Host determinants of reinfection with schistosomes in humans: a systematic review and meta-analysis

Abstract

Background: Schistosomiasis is still a major public health burden in the tropics and subtropics. Although there is an effective chemotherapy (Praziquantel) for this disease, reinfection occurs rapidly after mass drug administration (MDA). Because the entire population do not get reinfected at the same rate, it is possible that host factors may play a dominant role in determining resistance or susceptibility to reinfection with schistosomes. Here, we systematically reviewed and meta-analyzed studies that reported associations between reinfection with the principal human-infecting species (S. mansoni, S. japonicum and S. haematobium) and host socio-demographic, epidemiological, immunological and genetic factors.

Methodology/principal findings: PubMed, Scopus, Google Scholar, Cochrane Review Library and African Journals Online public databases were searched in October 2013 to retrieve studies assessing association of host factors with reinfection with schistosomes. Meta-analysis was performed to generate pooled odds ratios and standardized mean differences as overall effect estimates for dichotomous and continuous variables, respectively. Quality assessment of included studies, heterogeneity between studies and publication bias were also assessed. Out of the initial 2739 records, 109 studies were included in the analyses, of which only 32 studies with 37 data sets were eligible for quantitative data synthesis. Among several host factors identified, strong positive association was found with age and pre-treatment intensity, and only slightly for gender. These factors are major determinants of exposure and disease transmission. Significant positive association was found with anti-SWA IgG4 level, and a negative overall effect for association with IgE levels. This reconfirmed the concept that IgE/IgG4 balance is a major determinant of protective immunity against schistosomiasis. Other identified determinants were reported by a small number of studies to enable interpretation.

Conclusions: Our data contribute to the understanding of host-parasite interaction as it affects reinfection, and is a potential tool to guide planning and tailoring of community interventions to target high-risk groups.

Figure 1. Flow diagram for the search and systematic review process.

Figure 2. Association of younger age (<10 years old) with reinfection with schistosomes.

Presented here is the meta-analysis forest plot showing the pooled odds ratio and the corresponding 95% CI, subgroup analysis by species, and assessment of heterogeneity among studies. There was a strong statistically significant positive association between younger age (<10 years old) and reinfection with schistosomes.

Figure 3. Association of gender (male) with reinfection with schistosomes.

Presented here is the meta-analysis forest plot showing the pooled odds ratio and the corresponding 95% CI, subgroup analysis by species, and assessment of heterogeneity among studies. The observed positive association between reinfection and gender was only slightly significant.

Figure 4. Association of epidemiological factors with reinfection with schistosomes.

(A) Meta-analysis forest plot for the association of reinfection with high pretreatment intensity showing positive association with reinfection. (B) Forest plot for the association of reinfection with high rate of exposure showing association with reinfection without statistical significance. (C) Meta-analysis forest plot for the association of reinfection with residence in high transmission area did not show statistically significant association with reinfection.

Figure 5. Association of IgE levels with reinfection with schistosomes.

(A) Meta-analysis forest plot for the association of reinfection with anti-SWA IgE levels showing the pooled standardized mean difference and the corresponding 95% CI and assessment of heterogeneity among studies. The observed negative overall effect was not statistically significant. (B) Meta-analysis forest plot for the association of reinfection with anti-SEA IgE levels showing the pooled standardized mean difference and the corresponding 95% CI and assessment of heterogeneity among studies. The observed negative overall effect was not statistically significant.

Figure 6. Association of IgG4 levels with reinfection with schistosomes.

(A) Meta-analysis forest plot for the association of reinfection with anti-SWA IgG4 levels showing the pooled standardized mean difference and the corresponding 95% CI and assessment of heterogeneity among studies. IgG4 level was highly significantly associated with reinfection with schistosomes. (B) Meta-analysis forest plot for the association of reinfection with anti-SEA IgG4 levels showing the pooled standardized mean difference and the corresponding 95% CI and assessment of heterogeneity among studies. The observed positive overall effect was not statistically significant.

Figure 7. Funnel plots for assessment of publication bias.

(A) Funnel plots for assessment of publication bias for studies assessing association of reinfection with age showing lack of significant publication bias in the included studies. (B) Funnel plots for assessment of publication bias for gender.