Testing an infection model to explain excess risk of preterm birth with long-term iron supplementation in a malaria endemic area

Background In view of recent evidence from a randomized trial in Burkina Faso that periconceptional iron supplementation substantially increases risk of spontaneous preterm birth (< 37 weeks) in first pregnancies (adjusted relative risk = 2.22; 95% CI 1.39–3.61), explanation is required to understand potential mechanisms, including progesterone mediated responses, linking long-term iron supplementation, malaria and gestational age. Methods The analysis developed a model based on a dual hit inflammatory mechanism arising from simultaneous malaria and gut infections, supported in part by published trial results. This model is developed to understand mechanisms linking iron supplementation, malaria and gestational age. Background literature substantiates synergistic inflammatory effects of these infections where trial data is unavailable. A path modelling exercise assessed direct and indirect paths influencing preterm birth and gestation length. Results A dual hit hypothesis incorporates two main pathways for pro-inflammatory mechanisms, which in this model, interact to increase hepcidin expression. Trial data showed preterm birth was positively associated with C-reactive protein (P = 0.0038) an inflammatory biomarker. The malaria pathway upregulates C-reactive protein and serum hepcidin, thereby reducing iron absorption. The enteric pathway results from unabsorbed gut iron, which induces microbiome changes and pathogenic gut infections, initiating pro-inflammatory events with lipopolysaccharide expression. Data from the trial suggest that raised hepcidin concentration is a mediating catalyst, being inversely associated with shorter gestational age at delivery (P = 0.002) and positively with preterm incidence (P = 0.007). A segmented regression model identified a change-point consisting of two segments before and after a sharp rise in hepcidin concentration. This showed a post change hepcidin elevation in women with increasing C-reactive protein values in late gestation (post-change slope 0.55. 95% CI 0.39–0.92, P < 0.001). Path modelling confirmed seasonal malaria effects on preterm birth, with mediation through C-reactive protein and (non-linear) hepcidin induction. Conclusions Following long-term iron supplementation, dual inflammatory pathways that mediate hepcidin expression and culminate in progesterone withdrawal may account for the reduction in gestational age observed in first pregnancies in this area of high malaria exposure. If correct, this model strongly suggests that in such areas, effective infection control is required prior to iron supplementation to avoid increasing preterm births. Trial registration NCT01210040. Registered with Clinicaltrials.gov on 27th September 2010