Antioxidants and physical performance in elderly persons: the
Invecchiare in Chianti (InCHIANTI) study13
Matteo Cesari, Marco Pahor, Benedetta Bartali, Antonio Cherubini, Brenda WJH Penninx, G Rhys Williams,
Hal Atkinson, Antonio Martin, Jack M Guralnik, and Luigi Ferrucci
ABSTRACT
Background: Muscle strength and physical performance in old
age might be related to the oxidative damage caused by free
radicals.
Objective: The objective was to assess the correlation of plasma
concentrations and daily dietary intakes of antioxidants with skel-
etal muscle strength and physical performance in elderly persons.
Design: This study is part of the Invecchiare in Chianti (InCHI-
ANTI) study, which was conducted in 986 Italians aged 65 y.
Physical performance was assessed on the basis of walking speed,
ability to rise from a chair, and standing balance. Knee extension
strength was assessed with a hand-held dynamometer. The Euro-
pean Prospective Investigation into Cancer and Nutrition (EPIC)
questionnaire was used to evaluate the daily dietary intakes of
vitamin C, vitamin E, -carotene, and retinol. Plasma - and
-tocopherol concentrations were measured. Adjusted linear re-
gression analyses were used to calculate regression coefficients per
SD increase in plasma concentrations and daily dietary intakes.
Results: In adjusted analyses, plasma -tocopherol was signifi-
cantly correlated with knee extension ( 0.566, P 0.003) and
the summary physical performance score ( 0.044, P 0.008).
Plasma -tocopherol was associated only with knee extension
strength ( 0.327, P 0.04). Of the daily dietary intake
measures, vitamin C and -carotene were significantly correlated
with knee extension strength, and vitamin C was significantly
associated with physical performance ( 0.029, P 0.04).
Conclusions: Plasma antioxidant concentrations correlate posi-
tively with physical performance and strength. Higher dietary
intakes of most antioxidants, especially vitamin C, appear to be
associated with higher skeletal muscular strength in elderly
persons. Am J Clin Nutr 2004;79:28994.
KEY WORDS Antioxidants, dietary intake, physical perfor-
mance, elderly
INTRODUCTION
The assessment of physical function is becoming an increas-
ingly important component in the evaluation of elderly persons,
and standardized tests of physical performance are now widely
used both in research and clinical geriatric settings. In fact,
some of these performance measures, such as knee extension
strength and performanceas assessed by walking speed, bal-
ance, and ability to rise from a chair (chair stands)have been
shown to be useful in the prediction of institutionalization,
disability, and mortality (13).
Am J Clin Nutr 2004;79:289 94. Printed in USA. 2004 American Society for Clinical Nutrition
Recently, some authors suggested that age-related physical
decline might be related to oxidative damage perpetrated by
free radicals (4). A free radical has been defined as any
species capable of independent existence that contains one or
more unpaired electrons (5). This situation is energetically
unstable, making such species often highly reactive and short-
Downloaded from ajcn.nutrition.org by guest on October 3, 2017
lived. Free radicals disrupt the equilibrium of biological sys-
tems by damaging their major constituent molecules, leading
eventually to cell death (6, 7). Oxidative damage may play a
crucial role in the decline of functional activity in human
skeletal muscle with normal aging (8, 9). This hypothesis
suggests a primary role for antioxidant agents in the attempt to
avoid or reduce age-related physical decline. Antioxidants can
be defined as substances whose presence significantly inhibits
the rate of oxidation (10). The body contains an elaborate
antioxidant defense (11), and vitamin C, vitamin E, -carotene,
and retinol are the primary antioxidant vitamins (12). It has
been suggested that DNA damage, one of the most sensitive
biological markers of oxidative stress, might be caused by an
imbalance between the excessive generation of free radicals
and the deficient action of the antioxidant system (13).
We hypothesize that plasma antioxidants and the dietary
intake of antioxidants may correlate with strength and physical
1
From the Sticht Center on Aging, Wake Forest University Health
Sciences, Winston-Salem, NC (MC, MP, BWJHP, and HA); the Labora-
tory of Clinical Epidemiology, Italian National Research Council of Aging,
Geriatric Department, Florence, Italy (BB); the Department of Gerontology
and Geriatrics, University of Perugia, Perugia, Italy (AC); the Pharmaceu-
tical Research Institute, Bristol-Myers Squibb, Princeton, NJ (GRW); the
Human Nutrition Research Center on Aging, Tufts University, Boston
(AM); the Epidemiology, Demography, and Biometry Program, National
Institute on Aging, Bethesda, MD (JMG); the Longitudinal Studies Sec-
tion, Clinical Research Branch, National Institute on Aging, Baltimore
(LF).
2
Supported by Bristol-Myers Squibb Company (Princeton, NJ) and by
an unrestricted grant from BRACCO Imaging SpA, Italy. The InCHIANTI
study was supported as a targeted project (ICS 110.1/RS97.71) by the
Italian Ministry of Health and in part by the US National Institute on Aging
(NIA) (contracts 263 MD 9164 13 and 263 MD 821336). The work of MC
was supported by the Wake Forest University Claude D Pepper Older
Americans Independence Center (NIA grant P30-AG-021332-01).
3
Reprints not available. Address correspondence to M Cesari, Sticht Center
on Aging, Wake Forest University Health Sciences, 1 Medical Center Boule-
vard, Winston-Salem, NC 27157. E-mail: mcesari@wfubmc.edu.
Received April 17, 2003.
Accepted for publication July 10, 2003.
289
290 CESARI ET AL
performance. The purpose of this study was to assess the
correlation of plasma antioxidant concentrations and daily di-
etary intakes of antioxidants with skeletal muscle strength and
physical performance in elderly persons.
SUBJECTS AND METHODS
The present study is part of the Invecchiare in Chianti
(InCHIANTI) study, a prospective population-based study of
elderly people in Italy designed by the Laboratory of Clinical
Epidemiology of the Italian National Research Council of
Aging (Florence, Italy). The aims of the study are to identify
risk factors for the onset of disability in elderly persons, to
study physiologic subsystems critical for walking, and to de-
fine critical ranges for tests that evaluate the integrity of the
physiologic subsystems that are important for walking. The
study population included 1156 participants aged 65102 y,
who were randomly selected from residents in 2 towns of the
Chianti geographic area (Greve In Chianti and Bagno a Ripoli,
Tuscany, Italy) with the use of a multistage stratified sampling
method (14). The data collection started in September 1998 and
was completed in March 2000. A detailed description of the
sampling procedure and data collection method was previously
