Content uploaded by Nídia Celeste Horie
Author content
All content in this area was uploaded by Nídia Celeste Horie on Jan 02, 2016
Content may be subject to copyright.
Cognitive effects of intentional weight loss in elderly
obese individuals with mild cognitive impairment
Nidia Celeste Horie
1
, Valeria T Serrao
2
, Sharon Sanz Simon
3
,
Maria Rita Polo Gascon
2
, Alessandra Xavier dos Santos
4
,
Maria Aquimara Zambone
4
, Marta Merenciana del Bigio de Freitas
5
,
Edecio Cunha-Neto
6
, Emerson Leonildo Marques
1
, Alfredo Halpern
1
,
Maria Edna de Melo
1
, Marcio C Mancini
1
, Cintia Cercato
1
1- Obesity and Metabolic Syndrome Group- Sao Paulo University, School of Medicine; 2- Psychology
Division- Hospital das Clínicas, Sao Paulo University; 3- Institute of Psychiatry - Sao Paulo University,
School of Medicine; 4- Nutrition Division- Clinical Hospital - Sao Paulo University, School of Medicine; 5-
Discipline of Geriatrics - Sao Paulo University, School of Medicine; 6- Division of Clinical Immunology and
Allergy- Sao Paulo University, School of Medicine
Context: Obesity in midlife is a risk factor for dementia, but it is unknown if caloric restriction-
induced weight loss could prevent cognitive decline and therefore dementia in elderly patients
with cognitive impairment.
Objective: To evaluate the cognitive effect of intentional weight loss in obese elderly patients with
mild cognitive impairment (MCI), considering the influence of age, apolipoprotein E genotype
(APOE), physical activity, biochemical markers and diet.
Design: Single-center, prospective controlled trial.
Setting: Academic medical center
Participants: Eighty obese patients with MCI, aged 60 or older (68.1⫾4.9 years, body mass index
(BMI) 35.5⫾4.4kg/m
2
, 83.7% women, 26.3% APOE4 carriers).
Intervention: Random allocation to conventional medical care alone (n⫽40) or together with
nutritional counselling (n⫽40) in group meetings aiming to promote weight loss through caloric
restriction for 12 months.
Outcome: Measurements: Clinical data, body composition, neuropsychological tests (main out-
come), serum biomarkers, APOE genotype, physical performance, dietary recalls.
Results: Seventy-five patients completed the follow-up. BMI, on average, decreased 1.7 ⫾1.8kg/m
2
(p⫽0.021), and most of the cognitive tests improved, without difference between the groups. In
analysis with linear generalized models, the BMI decrease was associated with improvements in
verbal memory, verbal fluency, executive function and global cognition, after adjustment for
education, gender, physical activity and baseline tests: this association was strongest in younger
seniors (for memory and fluency) and in APOE4 carriers (for executive function). Changes in HOMA-
IR, C-reactive protein, leptin and intake of energy, carbohydrates and fats were associated with
improvement in cognitive tests.
Conclusions: Intentional weight loss through diet was associated with cognitive improvement in
patients with MCI.
ISSN Print 0021-972X ISSN Online 1945-7197
Printed in USA
Copyright © 2015 by the Endocrine Society
Received May 17, 2015. Accepted December 11, 2015.
Abbreviations:
ORIGINAL ARTICLE
doi: 10.1210/jc.2015-2315 J Clin Endocrinol Metab press.endocrine.org/journal/jcem 1
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
Dementia is a syndrome characterized mainly by dete-
rioration in cognitive function and is one of the ma-
jor causes of disability and dependency among older peo-
ple. Nearly50 milion people worldwide have dementia,
and it is caused by a variety of conditions that affect the
brain, of which the most common is Alzheimer’s Disease
(1). Studies of secondary prevention have been performed
with subjects with mild cognitive impairment (MCI), who
have increased risk of dementia, but few strategies, such as
physical activity, have been shown to reduce cognitive
decline in prospective studies (2).
According to several epidemiological studies, obesity in
midlife increases the risk of dementia later in life (3, 4), and
insulin resistance, dyslipidemia, hypertension, low grade
inflammation and leptin resistance are associated with
both obesity and cognitive decline. Caloric restriction is
one of the mainstays of obesity treatment, and its neuro-
protective effect has largely been demonstrated in both
animal models (5) and also in cognitively normal humans
(6, 7). In patients with cognitive impairment, however, its
effects have not yet been tested. Research on the effects of
a hypocaloric diet in the elderly faces major obstacles:
late-life obesity has been associated with decreased de-
mentia risk (8), and many studies show that weight loss is
associated with increased mortality and disability, espe-
cially in subjects over 70 years of age (9). These risks,
however, seem to be more important in underweight or
normal-weight subjects than in obese (10) ones.
We hypothesized that intentional weight loss via caloric
restriction in elderly obese subjects with MCI could slow
the cognitive decline. We also evaluated the influence of
other risk factors for dementia, such as age, presence of the
apolipoprotein E allele
⑀
4 (APOE4 carriers or APOE4
noncarriers), metabolic and inflammatory parameters,
and diet. Considering the risks of lean mass and strength
loss following weight loss, the safety of the intervention
was verified by measuring changes in lean mass and phys-
ical functioning.
Materials and Methods
This study was approved by the ethics committee of the institu-
tion and was conducted in adherence with the Helsinki Decla-
ration; the trial is registered with clinicaltrials.gov: NCT
01 286 389. Informed consent was obtained from all partici-
pants for being included in the study. A complete description of
the inclusion and exclusion criteria, participants selection, mea-
surements, and statistical analysis is in the Supplementary
Material.
Trial design
This was a prospective, 1:1, randomized study to assess the
cognitive effects of weight loss via caloric restriction in obese
subjects with MCI according to the European Consortium on
Alzheimer’s Disease (11), comparing groups who received con-
ventional care with or without nutritional counseling for 12
months (Figure 1). It was performed from June 2011 to May
2013 at the outpatient clinic of the of Hospital das Clinicas, São
Paulo University.
