. Clinical, pathological, and neurochemical changes in dementia: a subgroup with preserved mental status and numerous neocortical plaques. Ann Neurol. 1988 Feb;23(2):138-44. PubMed.


Please login to recommend the paper.


  1. The main finding that cholinergic deficits were not evident in early to
    mid-stage Alzheimer's disease challenges the rationale for current Alzheimer's therapy, which has been the premise that cholinergic deficits exist early in the illness and contribute to the clinical symptoms, e.g., memory impairment. Therefore, a new research focus will be to reexamine the current Alzheimer's drugs to see whether their benefit in early to moderate Alzheimer's patients involves mechanisms other than simple cholinergic enhancement. Because the drugs do have benefit, there is no immediate impact on clinical practice (i.e., the drugs should continue to be used) but there may be long-term effects if the basis of the efficacy of the drugs is better understood.

    Another finding that may be newsworthy is the apparent upregulation of cholinergic activity in the hippocampus in very early disease ("MCI"). The authors suggest that this may provide an opportunity to intervene pharmacologically at this stage and presumably delay or prevent progression to greater stages of dementia severity. I interpret this finding also to further indicate that the condition of mild cognitive impairment (MCI) involves the exact same neural systems as are involved in Alzheimer's disease, such that MCI (in most cases) is not distinct from Alzheimer's disease but simply an earlier stage of the same illness. If my interpretation is plausible, then there may be clinical implications in that Alzheimer disease may eventually be diagnosed and treated at earlier stages than is now commonly practiced.

    View all comments by John Morris
  2. Author Comment


    Increasing research efforts are directed at detecting Alzheimer's in its earliest stages, including those before it is clinically symptomatic. Understanding the brain changes that occur early in the disease will be key to determining the most effective interventions. A state of mild cognitive impairment (MCI) has been identified in late life as a risk state to the development of AD. However, limited data are available on the pathological state or the status of neurochemical systems, especially the cholinergic system, in this very early state. Since the earliest and most prominent symptom of mild impairment and Alzheimer's disease is memory loss, and the cholinergic system is associated with recent memory function, it has long been felt that cholinergic loss was the basis for the loss of short-term memory in Alzheimer's disease and perhaps in mild cognitive impairment.
    The present study is the first to report levels of the synthetic cholinergic enzyme choline acetyltransferase (ChAT) in the hippocampus of humans with very early Alzheimer's disease and mild cognitive impairment. ChAT is needed to synthesize the neurotransmitter acetylcholine, known to be important in memory and cognitive function.

    The analyses were done in the brains of nuns and priests from the Religious Orders Study (ROS), as one of a number of research projects involving this cohort. These members of the ROS have agreed to be serially followed and assessed medically and neuropsychologically every year, and have agreed to autopsy after death.

    Three Findings:

    First, we found that in MCI, over 60 percent of the subjects so diagnosed already had neuropathological evidence of Alzheimer's disease at autopsy. Second, cholinergic enzymes in the hippocampus, one of the most vulnerable areas in Alzheimer's disease and felt to be significantly involved in recent memory dysfunction, were elevated above normal in the patients with mild cognitive impairment and were (back) at the normal level in patients with early diagnosed clinical AD. This indicates that as the pathology of Alzheimer's disease is developing in the human brain, cholinergic enzymes are probably upregulated as a compensatory response, perhaps to try to maintain normal function. In the AD cases, the levels of the cholinergic enzyme in this region were equal to those of the normal controls, indicating a decline in the ability to compensate. However, since these patients are all symptomatic with MCI or AD, even though they have normal or above normal levels of the cholinergic enzymes, it is likely that the memory deficit of MCI and Alzheimer's disease is not due, as long thought, to cholinergic deficits. It is more likely is due to earlier degeneration of inputs from the entorhinal cortex to the hippocampus; the entorhinal cortex is where AD appears to begin, or exhibit cellular changes first. Thus, the difficulty in entorhinal cortex interferes with the hippocampus getting information from the rest of the brain about what it is supposed to remember.

