Managing MCI: Sifting Through the Unknowns

Published in the September 2007 issue of Applied Neurology

By Jordana Bieze Foster

In many ways, the frustration experienced by patients struggling with mild cognitive impairment (MCI) is matched by the frustration of clinicians facing the challenge of managing this heterogeneous condition. The prognosis can be variable, and no proven therapies exist.

Would it be less frustrating if a diagnosis of MCI could definitively identify patients who would eventually progress to Alzheimer disease (AD) or another form of dementia? What if identifying those patients would allow for early therapeutic intervention that would delay or prevent that progression? In reality, although dementia does ultimately develop in the majority of patients with MCI, little is known about how quickly MCI progresses to dementia. Some population-based studies suggest that as many as 44% of patients who meet diagnostic criteria for MCI will no longer meet those criteria on follow-up examination.1 To date, no therapies are approved specifically for use in patients with MCI, and the few clinical trials that have included patients with MCI have yielded little more than promising trends.2-11

In practice, these clinical unknowns must be balanced against the very real possibility that MCI will progress to dementia. This necessitates an approach to patient care that is both delicate and diplomatic.

"You obviously want to be cautious about what you tell patients. You don't want to give them the impression that, for certain, they're going to get Alzheimer disease," said Allan I. Levey, MD, PhD, professor and chair of neurology at Emory University in Atlanta, and director of the Emory Alzheimer's Disease Research Center. "But it's also important to educate the patient about what MCI is and that it does carry a risk; that's the reality. People want to know what they can do, so it's our job to try to answer that question directly."


With so much ambiguity already surrounding MCI, ensuring an accurate diagnosis is essential. Although experts have preferences for different neuropsychological tests, most advocate a battery of tests that can distinguish MCI from normal age-related deficits of cognitive impairment and that can distinguish MCI from dementia by verifying that any influence on activities of daily living has not affected the patient's functional independence. All emphasize that because MCI is usually diagnosed in patients with whom the physician has no previous relationship, a thorough patient history—obtained from both the patient and his or her caregivers—can be as important as test results for making the diagnosis.

"Ideally, you're most sure of the diagnosis when not just the patient is aware of the problem, but another person also confirms that a problem exists, that it is new, and that it is getting progressively worse," said Martin R. Farlow, MD, professor of neurology at Indiana University School of Medicine in Indianapolis, and associate director of the Indiana Alzheimer Disease Center.

A relatively new development in the study and management of MCI is the concept of narrowing the diagnosis to 1 of 4 subtypes: amnestic or nonamnestic, and affecting 1 or multiple cognitive domains. Before a 2003 international conference in Stockholm at which the narrower subtypes were outlined,12,13 discussion of MCI was largely focused on memory impairment and the amnestic subtype.

The amnestic subtype is considered to be the potential precursor to AD, with conversion rates estimated at 10% to 15% per year. Although it was initially thought that amnestic MCI would progress to AD and nonamnestic MCI would progress to other forms of dementia, more recent research14,15 suggests that there is quite a bit of crossover between the subtypes and their outcomes, further complicating the clinical picture.

"The [documented] conversion rates are most clear for the amnestic subtype. So in my practice, if someone meets those criteria, I'm able to give them pretty reliable information about conversion rates—that there is an approximately 50% chance of conversion to AD within 5 years," Levey said. "If the patient is not amnestic, it encourages me to look for other causes of cognitive deficits."

Those other causes can include underlying cerebrovascular disease, sleep apnea or other sleep disorders, and behavioral issues, such as depression—conditions that, perhaps ironically, may be managed more successfully than amnestic disorders. To assist in the differential diagnosis, some clinicians advocate performing an MRI scan on any patient with suspected MCI. Research also suggests that different patterns of medial temporal lobe atrophy on MRI can differentiate MCI subtypes16 and can predict progression, particularly to AD.17-19

"In my practice I get an MRI on anyone with objective memory loss," said Ranjan Duara, MD, medical director of the Wien Center for Alzheimer's Disease and Memory Disorders at Mount Sinai Medical Center in Miami Beach, Florida, and associate professor of medicine at the University of Miami. "If the MRI clarifies that the person doesn't have an underlying neurological condition, then that can be helpful in identifying other factors. You might find out that there are psychiatric or medical issues, or use of medications or alcohol that you didn't know about."


Unfortunately, even with an accurate diagnosis, the therapeutic options for MCI are limited. Few drugs approved by the FDA for mild to moderate AD have been studied in patients with MCI. Although the acetylcholinesterase inhibitor donepezil (Aricept) was associated with delayed conversion to AD in the first 12 months of the 2005 Memory Impairment Study,2 the statistically significant between-group difference in conversion rate had dissolved by the end of the 3-year study.