All patients were advised to engage in physical activity ac-
cording “The global recommendations on physical activity for
health” from the World Health Organization (12), briefly, they
should do at least 150 minutes of moderate-intensity aerobic
physical activity or walking throughout the week, or if limited
due to health conditions, they should be as physically active as
their abilities and conditions allow. All patients received con-
ventional medical care which was provided through consultation
with a geriatrician approximately every two months an which
focused on control of comorbidities. Half of the patients received
additionally nutritional counseling in groups conducted by nu-
tritionists (26 to 28 one-hour meetings held over the course of 12
months) that aimed to promote healthy eating habits and weight
loss through caloric restriction. The group meetings included
advice on eating a diet rich in fiber, fruits, vegetables, and whole
grains, and included at least 1g/kg of weight of protein per day,
with a recommended calorie deficit of approximately 500 kcal
per day (or to a minimum of 1200 kcal/d), and the meetings also
included lectures on food composition, meal preparation, eating
habits and self-monitoring techniques.
Participants and measurements
Eighty subjects with MCI (13), body mass index (BMI) ⱖ30
kg/m2, age ⱖ60 years, without conditions that interfere with
weight loss or cognition (eg, major depression [evaluated by the
module “Major Depressive Episode” of “Mini International
Neuropsychiatric Interview” (14)], hypothyroidism, heart fail-
ure, cancer, alcoholism, infectious diseases and auto-immune
activity; use of antiobesity drugs, benzodiazepines, neuroleptics
or estrogen replacement therapy in the past 2 months) were
selected.
Socio-demographic, anthropometric and clinical data were
recorded. Biochemical analysis included glucose, insulin, gly-
cated hemoglobin (HbA1c), HOMA-IR (homeostasis model as-
sessment-estimated insulin resistance), lipid profile, leptin, adi-
ponectin, interleukin 6 (IL6), tumor necrosis factor (TNF) alpha
(TNF
␣
), C-reactive protein (CRP) and apolipoprotein E (APOE)
genotype. As we had a small number of patients for some geno-
types, the patients were classified as APOE4 carriers (homozy-
gote or heterozygote for APOE4) or noncarriers. Physical per-
formance was measured with the Short Physical Performance
Battery (15) (SPPB). Food intake was estimated through 24-hour
diet recall collected by trained nutritionists. The calculations for
the energy and macronutrient intake were performed using the
Avanutri 4.0 software (16). As a reference table for nutritional
composition of foods was adopted the Brazilian Table of Food
Composition (TACO) (17) and when food was not listed in the
table, was used as a reference table of the United States Depart-
ment of Agriculture - USDA (18). We recorded the level of phys-
ical activity using the IPAQ (ie, the International Physical Ac-
tivity questionnaire [IPAQ] short version (19)) and classified the
patients as active or sedentary (active ⱖ150 minutes physical
activity/wk). All these measures, except genotyping, were re-
peated after 12 months.
The neuropsychological battery included measures of pre-
2Cognition and weight loss in obese elderly J Clin Endocrinol Metab
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
morbid Intelligence Quotient (IQ estimated by Vocabulary and
Matrix reasoning (20) tests), verbal memory (Rey Auditory Ver-
bal Learning Test- RAVLT (21); RAVLT-A7⫽delayed recall;
RAVLT兺A1A5⫽total learning; RAVLT A6⫽recall; RAVLT
recognition), attention [Digit Span forward (DF) and backwards
(DB) (22), Trail Making Test part A (TMA)], working memory
[DB, Trail Making Test part B (TMB)], psychomotor processing
speed (TMA, TMB (23)), executive function (Modified Wiscon-
sin Card Sorting Test – 48 cards version- MWCST (24),TMB
and verbal fluency), language [phonemic verbal fluency (25)
(measured with words starting with FAS) and semantic verbal
fluency (measured with number of animals)] and cognitive com-
plaints (Informant Questionnaire on Cognitive Decline in the
Elderly - IQCODE (26)). Two tests of global cognition were used
at the screening, Montreal Cognitive Assessment (MoCA) (27,
28) and CAMCog (29) (cognitive session of CAMDEX). All of
these tests except MoCA and IQ were repeated after 12 months.
Analysis
The statistical analyses were performed using the Statistical
Package for Social Science software version 20.0 (SPSS, Chicago,
Figure 1. Study flow chart from recruitment until 12-month follow-up.
doi: 10.1210/jc.2015-2315 press.endocrine.org/journal/jcem 3
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
IL). A p-value ⱕ0.05 was considered significant and all tests
were two-sided. Bivariate analysis was performed with two-sam-
ple t-tests or Mann-Whitney test for continuous variables and
2
test or Fisher’s test for categorical variables. The Generalized
Estimating Equation was used in the longitudinal analysis. The
independent variable group shows the main effect of intervention
group, the time variable shows the effect of time between pretest
and follow-up, while the interaction term group x time can be
interpreted as the effect of the intervention (Table 1, Supplemen-
tary material). As both groups lost similar amounts of weight
after the intervention period (mean delta BMI intensive group ⫽
–2.1 (-4.8; 0.7) kg/m
2
, conventional group ⫽–1.3 (-4.1; 1.4)
kg/m
2
,P⫽.428, Table 1, Supplementary material), we decided
to a posteriori pool the longitudinal analysis between groups,
and to test for the effects of change of BMI as a continuous
variable.
To evaluate the relationship between BMI change and vari-
ation in cognitive domains, considering the effect of covariates,
the regression with the generalized linear model with normal
distribution and identity link function was used. Cognitive tests
were grouped according to the following domains on the basis of
a generally accepted description: global cognition (CAMCOG
and IQCODE), verbal memory (RAVLT兺A1A5, A6, A7, rec-
ognition), executive ability (TMA, TMB, MWCST, digits back-
wards), language/fluency (phonemic verbal fluency and semantic
verbal fluency) and attention (digits forward and backwards).
