Effect of tobacco smoking on frontal cortical thickness development: A longitudinal study in a mixed cohort of ADHD-affected and -unaffected youth

Abstract

Smoking rates are particularly high during adolescence and young adulthood, when the brain is still undergoing significant developmental changes. Cross-sectional studies have revealed altered brain structure in smokers, such as thinner frontal cortical areas. Attention-deficit/hyperactivity disorder (ADHD) increases the risk of becoming nicotine-dependent, and has also been associated with abnormalities in frontal gray matter structure. The present study examines the relationships between smoking, cortical thickness and ADHD symptoms in a longitudinal design that compares adolescent and young adult smokers (n=44; 35 ADHD-affected) and non-smokers (n=45; 32 ADHD-affected) on frontal cortical thickness. Average frontal cortical thickness was estimated through structural magnetic resonance imaging (MRI) at two time points (mean ages 17.7 and 21.1 years), on average 3.4 years apart. Smokers had a 2.6% thinner frontal cortex than non-smokers and this difference was not explained by ADHD or other confounding factors. The rate of cortical thinning across the 3.4-year MRI measurement interval was similar in the total group of smokers compared to non-smokers. However, speeded thinning did occur in smokers who had started regular smoking more recently, in between the two measurements. These novel regular smokers did not differ significantly from the non-smokers at baseline. This suggests that the thinner frontal cortex was not a predisposing factor but rather a consequence of smoking. Although smokers had more ADHD symptoms overall, smoking did not influence the developmental course of ADHD symptoms.

Introduction

Smoking rates are particularly high during adolescence and young adulthood, when the brain is still undergoing significant developmental changes (Lydon et al., 2014). The presence of attention-deficit/hyperactivity disorder (ADHD) increases the risk of becoming nicotine-dependent (Groenman et al., 2013, Lee et al., 2011). In the current study, we examined the interplay between tobacco smoking, brain development, and ADHD symptoms by investigating the longitudinal effect of smoking on thickness of the frontal cortex in adolescents and young adults with and without ADHD.

Several structural magnetic resonance imaging (MRI) studies have explored links between smoking and deviations in cortical gray matter structure. A recent meta-analysis on studies using voxel-based morphometry (VBM) revealed that smokers had smaller gray matter volumes bilaterally in the frontal cortex and larger volumes in the right lingual cortex (Zhong et al., 2016). Other studies used cortical thickness (CT) as an outcome measure, which has been argued to be a more specific and sensitive measure for gray matter loss than VBM (Kühn et al., 2010). In accordance with the VBM studies, smokers exhibited lower CT in the orbitofrontal cortex (OFC), anterior cingulate cortex (ACC), and insula, and also more extensively across the frontal, temporal and parietal lobes (Durazzo et al., 2013, Karama et al., 2015, Kühn et al., 2010, Li et al., 2015). Taken together, less gray matter in various brain regions has been observed in relation to smoking, but most consistently in the prefrontal cortex (Wang et al., 2015).

Due to a lack of longitudinal studies it remains unclear whether the thinner frontal gray matter in smokers represents a pre-existing difference, making individuals more prone to develop smoking habits, or is a consequence of tobacco exposure. Only three longitudinal studies have been performed to date, all in middle-aged and elderly populations. Two of these studies investigated global gray matter volume and detected no accelerated volume loss in smokers over a period of four or five years respectively (Duriez et al., 2014, Van Haren et al., 2010). One study looked at regional volumes, demonstrating that elderly participants with a lifetime history of smoking displayed faster atrophy over two years in the OFC, middle frontal gyrus, and other frontal regions, as well as posterior and paralimbic areas (Durazzo et al., 2012). The above stresses the need for more longitudinal studies, especially in adolescence and young adulthood, the sensitive period in brain development when smoking habits are formed.

Here, we describe a study investigating longitudinal effects of smoking on the development of frontal cortex in youth with and without ADHD. The prefrontal cortex is among the last brain regions to mature and is thought to play a crucial role in exerting cognitive control over behaviour (Casey et al., 2005). It has been hypothesised that immature cognitive control abilities make adolescents and young adults more prone to drug use (Loth et al., 2011), and deficits in cognitive control have been associated with ADHD (Lee et al., 2011, Lipszyc and Schachar, 2010). These deficits in cognitive control coincide with the presence of smaller frontal volumes (of for example the OFC and ACC) in individuals with ADHD (Bralten et al., 2016, Frodl and Skokauskas, 2012, Valera et al., 2007). Accordingly, decreased CT in frontal areas may be a shared predisposing factor of individuals with ADHD and smokers, and may reflect immature cognitive control abilities.

To disentangle cause and consequence in the relation between smoking and CT, our first aim was to capture the progressive effect of smoking on frontal CT through adolescence and young adulthood, while controlling for ADHD as an alternative explanation.

As our second aim, we explored whether associations of smoking with frontal CT development depended on ADHD severity. We hypothesised that, considering the already vulnerable frontal cortical structure as reported in ADHD, smoking would have a larger impact on the frontal CT of individuals with more severe ADHD.

If smoking indeed speeds up thinning of the frontal cortex, this could in turn influence the development of ADHD symptoms. While smoking may have beneficial acute effects on ADHD symptoms (Gehricke et al., 2007), potentially used for self-medication purposes, it has been proposed that long-term smoking may cause amplification of impulsive behaviour (DeBry and Tiffany, 2008). Due to the lack of longitudinal studies, this hypothesis has been left largely unexplored. Therefore, our third aim was to investigate whether the developmental course of ADHD symptoms was different in smokers relative to non-smokers.