagnosis and treatment strategies are prioritized. Hence, investigating the implications of morbidities associated with polyparasitism is invaluable for healthcare workers. The aim of this study was to determine the prevalence and determinants of polyparasitism with S. haematobium, Plasmodium and STH following sustained control measures as well as evaluate the outcomes and the clinical correlates of infection in SAC living in the schistosomiasis endemic foci of Ba a, Ikata and Likoko in Muyuka-Cameroon

Study area and participants
This study was carried out in three rural localities of Ikata, Ba a and Mile 14-Lykoko in the Muyuka Health District. A detailed description of the study sites has been reported previously [25]. While there is potable water and streams in Ba a and Ikata, Likoko Native has no potable water. The villages have an Integrated Health Centre each except for Likoko. In addition to the IHC, Ba a has another health centre belonging to the Cameroon Baptist Convention. Previous studies in the area [26] revealed the presence of S. haematobium infections in school children along this path as well as the presence of the intermediate host. In the Mount Cameroon area, human malaria is meso-endemic during the dry season but becomes hyper-endemic in the rainy season, with incidence peaking in July-October [27].
This study was conducted among primary school children aged 4-14 years of both sexes whose parents consented to their participation in the study. Participation was voluntary and only children who had resided for at least three months in the study area took part in the study.

Study design, sample size estimation and sampling
This was a cross-sectional study which ran from March to June, 2015 as a follow-up of a cross -sectional study carried out earlier [25]. Prior to the commencement of study, regular visits were made to the various study sites, to educate the inhabitants on the importance, bene ts and protocol of the study. Children who presented signed consent forms were enrolled into the study and interviewed using a simple structured questionnaire to obtain information on both demography and factors that may be associated with the presence of the conditions. This was followed by clinical evaluation where weight, height and temperature were measured. The study involved the collection of venous blood, urine, and stool sample for haematological analysis, and microscopic detection of S. haematobium and STH eggs, respectively. Labelled blood and urine (placed on ice block) and stool samples preserved in 10% formalin were transported to the University of Buea Research Laboratory for further analysis.

Data Collection
A pre-tested questionnaire was administered to each participant with the aid of the teachers to obtain information on demography, personal hygiene and practices, health status and possible risk factors of Plasmodium and helminth infections as well as malnutrition and anaemia. The ages of participants were obtained from the school register. The axillary temperature was measured using a digital thermometer and a participant was considered febrile, if the body temperature was ≥ 37.5 °C. The height was measured to the nearest 0.1 cm using a graduated ruler of length 2 m. The body mass was measured to the nearest 0.5 Kg using a mechanical scale of capacity 120 Kg (KINLEE® model BR9310), and upper arm circumference was measured using a graduated tape. These measurements were used to calculate an array of anthropometric indices used as proxies for malnutrition: weight-for-age (under-weight); height-for-age (stunting); weight-for height (wasting). Anthropometric indices were computed as z-scores and the value of -2 was used as the critical point below which the participant was considered malnourished [28].
Malaria parasite diagnosis and full blood count Approximately 2 ml of venous blood was collected in ethylenediamine tetra-acetate tubes for malaria parasite detection and haematological analysis. Thick and thin blood lms were prepared in situ, following standard operational protocol [29]. Thin blood lms xed in methanol and thick blood lms were Giemsa stained and examined microscopically following standard procedures [29]. Slides were considered positive when asexual forms and/or gametocytes of any Plasmodium species were observed on the blood lm. All the slides were read twice by two independent microscopists. Malaria parasite per µL of blood was determined by counting the number of parasites per 200 leukocytes and multiplying by the individuals white blood cell (WBC) count. Parasitaemia was classi ed as low (≤ 500 parasite /µL of blood), moderate (501-5000 parasites/µL of blood) and high (> 5000 parasites/µL of blood).
A complete blood count was ran using a Beckman coulter counter (URIT 3300) that automatically gave values for red blood cell (RBC), WBC and platelet counts, haemoglobin (Hb), haematocrit (Hct), mean cell volume (MCV), mean cell haemoglobin (MCH), mean cell haemoglobin concentration (MCHC) following the manufacturer's instructions. The classi cation of anaemia (Hb concentration below the WHO reference values for age or gender) and its severity was done following WHO standards (mild anaemia = 10-10.9 g/dL, moderate anaemia = 7-9.9 g/dL and severe anaemia < 7 g/dL) [29,30].
Urine analysis for haematuria and schistosome eggs About 25 mL of midstream urine was collected into plastic screw cap vials after a brisk exercise between 10am and 2 pm. Gross haematuria was determined by visual observation while micro haematuria was determined with the aid of reagent strips (combistix) following the manufacturers guide (CYBOW™ 11M a series of Health Mate Ref 0974). Following agitation, 10 mL of urine was drawn using a syringe and ltered through a polycarbonate membrane lter (STERLITECH corporation). The lter membrane was examined microscopically for the presence of schistosome eggs as described by Cheesbrough [29]. Schistosome egg density was expressed as the number of eggs in 10 mL urine (e/10 mL) and the intensity of infection was categorised as either light (< 50 e/10 mL) or heavy infection (≥ 50 e/10 mL) [31,32].