With scores from 2 time points, principal-components analysis
with varimax rotation and Kaiser normalization was performed
on these domain-specific test groups to generate single compo-
nents for each domain. Standardized regression factor scores
were then generated from these components by Bartlett´s
method. The difference (⌬) between the cognitive tests results at
12 month follow up, subtracted from the baseline score, was
considered the dependent variable. The variables ⌬BMI, gender,
APOE4, physical activity at 12 months, age, education and base-
line test score were all considered predictors of change in cog-
nitive status. Age and APOE4 genotype (carrier) are two impor-
tant risk factors for dementia, and had been described as
influence for weight loss; therefore, the interaction between
weight loss and cognitive change was also verified. The result was
presented with tests of the main effects for age, APOE4 and
⌬BMI and the interaction between a risk factor and ⌬BMI, with
values of

(indicating the linear fit), SE (standard error), and
significance.
Partial linear correlation analysis between changes in cogni-
tive tests and clinical and laboratory variables was performed
(the adjustments are described at the Supplementary material).
Table 1. Baseline characteristics by intervention group
Groups conventional intensive p
Characteristics mean (SD) mean (SD)
n (%) n (%)
Age (years) 68.3 (5.3) 67.9 (4.5) 0.688
Gender (female) 34 (85%) 33 (82.5%) 0762
Education (years) 8.3 (4.3) 9.4 (4.8) 0.277
IQ (intelligence quotient) 97,6 (11) 98 (10.6) 0.845
MoCA (0 to 30) 19.1 (3.5) 19.2 (3.1) 0.920
IADL (9 to 27) 25.8 (2.1) 25.6 (1.6) 0.156
BMI class 0.280
BMI ⬍35 (kg/m
2
) 23 (57.5%) 24 (60%)
BMI 35–39,9 (kg/m
2
) 13 (32.5%) 8 (20%)
BMI ⱖ40 (kg/m
2
) 4 (10%) 8 (20%)
Comorbidities
Charlson comorbidity index 1.3 (1.1) 1.5 (1.4) 0.538
Hypertension (%) 34 (85%) 28 (70%) 0.108
Diabetes (%) 18 (45%) 15 (37,5%) 0.496
Pre-diabetes
a
(%) 14 (35%) 16 (40%) 0.644
Metabolic syndrome (%) 35 (87,5%) 32 (80%) 0.546
Previous smoking
b
(%) 11 (27.5%) 16 (40%) 0.237
Physical function
Physically active (%) 23 (57.5%) 27 (67.5%) 0.356
SPPB (0–12) 10.5 (1.8) 10.5 (1.5) 0.694
Balance (0–4) 3.8 (0.6) 3.8 (0.7) 0.748
4 meters walk (sec) 4.7 (1.0) 4.8 (0.9) 0.644
Sit/get up (sec) 12.3 (5.0) 11.4 (2.7) 0.298
Genotype 0.834
⑀
4
⑀
4 n (%) 0 1 (2.5%)
⑀
3
⑀
4 n (%) 7 (17.5%) 7 (17.5%)
⑀
2
⑀
4 n (%) 3 (7.5%) 3 (7.5%)
⑀
3
⑀
3 n (%) 27 (67.5%) 26 (65%)
⑀
2
⑀
3 n (%) 3 (7.5%) 3 (7.5%)
APOE4 carriers (%) 10 (25%) 11 (27.5%)
MCI type 0.993
amnestic multiple domain 18 (45%) 19 (47.5%)
amnestic single domain 8 (20%) 7 (17.5%)
nonamnestic mult.domain 2 (5%) 2 (5%)
nonamnestic single domain 12 (30%) 12 (30%)
n⫽number; SD ⫽standart deviation; MoCA ⫽Montreal Cognitive Assessment; IADL: instrumental activities of daily life, IPAQ: international
physical activity questionnaire, SPPB: short physical performance battery, BMI: body mass índex; MCI: mild cognitive impairment; a: pre-diabetes:
fasting glucose between 100 and 125 mg/dL or impaired glucose tolerance; b: just one patient was current smoker.
4Cognition and weight loss in obese elderly J Clin Endocrinol Metab
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
Results
Baseline characteristics of the study participants can be
found in Table 1. There were 80 patients, mean age 68.1 ⫾
4.9 years, without premorbid IQ deficit, BMI between 30
kg/m
2
and 49.5 kg/m
2
. The estimated energy intake was
1517 ⫾515 kcal/d (17.3 kcal/kg weight). The average diet
composition was 50.2% carbohydrate, 29.1% lipid, and
20.7% protein. Twenty one patients (26.3%) were
APOE4 carriers. There were no significant differences be-
tween the groups at baseline for age, gender, education,
comorbidities, genotype, measurements or laboratory
analysis.
Five women did not complete the follow-up: four
dropped out and one died of pneumonia (Figure 1).
BMI decreased by an average of 1.7 ⫾1.8 kg/m
2
(P⫽
.021), without significant change of lean mass, and 35
(43.8%) patients had weight loss greater than 5% of initial
body weight. The proportion of physically active patients
did not change (initial 62.5%, final 70.7%, P⫽.282). The
SPPB, gait speed and time to sit/get up improved signifi-
cantly. There was improvement on most of the cognitive
tests (Table 2, supplementary material, Table 2).
The generalized estimating equation was used for com-
parisons between groups in the longitudinal analysis. The
complete results for body composition, biochemical
markers, physical function, and diet can be found in the
Supplementary Material – Table 1, and for cognitive tests
in the Supplementary Material- Table 2. There was no
significant time-group interaction for most of the tests af-
ter 12 months, except for time to walk 4 m [conventional
group: –0.2 seconds (CI95%: – 0.7; 0.3), intensive group:
–0.9 seconds (CI95%:-1.4;-0.4), P⫽.012].
