International Journal of Clinical Medicine, 2012, 3, 352-360
http://dx.doi.org/10.4236/ijcm.2012.35068 Published Online September 2012 (http://www.SciRP.org/journal/ijcm) 1
Cognitive Profiles and Subtypes of Patients with Mild
Cognitive Impairment: Data from a Clinical Follow-Up
Study*
Kyung Won Park1#, Eun-Joo Kim2, Hwan Joo3, Sung-Man Jeon4, Seong-Ho Choi5, Jay C. Kwon6,
Byoung Gwon Kim7, Jae Woo Kim1
1Department of Neurology, Dong-A University College of Medicine, Dong-A University Medical Center, Busan, South Korea;
2Department of Neurology, Pusan National University Hospital, Pusan National University School of Medicine and Medical
Research Institute, Busan, South Korea; 3Department of Neurology, Busan Medical Center, Busan, South Korea; 4Department of
Neurology, Bong Seng Memorial Hospital, Busan, South Korea; 5Department of Neurology, Wallace Memorial Baptist Hospital,
Busan, South Korea; 6Department of Neurology, Changwon Fatima Hospital, Changwon, South Korea; 7Department of Preventive
Medicine, Dong-A University College of Medicine, Busan, South Korea.
Received May 30th, 2012; revised June 30th, 2012; accepted July 17th, 2012
ABSTRACT
Background: Mild cognitive impairment (MCI) is a heterogeneous condition with a variety of clinical outcomes, the
presence of which correlates with risk of Alzheimer’s disease as well as pre-clinical stages of other dementia subtypes.
The aims of this study were to assess the specific patterns of cognitive profiles an d to identify changes from baseline to
24 weeks in patients with MCI using detailed neuropsychological testing. Methods: We consecutively recruited 120
MCI patients at baseline according to the Petersen’s clinical diagno stic criteria, who were admitted to the Dementia and
Memory Clinics. We analyzed patients who fulfilled both in clusion and exclusion criteria for MCI and classified them
into four subtypes according to deficits in major cognitive domains; amnestic MCI single domain (aMCI-s), amnestic
multiple domain MCI (aMCI-m), non-amnestic single domain MCI (naMCI-s) and non-amnestic multiple domain MCI
(naMCI-m). Four groups of MCI were evaluated by a detailed neuropsychological battery test. Results: 83 patients with
MCI at the 24-week follow-up were classified into four subtypes. The most frequent subtype was amnestic multi-do-
main MCI, with the frequency of MCI subtypes as follows: aMCI-s (n = 21, 25.3%), aMCI-m (n = 53, 63.9%), naMCI-s
(n = 5, 6.0%) and naMCI-m (n = 4, 4.8%). In the major cognitive items of the SNSB-D, there were significant changes
between the initial and follow-up tests in the domains of language, memory and the frontal/executive function (p <
0.05), except for attention, in all MCI patient subtypes. At 24-weeks follow-up, the conversion rate to Alzheimer’s dis-
ease was 2.4% (n = 2) from a subtype of amnestic multi-domain MCI. Conclusions: Our study revealed the most fre-
quent subtype of MCI to be multiple domain amnestic MCI, with this subtype having a higher tendency of conversion
to Alzheimer’s disease.
Keywords: Mild Cognitive Impairment; Alzheimer’s Dementia; Neuropsycholog y; Conversion
1. Introduction
The recent increase of an elderly population in developed
countries like South Korea has led to a renewed focus on
degenerative disorders such as dementia. MCI refers to a
mild regression of cognitive functions, in particular
memory impairment in patients compared to that of nor-
mal people, although the condition is not severe enough
to be classified as dementia due to retainment of active-
ties of daily living, or ADL [1]. According to some epi-
demiological studies, people with MCI are in a high risk
group that may progress to Alzheimer’s disease [2-4].
Whereas 1 to 2 percent of normal control groups develop
dementia annually, 10 to 15 percent of MCI patients de-
velop dementia, in particular, Alzheimer’s disease [5].
This condition is clinically important because it is the
earliest stage that Alzheimer’s disease can be detected
and the effects of treatment could therefore be maxi-
mized. MCI is a syndrome involving heterogeneous
clinical manifestations and diverse causative diseases.
The discovery that MCI patients were likely to develop
dementia led to the recognition that early diagnosis of
MCI patients was crucial.
*The authors d eclare no conflicts of interest.
#Corresponding author. Furthermore, the classification of MCI patients ac-

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study 353
cording to manifestations of cognitive dysfunction re-
sulted in a renewed focus to verify which groups experi-
ence a high incidence of Alzheimer’s disease among dif-
ferent types of MCI patients [6-8]. Structural or func-
tional brain imaging techniques, such as Magnetic Reso-
nance Imaging (MRI), Positron Emission Tomography
(PET) and Single-Photon Emission Computed Tomo-
graphy (SPECT), were heavily employed for this re-
search, with efforts to verify differences in progression
patterns in accordance with the prognoses or types of
MCI [6].