published (14). The Italian National Research Council of Ag-
ing Ethical Committee approved the study protocol.
The present analyses were conducted in 986 participants; we
excluded subjects in whom daily nutrient intakes were not
assessed or in whom tests of physical performance and of knee
extension strength were not conducted (n 136). We also
excluded 34 participants who were vitamin supplement users.
Sociodemographic characteristics showed that the excluded
participants were elderly (79.8 compared with 75.3 y), more
likely to be female (67.1% compared with 55.3%), and less
educated (4.7 compared with 5.4 y) than were those eligible for
the present analyses.
Physical performance tests
Knee extension strength was measured with a hand-held
dynamometer (Nicholas Muscle Tester; Sammon Preston Inc,
Chicago). Participants, lying in lateral decubitus (opposite to
the examined limb) with the hip and knee in 45 and 60
flexed positions, respectively, were asked to perform the task
twice with each leg. The average of the best results obtained for
each leg was used for the present analyses. In our study we
found a high correlation between the 2 strength measures done
for each leg (r 0.896, P 0.001). This measurement of knee
extension strength has been proven to be reliable on the basis
of high test-retest and interrater reliabilities (0.85 and 0.74,
respectively) (15).
Performance of the lower extremities was assessed with the
use of 3 tests: walking speed, ability to stand from a chair, and
ability to maintain balance in progressively more challenging
positions. Walking speed was defined as the best performance
(time in s) of two 4-m walks at usual pace along a corridor. For
the chair-stand test, the participants were asked to rise and sit
down 5 times as quickly as possible with their hands folded
across the chest. The performance was expressed as total time
(in s) to complete the test. For the standing-balance test, the
subjects were asked to stand in 3 progressively more difficult
positions for 10 s each: a side-by-side feet standing position, a
semitandem position, and a full-tandem position. Good to
excellent reliability of these tests has been shown, which sup-
ports the utilization of these measures in aging research (16).
An arithmetic summary performance score was calculated to
obtain a continuous measure (17). The timed scores of the
performance tests were rescaled to values ranging from 0 to 1,
where 1 indicates the best performance and 0 the worst per-
formance. The worst performers were participants who were
unable to complete the task or those who had a performance
above the 99th percentile (walking time: 15.5 s; chair-stand
time: 27.3 s; standing-balance test: 30 s). The following for-
mulas were applied to rescale measures: 1) walking speed, 1
(15.5/speed in centimeters per second); 2) chair-stand test, 1
(time in seconds/27.3); and 3) standing-balance test, time in
seconds/30. A summary physical performance score ranging
from 0 to 3 was calculated by adding these 3 rescaled scores.
Such a summary score has been shown to have excellent
predictive validity for mortality, institutionalization, hospital-
ization, and incident disability (1, 3, 18, 19). The internal
consistency (Cronbachs ) of the summary scale in our study
Downloaded from ajcn.nutrition.org by guest on October 3, 2017
was 0.871.
Plasma antioxidants
Plasma vitamin E (- and -tocopherol) concentrations were
measured by reversed-phase HPLC as previously described
(20). Briefly, 100 mL plasma was mixed with 100 mL ethanol;
after being mixed by vortex, tocopherol was extracted into 500
mL hexane containing 0.002% butylated hydroxytoluene
(Sigma, St Louis). Tocol, a gift from Hoffmann-La Roche
(Nutley, NJ), was added to the mixture as an internal standard.
Samples were centrifuged at 180 g for 5 min at 4 C. The
supernatant fluid was collected and dried under a stream of
nitrogen gas and was reconstituted in 100 mL methanol. To-
copherols were separated by HPLC with a 3-m C18 reversed-
phase column (Perkin-Elmer, Norwalk, CT). The mobile phase,
delivered at a flow rate of 1.0 mL/min, consisted of 1% water
in methanol containing 10 mmol lithium perchlorate/L. Sam-
ples were injected with an autosampler (1100 series; Hewlett-
Packard, Palo Alto, CA). Eluted peaks were detected at an
applied potential of 0.6 V by an LC 4B amperometric elec-
trochemical detector (Bioanalytical System, West Lafayette,
IN). Peaks were integrated with ChemStation software
(Hewlett-Packard). Tocopherol concentrations were expressed
in mol/L. The reproducibility and accuracy of the procedures
used were tested by analyzing representative samples in trip-
licate from a sample provide by the American Association for
Laboratory Accreditation (Washington, DC), which contained
known concentrations of -tocopherol. Intra- and interbatch
CVs were 3% and 4.2%, respectively.
EPIC questionnaire
To determine daily nutritional intakes, we used the EPIC
(European Prospective Investigation into Cancer and Nutrition)
questionnaire (21, 22). This questionnaire was designed and
validated in a multicenter study performed in 10 European
countries, with the aim to evaluate the relation between diet
and the onset of cancer. Participants were asked how many
times in a year each food (or food category) and beverage were
consumed. All of the answers provided by the participants were
reported through specific software in a database, and the daily
 ANTIOXIDANTS AND PHYSICAL PERFORMANCE 291
intakes of micro- and macronutrients were calculated. The TABLE 1
software for the estimate of food composition was created and Participant characteristics1
verified by the EPIC study group according to food-composition Value
tables compiled in 1998 by researchers at the European Institute of
Oncology. Sociodemographic characteristics
Age (y) 75.3 0.22
In a previous study (23), data on dietary intake collected in
Female sex (%) 55.3
the InCHIANTI study were compared with data collected in a
Residents of Greve, Italy (%) 47.8
nationwide study in which a direct method of measure based on Education (y) 5.4 (0.1)
7-d weighted and registered food consumption was used. De- Mini-Mental State Examination Score 24.6 (0.2)
spite different populations and different methods used for data Smoking (%)
collection, the distributions of dietary intake from the 2 studies Never 59.0
were remarkably similar for most macro- and micronutrients, Former 30.1
both in men and women, which suggests that the EPIC food- Current 10.9
frequency questionnaire provides a good estimate of dietary BMI (kg/m2) 27.5 (0.1)
intake. In the present analyses, we considered the EPIC 20 (%) 1.9
2024.9 (%) 26.0
questionnairederived daily intakes of 4 antioxidant agents:
2530 (%) 43.5
vitamin C, vitamin E, -carotene, and retinol.
30 (%) 23.9
Covariates Missing (%) 4.7
Physical activity (%) 37.0
The covariates included sociodemographic variables (age, Plasma lipids (mg/dL)

