Neuropathologic Features of Amnestic Mild Cognitive Impairment

ORIGINAL CONTRIBUTION Neuropathologic Features of Amnestic Mild Cognitive Impairment

Ronald C. Petersen, PhD, MD; Joseph E. Parisi, MD; Dennis W. Dickson, MD; Kris A. Johnson, RN; David S. Knopman, MD; Bradley F. Boeve, MD; Gregory A. Jicha, MD, PhD; Robert J. Ivnik, PhD; Glenn E. Smith, PhD; Eric G. Tangalos, MD; Heiko Braak, MD; Emre Kokmen, MD†

Background: The neuropathologic substrate of amnes- Reagan criteria were used to analyze autopsy tissue from tic mild cognitive impairment (aMCI) is not known. 15 individuals who died while their clinical diagnosis was aMCI. For comparison, autopsy data on age-matched Objective: To determine the neuropathologic features groups of clinically healthy individuals and patients with of patients who died while their clinical classification was probable AD were analyzed. aMCI. Results: Most patients with aMCI did not meet the neu- Design: Cohort study. ropathologic criteria for AD, but their pathologic findings suggest a transitional state of evolving AD. All the patients Setting: Community based. with aMCI had pathologic findings involving medial temporal lobe structures, likely accounting for their memory Participants: Sixty-six individuals, including 15 who impairment. In addition, there were many concomitant had memory impairment beyond that allowed for aging pathologic abnormalities, including argyrophilic grain dis- but who were not demented, were studied along with 28 ease, hippocampal sclerosis, and vascular lesions. clinically healthy individuals and 23 patients with probable Alzheimer disease (AD) for comparison. Conclusions: The neuropathologic features of aMCI matched the clinical features and seemed to be interme- Main Outcome Measures: Standard neuropatho- diate between the neurofibrillary changes of aging and logic techniques and classification according to the pathologic features of very early AD. Khachaturian, Consortium to Establish a Registry for Alzheimer Disease, and National Institute on Aging– Arch Neurol. 2006;63:665-672

HE FIELD OF AGING AND DE- ment and likely progresses to AD, other sub- mentia is moving toward types, with different clinical criteria, have earlier identification of in- been proposed.10 One challenge in study- cipient disease states. The ing the transition between normal aging and concept of mild cognitive AD is the paucity of pathologic material on impairment (MCI) has evolved in recent patients with MCI. There are a few studies Author Affiliations: T years to represent the clinical transition be- in the literature on the neuropathologic fea- Alzheimer’s Disease Research tures of patients who are clinically mild and Center (Drs Petersen, Parisi, tween the cognitive changes found in normal aging and those of early Alzheimer dis- very few on patients who died while their Dickson, Knopman, Boeve, 11-17 Jicha, Ivnik, Smith, Tangalos, ease (AD).1,2 Several international clinical clinical diagnosis was MCI. and Kokmen and Ms Johnson) trials3-5 evaluating a variety of interven- and Departments of Neurology tions for MCI have recently concluded. The For editorial comment (Drs Petersen, Knopman, American Academy of Neurology has en- Boeve, Jicha, and Kokmen), see page 645 Laboratory Medicine and dorsed the concept of MCI with an evi- Pathology (Dr Parisi), dence-based medicine practice param- The present study draws on a longitu- Neurosciences (Dr Dickson), eter paper highlighting the importance of dinal, community-based study of aging and Psychiatry and Psychology MCI as a clinical entity.6 dementia.2,18 Some members of this co- (Drs Ivnik and Smith), and There is debate in the field regarding the hort have been followed for up to 18 years. Internal Medicine specific clinical criteria, longitudinal out- Between September 1, 1986, and Decem- (Dr Tangalos), Mayo Clinic come, and underlying neuropathologic fea- ber 31, 2004, 15 patients died and under- College of Medicine, Rochester, tures of MCI.7-9 It has been recognized that Minn, and Jacksonville, Fla; and went autopsy while their clinical classifi- Institute for Clinical Anatomy, the concept of MCI may be heterog- cation was still aMCI. As such, this is one J.W. Goethe University, eneous, and although the most common of the first studies to characterize the neu- Frankfurt, Germany (Dr Braak). subtype, that is, amnestic MCI (aMCI), pre- ropathologic substrate of patients still in †Deceased. sents with prominent memory impair- the clinical state of aMCI.

