Unveiling causal relationships between tobacco use phenotypes and neuroimaging: Insights from bidirectional Mendelian randomization and bibliometric analysis

Smoking is considered the most common addictive behavior worldwid. However, it is not systematically investigated whether there are bidirectional causal associations between tobacco use and neuroimaging, which might provide potential neural biomarkers or therapeutic neuro-targets for tobacco abuse or rehabilitation. In this study, using Mendelian randomization (MR), we explored both forward and reverse causal relationships between 4 tobacco use phenotypes (including Cigarettes per day, Smoking initiation, Age of initiation and Smoking cessation) and 209 neuroimaging features involving brain function and structure. Besides, we conducted a bibliometric analysis to find corresponding global trends of knowledge and specialized research areas. After Bonferroni correction, positive causal associations between cerebral global functional connectivity (FC) and smoking cessation, as well as between age of initiation and default mode network (DMN) structural connectivity (SC) were revealed; there were negative causal associations between mean diffusivity of left medial lemniscus and cigarette per day. Up to now, we also reveal that functional neuroimaging especially FC dominated smoking research field. Therefore, the discovery of causal relationships between smoking-related phenotypes and multi-modal cerebral alterations will promote a comprehensive understanding of tobacco addiction from cerebral function and structure perspective, facilitating the progression of personal treatment and risk prediction.