Downloaded from ajcn.nutrition.org by guest on October 3, 2017

sex, site, smoking habit, and education), cognitive status (Mini- Total cholesterol 217.6 (1.3)
Mental State Examination score), plasma lipid concentrations Triacylglycerols 110.5 (84.0150.0)3
(total cholesterol and triacylglycerols), comorbidity (self- Comorbid conditions (%)
reported diagnoses of hypertension, angina, myocardial infarc- Hypertension 44.4
tion, stroke, cancer, diabetes, congestive heart failure, and Angina or myocardial infarction 7.6
Stroke 5.8
chronic obstructive pulmonary disease), body mass index
Cancer 10.3
(computed as weight in kg/height in m2), the number of med- Diabetes 11.6
ications taken by participants at the baseline assessment, and Congestive heart failure 5.8
physical activity (defined as a moderate-to-high intensity ex- COPD 6.3
ercise performed for 12 h/wk or a light-intensity exercise Number of medications taken 2.2 (0.1)
performed for 4 h/wk). To adjust for overall food intake, we Plasma antioxidants (mol/L)
also adjusted our analyses for daily dietary energy intake, as -Tocopherol 29.96 (0.27)
assessed by the EPIC questionnaire. Participants were asked to -Tocopherol 1.45 (0.02)
report any medication taken in the last 2 wk. Drugs were coded EPIC questionnaire measures
according to the Anatomical Therapeutic and Chemical codes Vitamin C (mg) 110.78 (1.54)
Vitamin E (mg) 6.22 (0.06)
(24).
-Carotene (g) 2167.35 (36.91)
Statistical analyses Retinol (g) 492.76 (17.70)
Energy intake (kcal) 1920.17 (17.95)
Because the plasma concentrations of -tocopherol and - Performance tests
tocopherol, and daily dietary intakes of vitamin C, vitamin E, Physical performance4 2.18 (0.02)
-carotene, and retinol were not normally distributed, analyses Knee extension strength (kg) 15.17 (0.19)
were performed with the use of their log values. Spearman 1
n 986. COPD, chronic obstructive pulmonary disease; EPIC,
correlation tests were used to evaluate correlations of antioxi- European Prospective Investigation into Cancer and Nutrition.
dants plasma concentrations and daily dietary intakes with knee 2
x SE.
extension strength and summary physical performance score. 3
Median; interquartile range in parentheses.
All analyses were adjusted for age, sex, site, Mini-Mental State 4
Possible score ranges from 0 to 3.
Examination score, smoking habit, education, body mass in-
dex, total cholesterol, triacylglycerol (log value), hypertension,
angina or myocardial infarction, congestive heart failure, dia- RESULTS
betes, stroke, number of medications taken, and daily dietary The main characteristics of the participants are shown in
energy intake; all these variables showed a significant correla- Table 1. The mean (SE) age of the participants was 75.3
tion with outcome measures. Linear regression analyses were 0.2 y, and 55.3% were female. The most common diagnoses
used to identify regression coefficients for strength and perfor- reported by the participants were hypertension (44.4%), diabe-
mance outcomes per SD increase in plasma antioxidant con- tes (11.6%), cancer (10.3%), and angina or myocardial infarc-
centrations and daily dietary intakes. To calculate adjusted tion (7.6%).
knee extension strength and summary physical performance Mean (SE) plasma concentrations of - and -tocopherol
scores, analyses of covariance were performed for tertiles of were 29.96 0.27 and 1.45 0.02 mol/L, respectively.
antioxidants (plasma concentrations and dietary intakes) and Plasma concentrations of -tocopherol were correlated with
daily energy intake. Sex interactions were assessed by adding plasma concentrations of -tocopherol (r 0.373, P 0.001)
the interaction term for sex antioxidant concentration in the and with daily dietary intake of vitamin E (r 0.126, P
adjusted model. 0.001). No significant correlation was found between plasma
292 CESARI ET AL

