

with percentages obtained for the group of individuals originally
classified as normal by the POMA. When we used a sample that
contained only individuals classified as normal by MMSE during the
1st analysis, only the POMAwas able to identify a worsening of results
both in the average scores and in the percentage of individuals
classified as disabled in the MMSE in the 3rd and 4th evaluations in the
group classified as disabled by POMA in the 1st analysis.
Conclusion:
Particularly, the MMSE, as expected, seems to be the best
predictor of performance on the same scale after 4 and 6 years.
However, when only individuals classified as normal by the MMSE are
considered, which excludes, by definition, MMSE as a factor of the
analysis, the POMA becomes the most appropriate scale to predict
reductions in cognitive performance as indicated by the MMSE. These
results are further evidence that losses in motor activity can be useful
in early identification of increased risk of cognitive decline.
P-146
The association of sex-hormone binding globulin (SHBG) with
executive function in 70-year old community-dwellers: the PIVUS
study
A. Ticinesi
1
, F. Lauretani
1
, G.P. Ceda
1
, S. Masoni
1
, R. Aloe
2
, L. Lind
3
,
T. Cederholm
4
, T. Meschi
1
, M. Maggio
1
.
1
Department of Clinical and
Experimental Medicine, University of Parma,
2
Laboratory of Clinical
Chemistry and Hematology, University Hospital of Parma, Parma, Italy;
3
Department of Medical Sciences, Uppsala University,
4
Department of
Public Health and Caring Sciences, Uppsala University, Uppsala, Sweden
Introduction:
Changes in executive function are an early marker of
dementia. The metabolic determinants of executive function have
however not been fully elucidated. Sex-Hormone Binding-Globulin
(SHBG) is notmerelya transporter but also acts as ametabolicmediator.
Thus, our aim was to investigate the association between SHBG
levels and Trail-Making Test (TMT), a measure of executive function,
in older individuals.
Methods:
This cross-sectional analysis involved 534 seventy-year-old
community-dwellers (284 males) from the Prospective Vasculature in
Uppsala Seniors (PIVUS) Study. Univariate and multivariate regression
models were built to assess the association between SHBG and TMT
performance. Sex, comorbidities, smoking habits, BMI, total choles-
terol, fasting insulin, total testosterone, estradiol, Mini-Mental State
Examination (MMSE) and Endothelium-Independent Vasodilation
(EIDV) were considered as potential confounders.
Results:
Median SHBG levels were 47.3 nmol/L (IQR 35.4
–
60.5), while
TMT-A and TMT-B scores were 53 (IQR 43
–
67) and 130 (IQR 101
–
188),
respectively. TMT-A score was positively associated with EIDV, fasting
insulin, diabetes and smoking and negatively with MMSE, but was
not significantly associated with SHBG (
β
± SE 0.04 ± 0.04, unadjusted
p = 0.31). Conversely, TMT-B was positively associated with fasting
insulin, stroke and diabetes, and negatively associated with MMSE,
EIDV and SHBG (
β
± SE
−
0.59 ± 0.22, unadjusted p = 0.007). The
association between TMT-B and SHBG was confirmed in a multivariate
model considering sex and MMSE (
−
0.51 ± 0.21, p = 0.02), but not in a
fully-adjusted model including EIDV (
−
0.35 ± 0.20, p = 0.08).
Key conclusions:
SHBG levels are positively and independently asso-
ciated with executive functions, measured through TMT-B, in older
individuals. This association is however not independent of EIDV.
P-147
Integrating information from FGD-PET and amyloid-PET for
differentiating types of dementia in older persons with mild
cognitive symptoms: a case-series
F. Lauretani
1
, L. Ruffini
2
, M. Scarlattei
2
, A. Ticinesi
1
, T. Meschi
1
,
C. Ghetti
3
, G. Serreli
2
, P. Caffarra
4
, M. Maggio
1
.
1
Department of Clinical
and Experimental Medicine, University of Parma,
2
Nuclear Medicine Unit,
Parma University Hospital,
3
Medical Physics Unit, Parma University
Hospital,
4
Department of Neuroscience, University of Parma, Parma, Italy
Introduction:
The role of flourodeoxyglucose (FDG) and amyloid
positron-emission tomography (PET) in the differential diagnosis of
dementia is debated. However,
β
-amyloid load and diffuse glucose
hypometabolism in parietal, temporal and frontal lobes are typical of
Alzheimer
’
s disease (AD). Our aim was to assess the possible role
of both FDG- and amyloid-PET in establishing a differential diagnosis
for dementia in older patients with symptoms of mild cognitive
impairment (MCI).
Methods:
Three consecutive patients aged
≥
70 with mild memory or
language complaints (corrected MMSE 18
–
24), no other significant
comorbidity and undiagnostic cerebral CT or MRI were enrolled at the
Center for Cognitive Disorders of Parma University-Hospital. All of
themunderwent 18F-FDG and amyloid PET using awhole-body hybrid
system operating in three-dimensional detection mode on two
different days, following standardized protocols for image acquisition
and interpretation.
Results:
In one patient, FDG-PET showed frontal, parietal and temporal
lobe atrophy, with a significant presence of amyloid plaques at
amyloid-PET, allowing to diagnose probable AD. In another patient
with cerebral MRI showing small vascular lesions in peri-ventricular
and peri-trigonal areas, positive FDG-PET and negative amyloid-PET
allowed to suppose that vascular dementia was present. In the
third patient, amyloid-PET was negative, but FDG-PET showed
significant hypometabolism in the frontal lobe, allowing to establish
the diagnosis of probable frontotemporal dementia.
Key conclusions:
In patients with signs of MCI and undiagnostic first-
level brain imaging, the prescription of FDG-PET and amyloid-PET for
differential diagnosis of dementia type should be considered. Further
studies should validate the diagnostic algorithm proposed here.
P-148
Combined training of motor coordination and working memory
exercises is an effective tool to improve cognitive functions in the
elderly
N. Trushkova-Zhuravleva
1
, G. Zelano
2
.
1
Rehabilitation Research Center
“
Armonia
”
, Latina,
2
Department of Neuroscience, Catholic University,
Rome, Italy
Objectives:
Aging is associated with decline in cognitive functions
such as memory, attention, executive functions, and processing speed.
Many studies have focused on different cognitive training strategies to
improve cognitive functions in healthy older adults. Numerous studies
have provided evidence that physical activity promotes cortical
plasticity in the adult brain and in turn facilitates learning. There are
studies that showed that simultaneous physical activity during
vocabulary learning facilitates memorization of new items. The
purpose of our study is to investigate the effect of combined training
of motor coordination and working memory exercises focused on
novelty in healthy adults.
Methods:
We established combined physical and cognitive traning
that offers a great variety of exercises and does not aim at automating
but focuses on novelty. The study subjects were 42 in healthy adults,16
men and 26 women, aged
≥
76 years, cognitive status screen (MMSE
>24). We proposed to them to perform an hour of activity twice a
week for 24 weeks. At the beginning and at the end of treatment,
participants received a cognitive examination that included Mini-
Mental State Examination (MMSE), Forward and Backward Digit Span
Test, Rey
’
s Auditory Verbal Learning test, TMT-A, TMT-B.
Results:
All participants showed improvement in all tests, the most
significant improvement was in the Mini-Mental State Examination
(MMSE) (mean = 1,7 p < 0.001) Statistical analyses were performed by
t-test.
Conclusion:
Our findings suggest that combined training of motor
coordination and working memory exercises has proven to be effective
to improve cognitive functions in healthy adults.
Poster presentations / European Geriatric Medicine 7S1 (2016) S29
–
S259
S67