Causal associations between adiposity markers and nuclear magnetic resonance spectroscopy-measured lipids in plasma: a mendelian randomisation study in 110,000 mexican adults

Beryl Lin; Ramirez Diego Aguilar; Jesus Alegre-Diaz; Natalie Staplin; Herrington William G; Paulina Baca; Carballo Carlos Gonzalez; Raul Ramirez-Reyes; Fernando Rivas; Louisa Gnatiuc-Friedrichs; Michael Hill; Fredrik Romer; Jason Torres; Eirini Trichia; Rachel Wade; Center Regeneron Genetics; Rory Collins; Jonathan Emberson; Jaime Berumen; Pablo Kuri-Morales; Roberto Tapia-Conyer

doi:10.1530/endoabs.99.P71

P71

Prev Next

Section Contents Cite

Endocrine Abstracts (2024) 99 P71 | DOI: 10.1530/endoabs.99.P71

ECE2024 Poster Presentations Diabetes, Obesity, Metabolism and Nutrition (130 abstracts)

Causal associations between adiposity markers and nuclear magnetic resonance spectroscopy-measured lipids in plasma: a mendelian randomisation study in 110,000 mexican adults

Beryl Lin ^1,2 , Diego Aguilar Ramirez ¹ , Jesus Alegre-Diaz ^3,4 , Natalie Staplin ^1,5 , William G Herrington ^1,5 , Paulina Baca ^3,4 , Carlos González Carballo ^3,4 , Raul Ramirez-Reyes ^3,4 , Fernando Rivas ^3,4 , Louisa Gnatiuc-Friedrichs ^1,5 , Michael Hill ^1,5 , Fredrik Romer ^1,5 , Jason Torres ^1,5 , Eirini Trichia ^1,5 , Rachel Wade ^1,5 , Regeneron Genetics Center ⁶ , Rory Collins ¹ , Jonathan Emberson ^1,5 , Jaime Berumen ^3,4 , Pablo Kuri-Morales ^4,7 & Roberto Tapia-Conyer ⁴

Views

Author affiliations

¹Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health (NDPH), University of Oxford, Oxford, United Kingdom; ²Faculty of Medicine and Health, University of Sydney, Sydney, Australia; ³Experimental Research Unit from the Faculty of Medicine, National Autonomous University of Mexico, Mexico City, Mexico; ⁴School of Medicine, National Autonomous University of Mexico, Mexico City, Mexico; ⁵MRC Population Health Research Unit, Nuffield Department of Population Health (NDPH), University of Oxford, Oxford; ⁶Regeneron Genetics Center, Tarrytown, New york, United States; ⁷Instituto Tecnológico y de Estudios Superiores de Monterrey, Monterrey, Mexico

Background: The genotyped Mexico City Prospective Study (MCPS) cohort enables the causal effects of adiposity on lipids to be explored in an overweight Hispanic population with scarce statin use.

Methods: MCPS participants aged 35-84 years, with baseline nuclear magnetic resonance measurements of blood lipids, and not using lipid-lowering therapy at recruitment were included. Using the Mendelian randomisation one-sample ratio method, univariable and multivariable models with genetic scores for body-mass index (BMI) and waist-hip-ratio (WHR) as instruments estimated the unadjusted and mutually-adjusted effects of BMI and WHR on six nuclear magnetic resonance spectroscopy-measured plasma lipids: total cholesterol; low-density lipoprotein cholesterol (LDL-C); high-density lipoprotein cholesterol (HDL-C); triglycerides; apolipoprotein A1; apolipoprotein B. Interactions by age, sex and diabetes were explored.

Results: Among 110,669 included participants (mean age 52 years), mean (SD) BMI and WHR were 29.0 (4.9) kg/m² and 0.90 (0.07), respectively. Each 5 kg/m² higher genetically-predicted BMI (772 SNPs, 3.0% variation) was associated with 0.19SD higher triglycerides (95% CI 0.15-0.22), 0.14SD lower HDL-C (0.11-0.18), 0.11SD lower LDL-C (0.07-0.15), 0.06SD lower apolipoprotein A1 (0.02-0.09) and 0.03SD lower apolipoprotein B (-0.01-0.07). By contrast, each 0.075 unit higher genetically-predicted WHR (398 SNPs, 0.9% variation) was associated with 0.66SD higher triglycerides (0.57-0.74), 0.29SD lower HDL-C (0.21-0.37), 0.13SD lower LDL-C (0.05-0.20), 0.04SD lower apolipoprotein A1 (-0.03-0.12) and 0.10SD higher apolipoprotein B (0.03-0.18). Overall, higher BMI was associated with 0.09SD lower total cholesterol (0.05-0.12) and WHR was associated with 0.04SD lower total cholesterol (-0.03-0.12). Adjustment of BMI for WHR (or vice versa) had little impact on the effect size estimates. Associations were similar in men and women and irrespective of history of diabetes, but for several lipid indices the associations varied significantly depending on age. In particular, associations of BMI and WHR with triglycerides were substantially stronger at younger than at older ages.

Conclusion: In Mexican adults not on lipid-lowering therapy, genetically-predicted general and particularly central adiposity are associated with substantially higher triglycerides but moderately lower cholesterol levels. Triglycerides may be a more important lipid mediator of adiposity-associated atherosclerosis in this population than cholesterol.