Memory is the tie that binds together thoughts, impressions, and experiences. Memory function is dependent on several mental or cognitive abilities using several brain systems. Many disease processes can lead to compromise of these systems. When a patient presents to the neurologist with memory loss, the patient or the family is frequently concerned about a neurodegenerative process or dementia. Dementia is often defined as impairment of memory and at least one other cognitive domain that leads to a decline in ability to perform activities of daily living.  A more simple definition of dementia requires a decline in more than one cognitive domain (e.g., memory, language, visuospatial, or frontal executive function) that interferes with one's ability to function independently.
When evaluating a patient with a concern of memory loss, the following questions should be considered:
Does the patient truly have memory loss or is there another cognitive problem causing the memory disorder?
What is the localisation within the brain of the memory problem?
What aetiologies are responsible for the memory disorder?
What is the temporal profile for the memory loss: acute (seconds/minutes/hours), subacute (days/weeks), or chronic (many months to years)?
The first consideration is to determine whether the patient truly has memory loss or another cognitive problem.
Memory dysfunction can result from hippocampal lesions (short-term memory loss, e.g., Alzheimer's dementia) as well as from lesions of brain structures involved in long-term storage (e.g., semantic dementia). In many cases, memory is encoded properly in the hippocampus, but patients have trouble retrieving the stored memory. This retrieval deficit is typically due to problems with frontal lobe function, often caused by white matter disease.
Some degree of memory loss occurs normally with ageing. Normal ageing leads to decreased ability for retrieval from a decline in frontal lobe function, but does not impact the activities of daily living.  Memory loss becomes particularly concerning when it affects function or activities of daily living.
It is important to differentiate transient, fluctuating disturbances in consciousness due to a delirium from an underlying memory disorder. Many medications can induce confusion or even delirium in the elderly; for example, the greater the number of anticholinergic medications a patient is taking, the greater the risk of hospitalisation for confusion or dementia.  The history, examination, and neuropsychological testing can all be helpful in distinguishing a primary memory disorder from a delirium or impairment in retrieval. One relatively large study of hospitalised patients found that a combination of cognitive performance-based tests (Mini Mental State Examination [MMSE]; Mini-Cog) and informant-based tests (AD8; Dementia = [MC]^2) are useful in predicting the risk of delirium.  A Cochrane review of the MMSE for the detection of dementia in community and primary care populations concluded that the MMSE contributes to a diagnosis of dementia but should not be used in isolation to confirm or exclude disease.  A systematic review and meta-analysis of 149 studies of cognitive tests for detecting dementia found that the Mini-Cog and the Addenbrooke's Cognitive Examination-Revised (ACE-R) tests were comparable to the MMSE for detecting dementia, while the Montreal Cognitive Assessment (MoCA) was comparable to the MMSE for detecting mild cognitive impairment. 
A pseudo-dementia related to depression may result in a clinical presentation suggestive of a memory disorder. Consideration should be made regarding the patient's affective state, because depressed patients may experience diminished concentration, sleep disruption, and mild impairments on delayed recall, which may manifest as a memory disorder. 
Patients with memory problems or complaints who do not fit a diagnosis of dementia because they are not functionally impaired are often referred to as having mild cognitive impairment (MCI). The deficits in patients with MCI, by definition, are not severe enough to interfere with their normal activities of daily living. The most common form of MCI is amnestic MCI, in which memory is the primary problem. About 50% of patients diagnosed with amnestic MCI will progress to dementia within 5 years. The rate of progression to dementia is about 12% per year.  However, MCI can occur in cognitive domains other than memory, including language, visuospatial, and frontal executive (e.g., problems organising, planning, multi-tasking). Less is known about these other forms of MCI. Many feel that MCI is a precursor or a continuum of dementia and that it should be treated proactively, through prevention of cardiovascular risk factors, an active cardiovascular exercise programme, and an active and social lifestyle. 
The next step in evaluating a patient with memory loss is determining the localisation of the memory loss. Memory has traditionally been divided into the following memory systems: episodic, working, semantic, and procedural memory.  Separate brain structures are responsible for each of these memory types.
Episodic memory: lasts minutes to years, and localises to the hippocampus and limbic circuits. 
Working memory: a type of memory lasting seconds and involving active rehearsal of information; relies on the integrity of the prefrontal cortex, and the echoic memory in the angular gyrus. 
Semantic memory: typically consists of factual information (e.g., knowing what a certain object is and what it is used for, knowing who was the first President of the US), and localises to the inferolateral temporal lobes. 
Procedural memory: pertains to driving a car or riding a bike, and involves the basal ganglia, cerebellum, and supplementary motor area. 
The type of memory impairment manifested through the history, physical examination, and neuropsychological testing can give an indication of the localisation of the disease process.
The final consideration is the cause of the memory loss. Neurodegenerative, inflammatory/infectious, metabolic, vascular, traumatic, episodic, and endocrine processes can all produce memory impairment through compromise of the limbic and prefrontal circuits.
Professor of Neurology
Memory and Aging Center
University of California
MDG is an author of a number of references in this monograph. He serves on the board of directors for San Francisco Bay Area Physicians for Social Responsibility, on the editorial board of Dementia & Neuropsychologia, and serves or has served as a consultant for Best Doctors, Inc; the Gerson Lehrman Group, Inc; Guidepoint Global, LLC; Biohaven Pharmaceuticals, Inc; Lewis Brisbois Bisgaard & Smith LLP; Lundbeck; MEDACorp; NeuroPhage Pharmaceuticals; and Quest Diagnostics. He receives research support from CurePSP, the Michael J. Homer Family Fund, the National Institute on Aging (R01 AGAG031189), Quest Diagnostics, and the Tau Consortium. He receives honorarium from various medical centers for lectures in the field of dementia.
Memory and Aging Center
University of California
MOK declares that she has no competing interests.
Dr Michael D. Geschwind and Dr Mee-Ohk Kim would like to gratefully acknowledge Dr Michael H. Rosenbloom, Dr Steven Z. Chao, and Dr Julie N. Thai, previous contributors to this monograph. MHR, SZC, and JNT declare that they have no competing interests.
Consultant Psychiatrist and Visiting Professor
University of Hertfordshire
Mental Health Unit
OA declares that he has no competing interests.
Assistant Professor of Internal Medicine
Division of Geriatric Medicine
St Louis University
AB declares that he has no competing interests.
Use of this content is subject to our disclaimer