A classification of MCI patient subtypes along with
detailed neuropsychological tests, in addition to brain
imaging techniques, while observing changes in cogni-
tive patterns and prognoses can provide important clues
in determining whether to apply pharmacotherapy in a
practical clinic setting and make long-term follow-up
observations. In a report presenting the international
standards for, and classification of MCI [6] suggested
that memory disorder was not the only symptom of MCI
patients, and that diverse cognitive disabilities may be
present, such as visuospatial and language impairment, as
well as frontal lobe dysfunction even from an early state
of the disease. Therefore, they noted that MCI may be
classified into the subtypes of amnestic MCI and non-
amnestic MCI, according to whether patients have mem-
ory impairment or not. They may also be classified into
single and multiple domain disorders, according to the
presence of a failing single cognitive do main or of multi-
ple domains. In order to classify patients according to the
definitions of the four subtypes of MCI, a working crite-
rion in consideration of age and educational back-
ground were used, with neuropsychological test tools
representing each cognitive area. Based on the neuro-
psychological performances, MCI divided into the sub-
types: amnestic single domain MCI (aMCI-s), when the
patient lacked on the memory function; amnestic multi-
ple domain MCI (aMCI-m), when there were impair-
ments on several cognitive areas, including memory;
non-amnestic single domain MCI (naMCI-s), presence of
an impairment in another cognitive area, with normal
memory; and non-amnestic multiple domain MCI (naMCI-
m), with impairments in more than one cognitive domain,
with normal mem ory.
The frequency and pattern of progression of each sub-
type of MCI were inconsistent due to the presence of
diverse diagnostic criteria, as well as different sampling
and assessment methods. The authors performed this
study with the following objectives: First, to classify new
patients, who visited hospitals from September 2008 to
April 2009 that were participating in the study and who
were diagnosed with MCI into subtypes (amnestic single
domain, amnestic multiple domain, non-amnestic single
domain, and non-amnestic multiple domain), and inves-
tigate each subtype’s frequency and patterns. Second,
after six months elapsed, we observed whether the MCI
subtypes had changed and whether they had progressed
into dementia.
2. Subjects and Methods
This was a multi-center, observational study conducted
for 24 weeks across 13 hospitals located in Busan Met-
ropolitan City, Gyeongnam Province from September
2008 to February 2010. We performed this study with
full approval from the respective Institutional Review
Boards of each hospital, after inspection of the clinical
trial plan, written explanation for subjects, and after ob-
taining each subjects’ informed consent.
2.1. Subjects
We enrolled patients according to working criteria based
on the clinical diagnostic criteria for MCI [5,6]: 1) the
presence of subjective memory complaint as reported by
participants or informants; 2) intact ability to perform
activities of normal daily living; 3) normal general cog-
nitive function defined as cognitive performance above
the range of 1.0 standard deviation (SD) of normative
data in an extensive neuropsychological test; 4) abnor-
mal cognitive func tion including memory domain for ag e
and education documented by performance of at least 1.0
SD below mean normative data in cognitive tasks; 5)
non-demented according to DSM-IV criteria and ex-
cluded by fulfilling criteria (2) and (3). The participants
included in the present study were 55 to 85 years of age,
had not taken acetylcholinesterase inhibitors, memantine
or ginkgo bilboa for at least 4 weeks prior to the study
initiation, were ambulatory or ambulatory-aided (i.e.,
walker or cane), had brain MRI or CT scans revealing no
clinical evidence of other diseases capable of producing a
dementia syndrome, and had a reliable caregiver who
met the patient at least once a week and provide the in-
vestigator with accurate information. The following ex-
clusion criteria were adopted: 1) major depressive dis-
order, bipolar disorder, schizophrenia, substance use dis-
order, or mental retardation according to criteria of the
DSM-IV; 2) cerebrovascular disord ers, hydrocephalus or
intracranial mass, documented by CT or MRI within the
past 12 months; 3) abnormalities in serum folate and vi-
tamin B12, syphilis serology, or thyroid hormone levels;
4) history of traumatic brain injury or other neurologic
disease; and 5) significant medical problems (e.g. poorly
controlled diabetes or hypertension; cancer within the
past 5 years; clinically significant hepatic, renal, cardiac
or pulmonary disorders).
2.2. Neuropsychological Tests
The patients’ general cognitive functions were measured

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study
354
with the Korean version of the mini-mental state exam
(K-MMSE) [9]. The severity of dementia was evaluated
with a clinical dementia rating (CDR) [10] and the scores
in the six areas of CDR were combined to calculate the
sum of boxes (SOB) in each item. Overall severity was
expressed with global CDR. All patients underwent neu-
ropsychological tests using a standardized neuropsy-
chological battery called the Seoul Neuropsychological
Screening Battery—Dementia Version (SNSB-D) [11 ,12 ].
This screening battery contains tests for attention, lan-
guage, praxis, parietal function, visuoconstructive func-
tion, verbal and visual memory, and frontal executive
function. The neuropsychological tests were: digit span
(forward and backward), the Korean version of the Bos-
ton Naming Test (K-BNT) [13], ideomotor praxis, Rey-
Osterrieth Complex Figure Test (RCFT; copying, imme-
diate and 20-min delayed recall and recognition), the
Seoul Verbal Learning Test (SVLT; three learning-free
recall trials of 12 words, 20-min delayed recall trial for
these 12 items and a recognition test), phonemic and se-
mantic Controlled Oral Word Association Test (COWAT)
and the Stroop test (word an d color reading of 112 items
during a 2-min period). Age-, gender- and education-
specific norms for each test based on 447 norma l subjects
are available. The scores of these cognitive tests were
classified as abnormal when they were below the 16th
percentile of the norms for respective age-, gender- and
education-matched normal sub jects.