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 665

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 METHODS AUTOPSY PROCEDURE The brains were processed according to the protocol of the Mayo Alzheimer Disease Research Center Neuropathology Core. Af- CLINICAL EVALUATION ter external examination of the fresh specimen, the brain was divided by a sagittal cut into the right and left hemibrains. One During the past 18 years we followed more than 270 people hemisphere (usually the right) was serially sectioned into 1-cm who have received the diagnosis of MCI at some point in their coronal slabs, and alternate slices were frozen and fixed in para- clinical course. All were participants in the National Institute formaldehyde; the other hemisphere (usually the left) was fixed on Aging (NIA)–supported Mayo Clinic Alzheimer Disease for 7 to 10 days in 10% to 15% buffered formalin and was coro- Patient Registry.18 Participants were recruited prospectively in nally sectioned into 1-cm slabs and examined. The neuropatho- this project from primary care practices of the Division of logic protocol was performed according to the recommenda- Community Internal Medicine of the Mayo Clinic in Roches- tions of the Consortium to Establish a Registry for Alzheimer’s ter. The medical records of individuals 65 years and older Disease (CERAD)24 and the consensus guidelines for the clini- who had received a general medical examination were evalu- cal and pathological diagnosis of dementia with Lewy bodies.25 ated for any suggestion of a cognitive concern raised by the The following brain areas were routinely sampled: superior/ patient, a family member, or the examining physician. If a middle frontal gyri (plane just anterior to the temporal tip), su- concern was raised about cognitive function, permission was perior/middle temporal gyri (plane of the mammillary body), requested to approach the patient for evaluation. Patients then inferior parietal lobule (plane 1 cm behind the posterior pole underwent neurologic evaluation and neuropsychologic test- of the splenium), calcarine (primary visual) cortex, anterior cin- ing, neuroimaging, and laboratory studies. These studies have gulate (plane of the anterior commissure), hippocampus with been approved by the Mayo institutional review board. adjacent inferior temporal cortex (level of the lateral genicu- Patients were diagnosed as having aMCI if they fulfilled late body), amygdala and entorhinal cortex (level of the mam- the following criteria: (1) memory concerns, usually by the millary bodies), nucleus basalis, basal ganglia, cerebellum, thala- patient, preferably corroborated by an informant, (2) objective mus with subthalamic nucleus, midbrain (with substantia nigra), memory impairment for age, (3) essentially normal general pons (with locus ceruleus), and cerebellum, as well as repre- cognitive function as judged by the physician, (4) normal sentative sections of any lesions noted grossly. activities of daily living as judged by the physician, and (5) After routine processing in paraffin and cutting, sections were not demented. The diagnosis was made on a clinical basis; stained with hematoxylin-eosin, and selected sections were that is, a consensus committee comprising neurologists, a stained with modified Bielschowsky, Luxol fast blue/periodic geriatrician, neuropsychologists, nurses, and other study per- acid–Schiff, and thioflavin-S and immunostained for ␤-amy- sonnel adjudicated each case. Specific neuropsychologic cut- loid (clone 6F/3D) (Novocastra Laboratories Ltd, Newcastle off scores were not used; instead, the patients were diagnosed upon Tyne, England), phosphorylated tau (clone AT8) (En- as having aMCI if their memory performance was impaired dogen, Woburn, Mass), ␣-synuclein (clone LB509) (Zymed out of proportion to their other cognitive domains.2,10 These Laboratories, San Francisco, Calif), ubiquitin (polyclonal) individuals may have had isolated memory impairments (Dako, Glostrup, Denmark), neurofilament protein (clone 2F11) (single-domain aMCI) or may have mild impairments in non- (Dako), ␣-␤-crystallin (polyclonal) (Chemicon International memory domains (multiple-domain aMCI). Their activities of Inc, Temecula, Calif), and glial fibrillary acidic protein (poly- daily living were essentially intact except for some of the clonal) (Dako). minor impairments related to the memory disorder as documented by the Record of Independent Living and the Clinical Dementia Rating scale.19,20 NEUROPATHOLOGIC ASSESSMENT After initial assessment and enrollment, patients were reevaluated annually. At each reevaluation they were Regions in which the counting was performed were selected assessed clinically, and their performance was reviewed at macroscopically, according to the method of Duyckaerts et al,26 the consensus conference. The consensus conference com- as the areas of the slide where the cortical ribbon was the thin- mittee determined whether the patient continued to fulfill nest (ie, where the angle of section was the closest to 90° rela- the criteria for MCI, had reverted to normal, had progressed tive to the cortical pial surface). Quantitative pathologic find- to clinically probable AD, or had developed another form of ings were recorded, according to the method of Dickson et al,27 dementia. We used the Diagnostic and Statistical Manual of from ϫ200 microscopic fields, excluding fields at the crest of Mental Disorders, Third Edition,21 for the diagnosis of demen- the gyri or the depth of the sulci. Ten contiguous fields were tia and the National Institute of Neurologic and Communi- examined for all slides, except in the case of the amygdala, hip- cative Disorders and Stroke–Alzheimer’s Disease and Related pocampus, and entorhinal cortex, where pathologic findings Disorders Association22 criteria for AD. The clinical evalua- from 5 contiguous fields were recorded. In an attempt to limit tion of the patients with MCI has been discussed in detail the possible variability in staining techniques and in re- elsewhere.2,10 searcher interpretation, plaques were defined as diffuse (DP), For comparison, we also describe 28 individuals who were cored without dystrophic neurites (CP), and neuritic (NP). Dif- clinically healthy at the time of death (Clinical Dementia Rat- fuse plaques were identified as deposits of finely granular ma- ing of 0) and were age-matched at death to the patients with terial in the neuropil, sometimes with a more dense central core. MCI.23 We also describe a group of 23 age-matched patients Cored plaques lacked dystrophic neurites by silver staining or with clinically probable AD who had undergone autopsy. Both tau immunostaining but had distinct amyloid cores. Neuritic groups had been enrolled in and had undergone the same in- plaques were identified by the presence of dystrophic neu- vestigations using the same procedures as the patients with aMCI rites, arranged radially to form a discrete spherical lesion av- in the Mayo Alzheimer Disease Patient Registry/Alzheimer Dis- eraging approximately 30 µm in diameter. Using a combina- ease Research Center. These groups are described for the pur- tion of silver, anti-amyloid– and anti-tau–stained sections, plaque pose of comparison from a research group using the same pro- subtypes were separated and counted. Neurofibrillary tangles cedures for recruitment, evaluation, and postmortem evaluation (NFTs) were counted on Bielschowsky- and anti-tau–stained as used for the patients with aMCI. sections. All the results were normalized and expressed as num-

