Effect of vitamin e supplementation with exercise on cognitive functions and total antioxidant capacity in older people

Abstract

Objective

We investigated the effects of six months vitamin E administration on cognition evaluated by event-related potentials in exercising older subjects.

Design

Randomised controlled trial.

Setting

Retirement home in Antalya, Turkey.

Participants

Fifty-seven adults aged 60–85 years were randomly assigned to one of four groups: sedentary control (C), vitamin E (V), exercise training (E) and vitamin E under training (EV).

Intervention

V and EV groups were received vitamin E at a dose of 900 IU/day P.O. for 6 months. Trained groups were subjected to walking exercise involved 3 sessions per week for 6 months. Walking duration was gradually increased during 8 weeks, and stayed constant until the end of training period. Participants were begun walking at % 70 heart rate reserve for 20 min/day at the first two weeks, and walking duration was increased by 5 minutes/day of each week until subjects were reached a level of 50 min/day by week 8. Measurements Plasma vitamin E concentration, total antioxidant capacity and two parameters of event-related potentials namely P3 latency and amplitude were performed on all study groups both before and after training.

Results

Significant improvement in P3 latency was found in exercising groups. However, no significant differences were found between vitamin and other groups for P3 latency. Amplitude measurements were found unaltered among all groups.

Conclusion

We concluded that although six months training results improvement in P3 latency, vitamin E supplementation does not affect cognitive function evaluated by event-related potentials in older subjects.

Introduction

Physical exercise is an inexpensive treatment that could have substantial preventative and restorative properties for cognitive and brain function especially in older subjects (1, 2). In the first well-controlled training study of exercise and cognition, Dustman et al. found improvement in performance on some cognitive tasks are attenuated in subjects with high- compared to low-physical exercise (3). A previous study of 59 sedentary adults aged 60–79 years has shown that six months of aerobic training increased brain volume in regions of the brain associated with age-related decline in cognition, compared with non-aerobic training (4). Numerous observational studies have found that people who are physically active seem less likely than sedentary persons to experience cognitive decline and dementia in later life (5, 6). Weuve et al. reported that higher levels of physical activity over 2 years among the 18766 women in the Nurses' Health Study were associated with improved cognitive scores (7). Although fitness training broadly influenced a variety of cognitive processes, the largest positive effects were observed for executive control processes which include components of cognition such as planning, scheduling, working memory, inhibitory processes, and multitasking (8).

Electrophysiological studies using event-related brain potentials (ERPs) have been conducted during the last decade, and they have resulted in a deeper understanding of the relationship between exercise and cognitive functions. The advantage of the ERP approach is that it can provide information regarding the discrete cognitive processes between stimulus evaluation and response execution that are selectively influenced by exercise (9). ERPs are electrical potentials elicited by a series of repeated stimuli such as flashes of light, clicks, or tones that the subject expects to fail. Because ERPs reflect processes that occur in the CNS (central nervous system) between the stimulus and the response, they can provide information about the time course of cognitive processing in the brain. Correlation of latency of the P3 component of cognitive event-related potentials with cognitive dysfunction appears to be well established; the P3 latency is correlated negatively with mental function in normal subjects. Therefore, shorter latencies are related to superior cognitive performances and vice versa; that is, latency increments are observed as cognitive capability decreases because of aging (10). Moreover, P300 latency is reported to exhibit good correlation with the neuropsychological tests conventionally used to assess cognitive performance such as the Wechsler memory scale (11). Despite the implications of the finding that exercise can contribute to cognitive capability, surprisingly little is known about the mechanisms by which exercise or physical activity might influence cognition.

Oxidative mechanisms may play important roles in cellular processes associated with aging, and in neurodegenerative diseases, such as Alzheimer disease and Parkinson disease (12, 13, 14). The brain is especially vulnerable to free radical damage because of its high oxygen consumption rate, its abundance of easily peroxidized lipid membranes, and the presence of relatively few antioxidant enzymes (15). Oxidative reactions induced by ROS (reactive oxygen species) are thought to cause the degeneration of neurons.