Generalized linear models were used to analyze the re-
lationship between BMI change and cognitive change (Ta-
ble 3, Figure 2). Improvements in global cognition, verbal
memory, language and executive function were correlated
with a decrease in BMI. For verbal memory and language,
there was interaction between BMI change and age, mean-
ing that the effect of weight loss was more beneficial to
younger patients. For executive function, age had an in-
dependent effect (the older the patient, the worse the per-
formance), and there was interaction between APOE4 sta-
tus and BMI change: among APOE4 carriers, a decrease in
BMI was more beneficial. The change in attention did not
Table 2. Cognitive evaluation at baseline and variation after 12 months
Baseline Change
mean CI 95% mean CI 95% p
Global cognition
CAMCog (0–
107)
85.2 83.7, 86.8 2.7 1.7; 3.7 ⬍0.0001
IQCODE (1–5) 3.49 3.4, 3.56 ⫺0.28 ⫺0.38;
⫺0.18
⬍0.0001
Memory
RAVLT-A7(0–15) 6.6 6.0, 7.2 1.2 0.7, 1.8 ⬍0.0001
RAVLTSA1A5
(0–75)
38.5 36.8, 40.2 2.6 0.9; 4.3 0.002
Executive
function/
attention/
psychomotor
speed
Digits forward
(0–16)
6.7 6.4, 7.1 ⫺0.2 ⫺0.6, 0.2 0.357
Digits backward
(0–14)
4.2 3.8, 4.5 0.1 ⫺0.2, 0.5 0.371
Phonemic
fluency FAS
29.1 27.2, 30.9 0.9 ⫺0.4; 2.3 0.165
Semantic fluency 14.3 13.4, 15.2 1.5 0.5; 2.4 0.002
Wisconsin-
categories
2.9 2.6, 3.3 0.5 0.1; 0.9 0.017
TMA (sec) 65.1 59.3, 71.5 ⫺4.3 ⫺9.8; 1.1 0.119
TMB (sec) 185.1 163.4, 209.8 ⫺3.8 ⫺23.4; 15.9 0.709
CI95 ⫺95% confidence interval; difference between time (initial vs. final) tested by Generalized Estimation Equation. IQCODE: Informant
Questionnaire on Cognitive Decline in Elderly; CAMcog: cognitive session of the Cambridge Examination for Mental Disorders of the Elderly;
RAVLT-A7: Rey auditory verbal learning test - delayed recall; RAVLTSA1A5: Rey auditory verbal learning test - total learning; TMA: trail making test
part A; TMB: trail making test part B
doi: 10.1210/jc.2015-2315 press.endocrine.org/journal/jcem 5
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
correlate with BMI, age or APOE4 status (data not
shown). At figure 2, the correlation between BMI and
cognitive change (standardized scores) was illustrated,
and for didactical purposes the patients were separated in
2 age groups and 2 genotype groups.
A correlation analysis between changes in cognitive test
scores and clinical variables was performed (Table 4; Sup-
plementary material- Table 3 and 4). Leptin increase was
associated to attention improvement (DF). Reductions in
HOMA-IR correlated with improvements in global cog-
nition (CAMCog) and phonemic fluency. Decreased CRP
correlated with an increase in delayed memory. The de-
crease in caloric intake correlated with improvement in
verbal memory and executive function (TMB). The de-
crease in carbohydrate intake was associated with im-
provement in verbal memory, executive function (TMB)
and subjective complaints (IQCODE). Decreases in fat
intake were associated with improvements in verbal mem-
ory and TMB.
Table 3. Adjusted model for change in cognitive domains related to variation in BMI (body mass index), age,
APOE4 genotype (carrier or not) and interaction between BMI change and age (for language) or between BMI
change and APOE4 genotype (for executive function)
a
domains ⌬global cognition ⌬Verbal memory ⌬Language/ Fluency ⌬Executive
parameters

SE.
2p

SE.
2p

SE.
2p

SE.
2p
⌬BMI ⫺0.12 0.05 4.96 0.026 ⫺1.81 0.82 4.87 0.027 ⫺2.32 0.82 8.04 0.005 ⫺0.18 0.08 4.75 0.029
age (years) 0.00 0.02 0.00 .956 0.01 0.03 0.06 .811 0.01 0.02 0.09 .765 ⫺0.04 0.02 4.11 0.043
APOE4⫹0.13 0.21 0.39 .533 ⫺0.18 0.19 0.87 .352 ⫺0.01 0.19 0.00 .951 ⫺0.48 0.29 2.69 .101
⌬BMI* age 0.02 0.01 4.03 0.045 0.03 0.01 8.17 0.004
⌬BMI*E4⫹ ⫺0.20 0.10 3.88 0.049
Legend: BMI: body mass index. APOE4⫹: Carrier of apolipoprotein E4 genotype. SE.⫽standard error.
2
: Wald
-Square. a: Model adjusted for
age, the presence of APOE4, baseline testing, education (years), gender and level of physical activity; Global cognition: CAMcog (cognitive session
of CAMDEX) and IQCODE; ⌬Verbal memory: composite score of RAVLT SA1A5: Rey auditory verbal learning test - total learning; A6: recall, A7:
delayed recall, recognition; ⌬Executive: composite score of TMA: trail making test part A; TMB: trail making test part B, Wisconsin classification
cards and digits backwards; ⌬Language/ Fluency: composite score corresponding to phonemic and phonetic fluency. ⌬BMI*E4⫹: parameter of
interaction between
␦
BMI and APOE4 genotype; ⌬BMI* age: parameter of interaction between
␦
BMI and age in years.
Figure 2. Adjusted scatter plot showing the relation between BMI (kg/m
2
) change and cognitive change after 12 months intervention, by group
(aged below or above 70 years or APOE4 carrier or noncarrier): Legend: BMI: BMI. APOE4⫹: Carrier of apolipoprotein E4 genotype. Global
cognition: CAMcog (cognitive session of CAMDEX) and IQCODE; ⌬Verbal memory: composite score of RAVLT ⌺A1A5: Rey auditory verbal
learning test - total learning; A6: recall, A7: delayed recall, recognition; ⌬Executive: composite score of TMA: trail making test part A; TMB: trail
making test part B, WI classification cards and digits backwards; ⌬Language/Fluency: composite score corresponding to phonemic and phonetic
fluency. All the tests were adjusted to gender, education, baseline test and physical activity.