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 666

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 Table 1. Demographic, Clinical, and Apolipoprotein E 50 Features of 15 Patients With Mild Cognitive Impairment Healthy aMCI AD 40 Patient No./ Education, MMSE DRS AVLT DR, % Sex/Age, y y Score Score Score Retained APOE* 1/M/74 8 27 130 28 16 3/3 30 2/F/80 16 25 122 25 0 3/3

3/F/86 8 23 131 25 12 2/4 20

4/F/86 14 25 140 18 0 3/4 Participants, % 5/F/86 11 29 129 30 20 NA 6/M/86 12 26 137 27 66 NA 10 7/M/90 20 29 136 34 0 3/3 8/M/90 16 29 135 30 42 3/3 0 9/M/91 11 24 105 23 0 3/4 0 I II III IV V VI 10/F/93 16 28 111 22 50 3/3 Braak Stage 11/F/93 12 24 119 20 0 3/3 12/F/93 9 25 128 28 50 3/3 13/F/94 14 26 132 NA NA 2/3 Figure 1. Braak staging for healthy individuals, patients with amnestic mild cognitive impairment (aMCI), and patients with Alzheimer disease (AD). 14/F/94 16 24 116 19 0 3/3 15/F/98 20 27 125 26 0 3/3 of NPs and NFTs as being consistent with the neuropathologic Abbreviations: APOE, apolipoprotein E; AVLT, Auditory Verbal Learning diagnosis of AD and then make a final probabilistic assessment Test34 (sum of 5 learning trials); DRS, Mattis Dementia Rating Scale; of low, intermediate, or high probability of AD. DR, delayed recall on AVLT; MMSE, Folstein Mini-Mental State Examination; NA, not available. *Numbers indicate APOE genotypes. RESULTS

During the study, 15 patients died and underwent au- ber per square millimeter. Semiquantitative NP counts were per- topsy while their clinical classification was aMCI (single formed according to CERAD guidelines24,28; to ensure unifor- mity in estimates between evaluators, the following guidelines or multiple domain). Post mortem, their medical histo- were applied: sparse NPs consisted of 1 to 5 NPs/100ϫfield ries were reviewed to determine the condition of the pa- (grade 1); moderate NPs, 6 to 19 NPs/100ϫfield (grade 2); and tients between their last clinical evaluation by the re- frequent NPs, 20 or more NPs/100ϫfield (grade 3). Neuro- search team and their time of death. The median interval pathologic staging of NFT distribution was performed accord- between the last clinical evaluation and death was 0.72 ing to the scheme of Braak and Braak.29 The presence of Lewy year (263 days), with a range of 0.19 to 1.18 years (69- bodies in the substantia nigra, locus ceruleus, amygdala, cin- 430 days). Postmortem reviews confirmed that they had gulate cortex, and neocortex was noted using hematoxylin- ␣ remained mildly impaired. Six patients died of cardiac eosin, -synuclein, and ubiquitin immunostaining. Hematoxylin- causes, 3 of pulmonary complications, 4 of cancer, 1 of eosin–stained sections were used for the assessment of infarcts renal insufficiency, and 1 of trauma. and arteriolar disease. The size, location, and temporal relationship to clinical findings of large and small vessel infarcts The demographic, clinical, and apolipoprotein E fea- were recorded. Infarcts determined to represent acute or sub- tures of the 15 patients are given in Table 1. The mean acute processes proximal to death were not included in fur- (SD) age of the group at death was 88.9 (6.1) years, and ther analyses. Microvascular disease (arteriosclerosis) and mi- their education level was 13.5 (3.9) years. The aMCI sta- croinfarcts were assessed using a semiquantitative grading scale tus at the final clinical evaluation was documented by a mean (none, mild, moderate, and severe) according to the National (SD) Mini-Mental State Examination score of 26.1 (2.0); a Alzheimer Coordinating Center Neuropathology Data Manual mean (SD) Dementia Rating Scale score of 126.4 (10.1), Version 1.0 (available at https://www.alz.washington.edu with impaired learning documented on the Auditory Ver- /NONMEMBER/PDF/npded2004.pdf). The presence or ab- bal Learning Test with a cumulative mean (SD) learning sence of vascular contributions to cognitive decline in these pa- score across 5 trials of 25.4 (4.6) (approximately 10th- tients was then dichotomized to either present or absent according to consensus agreement between 2 examining neu- 19th percentile for age- and education-matched individu- ropathologists (J.E.P. and D.W.D.), as described previously.30 als); and mean (SD) delayed recall of 18.3% (23.6%) (0- 34 Argyrophilic grain disease (Braak disease) was characterized 9th percentile). There were 3 apolipoprotein E4 carriers by the presence of hippocampal and amygdaloid silver- and tau- among the 13 patients on whom apolipoprotein data were positive grains, as well as tau-positive coiled bodies in white available. The mean (SD) time since the patients were di- matter, and amygdaloid “ballooned” neurons as confirmed with agnosed as having aMCI was 2.8 (2.5) years. neurofilament or ␣-␤-crystallin immunoreactivity.31 Braak staging for the patients with aMCI compared After the neuropathologic assessments, each case was clas- with the healthy and AD groups is shown in Figure 1. sified according to current and past neuropathologic criteria: the 32 24 The modal stages for the patients with aMCI were II and Khachaturian criteria, the CERAD criteria, and the NIA- III, accounting for 10 of the 15 patients. The neuropatho- Reagan criteria.33 The Khachaturian criteria are most liberal with respect to meeting neuropathologic criteria for AD and focus on logic data according to Braak staging; DPs, CPs, and NPs; age-related counts of any type of plaque (DP or NP). The CERAD and quantitation of Lewy bodies and vascular disease are criteria require a semiquantitative rating of NP density as sparse, given in Table 2. The neuropathologic classifications intermediate, or frequent. Neither of these sets of criteria ac- of the data vary according to the specific criteria used. counts for NFTs. The NIA-Reagan criteria rate the probability For example, according to the Khachaturian criteria, 7

