Diagnosis of mild cognitive impairment (MCI)


MCI is the stage of cognition between the normal, ageing associated cognitive decline (compatible with a relatively independent life style) and dementia when the subject becomes totally dependent on external assistance. There is no single, diagnostic test for MCI; diagnosis is reached by exclusion of other possible treatable causes that may contribute to MCI-like symptoms (i.e. lack of vitamin B-12, low thyroid hormone levels, hypertension, depression and medication side effects).


The course of MCI is unpredictable; in can remain stable, progresses to dementia or the afflicted subjects regain their cognitive abilities. Therefore, for effective management, it is crucial to distinguish between:

  • normal, ageing associated cognitive decline and MCI and

  • measure the state of MCI since people with MCI have an increased risk of developing dementia.


There is no cure for MCI; in some patients, medications used to treat Alzheimer’s disease and attenuate anxiety do improve the patients’ problem solving and attention cognitive abilities (Neurology 2018, 16; 90(3):126–135; Clin Geriatr Med. 2013 29(4): 737–752; BMJ 2015; 350: h3029).

Benefits of early diagnosis of MCI are:

  • attenuation or reduced progression of MCI by managing conditions that contribute to MCI (e.g. depression, diabetes);

  • increased effectiveness of nonpharmacological treatment measures that attenuate or reduce the progression of MCI (i.e. physical exercise, cognitive interventions and increasing participation in mentally or socially stimulating activities);

  • diagnosis of the cause(s) of MCI is more accurate early in the disease process and

  • better quality of life and long term planning with early management of MCI symptoms.


Currently, there is no single reliable diagnostic marker for MCI. The diagnosis of MCI is usually based on the patient's symptoms; biomarker tests such as brain imaging and cerebrospinal fluid tests are performed, when the diagnosis of MCI is in doubt, to determine if the individual:

  • has MCI and

  • if the MCI is due to Alzheimer's or some other neurodegenerative brain disorder associated with cognitive impairment.


Brain changes in the amnestic type of MCI are often similar to brain changes seen in Alzheimer's disease or in other forms of dementia. For example, imaging studies have identified an association of the following brain changes with MCI: hippocampus atrophy, hypertrophy of the brain's ventricles and reduced use of glucose in specific brain regions (www.alz.org; BMJ 2015; 350: h3029; Comput Biol Med. 2018 Sep 15; 102: 30–39).


European and US guidelines recommend structural brain scanning with magnetic resonance imaging (MRI) for people with suspected dementia. Imaging is also embedded in modern diagnostic criteria for different neurodegenerative brain disorders including Alzheimer’s disease. Structural MRI can also evaluate the vascular disease’s contribution to cognitive impairment (BMJ 2015; 350: h3029). In addition to MRI, the potential of EEG, as a potential biomarker for MCI, was demonstrated. Event-related potentials, during sustained attention and memory tasks, tracked MCI progression (Alzheimer’s Dement (Amst). 2018 Jul 2; 10: 452–460).


There is no single brain imaging method (i.e. with a simultaneous good time and spatial resolution and the ability to measure metabolic activity of nerve cells only) appropriate for optimal development of biomarkers for MCI. Several imaging methods are combined, e.g. MRI, EEG, PET (positron-emission tomography) and SPECT (single photon emission computed tomography) to assist in MCI diagnosis. EEG directly measures nerve activity only, has a good time resolution and a low spatial resolution. Compared to EEG, MRI has a better spatial resolution and a lower time resolution. Neither EEG or MRI can measure local metabolic activity of nerve cells. MRI can measure changes in regional brain blood flow, reflecting regional changes in circulating oxygenated haemoglobin. Local metabolic activity of nerve cells can be estimated by PET, which has a lower time and spatial resolution then MRI, that measures local changes in brain metabolic activity. Therefore, EEG, MRI and PET are complementary brain imaging methods, and multimodal neuroimaging seems to be the way forward in developing more sensitive and specific (multimodal) markers for MCI. Integration of multimodal brain imaging data can improve the sensitivity and specificity of MCI biomarkers, provided differences in the neuronal and structural underpinnings of each brain imaging method are considered during data interpretation (NeuroImage 2014, 102: 3–10).