Cardiometabolic Risk Factors as Causal Mediators of the Relationship Between High Body Mass Index and Chronic Kidney Disease: A Two-Step Mendelian Randomization Study and Mediation Analyses

Introduction: Observational studies suggest that obesity is a risk factor of chronic kidney disease (CKD). It is unclear whether this reflects a direct causal effect of obesity or an effect mediated by obesity-related metabolic changes. Hypotheses: We hypothesised that 1) obesity is a causal risk factor of CKD defined as eGFR <60mL/min/1.73m 2 and 2) this causal effect is mediated through changes in low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterols, triglycerides, glucose, blood pressure and C-reactive protein (CRP) levels. Methods: To assess causality and investigate mediation, we used a combination of observational, two-step Mendelian randomization, and mediation analyses on individual-level data from two cohorts of the Danish general population (N=117,213). Results: Observationally, higher BMI was associated with a lower eGFR (change in eGFR; -0.81, confidence interval [CI]: -0.89- -0.73)) and a higher risk of CKD (odds ratio [OR]:1.19; 95% CI: 1.16-1.22) per 5kg/m 2 higher BMI (p for trend for both <2.9 x10 -40 ). Mendelian randomization analyses supported a causal effect of higher BMI on lower eGFR (change in eGFR -17.3 mL/min/1.73m 2 (95%CI: -19.3- -15.3), and risk of CKD (risk ratio [RR]6.99 (95%CI: 1.63-30.1) per 5kg/m 2 higher BMI. Measured BMI and the BMI allele score used in Mendelian randomization analyses were associated with changes in levels of LDL and HDL cholesterols, triglycerides, glucose, blood pressure, and CRP. Observational and Mendelian randomization analyses showed evidence to support a causal effect of changes in levels of systolic and diastolic blood pressure, LDL cholesterol and CRP with lower eGFR and higher risk of CKD. Of the excess risk of CKD coming from high BMI, elevated systolic and diastolic blood pressure mediated 18%(95%CI:14-22) and 9% (7-11), respectively. 9% (95%CI: 8-12) was mediated by elevated LDL cholesterol and 29% (95%CI: 23-35) by elevated CRP. Conclusion: Triangulating evidence from observational and Mendelian randomization analyses provided evidence to support a causal role of obesity on the risk of CKD and that this causal effect may be partly mediated by elevated systolic and diastolic blood pressure, CRP and LDL cholesterol levels.