Table 3.

Results of causal analysis of pulmonary embolism (exposure) and human blood metabolites (outcome)

Exposure group	Outcome group	Methods	Beta	SE	95%CI	P value
pulmonary embolism	1-stearoylglycerophosphoethanolamine	MR Egger	− 0.01012	2.555	(− 5.0179,4.9977)	0.9971
		Weighted median	−0.02833	0.9651	(−1.9199,1.8633)	0.9766
		Inverse variance weighted	0.2539	0.8356	(−1.4192,1.8917)	0.7612
pulmonary embolism	X-12100--hydroxytryptophan	MR Egger	−1.129	1.59	(−4.2454,1.9874)	0.5288
		Weighted median	−0.3097	0.6186	(−1.5222,0.9028)	0.6166
		Inverse variance weighted	0.08013	0.5019	(−0.9036,1.0639)	0.8732
pulmonary embolism	X-12029	MR Egger	−1.365	1.078	(−3.4779,0.7479)	0.2946
		Weighted median	0.007778	0.4192	(−0.8139,0.8294)	0.9852
		Inverse variance weighted	0.3129	0.4308	(−0.5315,1.1573)	0.4675
pulmonary embolism	X-11412 subunit beta-2	MR Egger	−1.613	1.609	(−4.7666,1.5406)	0.3899
		Weighted median	−0.456	0.6179	(−1.6671,0.7551)	0.4605
		Inverse variance weighted	−0.1042	0.5212	(−1.1258,0.9174)	0.8415
pulmonary embolism	N1-methyladenosine	MR Egger	0.1411	0.9921	(−1.8034,2.0856)	0.8959
		Weighted median	−0.3065	0.3746	(−1.0407,0.4277)	0.4133
		Inverse variance weighted	−0.3298	0.3075	(−0.9325,0.2729)	0.2835
pulmonary embolism	valine	MR Egger	−0.2964	1.072	(−2.3975,1.8047)	0.8002
		Weighted median	0.4962	0.3983	(−0.2844,1.2769)	0.2129
		Inverse variance weighted	0.5772	0.3468	(−0.1025,1.2569)	0.09607

Tests were considered statistically significant at P values < 0.05. Horizontal lines denote 95% confidence intervals. The MR base platform (http://app.mrbase.org/) was used for MR analyses