2.3. Classification of MCI Patients According to
Subtypes
In order to classify patients according to the definitions
of the four subtypes of MCI, a working criteria in con-
sideration of age and educational background were used,
with neuropsychological test tools representing each
cognitive area that was specified by the standard estab-
lished through data research. An experienced neuropsy-
chologist performed the neuropsychological tests on all
patients, and based on the results, divided them into the
subtypes: 1) amnestic single domain MCI (aMCI-s),
when the patient lacked disabilit y in other co gn itive areas,
except for degraded memory; 2) amnestic multiple do-
main MCI (aMCI-m), when there were disabilities in
other cognitive areas, including memory; 3) non-amnes-
tic single domain MCI (naMCI-s), presence of a disabil-
ity in another cogn itive area, with normal memory; and 4)
non-amnestic multiple domain MCI (naMCI-m), with
disabilities in more than one cogn itive domain, with nor-
mal memory.
2.4. Statistical Analysis
STATA/SE 11.2 (Stata Corp. 2009, College Station, TX,
USA) was used for all statistical analyses, and a two-
tailed test was performed with the level of significance
set at 0.05.
A frequency analysis was done of the subtypes, and all
data collected through the neuropsychological tests were
analyzed using descriptive statistics. Mean values of con-
tinuous variables of the data were compared with a
paired t-test and non-continuous variables were com-
pared with a chi-square test. When the P value was less
than 0.05, a difference was considered to be statistically
significant. Based on the findings from the neuropsy-
chological tests performed again six months later and the
clinicians’ own judgments, we calculated the frequency
of patients whose MCI progressed to dementia and cal-
culated the rate of conversion of MCI to dementia in
each subtype. We ascertained the frequency of each sub-
type of MCI and the number of MCI patients in each
subtype who had changed since the initial diagnosis. In
order to verify which cognitive domain underwent the
most and least changes, the scores of each of the four
cognitive domains (memory, visuospatial ability, linguis-
tic ability and frontal lobe function) were measured, and
the scores in the beginning and in the 24th week were
compared.
3. Results
3.1. Frequency of Subtypes of MCI at Baseline
Among the 120 subjects of this study, the follow-up tests
of a total of 83 patients were completed. In total, 23 pa-
tients dropped out (follow-up loss), 9 patients withdrew
their consent to participate in the study, 2 patients were
relocated to other hospitals, 1 patient died, 1 patient was
excluded due to use of a banned medication and 1 patient
was excluded due to the onset of cerebral infarction, all
during the follow-up test period. The most common sub-
type among MCI patients was found to be amnestic mul-
tiple domain MCI (aMCI-m). Each subtype’s frequency
was as follows: aMCI-s is 25.3% (n = 21), aMCI-m is
63.9% (n = 53), naMCI-s is 6.0% (n = 5), and naMCI-m
is 4.8% (n = 4) (Table 1).
The rate of male to female patients among the subjects,
their average age, and their number of years of education
were 34.2% to 65.8%, 68.5 ± 7.5 years, and 7.0 ± 4.3
years, respectively (Table 1).
3.2. Changes of Neuropsychological Test Results
in Each Subtype at 24 Weeks Follow-Up
The total score and follow-up sco re changes of the Seoul
Neuropsychological Screening Battery-Dementia Ver-
sion (SNSB-D) in each subtype are shown in detail in
Tables 2-5. After the 24 weeks follow-up period, all
MCI patients showed sign ificant improvements in cogni-
tive function on the total score of SNSB-D compared
with the baseline assessment (P < 0.05). Furthermore,

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study
355
Table 1. Demographic paramters and frequency in patients with MCI according to subtypes.
Total MCI patients
(n = 120) Amnestic single MCI patients (n = 29) Amnestic multiple MCI patients (n = 74) Non-amnestic single MCI patients (n = 11) Non-amnestic multiple MCI patients (n = 6)P value
Female, n (%) 79 (65.8) 22 (75.9) 45 (60.8) 7 (63.6) 5 (83.3) 0.397
Age, years 68.5 ± 7.5 65.4 ± 7.8 69.0 ± 7.2 69.2 ± 6.9 75.0 ± 5.9 0.015
Education, years 7.0 ± 4.3 6.9 ± 3.8 7.0 ± 4.7 7.5 ± 3.3 6.5 ± 4.0 0.909
BMI 24.1 ± 3.0 24.1 ± 2.7 24.1 ± 3.3 24.3 ± 2.5 23.7 ± 2.9 0.973
Hypertension, n (%) 42 (35.0) 10 (23.8) 22 (52.4) 6 (14.3) 4 (9.5) 0.145
Diabetes mellitus, n (%) 23 (19.1) 5 (21.7) 14 (60.9) 2 (8.7) 2 (8.7) 0.842
Hyperlipidemia, n (%) 13 (10.0) 4 (30. 7) 7 (53.9) 0 (0.0) 2 (15.4) 0.180
Heart disease, n (%) 19 (15.8) 7 (36. 8) 9 (47.4) 1 (5.3) 2 (10.5) 0.261
Alcohol drinking, n (%) 3(2.5) 0 (0.0) 3 (4.1) 0 (0.0) 0 (0.0) 0.431
Smoking, n (%) 29 (24.4) 4 (13.8) 22 (31.1) 2 (18.1) 1 (16.7) 0.557
Depression, n (%) 4 (3.3) 2 (6.0) 2 (2.0) 0 (0.0) 0 (0.0) 0.611
Family history of
dementia, n (%) 23 (19.3) 3 (10.3) 14 (19.2) 4 (36.4) 2 (33.3) 0.230
ApoE4 genotype
(n = 22), positive, n 8 3 3 1 1 NS
Values a re presented as nu mber (%) and mean ± SD. MCI, mild cognitive impairment; BMI, body mass index.
