- Top of page
- Materials and methods
Objective To identify possible synergistic associations of hookworm and other helminths.
Method Cross-sectional survey of all households within 10 km2 of Americaninhas, a rural community in Minas Gerais, Brazil. We determined the prevalence and intensity of single and multiple helminth species infection in an age-stratified sample of 1332 individuals from 335 households.
Results Hookworm was the most prevalent helminth infection (68.2%), followed by Ascaris lumbricoides (48.8%) and Schistosoma mansoni (45.3%). Overall, 60.6% of individuals harboured mixed helminth infections. Multivariate analysis indicated significant positive associations for co-infection with hookworm and S. mansoni and for co-infection with hookworm and A. lumbricoides. Co-infections with hookworm and A. lumbricoides resulted in higher egg counts for both, suggesting a synergistic relationship between these species, although, we found important age differences in this relationship. However, the intensity of S. mansoni or A. lumbricoides co-infection did not differ from that of mono-infection.
Conclusion These results have implications for the epidemiology, immunology and control of multiple helminth infections. More research is needed to examine the rates of re-infection and immune responses after chemotherapy, and to what extent the effects of polyparasitism are altered by chemotherapy.
Objectif Identifier les associations synergétiques possibles entre l'ankylostome et les autres helminthes.
Méthode Une étude transversale incluant toutes les habitations sur 100 km2 autour de Americaninhas, une communauté rurale à Minas Gerais au Brésil. Nous avons déterminé la prévalence et l'intensité de l'infection à une ou plusieurs espèces d'helminthes dans un échantillon d’âge stratifié comportant 1332 individus provenant de 335 habitations.
Résultats L'ankylostome était l'helminthe le plus prévalant dans les infections (73.3%), suivi du Schistosoma mansoni (45.3%) et Ascaris lumbricoides (48.8%). Au total, 60.6% des individus étaient porteurs d'infections mixtes à helminthes. Une analyse multivariée a indiqué des associations significatives pour la coinfection ankylostome et S. mansoni et pour la coinfection ankylostome et A. lumbricoides. Les coinfections à ankylostome et A. lumbricoides résultaient en un nombre élevé des œufs des deux espèces, quoique nous avons trouvé une différence dans les ages pour cette relation. Cependant, l'intensité de la coinfection avec S. mansoni ou A. lumbricoides ne différait pas de celle de la monoinfection.
Conclusion Ces résultats ont des implications dans l’épidèmiologie, l'immunologie et le contrôle des infections mixtes aux helminthes. Des recherches plus poussées sont nécessaires pour examiner le taux de réinfection et les réponses immunitaires suite au traitement, et à quel point les effets de polyparasitisme sont altérés par la chemothérapie.
Objetivo Identificar posibles asociaciones sinergísticas entre uncinarias y otros elmintos.
Método Estudio de corte transversal de todas las casas dentro de 100 km2 de Americaninhas, una comunidad rural de Minas Gerais, Brasil. Determinamos la prevalencia y la intensidad de infecciones únicas o múltiples en una muestra estratificada por edad de 1332 individuos provenientes de 335 hogares.
Resultados Las uncinarias fueron la infección helmíntica más prevalente (73.3%), seguidas por Schistosoma mansoni (45.3%) y Ascaris lumbricoides (48.8%). En total, 60.6% de los individuos portaban infeciones helminticas mixtas. El análisis multivariado indica una asociación positiva para la co-infección con uncinarias y S. mansoni así como para la co-infección con uncinarias y A. lumbricoides. Co-infecciones con uncinarias y A. lumbricoides resultaban con conteos de huevos más altos para ambos, sugiriendo una relación sinergística entre estas dos especies, aunque encontramos diferencias en edad importantes en esta relación. Sin embargo, la intensidad de co-infección de S. mansoni o A. lumbricoides no se diferenciaba de la de la mono-infección.
Conclusión Estos resultados tienen implicaciones en la epidemiología, inmunología y el control de infecciones helmínticas múltiples. Es necesario realizar más estudios para determinar las tasas de re-infección y respuesta inmune después de recibir quimioterapia, así como para averiguar hasta que punto los efectos del poli-parasitismo son alterados por la quimioterapia.
