Association between dried fruit intake and kidney function: research from univariate and multivariate Mendelian randomized studies

Abstract

Objectives: Observational studies have identified an association between dried fruit intake and kidney function. However, these studies have limitations such as vulnerability to confounders and reverse causality bias. Therefore, this study aimed to explore the potential causal relationship between dried fruit intake and kidney function.

Methods: A two-sample Mendelian randomization (MR) study was conducted using a large-scale genome-wide association study dataset to investigate the causal relationship between dried fruit intake and kidney function markers (blood urea nitrogen (BUN), creatinine (CR), uric acid (UA), cystatin C (CyC), hematuria, microalbuminuria). The main analytical method was inverse variance weighting. In addition, we applied the MR Egger and weighted median to assess the robustness of the results. Finally, Multivariate Mendelian randomization (MVMR) was used to estimate the direct effect of dried fruit intake on kidney function markers.

Results: The univariate MR analysis showed that increased dried fruit intake was associated with lower kidney function markers, including BUN (β: -0.171, 95% confidence interval (CI): -0.239 to -0.102, p = 1.063 × 10^-6), CR (β: -0.205, 95% CI: -0.311 to -0.099, p = 1.455 × 10^-4), UA (β = -0.317, 95% CI: -0.384 to -0.249, p = 4.439 × 10^-20), and CysC (β = -0.323, 95% CI: -0.384 to -0.249, p = 1.074 × 10^-11); however, it was unrelated to hematuria and microalbuminuria. Causality persisted after performing MVMR analysis; however, with the addition of alcohol consumption and smoking as exposure factors, the causality for UA (β = -0.296, 95% CI: -0.523 to -0.068, p = 1.094 × 10^-2) and CysC (β = -0.238, 95% CI: -0.465 to -0.011, p = 4.024× 10^-2) weakened, while the causality for BUN (β = -0.038, 95% CI: -0.215 to 0.138, p = 6.698 × 10^-1) and CR (β = -0.038, 95% CI: -0.431 to 0.046, p = 1.347 × 10^-1) disappeared.

Conclusion: Increased dried fruit intake was associated with lower kidney function markers (BUN, CR, UA, and CysC) in the absence of smoking and alcohol consumption; however, the causal relationship between dried fruit intake and BUN and CR disappeared in the presence of smoking and alcohol consumption. These results provide a promising avenue for delaying the course of chronic kidney disease.

Keywords: Mendelian randomization; causal relationship; dried fruit intake; genome-wide association study; kidney function.

Figure 1

Study design.

Figure 2

Scatter plot (A), funnel plot (B), forest plot (C), and leave-one-out analysis (D) of single nucleotide polymorphisms (SNPs) associated with dried fruit intake based on blood urea nitrogen (BUN) level.

Figure 3

Scatter plot (A), funnel plot (B), forest plot (C), and leave-one-out analysis (D) of single nucleotide polymorphisms (SNPs) associated with dried fruit intake based on serum creatinine (CR) level.

Figure 4

Scatter plot (A), funnel plot (B), forest plot (C), and leave-one-out analysis (D) of single nucleotide polymorphisms (SNPs) associated with dried fruit intake based on serum uric acid (UA) level.

Figure 5

Scatter plot (A), funnel plot (B), forest plot (C), and leave-one-out analysis (D) of single nucleotide polymorphisms (SNPs) associated with dried fruit intake based on cystatin C (CysC) level.