the aim was to examine cross-sectional association between moderate alcohol consumption and total brain volume in a cohort of participants in early middle-age, unconfounded by age-related neuronal change. 353 participants aged 39 to 45 years reported on their alcohol consumption using the AUDit-c measure. participants with alcohol abuse were excluded. Brain MRi was analyzed using a fully automated method. Brain volumes were adjusted by intracranial volume expressed as adjusted total brain volume (aTBV). AUDIT-C mean of 3.92 (SD 2.04) indicated moderate consumption. In a linear regression model, alcohol consumption was associated with smaller atBV (B = − 0.258, p < .001). When sex and current smoking status were added to the model, the association remained significant. Stratified by sex, the association was seen in both males (B = − 0.258, p = 0.003) and females (B = − 0.214, p = 0.011). Adjusted for current smoking, the association remained in males (B = − 0.268, p = 0.003), but not in females. When alcohol consumption increased, total brain volume decreased by 0.2% per one AUDIT-C unit already at 39-45 years of age. Moderate alcohol use is associated with neuronal changes in both males and females suggesting health risks that should not be overlooked. Alcohol abuse is harmful to the brain, but the effects of moderate consumption, and the earliest age when the harmful effects appear, are less known. Volume changes of small parts of the brain, e.g. hippocampus 1 and the ventricles 2,3 , have been associated with moderate consumption, but the relationship of alcohol consumption and total brain volume is contradictory. Methodological differences in imaging technology and in estimating alcohol consumption may make the results difficult to compare and interpret. In addition, most previous studies of global brain values have included participants in late middle-age or old age. However, the rate of age-related brain volume does not occur in a linear fashion, but the decrease accelerates towards old age making the effect of age difficult to control in statistical analysis. Studying younger participants and groups with less age heterogeneity would be helpful. We found ten studies 4-13 which used volumetry, voxel-based morphometry or brain age estimates for analyzing the association of large brain structures and alcohol consumption in moderate drinkers (Table 1). Two of these studies included middle-aged participants in their forties to sixties 4,5 , two mainly over 65-year-olds 6,7 , and six report on participants of a wide age range 8-13. Findings in the middle-aged were conflicting. Decreased 4 and increased 5 volume of white matter, and increased volume of gray matter and ventricular size were associated with moderate drinking in males, but no associations were found in females. Studies of participants over 65 years found either larger 6 or smaller 7 whole brain volumes associated with drinking. Studies of a wider age range have found decreased volumes 8-10 or no association 11,12 with consumption. The most recent stud...

No abstract

Background Attention-deficit/hyperactivity disorder (ADHD) is associated with negative life outcomes and recent studies have linked it to increased mortality. These studies have examined nationwide registers or clinic-referred samples and mostly included participants up until the age of 30. No studies have investigated mortality associated with subthreshold levels of ADHD symptoms. Our aim was to analyze mortality in a perinatal risk cohort of 46-year-old adults with childhood ADHD (cADHD) and milder childhood attention problems (including hyperactivity and inattention; cAP) compared with a group with similar birth risks but no or low levels of childhood ADHD symptoms (Non-cAP). Causes of death obtained from a national register were examined. Methods Mortality was analyzed with Cox proportional hazard models for all-cause mortality, cause-specific mortality (natural and unnatural causes), and age-specific mortality (under and over age 30). All models were adjusted with gender. The total n in the study was 839 (cADHD n = 115; cAP n = 216; Non-cAP n = 508). Results By the age of 46, 11 (9.6%) deaths occurred in the cADHD group, 7 (3.2%) in the cAP group, and 20 (3.9%) in the Non-cAP group. The cADHD group had the highest mortality risk (adjusted hazard ratio = 2.15; 95% CI 1.02, 4.54). Mortality was not elevated in the cAP group (adjusted hazard ratio = 0.72; 95% CI .30, 1.72). Mortality in the cADHD group was mainly attributed to unnatural causes of death (adjusted hazard ratio = 2.82; 95% CI 1.12, 7.12). The mortality risk in the cADHD group was sixfold before age 30 (adjusted hazard ratio = 6.20; 95% CI 1.78, 21.57). Conclusions Childhood ADHD was associated with a twofold risk of premature death by the age of 46 in this prospective longitudinal cohort study. Our results corroborate previous findings and the morbidity of ADHD. Subthreshold levels of childhood ADHD symptoms were not linked to increased mortality. Our results suggest that mortality risk is higher in young than middle adulthood. Future studies should examine mortality associated with ADHD in different ages in adulthood to identify those in greatest risk of premature death.

Maternal diabetes mellitus in pregnancy is associated with impairments in memory functions of the offspring in childhood and adolescence but has not been studied in adulthood. The association of perinatal hypoglycemia with memory has not been studied in adulthood either. The combined sequelae of these two risk factors have not been directly compared. We studied general cognitive ability and memory functions in a prospective follow-up of a cohort born in 1971 to 1974. The sample included participants exposed to prenatal hyperglycemia (n = 24), perinatal hypoglycemia (n = 19), or both (n = 7). It also included controls with no early risks (n = 82). We assessed the participants' Intelligence quotient (IQ), working memory, and immediate and delayed recall of both verbal and visual material at the age of 40. We did not find significant differences in IQ or the memory tests between the groups. We did identify an interaction (p = 0.03) of the early risk with the type of digit span task: compared to the controls, the participants exposed to perinatal hypoglycemia had a larger difference between the forward digit span, a measure of attention, and the backward digit span, a measure of working memory processing (p = 0.022). The interaction remained significant when birth weight was controlled for (p = 0.026). Thus, in this small cohort, prenatal hyperglycemia, perinatal hypoglycemia, and their combination appeared relatively benign disorders. The association of these conditions with neurocognitive impairments in adulthood remains unconfirmed. The significance of the working memory difference needs to be verified with a larger sample.