Table 2. Mean changes in cognitive function on neuropsychological tests in patients with all subtypes of MCI.
Outcome measure Baseline mean (SD) (n = 83) 24-week F/U mean (SD) (n = 83)P value
MMSE 26.3 (2.3) 26.4 (2.4) 0.4238
Attention 9.1 (1.9) 9.4 (2.1) 0.1796
Forward 5.8 (1.5) 6.0 (1.4) 0.0494
Backward 3.4 (0.9) 3.3 (1.1) 0.7313
Language & related func ti on 20.0 (3.8) 20.8 (3.8) 0.0012
Naming (K-BNT) 10.4 (2.9) 10.9 (2.7) 0.0279
Calculation 9.6 (2.4) 9.9 (2.4) 0.0221
Visuospatial function 27.6 (8.0) 29.1 (6.9) 0.0394
Rey figure copy 27.6 (8.0) 29.1 (8.0) 0.0394
Memory 53.8 (17.4) 61.9 (21.5) <0.001
Orientation 5.6 (0.6) 5.6 (0.7) 0.7650
Verbal immediate recall 15.8 (5.7) 16.9 (6. 1 ) 0.0319
Verbal delayed reca l l 3.5 (2.4) 5.1 (2.8) 0.0000
Verbal recognition index 7.4 (2.1) 7.6 (2.6) 0.5153
Visual immediate / d e l a yed recall 15.2 (11.7) 20.2 (13.8) <0.001
Visual recognition 6.2 (2.4) 6.5 (2.5) 0.2782
Frontal/executive 44.4 (10.2) 46.3 (9.8) 0.0039
Impersistence 3.0 (0.0) 3.0 (0.0) N/A
Contrasting program 2.6 (0.7) 2.7 (0.5) 0.1316
Go-no-go test 2.0 (1.0) 2.3 (0.8) 0.0011
Fist-edge-palm 2.7 (0.5) 2.7 (0.5) 0.8587
Luria loop 2.6 (0.9) 2.8 (0.7) 0.0356
Word fluency-animal 13.1 (3.9) 12.8 (3.6) 0.3324
Word fluency- ph one mic 5.6 (3.9) 5.9 (3.4) 0.4209
Stroop test-color 12.8 (4.5) 14.0 (4.7) 0.0010
SNSB-D 154.9 (32.6) 167.4 (35.9) <0.001
CDR sum of box 1.3 (0.8) 1.2 (0.8) 0.0357
IADL 3.5 (3.6) 2.7 (3.8) 0.0362
NPI 4.0 (8.3) 3.8 (8.0) 0.7711
Geriatric dep ression scale 16.9 (6.8) 15.6 (7.0) 0.0555
MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; SNSB-D, Seoul Neuropsychological Screening Battery-Dementia Version; K-BNT,
Korean Boston Naming Te st; CDR, Clinical Dementia Rating; IADL, Instrumental Activities of Daily L iving; NPI, Neuropsychiatric inventory.

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study
356
Table 3. Mean changes in cognitive function on neuropsychological tests in patients with aMCI-s.
