Genetic association between celiac disease and chronic kidney disease: a two-sample Mendelian randomization study

Figures & data

Table 1. Detailed data information of the Mendelian randomization study.

Phenotype	Sample size	Sample Case	Sample Control	Ancestry	Number of GVs	Reference
Celiac disease	24269	12041	12228	European	38037	[21]
Chronic kidney disease	117165	12385	104780	European	2179497	[23]
Glomerular filtration rate	133,413	–	–	European	2116469	[23]

Figure 1. Flowchart of the bidirectional two-sample Mendelian randomization study. The design hypotheses are that the genetic variants are associated with celiac disease traits, but not with confounders, and the genetic variants are associated with the risk of CKD and the eGFRcrea only through celiac disease traits. Similar hypotheses are applicable for the reverse MR analysis.

Assumption i: IVs are significantly associated with celiac disease at a genome-wide level; assumption ii: IVs must remain independent of any confounders; assumption iii: IVs solely influence CKD and eGFRcrea through celiac disease.

GV, genetic variants; CKD, chronic kidney disease; eGFRcrea, estimated glomerular filtration rate levels based on serum creatinine; IV, instrumental variables.

Table 2. Summary of genetic variants used to instrument celiac disease.

GV	CHR	POS	EA	OA	EAF	BETA	SE	P-value	F statistic
rs10752747	1	2524915	T	G	0.309	−0.116	0.020	5.05E-09	34.176
rs10892258	11	118579865	A	G	0.233	−0.150	0.022	1.73E-11	45.253
rs11979905	7	37437877	G	A	0.108	0.163	0.029	2.49E-08	31.070
rs12527282	6	137967252	T	C	0.280	−0.155	0.021	1.69E-13	54.346
rs12663317	6	32742827	C	A	0.085	−0.616	0.043	3.53E-47	208.225
rs130078	6	31118565	G	C	0.781	−0.577	0.024	3.35E-125	566.441
rs13030124	2	204694263	A	G	0.450	−0.105	0.019	2.40E-08	31.136
rs13119723	4	123218313	G	A	0.153	−0.309	0.028	7.60E-29	124.100
rs13198474	6	25874423	A	G	0.054	0.946	0.031	1.00E-200	920.516
rs2441467	2	61374149	C	T	0.418	0.105	0.019	1.70E-08	31.801
rs6498114	16	10964118	T	G	0.756	−0.131	0.021	5.83E-10	38.378
rs7162232	15	75115895	A	G	0.681	−0.117	0.020	7.97E-09	33.282
rs9296009	6	32114515	T	A	0.234	−0.954	0.029	1.00E-200	1087.020
rs931	6	33054619	A	G	0.307	0.618	0.020	1.00E-200	979.696

Figure 2. Scatter plots of Mendelian randomization. Scatter plot estimations of celiac disease as the expositional factor on CKD (A) and eGFRcrea (B). Each point in the scatter plot represents an IV. The line on each point reflects the 95% CI. A positive slope indicates that celiac disease had a positive effect on CKD. A negative slope indicates that celiac disease has a detrimental impact on eGFRcrea levels.

GV, genetic variants; CKD, chronic kidney disease; eGFRcrea, estimated glomerular filtration rate levels based on serum creatinine; IV, instrumental variables; CI, confidence interval.

Table 3. Causal effects of bidirectional Mendelian randomization.

Exposure	Methods	nGVs	OR	95%CI	P-value
CD versus CKD	MR Egger	14	1.034	0.999-1.070	0.07514414
	Weighted median	14	1.028	1.000-1.057	0.04958771
	Inverse variance weighted	14	1.027	1.003-1.050	0.02469948
	Weighted mode	14	1.030	1.003-1.057	0.04401706
CD versus eGFRcrea	MR Egger	14	0.996	0.994-0.998	5.00E-03
	Weighted median	14	0.996	0.994-0.998	1.68E-05
	Inverse variance weighted	14	0.997	0.995-0.998	2.94E-06
	Weighted mode	14	0.996	0.994-0.997	3.11E-04

Figure 3. Leave-one-out tests of the genetic risks of celiac disease on CKD and eGFRcrea. The red lines are the analysis results of random effects IVW.

CKD, chronic kidney disease; eGFRcrea, estimated glomerular filtration rate levels based on serum creatinine; IVW, inverse variance weighted.

Table 4. Pleiotropy and heterogeneity tests of Mendelian randomization.

Exposure	IVW_Q_pval	MR Egger_Q_pval	egger_intercept	egger_intercept _pval	RSSobs	Global P value
CD versus CKD	0.894	0.870	−4.36E-03	0.561	8.477	0.889
CD versus eGFRcrea	0.408	0.352	0.000	0.619	18.949	0.315

Trynka G, Hunt KA, Bockett NA, et al. Dense genotyping identifies and localizes multiple common and rare variant association signals in celiac disease. Nat Genet. 2011;43(12):1193–1201. doi:10.1038/ng.998.

 PubMed Web of Science ®Google Scholar

Pattaro C, Teumer A, Gorski M, et al. Genetic associations at 53 loci highlight cell types and biological pathways relevant for kidney function. Nat Commun. 2016;7(1):10023. doi:10.1038/ncomms10023.

 PubMedGoogle Scholar

Data availability statement

The data underlying this article were accessed from the IEU OpenGWAS consortium (https://gwas.mrcieu.ac.uk/).