Chronic kidney disease of undetermined cause (CKDu): an exploratory study analysing associations between an emerging epidemic and environmental risk factors

Abstract

An epidemic of chronic kidney disease is occurring in rural communities in low-and middle-income countries. Characterised by chronic impairment of kidney function, this disease is not associated with known risk factors diabetes or hypertension and occurs primarily in communities characterised by a hot climate with reliance on agricultural work. Hypothesised environmental risk factors include heat stress, heavy metals and pesticide exposure, and low altitude. As it is unclear whether there is a unified aetiology, this disease has been termed chronic kidney disease of undetermined cause (CKDu). In this thesis I aimed to investigate whether the above environmental exposures are CKDu risk factors. Firstly, I conducted a global geographical analysis of CKDu using the global burden of disease database, modelling age-standardized death rates to highlight potential high-risk regions. To analyse CKDu at the individual level I developed a methodological framework to analyse satellite imagery and measure potable water sources and blood for heavy metals. I use Bayesian kernel machine regression (BKMR) to investigate associations between kidney function and heavy metals in blood. I analysed data from Sri Lanka and India, which are CKDu endemic, but aetiologies are unknown. Using satellite-derived data, associations between kidney function temperature, altitude, and vicinity to cropland were investigated. I then analysed data in Malawi which is suspected as high-risk based on similarities of climate and agriculture typical of endemic regions. Via a nested cross-sectional study, associations between kidney function and satellite-derived data were conducted. Heavy metals concentrations in potable water and blood samples were analysed using BKMR. Results suggest CKDu is not present in Malawi. There were no associations between temperature, altitude, heavy metals, or vicinity to cropland and reduced kidney function. BKMR showed potential interactions between reduced kidney function and low blood Cadmium concentrations. In Sri Lanka and India, elevated temperature was not associated with CKDu. However, in Sri Lanka associations were observed with agricultural work; in India vicinity to cropland was associated. My findings challenge the current hypothesis that CKDu is a heat-related disease. My environmental exposure framework can be used as a foundation for further research facilitating international comparisons of environmental risk factors for CKDu.