More on Mendelian Randomisation

News Blog readers know that the CLAHRC WM Director loves Mendelian randomisation studies, originally proposed by his erstwhile colleagues, Gray and Whitley.[1] The method has been used to crack open the story regarding lipids and coronary artery disease.[2] Everyone knows that low density lipoproteins are bad news – these fats clog up arteries. The association is confirmed by Mendelian studies. But what about those two old chestnuts, high density lipoproteins (HDLs) and triglycerides? In observational studies HDLs are consistently associated with reduced risk of coronary disease.[3] While triglyceride levels are associated with increased coronary risk, this effect disappears once confounders have been controlled in multi-variable analysis.[3] However, Mendelian randomisation tells a completely different story – HDLs are not associated with coronary risk, while triglycerides are.[4] [5] What is going on here? That is to say, why do the observational studies and the Mendelian studies give such different answers with respect to HDLs and triglycerides? More curious still, why does the association between triglyceride and coronary artery disease confirmed by Mendelian randomisation disappear after controlling for confounders? This is not entirely clear, but as HDL levels drop, so triglycerides tend to rise. Hence controlling for triglyceride levels when examining HDLs, and vice-versa, will give the wrong result. This may be yet another example of ‘over controlling’ but including in a multi-variable analysis / logistic regression variables that have a causal interaction with the explanatory variable of interest.[6]

— Richard Lilford, CLAHRC WM Director

References:

1. Gray R & Wheatley K. How to avoid bias when comparing bone marrow transplantation with chemotherapy. Bone Marrow Transplant. 1991;7(s3):9-12.
2. Emdin CA, Khera AV, Kathiresan S. Mendelian Randomization. JAMA. 2017; 318(19): 1925-6.
3. Di Angelantonio E, Sarwar N, Perry P, et al.; Emerging Risk Factors Collaboration. Major lipids, apolipoproteins, and risk of vascular disease. JAMA. 2009; 302(18): 1993-2000.
4. Do R, Willer CJ, Schmidt EM, et al. Common variants associated with plasma triglycerides and risk for coronary artery disease. Nat Genet. 2013; 45(11): 1345-52.
5. Frikke-Schmidt R, Nordestgaard BG, Stene MCA, et al. Association of loss-of-function mutations in the ABCA1 gene with high-density lipoprotein cholesterol levels and risk of ischaemic heart disease. JAMA. 2008; 299(21): 2524-32.
6. Lilford RJ. A Very Interesting Paper Using Mendelian Randomisation to Determine the Effect of Extra Years of Education on Heart Disease. NIHR CLAHRC West Midlands News Blog. 10 November 2017.