Brain metabolites of choline (Ch) and myo-Inisotol (mI) have been reported as elevated among geriatric depressed patients. Twodimensional (2D) magnetic resonance spectroscopy (MRS) provides estimates of Ch, mI, and creatine (Cr) similar to one-dimensional MRS, and it also estimates the resonances of the Ch-containing compounds of phosphoethanolamine (Pe) and phosphocholine (PCh). In this cross-sectional geriatric study, 14 depressed patients and 14 healthy volunteers who were comparable in age, gender, education, comorbid medical burden, and Mini-Mental State Examination (MMSE) scores completed 2D MRS and a neurocognitive battery. A voxel in the left dorsolateral cortex, which was comprised of approximately 60% white matter, was used to estimate the CR ratios of Ch, PCh, Pe, and mI. Composite scores for cognitive function were developed for verbal learning, recall, recognition, executive function, hypothesis generation, and processing speed. Among nondepressed subjects, cognition was positively correlated with Ch/Cr and mI/Cr and negatively correlated with PCh/Cr in four domains of verbal learning, recognition, recall, and hypothesis generation. In contrast, depressed patients did not have consistent relationships between Ch/Cr, mI/Cr, and PCh/Cr and cognition. There was a significant difference in the overall pattern of associations between the four metabolites and verbal learning and processing speed in depressed patients compared to healthy controls. The attenuated relationship between metabolites and specific cognitive domains in patients with late-life MDD suggests that the level of cognitive performance observed during depressive episodes may be associated with changes in biochemistry within the frontostriatal neuronal circuitry.

Background Relapses in major depression are frequent and are associated with a high burden of disease. Although short-term studies suggest a normalisation of depression-associated brain functional alterations directly after treatment, long-term investigations are sparse. Aims To examine brain function during negative emotion processing in association with course of illness over a 2-year span. Method In this prospective case–control study, 72 in-patients with current depression and 42 healthy controls were investigated during a negative emotional face processing paradigm, at baseline and after 2 years. According to their course of illness during the study interval, patients were divided into subgroups (n = 25 no-relapse, n = 47 relapse). The differential changes in brain activity were investigated by a group × time analysis of covariance for the amygdala, hippocampus, insula and at whole-brain level. Results A significant relapse × time interaction emerged within the amygdala (PTFCE-FWE = 0.011), insula (PTFCE-FWE = 0.001) and at the whole-brain level mainly in the temporal and prefrontal cortex (PTFCE-FWE = 0.027), resulting from activity increases within the no-relapse group, whereas in the relapse group, activity decreased during the study interval. At baseline, the no-relapse group showed amygdala, hippocampus and insula hypoactivity compared with healthy controls and the relapse group. Conclusions This study reveals course of illness-associated activity changes in emotion processing areas. Patients in full remission show a normalisation of their baseline hypo-responsiveness to the activation level of healthy controls after 2 years. Brain function during emotion processing could further serve as a potential predictive marker for future relapse.

Impulse control disorders (ICD) in Parkinson’s disease (PD) frequently occur, not always as a direct consequence of dopaminergic medication. This study investigated premorbid personality traits and behavioural characteristics in non-demented PD patients with self-reported symptoms of ICD (PD-srICD). From a total of 200 non-demented PD patients who filled out questionnaires assessing symptoms and severity of ICD, those were classified as PD-srICD (n = 32) who reported current occurrence of at least one compulsive behaviour (gambling, sexual behaviour, buying behaviour, or eating). As a control group, 32 patients with no self-reported ICD symptoms were matched for levodopa equivalent daily dose. The demographic, clinical, and premorbid personality profiles were compared between both groups. Frequency of psychological characteristics indicating substance use disorder was evaluated in patients with PD-srICD. Patients with PD-srICD were more frequently male, younger at examination, had earlier PD onset, more depression, higher non-motor burden, less quality of life (p < 0.05, respectively), and more frequently reported premorbid sensation seeking/novelty orientation (p = 0.03) and joyful experience of stress (p = 0.04) than patients in the control group. Of patients with PD-srICD, 90.6% reported at least one behavioural characteristic of substance use disorder, most frequently positive expectations following ICD behaviour and illusional beliefs about its behavioural control. Signs of addiction were common among patients with PD-srICD. Therefore, the profile of psychological characteristics in patients with PD-srICD resembled that of patients with substance use disorder. It can be concluded that dopamine replacement therapy (DRT) alone does not account for PD-srICD and that thorough psychological diagnostics are recommended.

No abstract