Does Mild Cognitive Impairment (MCI) always turn into dementia?

/Home/About Dementia/Mild Cognitive Impairment (MCI)

Your loved one has been diagnosed with "Mild Cognitive Impairment (MCI)". Does that mean that your loved one will inevitably develop Alzheimer's?

No.

In fact, the most recent papers have concluded that most people with MCI do not develop dementia within the first 10 years after diagnosis. (Longer times have not yet been studied.)

1. How many MCI loved ones will go on to develop dementia?

The most compelling papers that concluded most MCI patients will never develop dementia include what are called “meta-analyses”, that is, they combined and reanalyzed the results from a number of different studies that the researchers considered to meet criteria for being well-designed and -executed.

For example, Mitchell and Shiri-Feshki (2009) analyzed 41 high-quality studies, some done on community populations and some in clinical trials. They concluded that the annual conversion rate from MCI to a dementia is ~5-10%; and that even after 10 years, more than 60% of MCI patients will not progress to Alzheimer’s or any other dementia. In fact, a substantial percentage actually revert to normal. Other meta-analyses of long-term (5-10 years) studies reported even lower annual conversion rates, of 3.3 – 4.2%, and cumulative conversion rates of ~31% over 10 years.

Now, you may find individual studies which indicated a much higher annual conversion rate than these — and because of them, your doctor may have told you that the chances your loved one will develop dementia are pretty high. However, there were some very basic flaws in the way the data was analyzed. Now that more extensive studies have been done, we know that the chances an MCI patient will remain stable, or even revert to normal, are very good. Some individual studies have found that as many as 37 – 55% actually improve and no longer have MCI, let alone dementia.

Some of the basic flaws in the earlier interpretation of the conversion data include:

>> Length of the study

The currently-accepted definition of MCI is only a dozen years old, and the definition has changed several times, so until fairly recently, it has been impossible to accrue enough prospective data to allow statements on the prognosis to be made with any kind of confidence.

Since the concept is so new, the first studies to be published did not last for very long periods of time. They sometimes reported a pretty hefty “annual conversion rate”. However, when the meta-analyses compared the cumulative (not annual) conversions reported from short-term studies with those from more recent, long term studies, the results were very similar. In a 2008 meta-analysis of 15 studies, for example, the total number of patients who had progressed to a dementia in studies lasting less than 5 years was 27.4%, while the total number of patients who had progressed to dementia by the end of studies lasting up to 10 years was 31.4%. Meaning … if the loved one is going to develop dementia, it usually happens within the first 3 years after diagnosis, and the conversion rate drops dramatically in later years.

Why would that be? None of the risk factors for developing dementia that were considered would be expected to diminish substantially with time — and the effect of age has been to increase the rate at which dementia develops. So why the rapid drop-off in conversion rates? One hypothesis was that patients diagnosed with MCI are a heterogeneous group, who have varying neuropathologies or don’t have any neuropathology at all. Those with “aggressive” conditions will convert early, leaving a surviving cohort with relatively high resilience. Another possibility is “diagnostic error”, i.e., some cases of early dementia are mislabeled as MCI, and these will deteriorate quickly.

There is growing evidence that the concept of MCI patients as a heterogeneous group is correct — just as there is growing evidence that mixed pathologies are also frequently found in probable Alzheimer’s. The Religious Orders Study, for example, performed postmortem biopsies and found that of the 134 patients diagnosed with MCI, 54.4% had pathologically diagnosed Alzheimer’s, 19.4% had mixed pathologies, and 39.1% had macroscopic infarcts. Of 179 persons diagnosed with probable Alzheimer’s, 45.8% had mixed pathologies, most commonly Alzheimer’s with macroscopic infarcts, neocortical Lewy body disease, or both.

>> Age

It should be noted that there is a big difference between conversion rates over the years that a study lasts, and conversion rates associated with the age at which the loved one is diagnosed. A study that included MCI patients ages 40 to 85 years concluded that age strongly influences the rate at which MCI may progress to dementia. As in other long-term studies, the majority of patients in this study did not progress to dementia over the ten years of the study. However, among those that did progress, the risk of converting from MCI to dementia was much higher in loved ones ages 70 to 85, whereas conversion was rare in loved ones ages 40 to 54. The Nun Study also concluded that age was a risk factor for converting from MCI to dementia.