Vitamin E, the major lipophilic antioxidant in humans, is a physiological scavenger of ROS produced during lipid peroxidation (16, 17). Low levels of vitamin E have been associated with cognitive impairment (18) and poor lower extremity muscle strength (19, 20). However, only a few studies have examined the interaction between vitamin E and exercise in older humans (21, 22, 23). To define a relationship between vitamin E administration and cognitive performance, it is important to consider the changes in vitamin E status. Ortega et al. investigated the effect of vitamin E status on cognitive function in older people aged 65-91 years with free of significant cognitive impairement. They tested cognitive capacity using by Pfeiffer's Mental Status Questionnaire (PMSQ). They found that a relationship exists between vitamin E status and cognitive function, although they do not provide proof of a causative effect (24). Morris et al. described the association between dietary antioxidants and 3-year change in cognitive function in a longitudinal population-based study conducted from 1993 to 2000. They found a 36 % reduction in the rate of cognitive decline among persons in the highest quintile of total vitamin E intake compared with those in the lowest quintile (25).

Vitamin E might help in the prevention of cognitive decline through its powerful antioxidant action, which provides protection against nervous tissue damage, but also through its role in the prevention of vascular damentia, a disorder responsible for cognitive decline in many people (12). Foy et al. have found lower blood concentrations of vitamin E in these patients than in control subjects, but not at the level of nutritional deficiency (26). Thus, lower antioxidant status in patients with dementia may indicate an increased level of oxidative stres, an important factor in the impairment of cognitive function.

The aim of this study was to investigate the effect of vitamin E supplementation on cognitive preformance in exercising older people. We hypothesized that supplementation with vitamin E would augment the increased activity of the antioxidant mechanisms resulting from aerobic exercise training and would increase cognitive function, which declines with age in sedentary older adults.

Materials and methods

Participants

One hundred twenty elderly subjects living in a retirement home were assessed for eligibility. Forty-one subjects (26 women and 15 men) did not meet the inclusion criteria, and 22 (6 women and 16 men) refused to participate. Fifty-seven sedentary adults (11 women and 46 men) were included in the trial. The inclusion criteria were standardized Mini-Mental State Examination score >24, nonsmoker, nonuser of alcohol, without medical illness, not receiving medication or vitamins, sedentary, and willing to participate. The exclusion criteria were sudden illness, leaving the retirement home, and unwillingness to continue the program. The participants were informed that they would be allocated randomly to either an exercise group or a control (nonexercise) group. The subjects provided informed consent, and a standardized individual information session was organized to allow a dietitian to instruct the participants on how to record daily food intake.

Participants were stratified by sex and then randomly assigned to a control group (n=28; 5 women, 23 men) and an exercise group (n=29; 6 women, 23 men) one week before baseline measurements (Figure 1). Individuals assigned to take vitamin E were randomly assigned to one of two groups; 14 to a control group, whose members did not exercise but who took vitamin E (V group; 3 women, 11 men), and 14 to an exercise group, whose members took vitamin E (EV group; 3 women, 11 men). Fourteen subjects were assigned to the non-vitamin control group, whose members did not exercise (C group; 2 women, 12 men), and 15 to the exercise group, who exercised but did not receive vitamin E (E group, 3 women, 12 men). Six subjects in the exercise group and one in the control group were excluded during measurements after they declined to continue, four of these were in the EV group (one because of surgery, two who were unwilling to continue, and one for no given reason), two in the E group (both for no given reason) and one in the C group (who left the retirement home). In the first five weeks of the training period, seven more participants withdrew from the study: two subjects from the E group (one had a transient ischemic attack, one was unwilling to continue); four from the V group (two left the retirement home, one sustained a fracture, one for no given reason), and one from the C group (myocardial infarction). The final analysis included the 43 participants who completed the exercise training program: 12 in the C group, 10 in the V group, 11 in the E group, and 10 in the EV group. The study was approved by the Republic of Turkey Ministry of Health, General Directorate of Drug and Pharmacy (approval number 013551/2005) and Akdeniz University School of Medicine Ethical Committee of Drug Research (approval number 33/2003). Participants were all asked to maintain their normal behavior during the six-month follow- up, and the project supervisor followed them.

Figure 1.

Procedure for randomized clinical trial

Assessment

Height was measured using an ultrasonic height measure (Soehnle-Waagen GmbH & Co. KG). Body weight was measured with a Tanita Body Composition Analyzer (Model TBF-300 TANITA, Tokyo, Japan). Body mass index (BMI) was calculated from weight and height.