6Cognition and weight loss in obese elderly J Clin Endocrinol Metab
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
Discussion
As far as we know, this is the first clinical trial that inves-
tigated the effects of intentional weight loss via caloric
restriction in patients with MCI. Cognitive tests have
proven to be good predictors of the conversion of MCI to
dementia and are suitable to evaluate treatment response
(30); for this study we evaluated multiple cognitive do-
mains. Delayed recall is considered one of the main tests
for predicting the progression of cognitive impairment
(31). Worsened performances on the TMB have already
been associated with obesity (32), metabolic syndrome
and type 2 Diabetes (33), and it is considered useful for
predicting dementia (34). One previous study has demon-
strated its improvement associated with intentional
weight loss in patients without cognitive impairment (35).
In our study, for tests of verbal memory, language, exec-
utive function and global cognition, improved scores were
correlated with a decrease in BMI. Even considering that
it is a secondary analysis, it is interesting that it endorses
the main study´s hypothesis. For younger patients, the ef-
fect was more pronounced particularly to memory and
language. The concept of a better therapeutic window for
cognitive protection was already discussed at the context
of hormone therapy (HT) (36): apparently, starting some
hormone combinations nearby menopause is beneficial,
but starting many years after menopause increases demen-
tia risk. Other therapies, like monoclonal antibodies
against B amyloid (37), also had shown mixed results, and
one of the possible influences was considered the treat-
ment window: maybe after the amyloid plaques are wide
spread, treatments that target the amyloid are no longer
enough to prevent the development of dementia. It is pos-
sible that older patients, having a more advanced neuro-
pathology, were less responsive to metabolic changes and
then less responsive to our intervention.
Among APOE4 carriers, a decrease in BMI was asso-
ciated with executive improvements, showing greater ben-
efits in patients with greater dementia risk. The effect of
APOE polymorphism on treatment response has been
demonstrated in other studies; APOE4 carriers with de-
mentia appear to have a worse response to treatment than
noncarriers (38); yet in patients with MCI or mild demen-
tia, this effect varies: cognitive improvement with rosigli-
tazone was demonstrated only among APOE4 noncarriers
(39), and carriers showed better response with bapineu-
zumabe (40) and donepezil (41). A former study (42) re-
ported that APOE4 carriers who consumed a high-fat diet
had a greater risk of developing AD compared with non-
carriers, effect probably linked with the function of APOE
at the lipoprotein transport. In our study, APOE4 carriers
showed a slight better cognitive response to caloric re-
striction; although this pattern did not repeated in all cog-
nitive areas, this finding deserves attention in future
research.
The value of plasma leptin for predicting cognitive
change is uncertain, since a prior longitudinal study with
a community-based sample found that higher leptin levels
were associated with lower dementia risk (43), however a
study of elderly obese individuals found that higher leptin
levels were associated with brain atrophy (44). The use of
leptin in animal models has been associated with memory
improvement (45). In our group, an increase in leptin was
associated with greater improvement in attention (DF),
and in executive function (phonemic fluency) there was
the same trend. Analysis of the correlation between met-
abolic variables and changes in cognitive performance re-
vealed associations between a decrease in insulin resis-
tance and an improvement in cognition, and a decrease in
CRP and improvement in memory. Insulin resistance is
associated with increased risk of cognitive decline and the
Table 4. Partial linear correlation between change in cognitive tests and change in leptin, HOMA-IR, C-reactive
protein and diet, adjusted to age, gender, baseline test, education, presence of apolipoprotein
⑀
2or
⑀
4 allele and
physical activity level
⌬leptin ⌬HOMAIR ⌬CRP ⌬kcal ⌬carbo-hydrate ⌬lipid
⌬CAMCog ⫺0.085 -0.393** 0.002 ⫺0.154 ⫺0.110 ⫺0.113
⌬SA1A5 ⫺0.074 ⫺0.212 0.049 -0.346** -0.253*-0.318**
⌬A7 ⫺0.151 -0.238#-0.299*-0.303*-0.336** -0.312**
⌬Digits
Forward
0.383** ⫺0.143 0.212#⫺0.167 ⫺0.137 ⫺0.166
⌬Phonemic
fluency
0.227#-0.273*⫺0.038 ⫺0.184 0.027 -0.218#
⌬TMB
&
0.176 0.090 0.047 0.267*0.290*0.241*
⌬IQCODE
&
⫺0.004 ⫺0.039 ⫺0.056 0.216#0.315** 0.139
#P⬍0,1; * P⬍0,05; **P⬍0,01. &: greater the score, worst de performance, for the other tests, greater the score, better the performance;
CAMcog: cognitive session of CAMDEX; RAVLT SA1A5: Rey auditory verbal learning test - total learning; RAVLT-A7: Rey auditory verbal learning
test - delayed recall; TMB: trail making test part B; IQCODE: Informant Questionnaire on Cognitive Decline in Elderly. Leptin and dietary
parameters: adjusted also to initial body mass índex.
doi: 10.1210/jc.2015-2315 press.endocrine.org/journal/jcem 7
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
correlation between improvement in memory and a de-
crease in insulin resistance and CRP (6) has already been
described.
In a previous study, we found that weight loss after
bariatric surgery for severely obese women (mean age
40.5, mean BMI 51.1 kg/m
2
) was associated with a reverse
in hypermetabolism in the posterior cingulate gyrus and
an improvement in executive function (7). A previous
study with healthy subjects (6) (mean age 60.5) also dem-
onstrated an improvement in memory after a caloric re-
striction of only 11.6%. The decrease in reported caloric
intake of the patients in our study was only 8.7% (-132
kcal/d, P⫽.061); even so, improvements in memory and
executive function were correlated with decreased con-
sumption of energy, carbohydrates, and fat.