>> Size of the study

Some of the studies were very small … and results can be skewed under such circumstances. The largest studies conducted to date have found low to very low conversion rates. For example, the “Three Cities” study, which followed 2,882 persons with MCI for four years, found that only 6.6% progressed to dementia during those four years, 56.5% remained stable and 37% actually improved.

A long-term study that followed 949 members of a religious order who had evidence of MCI at baseline found that only 7.5% developed dementia over an 8-year period (note how close this percentage is to that of the Three Cities study), which is an annual conversion rate of 0.94%. In this latter study, ~15% recovered and no longer showed any signs of MCI. In the first three years of the Descripa study (881 elderly patients), the cumulative conversion is 29.7%. The very recent Nakayama Study found that, over five years, 10.6% progressed to Alzheimer’s, 4.8% to vascular dementia, and 5.8% to dementia of other etiology.

>> Method of recruitment

Striking differences are seen when the results of studies done on “community subjects” (i.e., people who volunteered to be in a “clinical trial”, and were then tested to identify those who had MCI) are compared to those from studies done on patients who had visited a memory disorder clinic complaining of problems. In the 2008 meta-analysis, 9 of the studies were community based, and 6 took place in “specialist clinical settings”. The annual conversion rate was significantly higher for studies on patients who sought a diagnosis at a memory disorder clinic — 6.7% compared to 2.6% for studies done on community subjects.

Why would recruitment method make a difference? It is thought to be due to sampling effects. People who have definite memory complaints, especially when combined with other risk factors, tend to seek help from specialist centers — i.e., they tend to have more severe cognitive problems. Those who volunteer to be part of a longitudinal study often are unaware that they have problems, or their problems are so mild that they brush them off as “typical old age”.

>> Potentially treatable causes of MCI

Authors of the meta-analyses have speculated that one possible reason for the conversion rate dropping in outlying years is that some patients who are diagnosed with MCI do not have any neuropathology. One recent study concluded that non-degenerative, potentially treatable causes of cognitive decline are found in as many as 30% of people diagnosed with MCI. The most common of these include hypothyroidism, vitamin B12 deficiency, vascular disease, normal pressure hydrocephalus, and subdural hematoma. Another study concluded that changing the risk factors for stroke and treating depression may have contributed to those with MCI who reverted to normal.

>> Appropriate controls

Progression to dementia over time needs to be placed into context. Loved ones have an increasingly greater risk of developing dementia as they get older, whether or not they have MCI. Few studies have considered how much of the “conversion” — especially the persisting low-level conversion rate in outlying years — is not conversion from MCI per se, but rather the likelihood that any given person will develop dementia. One study, for example, found that 38% of subjects who had MCI at baseline progressed to dementia within two years. The same thing also happened to 15% of test subjects who did not have MCI at baseline. According to study results, MCI patients progress to dementia in a period of two years more than twice as frequently as normal people, but only one third of the patients becoming demented during this period showed MCI at baseline. On the other hand, 33% of those who were diagnosed with MCI at baseline had improved within the two-year period.

2. Which MCI loved ones are likely to develop dementia?

Unfortunately, there is currently no way to predict with any certainty which patients with MCI will eventually progress to dementia. Certain “risk factors” have been tentatively identified, but the data from different studies is sometimes contradictory and confusing. It is generally agreed that no single factor is strong enough to be used as the sole predictor. Certain combinations of factors seem to be a bit more reliable. However, even when a patient has one of these combinations, that only means that the patient is more likely to develop dementia — it does not mean that the patient will develop dementia. Risk factors that can be measured by clinical tests include atrophy of the medial temporal lobe, certain cerebrospinal fluid biomarkers (e.g., lower Abeta[42] and higher total tau), APOE4, glucose hypometabolism patterns (“PET scan”), olfactory identification deficits, and neuropsychological markers. More subjective risk factors that have been tentatively suggested include confirmation of memory difficulties by a knowledgeable informant (such as a spouse, child, or close friend), poor performance on objective memory testing, and any changes in the ability to perform daily tasks, such as hobbies or finances, handling emergencies, or attending to one’s personal hygiene.

Some studies have concluded that the presence of more severe and widespread cognitive deficits, as indicated by a diagnosis of multiple-domain amnestic MCI, may be a better predictor of Alzheimer’s than single-domain amnestic or non-amnestic deficits. Others produced a less consistent pattern of conversion.