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 667

Patient No./ Diffuse Amyloid Cored Neuritic Braak and Lewy Sex/Age, y Plaques* Plaques* Plaques* Braak Stage29 Bodies† Comments Patients With Amnestic Mild Cognitive Impairment 1/M/74 0 0 0 0 0 Normal 2/F/80 0 0 0 II 0 AGD 3/F/86 2 1 0 III 0 NFT, AGD, chronic left posterior cerebral infarct 1.5 ϫ 2.5 ϫ 5.0 cm involving lingual gyrus, splenium, to occipital pole 4/F/86 2 1 0 V 0 Early AD 5/F/86 0 0 0 III 0 NFT, AGD, right HS 6/M/86 0 0 0 II 0 Early AGD/normal, chronic lacunar infarcts right thalamus 0.8 ϫ 0.3 cm, left caudate 0.4 ϫ 0.3, and right parietal centrum semi-ovale (minute) 7/M/90 3 2 0 III 0 Incipient AD, chronic left parietal lobe microinfarct 8/M/90 3 2 0 II 0 PA, chronic left inferior midfrontal gyrus infarct 3 ϫ 2 ϫ 2cm 9/M/91 0 0 0 III 0 NFTs, AGD 10/F/93 1 1 0 III 0 AGD 11/F/93 0 0 0 II 0 AGD, left HS extending anteriorly to involve left amygdala 12/F/93 3 2 1 IV T DLB 13/F/94 3 2 2 V 0 Early AD, chronic infarct left superior bank of calcarine cortex 2 ϫ 1 ϫ 1 cm, subacute infarct left lateral occipital cortex 14/F/94 3 2 2 IV 0 Early AD 15/F/98 0 0 0 II 0 HS, 2 meningiomas involving left supraclinoid internal carotid artery region (anteromedial aspect of left middle cranial fossa); a) meningothelial, 2.0-cm diameter; b) transitional, 0.6-cm diameter Cognitively Healthy Individuals 16/F/74 3 3 2 V 0 Early AD 17/M/74 0 0 0 II 0 AGD 18/F/80 0 0 0 0 0 Normal 19/M/80 1 0 0 I 0 Frontal infarct 20/F/80 1 0 0 I 0 Caudate infarct, temporal pole contusions 21/M/84 3 2 0 I 0 PA 22/M/84 2 1 1 I 0 Normal 23/F/84 3 2 1 I 0 Normal 24/F/84 1 0 0 II 0 Normal 25/F/84 3 0 0 II 0 Neuronal loss, pallor, and gliosis in substantia nigra 26/F/84 3 1 1 II 0 Normal 27/M/85 3 1 0 I 0 Caudate infarct 28/F/85 1 1 1 I 0 Caudate infarct 29/M/85 1 0 0 II T DLB, frontal infarct 30/M/86 3 1 0 I B PD 31/M/86 3 1 0 II 0 Normal 32/F/89 0 0 0 III 0 AGD, temporal lobe telangiectasia 33/M/89 0 1 0 III 0 Caudate infarct 34/F/89 3 1 1 V 0 PA 35/F/90 2 0 0 I 0 Normal 36/F/90 3 1 1 III 0 Normal 37/F/91 2 1 1 III 0 Normal 38/M/92 1 0 0 I 0 Normal 39/F/93 0 0 0 II 0 Normal 40/F/93 0 0 0 II 0 Caudate infarct 50/M/93 1 0 0 II 0 Normal 51/F/93 1 1 0 III 0 Parietal infarct 52/F/95 1 1 1 I T DLB

(continued)

of 15 patients with aMCI were classified as fulfilling the Figure 2 shows the frequency of participants in all 3 criteria for AD, whereas 8 were not. Assuming that the groups meeting the criteria for low, intermediate, or high aMCI diagnosis was sufficient to fulfill the clinical im- likelihood of having AD by NIA-Reagan criteria. Most of pairment criteria required by the CERAD, 3 patients with the patients with MCI (n=11) had a low likelihood of aMCI were classified by CERAD criteria as having prob- having AD, whereas 3 had an intermediate and 1 a high able AD, and another 2 had possible AD. However, 10 likelihood of having AD. of the patients had insufficient neuropathologic find- For comparison, the neuropathologic data for the ings to classify them as having AD by CERAD criteria. healthy and probable AD groups are also given in Table 2

