Affiliation:
1. Systems Biology of Complex Diseases Centro de Altos Estudios en Ciencias Humanas y de la Salud (CAECIHS), Universidad Abierta Interamericana, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) Buenos Aires Argentina
2. Clinical and Molecular Hepatology Centro de Altos Estudios en Ciencias Humanas y de la Salud (CAECIHS), Universidad Abierta Interamericana, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) Buenos Aires Argentina
3. Universidad Maimónides Buenos Aires Argentina
Abstract
AbstractBackgroundIt has been consistently shown that obesity contributes directly to arterial hypertension and cardiovascular disease (CVD), independently of other risk factors. Likewise, non‐alcoholic fatty liver disease (NAFLD) is acknowledged as a contributor and a risk enhancer for CVD.ObjectivesWe tested the hypothesis of a causal role of NAFLD in the effect of obesity on arterial hypertension.MethodsUsing causal mediation analysis, we quantified the magnitude of the body mass index (BMI) effect on arterial hypertension and CV‐traits mediated by NAFLD. First, we analysed data from 1348 young adults in the Bogalusa Heart Study (BHS), a cohort aimed at assessing the natural history of CVD. Then, we used data from 3359 participants of the National Health and Nutrition Examination Survey (2017–2018 cycle, NHANES) to replicate the findings.ResultsWe found that roughly 92% of the effects of BMI on arterial hypertension in the BHS and 51% in the NHANES population are mediated by NAFLD. In addition, indirect effects of BMI on systolic (SBP) and diastolic (DBP) blood pressure, and heart rate (HR) through NAFLD explained up to 91%, 93%, and 100% of the total effect, respectively, in the BHS. In the NHANES survey, indirect effects of BMI through NAFLD on CV traits explain a significant proportion of the total effects (SBP = 60.4%, HR = 100%, and pulse pressure = 88%).ConclusionNAFLD mediates a substantial proportion of the effect of obesity on the presence of hypertension and CV‐parameters independently of relevant covariates. This conclusion has implications for clinical management.
Funder
Agencia Nacional de Promoción Científica y Tecnológica