General cognitive function was measured using the modified Turkish version of the Standardized Modified Mini-Mental State Examination (SMMMSE) (27). Physical activity level was assessed by the Turkish version of the Modified Baecke Questionnaire (28) and a score <9 in the questionnaire was used as the definition of physical inactivity. To determine whether nutritional intake was different between groups, 3-day dietary record forms were provided to each participant with standard instructions on how to complete the record before visit. Participants were instructed to estimate servings of foods using household measurements (volume) as described in national dietary guidance documents as previously described (29). Each participant received individual training sessions with the same investigator with regard to measuring technique and volume estimation. Picture books of portion sizes were also provided after the dietary estimation training session (29). Diet records were assessed by the same investigator using the Bebis program (Ebispro for Windows, Stuttgart, Germany; Turkish version Bebis, Version 5, 2006; Data bases: Bundeslebenmittelschlüssell, 11.3).

Vitamin E dose

Vitamin E was given to each subject by nurses at the retirement home. The dose was one 300-mg capsule (Ephynal soft-gel 300-mg capsules, Roche Pharmaceuticals, France; 300 IU vitamin E equivalent to 300 mg DL-a-tocopherol acetate per capsule) after each meal, giving a total dose of 900 mg/day/subject, for six months. At the beginning of each month, the project supervisor gave the nurses packages of vitamin E capsules. In the first month, after randomization and before the measurements, 84 packages were given (total of 2520 capsules or 90 capsules for each of the 28 subjects). The names of the subjects were written on the packages. In the second month, 66 packages were given, in the third month, 60 packages were given, and this number was maintained each month until the end of the study. Each week, the project supervisor interviewed the nurses and the subjects to check that the participants had taken the vitamin E and to record the occurrence and frequency of any side effects.

Vitamin E concentration and total antioxidant capacity (TAC)

Venous blood was sampled from an antecubital vein into a Vacutainer after a 12-h overnight fast. Plasma vitamin E concentration and total antioxidant capacity (TAC) were evaluated at baseline and after six months. Plasma vitamin E concentration was measured as described by Desai (30). Briefly, 1 mL of absolute ethanol and 500 µl of tocopherol were added to 500 µl of plasma. After thorough mixing, the tubes were incubated at 70 °C for 5 min, 1 mL of saturated KOH was added, and the tubes were incubated at 70 °C for 30 min. The tubes were cooled on ice, and 3 mL of hexane was added. After thorough vortexing for 5 min, the samples were centrifuged at 2500 × g for 10 min. Vitamin E concentration was estimated spectrofluorometrically (PerkinElmer, LS50-B, Norwalk, CT) using 286 nm and 330 nm as the excitation and emission wavelengths, respectively. A standard curve was obtained using 1–10 µg of a-tocopherol/mL in absolute ethanol. The concentration of vitamin E in plasma is expressed in mg/dL.

A colorimetric assay kit (TAC assay kit, NX 2332, Randox Laboratories, Crumlin Co., Antrim, UK) was used to measure the resting TAC in plasma. The Randox total antioxidant capacity assay is based on the reaction between metmyoglobin and hydrogen peroxide, which generates a free radical and was developed in direct response to increased interest concerning the role of free radicals in disease. The kinetics was measured at 600 nm automatically. TAC in plasma is expressed as mmol/L.

ERP recordings

Ag-AgCl electrodes were placed at Fz and Cz actively according to the international 10-20 system, referenced to linked earlobe electrode, with a forehead electrode as a ground electrode. Impedances were maintained below 5kO and were measured from each lead at the beginning and end of the each session. P3 potentials were recorded with a Nihon-Kohden® Neuropack® 8 EMG/Evoked response measuring system, MEB-4200K with a band-pass of 0.1-50 Hz. Recordings were made for 1000 ms, beginning 100 ms prior to stimulus onset with an amplification of 50 µV / unit sensitivity. P3 potentials were obtained from an auditory oddball paradigm. The 2 kHz target tones were presented with a probability of 20 %, while 1 kHz non-target tones were presented with a probability of 80 % binaurally over headphones at an intensity of 90 dB sound level. Tone bursts were presented with an inter-stimulus interval of 2 seconds with 10 ms rise / fall times. Rare tones were presented randomly during the task not being presented more than three times consecutively. Responses of targets and non-targets were averaged separately. At the end of the each session, the subjects count was compared with the actual number of target tones given, in order to assess accuracy of task performance. All subjects performed the tasks with a less than 5 % errors rate in all trials.

