SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Subjects and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Oxidative stress may contribute to the pathogenesis of amyotrophic lateral sclerosis (ALS). We therefore examined prospectively whether individuals who regularly use supplements of the antioxidant vitamins E and C have a lower risk of ALS than nonusers. The study population comprised 957,740 individuals 30 years of age or older participating in the American Cancer Society's Cancer Prevention Study II. Information on vitamin use was collected at time of recruitment in 1982; participants then were followed up for ALS deaths from 1989 through 1998 via linkage with the National Death Index. During the follow-up, we documented 525 deaths from ALS. Regular use of vitamin E supplements was associated with a lower risk of dying of ALS. The age- and smoking-adjusted relative risk was 0.99 (95% confidence interval [CI], 0.69–1.41) among occasional users, 0.59 (95% CI, 0.36–0.96) in regular users for less than 10 years, and 0.38 (95% CI, 0.16–0.92) in regular users for 10 years or more as compared with nonusers of vitamin E (p for trend = 0.004). In contrast, no significant associations were found for use of vitamin C or multivitamins. These results suggest that vitamin E supplementation could have a role in ALS prevention. Ann Neurol 2004

An important role of oxidative stress in the pathogenesis of amyotrophic lateral sclerosis (ALS) is supported by the presence of biomarkers of increased oxidative damage in sporadic ALS patients1–6 and by the fact that several distinct mutations in the Cu-Zn superoxide dismutase (SOD1) gene, which is a critical component of cellular antioxidant defense mechanisms, cause an autosomal dominant form of ALS.3, 7, 8 Furthermore, supplementation with vitamin E, which is the major cellular antioxidant defense against lipid peroxidation, significantly delayed the onset of clinical disease in mice carrying one of these human SOD1 mutants.9, 10 Whether or not dietary supplementation with vitamin E reduces the risk of ALS is not known. A lack of significant associations between intakes of antioxidant vitamins and ALS risk was reported in two case–control studies, but both were too small to estimate the effects of use of specific vitamin E supplements.11, 12

We therefore have examined prospectively whether individuals who regularly use supplements of the antioxidant vitamins E or C have a lower risk of developing ALS. The study was conducted among participants in the Cancer Prevention Study–II (CPS-II).13

Subjects and Methods

  1. Top of page
  2. Abstract
  3. Subjects and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Study Population

The CPS II participants were recruited in 1982 by American Cancer Society volunteers in 50 states, the District of Columbia, and Puerto Rico.13 Families with at least one member older than age 45 years were eligible for the study; in eligible families, all family members older than age 30 years were invited to participate. The median age at cohort entry in 1982 was 57 years for men and 56 years for women. In total, 508,334 men and 676,288 women, including 448,590 married couples, completed a four-page questionnaire. Because deaths from ALS before January 1, 1989 were not coded separately, we excluded follow-up time (and 85,896 deaths) before this date. In the baseline self-administered questionnaire, participants were asked whether they had a diagnosis of selected diseases (not including ALS) or “any other serious disease.” We excluded 44,218 men and 65,836 women who reported “any other serious disease” (approximately 11% of the cohort) because they may have had ALS at the time of completing the baseline questionnaire. Furthermore, as in previous studies,14 we excluded individuals who reported identical frequency and duration for all vitamin supplements (13,146 men and 17,786 women), because it is likely that these participants were reporting components of their multivitamins rather than use of individual vitamin supplements. Thus, the final population for analyses included a total of 957,740 individuals (555,744 women), who were followed up for ALS mortality from January 1, 1989 to December 31, 1998.

Case Ascertainment

Vital status of the study participants has been determined by automated linkage with the National Death Index (NDI) through December 31, 1998.15 Death certificates or codes for cause of death have been obtained for more than 98% of known deaths. The underlying and contributing causes of death were coded according to the 9th revision of the International Classification of Disease (ICD)–9. All individuals with ICD-9 code 335.2 (motor neuron disease) listed as either the underlying or a contributing cause of death were considered as cases.16 As expected, ALS mortality in this cohort was higher in men than in women and increased with age.16

Assessment of Exposure and Other Covariates

The 1982 questionnaire included questions about current use of vitamin A, vitamin C, vitamin E, and multivitamins. Participants were asked to report how many times they used each supplement in the past month, and the number of years of use. Participants who used a vitamin only occasionally were instructed to report times per month as “1/2.”14 For each vitamin supplement, we categorized participants as nonusers, occasional users (<15 times per month), regular users (≥15 times per month) for less than 10 years, regular users for 10 years or longer, and missing or incomplete data. These categories were chosen a priori before the beginning of this study. Individuals who left blank the frequency of use of one or more (but not all) vitamins were considered as nonusers of that vitamin, whereas those who left the entire vitamin and medication section blank or reported regular use of a vitamin but no duration were classified as missing or incomplete data. A subset of 184,190 participants in the CPS-II cohort joined the CPS-II Nutrition cohort by completing a follow-up questionnaire in 1992.17 Data from this subset were used to estimate the consistency of vitamin use over 10 years. At baseline, all participants were also asked to report several demographic and lifestyle variables and the frequency of consumption of 32 common food items.18