Researchers, nonetheless, have been encouraged by the results of the study, not only because of the early success in the entire group treated with donepezil, but also because the results showed even stronger evidence of the drug's effectiveness in carriers of apolipoprotein (APOE) e4 alleles, which have been associated with more rapid progression from amnestic MCI to AD.20

Of the 769 patients with amnestic MCI in the study, possible or probable AD developed after 12 months in only 16 of the 253 patients who received donepezil (5 mg/d, increased to 10 mg/d after 6 weeks), compared with 33 of the 257 patients who received vitamin E (1000 IU daily, increased to 2000 IU daily after 6 weeks) and 38 of the 259 patients who received placebo. Although the conversion rates for the 3 treatment groups were not significantly different at 3 years, when researchers looked specifically at patients who tested positive for the APOE e4 allele, the conversion rate in the donepezil group was significantly lower than in the placebo group for the entire study period.

An MRI analysis of a 131- patient subset of that study population, published in 2007, found that treatment had no effect on annual percent-volume change in the hippocampus, entorhinal cortex, whole brain, and ventricle; however, the researchers did report a non-significant trend suggesting reduced hippocampal atrophy in APOE e4 allele carriers treated with donepezil.3

An earlier, smaller-scale trial of donepezil also failed to demonstrate efficacy in its primary end points and improvement on the New York University (NYU) Paragraph Recall Test and the Clinical Global Impression of Change scale in the intent-to-treat population.4 In that 24-week trial, the 133 patients with amnestic MCI who were randomly selected to receive donepezil (5 mg/d for 42 days, increased to 10 mg/d) were significantly more likely to demonstrate improvement on the cognitive subscale of the AD Assessment Scale than were the 137 patients who received placebo.

Other drugs that are FDA- approved for mild to moderate AD have fared less well in clinical trials of the MCI population. The most recent trial to be published, the Investigation into the Delay to Diagnosis of AD with Exelon (InDDEx) study,5 found no significant benefit of the cholinesterase inhibitor rivastigmine (Exelon) on the time to clinical diagnosis of AD in more than 1000 patients with MCI. During the 48-month study, AD was diagnosed in 17.3% of patients who were randomly selected to receive rivastigmine (0.5 mg bid, gradually increased to a maximum of 12 mg/d) and in 21.4% of those who received placebo.

Researchers also found no significant between-group differences regarding change from baseline in performance on neuropsychological tests. Interestingly, treatment benefits were not seen even in a subpopulation of APOE e4 allele carriers. However, the authors noted that their inclusion criteria from the NYU Delayed Paragraph Recall Test were not adjusted for age and education. By contrast, delayed recall inclusion criteria for the Memory Impairment Study were adjusted for education level.

Two earlier studies of the cholinesterase inhibitor galantamine (Razadyne),6,7 which is also FDA-approved for mild to moderate AD, found no statistically significant benefits at 24 months in improving cognition or delaying conversion to AD in an MCI population of more than 1000 patients. Some statistically significant treatment-related differences were seen in population subgroups based on NYU delayed or immediate recall test performance.

Of 898 persons who underwent MRI, those randomly selected to receive galantamine (4 mg bid for 4 weeks, gradually increased to 8 mg or 12 mg bid) had significantly lower rates of whole brain atrophy than those who received placebo. Hippocampal atrophy rates were not significantly different.

An additional concern regarding the use of cholinesterase inhibitors in patients with MCI is the number of adverse events reported in the aforementioned clinical trials. In all trials, adverse events, primarily GI, occurred more frequently in treatment groups than in placebo groups. In the galantamine trials, more deaths caused by treatment-emergent adverse events occurred in patients taking the drug than in those taking placebo, although that imbalance was normalized to some extent when retrieved dropout patients were included in the analysis.

Despite the lack of efficacy of cholinesterase inhibitors in primary trial outcomes thus far, clinicians and researchers are encouraged by the secondary findings. Some are including the drugs in discussions of therapeutic options for patients with MCI who are willing to pay out-of-pocket.

"There have been a number of trials, and many have been negative, but there are also trials that have suggested that some of the medications developed for AD may have some limited but real utility in treating the memory deficits associated with at least the amnestic form of MCI for a period of time," Farlow said.


Similar results—falling short of efficacy regarding a primary end point but suggesting promise in other measures—were seen in a 2004 phase 2 trial of the ?-aminobutyric acid-B receptor antagonist SGS742 in 110 patients with MCI.8 After 8 weeks of treatment, no statistically significant difference in improvement on the Hopkins Verbal Learning Test-Revised was seen between patients who received SGS742 and those who received placebo. However, the test group did demonstrate greater improvement on the NYU Paragraph Recall Test, the Weschler Adult Intelligence Scale Digit Symbol test, and the Cambridge Automated Neuropsychological Test Battery.