P3 latencies and amplitudes were measured from peak to peak points of N2 and P3 waves at Cz and Fz recordings of subjects. The P3 potentials obtained from target tones were evaluated. Principal peaks and their identification were made according to the standard recommendations for long latency auditory event related potentials of The International Federation of Clinical Neurophysiology and principal component analysis technique (31). The neurophysiological values were averaged across Cz and Fz electrode positions in all subjects.

Exercise training program

All exercise sessions were supervised. Heart rate was monitored and recorded for each subject using a heart rate monitor (Sport Tester PE 300, Helsinki, Finland), and arterial blood pressure was measured with a sphygmomanometer. Trained supervisors were recorded all heart rate and blood pressure values for every participant for the duration of every training session. Participants were encouraged to maintain exercise intensity whenever possible to ensure that heart rate was within the target range throughout the six months.

Subjects participated in a walking program three times per week between 08:30 and 10:00 h for six months (between March 17 and September 13). Each training session began with a 10-min warm-up and ended with a 15-min cool-down, which both included slow walking followed by slow static stretching. Each session was led by trained fitness instructors and supervised by the main researcher. The control groups (V and C) did not train and participated only in the measurement procedures.

The target heart rate range was established for each participant before and during the training program using the Karvonen method (32).The ultimate goal was for training intensity to be higher than 60% of heart rate reserve but not to exceed 70% of heart rate reserve. We selected the intensity, frequency, and duration of each training session to improve cardiovascular endurance and physical performance, as recommended for adults over age 65 years by the American College of Sports Medicine (33). We limited the intensity of training to not above 70% of heart rate reserve because it is thought that exercise intensity above 70% of maximum oxygen uptake may surpass the capacity of the endogenous antioxidant system and consequently may promote oxidative stress (34).

The training duration was 20 min in the first two weeks. After the first two weeks, the duration of training was increased by 5 min every week until week 8 and then was maintained at 50 min until the end of the study.

Statistical analysis

Data are expressed as means ± S.D., and the statistical significance of the data of the four groups were assessed by repeated measures of analysis of variance (ANOVA), and post-hoc Tukey test. Pre- and post-test measurements were assessed by t- test. A level of p<0.05 was accepted as statistically significant.

Results

The baseline demographic, cognitive, physical function and physical activity characteristics and vitamin E intakes of the groups are presented in Table 1. The groups were not different in terms of their pretest body mass index, SMMMSE, physical activity score and functional fitness test scores and daily vitamine E intake. All the participants were classified as having low levels of physical activity.

Table 1.

Descriptive information for the randomised study groups

	C (n= 12)	E (n = 11)	V (n=10)	EV (n=10)
Age (years)	71 ±9.1	69.6 ±8.6	73.1 ±4.5	72.8 ±7.1
Body Mass Index (kg/m2)	23.8 ±4.5		27.4 ±4.1	30.7 ±6
Education (years)	7.42 ±2.84	7 09 ± 3 21	9.0 ±3.92	9.90 ±3.38
Standardized modified mini mental state examination (score)	27.8 ±3.2	26.7 ±3.4	27.5 ±3.1	28.2 ±1.6
Physical activity score (points)	5 ± 3	65 ± 46	4.2 ±2.4	4.4 ±2.6
Daily vitamine E intake (mg)	175.6 ±52.2	166 ±53.4	166.5 ±53.3	174 ±41.8

All the participants in the training group completed the 6 months aerobic training successfully. None experienced angina or arrythmias during the tests and training program.

Table 2 shows plasma vitamin E concentrations of each group. As shown in the table, plasma vitamin E levels were increased in the V group with respect to the C group (p=0.045). On the other hand, plasma vitamin E concentrations were increased in E (p=0.05), V (p=0.012) and EV (p=0.004) groups compared to the pretest measurements.

Table 2.

Plasma vitamin E concentrations and total antioxidant capacity of the four groups

	C		E		V		EV
	Pretest	Post-test	Pretest	Post-test	Pretest	Post-test	Pretest	Post-test
Plasma Vitamin E	2,002±0,30	2,262±0,33	1,862±0,70	2,506±0,94∗	2,090±0,43	3,742±1,59∗∗#	1,805±0,46	3,631±1,20∗∗∗##
Concentration (mg.dl-1)
Plasma Total Antioxidant	1,613±0,23	1,612±0,28	1,623±0,19	1,668±0,34	1,622±0,29	1,522±0,29	1,939±0,41	1,548±0,51
Capacity (mmol.l-1)

Values are mean ± SD. C; sedentary control group, E; exercise training group, V; vitamin E group (900 IU/day) and EV; vitamin E under training group. ∗ p=0.05, ∗∗ p=0.012, ∗∗∗ p=0.004; significant pre-test difference. # p=0.045, ## p=0.032; significant difference vs C group.

