# Association between Apolipoprotein B and diabetic nephropathy: insights from the National Health and Nutrition Examination Survey 2007–2016 and Mendelian randomization analysis

## TL;DR

This study explores the link between Apolipoprotein B and diabetic nephropathy using data from a national health survey and genetic analysis, finding a correlation but no direct causal relationship.

## Contribution

The study combines epidemiological and genetic approaches to investigate ApoB's role in diabetic nephropathy, revealing a correlation but no causal link.

## Key findings

- Elevated ApoB levels (>1.2 g/L) are significantly associated with increased diabetic nephropathy risk.
- A positive linear association exists between ApoB levels exceeding 1.12 g/L and diabetic nephropathy risk.
- Mendelian randomization analysis found no direct causal effect between ApoB and diabetic nephropathy.

## Abstract

This study aimed to investigate the role of Apolipoprotein B (Apo B) in diabetic nephropathy (DN) from epidemiological and genetic perspectives.

We employed weighted multivariable-adjusted logistic regression to assess the relationship between ApoB and DN risk, utilizing data from the National Health and Nutrition Examination Survey spanning 2007–2016. Then, we used restricted cubic splines (RCS) to flexibly model and visualize the relation of predicted ApoB levels with DN risk. Subsequently, a bidirectional two-sample Mendelian randomization study using genome-wide association study summary statistics was performed. The primary Inverse Variance Weighted method, along with supplementary MR approaches, was employed to verify the causal link between ApoB and DN. Sensitivity analyses were conducted to confirm the robustness of the results.

Our observational study enrolled 2242 participants with diabetes mellitus from NHANES. The multivariable logistic regression model indicated that elevated ApoB levels (>1.2 g/L), compared to low levels (<0.8 g/L), were significantly associated with DN risk (P < 0.05). The RCS model revealed a positive linear association with the risk of DN when ApoB levels exceeded 1.12 g/L (OR = 1.29, 95% CI: 1.07–1.57, P = 0.008). However, the MR IVW method did not reveal a direct causal effect of DN on ApoB (OR: 0.976; 95% CI: 0.950–1.004; P = 0.095), nor a direct causal effect of ApoB on DN (OR: 0.837; 95% CI: 0.950–1.078; P = 0.428).

The evidence from observational studies indicates a positive correlation between ApoB levels exceeding 1.12 g/L and the onset of DN. However, the causal effects of ApoB on DN and vice versa were not supported by the MR analysis.