Of note, most tests involving memory and executive
function showed improvements. As the interval between
baseline and final evaluation was long (ie, 12 months), it
is unlikely that this improvement is attributable only to a
learning effect, especially considering that patients with
cognitive deficit suffer less effect learning (46) and cogni-
tive change showed a dose-response relationship with the
variation in BMI. As variation in some tests were related
to clinical and metabolic variables, it is possible that the
clinical management of comorbidities contributed to this
improvement. Even though loss of muscle mass, strength,
and functional capacity are potential risks associated with
of weight loss in the elderly (47), our results actually dem-
onstrated a functional improvement and stability of lean
mass.
Limitations
The weight loss difference between the groups was
lower than expected, limiting the analyses by the initial
study design. Thus, the initial goal, which depended on
comparing cognition variations in groups with different
degrees of weight loss, was not fully achieved. It is possible
that the long period of intervention and the mobility dif-
ficulties of the elderly decreased adherence to group meet-
ings, and the nutritional counseling was a mild interven-
tion. As the process of recruiting volunteers included
information on the risks of obesity, it is also possible that
this may have increased motivation to lose weight in both
groups. We therefore considered it most useful to make an
exploratory analysis considering the decrease of BMI as a
continuous variable and to observe its influence on cog-
nitive tests. Also, due to the exploratory nature of the
analyses and the relatively small sample, it was not pos-
sible to carry a multiple comparisons correction.
Our group was predominantly women, so the results
may not be generalizable. The number of patients might
have been too small for a detailed analysis by genotype.
Conclusions
Intentional weight loss through caloric restriction in obese
subjects with MCI was safe and correlated with improve-
ments in memory, executive function, global cognition
and language, and this association was strongest in
younger seniors and in APOE4 carriers. Changes in met-
abolic markers and diet were also associated with im-
provement in cognitive tests.
Acknowledgments
This work was supported by FAPESP (Fundação de Amparo à
Pesquisa do Estado de São Paulo) (2011/06194 –6) and Hospital
das Clinicas - Universidade de São Paulo.
Address all correspondence and requests for reprints to:
Dr.Nidia Celeste Horie Ph.D. nidiachorie@yahoo.com.br Fac-
uldade de Medicina da Universidade de São Paulo - Endocrinol-
ogy R Dr Ené as de Carvalho Aguiar, 155 8andar, Endocrino-
logia Sao Paulo SP BRAZIL 05403–000 55– 011–99640 6475.
Corresponding author and person to whom reprint requests
should be adressed: Nidia Celeste Horie,
nidiachorie@yahoo.com.br, Rua Dr Enéas de Carvalho Aguiar,
155, 8o. Andar, bloco 3 (Endocrinologia), CEP: 05403–000,
São Paulo – SP - Brasil
Disclosure Summary: The authors have nothing to disclose.
Clinical Trial Registration Number: NCT01286389.
This work was supported by This research was supported by
FAPESP (Fundação de Amparo à Pesquisa do Estado de São
Paulo) (2011/06194– 6) and Hospital das Clinicas - Universi-
dade de São Paulo. The funding agency was not involved in any
aspects of the planning or execution of this study.
References
1. World Health Organization and Alzheimer’s Disease International,
Dementia: A Public Health Priority, Geneva, Switzerland, 2012.
http://www.who.int/mental health/publications/dementia re-
port 2012/en/
2. Langa KM, Levine DA. The diagnosis and management of mild
cognitive impairment: a clinical review. JAMA. 2014 Dec 17;
312(23):2551–61. doi: 10.1001/jama.2014.13806.
3. Gorospe CE, Dave JK. The risk of dementia with increased body
mass index. Age and Ageing 2007. Age Ageing 36: 23–29.
4. Whitmer RA. Gunderson EP. Barrett-Connor E. Quesenberry CP Jr.
Yaffe K. Obesity in middle age and future risk of dementia: a 27 year
longitudinal population based study. BMJ. 2005 Jun ;330(7504):
136011.
5. Murphy T, Dias GP, Thuret S. Effects of diet on brain plasticity in
animal and human studies: mind the gap. Neural Plast. 2014;2014:
563160. doi:10.1155/2014/563160.
6. Witte AV, Fobker M, Gellner R, Knecht S, Flöel A. Caloric restric-
tion improves memory in elderly humans. Proc Natl Acad Sci USA.
2009 Jan 27;106(4):1255–60. doi: 10.1073/pnas.0808587106.
7. Marques EL, Halpern A, Corrêa Mancini M, de Melo ME, Horie
NC, Buchpiguel CA, Martins Novaes Coutinho A, Ono CR, Prando
S, Santo MA, Cunha-Neto E, Fuentes D, Cercato C. Changes in
8Cognition and weight loss in obese elderly J Clin Endocrinol Metab
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
Neuropsychological Tests and Brain Metabolism After Bariatric
Surgery. J Clin Endocrinol Metab. 2014 Aug 26:jc20142068.
8. Power BD, Alfonso H, Flicker L, Hankey GJ, Yeap BB, Almeida OP.
Body adiposity in later life and the incidence of dementia: the health
in men study. PLoS One. 2011 Mar 25;6(3):e17902. doi: 10.1371/
journal.pone.0017902.
9. Murphy RA, Patel KV, Kritchevsky SB Houston DK, Newman AB,
Koster A, Simonsick EM, Tylvasky FA, Cawthon PM, Harris TB;
Health, Aging, and Body Composition Study. Health, Aging, and
Body Composition Study. Weight change, body composition, and
risk of mobility disability and mortality in older adults: a popula-
tion-based cohort study. J Am Geriatr Soc. 2014 Aug;62(8):1476–
83. doi: 10.1111/jgs.12954.
10. Harrington M, Gibson S, Cottrell RC. A review and meta-analysis
of the effect of weight loss on all-cause mortality risk. Nutr Res Rev.
