Study design
To determine the cause-and-effect relationship between hypothyroidism and preeclampsia or eclampsia, we conducted a two-sample MR. SNP was used as an instrumental variable. There are three crucial presumptions that must always be true in order to optimize the accuracy of the results. First, the selected IVs are associated with risk factors (hypothyroidism). Second, IVs is not associated with confounders between exposure (hypothyroidism) and outcome(preeclampsia or eclampsia). Third, IVs has no direct effect on outcome (preeclampsia or eclampsia), but only affect outcome through exposure (Hypothyroidism). (Figure 1)
Data sources
The Finnish database (FinnGen Consortium) provides a reliable tool for MR analyses [9]. In our study, summary-level statistics for hypothyroidism were obtained from a large-scale meta-analysis of the publicly available Finnish database (FinnGen Consortium) version R9, which involves exposure data hypothyroidism (26,306 cases and 187,684 controls,16,380,461SNPs), and outcome data pre-eclampsia or eclampsia ( 3903 cases and 114,735 controls,16,379,723SNPs).A threshold of P < 0.05 was used to select SNPs and all SNPs and associated data were obtained from studies that analysed only populations of European descent separately in order to remove population stratification bias.
Genetic instrumental variable selection criteria
SNPs with strong associations (P<0.05) and independent inheritance (R2 < 0.001, kb = 10, 000) were selected from the hypothyroidism and pre-eclampsia or eclampsia GWAS databases with no evidence of linkage disequilibrium (LD) in the pooled RA or infertility GWAS data.In addition, it is of genome-wide significance to see if these SNPs are associated with potential risk factors from the PhenoScanner database and to remove SNPs that are associated with these potential confounders. (http://www.phenoscanner.medschl.cam.ac.uk/). We estimate IVs strength using the F-statistic to prevent bias brought on by weak IVs. F>10 indicates that the instrument has sufficient strength, which means that the IVs has sufficient estimation effect on the subsequent MR analysis without weak instrumental bias. When F<10, it indicates a weak SNP and should be eliminated.(R2 stands for the exposure variance explained by each IVs, N for the GWAS sample size, and K for the quantity of SNPs examined by MR.)
![](data:image/png;base64,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)
MR analysis
We employed the MR-Egger, weighted median, and random effects inverse variance weighting (IVW) in the MR analysis.The principal method of MR analysis used the IVW methodology to evaluate the causal effects between the exposure and the result since it gave accurate causal estimates in the absence of directional pleiotropy[10]. MR-Egger and weighted median were used to complement the analysis. For IVs with P < 0.05 from the IVW analysis, we performed MR-PRESSO with the MRPRESSO package, which detects, removes potentially polymorphic IVs (outliers) and provides estimates adjusted for outliers.
Sensitivity analysis
We conducted tests for heterogeneity and pleiotropy to ensure the stability of the results of the MR analysis. Cochran's Q test was used to calculate the level of heterogeneity, and MR-Egger regression and IVW methods were used to detect heterogeneity in IVs, with P<0.05 reflecting the presence of considerable heterogeneity. We checked for pleiotropy by MR-Egger intercept regression, and P < 0.05 was considered significant pleiotropy, and then the study results were unreliable. In addition, we also conducted a "leave-one-out" sensitivity analysis, in which one SNP was sequentially missing from the MR reanalysis to singling out the potential effects of SNPs.
Finally, the visualization results of the MR analysis are presented through scatter plots and forest plots. The analysis was performed using R software (4.2.3) and RStudio (2023.03.0), as well as the R packages TwoSampleMR and MR-PRESSO.