TABLE 2 -carotene were significantly correlated with knee extension

Spearman correlations between physical performance and muscular strength (vitamin C: 0.383, SE 0.162, P 0.02; -caro-
strength tests and plasma concentrations and dietary intakes of tene: 0.311, SE 0.159, P 0.05). Only the daily intake of
antioxidants
vitamin C was significantly associated with physical performance
Knee extension Physical ( 0.029, SE 0.014, P 0.04). No significant sex interaction
strength performance score was found between dietary intake and physical performance (all
P values 0.05).
kg
Plasma antioxidants Adjusted means for knee extension strength and summary
-Tocopherol 0.054 0.1151 physical performance score according to tertiles of plasma
-Tocopherol 0.057 0.1071 antioxidant concentrations and daily dietary intakes were as-
Daily dietary intakes sessed. Greater knee extension strength was found at higher
Vitamin C 0.2101 0.2191 concentrations of -tocopherol (lowest compared with highest
Vitamin E 0.2811 0.2431 tertile: P 0.003 and P for trend 0.01). Higher plasma
-Carotene 0.2161 0.2021 concentrations of -tocopherol were found at higher physical
Retinol 0.0951 0.0732 performance scores (lowest compared with highest tertile: P
1
P 0.01. 0.009 and P for trend 0.003). Similar results were found for
2
P 0.05. higher daily intakes of vitamin C and greater knee extension
strength (lowest compared with highest tertile: P 0.008 and
-tocopherol and daily dietary intake of vitamin E. This finding P for trend 0.03) and physical performance (lowest com-
could have been due to a higher consumption of olive oil (x pared with highest tertile: P 0.03 and P for trend 0.08).
Higher daily intakes of -carotene tended to be associated with

Downloaded from ajcn.nutrition.org by guest on October 3, 2017

SD: 26.1 13.0 g/d) than of other vegetable oils (x SD: 1.0
1.6g/d) and animal fats (x SD: 0.7 1.0 g/d) in our
sample. Confirming our hypothesis, Spearman correlations
showed a significant positive relation between olive oil con-
sumption and -tocopherol (r 0.073, P 0.02) but not with
-tocopherol.
In unadjusted analyses (Table 2), daily dietary intakes of
vitamin C, vitamin E, and -carotene were significantly and
positively correlated with knee extension strength. Plasma con-
centrations of - and -tocopherol and daily dietary intakes of
vitamin C, vitamin E, and -carotene were similarly correlated
with summary physical performance scores.
Adjusted regression coefficients per SD increase in plasma
antioxidants for knee extension strength and physical perfor-
mance are described in Table 3. -Tocopherol was signifi-
cantly correlated with both performance measures (knee exten-
sion: 0.566, SE 0.193, P 0.003; physical performance:
0.044, SE 0.017, P 0.008). -Tocopherol was associated
only with knee extension strength ( 0.327, SE 0.165, P
0.04). No significant sex interaction was found between plasma
antioxidants and knee extension strength or physical performance
(all P values 0.05). Daily dietary intakes of vitamin C and
greater knee extension strength (lowest compared with highest
tertile: P 0.05 and P for trend 0.06).
DISCUSSION
This study evaluated the relation between antioxidants and
objective performance assessments in elderly persons. We in-
vestigated whether plasma concentrations and daily dietary
intakes of antioxidants were associated with physical perfor-
mance and muscular skeletal strength in an elderly population.
To our knowledge, this is the first study that evaluated simul-
taneously the effects of dietary antioxidant intakes and plasma
antioxidant concentrations on physical performance in an el-
derly population. Previous studies were focused on younger
samples, especially athletes, and principally aimed at assessing
whether antioxidant supplementation reduces the rate of oxi-
dation or helps promote faster recovery from the damage
generated by exercise (11, 12, 2527). Our analyses suggest a
relation between plasma concentrations of antioxidants and
physical performance and strength. A significant positive as-
sociation, especially on muscular skeletal strength, was also

TABLE 3
Adjusted regression coefficients for knee extension strength and summary physical performance scores per SD increase (log) in plasma antioxidants
and in dietary intakes of antioxidants1