- Top of page
- Materials and methods
In 1978, Buck and colleagues published a landmark study of polyparasitism in rural communities in Chad, Peru, Afghanistan and Zaire, which indicated the frequent occurrence of concomitant parasitic infections (Buck et al. 1978). An increasing number of studies of helminth epidemiology have since shown that multiple helminth infections are extremely widespread (Booth et al. 1998; Needham et al. 1998; Petney & Andrews 1998; Brooker et al. 2000; Howard et al. 2002; Tchuem Tchuentéet al. 2003; Raso et al. 2004). Evidence from laboratory model systems suggests that infection with one helminth may influence the outcome of infection with others (Cox 2001), with evidence suggestive of both synergism and antagonism in concurrent intestinal nematode and schistosome infections (Pritchard et al. 1991; Webster et al. 1997; Corrêa-Oliveira et al. 1988,2002; Cox 2001). A number of epidemiological studies have indicated that individuals with multiple helminth infections often harbour heavier infections than individuals with single species infections (Booth et al. 1998; Needham et al. 1998; Brooker et al. 2000; Tchuem Tchuentéet al. 2003). Other studies have reported that hookworm infection is positively associated with Schistosoma mansoni infection (Chamone et al. 1990; Keiser et al. 2002a,b; Raso et al. 2004) and with filarial nematode infection (Faulkner et al. 2005). It has also been speculated that helminth infections may adversely influence host immune responses to other parasites, especially to intracellular pathogens such as malaria (Nacher 2004). To date, most studies of polyparasitism have been undertaken among populations of school-aged children. To our knowledge, there have been only three epidemiological studies, which have investigated multiple helminth infections, both in terms of prevalence and intensity, within an entire community (Buck et al. 1978; Keiser et al. 2002b; Raso et al. 2004). Here, we present age-stratified data from an ongoing longitudinal study of the epidemiology and immunology of intestinal nematode and schistosome infections in Brazil. The present analysis describes the epidemiology of single and multiple helminth infection. Emphasis is placed on age-related changes in polyparasitism and with associations that occur between the species.
- Top of page
- Materials and methods
A total of 1332 individuals from 335 households were included in the study, of these individuals 82.7% were infected with at least one species of helminth. The overall prevalence of hookworm infection was 68.2%, the prevalence of S. mansoni was 45.3%, the prevalence of A. lumbricoides was 48.8%, the prevalence of Trichuris trichiura was 1.1% and the prevalence of infection of Enterobius vermicularis, Hymenolepis nana and Taenia spp. were respectively, 0.8%, 0.2% and 0.2%. The prevalence of heavy hookworm infection was 8.3%, the prevalence of heavy S. mansoni infection was 8.1%, and the prevalence of heavy A. lumbricoides infection was 18.2%.
Prevalence and intensity of A. lumbricoides rose to a peak in the 5–10 years age group and declined thereafter (Figure 1a). Prevalence of hookworm infection rose with age until the 15–20 years age group and stayed constant thereafter. Prevalence of S. mansoni infection also rose with age until the 10–15 years age group where it remained constant until peaking in the 40–45 years age group. Intensity of S. mansoni infection peaked in the 10–15 years age group and again in the 20–25 years age group and declined thereafter, whereas intensity of hookworm infection rose with increasing age (Figure 1b, c).
The prevalence of single and multiple species infection overall and by age group is shown in Table 1. Overall, of those infected with a helminth, 73.3% of individuals harboured mixed infections. The most common combinations were infection with hookworm and A. lumbricoides and with hookworm and S. mansoni. Only 4.6% of infected individuals were concurrently infected with A. lumbricoides and S. mansoni, which was significantly lower than would be expected by chance alone (6.5%) (χ2 = 9.9, P = 0.001). The observed percentage of individuals with all three species of infection was 19%, significantly higher than the expected value of 13.8% (χ2 = 13.0, P = 0.0003), and the prevalence of dual infection with hookworm and A. lumbricoides was also significantly higher than expected by chance (19.9%vs. 16.7%, χ2 = 4.6, P = 0.03). The prevalences of different species combinations are shown to vary by age. Table 2 presents the results of the logistic regression analysis and shows that, while controlling for age-group, sex and sector, hookworm was significantly positively associated with A. lumbricoides and with S. mansoni. This was true both for infections per se and for heavy infections. In contrast, no significant associations were observed between co-infections A. lumbricoides and S. mansoni.
Table 1. Observed and expected prevalence of single and multiple helminth infection among 1332 individuals in Americaninhas, Minas Gerais, Brazil in 2004.
|Species combination||Number positive||Percentage of infections||Percentage of population||χ2||P value||Observed by age groups (years)|
|Any single infection||294||26.7||22.1|| || || ||24.8||19.8||18.7||24.2|
|A. lumbricoides only||81||7.4||6.1||7.8||3.1||0.080||12.4||4.3||1.5||3.9|
|S. mansoni only||61||5.5||4.6||8.0||13.4||<0.0001||3.6||4.8||5.6||5.1|
|Any double infection||554||50.3||41.6|| || || ||37.6||48.9||39.3||38.9|
|Hookworm and A. lumbricoides||265||24.1||19.9||16.7||4.6||0.03||26.2||22.7||12.1||13.8|
|Hookworm and S. mansoni||238||21.6||17.9||17.1||0.3||0.610||6.0||21.4||24.2||23.9|
|A. lumbricoides and S. mansoni||51||4.6||3.8||6.5||9.9||0.001||5.4||4.8||3.0||1.2|
|Any mixed infection||807||73.3||60.6|| || || ||53.7||71.0||59.0||56.8|
Table 2. Associations between helminth species among 1332 individuals in Americaninhas, southeastern Brazil in 2004.
