Effects of Ncl. Basalis Meynert volume on the Trail‐Making‐Test are restricted to the left hemisphere

Abstract Background Cortical acetylcholine released from cells in the basal forebrain facilitates cue detection and improves attentional performance. Cholinergic fibres to the cortex originate from the CH4 cell group, sometimes referred to as the Nucleus basalis of Meynert and the Nucleus subputaminalis of Ayala. The aim of this work was to investigate the effects of volumes of cholinergic nuclei on attention and executive function. Methods The volumes of CH4 and CH4p subregions were measured in a subgroup of 38 subjects (33.5 ± 11 years, 20 females) from a population‐based cohort study of smokers and never‐smokers who have undergone additional MR imaging. To define regions of interest, we applied a DARTEL‐based procedure implemented in SPM8 and a validated probabilistic map of the basal forebrain. Attention and executive function were measured with Trail‐Making Test (TMT A+B) and Stroop‐Task. Results We found a quadratic effect of the left CH4 subregion on performance of the TMT. Extremely small as well as extremely large volumes are associated with poor test performance. Conclusions Our results indicate that a small CH4 volume predisposes for a hypocholinergic state, whereas an extremely large volume predisposes for a hypercholinergic state. Both extremes have detrimental effects on attention. Comparable nonlinear effects have already been reported in pharmacological studies on the effects cholinergic agonists on attention.


Nicotine dependence and the cholinergic system
Affection of the central cholinergic system is considered to play a role in the pathogenesis of nicotine dependence (Winterer et al. 2010). An fMRI study on visual attention revealed increased activation of the basal forebrain in smokers, suggesting an affection of the NBM in particular (Vossel et al. 2011). We considered it therefore possible, that the effect of NBM volume on attention is modulated by a history of nicotine dependence. On the other hand, the sample composition in our study (heavy and never-smokers) might affect the generalisability of our results. The study, from which this sub-sample was drawn, has shown that smokers have cognitive deficits that are not attributable to withdrawal. These deficits especially concern the performance in the TMT and other tests of visual attention (Wagner et al. 2013). We performed additional statistical analyses to address the questions if (1) smoking habits affects CH4 volumes and (2) the effect of the left CH4 volume on test performance interacts with the effect of nicotine dependence.

Baseline data according to smoking habits
Baseline data did not differ significantly between smokers and never-smokers. Table S1 provides details about the demographic data in both groups.
Smokers consumed at least two and up to 35 cigarettes per day with an average of 15.5 at the time of data acquisition. They reported a smoking history of up to 41.2 packyears, with a mean of 12.5±13.9 packyears. Blood samples for measurement of cotinine plasma levels were taken at the end of the test day, two hours after the participant had smoked a second cigarette. Plasma samples were kept at -80°C pending the chemical analysis. Mean cotinine levels in smokers were 85.2±96.3ng/mL, levels in never-smokers were below the detection threshold.

Statistical Analysis
The left and the right normalised CH4 volume were used as the dependent variables in two analyses of covariance (ANCOVA) of which each included current smoking (yes/no) as the covariate of interest, and age and sex as covariates of no interest. Interactions among the independent variables were rejected from the model, if they were insignificant at p≥0.1.
To check effects of smoking on our analysis of TMT performance, we added smoking (yes/no) and the interaction effect of smoking and left CH4 volume and squared left CH4 volume to the regression model. Smokers were coded as 0.5, never-smokers as -0.5. We report significant results at an uncorrected threshold of p<0.05.
Smoking affected neither the left (F=0.006, df=33, p=0.940), nor the right CH4 volume (F=0.225, df=33, p=0.638). For Part A of the TMT, there was a significant interaction of smoking and the squared volume (B=-0.002±0.001, p=0.031), but the main effects of the volume also remained significant (quadratic: B=0.001±0.000, p=0.003; linear: B=0.069±0.022, p=0.004). There was no significant interaction of the linear effect of left CH4 volume with smoking (p=0.578). For Part B of the TMT, we found no significant interactions of smoking and the left CH4 volume (quadratic: p=0.165; linear: p=0.482). In the regression analysis of the TMT Part B, the VIFs exceeded the threshold of 2.5 (3.118 maximum).

Conclusions
The results of our additional analyses showed no effect of nicotine dependence on the normalised CH4 volumes.
Results from the analysis of interaction terms of smoking and the left CH4 volume are somewhat difficult to interpret: There was an interaction of the quadratic volume effect and smoking on TMT Part A, indicating that volume differences in smokers affected performance in Part A less than in never-smokers, which might reflect decreased sensitivity of smokers to ACh release. Decreased ACh sensitivity has been suggested to predispose to nicotine dependence due to its effect on cognitive performance (Ernst et al. 2001). Nevertheless, we could not reproduce this finding in the analysis of TMT Part B, although one would expect that performance in both tests depends on similar cognitive mechanisms. To fully elucidate this point will need further studies. Since VIFs slightly exceeded the threshold value in the analysis of TMT Part B, multicollinearity might have affected our results.
Since smokers did not differ in regards of CH4 volumes, and interactions of left CH4 volume that could affect TMT performance were either insignificant (Part B) or did not affect the significance of main effects (Part A), we conclude that our results can be generalised for other populations.