Plasma total antioxidant capacity of the four groups are presented in the Table 2. Neither exercise, nor vitamin E administration were resulted an alteration on plasma TAC.

P3 latencies of Fz and Cz sites for all groups are presented in Table 3. Although baseline P3 latencies were similar in all groups, shorter P3 latencies in E group both at the Fz (p=0.018) and Cz (p=0.010) sites, and in EV group at Fz (p=0.007) and Cz (p=0.023) sites were recorded compared to the pretest measurements. On the other hand, in the EV group, the mean latencies of P3 at Fz (p=0.037) and Cz (p=0.000) were reduced with respect to the C group. Therefore, we concluded that exercise training was resulted shortened mean latencies of P3 component instead of vitamin E usage in older people.

Table 3.

Mean P3 latencies and amplitude values of ERPs recorded from Fz and Cz sites of the four groups

	C				E				V				EV
	Prtes		Post-test		Pretest		Post-test		Pretest		Post-test		Pretest		Post-test
	Fz	Cz	Fz	Cz	Fz	Cz	Fz	Cz	Fz	Cz	Fz	Cz	Fz	Cz	Fz	Cz
P 3 latency (ms)	433,3± 15,28	433,8 ± 14,68	406,5 ± 62,45	433,3 ± 15,28	419,7 ± 42,18	416,3± 25,56	347,8 ± 24,83 ∗#	347,5 ± 25,07 ∗∗##	407,2 ± 35,37	407,2 ± 35,38	397,0 ± 22,09	384,2± 26,24 ∗∗∗ ###	405,5 ± 17,12	406,0 ± 30,86	343,0 ± 28,43	343,0± 28,43
P3amplitude (μV)	11,76 ± 4,40	13,25 ± 2,85	14,09 ± 4,95	15,90± 4,80	11,69 ±2,82	12,10±3,37	11,00 ± 4,52	13,25±5,10	14,97 ± 3,81	13,43 ± 3,39	13,00± 3,59	12,44 ±2,30	14,87± 8,15	15,80 ±7,06	12,97± 9,09	#### 13,67± 8,13

Values are mean ± SD. C; sedentary control group, E; exercise training group, V; vitamin E group (900 IU/day) and EV; vitamin E under training group. ∗ p=0.018, ∗∗ p=0.010, ∗∗∗p=0.027, ∗∗∗∗p=0.007, ∗∗∗∗∗p=0.023; significant pre-test difference. #p=0.037, ## p=0.002, ### p=0.024, #### p=0.0000; significant difference vs C group

Pre- and post-test measurements of the peak to peak amplitude values of all groups were unaltered (Table 3).

Discussion

This study evaluated the synergistic effect of vitamin E and six months of aerobic exercise training on cognitive function evaluated by event-related potentials in sedentary older adults. To our knowledge, this is the first study evaluating the effects of simultaneous antioxidant intake and aerobic exercise training on cognitive performance measures such as P3 latency and amplitude values in an elderly population.

Baseline plasma concentrations of alpha tocopherol in our study were too high compared with the findings of previous studies in elderly populations (35). This could be explained by the regional differences of dietary habbits in our subjects. Wittin et al. stated that it is difficult to set a rational RDA (recommended dietary allowance) for vitamin E and that in contrast to the RDA's for other vitamins, there can be no single recommended value for each age and sex group (36). The two groups that took vitamin E (EV and V groups) showed significant post-test increases in plasma vitamin E concentration (101% and 79%, respectively), whereas the E and C groups showed no significant change (34% and 13%, respectively). These findings indicate that the participants complied with the six-month vitamin E supplementation protocol.