    Third, levels of the cholinergic enzymes in the cerebral cortex (on the brain surface) in MCI are normal in the several regions examined, with the exceptions of frontal cortex, in which like the hippocampus it was also elevated above normal. In the mild AD cases, levels of the enzyme were normal in all areas, with return of the levels in frontal cortex to normal. This also suggests that there is no major structural loss of the synthetic enzyme for the neurotransmitter acetylcholine in either MCI or early, mild AD.

    Implications for Clinical Practice and Further Research:

    First, controlled studies of the cholinesterase inhibitors should be undertaken in more advanced AD; they have thus far only been tested and approved for use in mild to moderate dementia. Second, we found the earliest symptoms of cognitive impairment do not appear due, as previously thought, to cholinergic deficits but more likely to earlier changes of AD, dysfunction and death of the entorhinal cortex neurons that supply the hippocampus with its cortical information. Efforts to slow down or stop this neurodegeneration of the inputs to the hippocampus, the brain's RAM chip, should be accelerated. This also provides strong rationale for development of even earlier diagnostic techniques for identification of subjects at risk to develop Alzheimer's disease.

    Research into how the regenerative efforts may delay clinical symptoms of the disorder, and how such compensatory efforts can be aided, is also a goal in these studies. Furthermore, since the pathology underlying the early cognitive changes in Alzheimer's disease may evolve from neurofibrillary pathology in the entorhinal cortex, these studies heighten the need to find ways to stop or slow down the neurofibrillar pathological alterations in Alzheimer's brain from its earliest beginnings in the temporal lobe.

    Additional studies to determine the status of the cholinergic system in mild cognitive impairment are needed. We need better methods for earlier detection of at-risk people and more clear-cut knowledge of when the irreversible neurodegenerative changes begin.

    The study was funded as part of a Program Project Grant from the National Institutes on Aging (NIA), which also funds the ROS, and was conducted in collaboration with Dr. Elliott Mufson and others at Rush Medical School in Chicago.

    View all comments by Steven DeKosky
  3. The study by DeKosky does not allow for conclusions about whether or not the cholinergic deficit is present at early stages of AD. The data show that there are differences in ChAT activity between a "no cognitive impairment" (NCI) group and a "mild cognitive impairment" group (MCI) but these groups should not be equated with "no Alzheimer's disease" and "mild Alzheimer's disease" since in both groups, about 60 percent of the cases had CERAD diagnoses of possible AD.

    Since even in the MCI group 40 percent of the cases do not have significant AD pathology, the study is really about the clinical groups (NCI and MCI), not about AD stages. This error was also made by Davis et al. [3] in a 1999 paper. To examine the question of whether or not the cortical cholinergic deficit is present at early stages in AD, we have divided cases on histopathologic grounds into a "normal" group (no plaques, with low Braak stage) and a "mild AD" group (diffuse plaques and/or sparse neuritic plaques, also with low Braak stage). Our published results [1;2] showed that the mild AD group had significantly lower cortical ChAT activities and cholinergic fiber densities. Katzman et al had similar findings in 1988 [4].


    . Cholinergic fibre loss associated with diffuse plaques in the non-demented elderly: the preclinical stage of Alzheimer's disease?. Acta Neuropathol. 1997 Feb;93(2):146-53. PubMed.

    . The cholinergic deficit coincides with Abeta deposition at the earliest histopathologic stages of Alzheimer disease. J Neuropathol Exp Neurol. 2000 Apr;59(4):308-13. PubMed.

    . Cholinergic markers in elderly patients with early signs of Alzheimer disease. JAMA. 1999 Apr 21;281(15):1401-6. PubMed.

