Non-alcoholic fatty liver disease and gestational diabetes mellitus: a bidirectional two-sample mendelian randomization study

Abstract

Purpose: Previous observational studies have revealed a potential link between non-alcoholic fatty liver disease (NAFLD) and gestational diabetes mellitus (GDM), but their causal relationship remains unclear. Thus, this study aimed to examine whether a causal link exists between genetically determined NAFLD and GDM.

Methods: Utilizing publicly accessible genome-wide association studies (GWAS), a two-sample bidirectional Mendelian randomization (MR) analysis was conducted. The GWASs data pertaining to NAFLD and GDM were obtained from the UK Biobank Consortium and FinnGen database in primary analysis, respectively. The random-effects inverse variance weighted (IVW) method was utilized as primary analysis method. Several sensitivity analyses were utilized to verify the robustness of the results. Additionally, we also analyzed the causal effect of potential shared influencing factors on these two conditions.

Results: The result of the IVW method showed that there was no significant causal relationship between genetically determined NAFLD and GDM (OR = 0.98, 95% CI: 0.90-1.07, P = 0.691). Similarly, our reverse MR analysis failed to detect a significant causal effect of GDM on NAFLD (OR = 1.14, 95% CI: 0.97-1.36, P = 0.118). Sensitivity analyses further confirmed the robustness of the results. Moreover, we found that genetically determined body mass index, waist-to-hip ratio, triglycerides, and television viewing time may be positively correlated with NAFLD and GDM, while high-density lipoprotein cholesterol and apolipoprotein A-I may both be negatively correlated with NAFLD and GDM.

Conclusions: The current bidirectional MR study failed to provide sufficient genetic evidence for the causal relationship between NAFLD and GDM.

Keywords: Causality; Gestational diabetes mellitus; Mendelian randomization; Nonalcoholic fatty liver disease.

Fig. 1

The flow chart of bidirectional primary MR analyses. IVs, instrumental variants; SNP, single-nucleotide polymorphism. NAFLD, non-alcoholic fatty liver disease; GDM, gestational diabetes mellitus; ① relevance assumption; ② independence assumption; ③ exclusion-restriction assumption

Fig. 2

The flow chart of the causal effect of potential shared influencing factors on NAFLD or GDM in the secondary MR analysis. IVs, instrumental variants; SNP, single-nucleotide polymorphism. NAFLD, non-alcoholic fatty liver disease; GDM, gestational diabetes mellitus; BMI, body mass index; WHR, waist-to-hip ratio; TG, triglycerides; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; Apo A-1, apolipoprotein A-I; Apo B, apolipoprotein B; HOMA-B, homeostasis model assessment of beta-cell function; HOMA-IR, homeostasis model assessment of insulin resistance; TV, Time spent watching television; Computer, Time spent using computer; Driving, Time spent driving

Fig. 3

The results of bidirectional primary MR analyses

Fig. 4

The results of causal effect of potential shared influencing factors on NAFLD or GDM