aMCI-s (n = 21)
Outcome measure Baseline mean (SD) 24-week F/U mean (SD) P value
MMSE 26.9 (1.7) 27.2 (1.7) 0.4127
Attention 9.8 (2.4) 10.2 (2.2) 0.4123
Forward 6.1 (1.7) 6.4 (1.4) 0.4361
Backward 3.7 (0.9) 3.8 (1.3) 0.7152
Language & related func ti on 21.0 (3.4) 21.7 (3.5) 0.0784
Naming (K-BNT) 11.1 (2.1) 11.5 (2.5) 0.3081
Calculation 9.8 (2.5) 10.1 (2.2) 0.0571
Visuospatial function 31.8 (4.8) 31.4 (4. 5) 0.8171
Rey figure copy 31.8 (4.8) 31.5 (4.5) 0.8171
emory 57.8 (22.2) 67.9 (22.0) 0.0063
Orientation 5.7 (0.6) 5.6 (0.7) 0.5402
Verbal immediate recall 14.9 (5.7) 16.8 (6.80.) 0.1290
Verbal delayed reca l l 3.2 (2.5) 5.3 (2.8) 0.0022
Verbal recognition index 7.1 (1.9 7.9 (2.3) 0.0725
Visual immediate/d el a yed recall 20.2 (15.9) 25.1 (13.7) 0.0539
Visual recognition rey 6.6 (2.2) 7.2 (2.4) 0.3000
Frontal/executive 5 0.7 (9.9) 51.4 (9.3) 0.5190
Impersistence 3.0 (0.0) 3.0 (0.0) -
Contrasting program 2.8 (0.5) 2.9 (0.4) 0.7477
Go-no-go test 2.0 (0.9) 2.5 (0.7) 0.0466
Fist-edge-palm 2.8 (0.5) 2.8 (0.5) 1.0000
Luria loop 2.7 (0.7) 3.0 (0.0) 0.0829
Word fluency-animal 14.2 (4.5) 14.0 (3.8) 0.6799
Word fluency- ph one mic 6.4 (3.8) 6.7 (3.8) 0.6708
Stroop test-color 16.7 (3.4) 16.7 (4.0) 0.9157
SNSB-D 171.0 (33.8) 152.7 (32.6) 0.0127
CDR sum of box 1.1 ( 0.6) 1.0 (0.6) 0.0829
IADL 2.8 (3.9) 1.6 (2.7) 0.0146
NPI 4.0 (10.9) 4.0 (12.1) 0.9114
Geriatric dep ression scale 19.0 (5.9) 16.4 (8.6) 0.0494
aMCI-s, single domain amnestic mild cognitive impairment; MMSE, Mini-Mental State Examination; SNSB-D, Seoul Neuropsychological Screening Bat-
tery-Dementia Version; K-BNT, Korean Boston Naming Test; CDR, Clinical Dementia Rating; IADL, Instrumental Activities of Daily Living; NPI, Neuro-
psychiatric inventory.
patients with each subtype of MCI showed significant
improvement on the total score of SNSB-D compared
with the baseline assessment (P < 0.05). The mean dif-
ferences of total scores of SNSB-D in patients with MCI
were as follows; entire MCI (12.5 ± 18.7), aMCI-s
(11.7 ± 19.6), aMCI-m (12.6 ± 19.9), naMCI-s (13.0 ±
9.7), naMCI-m (15.3 ± 5.3). The other changes in the
cognitive outcome measures are shown in detail in Ta-
bles 2-5.
3.3. Progression of Subjects with MCI
In the follow-up test after 24 weeks, two MCI patients
(2.4%) progressed to Alzheimer’s disease; while their
subtype during the initial test was aMCI-m. No patient
progressed to a type of dementia other than Alzheimer’s
disease. The number of MCI patients who underwent
conversion to a normal condition was five (6%) and their
MCI subtypes were aMCI-s (1 person), aMCI-m (3 per-
sons) and naMCI-s (1 person). After 24 weeks, the fre-
quencies of each subtype were as follows: aMCi-s was
25.5%, aMCI-m was 42.1%, naMCI-s was 16.9%, and
naMCI-m was 7.2%.
4. Discussion
Previous research on MCI has revealed that it not only
consists of the amnestic type, but rather a range of dif-

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study 357
Table 4. Mean changes in cognitive function on neuropsychological tests in patients with aMCI-m.
aMCI-m (n = 53)
Outcome measure Baseline mean (SD) 24-week F/U mean (SD) P value
MMSE 25.9 (2.4) 26.0 (2.7) 0.6770
Attention 9.0 (1.8) 9.1 (2.0) 0.3503
Forward 5.7 (1.4) 6.0 (1.4) 0.0795
Backward 3.2 (0.9) 3.1 (1.1) 0.5603
Language & related func ti on 19.3 (3.9) 20.1 (3.9) 0.0159
Naming (K-BNT) 9.9 (3.2) 10.4 (2.7) 0.0751
Calculation 9.4 (2.6) 9.6 (2.5) 0.1593
Visuospatial function 25.9 (8.9) 28.0 (7.9) 0.0287
Rey figure copy 25.9 (8.9) 28.0 (7.8) 0.0287
Memory 50.5 (15.1) 57.7 (21.2) 0.0004
Orientation 5.5 (0.6) 5.6 (0.8) 0.3742
Verbal immediate recall 16.0 (5.6) 16.5 (5.6) 0.3862
Verbal delayed reca l l 3.4 (2.4) 4.9 (2.8) 0.0000
Verbal recognition index 7.3 (2.3) 7.2 (2.7) 0.6062
Visual immediate / d e l a yed recall 12 .2 (9.3) 17.4 (13.6) 0.0009
Visual recognition rey 6.0 (2.6) 6.2 (2.5) 0.6641
Frontal/executive 41.9 (9.6) 44.1 (9.6) 0.0126
Impersistence 3.0 (0.0) 3.0 (0.0) -
Contrasting program 2.5 (0.8) 2.7 (0.5) 0.1066
Go-no-go test 1.9 (1.0) 2.2 (0.9) 0.0177
Fist-edge-palm 2.7 (0.5) 2.8 (0.5) 0.6590
Luria loop 2.4 (1.0) 2.7 (0.9) 0.0963
Word fluency-animal 12.4 (3.6) 12.3 (3.6) 0.7112
Word fluency- ph one mic 5.4 (3.8) 5.5 (3.1) 0.7719
Stroop test-color 11.5 (4.2) 13.0 (4.9) 0.0053
SNSB-D 146.5 (31.0) 159.0 (36.9) 0.0000
CDR sum of box 1.5 (0.9) 1.3 (0.8) 0.1212
IADL 3.8 (3.7) 3.3 (4.3) 0.3476
NPI 4.5 (7.8) 4. 0 (6.4) 0.5156
Geriatric dep ression scale 16.1 (7.1) 1537 (6.4) 0.6306
aMCI-m, multiple domain amnestic mild cognitive impairment; MMSE, Mini-Mental State Examination; SNSB-D, Seoul Neuropsychological Screening Bat-
tery-Dementia Version; K-BNT, Korean Boston Naming Test; CDR, Clinical Dementia Rating; IADL, Instrumental Activities of Daily Living; NPI, Neuro-
psychiatric inventory.
