A bidirectional Mendelian randomization study supports causal effects of kidney function on blood pressure

Abstract

Blood pressure and kidney function have a bidirectional relation. Hypertension has long been considered as a risk factor for kidney function decline. However, whether intensive blood pressure control could promote kidney health has been uncertain. The kidney is known to have a major role in affecting blood pressure through sodium extraction and regulating electrolyte balance. This bidirectional relation makes causal inference between these two traits difficult. Therefore, to examine the causal relations between these two traits, we performed two-sample Mendelian randomization analyses using summary statistics of large-scale genome-wide association studies. We selected genetic instruments more likely to be specific for kidney function using meta-analyses of complementary kidney function biomarkers (glomerular filtration rate estimated from serum creatinine [eGFRcr], and blood urea nitrogen from the CKDGen Consortium). Systolic and diastolic blood pressure summary statistics were from the International Consortium for Blood Pressure and UK Biobank. Significant evidence supported the causal effects of higher kidney function on lower blood pressure. Based on the mode-based Mendelian randomization method, the effect estimates for one standard deviation (SD) higher in log-transformed eGFRcr was -0.17 SD unit (95 % confidence interval: -0.09 to -0.24) in systolic blood pressure and -0.15 SD unit (95% confidence interval: -0.07 to -0.22) in diastolic blood pressure. In contrast, the causal effects of blood pressure on kidney function were not statistically significant. Thus, our results support causal effects of higher kidney function on lower blood pressure and suggest preventing kidney function decline can reduce the public health burden of hypertension.

Keywords: Mendelian randomization; biomarkers; blood pressure; chronic kidney disease; genome-wide association study; kidney function.

Figure 1 ∣. (a) The hypothesized bidirectional relations between kidney function and blood pressure (BP) depicted in a directed acyclic graph (DAG) in which the black arrows represent causal relations.

In our study, the primary kidney function measures were glomerular filtration rate (GFR) estimated from serum creatinine and not measured directly. Adapted from Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23:R89–R98. © The Author 2014. Published by Oxford University Press. (b) Selection of genetic instruments for GFR estimated from serum creatinine (eGFRcr) and systolic blood pressure (SBP), our primary traits. The centered graph represents the conceptual causal relations (black arrows) between estimated GFR and SBP, similar to the DAG in (a), except that estimated GFR levels were calculated from serum creatinine and therefore has 2 parts: a GFR and a biomarker-level component. The diagrams on the 2 sides with gray arrows represent the process of genetic instrument selection. LD, linkage disequilibrium; SNP, single-nucleotide polymorphism.

Figure 2 ∣. Estimates of causal effect of log(glomerular filtration rate estimated from serum creatinine [eGFRcr]) and log(blood urea nitrogen [BUN]) on systolic blood pressure (SBP) and diastolic blood pressure (DBP) calculated using the weighted mode method.

CI, confidence interval.

Figure 3 ∣. Estimates of causal effect of systolic blood pressure (SBP) and diastolic blood pressure (DBP) on log(glomerular filtration rate estimated from serum creatinine [eGFRcr]), log(blood urea nitrogen [BUN]), and chronic kidney disease (CKD) calculated using the weighted mode method.

CI, confidence interval.