Data analysis
Descriptive measures like the mean and standard deviation (SD), geometric means, frequencies, and proportions were used to summarize data.
Differences in proportions between populations were compared using Chi (χ 2 ) test. Geometric mean parasite density (GMPD) of P. falciparum and schistosome egg counts by age and sex were compared using analysis of variance (ANOVA) and the Student's t-test, respectively. Geometric means were computed for those positive only and the log transformed counts were used in the analysis. The adjusted odds ratio (AOR) in the multivariate analysis was used to see the strength of the association of the risk factors with infection. The 95% con dence interval (CI) was reported and p-values < 0.05 were considered indicative of statistical signi cance. The ability of a measurable morbidity to discriminate between infections and the diagnostic performance was evaluated using the receiver operating characteristics (ROC) curve analysis and the strength of agreement was determined by estimating the Kappa value. Kappa (к) was calculated using a Graphpad calculator [34] and the values interpreted as stated by Landis [35]. All data was analysed using IBM-Statistical Package for Social Science (SPSS) version 21 (IBM-SPSS Inc., Chicago, IL, USA).

Ethical considerations
The study protocol was reviewed and approved by the Institutional Ethical Review Board hosted by the Faculty of Health Sciences, University of Buea (2014/243/UB/FHS/IRB). The population was sensitized in the various communities at the beginning of the study. Written informed consent was obtained from all parents/ caregivers whose child/children participated in the study after explaining the purpose and bene ts of their participation. Participation was totally voluntary, and a participant could opt out of the study at any time without any penalty. Participants who had malaria and or helminths were given rst line treatment as recommended by the national treatment guideline policy for malaria and helminths.

Risk factors of polyparasitism
As shown in Table 3 the multivariate analysis revealed no signi cant demographic or behavioural factors associated with S. haematobium + P. falciparum + STH and S. haematobium + STH infections. Being female (AOR = 2.38, P = 0.009) was the only signi cant risk factor associated with S. haematobium + Plasmodium infection as they were 2.38 times at odds of having the co-infection. On the other hand, living in Ikata (AOR = 0.04 P < 0.001) and not practicing open defecation behaviour demonstrated signi cant protection against S. haematobium + Plasmodium co-infection.