2009 Jun;22(1):93–108. doi: 10.1017/S0954422409990035.
11. Portet F, Ousset PJ, Visser PJ, Frisoni GB, Nobili F, Scheltens P,
Vellas B, Touchon J; MCI Working Group of the European Con-
sortium on Alzheimer’s Disease (EADC). Mild cognitive impairment
(MCI) in medical practice: a critical review of the concept and new
diagnostic procedure. Report of the MCI Working Group of the
European Consortium on Alzheimer’s Disease. J Neurol Neurosurg
Psychiatry. 2006 Jun;77(6):714– 8. Epub 2006 Mar 20. Review.
12. WHO. Global recommendations on physical activity for health.
2010. http://whqlibdoc.who.int/publications/2010/
9789241599979 eng.pdf.
13. Portet F, Ousset PJ, Visser PJ Frisoni GB, Nobili F, Scheltens P,
Vellas B, Touchon J; MCI Working Group of the European Con-
sortium on Alzheimer’s Disease (EADC). Mild cognitive impairment
(MCI) in medical practice: a critical review of the concept and new
diagnostic procedure. Report of the MCI Working Group of the
European Consortium on Alzheimer’s Disease. J Neurol Neurosurg
Psychiatry. 2006 Jun;77(6):714– 8.
14. Amorim P. Mini International Neuropsychiatric Interview (MINI):
validação de entrevista breve para diagnóstico de transtornos men-
tais. Rev. Bras. Psiquiatr. [online]. 2000, vol.22, n.3, pp. 106–115.
ISSN 1516– 4446. doi: 10.1590/S1516 – 44462000000300003.
15. Nakano MM. Versão Brasileira da Short Physical Performance Bat-
tery-SPPB: Adaptação Cultural e Estudo da Confiabilidade.
Monografia (Mestrado) – UNICAMP, Campinas, 2007;181.
16. Avanutri, Nutrição Serviços e Informática ME, Avanutri Revolution
– versão 4.0.
17. Universidade de Campinas. Núcleo de Estudos e Pesquisas em Ali-
mentação. Tabela brasileira de composição de alimentos – TACO.
4ª. Ed rev. e ampl. Campinas: UNICAMP, 2011.
18. Gebhardt SE, Thomas RG. Home and Garden Bulletin 72: Nutritive
Value of Foods. Washington, DC: US Department of Agriculture;
2002. [www.ars.usda.gov/Services/docs.htm?docid⫽6282].
19. Matsudo SM, Araújo TL, Matsudo VKR, Andrade DR, Andrade
EL, Oliveira LC, et al. Questionário Internacional de Atividade
Física (IPAQ): estudo de validade e reprodutibilidade no Brasil. Rev
Bras Ativ Saude. 2001;10:5–18.
20. Wechsler D. Wechsler Adult Intelligence Scale. 3rd ed. San Antonio,
TX: Psychological Corporation; 1997.
21. Diniz LFM, Cruz MF, Torres VM, Cosenza RM. O teste de apren-
dizagem auditivo-verbal de Rey: normas para uma população brasil-
eira. Rev Bras Neurol. 2000;36:79– 83.
22. Weschler D. Weschler Memory Scale - revised manual. New York:
Psychological Corporation, 1987.
23. Spreen O, Strauss E. A Compendium of neuropsychological test –
administration, norms and commentary. New York, Oxford Uni-
versity Press. 2nd ed., 1998.
24. Nelson HE. A modified card sorting test sensitive to frontal lobe
defects. Cortex,. 1976;12:313–324.
25. Caramelli P, Carthery MT, Charchat-Fichman H, Porto CS, Nitrini
R. Teste de fluência verbal no diagnóstico da doença de Alzheimer
leve: notas de corte em função da escolaridade. Arq Neuropsiquiatr
2003; 61(Supl 2):S32.
26. Sanchez MA, Lourenço RA. Informant Questionnaire on Cognitive
Decline in the Elderly (IQCODE): cross-cultural adaptation for use
in Brazil. Cad Saude Publica. 2009;25:1455–1465.
27. Nasreddine ZS, Phillips NA, Bedirian V, Charbonneau S, White-
head V, Collin I, Cummings JL, Chertkow H. The Montreal Cog-
nitive Assessment, MoCA: a brief screening tool for mild cognitive
impairment. J Am Geriatr Soc. 2005;53(4):695–699.
28. Memória CM, Yassuda MS, Nakano EY, Forlenza OV. Brief screen-
ing for mild cognitive impairment: validation of the Brazilian ver-
sion of the Montreal cognitive assessment. Int J Geriatr Psychiatry.
2013 Jan;28(1):34– 40. doi:10.1002/gps.3787.
29. Bottino CMC, Almeida OP, Tamai S, Forlenza OV, Scalco MZ,
Carvalho IAM. Entrevista estruturada para o diagnóstico de tran-
stornos mentais em idosos – CAMDEX – The Cambridge exami-
nation for mental disorders of the elderly. Brazilian version (trans-
lated and adapted on behalf of the editors). Cambridge: Cambridge
University Press; 1999.
30. Gomar JJ, Bobes-Bascaran MT, Conejero-Goldberg C, Davies P,
Goldberg TE; Alzheimer’s Disease Neuroimaging Initiative. Utility
of combinations of biomarkers, cognitive markers, and risk factors
to predict conversion from mild cognitive impairment to Alzheimer
disease in patients in the Alzheimer’s disease neuroimaging initia-
tive. Arch Gen Psychiatry. 2011 Sep;68(9):961–9. doi: 10.1001/
archgenpsychiatry.2011.96.
31. Riley KP, Jicha GA, Davis D, Abner EL, Cooper GE, Stiles N, Smith
CD, Kryscio RJ, Nelson PT, Van Eldik LJ, Schmitt FA. Prediction of
preclinical Alzheimer’s disease: longitudinal rates of change in cog-
nition. J Alzheimers Dis. 2011;25(4):707–17. doi: 10.3233/JAD-
2011–102133.