|Parasite||Association||Adjusted OR (95% CI)*||P-value|
|Hookworm||A. lumbricoides||3.59 (2.67, 4.82)||<0.001|
|S. mansoni||2.95 (2.19, 3.98)||<0.001|
|Heavy hookworm||Heavy A. lumbricoides||2.29 (1.35, 3.90)|| 0.002|
|Heavy S. mansoni||4.13 (2.43, 6.99)||<0.001|
|A. lumbricoides||Hookworm||3.65 (2.71, 4.91)||<0.001|
|S. mansoni||0.99 (0.75, 1.29)|| 0.917|
|Heavy A. lumbricoides||Heavy hookworm||2.29 (1.34, 3.92)|| 0.002|
|Heavy S. mansoni||0.59 (0.32, 1.11)|| 0.105|
|S. mansoni||Hookworm||3.01 (2.22, 4.06)||<0.001|
|A. lumbricoides||0.98 (0.75, 1.28)|| 0.881|
|Heavy S. mansoni||Heavy hookworm||4.19 (2.45, 7.14)||<0.001|
|Heavy A. lumbricoides||0.59 (0.32, 1.09)|| 0.091|
Overall, the mean intensity of hookworm infection significantly increased according to the multiplicity of species infection: hookworm only = 1220 epg; hookworm and A. lumbricoides = 1827 epg; hookworm and S. mansoni = 1888 epg; all three species = 2023 epg (P = 0.0002). Intensity of A. lumbricoides also significantly differed by multiple species infections, with individuals infected with A. lumbricoides and S. mansoni harbouring the lightest infections: A. lumbricoides only = 6323 epg; A. lumbricoides and hookworm and = 6551 epg; A. lumbricoides and S. mansoni = 4304 epg; all three species = 5330 epg (P = 0.003). Individuals infected with A. lumbricoides and S. mansoni also had the lowest intensities of S. mansoni, although group differences had only borderline significance: S. mansoni only = 208 epg; S. mansoni and hookworm = 338 epg; S. mansoni and A. lumbricoides = 163 epg; all three species = 266 epg (P = 0.07). When investigating these patterns by age group important differences were revealed. Although, mono-infected individuals had lower hookworm intensities in each age group, hookworm intensity differed significantly by species combinations only in the age categories 10–24 and 25–39 years (Kruskall–Wallis test of equality of populations: P = 0.027 and P = 0.015) (Figure 2c). Similarly, although individuals infected with S. mansoni and A. lumbricoides had the lowest intensity of A. lumbricoides infection, intensity only differed by species combinations among 25–39 years old (P = 0.009) (Figure 2b). For A. lumbricoides there were no significant differences in intensity by species combinations between the age groups, although individuals infected with S. mansoni and A. lumbricoides consistently had the lowest intensities (Figure 2a).
Figure 2. Mean intensity of infection among 1332 individuals in Americaninhas, Minas Gerais, Brazil, stratified by multiplicity of infection and age group. (a) A. lumbricoides, (b) S. mansoni and (c) hookworm. Columns indicate arithmetic mean intensity of infection and vertical bars indicate standard error of the mean.
Download figure to PowerPoint
- Top of page
- Materials and methods
Our analysis showed that there was a significant positive association between co-infection with hookworm and S. mansoni and with hookworm and A. lumbricoides. The positive association for co-infection with hookworm and S. mansoni is in keeping with previous studies in Brazil (Chamone et al. 1990) and Côte d'Ivoire (Keiser et al. 2002a,b; Raso et al. 2004). We further show that intensity of hookworm infection typically increased with the multiplicity of infection. Taken together, these results suggest a synergistic relationship between hookworm and other helminths, as has been observed elsewhere (Chamone et al. 1990). This observation is also compatible with immunological data from hookworm patients showing that antibodies reacting with crude antigen extracts from hookworms cross-react with egg and adult worm antigen extracts from S. mansoni (Pritchard et al. 1991; Corrêa-Oliveira et al. 1988, 2002). Similar antibody cross-reactivity between N. americanus and S. mansoni has been observed in murine models (Timothy et al. 1992). However, even though antigenic cross-reactivity was reported in this experimental model, no protective immunity was achieved by heterologous challenge infection (Timothy et al. 1992). Corrêa-Oliveira et al. (1988,2002) found that individuals from areas co-endemic for S. mansoni and hookworm had significantly lower proliferative responses to crude hookworm antigen extracts than individuals from hookworm mono-endemic areas.