Infection outcomes and clinical correlates
The most common clinical morbidity measured was anaemia (74.3%) followed by microcytosis (45.3%), malnutrition (26.5%) and the least was hypochromasia (1.6%). Apart from anaemia, the most common symptoms associated with S. haematobium infection were haematuria (46.5%), microcytosis (41.4%) and malnutrition (27.3%); Plasmodium sp infection: microcytosis (45.9%), malnutrition (16.3%) and fever (14.4%), while for Schistosoma and Plasmodium co-infection was haematuria (54.0%), microcytosis (50%) and fever (36.0%) as shown in Table 4.   Analysis of the suitability of the clinical signs measured to determine the symptoms of the disease revealed the speci city of haematuria in predicting S. haematobium infection was 100% (95% CI = 98.9-100%), with a sensitivity of 46.5% (95% CI = 37.0-56.2%) and a moderate agreement (к = 0.576) with microscopy while, the sensitivity and speci city of presumptive use of anaemia and microcytosis were 80.8% and 29.8% vs 41.4% and 54.3% with a slight and no agreement respectively with microscopy. In relation to Plasmodium infection the sensitivity and speci city of presumptive use of anaemia and microcytosis to predict infection was 79.6% and 29.8 vs 45.9% and 54.3% with a slight agreement (к = 0.051 and к= 0.002) respectively with microscopy as shown in Table 5. Overall, the most common unmeasurable clinical outcome reported was fever in the past 3 days (51.7%), followed by lower abdominal pain (43.4%) and fever in the last 3 months (40%). As shown in Table 6, SAC who had single infection with S. haematobium were 1.83 times and 1.68 times at odds of reporting fever in the last 3 days (AOR = 1.83, P = 0.015) and headaches (AOR = 1.68, P = 0.045) respectively. Similarly, co-infection with S. haematobium and malaria parasite was signi cantly associated with 3 fold odds of history of fever in the last three days (AOR = 3.02, P = 0.001) and in addition 2.84 times at odds of having lower abdominal pain (AOR = 2.84, P = 0.002). While SAC with S. haematobium + STH and those with MP and STH infections were 3.32 times and 2.12 times at odds of reporting diarrhoea and vomiting respectively, the risk was not statistically signi cant.  Polyparasitism occurred in 19.9% of the children although the prevalence of single infection was more common with similar occurrence of S. haematobium and P. falciparum infection. This polyparasitism prevalence in SAC is higher than the 7.6% observed in Mbam and Inoubou Division, within the Centre Region of Cameroon [38], 11.2% in Ghana [39] and lower than the 30% and 28% observed in Kenya [40,41]. When compared with previous studies in the same locality [25,42], a decline in infections with S. haematobium and P. falciparum following MDA was observed in SAC. However, the prevalence of polyparasitism is likely to remain a signi cant public health problem in the Ikata-Likoko area where environmental (streams near homes, high rainfall) and socio-economic (farming and shing activities, inadequate health care services, low level of education) characteristics are likely to favour the transmission of these infections. Again, while the national control strategy for helminth infection in SAC may curb transmission, infected individuals not included in the programme are likely to serve as a source of re-infection due to their common exposure to snail infested streams serving the communities.
The predominance of S. haematobium and P. falciparum (7.8%) co-infection when compared with S. haematobium and STH (0.8%), MP and STH (0.8%) and S. haematobium, P. falciparum and STH (0.9%) is not unusual. This may be attributed to the signi cant decline in STH infections in the Mount Cameroon area following the school-based deworming (SBDW) strategy with mebendazole adapted by Cameroon in 2004 and has been implemented annually since 2007 in both enrolled and unenrolled children [43; 44]. This S. haematobium and P. falciparum co-infection is of public health importance as the prevalence is higher than the 0.9% observed in Accra Ghana [45], comparable to the 9.0% in Gabon [46], lower than the 10.9% and 13.6% reported in Mvomero-Tanzania and West Region of Cameroon respectively [47,48] and within the 2.84 to 57.1% range reported in Africa [16,49,50].
Findings from the univariate analysis revealed being female, site (Likoko), children who did not use potable water, usually bathed in streams and carried out open defecation were more likely to have S. haematobium and Plasmodium co-infection with interchangeable factors affecting the prevalence of P. falciparum and STH. Similar factors have been reported elsewhere [9,46,51]. However, the multivariate analysis demonstrated being female was the only signi cant risk factor with 2.38 times likelihood of having the S. haematobium and Plasmodium co-infection. This is not surprising as females spend more contact time in infested streams washing clothes, playing, swimming, and when bathing hence, the likelihood to be re-infected after treatment is higher [52,53]. Albeit S. haematobium and P. falciparum have distinct transmission patterns, ndings from the study (Additional le 2) demonstrated similar drivers of the infections. This probably asserts the in uence of environmental and host factors on the epidemiological and geographical patterns of infections and diseases [54]. Hence, sustainable multidisciplinary intervention that integrates preventive chemotherapy with education on water, sanitation and hygiene (WASH), provision of potable water supply to communities, appropriate faecal disposal methods and improvement in health facilities and care is desired to reduce the burden of parasitic infections.
Worthy of note is the abundance of light infections with S. haematobium and low-density malaria parasite infections observed. In addition, all infections with STHs (Ascaris lumbricoides and Trichuris trichuria) were light and occurred mostly in SAC of Likoko area. The consequences of the absence of potable water supply and an integrated health centre in the Likoko community is undoubtedly demonstrated here by the presence and high occurrence of all the identi ed parasites, suggestive of contaminated environment than the other localities. The high prevalence of light infection is consistent with similar studies in Nigeria, Malawi and Ghana [45,52,55]. Light infections can occur in populations previously targeted for schistosomiasis control [56] on the other hand, high prevalence of heavy intensity infection suggestive of long-term transmission and attributable to poor sanitation and water supply facilities have also been reported [57]. Most likely, the MDA with an anti-helminthic each year and the ineffective use of the LLIN were not successful in preventing reinfections but probably aided in maintaining lower grade parasite intensities in the population.