Also in 2004, a 4-week trial in which 175 patients with MCI were randomly selected to receive either placebo or the ampakine compound, CX516, failed to demonstrate improvement in delayed recall of a 15-item word list.9,10 A treatment effect of CX516 was seen, however, in those patients with the greatest degree of memory impairment at baseline. A French safety and efficacy trial of a more potent ampakine compound, S18986, is currently recruiting patients with MCI.

The anti-inflammatory agent rofecoxib (Vioxx), which was FDA- approved for treatment of osteoarthritis and rheumatoid arthritis before being voluntarily withdrawn in 2004 in the wake of concerns about adverse cardiovascular events, was associated with statistically insignificant treatment benefits in a 2005 randomized controlled trial of 1457 patients with MCI.11 Although unexpectedly low in both placebo and treatment groups, the conversion rate was higher in the rofecoxib group (6.4%) than in the placebo group (4.5%). Secondary end points also did not differ significantly between groups.

Several trials of other potential MCI therapies—including the diabetes drug rosiglitazone (Avandia), nicotine, and physical exercise—are currently under way. Many suspect, however, that the best outlook for patients with MCI will follow the FDA approval of 2 next-generation, disease-modifying AD drugs: tramiprosate (Alzhemed), which works to prevent amyloid from forming plaque; and R-flurbiprofen (Flurizan), which lowers levels of toxic ß-amyloid 42. Tramiprosate was designated as a fast-track product by the FDA in July 2007; however, updated phase 3 clinical trials data were still pending at press time. R-flurbiprofen phase 3 data are expected to be released in 2008.

In the meantime, because pharmacotherapeutic options for patients with MCI are limited, clinicians are focusing on treating patients' symptoms while monitoring their cognitive function. It is recommended that, in particular, patients should be treated for cardiovascular disease and depression and other behavioral disorders that complicate and exacerbate cognitive impairment.

"We should put this into the context of optimal aging: how can you optimize the aging process—especially cognitive aging—and what are the things that can protect you?" Duara said. "The big factors are lifestyle changes that optimize your ability to age successfully. That's the way I approach MCI, lacking any treatment."


Without any definitive data in support of MCI therapies, clinicians looking to provide their patients with factual information may be equally frustrated by the variability in published conversion rates. A recently published study from the University of Pittsburgh21 found that 51% of 136 patients with MCI progressed to dementia after a mean of 4.3 years, which is consistent with some previous studies conducted in referral clinics22,23; however, epidemiological studies have reported conversion rates ranging widely—from 3.5% after 2 years of follow-up24 to 43% after 5 years of follow-up.25 In the Pittsburgh study, 18% of patients with MCI returned to normal; in earlier studies, that figure ranged from 15% to 44%.1

Some of this variability is related to how MCI is being defined, and some of it is related to the populations being studied. "If someone comes into a memory disorder clinic or an AD center, the likelihood that the person just walking through the door has AD is about 80%. Someone probably brought him or her there and has been observing the person," Duara said. "If you're doing an epidemiological survey, and you find out that 5% of people over age 65 have MCI and that some of those people go back to normal, that's not surprising. Those persons didn't have dementing disease. So when you compare studies, you have to consider that the diagnosis or the rate of progression from MCI to dementia is determined by the likelihood that the patients actually have a dementia."

It is not only difficult to predict whether and how quickly MCI will progress but it also appears equally difficult to predict the outcome of progression. Although it has generally been thought that amnestic forms of MCI progress to AD and nonamnestic forms to other types of dementia, research from Vienna and from Leipzig, Germany, suggest that this is not necessarily the case.14,15 In the Austrian study, 26.8% of nonamnestic and 48.7% of amnestic patients converted to AD within 30 months. Several patients with amnestic MCI also went on to convert to vascular dementia or Lewy body dementia.

In the German community-based study of 980 nondemented persons, patients with nonamnestic-single domain MCI were more likely to progress to AD than to vascular or other dementias. However, those patients with nonamnestic-multiple domain MCI were the most likely to progress to non-AD dementias.


Despite all that is not yet known about MCI and despite the lack of available therapies, clinicians agree that making the diagnosis can be valuable.

"We often overlook how distressing memory impairment is to the patient and his or her family, how long it takes to get a diagnosis, and how infrequently primary care doctors arrive at a diagnosis in the early stages of dementia," Levey said. "Patients may seem to circle around the health care system for years looking for answers. You're doing good things by making an accurate early diagnosis."

An early diagnosis gives patients time to make decisions about the future of their care in the event that the disease does progress. It also gives them the opportunity to maximize the effect of any lifestyle changes they may make or any promising therapies they may be willing to try.