Knee extension strength (kg) Physical performance score

Regression coefficient (SE) P Regression coefficient (SE) P

Plasma
-Tocopherol (mol/L) 0.566 (0.193) 0.003 0.044 (0.017) 0.008
-Tocopherol (mol/L) 0.327 (0.165) 0.04 0.004 (0.015) 0.80
Daily intake
Vitamin C (mg) 0.383 (0.162) 0.02 0.029 (0.014) 0.04
Vitamin E (mg) 0.277 (0.204) 0.17 0.013 (0.018) 0.47
-Carotene (g) 0.311 (0.159) 0.05 0.015 (0.014) 0.29
Retinol (g) 0.012 (0.147) 0.94 0.010 (0.013) 0.31
1
The coefficients were adjusted for age, sex, site, Mini-Mental State Examination score, smoking, education, BMI, physical activity, total cholesterol,
triacylglycerol (log value), hypertension, coronary heart disease, stroke, congestive heart failure, number of medications taken, and daily dietary energy
intake. The -tocopherol SD 0.278, -tocopherol SD 0.441, daily vitamin C intake (log) SD 0.449, daily vitamin E intake (log) SD 0.323, daily
-carotene intake (log) SD 0.508, and daily retinol intake (log) SD 0.910.
 ANTIOXIDANTS AND PHYSICAL PERFORMANCE
found for the daily dietary intakes of several antioxidants,
including vitamin C and -carotene.
Free radical oxidative stress has been implicated in the
pathogenesis of a variety of diseases, and natural antioxidant
defenses have been found to be defective in many diseases
(10). A free radical is a molecule that contains an unpaired
electron in its outer orbit and that can exist independently (5,
12). Stability is achieved by the removal of electrons from
surrounding molecules to produce an electron pair. The re-
mainder of the attached molecule then possesses an unpaired
electron and therefore becomes a free radical. In this way, the
presence of a single radical may initiate a chain sequence of
electron transfer (redox) reactions. Several studies suggest an
important role of oxidation in physiologic decline with normal
aging (8, 9, 13). Oxidative stress can come from a variety of
sources, from abnormal metabolic pathways to inflammatory
cells (eg, macrophages and neutrophils producing hydrogen
peroxide and hypochlorus acid as means of bacterial killing)
(28), smoking, and radiation (10).
Our results confirmed previous findings that suggest a strong
correlation of plasma antioxidant concentrations with physical
performance and strength. In fact, we found a significantly high
correlation of -tocopherol, the predominant form of circulat-
ing vitamin E (29), with physical performance and strength
tests. We also reported a strong correlation between skeletal
muscular strength and -tocopherol, a vitamin E form indicated
by some studies to be a more effective trap for lipophilic-free
radicals than -tocopherol (29). In our study, we found no
significant sex interaction between performance and antioxi-
dants.
We also examined the effects of the dietary intake of major
antioxidants on physical performance and strength. Vitamin C
is a water-soluble vitamin that represents a first-line antioxi-
dant defense in plasma. It is a powerful inhibitor of lipid
peroxidation and regenerates vitamin E in lipoproteins and
membranes. -Carotene is a lipid-soluble antioxidant and a
precursor of retinol that works synergistically with vitamin C.
Vitamin E is a major lipid-soluble chain-breaking antioxidant
that prevents lipid peroxidation in lipoproteins and biological
membranes. Its antioxidant function is strongly supported by
regeneration promoted by vitamin C (10, 12).
Our findings suggest an association between higher daily
dietary intakes of vitamin C and -carotene and skeletal muscle
strength. Whether higher antioxidant intakes are beneficial in
promoting better physical performance and muscular strength
is controversial. Although the findings of some studies do not
support a beneficial effect of increased antioxidant intakes on
physical performance (11, 12, 30), other studies have shown
improvements in the recovery from exercise with antioxidant
intake and a preventive role of antioxidant supplements against
oxidative damage (10, 27, 31).
In our study we showed a nonsignificant correlation between
the dietary intake of vitamin E and -tocopherol but a signif-
icant correlation with -tocopherol. The nonsignificant associ-
ation between vitamin E and -tocopherol may have been due
to the participants high consumption of olive oil, which was
the major source of vitamin E in the study. Dietary intake of
vitamin E includes the sum intake of all tocopherol and toco-
trienols with vitamin activity. Therefore, such a summary mea-
sure is probably not precise and specific and may explain the
293
nonsignificant association between the dietary intake of vita-
min E and physical performance and strength measures.
Even though antioxidant supplementation has been sug-
gested as a mean to perform better and recover more quickly
and fully from vigorous exercise, the theoretical basis for why
antioxidants may enhance performance is not clear. An im-
proved antioxidant status is one adaptation to regular exercise
training (32), and an acute increase in plasma antioxidant
activity has been observed after exercise (33). This finding
suggests that antioxidant status may be important in healthy
muscle function, especially during the recovery phase after
exercise. To our knowledge, no evidence is available that
documents the suppression of oxidative damage in muscle by
antioxidant supplements, but it seems reasonable that an ade-
quate antioxidant intake is needed to maintain healthy muscular
activity (10, 34).
Several limitations of our study should be mentioned. The
cross-sectional design of the study did not permit us to evaluate
the effect of daily nutrient intakes on changes in physical
performance. Further longitudinal studies are needed to eval-
Downloaded from ajcn.nutrition.org by guest on October 3, 2017
uate the role that antioxidant intakes may have on physical
performance in the elderly. The study population was charac-
terized by high nutritional indexes and tended to be over-
weight. These characteristics did not allow us to assess the
effect of very poor nutrition. The general overweight status of
participants may have influenced our findings. To better iden-
tify the effects of antioxidants on physical performance, it may
be useful to evaluate a population in which under- and normal-
nourished persons are used as comparison groups. Finally,
another limitation of our study was its limited assessment of
biological measures. The evaluation of both dietary intake
estimates and biological measures would have been the ideal
combination, especially given the significant association found
between dietary intakes of vitamin C and -carotene and knee
extension strength.
In conclusion, our study indicates a significant positive cor-
relation between plasma antioxidant concentrations and phys-
ical performance and strength. Higher dietary intakes of anti-
oxidants, especially of vitamin C, in the elderly were also
found to be associated with greater skeletal muscular strength.
However, the potentially positive effects of antioxidant intakes