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 668

Patient No./ Diffuse Amyloid Cored Neuritic Braak and Lewy Sex/Age, y Plaques* Plaques* Plaques* Braak Stage29 Bodies† Comments Patients With AD 53/M/78 3 2 1 IV 0 Incidental pontine cavernous angioma 54/F/79 3 2 3 V N DLB 55/F/83 3 2 0 III 0 Pure AD 56/F/85 3 3 2 VI N DLB, AGD 57/F/85 3 2 2 IV 0 Thalamic infarct 58/M/85 3 3 3 V T DLB 59/F/86 3 3 2 IV 0 Severe arteriosclerosis and atherosclerosis of the circle of Willis, without evidence of amyloid angiopathy 60/M/86 3 3 3 VI T DLB 61/F/88 3 3 3 VI 0 AGD 62/F/88 3 2 2 VI 0 Pure AD 63/M/90 3 3 3 IV 0 NPH, Chiari malformation 64/F/90 3 3 2 IV 0 AGD 65/M/92 3 3 1 IV 0 Parietal and cingulate infarcts 66/F/92 3 3 3 VI T DLB 67/F/92 3 3 1 IV 0 Pure AD 68/F/92 3 2 2 IV 0 Pure AD 69/F/93 3 3 3 VI 0 AGD 70/F/94 3 1 0 III 0 Pure AD 71/F/95 2 3 3 VI 0 Incidental posterior falx meningioma 72/F/96 3 3 1 IV 0 AGD 73/F/96 3 2 3 V 0 Caudate infarct 74/F/103 3 3 1 V 0 Caudate infarct 75/F/103 2 2 0 V 0 HS, AGD

Abbreviations: AD, Alzheimer disease; AGD, argyrophilic grain disease; DLB, dementia with Lewy bodies; HS, hippocampal sclerosis; NFT, neurofibrillary tangle dominant abnormality; NPH, normal pressure hydrocephalus; PA, pathologic aging; PD, Parkinson disease. *Semiquantitative plaque scoring system: 0 indicates none; 1, sparse; 2, moderate; and 3, frequent. †Lewy body classification: B indicates brainstem/Braak stages I and II; T, transitional/limbic/Braak stages III and IV; and N, neocortical/Braak stages V and VI.

100 3.0 Healthy Healthy aMCI aMCI AD AD 80 2.5

2.0 60

1.5 40 Participants, % 1.0 Scaled Plaque Scores 20

0.5

0 Low Intermediate High 0 Likelihood of Having AD Diffuse Plaques Cored Plaques Neuritic Plaques

Figure 2. Distribution of diagnoses by National Institute on Aging–Reagan Figure 3. Mean plaque scores for diffuse, cored, and neuritic plaques for criteria33 for healthy individuals, patients with amnestic mild cognitive healthy individuals, patients with amnestic mild cognitive impairment (aMCI), impairment (aMCI), and patients with Alzheimer disease (AD). and patients with Alzheimer disease (AD). Error bars represent SD.

and Figures 1 and 2. In general, for the healthy individu- Some patients with aMCI had other concomitant neu- als, most had Braak stages of I to III and a 0 (none) or 1 ropathologic features, which likely contributed to the clini- (sparse) rating of DPs. Cored plaques were uncommon, cal presentations of the patients. In particular, there was with a mean (SD) score of 0.68 (0.77), and NPs were very argyrophilic grain disease in 7 patients with aMCI,31,35 uncommon, with a mean (SD) score of 0.36 (0.56); the with the extent of involvement ranging from occasional modal number of NPs was 0 (13 of 28) (Table 2 and grains in the medial temporal lobe to argyrophilic grains Figure 3). In contrast, the age-matched patients with constituting the major pathologic abnormality (patient AD showed a Braak stage of IV to V, with a mean DP score 2). In other patients, argyrophilic grains were seen in con- of 2.91 (0.29) and frequent CPs (mean [SD], 2.57 [0.59]) junction with NFTs (patients 6, 9, and 10), DPs (patient and NPs (mean [SD], 1.91 [1.08]) (Table 2 and Figure 3). 3), or hippocampal sclerosis (patients 5 and 11) (Table 2).

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 669

Middle Frontal Cortex, β-Amyloid

Hippocampal Sector CA1, Tau AT8

Figure 4. Representative images of the middle frontal cortex and hippocampal sector CA1 from a healthy individual (patient 29), a patient with amnestic mild cognitive impairment (aMCI) (patient 3), and a patient with Alzheimer disease (AD) (patient 51) stained with antibodies to ␤-amyloid and tau (AT8), respectively (original magnification, ϫ100).

