Genetic determinants of increased body mass index mediate the effect of smoking on increased risk for type 2 diabetes but not coronary artery disease

Abstract Clinical observations have linked tobacco smoking with increased type 2 diabetes risk. Mendelian randomization analysis has recently suggested smoking may be a causal risk factor for type 2 diabetes. However, this association could be mediated by additional risk factors correlated with smok...

Full description

Saved in:
Bibliographic Details
Published in:Human molecular genetics 2020-11, Vol.29 (19), p.3327-3337
Main Authors: Thom, Christopher S, Ding, Zhuoran, Levin, Michael G, Damrauer, Scott M, Lee, Kyung Min, Lynch, Julie, Chang, Kyong-Mi, Tsao, Philip S, Cho, Kelly, Wilson, Peter W F, Assimes, Themistocles L, Sun, Yan V, O’Donnell, Christopher J, Vujkovic, Marijana, Voight, Benjamin F
Format: Article
Language:English
Subjects:
Association Studies
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Clinical observations have linked tobacco smoking with increased type 2 diabetes risk. Mendelian randomization analysis has recently suggested smoking may be a causal risk factor for type 2 diabetes. However, this association could be mediated by additional risk factors correlated with smoking behavior, which have not been investigated. We hypothesized that body mass index (BMI) could help to explain the association between smoking and diabetes risk. First, we confirmed that genetic determinants of smoking initiation increased risk for type 2 diabetes (OR 1.21, 95% CI: 1.15–1.27, P = 1 × 10−12) and coronary artery disease (CAD; OR 1.21, 95% CI: 1.16–1.26, P = 2 × 10−20). Additionally, 2-fold increased smoking risk was positively associated with increased BMI (~0.8 kg/m2, 95% CI: 0.54–0.98 kg/m2, P = 1.8 × 10−11). Multivariable Mendelian randomization analyses showed that BMI accounted for nearly all the risk smoking exerted on type 2 diabetes (OR 1.06, 95% CI: 1.01–1.11, P = 0.03). In contrast, the independent effect of smoking on increased CAD risk persisted (OR 1.12, 95% CI: 1.08–1.17, P = 3 × 10−8). Causal mediation analyses agreed with these estimates. Furthermore, analysis using individual-level data from the Million Veteran Program independently replicated the association of smoking behavior with CAD (OR 1.24, 95% CI: 1.12–1.37, P = 2 × 10−5), but not type 2 diabetes (OR 0.98, 95% CI: 0.89–1.08, P = 0.69), after controlling for BMI. Our findings support a model whereby genetic determinants of smoking increase type 2 diabetes risk indirectly through their relationship with obesity. Smokers should be advised to stop smoking to limit type 2 diabetes and CAD risk. Therapeutic efforts should consider pathophysiology relating smoking and obesity.
ISSN:0964-6906
1460-2083
DOI:10.1093/hmg/ddaa193