3. What happens to MCI loved ones who improve?

Like many other studies, the Kungsholmen Project found that only 35% of patients diagnosed with MCI at baseline progressed to dementia within the first three years of the study. However, they also specifically looked at the subsequent fate of MCI patients who improved during those first three years. They found that persons who improved did not have a significantly higher risk of later progressing to dementia than persons who had never been impaired.

4. Does treatment alter the course of MCI?

The jury is still out … and quite possibly because the underlying causes of MCI are so many and so diverse that it is difficult to detect a benefit to a subpopulation.

The first meta-analysis of three published and five unpublished trials (three on donepezil, two on rivastigmine, and three on galantamine), ranging in length from 6 months to 3 years, did not find significant differences compared with placebo groups. An apparently unpublished reanalysis of the four largest trials (one donepezil, one rivastigmine, and two galantamine) did find a reduced risk of progression in the short term; and two individual trials indicated a trend toward efficacy even though the results were not statistically significant. However, the evidence has not been strong enough for any of these drugs to be approved for treating MCI.

Moreover, the risks associated with these drugs are not negligible, especially for persons who do not have an MCI leading to Alzheimer’s itself. Given that the percentage of MCI patients who remain stable or revert to normal on their own is so high, and that those who do progress may develop a non-Alzheimer’s dementia that may be harmed rather than helped by these drugs, the risks may well be significantly higher than the rewards.

What is not risky, and may indeed be very beneficial, is the use of non-drug interventions. In fact, implementing these would be a good idea for the family and friends of MCI patients as well as for the patients themselves. Evidence is mounting that they can help prevent or delay the onset of MCI and dementia, as well as slow their progression. They include (1) sticking as closely to a Mediterranean diet as possible; (2) getting lots of exercise; (3) giving your brain lots of exercise, too, from activities that involve learning something new; (4) socializing, doing things with family and friends; (5) getting plenty of rest; (6) doing anything and everything you can to minimize stress; and (7) no smoking (in fact, avoid all types of air pollution as much as possible.)

References:

- General overview of MCI — what is and is not known (excellent)

Petersen RC, Roberts RO, Knopman DS, Boeve BF, Geda YE, Ivnik RJ, Smith GE, Jack CR Jr. Mild cognitive impairment: ten years later. Arch Neurol. 2009 Dec;66(12):1447-55.

http://archneur.ama-assn.org/cgi/content/full/66/12/1447

- Meta-analyses of conversion to dementia

Mitchell AJ, Shiri-Feshki M. Rate of progression of mild cognitive impairment to dementia – meta-analysis of 41 robust inception cohort studies. Acta Psychiatr Scand 2009;119(4):252-65.

http://www.ncbi.nlm.nih.gov/pubmed/19236314

Burke D. Review: long-term annual conversion rate to dementia was 3.3% in elderly people with mild cognitive impairment. Evid Based Med. 2009 Jun; 14(3):90.

http://ebm.bmj.com/content/14/3/90.extract

Mitchell AJ, Shiri-Feshki M. Temporal trends in the long term risk of progression of mild cognitive impairment: a pooled analysis. Journal of Neurology, Neurosurgery, and Psychiatry 2008;79:1386-91.

http://jnnp.bmj.com/content/79/12/1386.abstract

Roberto Raschetti R, Albanese E, Vanacore N, Maggin M. Cholinesterase Inhibitors in Mild Cognitive Impairment: A Systematic Review of Randomised Trials. PLoS Medicine November 2007;4(11):e338.

http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.0040338

- Individual studies

Three Cities: Artero S, Ancelin M-L, Portet F. et al. Risk profiles for mild cognitive impairments and progression to dementia are gender specific. J Neurol Neurosurg Psychiatry. 2008;79(9):979-84.

http://prunel.ccsd.cnrs.fr/docs/00/29/49/05/PDF/Artero_JNNP_2008-1.pdf

Nun Study: Tyas SL, Salazar JC, DA, et al. Transitions to mild cognitive impairments, dementia, and death: findings from the Nun Study. Am J Epidemiol 2007;165:1231–8.

http://aje.oxfordjournals.org/content/165/11/1231.full

NOTE: this paper was referenced by one of the meta-analyses as the largest study conducted to date, looking at “members of a religious order (nuns and brothers), of whom 949 had evidence of MCI at baseline.” The referenced paper, itself, however, only reports on results from 470 nuns. It is not clear whether the meta-analysis used unpublished data (which has been presented at symposia), data published in another paper (a study this large yields many papers on data analyzed in different ways), or, perhaps, actually intended to refer to the “Religious Orders” study, a very large, ongoing longitudinal study involving Catholic clergy, monks, and nuns.