Concomitant vascular disease was seen in 5 patients, with Examination and the Dementia Rating Scale, with rela- definite contributions to cognitive decline in only a single tive impairment in learning and recall as characterized patient (patient 6) and another 4 patients with possible by Auditory Verbal Learning Test performance. contributions (patients 3, 7, 8, and 13). Hippocampal scle- The predominant location of 1 or more of several neu- rosis was seen in 3 patients, alone in patient 15 and with ropathologic features was in medial temporal lobe struc- argyrophilic grains in patients 5 and 11. Finally, a few tures, and that likely accounted for the memory impair- patients were classified as having neuropathologic AD, ments seen in these patients clinically. Typically, these including 3 who met intermediate probability (patients changes consisted of NFTs in the entorhinal cortex and hip- 4, 7, and 14) and 1 who met high probability (patient pocampal formation that can be seen in early incipient AD. 13) for AD by NIA-Reagan criteria (Table 2). None of the Many cases had the deposition of diffuse amyloid in the patients had definite AD abnormalities with a high Braak neocortex, with relatively few NPs and a Braak stage of III stage and frequent NPs (by CERAD criteria). Figure 4 or less. This constellation of findings has been referred to shows amyloid and neurofibrillary pathologic features for as pathologic aging by some investigators, but its true role typical healthy, aMCI, and AD cases. in the distinction between normal aging and AD is uncer- tain.27 Most of the cases were characterized by the neuro- COMMENT pathologists as having prodromal or incipient AD, mean- ing that they did not fulfill the criteria for AD but were These data describe in detail the neuropathologic find- suggestive of being in transition (diffuse amyloid in the neo- ings of prospectively characterized and longitudinally fol- cortex and frequent NFTs in medial temporal lobe struc- lowed patients who died while their clinical diagnosis was tures). The role of argyrophilic grains in AD development aMCI compared with similar patients with AD and in- is not known, but a recent study36 suggests that more than dividuals with no cognitive impairment. We found that 25% of patients with AD have argyrophilic grains, and in- the regional involvement by NFTs correlated best with creased frequency of argyrophilic grain disease in AD with the degree of clinical impairment across the spectrum of 4R tau-specific immunohistochemical findings has also been healthy to aMCI to AD. In contrast, the amyloid plaque reported. Hippocampal sclerosis is not believed to be on burden was less discriminating. In general, patients with the AD spectrum, but it may be related to ischemic factors AD had more CPs and NPs, whereas the amyloid bur- and neurodegeneration in others.37,38 den of patients with aMCI was more similar to that of Because the sets of neuropathologic criteria for AD the healthy individuals. Thus, it may be that the transi- range from the most liberal (Khachaturian) to interme- tion to dementia occurs when neurofibrillary abnormali- diate (CERAD) to the most restrictive (NIA-Reagan), the ties spread beyond the medial temporal lobe. proportion of patients with aMCI classified as AD di- Several demographic features of these patients are note- minished. That is, if the relatively loose Khachaturian cri- worthy. They were quite old (mean age, 89 years). The teria for AD were invoked recognizing age-adjusted counts prevalence of apolipoprotein E4 carriers was low at 3 of of any form of amyloid plaque, 7 of 15 patients with aMCI 13, and 1 of these had the 2/4 genotype. This likely re- met the criteria for AD. However, even with this most flects the advanced age of this group. These patients met liberal set of criteria, fewer than 50% of the group would the clinical criteria for aMCI and had normal indices of be classified as having AD. Using CERAD criteria, which general cognitive function for age on the Mini-Mental State emphasizes NPs, only 5 of 15 patients met the criteria

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 670

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 for possible or probable AD. Using the currently ac- of the present study because the patients in the Wash- cepted NIA-Reagan criteria for AD, including NPs and ington University study were demented at the time of NFTs, only 1 patient was believed to have a high prob- death, and this group uses modified Khachaturian cri- ability of representing AD changes, 3 had intermediate teria for the neuropathologic diagnosis of AD, which are probability, and the rest had low or no probability. The far more liberal than those used herein. This research Braak staging corroborated this classification, with most group does not use the MCI classification and focuses patients being between Braak transentorhinal and lim- on patients with very mild clinically probable AD. In so bic stages. Typically, a Braak stage IV is believed to be doing, they demonstrated that when most of their pa- compatible with AD. Therefore, most patients had “AD tients with mild AD underwent autopsy, most had fea- tendencies” but did not meet the criteria for definite AD. tures of AD neuropathologically. It should be noted, however, that there was a consid- A study15 investigating the distribution of NFTs and erable amount of heterogeneity in the neuropathologic