32. Cserjési R, Luminet O, Poncelet AS, Lénárd L. Altered executive
function in obesity. Exploration of the role of affective states on
cognitive abilities. Appetite. 2009 Apr;52(2):535–9.
33. García-Casares N, Jorge RE, García-Arnés JA, Acion L, Berthier
ML, Gonzalez-Alegre P, Nabrozidis A, Gutiérrez A, Ariza MJ, Rioja
J, González-Santos P. Cognitive Dysfunctions in Middle-Aged Type
2 Diabetic Patients and Neuroimaging Correlations: A Cross-Sec-
tional Study. J Alzheimers Dis. 2014 Jul 7.
34. Ewers M, Walsh C, Trojanowski JQ, Shaw LM, Petersen RC, Jack
CR Jr., Feldman HH, Bokde AL, Alexander GE, Scheltens P, Vellas
B, Dubois B, Weiner M, Hampel H; North American Alzheimer’s
Disease Neuroimaging Initiative (ADNI). Prediction of conversion
from mild cognitive impairment to Alzheimer’s disease dementia
based upon biomarkers and neuropsychological test performance.
Neurobiol Aging. 2012 Jul;33(7):1203–14. doi: 10.1016/j.neuro-
biolaging.2010.10.019.
35. Siervo M, Nasti G, Stephan BC, Papa A, Muscariello E, Wells JC,
Prado CM, Colantuoni A. Effects of intentional weight loss on phys-
ical and cognitive function in middle-aged and older obese partici-
pants: a pilot study. J Am Coll Nutr. 2012 Apr;31(2):79– 86.
36. Maki PM. Critical window hypothesis of hormone therapy and cog-
nition: a scientific update on clinical studies. Menopause. 2013 Jun;
20(6):695–709. doi: 10.1097/GME.0b013e3182960cf8.
37. Prins ND, Scheltens P. Treating Alzheimer’s disease with monoclo-
nal antibodies: current status and outlook for the future. Alzheimers
Res Ther. 2013 Nov 11;5(6):56. doi: 10.1186/alzrt220.
38. Villeneuve S, Brisson D, Marchant NL, Gaudet D. The potential
applications of Apolipoprotein E in personalized medicine. Front
Aging Neurosci. 2014 Jul 8;6:154. doi: 10.3389/fnagi.2014.00154.
eCollection 2014.
39. Risner ME, Saunders AM, Altman JF, Ormandy GC, Craft S, Foley
IM, Zvartau-Hind ME, Hosford DA, Roses AD; Rosiglitazone in
Alzheimer’s Disease Study Group. Efficacy of rosiglitazone in a ge-
netically defined population with mild-to-moderate Alzheimer’s dis-
ease. Pharmacogenomics J. . 2006 Jul-Aug;6(4):246–54.
40. Salloway S, Sperling R, Fox NC, Blennow K, Klunk W, Raskind M,
Sabbagh M, Honig LS, Porsteinsson AP, Ferris S, Reichert M, Ketter
N, Nejadnik B, Guenzler V, Miloslavsky M, Wang D, Lu Y, Lull J,
Tudor IC, Liu E, Grundman M, Yuen E, Black R, Brashear HR;
doi: 10.1210/jc.2015-2315 press.endocrine.org/journal/jcem 9
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
Bapineuzumab 301 and 302 Clinical Trial Investigators. Two phase
3 trials of bapineuzumab in mild-to-moderate Alzheimer’s disease.
N Engl J Med. 2014 Jan 23;370(4):322–33. doi: 10.1056/NEJ-
Moa1304839.
41. Petersen RC, Thomas RG, Grundman M, Bennett D, Doody R,
Ferris S, Galasko D, Jin S, Kaye J, Levey A, Pfeiffer E, Sano M, van
Dyck CH, Thal LJ; Alzheimer’s Disease Cooperative Study Group.
Vitamin E and donepezil for the treatment of mild cognitive impair-
ment. N Engl J Med. 2005 Jun 9;352(23):2379– 88.
42. Wakimoto P, Block G. Dietary intake, dietary patterns, and changes
with age: an epidemiological perspective. J Gerontol A Biol Sci Med
Sci. 2001;56:65–80.
43. Lieb W, Beiser AS, Vasan RS, Tan ZS, Au R, Harris TB, Roubenoff
R, Auerbach S, DeCarli C, Wolf PA, Seshadri S. Association of
plasma leptin levels with incident Alzheimer disease and MRI mea-
sures of brain aging. JAMA. 2009 Dec 16;302(23):2565–72. doi:
10.1001/jama.2009.1836.
44. Rajagopalan P, Toga AW, Jack CR, Weiner MW, Thompson PM;
Alzheimer’s Disease Neuroimaging Initiative. Fat-mass-related hor-
mone, plasma leptin, predicts brain volumes in the elderly. Neu-
roreport. 2013 Jan 23;24(2):58– 62. doi:10.1097/
WNR.0b013e32835c5254.
45. Gisou M, Soheila R, Nasser N. Evaluation of the effect of intrahip-
pocampal injection of leptin on spatial memory. Afr J. Pharm.P-
harmacol. 2009;3(9):443–448.
46. Schrijnemaekers AM, de Jager CA, Hogervorst E, Budge MM. Cases
with mild cognitive impairment and Alzheimer’s disease fail to ben-
efit from repeated exposure to episodic memory tests as compared
with controls. J Clin Exp Neuropsychol. 2006 Apr;28(3):438–55.
47. Villareal DT, Apovian CM, Kushner RF, Klein S; American Society
for Nutrition; NAASO, The Obesity Society. Obesity in older adults:
technical review and position statement of the American Society for
Nutrition and NAASO, The Obesity Society, Am J Clin Nutr 2005;
82:923–34.
10 Cognition and weight loss in obese elderly J Clin Endocrinol Metab
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 02 January 2016. at 08:13 For personal use only. No other uses without permission. . All rights reserved.