ferent subtypes. As proposed by Winbald et al. [6], MCI
can be classified into four different types, according to
disabilities in domains identified by neuropsychological
tests. We consider that each subtype has different causes
and prognoses. Recently published research has reported
a classification of MCI patients into subtypes according
to neuropsychological tests with the observation of each
subtype’s clinical manifestations and prognoses. This
resulted in the finding that MCI was not a pre-stage of
Alzheimer’s disease, but rather consisted of a group of
heterogeneous diseases with various kinds of prognoses
[1,14]. According to further research, aMCI-m is likely
to progress to Alzheimer’s disease, while naMCI-s or
naMCI-m are likely to progress to vascular dementia,
frontotemporal dementia or dementia with Lewy bodies
[1,15].
To our knowledge, this is the first regional study of its
kind performing neuropsychological assessments on 120
MCI patients, calculating the frequency of each subtype
of MCI, observing changes in each subtype’s frequency
after a 24-week follow-up test and analysis of the
changes in the cognitive fun ctions of each typ e of patient.
We found that the most common subtype of MCI was
aMCI-m (63.9%), followed by aMCI-s (25.3%),
naMCI-s ( 6.0%), and na MCI-m (4.8%). Our resul ts d if fer
from that of a prior study (Fischer et al., 2007), where the

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study
358
Table 5. Mean changes in cognitive function on neuropsychological tests in patients with naMCI.
naMCI ( n = 9)
Outcome measure Baseline mean (SD) 24-week F/U mean (SD) P value
MMSE 26.8 (2.5) 26.9 (1.8) 0.8695
Attention
Forward 5.1 (1.3) 5.4 (1.7) 0.4714
Backward 3.5 (0.5) 3.4 (0.5) 0.3466
Language & rela ted function
Naming (K-BNT) 11.3 (2.4) 11.7 (2.6) 0.3466
Calculation 10.7 (1.3) 11.1 (1.2) 0.1950
Visuospatial function
Rey figure copy 28.0 (4.4) 29.8 (3.4) 0.3321
Memory
Orientation 5.7 (0.5) 5.6 (0.5) 0.5943
Verbal immediate recall 16.8 (6.1) 19.3 (7.4) 0.0171
Verbal delayed reca l l 5.0 (1.7) 6.1 (2.8) 0.0619
Verbal recognition index 8.6 (1.1) 9.3 (1.3) 0.1411
Visual immediate / d e l a yed recall 21 .6 (6.6) 25.6 (10.3) 0.1577
Visual recognition rey 6.4 (2.1) 6.8 (1.8) 0.6454
Frontal/executive
Contrasting program 2.6 (1.0) 2.7 (0.7) 0.7995
Go-no-go test 2.3 (1.0) 2.6 (0.7) 0.3466
Fist-edge-palm 2.7 (0.7) 2.6 (0.7) 0.3466
Luria loop 3.0 (0.0) 2.9 (0.3) 0.3466
Word fluency-animal 14.8 (3.5) 13.2 (3.3) 0.2021
Word fluency- ph one mic 4.9 (4.4) 6.0 (4.2) 0.1786
Stroop test-color 11.1 (3.4) 14.0 (2.8) 0.0064
SNSB-D 167.2 (22.1) 181.2 (21.2) 0.0006
CDR sum of box 0.8 (0.3) 0.8 (0.4) 0.5943
IADL 3.2 (2.0) 1.9 (1.8) 0.2572
NPI 0.3 (1.0) 2. 2 (3.5) 0.1276
Geriatric dep ression scale 17.0 (7.0) 13.4 (7.0) 0.1354
naMCI, non-amnestic mild cognitive impairment; F/U, follow-up; MMSE, Mini-Mental State Examination; SNSB-D, Seoul Neuropsychological Screening
Battery-Dementia Version; K-BNT, Korean Boston Naming Test; CDR, Clinical Dementia Rating; IADL, Instrumental Activities of Daily Living; NPI, Neu-
ropsychiatric inventory.
frequency of subtypes of MCI patients were 48 out of
141 (34%) patients with aMCI and 93 out of 141 (66%)
patients with naMCI. The frequency of MCI subtypes in
further studies showed that the prevalence of each sub-
type of MCI in research, where the subjects were from
the general population, was different to that of the cur-
rent study [16,17]. They reported that the nonamnestic
MCI type was as frequent as the amnestic MCI type,
while the single domain amnestic MCI was more preva-
lent than multiple domain amnestic MCI. Another study
reported that the prevalence of aMCI and naMCI was
around 9% and 15%, respectively. The overall rate of
naMCI was even higher than that of aMCI in the com-
munity-based study [18].