amyloid plaques in 5 healthy individuals and 3 patients findings. Many patients had some argyrophilic grain dis- with MCI reached conclusions similar to ours. These in- ease. This has been noted by Braak and Braak,31 Tolnay vestigators found that the NFT distribution followed a et al,35 and others,39 but the relationship of the grain ab- hierarchical pattern and correlated with memory perfor- normality to clinical symptoms is unclear. It is likely that mance, whereas the density of amyloid plaques did not. argyrophilic grains were an incidental finding in some There are limitations to the present study. The num- patients and, in a few, may have been the primary patho- ber of patients who died while their clinical classifica- logic abnormality. Hippocampal sclerosis was also pres- tion was aMCI was small. Our patients were quite old. ent in a few patients, and it is accepted that this entity Individuals who die and undergo autopsy while in the can produce clinical features similar to those of incipi- aMCI state may not represent the universe of all pa- ent AD.37,40 Finally, several patients had features of sub- tients with MCI. A study that evaluates the neuropatho- cortical ischemic-vascular disease.41 logic outcome of patients who once were classified as The healthy and AD comparison groups provide an in- aMCI but subsequently progressed to another diagnosis teresting contrast for the interpretation of the neuropatho- may represent the more typical MCI case.44 logic features of aMCI. Comparing the various types of In summary, although the concept of aMCI remains plaques (DP, CP, and NP), it is apparent that the patients under study, data are converging on the clinical charac- with MCI seem more like the healthy individuals than those terization, rates of progression, clinical predictors, and with AD. With respect to NFT distribution as measured now the neuropathologic substrate. This study provides by Braak staging, the MCI group seemed more advanced. data on individuals who died while their clinical classi- These findings complement neuropathologic studies fication was aMCI and contributes to our understand- on early AD. An important recent article from the Nun ing of the neuropathologic substrate of these patients. Study42 explored the relationship between AD neurofibrillary pathologic features and intermediate stages of cog- Accepted for Publication: September 8, 2005. nitive impairment. These investigators found a strong re- Correspondence: Ronald C. Petersen, PhD, MD, Mayo lationship between neurofibrillary pathologic features and Clinic College of Medicine, 200 First Street SW, Roch- cognitive state across the clinical spectrum, but they also ester, MN 55905 ([email protected]). noted that by excluding other non-AD abnormalities, they Author Contributions: Study concept and design: Petersen, may not be able to explain the total spectrum of find- Parisi, Knopman, and Smith. Acquisition of data: Petersen, ings. Approximately 50% of their total group was ex- Parisi, Dickson, Johnson, Knopman, Boeve, Jicha, Ivnik, cluded for having non-AD abnormalities. They also com- Smith, Tangalos, and Kokmen. Analysis and interpreta- mented that the clinical classification system used in the tion of data: Petersen, Parisi, Dickson, Knopman, Boeve, Nun Study may have underestimated the degree of im- Jicha, and Braak. Drafting of the manuscript: Petersen, pairment in their patients, who were slightly more im- Parisi, and Johnson. Critical revision of the manuscript for paired than the patients with aMCI in the present study. important intellectual content: Petersen, Parisi, Dickson, The Religious Orders Study found that 44% of their Knopman, Boeve, Jicha, Ivnik, Smith, Tangalos, Braak, patients with MCI had a low likelihood of AD patho- and Kokmen. Statistical analysis: Petersen. Obtained fund- logic abnormalities, and another 44% had an intermedi- ing: Petersen and Kokmen. Administrative, technical, and ate likelihood according NIA-Reagan criteria.13 How- material support: Petersen, Parisi, Dickson, Johnson, Boeve, ever, the MCI diagnosis in the Religious Orders Study Jicha, Ivnik, Tangalos, and Braak. Study supervision: was different from the current aMCI criteria and al- Petersen, Parisi, and Knopman. lowed for impairment in multiple cognitive domains and Funding/Support: This work was funded by the may have included patients who were more impaired than National Institute on Aging, grant U01 AG06786 from the the present group. More recently, data from the Reli- Mayo Alzheimer’s Disease Patient Registry, and grant P50 gious Orders Study have documented an intermediate level AG16574 from the Mayo Alzheimer’s Disease of AD and vascular abnormalities in their patients with Research Center and the Robert H. and Clarice Smith and MCI using a broader definition of multiple-domain MCI.11 Abigail van Buren Alzheimer’s Disease Research Program. A study from Washington University describing the Acknowledgment: We thank Donna Asleson and Dorla neuropathologic features of patients with very mild AD Burton for their expert secretarial assistance in the prepa- (Clinical Dementia Rating of 0.5) demonstrated that 84% ration of this manuscript and our other colleagues in the of these patients had neuropathologic features of AD.43 Alzheimer’s Disease Research Center, in particular, Ruth This study does not necessarily contradict the findings Cha, MS, for statistical analysis.