on performance and strength need to be confirmed in future
prospective studies and clinical trials.
The EPIC-Italy food-frequency questionnaire was developed as a coop-
erative project supported by the Italian Association for Cancer Research.
We are particularly grateful to Domenico Palli and Simonetta Salvini
(Molecular and Nutritional Epidemiology Unit, CSPO, Scientific Institute
of Tuscany, Florence, Italy) and Vittorio Krogh (National Cancer Institute,
Milan, Italy) for their help in processing and developing the InCHIANTI
nutritional database.
MC helped design the study, analyze the data, and write the manuscript.
MP and BWJHP helped design the study, write the manuscript, and
critically review the manuscript. BB helped collect and analyze the data
and critically review the manuscript. AC and AM helped collect the data
and critically review the manuscript. GRW and HA helped critically review
the manuscript. JMG helped design the study and critically review the
manuscript. LF helped design the study, collect and manage the data, and
critically review the manuscript. None of the authors had any financial or
personal conflict of interest. Bristol-Myers Squibb sponsored the data
analyses but had no influence on the design of the study, the data collec-
tion, the analyses and interpretation of the data, or the decision to publish
the results.
294 CESARI ET AL
REFERENCES
1. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical
performance battery assessing lower extremity function: association
with self-reported disability and prediction of mortality and nursing
home admission. J Gerontol A Biol Med Sci 1994;49:M8594.
2. Laukkanen P, Heikkinen E, Kauppinen M. Muscle strength and mo-
bility as predictors of survival in 75 84-year-old people. Age Aging
1995;24:468 73.
3. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB.
Lower-extremity function in persons over the age of 70 years as a
predictor of subsequent disability. N Engl J Med 1995;332:556 61.
4. De La Fuente M. Effects of antioxidants on immune system ageing.
Eur J Clin Nutr 2002;56:S5 8.
5. Halliwell B, Gutteridge JMC. Free radicals in biology and medicine.
Oxford, United Kingdom: Clarendon Press, 1989.
6. Aust SD, Morehouse LA, Thomas CE. Role of metals in oxygen
radical reactions. J Free Radic Biol Med 1985;1:325.
7. Sevanian A, Hochstein P. Mechanisms and consequences of lipid
peroxidation in biological systems. Annu Rev Nutr 1985;5:36590.
8. Pansarasa O, Castagna L, Colombi B, Vecchiet J, Felzani G, Marzatico
F. Age and sex differences in human skeletal muscle: role of reactive
oxygen species. Free Radic Res 2000;33:28793.
9. Fano G, Mecocci P, Vecchiet J, et al. Age and sex influence on
oxidative damage and functional status in human skeletal muscle.
J Muscle Res Cell Motil 2001;22:34551.
10. Maxwell SRJ. Prospects for the use of antioxidant therapies. Drugs
1995;49:345 61.
11. Clarkson PM. Antioxidants and physical performance. Crit Rev Food
Sci Nutr 1995;35:131 41.
12. Clarkson PM, Thompson S. Antioxidants: what role do they play in
physical activity and health? Am J Clin Nutr 2000;72(suppl):637S
46S.
13. Mendoza-Nunez VM, Sanchez-Rodriguez MA, Retana-Ugalde R,
Vargas-Guadarrama LA, Altamirano-Lozano MA. Total antioxidant
levels, gender, and age as risk factors for DNA damage in lymphocytes
of the elderly. Mech Age Dev 2001;122:835 47.
14. Ferrucci L, Bandinelli S, Benvenuti E, et al. Subsystems contributing
to the decline in ability to walk: bridging the gap between epidemiol-
ogy and geriatric practice in the InCHIANTI study. J Am Geriatr Soc
2000;48:1618 25.
15. Bandinelli S, Benvenuti E, Del Lungo I, Baccini M, Benvenuti F, Di
Iorio A. Measuring muscular strength of the lower limbs by hand-held
dynamometer: a standard protocol. Aging 1999;11:28793.
16. Ferrucci L, Guralnik JM, Salive ME, et al. Effect of age and severity
of disability on short-term variation in walking speed: the Womens
Health and Aging Study. J Clin Epidemiol 1996;49:1089 96.
17. Onder G, Penninx BWJH, Lapuerta P, et al. Change in physical
performance over time in older women: the Womens Health and
Aging Study. J Gerontol A Biol Med Sci 2002;57A:M289 93.
18. Ostir GV, Markides KS, Black SA, Goodwin JS. Lower body func-
tioning as a predictor of subsequent disability among older Mexican
Americans. J Gerontol A Biol Sci Med Sci 1998;53:M4915.
19. Penninx BWHJ, Ferrucci L, Leveille SG, Rantanen T, Pahor M,
Guralnik JM. Lower extremity performance in nondisabled older per-
sons as a predictor of subsequent hospitalization. J Gerontol A Biol Sci
Med Sci 2000;55:M6917.
20. Martin A, Zulueta J, Hassaun P, Blumberg J, Meydani M. The effect
of vitamin E on human vascular endothelial cells response to oxidative
injury induced by hypoxia/reperfusion. Free Radic Biol Med 1995;20:
99 105.
21. Kaaks R, Riboli E. Validation and calibration of dietary intake mea-
surements in the EPIC Project: methodological considerations. Int J
Epidemiol 1997;26(suppl):S1525.
22. Kaaks R, Slimani N, Riboli E. Pilot phase studies on the accuracy of
dietary intake measurements in the EPIC Project: overall evaluation of
results. Int J Epidemiol 1997;26(suppl):S26 36.
23. Bartali B, Turrini A, Salvini S, et al. Dietary intake estimated using
different methods in two Italian older populations. Arch Gerontol
Geriatr (in press).
24. Pahor M, Chrischilles EA, Guralnik JM, Brown SL, Wallace R,
Carbonin PU. Drug data coding and analysis in epidemiologic studies.
Eur J Clin Epidemiol 1994;10:40511.
25. Barnett DW, Conlee RK. The effects of a commercial dietary supple-
ment on human performance. Am J Clin Nutr 2003;40:586 90.
26. Balakrishnan SD, Anuradha CV. Exercise, depletion of antioxidants
Downloaded from ajcn.nutrition.org by guest on October 3, 2017
and antioxidant manipulation. Cell Biochem Funct 1998;16:269 75.
27. Takanami Y, Iwane H, Kawai Y, Shimomitsu T. Vitamin E supple-
mentation and endurance exercise: are there benefits? Sports Med
2000;29:73 83.
28. Weiss SJ. Tissue destruction by neutrophils. N Engl J Med 1989;320:
36576.
29. Jiang Q, Christen S, Shigenaga MK, Ames BN. Gamma-tocopherol,
the major form of vitamin E in the US diet, deserves more attention.
Am J Clin Nutr 2001;74:714 22.
30. Oostenbrug GS, Mensink RP, Hardeman MR, De Vries T, Brouns F,
Hornstra G. Exercise performance, red blood cell deformability, and
lipid peroxidation: effects of fish oil and vitamin E. J Appl Physiol
1997;83:746 52.
31. Jakeman P, Maxwell SRJ. Effect of antioxidant vitamin supplementa-
tion on muscle function after eccentric exercise. Eur J Appl Physiol
1993;67:426 30.
32. Robertson JD, Maugham RJ, Duthie GG, Morrice PC. Increased blood
antioxidant systems of runners in response to training load. Clin Sci
1991;80:611 8.
33. Maxwell SRJ, Jakeman P, Thomason H, Leguen C, Thorpe GH.
Changes in plasma antioxidant status during eccentric exercise and the
effect of vitamin supplementation. Free Radic Res Commun 1993;19:
191202.
34. Jackson M, Edwards RH. Free radicals and trials of antioxidant ther-
apy in muscle disorders. Adv Exp Biol Med 1990;264:48591.

