Smoking in the scanner?

There is a new paper out in Scientific Reports, titled “Investigating the neural correlates of smoking: Feasibility and results of combining electronic cigarettes with fMRI”. This is a study that have managed to combine actual smoking with functional MRI (fMRI).

Most studies looking at brain processing of smoking run into trouble with MRI. This is because smoking and scanning do not go well together. Hospitals don’t allow smoking, things should generally not be on fire in the MRI scanner, and ventilation is an issue when you’re lying in a narrow bore. Because of this, we haven’t been able to properly look at the sensations and behaviour of smoking alongside the effects of nicotine (and other active products in cigarette smoke). This study tries to get around these practical problems and also look at the brain response to real-time smoking.

For the practical part, the study used e-cigarettes. E-cigarettes solve some of the problems with smoking in the scanner (fire and ventilation to some extent), but can cause image artifacts and may also contain metal. The paper shows how smaller types of e-cigarettes did not cause image artifacts plus were safe to use in the scanner from a metal point of view. E-cigarette smoking is a good mimic for traditional smoking, so this is a workable model of ‘the real thing’ that fits with MRI.

In terms of brain responses, the authors found activation in several brain regions associated with smoking e-cigarettes. These regions included motor cortex, insula, cingulate, amygdala, putamen, thalamus, globus pallidus and cerebellum. There were also (relative) deactivations in the ventral striatum and orbitofrontal cortex associated with smoking.


Image from the paper showing brain responses when participants were instructed to smoke. Red-yellow is activation and blue is deactivation.

Some of this activation is (unsurprisingly) linked to movement. The motor cortex activation (stronger on the left hand side, which correspond to right-hand side motion) is most likely due to movements associated with smoking. Similarly, cerebellar activation is often related to motion. Other regions are associated more with the effects of smoking. The putamen is part of a brain region called the striatum, which plays a role in reward and in supporting addiction. The ventral striatum (and orbitofrontal cortex) are associated with drug craving.

From a personal point of view, having worked a great deal with breathing, I am excited that the paper showed activation in the insula and cingulate. Both are structures involved in breathing and breathlessness tasks. However, without behavioural measures to link the findings to, it is hard to say what this activation means in this setting. It is important to remember that just because a similar activation pattern occurs with two different tasks, it doesn’t necessarily follow that the activation means the same. Each region of the brain typically handle more than one thing, particularly cortical regions.

The authors also found that activation patterns was similar both if the participants were told when to smoke and when to stop (first scan), and if they could smoke at will (second scan). However, in the second scan, the activation was weaker. The authors suggest that this could be because this task was more variable, meaning more between-subject variance and poorer timing (from a fMRI point of view). It could also be an order effect, as the subjects had more nicotine in their system in the second scan. This fits with lower activation in reward-related brain regions in the second scan. Or it could simply be because to smoke on command or whenever one wants to are different situations. Again, it is hard to tell why without other measures.

Nevertheless, this is an interesting paper, both from a methods point of view and for those interested in smoking processing and effects on the brain. It’s also written in a nice and easily accessible way. I’d recommend looking it up:

Reference: Matthew B. Wall, Alexander Mentink, Georgina Lyons, Oliwia S. Kowalczyk, Lysia Demetriou & Rexford D. Newbould. Investigating the neural correlates of smoking: Feasibility and results of combining electronic cigarettes with fMRI. Scientific Reports 7, Article number: 11352 (2017)
DOI: 10.1038/s41598-017-11872-z


Approaches to COPD

When I say I work on Chronic Obstructive Pulmonary Disease (COPD), I usually get one out of two responses: sage nodding or confused smiles. Invariably, the nodders are those who know someone with COPD or work within the field. If you have ever seen anyone suffer from the disease, you’ll remember it. But if you have never encountered COPD directly, you’d be forgiven for not knowing too much about it. It has a low profile compared to diseases of similar impact. Unfairly so.

By way of background, chronic respiratory disease kills almost as many people as lung cancer in England and Wales each year, with the total annual toll being approximately 29,000. COPD makes up a large part of these numbers. The WHO estimates that COPD will be the third leading cause of death worldwide in 2030. Those numbers alone should warrant attention.


In the UK, COPD is predominantly caused by smoking. The British Lung Foundation states that 80% of all cases are caused by long-term cigarette smoking, and that about 25% of all long-term smokers will develop the disease. Therein lies perhaps part of the problem with the public profile of COPD. There’s a disproportionate amount of blame to go around in ‘self-inflicted’ lung diseases, COPD included. This stigma is a barrier both to resource allocation (treatment availability, research funding) and to patients (treatment seeking behaviour), and contributes to making COPD a greater problem than it needs to be. Smoking cessation is important, but if we truly want to improve outcomes in COPD, we must let go of the moralising and focus on the medicine.


Furthermore, COPD is typically diagnosed late, its primary symptoms often confused for signs of aging or smoker’s cough. The treatment that many patients receive is too little too late. While it is not curable, it can be treated, and the effectiveness of treatment depends on early diagnosis. Increasing the profile of COPD, both with the public and healthcare professionals, is a way to remedy this.

I’ll let Mr Spock have the last word, as is appropriate:

nimoy(Leonard Nimoy, 1931-2015)