We measured TAC to evaluate the change in antioxidant status after the six months of exercise training and vitamin E supplementation, but we found no effects of time or group. This contrasts with a previous study that found that four months of endurance training increased TAC significantly (37). The fact that we did not find a statistically significant increase during exercise in the V and EV groups suggests that although some components of TAC may have been improved with vitamin E supplementation, other components may have been down-regulated. Several methods have been developed to assess the TAC in human serum or plasma, and the total antioxidative serum capacity is not a simple sum of the activities of the various antioxidative substances. For example, Brennan et al. (38) demonstrated that 42 days of combined vitamin C and E supplementation significantly reduced erythrocyte superoxide dismutase and glutathione peroxidase in healthy adults by 35.7% and 22.3%, respectively. Alternatively, antioxidant mobilization in response to exercise is dynamic, and different TAC components like vitamin E might increase (39), whereas other components, such as glutathione, decrease (40), thereby minimizing the potential improvement in an overall TAC measurement. Further research is required to determine what specific enzymatic, dietary, and non-enzymatic antioxidant are acutely affected in overweight persons during exercise with and without supplementation. The lack of response of TAC may also be due to methodological issues such as the effect of dilution of the plasma samples (as instructed by the manufacturer) or the use of the inhibition percentage at a fixed time without considering the length of the inhibition time when using the Randox assay kit (41).

In our study, shortened P3 latency values were found in both exercise trained groups, at both Fz and Cz sites, while the vitamin E usage had no additive effect on P3 latency. P3 amplitude values were unaltered among four groups. Given the sensitivity of P3 measures (i.e., P3 latency being sensitive to perceptual/central processing and P3 amplitude reflecting attentional allocation to the stimuli in the service of memory updating), these data suggest that physical activity influences perceptual/central processing for older adults, and indeed does so in a reasonably challenging task known to reflect multiple components of executive control (1).

The mechanisms responsible for exercise training effects on cognitive function remain unclear. There are a several factors influenced the relationship between exercise training and cognition such as type and intensity of exercise training, and the use of diferential cognitive assessments. For example, aerobic exercise programs combined with strength and flexibility training regimens had a greater positive effect on cognition than aerobic components alone. We have previously shown that, although P3 latency and amplitude values remained unaltered, shorter N2 and P2 latencies and increased amplitude measurements at both Fz and Cz sites of event-related potentials were obtained in strength, compared to the endurance trained older people (42). Previous studies suggest that this effect may result from increases in insulin-like growth factor 1 (IGF-1), which is known to increase in response to strength training. IGF-1 is a neuroprotective factor involved in neuronal growth and differentiation (43). On the other hand, in a previous study by Kamijo et al. (44), the influence of exercise intensity on arousal level was investigated using contingent negative variation (CNV), which is a slow negative ERP shift reflective of arousal level (45). This study suggested that differences in exercise intensity influenced arousal level through changes in CNV amplitudes showing a curvilinear (inverted U shaped) behavior (43).

According to previous literature on the effects of antioxidants in Alzheimer Diasease (46), vitamin E might have slowed the progression of disturbances. Masaki et al. examined the associations of vitamin E and vitamin C supplement use in 1988 with cognitive function and dementing illness determined in 1991 to 1993 among elderly Japanese men participating in the Honolulu-Asia Aging Study (HAAS). In their study, cognitive performance was assessed by the Cognitive Abilities Screening Instrument. They demonstrated that vitamin E and C supplements might protect against vascular dementia and might improve cognitive function (47).

The present study is not intended to, and can not, draw definitive conclusions on the effect of vitamin E in Alzheimer Diasease, because Alzheimer Disease is a multifaceted progressive disorder characterized by both cognitive and behavioral dysfunctions (48). In our study, the participants were selected in older people whose SMMMS scores were 24 and higher.

Several limitations of our study should be mentioned. Study sample was relatively small, considering that four treatment groups were involved. The general overweight status of the participants may have influenced our findings. Another limitation of our study was its limited assessment of endogenous antioxidant capacity. Finally, giving the placebo capsules to the C and E groups would have increased the awareness of the participants.

Conclusion

This study showed that vitamin E supplementation has no additive effect to that of aerobic exercise training on cognitive performance measures such as P3 latency and amplitude measurements of event-related potentials. A causal interpretation of the observed findings would require additional evidence by other longitudinal epidemiologic studies and /or primary prevention trials.

Acknowledgements: This study was supported by Akdeniz University Research Foundation (2005.01.0122.001), and Turkish Regional League Against Rheumatism. We thank to ROCHE Pharmaceuticals as the donor of the vitamin E, the physiatrist and nurses at Old Retirement Home, Akdeniz University Hospital, and fitness instructors in the School of Physical Education and Sports, and a very special thanks go to all the participants who volunteered for this study.