Religious Orders: (Note: this very large, very long — and still ongoing — study has produced dozens of papers on dozens of narrowly-defined subjects. More than 2,000 priests, nuns, and monks enrolled in the Religious Orders Study or related Memory and Aging Project.)
- Schneider JA, Arvanitakis Z, Leurgans SE, Bennett DA. The neuropathology of probable Alzheimer disease and mild cognitive impairment. Ann Neurol. 2009 Aug;66(2):200-8.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2812870

- Aggarwal T, Wilson RS, Beck TL, et al. Mild cognitive impairment in different functional domains and incident Alzheimer’s disease. Neurology. 2005;64(suppl1):A166.

http://jnnp.bmj.com/content/76/11/1479.full

- Bennett DA, Wilson RS, Schneider JA, Evans DA, Beckett LA, Aggarwal NT, Barnes LL, Fox JH, Bach J. Natural history of mild cognitive impairment in older persons.
Neurology. 2002 Jul 23;59(2):198-205. http://www.ncbi.nlm.nih.gov/pubmed/12136057

DESCRIPA: Visser PJ, Verhey FRJ et al. Development of Screening Guidelines and Clinical Criteria for Predementia Alzheimer’s Disease. The DESCRIPA Study. Neuroepidemiology 2008;30:254-65.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2821431/

Nakayama Study: Sonobe N, Hata R, Ishikawa T, Sonobe K, Matsumoto T, Toyota Y, Mori T, Fukuhara R, Komori K, Ueno S-I, Tanimukai S, Ikeda M. Risk of progression from mild memory impairment to clinically diagnosable Alzheimer’s disease in a Japanese community (from the Nakayama Study). Internat Psychogeriatrics FirstView 16 Dec 2010.

http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=7947664

Kungsholmen Project: Palmer K, Wang HX, Backman L, Fratiglioni L. Differential evolution of cognitive impairment in nondemented older persons: results from the Kungsholmen project. Am J Psychiatry 2002; 159: 439-42.

http://ajp.psychiatryonline.org/cgi/content/full/159/3/436

Influence of age: Visser PJ, Kester A, Jolles J, et al. Ten-year risk of dementia in subjects with mild cognitive impairment. Neurology 2006;67:1201–7.

http://www.ncbi.nlm.nih.gov/pubmed/17030753/

Potentially treatable causes of MCI: Jicha GA, Abner E, Schmitt FA et al. Clinical features of mild cognitive impairment differ in the research and tertiary clinic settings. Dementia and Geriatric Cognitive Disorders 2008;26(2):187-92.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2667338/

Risk factors for progression: Devanand DP, Liu X, Tabert MH, Pradhaban G, Cuasay K, Bell K, de Leon MJ, Doty RL, Stern Y, Pelton GH. Combining early markers strongly predicts conversion from mild cognitive impairment to Alzheimer’s disease. Biol Psychiatry. 2008 Nov 15;64(10):871-9.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2613777/

MCI subtypes:
- Forlenza OV, Diniz BS, Nunes PV, Memória CM, Yassuda MS, Gattaz WF. Diagnostic transitions in mild cognitive impairment subtypes. Int Psychogeriatr. 2009 Dec;21(6):1088-95.

http://www.ncbi.nlm.nih.gov/pubmed/19691909

- Issa AM, Rountree S, Waring SC, Yeh SH, Doody RD. Mild cognitive impairment types and their outcomes. Neurology. 2005;64(suppl1):A168. I’m not finding this at the Neuro website … the study is summarized in the Medscape article at:

http://www.medscape.org/viewarticle/506112

Should screening for MCI even be done? Buntinx F, Paquay L, Ylieff M, De Lepeleire J. Progression and improvement after mild cognitive impairment. Arch Public Health 2009, 67:7-14.

https://lirias.kuleuven.be/bitstream/123456789/236256/1/2009FBmildcognAPH.pdf