Low levels of parasite loads represent chronic parasite infections which may play a major role in clinical morbidity [58]. The effects of polyparasitism which are often clinically inapparent may lead to multiple morbidities. Nevertheless, in some situations, co-infections may exacerbate disease symptoms due to one of the pathogens. Observations from the study revealed anaemia as the most common (74.3%) clinical morbidity measured and its occurrence was exacerbated in co and triple infections with Plasmodium and helminths in line with Nyarko et al [45]. Furthermore, a slight agreement in sensitivity and speci city of anaemia with microscopy in predicting the presence of both S. haematobium and P. falciparum infections was proven. While the spectrum of anaemia is broad and complex in resource-limited settings, these ndings assert the signi cant contributions of urogenital schistosomiasis and malaria to the burden of anaemia in endemic areas accentuated by several studies [50,59,60] and could be a valuable presumptive diagnostic marker of both infections even if the speci city is low.
Morbidity associated with urogenital schistosomiasis is caused by granulomatous reactions formed in response to egg deposition in the walls of the urinary tract, triggering in ammatory reaction, haematuria, proteinuria, brosis with ensuing obstruction and bladder carcinogenesis (61,62). Haematuria or bloody urine is a classic sign of urogenital schistosomiasis and ndings from the study revealed an overall prevalence of 13.0%. This is lower than the 16.6% observed in SAC in Northern Angola [63]. Haematuria was the second most common morbidity associated with urogenital schistosomiasis with 100% speci city, sensitivity of 46.5% and a moderate kappa agreement with microscopy in predicting the presence of the infection. The high speci city and low sensitivity observed is not atypical even though the sensitivity is lower than the 65% reported in populations with lower intensity infections [64]. Nonetheless, this is congruent with synthesis of previous ndings that highlight dipstick sensitivity to decrease while speci city increases when compared to dipstick performance in high prevalence areas. This lends support for the need of a combination of diagnostic tools including clinical criteria as light and old infections may be missed by microscopy [65].
Other morbidities of signi cance observed in the study were microcytosis (45.3%) and malnutrition (26.5%). The prevalence of malnutrition (24.4%), with the most common being stunting, is comparable to those of SAC in localities close by [66] and lower than the 29.7% in SAC in Rural Senegal [67]. Observation from the study showed a general inclination of SAC with P. falciparum to have predominance of microcytosis while those with S. haematobium had a higher occurrence of malnutrition. Unlike the increase in prevalence of microcytosis observed in P. falciparum and S. haematobium co-infection, increase in malnutrition prevalence was observed in triple infections of S. haematobium, P. falciparum and STH only.
Although the directionality of causality of these morbidities are not very speci c, microcytosis have been previously associated with protection against erythrocytic stage Plasmodium infection and severe malarial anaemia [68,69]. On the other hand, the growth faltering and malnutrition attributed to urogenital schistosomiasis has been linked to chronic anti parasite in ammation which persists during childhood [67,70,71,72].
A history of fever in the past 3 days was the most common unmeasurable clinical outcome reported while fever pervasiveness was lower. In addition, co-infection with S. haematobium and malaria parasite was signi cantly associated with 3-fold odds of history of fever in the last three days. Fever is a non-speci c marker of infection that is often considered as a symptom of malaria in endemic areas. It results from endogenous pyrogen molecules activities, notably pro-in ammatory cytokine tumour necrosis factor (TNF)-α. However, S. haematobium infection could further augment antiin ammatory responses induced by asymptomatic P. falciparum infection reducing the risk of fever probably accounting for the low occurrence of fever in the population [46,73]. Other common morbidities of signi cance reported associated with co-infections include lower abdominal pain, diarrhoea and vomiting.
The study is not without limitations. The use of a single stool and urine sample for the detection of helminth infection may have led to underestimation of the prevalence of polyparasitism as well as the intensities of the infections considering the variation in day to day excretion of eggs of some of these parasites. Other intestinal parasites may have gone undetected due to the insensitivity of the Kato-Katz technique used. Despite this underestimation, we consider the data meaningful to reveal implications on disease-related outcome and clinical correlates.

Conclusions
Polyparasitism is a public health problem in the Ikata-Likoko area in Muyuka even though single infection with either Plasmodium or S. haematobium was more common. Similar behavioural and environmental drivers of co-infections were observed with females most at risk hence, more sustainable, multidisciplinary, aggressive intervention control strategy is needed. Anaemia was the most common clinical morbidity measured and its occurrence was exacerbated in co-and triple-infections hence, anaemia could be of value as a presumptive diagnostic marker of urogenital schistosomiasis and malaria in resource-limited endemic areas. Haematuria was speci c to urogenital schistosomiasis while there was a general inclination of SAC with P. falciparum to have predominance of microcytosis while those with S. haematobium had a higher occurrence of malnutrition. While lower abdominal pain, diarrhoea and vomiting were commonly reported, fever and principally a history of it is of value in predicting polyparasitism.

Declarations
Ethics approval and consent to participate