Vegetarian diets and exposure to organochlorine
pollutants, lead, and mercury
Dear Sir:
The practice of feeding animal meat and bone meal to cattle has
been linked to the recent epidemic of bovine spongiform encepha-
lopathy. Bovine spongiform encephalopathy, in turn, has been
linked to the human prion disease variant Creutzfeldt-Jakob disease
(1). In that recent issue of the Journal, Trevitt and Singh introduced
an interesting issue that has escaped previous discussions regarding
vegetarian diets and human health. Not only are harmful prions
found in the meats of animals that have consumed animal products,
but also found in these meats are persistent lipophilic organic pol-
lutants (PLOP), such as pesticides, polychlorinated dibenzo-p-
dioxins, dibenzofurans, polychlorinated biphenyls, and polybromi-
nated compounds. Animal products can also contain other
neurotoxic substances, such as monomethyl mercury (MMHg) and
lead.
Substances that are resistant to degradation and that are released
in the environment accumulate in the food web and often end up
being consumed by humans. The chemical characteristics of these
persistent pollutants determine the location of accumulation, the
metabolism, and the half-time within animal tissues. Lipophilic sub-
stances such as PLOP accumulate in fat tissues, whereas lead accu-
mulates in bones as a function of age and diet. Although organo-
chlorine pesticides are a significant contaminant of crops, animal
byproducts are the major carriers of these harmful substances (2).
MMHg is also a fat-soluble substance, but it is complexed with
hydrophilic sulfhydril-containing molecules of protein matrices in
animal tissues. A series of chemical reactions involving mercury and
sulfate-reducing bacteria produce MMHg, which is then introduced
in the aquatic food chain and is thereby acquired by humans through
the consumption of fish and shellfish. Animals and humans share
some of the bioaccumulation features of these persistent substances,
but, because of different dietary habits, humans may or may not be
exposed to them.
The trend among affluent countries is to feed animal byproducts
to poultry and ruminants (herbivores such as cows, sheep, and goats).
These animals are used in milk production or are consumed as meat.
Fishmeal is mainly used as feed for dairy cows (3). Fishmeal con-
sumption results in an increase in milk production and stimulates an
increase in docosahexaenoic and eicosapentaenoic acids (4). Cow
feed may also contain significant proportions of bone meal (5).
The health benefits of vegetarian diets have been a topic of health
and nutrition research since long before the emergence of modern
environmental concerns. However, because of increases in environ-
mental pollution, dietary habits have become an increasingly important
public health issue. Studies have shown that a primarily vegetarian diet
results in decreased exposures to bioconcentrated environmental pol-
lutants. Vegetarian mothers have been shown to secrete substantially
fewer organochlorine substances in breast milk than do nonvegetarian
mothers (6). Environmental estrogenic substances (polychlorinated
Am J Clin Nutr 2004;80:237 8. Printed in USA. 2004 American Society for Clinical Nutrition
Letter to the Editor
biphenyls and phthalate esters) have been found in human serum in the
following relative concentrations: urban fish eaters rural fish eaters
urban vegetarians rural vegetarians (7). Vegetarian vegans from
Hong Kong who consumed no fish or shellfish were shown to have very
low concentrations of hair mercury compared with nonvegetarians (8).
Passos et al (9) suggest that an increased consumption of fruit can lower
the body burden of mercury in Brazilian Amazon fish-eating popula-
tions. After 12 mo of changing from an omnivorous diet to a lactoveg-
etarian diet, healthy adults had decreased hair concentrations of mercury
and lead (10). Therefore, a predominance of plant food in human diets
is an important step in lowering the body burden of harmful substances
in populations exposed to pollutants.
Strict vegetarianism, or veganism, is defined as the practice of
consuming plant food only (no animal products). However, variants
of vegetarianismsuch as lactoovovegetarianisminvolve the
practice of consuming eggs and dairy products. Although a vegetar-
ian lifestyle can lower the body burden of PLOP, MMHg, and lead,
such benefits can be undermined by the consumption of contami-
nated milk and egg products. Farm animals that have been fed con-
taminated animal products produce contaminated milk and egg
products.
In summary, the inclusion of animal products in some forms of
vegetarianism can increase the exposure of humans to persistent
pollutants. Vegetarians with a desire to reduce their exposure to
neurotoxic substances (eg, PLOP, MMHg, and lead) should be
aware of farming practices that introduce these pollutants into the
human diet.
Jos G Dorea
Department of Nutrition
CP 04322
Universidade de Brasilia
70919.970 Brasilia DF
Brazil
E-mail: dorea@rudah.com.br
REFERENCES
1. Trevitt CR, Singh PN. Variant Creutzfeldt-Jakob disease: pathology,
epidemiology, and public health implications. Am J Clin Nutr 2003;
78(suppl):651S 6S.
2. Singh PP, Chawla RP. Insecticide residues in total diet samples in Pun-
jab, India. Sci Total Environ 1988;76:139 46.
3. Yeo JM, Knight CH, Chamberlain DG. Effects of changes in dietary
amino acid balance on milk yield and mammary function in dairy cows.
J Dairy Sci 2003;86:1436 44.
4. Gulati SK, McGrath S, Wynn PC, Scott TW. Preliminary results on the
relative incorporation of docosahexaenoic and eicosapentaenoic acids
into cows milk from two types of rumen protected fish oil. Int Dairy J
2003;13:339 43.
5. Akayezu JM, Hansen WP, Otterby DE, Crooker BA, Marx GD. Yield
237
238 ERRATUM