One of the main reasons for the different rates in the
frequency of MCIs may be due to the difference between
hospital-based studies and community-based epidemiol-
ogical studies. The other reason for differing rates of
frequency in MCI subtypes is a dissimilar adaptation of
operational criteria for defining MCI. Our study defined
memory impairment in the Seoul Verbal Learning Test
(SVLT) as 1 standard deviation (SD) below age- and
education-matched control subjects, unlike other studies
[5,8], therefore a larger number of patients were classi-
fied into the aMCI subtype. It is also possible that cul-
tural differences between Korea and other countries af-
fected the scores; for example in the United States, visu-
ospatial impairment displayed during driving, or disabili-

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study 359
ties in planning or decision making, may have been con-
sidered important in addition to memory impairment.
Furthermore, this study included cases where patients’
guardians or informants reported memory impairment
even without subjective memory complaints from the
patients, which resulted in a higher rate of MCI cases
relative to those of other studies.
During the 24 weeks follow-up period, all patients of
MCI with each subtype of MCI showed significant im-
provements in cognitive fun ction on the to tal score of the
SNSB-D compared with the baseline assessment. In the
major cognitive items of the SNSB-D, there was signifi-
cant change between the initial and the follow-up as-
sessments in the domains of language, memory and the
frontal/executive function in all subtypes of MCI. This
result is probably due to a relatively large number of pa-
tients with high MMSE scores and mild stag e of patients
being included, and learning effect may have resulted
during the relatively short follow-up period. The condi-
tion of depression may also have had an influence on the
test scores, given that the patients with high depression
scores in the initial tests sh owed improvement in the fol-
low-up tests.
Our study showed that two MCI patients (2.4%) pro-
gressed to Alzheimer’s disease; their subtype during the
initial test was aMCI-m. No patient progressed to other
types of dementia other than Alzheimer’s disease. A
3-year follow-up research study reported that rates of
conversion to Alzheimer’s disease for the MCI subtypes
were 38% for amnestic MCI , 20% for non-amnestic MCI
and 16% for amnestic multiple domain MCI [7]. The
other long-term follow-up study revealed a conversion
rate to AD at 49% for amnestic MCI and 27% for non-
amnestic MCI [8]. Our study did not show in-depth con-
version rates due to a shorter follow-up period. However,
our results show a tendency for progression to AD from
amnestic MCI rather than non-amnestic MCI.
We identified some shortcomings in our study. First,
although it was a prospective follow-up study, we did
observe changes in the MCI patients’ subtypes and their
cognitive functions during the short period of 24 weeks,
therefore making it difficult to form a precise evaluation
of the patients’ disease progression or an estimation of
the annual rate of conversion from MCI to dementia.
Second, the number of patients in each subtype was un-
evenly distributed, making it difficult to determine true
statistical significance. Also, differences in levels of
apolipoprotein E4 were not obtained from all patients,
nor the analysis of differences in MCI in accordance with
the existence or non-existence of cerebrovascular lesions,
such as white matter changes or lacun ar infarction.
In conclusion, we were able to verify that the SNSB-D
is a useful tool to classify MCI subtypes and follow their
progression in detail, suggesting that the routine classifi-
cation of MCI into subtypes and observation of progres-
sion may be conducive to predicting a transition of MCI
to Alzheimer’s disease or other types of dementia. In our
study, aMCI-m was the most common subtype among
the four subtypes of MCI, and its rate of conversion into
Alzheimer’s disease was statistically significant. For a
more comprehensive investigation, it will be necessary to
perform further prospective multi-center research for a
period of a few years, with many hundreds of patients as
subjects, in region-based cohort studies as well as hospi-
tal-based cohort studies.
5. Acknowledgements
This study was supported by a grant from the Korea
Healthcare Technology R&D Project, Ministry of Health
and Welfare, Republic of Korea (A102065) and Eisai
Korea Inc.
REFERENCES
[1] R. C. Petersen, R. Doody, A. Kurz, R. C. Mohs, J. C.
Morris, P. V. Rabins, et al., “Current Concepts in Mild
Cognitive Impairment,” Archives of Neurology, Vol. 58,
No. 12, 2001, pp. 1985-1992.
doi:10.1001/archneur.58.12.1985
[2] H. Amieva, L. Letenneur, J. F. Dartigues, I. Rouch-
Leroyer, C. Sourgen, F. D’Alchee-Biree, et al., “Annual
Rate and Predictors of Conversion to Dementia in Sub-
jects Presenting Mild Cognitive Impairment Criteria De-
fined According to a Population-Based Study,” Dementia
and Geriatric Cognitive Disorders, Vol. 18, No. 1, 2004,
pp. 87-93. doi:10.1159/000077815
[3] P. Alexopoulos, T. Grimmer, R. Perneczky, G. Domes
and A. Kurz, “Progression to Dementia in Clinical Sub-
types of Mild Cognitive Impairment,” Dementia and
Geriatric Cognitive Disorders, Vol. 22, No. 1, 2006, pp.
27-34. doi:10.1159/000093101
[4] The Canadian Study of Health and Aging Working Group,
“The Incidence of Dementia in Canada,” Neurology, Vol.