Immunological mechanisms, such as the differential activation of T helper cell subsets by parasitic stages, may explain the synergisms and antagonisms observed during helminth co-infections in this study. Mice with an underlying S. mansoni infection were found to have a T helper 2-dominant immune response that could alter the disease outcome of different intestinal nematode species, e.g., Strongyloides venezuelensis (Yoshida et al. 1999) and T. muris (Curry et al. 1995). Protective effects might vary between different parasitic species, acting at different sites (skin, lung and gut) and different developmental stages such as third stage infective larvae (L3) (Yoshida et al. 1999) or adult worms (Curry et al. 1995). In N. americanus infection, increased levels of interleukin (IL)-5 secreted by peripheral blood mononuclear cells upon stimulation with crude adult hookworm antigen extract correlated with resistance to reinfection, indicating a role for Th2 cytokine responses in protection against hookworm reinfection (Quinnell et al. 2004a).
Interestingly, individuals in our study co-infected with hookworm and other strong Th2 cytokine (including IL-5) inducing helminth such as A. lumbricoides or S. mansoni had higher intensities of infection than individuals mono-infected with hookworm, pointing to a possible synergistic effect on the immune response to these helminths during hookworm co-infection. As such, it is possible that co-infection with hookworm, with a reduced cellular reactivity (Corrêa-Oliveira et al. 2002) and the secretion of immunomodulatory molecules (Chow et al. 2000; Loukas & Prociv 2001; Hsieh et al. 2004), subverts or overrides the mechanisms which would enable an age-acquired decline in infection intensity seen in other helminth infections. In this context, it is important to mention the possible role of recently described T-regulatory cells (Tr1), and their secretion of downmodulatory cytokines, e.g. IL-10 or TGF-β, which seem to participate in the regulation of the host's immune response (Maizels et al. 2004). For experimental schistosomiasis, these T-regulatory cells were found to be the major source of IL-10 (Hesse et al. 2004). Whether the observed production of IL-10 is stimulated directly by the presence of the parasites and/or their products, or represents a natural cross-regulatory mechanism initiated by the host remains to be established.
Here we also show that that individuals co-infected with A. lumbricoides and S. mansoni have lower infection intensities for each species than individuals mono-infected with each species, suggestive of antagonism between A. lumbricoides and S. mansoni infection. Previous studies of A. lumbricoides infections in humans have shown that Th-2-type immune responses were associated with reduced worm burden and age-related protective immune responses (Turner et al. 2003). A similarly strong induction of Th-2 responses has been observed with the start of S. mansoni egg deposition in intestinal tissues in experimental animal models (Grzych et al. 1991; Pearce et al. 1992). As such, this markedly similar immune modulation may account for the decrease in intensity of infection observed in individuals co-infected with S. mansoni and A. lumbricoides compared to individuals mono-infected with each helminth. This has also been proposed for experimental infections with A. suum (Frontera et al. 2003). However, many other factors may also contribute to these parasite interactions as well, and we plan to further investigate the immune responses of individuals who are mono- and co-infected for hookworm, A. lumbricoides, and S. mansoni in future studies.
An alternative explanation for the observed associations is the similar and different transmission routes of the different helminth species. Transmission of hookworm and A. lumbricoides occurs through exposure to soil contaminated with free-living infective stages and is influenced by several factors, including micro-climatic suitability, sanitation and hygiene, and environmental contamination with human excreta (Olsen et al. 2001; Keiser et al. 2002a; Traub et al. 2004). Schistosomiasis, in contrast, is a water-borne infection and transmission is based on infective water bodies and human water contact. Information on individual and household specific risk factors was not collected and this hypothesis remains to be further investigated.
In the tropics polyparasitism is extremely common. Our results show that infection with one helminth can profoundly affect the intensity of infection with another helminth, e.g., heavy hookworm infection results in heavy infection with S. mansoni and vice versa. It may be that the powerful immunomodulatory mechanisms of helminths have important effects on susceptibility not only to other helminth infections but to other infectious diseases in general, including implication for the response to vaccination: e.g., the response to Bacillus Calmette-Guerin (BCG) vaccination in the tropics (Elias et al. 2001; Brooker et al. 2004; Quinnell et al. 2004b) and the live oral cholera vaccine CVD 103-HgR (Cooper et al. 2001). Our cross-sectional study provides an indication of the presumably cumulative effect of single and multiple species infection. Longitudinal studies are now underway in the study area to examine the rates of re-infection and immune responses following chemotherapy, and thus will provide information on how much of the effects of polyparasitism is altered by chemotherapy.