Decline in infection-related morbidities following drug-mediated reductions in the intensity of Schistosoma infection: A systematic review and meta-analysis

Abstract

Background: Since 1984, WHO has endorsed drug treatment to reduce Schistosoma infection and its consequent morbidity. Cross-sectional studies suggest pre-treatment correlation between infection intensity and risk for Schistosoma-related pathology. However, evidence also suggests that post-treatment reduction in intensity may not reverse morbidity because some morbidities occur at all levels of infection, and some reflect permanent tissue damage. The aim of this project was to systematically review evidence on drug-based control of schistosomiasis and to develop a quantitative estimate of the impact of post-treatment reductions in infection intensity on prevalence of infection-associated morbidity.

Methodology/principal findings: This review was registered at inception with PROSPERO (CRD42015026080). Studies that evaluated morbidity before and after treatment were identified by online searches and searches of private archives. Post-treatment odds ratios or standardized mean differences were calculated for each outcome, and these were correlated to treatment-related egg count reduction ratios (ERRs) by meta-regression. A greater ERR correlated with greater reduction in odds of most morbidities. Random effects meta-analysis was used to derive summary estimates: after treatment of S. mansoni and S. japonicum, left-sided hepatomegaly was reduced by 54%, right-sided hepatomegaly by 47%, splenomegaly by 37%, periportal fibrosis by 52%, diarrhea by 53%, and blood in stools by 75%. For S. haematobium, hematuria was reduced by 92%, proteinuria by 90%, bladder lesions by 86%, and upper urinary tract lesions by 72%. There were no consistent changes in portal dilation or hemoglobin levels. In sub-group analysis, age, infection status, region, parasite species, and interval to follow-up were associated with meaningful differences in outcome.

Conclusion/significance: While there are challenges to implementing therapy for schistosomiasis, and praziquantel therapy is not fully curative, reductions in egg output are significantly correlated with decreased morbidity and can be used to project diminution in disease burden when contemplating more aggressive strategies to minimize infection intensity.

Fig 1. Flow chart of study search and selection strategy.

The flow diagram indicates the numbers of titles and studies reviewed in preparation of the current systematic review and meta-analysis of chemotherapy treatment effects on infection-related morbidities in Schistosoma-endemic areas.

Fig 2. Odds ratios for morbidities related to schistosomiasis after treatment.

Color circles indicate summary odds ratios estimated by random effects meta-analysis for morbidity prevalences after treatment, as compared to pre-treatment levels. The (N) by each line indicates the number of studies included in that meta-analysis. *These numbers indicate, respectively, the odds ratio and its 95% confidence interval, the Cochran χ² value (where all χ² values had P < 0.001), and Higgin's and Thompson's I² statistic for heterogeneity estimation among the studies in each meta-analysis. Corresponding Forest plots of included studies and their summary statistics are included in Supplemental Information file S1 Fig.

Fig 3. Forest plot of the effect of anti-schistosomal treatment on hemoglobin levels.

Panel A, Forest plot and meta-analysis for the difference in mean hemoglobin levels, pre-intervention vs. post-intervention, for school-age subjects. Panel B, Forest plot and meta-analysis for the difference in mean hemoglobin levels, pre-intervention vs. post-intervention, for adult males.

Fig 4. Log odds ratio for post-treatment hepatomegaly, according to egg reduction rates achieved, by hepatic lobe.

The figure panels show the log10 of the post-treatment/pre-treatment odds ratio of morbidity, according to each study’s treatment mediated egg reduction rate (ERR), as related to hepatomegaly in the right lobe (upper left panel), n = 11, in the left lobe (upper right panel), n = 9, or where the lobe was not specified (lower panel), n = 5. N.B. the very low ERRs (< 0 to 20%) reflect observed post-treatment intensity outcomes seen among placebo-treated subgroups.

Fig 5. Log odds ratio of periportal fibrosis according to post-treatment egg reduction rate.

The graph shows the log10 of the post-treatment/pre-treatment odds ratio for periportal fibrosis according to each study’s post-treatment egg reduction rate (ERR), n = 10.

Fig 6. Log odds ratio of hematuria, proteinuria, and bladder abnormalities according to post-treatment egg reduction rate.

The panels show the log10 of the post-treatment/pre-treatment odds ratio of morbidity according to post-treatment egg reduction rate (ERR) as related to hematuria (upper left panel), n = 14, proteinuria (upper right panel), n = 9, or urinary bladder abnormalities detected on ultrasound (lower panel), n = 12.