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 671

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 22. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical REFERENCES diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA work group under the auspices of Department of Health and Human Services Task Force on 1. Petersen RC, Doody R, Kurz A, et al. Current concepts in mild cognitive impairment. Alzheimer’s Disease. Neurology. 1984;34:939-944. Arch Neurol. 2001;58:1985-1992. 23. Knopman DS, Parisi JE, Salviati A, et al. Neuropathology of cognitively normal 2. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cog- elderly. J Neuropathol Exp Neurol. 2003;62:1087-1095. nitive impairment: clinical characterization and outcome. Arch Neurol. 1999; 24. Mirra SS, Heyman MA, McKeil D, Sumi SM, Crain BJ. The Consortium to Establish 56:303-308. a Registry for Alzheimer’s Disease (CERAD), part II: standardization of the neuro- 3. Geda YE, Petersen RC. Clinical trials in mild cognitive impairment. In: Gauthier pathologic assessment of Alzheimer’s disease. Neurology. 1991;41:479-486. S, Cummings JL, eds. Alzheimer’s Disease and Related Disorders Annual 2001. 25. McKeith IG, Galasko D, Kosaka K, et al. Consensus guidelines for the clinical and London, England: Martin Dunitz; 2001:69-83. pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consor- 4. Knopman DS. Treatment of mild cognitive impairment and prospects for preven- tium on DLB international workshop. Neurology. 1996;47:1113-1124. tion of Alzheimer’s disease. In: Petersen RC, ed. Mild Cognitive Impairment: Aging 26. Duyckaerts C, Bennecib M, Grignon Y, et al. Modeling the relation between neu- to Alzheimer’s Disease. New York, NY: Oxford University Press Inc; 2003:243-258. rofibrillary tangles and intellectual status. Neurobiol Aging. 1997;18: 5. Petersen RC, Thomas RG, Grundman M, et al. Donepezil and vitamin E in the 267-273. treatment of mild cognitive impairment. N Engl J Med. 2005;352:2379-2388. 27. Dickson DW, Crystal HA, Mattiace LA, et al. Identification of normal and patho- 6. Petersen RC, Stevens JC, Ganguli M, Tangalos EG, Cummings JL, DeKosky ST. logical aging in prospectively studied nondemented elderly humans. Neurobiol Practice parameter: early detection of dementia: mild cognitive impairment (an Aging. 1992;13:179-189. evidence-based review): report of the Quality Standards Subcommittee of the 28. Mirra SS, Hart MN, Terry RD. Making the diagnosis of Alzheimer’s disease: American Academy of Neurology. Neurology. 2001;56:1133-1142. a primer for practicing pathologists. Arch Pathol Lab Med. 1993;117:132-144. 7. Bozoki A, Giordani B, Heidebrink JL, Berent S, Foster NL. Mild cognitive impair- 29. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta ments predict dementia in nondemented elderly patients with memory loss. Arch Neuropathol (Berl). 1991;82:239-259. Neurol. 2001;58:411-416. 30. Knopman DS, Parisi JE, Boeve BF, et al. Vascular dementia in a population- 8. Bennett DA, Wilson RS, Schneider JA, et al. Natural history of mild cognitive im- based autopsy study. Arch Neurol. 2003;60:569-575. pairment in older persons. Neurology. 2002;59:198-205. 31. Braak H, Braak E. Cortical and subcortical argyrophilic grains characterize a dis- 9. Larrieu S, Letenneur L, Orgogozo JM, et al. Incidence and outcome of mild cog- ease associated with adult onset dementia. Neuropathol Appl Neurobiol. 1989; nitive impairment in a population-based prospective cohort. Neurology. 2002; 15:13-26. 59:1594-1599. 32. Khachaturian ZS. Diagnosis of Alzheimer’s disease. Arch Neurol. 1985;42:1097-1105. 10. Petersen RC. Conceptual overview. In: Petersen RC, ed. Mild Cognitive Impair- ment: Aging to Alzheimer’s Disease. New York, NY: Oxford University Press Inc; 33. Ronald and Nancy Reagan Research Institute of the Alzheimer’s Association and 2003:1-14. the National Institute of Aging Working Group. Consensus report of the Work- 11. Bennett DA, Schneider JA, Bienias JL, Evans DA, Wilson RS. Mild cognitive im- ing Group on: “Molecular and Biochemical Markers of Alzheimer’s Disease.” Neu- pairment is related to Alzheimer disease pathology and cerebral infarctions. robiol Aging. 1998;19:109-116. Neurology. 2005;64:834-841. 34. Ivnik RJ, Malec JF, Tangalos EG, et al. The Auditory-Verbal Learning Test (AVLT): 12. Braak H, del Tredici K, Braak E. Spectrum of pathology. In: Petersen RC, ed. Mild norms for ages 55 years and older. Psychol Assess. 1990;3:304-312. Cognitive Impairment: Aging to Alzheimer’s Disease. New York, NY: Oxford Uni- 35. Tolnay M, Spillantini MG, Goedert M. Argyrophilic grain disease: widespread hy- versity Press Inc; 2003:149-189. perphosphorylation of tau protein in limbic neurons. Acta Neuropathol (Berl). 13. DeKosky ST, Ikonomovic M, Styren S, et al. Upregulation of choline acetyltrans- 1997;93:477-484. ferase activity in hippocampus and frontal cortex of elderly subjects with mild 36. Fujino Y, Wang DS, Thomas N, Espinoza M, Davies P, Dickson DW. Increased cognitive impairment. Ann Neurol. 2002;51:145-155. frequency of argyrophilic grain disease in Alzheimer disease with 4R tau- 14. Gomez-Isla T, Hyman BT. Neuropathological changes in normal aging, minimal specific immunohistochemistry. J Neuropathol Exp Neurol. 2005;64:209-214. cognitive impairment, and Alzheimer’s disease. In: Petersen RC, ed. Mild Cog- 37. Dickson DW, Davies P, Bevona C, et al. Hippocampal sclerosis: a common patho- nitive Impairment: Aging to Alzheimer’s Disease. New York, NY: Oxford Univer- logical feature of dementia in very old (Ն80 years of age) humans. Acta Neuro- sity Press Inc; 2003:191-204. pathol (Berl). 1994;88:212-221. 15. Guillozet AL, Weintraub S, Mash DC, Mesulam MM. Neurofibrillary tangles, amy- 38. Lippa CF, Dickson DW. Hippocampal sclerosis dementia: expanding the pheno- loid, and memory in aging and mild cognitive impairment. Arch Neurol. 2003; types of frontotemporal dementias? Neurology. 2004;63:414-415. 60:729-736. 39. Saito Y, Ruberu NN, Sawabe M, et al. Staging of argyrophilic grains: an age- 16. Mitchell TW, Mufson EJ, Schneider JA, et al. Parahippocampal tau pathology in associated tauopathy. J Neuropathol Exp Neurol. 2004;63:911-918. healthy aging, mild cognitive impairment and early Alzheimer’s disease. Ann Neurol. 40. Leverenz JB, Agustin CM, Tsuang D, et al. Clinical and neuropathological char- 2002;51:182-189. acteristics of hippocampal sclerosis: a community-based study. Arch Neurol. 2002; 17. Mufson EJ, Chen EY, Cochran EJ, Beckett LA, Bennett DA, Kordower JH. Ento- 59:1099-1106. rhinal cortex ␤-amyloid load in individuals with mild cognitive impairment. Exp 41. DeCarli C. Mild cognitive impairment: prevalence, prognosis, etiology, and Neurol. 1999;158:469-490. treatment. Lancet Neurol. 2003;2:15-21. 18. Petersen RC, Kokmen E, Tangalos E, Ivnik RJ, Kurland LT. Mayo Clinic Alzhei- 42. Riley KP, Snowdon DA, Markesbery MD. Alzheimer’s neurofibrillary pathology mer’s Disease Patient Registry. Aging. 1990;2:408-415. and the spectrum of cognitive function: findings from the Nun Study. Ann Neurol. 19. Weintraub S, Baratz R, Mesulam MM. Daily living activities in the assessment of 2002;51:567-577. dementia: Alzheimer’s disease: a report of progress. Aging. 1982;19:189-192. 43. Morris JC, Storandt M, Miller JP, et al. Mild cognitive impairment represents early- 20. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. stage Alzheimer’s disease. Arch Neurol. 2001;58:397-405. Neurology. 1993;43:2412-2414. 44. Jicha GA, Parisi JE, Dickson DW, et al. Neuropathological outcome of mild cog- 21. American Psychiatric Association. Diagnostic and Statistical Manual of Mental nitive impairment following progression to clinical dementia. Arch Neurol. 2006; Disorders, Third Edition. Washington, DC: American Psychiatric Association; 1980. 63:674-681.

(REPRINTED) ARCH NEUROL / VOL 63, MAY 2006 WWW.ARCHNEUROL.COM 672