By Dr. Priyom Bose, Ph.D.Mar 27 2024Reviewed by Benedette Cuffari, M.Sc.

Previously, Mendelian randomization (MR) studies have explored the causal relationship between smoking and abdominal obesity using a single genetic variant for smoking heaviness. Similarly, a recent Addiction study uses multiple genetic instruments to estimate the causal relationship between smoking and abdominal obesity.

Study: Estimating causality between smoking and abdominal obesity by Mendelian randomization. Image Credit: kong-photo / Shutterstock.com

How does smoking affect obesity?

Smoking leads to several chronic disorders, particularly cardiovascular and respiratory diseases. In fact, smokers often have more abdominal fat as compared to non-smokers, which further increases their risk of cardiometabolic diseases.

It remains unclear whether the association between body fat distribution and smoking is causal. Genetic variants associated with exposure traits have been used as instrumental variables by MR studies to assess this potentially causal relationship. MR is similar to a naturally randomized controlled trial as, during conception, paternal and maternal alleles are randomly allocated.

Previously, MR studies have explored the causal relationship between the heaviness of smoking and abdominal obesity using a single genetic variant. Two studies noted no causal relationship, whereas a third suggested a causal link between the number of cigarettes smoked each day and the waist-hip ratio (WHR), even after controlling for the body mass index (BMI). 

About the study

The current causal analysis using summary effect estimates (CAUSE) study involved two-sample MR analyses to quantify the effect of smoking initiation, heaviness, and life-time smoking on abdominal adiposity. To this end, genome-wide association studies (GWAS) summary statistics were obtained from the GWAS and Sequencing Consortium of Alcohol and Nicotine Use (GSCAN), United Kingdom Biobank, and Genetic Investigation of Anthropometric Traits (GIANT) Consortium.

All study participants were of European ancestry. Exposure traits included smoking initiation, heaviness, and lifetime smoking were used, whereas outcome traits including WHR, as well as waist and hip circumferences (WC and HC) were used. The outcome traits were considered with and without adjustment for BMI.

Study findings

Lifetime smoking and smoking initiation causally increased abdominal adiposity, independent of socio-economic status, alcohol consumption, and other factors. Visceral fat or visceral adipose tissue (VAT) increased more than abdominal subcutaneous fat (ASAT).

The causal relationship between abdominal fat and smoking heaviness could not be established. However, reverse causality analyses indicated that smoking heaviness could be increased causally by abdominal adiposity.

Previous studies have used a single genetic variant in the CHRNA3/5 locus to establish causality between abdominal adiposity and smoking heaviness, both of which did not identify a causal relationship between these two factors. The Wald ratio estimates for the CHRNA3/5 locus in the current study observed causal effects; however, upon instrumenting smoking heaviness with all known genetic loci, this effect was not present. 

Consistent with previous studies, two-sample MR analyses with 13 smoking heaviness variants showed negative causality between BMI and smoking heaviness due to the CHRNA3/5 locus. Moreover, LHC-MR and CAUSE analysis failed to establish causality between lower BMI and smoking heaviness, thus suggesting a pleiotropic effect of the CHRNA3/5 locus on BMI and smoking heaviness, rather than a causal effect.

Smoking could lead to higher abdominal fat by increasing ASAT or visceral fat. The results for magnetic resonance imaging (MRI)-based adipose depot volumes suggested increased VAT to be primarily responsible for higher abdominal adiposity, rather than ASAT, which is consistent with the findings in existing research.

Conclusions

Lifetime smoking and smoking initiation may causally lead to higher abdominal and particularly visceral fat. Thus, public health efforts to reduce and prevent smoking could aid in lowering abdominal fat and the associated risk of chronic illnesses.

The strengths of the current study include the application of different complementary MR methods and sensitivity analyses, as well as the use of large-scale GWAS summary-level data to reduce reverse causality, sample overlap, and pleiotropic effects.

The main limitation of the present study involves the presence of residual pleiotropic effects and their influence on causal estimates. These effects could not be completely removed, despite performing multiple sensitivity analyses.

Additionally, in the sub-sample of current smokers, the sample size for body fat distribution was small. This restricted the statistical power of the analysis of smoking heaviness. Another limitation involved the inability to evaluate the effect of smoking cessation on body fat distribution.

Importantly, cigarettes depict unstandardized tobacco doses, which could have had a non-negligible impact on the accuracy of the estimates for smoking heaviness. Furthermore, the study population, which was restricted to individuals of European genetic ancestry, limits the generalizability of the study findings to other diverse populations.

Journal reference:

Carrasquilla, G. D., Garcia-Urena, M., Romero-Lado, M. J., & Kilpelainen, T. O. (2024). Estimating causality between smoking and abdominal obesity by Mendelian randomization. Addiction. doi:10.1111/add.16454