A major focus of research in aging and dementia is the earliest identification of cognitive impairment. Most clinicians believe that the degenerative dementias (e.g., Alzheimer's disease [AD], frontotemporal dementia) begin very gradually and progress in an insidious fashion over many years and, perhaps, decades. If that course of progression is reasonable, then it is quite likely that there is a transitional period during which people are not cognitively normal but do not meet criteria for dementia. Mild cognitive impairment (MCI) is the term that has come to represent this transitional state.

Since most dementias require a memory deficit to fulfill clinical criteria, individuals with MCI have a memory impairment beyond that which is felt to constitute normal aging, but other aspects of cognitive function are generally preserved. Table 1 represents the current set of criteria used by the Mayo Clinic for the amnestic form of MCI. The first criterion requires a memory complaint, since most people at this stage of impairment are aware of their difficulties. Preferably, this memory complaint should be corroborated by an informant. The clinician must consider incipient depression in people with an isolated memory complaint, since this can be a common presentation of depression. It is also possible, of course, that depression may be one of the earliest features of a progressive dementia (i.e., depression and dementia need not be mutually exclusive) (Cummings, 2003).

The second criterion refers to an objective memory impairment. In order to corroborate the memory complaint, the clinician should document a memory impairment to be certain that the complaint does not simply represent excessive worry about memory and aging, which might be quite normal. Generally, neuropsychological testing is quite helpful in documenting this impairment, but the precise reference standard for normal aging can be problematical (Sliwinski et al., 2003; Smith and Ivnik, 2003). Typically, subjects with MCI will perform between one and two standard deviations below their age mates on normative neuropsychology data.

The third criterion concerns the relatively normal general cognitive function. That is, subjects with MCI generally function quite well in other cognitive domains such as language, attention, executive function and visuospatial skills. Technically, these functions are not absolutely normal, but to most clinicians they are not sufficiently impaired to warrant the diagnosis of dementia. Statistically, subjects with MCI do not perform as well as normal subjects, but their performance in these domains is nowhere near that of subjects suffering from dementia.

In a similar fashion, the activities of daily living of subjects with MCI are essentially normal. Again, if one investigates complex activities of daily living, these subjects might have slight impairments in performance, but of insufficient severity to constitute dementia. Generally, these individuals might be slower at performing complex activities but are still able to complete them without difficulty.

Finally, these people do not meet the standard Diagnostic and Statistical Manual of Mental Disorders, Vol. IV criteria for dementia. They appear more normal than not and do not look like subjects who are demented clinically. They are not functionally impaired as is required for the diagnosis of dementia. To the casual observer, subjects with MCI appear quite normal.

It is important to note that the fulfillment of these criteria is determined clinically. That is, there is no "MCI test" or cutoff score on instruments that determine this diagnosis. Rather, clinicians make the determination in much the same way as they make the clinical diagnoses of depression, dementia, AD or other neuropsychiatric conditions. This diagnosis requires clinicians to adjust their thresholds for what constitutes a cognitive impairment. If clinicians are willing to accept more subtle degrees of cognitive dysfunction that are corroborated by an informant, this condition is quite recognizable clinically.

Most of the research on MCI has been done on the amnestic form. However, as the concept of an intermediate state of impairment between normal aging and fully developed dementia has evolved, it has become apparent that there is more heterogeneity to MCI (Petersen et al., 2001). For example, individuals may develop an impairment in a cognitive domain other than memory. Alternatively, a person may have a slight impairment in memory plus, for example, executive function--yet of insufficient severity to constitute dementia. Therefore, the concept of MCI recently has been expanded to include at least three subtypes: 1) amnestic MCI; 2) multiple domain MCI in which a person may have mild impairments in several cognitive domains with or without a memory impairment; and 3) single non-memory domain MCI in which a person is impaired in a non-memory area such as executive function or language. The other cognitive domains, including memory, are essentially normal. The multiple domain type of MCI may progress to AD or it may be a phenotype of incipient vascular dementia. The single non-memory domain MCI may be the harbinger of other dementing illnesses, such as frontotemporal dementia or dementia with Lewy bodies.

Finally, the descriptions of these clinical subtypes yield various phenotypes but do not imply a cause for the clinical problem. Consequently, one can superimpose an array of suspected etiologies on each of these clinical subtypes as is shown in the Figure. (Due to copyright concerns, Geriatric Times is unable to publish this figure online. Please refer to p16 of the print edition--Ed.) With this scheme, each clinical presentation could be caused by one or more entities, such as degeneration or vascular disease. Psychiatric conditions such as depression could also cause these clinical phenotypes and need to be considered and addressed since they are potentially treatable. Therefore, if one were interested in investigating prodromal AD, one would focus attention on the amnestic subtype of MCI of putative degenerative etiology. Studies have shown that these people overwhelmingly progressed to AD (Bennett et al., 2002; Larrieu et al., 2002; Petersen et al., 1999).

