Heavy smoking linked to atrophy in Alzheimer’s brain regions

By Dr. Priyom Bose, Ph.D.Reviewed by Lauren HardakerJul 25 2025

New MRI research reveals that heavy smoking can shrink key brain regions tied to memory and cognition, and excess weight might intensify the damage, raising fresh questions about dementia prevention.

Study: Smoking predicts brain atrophy in 10,134 healthy individuals and is potentially influenced by body mass index. Image credit: fongbeerredhot/Shutterstock.com

A recent study in Npj Dementia investigated the relationship between smoking and brain atrophy and whether body mass index (BMI) influences this association. The study found that smokers had significantly lower gray and white matter brain volumes than non-smokers. When BMI was included in statistical models, the relationship between smoking pack-years and brain volume loss was reduced, suggesting a possible mediating effect rather than direct causation.

Neurodegenerative disorder: Prevalence and risk factors

A neurodegenerative disorder occurs when nerve cells in the brain and nervous system progressively lose their function, resulting in a decline in physical and cognitive abilities. Alzheimer’s disease (AD) is the most common type of dementia, affecting memory, cognitive function, and behavior.

An increased prevalence of dementia has been recorded worldwide. A recent study has estimated that approximately 47 million people across the world have been diagnosed with dementia. This number is predicted to increase by 10 million new cases each year.

Multiple studies have identified early, midlife, and late-life risk factors for dementia. Smoking is a risk factor that contributes to up to 14% of dementia cases worldwide. Toxins present in cigarette smoke may induce neuroinflammation, a mechanism closely linked to AD. In addition to dementia, previous research has also shown that cigarette smokers are at a high risk of many diseases, such as cerebrovascular disease and respiratory ailments.

While prior meta-analyses have linked smoking to increased dementia risk, few large-scale studies have examined how smoking history and intensity directly relate to Magnetic resonance imaging (MRI)-measured brain atrophy, a biomarker of neurodegeneration. To assess this, the association between smoking and brain atrophy, and loss of brain tissue from shrinkage or death of neurons with reduced neuronal connections, must be evaluated.

Researchers generally track brain atrophy for AD and other neurodegenerative disorders through neuroimaging by volume loss on T1-weighted structural imaging, which is distinct from aging. MRI is performed to assess cerebral volume loss, a neurodegeneration biomarker.

Not many large-scale studies have investigated the association between smoking and brain atrophy based on MRI-measured brain volume loss, which could play a crucial role in determining how significantly smoking contributes to cognitive decline and AD.

About the study

The current study tested the hypothesis that individuals with a history of smoking experience higher brain atrophy at the whole brain and regional lobar levels than non-smokers.

A total of 10,134 participants from four study sites, aged 18 to 97, were selected for this study. All participants underwent a non-contrast whole-body MRI scan. Before imaging, they completed questionnaires, from which their demographic information, medical history, and smoking status were obtained. Each participant provided information about the number of packs they smoked per day and the number of years they smoked.

Based on the questionnaire responses, participants were grouped as the smoking group (a non-zero pack-year value) and the non-smoking group (zero pack years). Pack-years corresponds to a measure of tobacco exposure for assessing smoking history and associated risks. The smoking group comprised 3,292 participants, while the non-smoking group comprised 6,842 individuals.

The current study used the FastSurfer network, an extensively validated deep-learning pipeline, to quantify brain volumes from 3D T1 scans. A deep-learning model was also used to segment the intracranial volume (ICV).

A regression analysis was performed in smokers to investigate the relationship between Pack Years of Smoking and brain regions in two different models: Model 1 (adjusted for age, sex, and study site) and Model 2 (adjusted for age, sex, site, and BMI).

Study findings

In comparison to the non-smoking group, participants belonging to the smoking group were more frequently women, Caucasian, had higher BMI, were older, and had higher rates of type 2 diabetes mellitus and hypertension. The smoking group had a mean pack year of 11.93.

Groupwise regional comparisons revealed lower brain volumes in smoking versus non-smoking groups. A Pearson bivariate correlation indicated a modest positive correlation between higher BMI and increased smoking pack years. By comparing model 1 and model 2, the current study observed a weakening of statistical significance and effect sizes across 11 brain regions when BMI was added, suggesting a possible, but not definitively proven, mediating effect of BMI in the association between the increase in smoking pack years and reduced brain volumes.

Importantly, smokers still showed significant atrophy in multiple regions, including Alzheimer’s disease-relevant areas such as the hippocampus, posterior cingulate, and precuneus, even when adjusting for BMI.

Conclusions

The current study revealed that individuals with a history of smoking and higher pack years of smoking had brain atrophy. The preliminary findings also indicate that BMI could play a potential and exploratory role in the association between cigarette smoking and brain volume loss. Therefore, obesity and smoking are two risk factors that could be exploited in the future to prevent dementia, including AD.

In the future, more research is required to examine the potential mediating effects of white matter hyperintensity volume and brain atrophy concerning smoking history and pack-years.

The key strength of this study lies in the analysis of a large cohort with a smoking history and quantitative structural brain imaging.  Furthermore, it enabled the measurement of regional brain volume at risk for AD pathology, such as the hippocampus, posterior cingulate, and precuneus.

Despite the strengths, the cross-sectional design limited the authors' ability to conclude causation. The study design lacked the temporal resolution necessary for reliable mediation or moderation testing. Additionally, the study did not include cognitive testing or Alzheimer’s biomarkers, such as amyloid or tau, which limits the ability to link observed brain atrophy to dementia directly. Therefore, the role of BMI in the relationship between smoking and brain atrophy requires more longitudinal analysis for validation.

Download your PDF copy now!