    View all comments by Thomas Beach
  4. Authors' Reply
    We appreciate Dr. Beach's comment on our publication, and his comments concerning a suggested relationship between cholinergic deficits and amyloid pathology. However, he has misinterpreted our (DeKosky et al., 2002) and Davis et al. 1999 studies of cholinergic changes in clinically diagnosed MCI and mild AD. Our studies examined clinical diagnostic groups; that was clearly stated in both publications. This is not to be considered an "error": this is what the subjects manifested in behavior. Carefully studying clinically diagnosed subjects is the only logical approach to address the hypothesis of cholinergic deficits in early stages of AD dementia.

    Dr. Beach's argument compares apples and oranges. It is worth noting that one cannot diagnose Alzheimer's disease on neuropathological grounds alone; in the absence of cognitive symptoms prior to death, even clear-cut neuropathological evidence of plaques and tangles can only gain a diagnosis of possible AD. But many people die without ever manifesting clinical dementia, so stratifying on the basis of histopathology alone is not the answer. Neither Beach's NOR our data answer the question about whether there is ANY cholinergic deficit in early disease, since we measured ChAT and not cholinergic function. What our study shows, as did Ken Davis' study in cortex (Davis et al's 1999), is that significant loss of ChAT does NOT occur early in the disease and is not the correlate of the cognitive loss in early dementia, as we have believed since the 1970s.

    In contrast, Dr. Beach refers in his work (Beach et al., 2000; Beach et al., 1997) to people who came to autopsy without having been cognitively evaluated prior to death but who displayed some degree of AD-like pathology. Despite such changes, it cannot be foreseen that this population of elderly individuals would ever develop clinical dementia (note the advanced age of the subjects in our study, who had some degree of AD pathology but no clinical impairment). Thus, the work of Beach does not clearly distinguish between neuropathological and clinical diagnoses of AD as it related to cholinergic dysfunction.

    The second study of Dr Beach (Beach et al., 1997) does not define a cholinergic deficit but an alteration in the hydrolyzing enzyme AChE. It is possible that this is an attempt by the brain to enhance cholinergic function in cortex, similar to the effects of anti-cholinesterase drugs designed to boost cholinergic neurotransmission. In the absence of clinical information about these subjects it is an interesting finding but does not address the issues that are addressed in our paper and that of Davis.

    We disagree strongly with his contention that we should not equate NCI with "no AD". As the name implies ("no cognitive impairment"), NCI had no clinical signs of AD dementia. Likewise, the MCI cases were clinically distinguished from "mild AD" on clinical grounds. Our findings, as well as that of Davis and colleagues, demonstrate that ChAT deficits are absent in the early clinical stages of AD. Although the effect of pathology is important (the extent of pathologic changes by NIA-Reagan criteria was provided for each diagnostic group), the central focus of our study was to determine if there was loss of ChAT in people with clinical cognitive impairment, either MCI or mild dementia. The answer is clearly no. Future studies need to determine the relationship of loss of ChAT with AD pathologic changes, interactions with neurotransmitter and metabolic alterations in all the clinical diagnostic groups, and why widely varying degrees of pathological change may occur in absence of clinical dementia.


    . Upregulation of choline acetyltransferase activity in hippocampus and frontal cortex of elderly subjects with mild cognitive impairment. Ann Neurol. 2002 Feb;51(2):145-55. PubMed.

    . Cholinergic markers in elderly patients with early signs of Alzheimer disease. JAMA. 1999 Apr 21;281(15):1401-6. PubMed.

    . The cholinergic deficit coincides with Abeta deposition at the earliest histopathologic stages of Alzheimer disease. J Neuropathol Exp Neurol. 2000 Apr;59(4):308-13. PubMed.

    . Cholinergic fibre loss associated with diffuse plaques in the non-demented elderly: the preclinical stage of Alzheimer's disease?. Acta Neuropathol. 1997 Feb;93(2):146-53. PubMed.

    View all comments by Steven DeKosky

Make a Comment

To make a comment you must login or register.