If the outline for the approach to MCI as depicted in the Figure (Due to copyright concerns, Geriatric Times is unable to publish this figure online. Please refer to p16 of the print edition--Ed.) is fulfilled, what is the outcome of these individuals if followed longitudinally? Most of the literature on MCI pertains to the amnestic subtype, since most investigators have been concerned about the prodromal form of AD (Petersen and Morris, 2003). A large body of literature indicates that this subtype tends to progress to AD at a rate of 10% to 15% per year (Petersen et al., 1999). There is some variability in the outcome rates among studies depending on factors such as clinical setting (e.g., memory disorder, clinical versus an epidemiologic study, specific criteria used, reference groups for comparing normal subjects and length of follow-up). However, across multiple sites, the rates of progression to AD are clearly accelerated relative to the 1% to 2% incidence rate for AD in the general population.

Ideally, prevention of AD is the goal; therefore, clinical trials are underway to determine if any of the therapeutic strategies for AD might be effective in MCI (Petersen, 2003). Clinical trials currently underway are outlined in Table 2. They all use the amnestic form of MCI criteria. The operationalization of these criteria varies among the studies, but all are based on the premise of amnestic MCI being a prodromal form of AD. The therapeutic agents being tested are cholinesterase inhibitors, antioxidants, anti-inflammatories, glutamate receptor modulators and nootropics. The length of the trials varies for these investigations, and the end points include the progression to dementia or the progression of symptoms. The duration of the studies ranges from symptomatic trials of only four weeks' duration to the AD outcome studies of three or more years. Only preliminary data are available at present, but several of these trials will be concluding in the next 12 months, at which time we will have considerably more information as to whether amnestic MCI is treatable. This possibility could have a significant impact on the recognition and attention being paid to MCI.

While the essence of MCI is focused on cognitive impairment in the subjects, one could also entertain a prodromal psychiatric presentation for dementia. For example, people with new-onset, late-life depression may have mild affective symptoms with very subtle cognitive deficits. This could be a prodromal presentation of AD. Consequently, clinicians need to be vigilant for subtle presentations of developing psychiatric conditions. As noted, depression can be a possible etiology for variable clinical presentations of MCI.

The landscape of the clinical diagnostic entities in aging and dementia is shifting. Clinicians are becoming increasingly aware of more subtle presentations of a variety of these disorders. Hopefully, this will translate into earlier and possibly more effective treatments for many of these disorders. Presumably, at least for degenerative conditions, the sooner the clinician is able to intervene, the more likely the treatments will have an impact on the underlying pathophysiology. As such, it becomes incumbent for clinicians to become familiar with these mild presentations of dementing disorders.

Dr. Petersen is professor of neurology, the Cora Kanow Professor of Alzheimer's Disease Research and director of the Alzheimer's disease research center at the Mayo Clinic College of Medicine. He is the editor of Mild Cognitive Impairment: Aging to Alzheimer's Disease, published by Oxford University Press.

Bennett DA, Wilson RS, Schneider JA et al. (2002), Natural history of mild cognitive impairment in older persons. Neurology 59(2):198-205.

Cummings JL (2003), Neuropsychiatric symptoms. In: Mild Cognitive Impairment: Aging to Alzheimer's Disease, Petersen RC, ed. New York: Oxford University Press Inc., pp41-61.

Larrieu S, Letenneur L, Orgogozo JM et al. (2002), Incidence and outcome of mild cognitive impairment in a population-based prospective cohort. Neurology 59(10):1594-1599.

Petersen RC (2003), Mild cognitive impairment clinical trials. Nat Rev Drug Discov 2(8):646-653.

Petersen RC, Doody R, Kurz A et al. (2001), Current concepts in mild cognitive impairment. Arch Neurol 58(12):1985-1992.

Petersen RC, Morris JC (2003), Clinical features. In: Mild Cognitive Impairment: Aging to Alzheimer's Disease, Petersen RC, ed. New York: Oxford University Press Inc., pp15-40.

Petersen RC, Smith GE, Waring SC et al. (1999), Mild cognitive impairment: clinical characterization and outcome. Arch Neurol 56(3):303-308.

Sliwinski M, Lipton R, Buschke H, Wasylyshyn C (2003), Optimizing cognitive test norms for detection. In: Mild Cognitive Impairment: Aging to Alzheimer's Disease, Petersen RC, ed. New York: Oxford University Press Inc., pp89-104.

Smith GE, Ivnik RJ (2003), Normative neuropsychology. In: Mild Cognitive Impairment: Aging to Alzheimer's Disease Petersen RC, ed. New York: Oxford University Press Inc., pp63-88.

MCI as a Useful Clinical Concept