"This is exactly the stage when you want to start intervention," Farlow said. "In that sense, it is good to start seeing symptoms at this stage, when there is a possibility of a reversible cause."


1. Palmer K, Fratiglioni L, Windblad B. What is mild cognitive impairment? Variations in definitions and evolution of nondemented persons with cognitive impairment. Acta Neurol Scand Suppl. 2003;179:14-20.

2. Petersen RC, Thomas RG, Grundman M, et al. Vitamin E and donepezil for the treatment of mild cognitive impairment. N Engl J Med. 2005;352: 2379-2388.

3. Jack CR, Petersen RC, Grundman M, et al. Longitudinal MRI findings from the vitamin E and donepezil treatment study for MCI. Neurobiol Aging. 2007 Apr 21 [Epub ahead of print].

4. Salloway S, Ferris S, Kluger A, et al. Efficacy of donepezil in mild cognitive impairment: a randomized placebo-controlled trial. Neurology. 2004; 63:651-657.

5. Feldman HH, Ferris S, Winblad B, et al. Effect of rivastigmine on delay to diagnosis of Alzheimer's disease from mild cognitive impairment: the InDDEx study. Lancet Neurol. 2007;6:501-512.

6. Scheltens P, Fox NC, Barkhof F, Gold M. Effect of galantamine treatment on brain atrophy as assessed by MRI in patients with mild cognitive impairment. In: Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia. Abstract P2-158.

7. Gold M, Nye JS, Goldstein HR, Truyen L. Initial evaluation of galantamine for mild cognitive impairment: results from two double-blind, randomized, placebo-controlled studies. In: Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia. Hot Topics Presentation S5-04.

8. Tomlinson J, Cummins H, Wendt J, et al. SGS742, a novel GABAB receptor antagonist, improves cognition in patients with mild cognitive impairment. Neurology. 2004;62(suppl 5). Abstract P02.053:A128.

9. Cortex Pharmaceuticals Ampakine® fixed-dose CX516 cross national study of mild cognitive impairment (MCI) fails to meet primary end- point. Press release. Available at: http://www. Accessed August 8, 2007.

10. Johnson SA, Simmon VF. Randomized, double-blind, placebo-controlled international clinical trial of the ampakine CX516 in elderly participants with mild cognitive impairment: a progress report. J Mol Neurosci. 2002;19:197-200.

11. Thal LJ, Ferris SH, Kirby L, et al. A randomized, double-blind study of rofecoxib in patients with mild cognitive impairment. Neuropsycho- pharmacology. 2005;30:1204-1215.

12. Winblad B, Palmer K, Kivipelto M, et al. Mild cognitive impairment—beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med. 2004;256:240-246.

13. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med. 2004;256:183-194.

14. Busse A, Hensel A, Gühne U, et al. Mild cognitive impairment: long-term course of four clinical subtypes. Neurology. 2006;67:2176-2185.

15. Fischer P, Jungwirth S, Zehetmayer S, et al. Conversion from subtypes of mild cognitive impairment to Alzheimer dementia. Neurology. 2007; 68:288-291.

16. Bell-McGinty S, Lopez OL, Meltzer CC, et al. Differential cortical atrophy in subgroups of mild cognitive impairment. Arch Neurol. 2005;62:1393-1397.

17. Korf ES, Wahlund LO, Visser PJ, Scheltens P. Medial temporal lobe atrophy on MRI predicts dementia in patients with mild cognitive impairment. Neurology. 2004;63:94-100.

18. Geroldi C, Rossi R, Calvagna C, et al. Medial temporal atrophy but not memory deficit predicts progression to dementia in patients with mild cognitive impairment. J Neurol Neurosurg Psychiatry. 2006;77:1219-1222.

19. Jack CR Jr, Shiung MM, Weigand SD, et al. Brain atrophy rates predict subsequent clinical conversion in normal elderly and amnestic MCI. Neurology. 2005;65:1227-1231.

20. Tierney MC, Szalai JP, Snow WG, et al. A prospective study of the clinical utility of ApoE genotype in the prediction of outcome in patients with memory impairment. Neurology. 1996;46: 149-154.

21. Lopez OL, Kuller LH, Becker JT, et al. Incidence of dementia in mild cognitive impairment in the cardiovascular health study cognition study. Arch Neurol. 2007;64:416-420.

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

23. Morris JC, Storandt M, Miller JP, et al. Mild cognitive impairment represents early-stage Alzheimer disease. Arch Neurol. 2001;58:397-405.

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

25. Tuokko H, Frerichs R, Graham J, et al. Five-year follow-up of cognitive impairment with no dementia. Arch Neurol. 2003;60:577-582.

Copyright 2008 Jordana Foster – 24 Kirkland Dr, Stow, MA – Email: – Fax: (815) 346-5239