The use of enteral tube feeding [by a nasogastric tube or percutaneous endoscopic gastrostomy (PEG)] for patients with advanced dementia who develop problems with eating or swallowing or have poor nutritional intake is common. In one US survey, one third of 186,835 nursing home residents with advanced dementia were tube fed but, despite its wide use, potential benefits (prolonging survival, improving quality of life, better nourishment, decreased risk of pressures sores) and harms (increased risk of developing pneumonia due to inhaling small food quantities and even death) of this procedure remain unclear [7]. No randomized clinical trial has so far been carried out in this field and available data are limited to a few observational controlled studies. Heterogeneity in primary outcome measures, with some studies assessing survival and others only nutritional status or quality of life, has contributed to insufficient evidence.

Delirium in the elderly predicts sustained poor cognitive and functional status and increased likelihood of nursing home placement after a medical admission. Delirium is an independent predictor of adverse outcomes in elderly hospitalized patients, particularly in the presence of cognitive impairment or dementia at baseline, and it is associated with an increased mortality rate and acceleration of cognitive decline [8]. Delirium in dementia is often unrecognized. It should be highlighted that disruptive behavior in dementia may be due to underlying delirium, and medications for treating behavioral disturbances may worsen or further mask the problem. Unfortunately, behavioral changes in dementia are often misattributed to fluctuations, diurnal variations of symptoms (sundowning), or to the underlying dementing illness itself rather than to a superimposed delirium. A delay in diagnosis of delirium and its underlying cause contributes to the poor outcomes associated with delirium superimposed on dementia. Hypoactive delirious patients appear to be at particular risk because of complications from aspiration and inadequate oral nutrition as well as falls and pressure sores. The prevalence of delirium superimposed on dementia is high. In a review where 14 studies were considered, including 7 prospective studies, 3 retrospective studies, 2 cross-sectional studies, and 2 clinical trials, the prevalence of delirium superimposed on dementia ranged from 22 to 89 % of hospitalized or community population people aged 65 and older [9]. In the year following hospitalization, AD patients manifest a more rapid progression of cognitive impairment, increased mortality, and greater chance of definitive institutionalization. Approximately one out of eight hospitalized patients with AD who develop delirium will have at least one adverse outcome, including death, institutionalization, or cognitive decline, associated with delirium [10]. Specifically, for hospitalized patients with AD, approximately 1 in 16 deaths, 1 in 7 institutionalizations, and 1 in 5 cases of accelerated cognitive decline within 1 year can specifically be attributed to delirium [10]. However, if the causative factor is rapidly corrected, recovery can be complete. Delirium present at discharge from the hospital is associated with a 2.6-fold increased risk of death or nursing home placement, and delirium persisting after hospital discharge is associated with a 2.2-fold risk of death within the following year [8].

These figures emphasize the importance of early recognizing and preventing delirium in persons with dementia, through the implementation of a risk factor reduction strategy targeting sleep deprivation, immobility, sensory impairment, dehydration, metabolic disorders, abuse of psychoactive drugs and the elimination of unnecessary medications.

This term originally referred to a transitional zone in subjects with memory complaints but without any evidence of functional decline. These subjects, although not fulfilling criteria for dementia, are at risk of developing AD or other dementing disorders. Although other subtypes of MCI have been described later, the most typical is characterized by memory impairment but with all other cognitive domains essentially spared (amnestic MCI). The original clinical criteria for MCI were: (a) memory complaint, preferably corroborated by an informant; (b) objective memory impairment for age; (c) essentially normal general cognition; (d) largely intact activities of daily living; (e) absence of dementia [11]. However, any single cognitive domain other than memory (non-amnestic MCI single domain) or multiple domains simultaneously (amnestic or non amnestic-MCI multiple domain) can be affected.

It is commonly believed that amnestic MCI typically evolves to AD, while non-amnestic MCI progresses to dementing disorders other than AD. However, each form of MCI is highly heterogeneous, with many MCI subjects who actually do not progress to dementia, but stop declining or even improve. Despite this variability in clinical outcomes, it is undisputed that MCI altogether represents an “at risk” condition, as indicated by the fact that, compared to normal elderly adults, those with amnestic MCI are more likely to evolve to AD (10–15 % vs. 1–2 % per year) [11]. Predictors of conversion to AD include particularly bad performances on measures of delayed recall, presence of apolipoprotein epsilon 4 allele, hippocampal atrophy on cerebral MRI, parietal and temporal hypometabolism on FDG-PET, and low Abeta 42/high t-tau proteins in CSF.

There is insufficient evidence that drugs used in dementia due to AD, including cholinesterase inhibitors, affect progression to dementia or cognitive test scores in MCI.