Neuroimaging is now used more widely in delirium research due to advances in technology. However, imaging delirious patients presents challenges leading to methodological limitations and restricted generalisability. The findings that atrophy and WMH burden predict delirium replicates findings from the original review, while advanced techniques have identified other substrates and mechanisms that warrant further investigation.

Background Delirium is an extremely common hospital complication. No study to date has assessed whether a priori defined covariates; type of hospital setting and year of study publication, influence the relationship between delirium and mortality. This is also the first study to examine the longitudinal trend of delirium-associated mortality over recent decades, to analyse the trajectory of our efforts in combating this disease. Methods MEDLINE, EMBASE and PsycINFO, were searched from January 1981 to May 2018 for English-language primary articles. Rigorous title and abstract screen and full-text screen were conducted independently by two reviewers. This paper adhered to MOOSE guidelines. Data was extracted independently by one reviewer using standardised data-collection sheets, with a separate reviewer verifying for accuracy. The quality of included studies was assessed using the Newcastle-Ottawa Quality Assessment Scale. Unadjusted effect sizes and event counts were analysed with a random effects model in primary meta-analysis and meta-regression, whereas a mixed effect model was used in secondary sub-group analysis. Mortality data at longest follow-up and cumulative mortality (hospital mortality combined with mortality at longest follow-up) data were analysed. Results As part of a larger project, 446 of 6790 articles were retrieved, including 71 studies that measured mortality. Our results demonstrate that elderly inpatients with delirium had significantly greater odds of mortality (OR 3.18 [95%CI: 2.73, 3.70]) compared to non-delirious controls. Patients with delirium in the ICU had the highest odds for mortality (OR: 7.09 [95%CI: 3.60, 14.0]); double the risk compared to the average. Curiously, despite advancements in delirium research, delirium associated in-hospital odds of mortality has not changed in 30 years. Conclusion This is the largest meta-analysis to confirm the association between delirium and mortality, in older (age ≥ 65) hospital inpatients. The current meta-analysis highlights the significant odds of mortality after an episode of delirium, and these odds are much higher for ICU patients. However, in contrast to other medical conditions that have seen a decrease in associated mortality over the past few decades, delirium associated mortality remains unchanged. These findings underscore the urgent need for better delirium treatments. PROSPERO Registration Number: CRD42018098627, https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=98627

Objectives Observational studies have examined the association between delirium and development of new dementia. However, no recent review has collectively assessed the available evidence quantitatively and qualitatively. We systematically reviewed and critically evaluated the literature regarding the association between delirium and dementia, and calculated the odds of developing new dementia after having delirium. Methods This systematic review and meta‐analysis was conducted according to Preferred reporting items for systematic reviews and meta‐analyses guidelines. MEDLINE, EMBASE and PsycINFO, were searched for English‐language articles that compared the incidence of new dementia in older adult (≥65) inpatients with delirium, to inpatients without delirium. A random effects model was used for meta‐analysis, and overall effect size was calculated using reported raw data of event counts. The Newcastle‐Ottawa Quality Assessment scale assessed risk of bias. Results Six observational studies met eligibility criteria, with follow‐up times ranging from six months to five years. Four looked at hip fracture surgical patients; one was on cardiac surgery patients and one examined geriatric medical patients. All studies excluded patients with pre‐existing dementia. Pooled meta‐analysis revealed that older adult inpatients who developed delirium had almost twelve times the odds of subsequently developing new dementia compared to non‐delirious patients (OR = 11.9 [95% CI: 7.29–19.6]; p < 0.001). Conclusions Older adult inpatients who develop delirium are at significant risk of subsequently developing dementia. This emphasises the importance of delirium prevention and cognitive monitoring post‐delirium. The included studies mainly examined post‐surgical patients—further research on medical and intensive care unit cohorts is warranted. Future studies should assess whether delirium duration, severity and subtype influence the risk of developing dementia.