55, No. 1, 2000, pp. 66-73. doi:10.1212/WNL.55.1.66
[5] R. C. Petersen, G. E. Smith, S. C. Waring, R. J. Ivnik, E.
G. Tangalos and E. Kokmen, “Mild Cognitive Impair-
ment: Clinical Characterization and Outcome,” Archives
of Neurology, Vol. 56, No. 3, 1999, pp. 303-308.
doi:10.1001/archneur. 56.3.303
[6] B. Winblad, K. Palmer, M. Kivipelto, V. Jelic, L. Fratig-
lioni, L. O. Wahlund, et al., “Mild Cognitive Impair-
ment—Beyond Controversies, Towards a Consensus:
Report of the International Working Group on Mild Cog-
nitive Impairment,” Journal of Internal Medicine, Vol.
256, No. 3, 2004, pp. 240-246.
doi:10.1111/j.1365-2796.2004.01380.x
[7] F. Maioli, M. Coveri, P. Pagni, C. Chiandetti, C. Mar-
chetti, R. Ciarrocchi, et al., “Conversion of Mild Cogni-
tive Impairment to Dementia in Elderly Subjects: A Pre-
liminary Study in a Memory and Cognitive Disorder
Unit,” Archives of Gerontology and Geriatrics, Vol. 44,

Cognitive Profiles and Subtypes of Patients with Mild Cognitive Impairment: Data from a Clinical Follow-Up Study
360
No. 1, 2007, pp. 233-241.
doi:10.1016/j.archger.2007.01.032
[8] P. Fischer, S. Jungwirth, S. Ze hetmay er, S. We issgram, S.
Hoenigschnabl, E. Gelpi, et al., “Conversion from Sub-
types of Mild Cognitive Impairment to Alzheimer De-
mentia,” Neurology, Vol. 68, No. 4, 2007, pp. 288-291.
doi:10.1212/01.wnl.0000252358.03285.9d
[9] Y. Kang, D. L. Na and S. Hahn, “A Validity Study on the
Korean Mini-Mental State Examination(K-MMSE) in
Dementia Patients,” Journal of the Korean Neurological
Association, Vol. 15, 1997, pp. 300-308.
[10] J. C. Morris, “The Clinical Dementia Rating (CDR):
Current Version and Scoring Rules,” Neurology, Vol. 43,
No. 11, 1993, pp. 2412-2414.
doi:10.1212/WNL.43.11.2412-a
[11] Y. Kang and D. L. Na, “Seoul Neuropsychological Scr-
eening Battery,” Human Brain Research & Consulting
Co., Seoul, 2003.
[12] H. J. Ahn, J. Chin, A. Park, B. H. Lee, M. K. Suh, S. W.
Seo, et al., “Seoul Neuropsychological Screening Bat-
tery-Dementia Version (SNSB-D): A Useful Tool for
Assessing and Monitoring Cognitive Impairments in De-
mentia Patients,” Journal of Korean Medical Science, Vol.
25, No. 7, 2010, pp. 1071-1076.
doi:10.3346/jkms.2010.25.7.1071
[13] H. Kim and D. L. Na, “Normative Data on the Korean
Version of the Boston Naming Test,” Journal of Clinical
and Experimental Neuropsychology, Vol. 21, No. 1, 1999,
pp. 127-133. doi:10.1076/jcen.21.1.127.942
[14] J. T. Tschanz, K. A. Welsh-Bohmer, C. G. Lyketsos, C.
Corcoran, R. C. Green, K. Hayden, et al., “Conversion to
Dementia from Mild Cognitive Disorder: The Cache
County Study,” Neurology, Vol. 67, No. 2, 2006, pp. 229-
234. doi:10.1212/01.wnl.0000224748.48011.84
[15] M. Zanetti, C. Ballabio, C. Abbate, C. Cutaia, C. Vergani
and L. Bergamaschini, “Mild Cognitive Impairment Sub-
types and Vascular Dementia in Community-Dwelling
Elderly People: A 3-Year Follow-Up Study,” Journal of
the American Geriatrics Society, Vol. 54, No. 4, 2006, pp.
580-586. doi:10.1111/j.1532-5415.2006.00658.x
[16] A. Busse, A. Hensel, U. Guhne, M. C. Angermeyer and S.
G. Riedel-Heller, “Mild Cognitive Impairment: Long-
Term Course of Four Clinical Subtypes,” Neurology, Vol.
67, No. 12, 2006, pp. 2176-2185.
doi:10.1212/01.wnl.0000249117.23318.e1
[17] J. J. Manly, M. X. Tang, N. Schupf, Y. Stern, J. P. Von-
sattel and R. Mayeux, “Frequency and Course of Mild
Cognitive Impairment in a Multiethnic Community,”
Annals of Neurology, Vol. 63, No. 4, 2008, pp. 494-506.
doi:10.1002/ana.21326
[18] S. Jungwirth, S. Weissgram, S. Zehetmayer, K. H. Tragl
and P. Fischer, “VITA: Subtypes of Mild Cognitive Im-
pairment in a Community-Based Cohort at the Age of 75
Years,” International Journal of Geriatric Psychiatry,
Vol. 20, No. 5, 2005, pp. 452-458. doi:10.1002/gps.1311