response of lactating Holstein dairy cows to dietary fish meal and bone
meal. J Dairy Sci 1997;80:2950 63.
6. Noren K. Levels of organochlorine contaminants in human milk in
relation to the dietary habits of the mothers. Acta Paediatr Scand 1983;
72:811 6.
7. Rozati R, Reddy PP, Reddanna P, Mujtaba R. Role of environmental
estrogens in the deterioration of male factor fertility. Fertil Steril 2002;
78:118794.
8. Dickman MD, Leung CK, Leong MK. Hong Kong male subfertility links
to mercury in human hair and fish. Sci Total Environ 1998;214:16574.
9. Passos CJ, Mergler D, Gaspar E, et al. Eating tropical fruit reduces
mercury exposure from fish consumption in the Brazilian Amazon.
Environ Res 2003;93:12331.
10. Srikumar TS, Johansson GK, Ockerman PA, Gustafsson JA, Akesson B.
Trace element status in healthy subjects switching from a mixed to a
lactovegetarian diet for 12 mo. Am J Clin Nutr 1992;55:88590.

Erratum

Cesari M, Pahor M, Bartali B, et al. Antioxidants and physical performance in elderly persons: the Invecchiare
in Chianti (InCHIANTI) study. Am J Clin Nutr 2004;79:289 94.

On page 290, column 2, paragraph 3, the unit mL is used incorrectly in 4 instances; the correct unit is L.
The text should have been presented as follows: Briefly, 100 L plasma was mixed with 100 L ethanol; after
being mixed by vortex, tocopherol was extracted into 500 L hexane containing 0.002% butylated hydroxy-
toluene (Sigma, St Louis). Tocol, a gift from Hoffmann-La Roche (Nutley, NJ), was added to the mixture as an
internal standard. Samples were centrifuged at 180 g for 5 min at 4 C. The supernatant fluid was collected
and dried under a stream of nitrogen gas and was reconstituted in 100 L methanol.