Statistical Analyses

Participants contributed follow-up time from January 1, 1989 to the date of death or December 31, 1998, whichever came first. Age-adjusted (in 5-year age groups) relative risks (RR) were calculated by dividing the ALS mortality rate among participants in each category of vitamin use by the corresponding mortality rate in nonusers, using Mantel–Haenszel weights.19 We used Cox proportional hazards regression to estimate RR and 95% confidence intervals (CIs) when adjusting simultaneously for use of different vitamin supplements and additional variables. To obtain a better age-adjustment, we stratified the Cox models by age in single years. Significance of trends was assessed by including vitamin use in the regression models as a continuous variable, with the following weight: 0 (nonusers), 1 (occasional use), 2 (regular use, <10 years), and 3 (regular use, 10 or more years). Interactions with age (below or above 65 at beginning of follow-up) and smoking (never or ever) were entered as multiplicative terms in the Cox models, and their significance was ascertained based on the likelihood ratio test. Deviations from the proportional hazard assumptions were tested by examining whether there was a significant interaction between time of follow-up (0–5 years vs >5 years) and the exposure of interest. SAS version 8.2 was used for all analyses.

Results

  1. Top of page
  2. Abstract
  3. Subjects and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

During the follow-up, 525 participants (277 men and 248 women) died of ALS; the median age at death was 71 years in men and 72 in women. Individuals who reported regular long-term use of vitamin E supplements were less likely to be smokers, exercised more, were more educated, and consumed more multivitamins, other vitamin supplements, fruit and vegetables, and high-fiber cereals than nonusers; furthermore, women were more likely to have used estrogen replacement therapy (Table 1). These variables, some of which have been related to risk of ALS in previous studies,12, 20 therefore were considered in the analyses as potential confounders. Furthermore, as estimated from participants in the CPS-II Nutrition cohort, most regular users of vitamin E in 1982, but only a small proportion of never users, reported regular use of this vitamin in 1992 (see Table 1). Long-term users of vitamin E supplements had slightly lower all causes mortality than that of nonusers (RR adjusted for age and smoking = 0.90).

Table 1. Baseline (1982) Characteristicsa Among Men and Women According to Vitamin E Use
CharacteristicMen (N = 401,996)Women (N = 555,744)
NoneRegularb <10 yrRegularb ≥10 yrNoneRegularb <10 yrRegularb ≥10 yr
  • a

    All variables, except age, are age-adjusted by direct standardization to the entire cohort.

  • b

    Fifteen or more times per month.

  • c

    Based on a subset of 86,404 men and 97,786 women who completed a follow-up questionnaire.

Participants (%)78.24.32.275.76.02.1
Age (mean)63.164.366.562.762.265.3
Race, white (%)949697939697
Current smokers (%)211617211718
Past smokers (%)303230202527
Body mass index (mean, kg/m2)26.025.825.624.824.023.9
Heavy physical activity (%)121213667
Alcohol (mean, gm/day)17.517.819.28.69.910.4
Education      
 Some high school or less (%)16111014109
 High school grad (%)211815313028
 Vocational/Trade (%)676666
 Some college (%)202224232729
 College graduate or higher (%)374246252728
Married (%)959493777878
Ever estrogen replacement (%)34.844.045.1
Vitamin C use      
 Regular, <10 yrs (%)3.258.22.33.853.23.4
 Regular, ≥10 yrs (%)1.88.775.61.99.676.3
Vitamin A use      
 Regular, <10 yrs (%)0.217.22.80.316.44.3
 Regular, ≥10 yrs (%)0.01.323.30.11.323.7
Multivitamin use      
 Regular, <10 yrs (%)6.137.73.37.836.34.4
 Regular, ≥10 yrs (%)5.513.253.66.416.556.6
Diet      
 Fat index (mean, gm/day)403939353535
 Vegetables and fruit (mean, servings/day)2.02.32.42.22.52.6
High-fiber cereals (mean, servings per day)0.550.710.760.550.710.79
Regular use of vitamin E in 1992 (%)c85069115169

ALS mortality was 62% lower among long-term users of vitamin E than among nonusers (Table 2). In contrast, no significant associations were found for use of vitamin C or multivitamin supplements (see Table 2). Further adjustment for education (categories as in Table 1), level of physical activity (none, slight, moderate, heavy), alcohol consumption (quintiles), servings of fruits and vegetables and high-fiber cereals (quintiles), or, among women, use of estrogen replacement therapy (never, ever), did not materially change the results. The inverse association between use of vitamin E intake and ALS mortality was similar in men (RR for regular users vs nonusers: <10 years. 0.68; ≥10 years, 0.42) and women (<10 years, 0.51; ≥10 years, 0.33). No significant interactions were found between use of vitamin E supplements and smoking history (ever vs never smoker), age at beginning of follow-up (below or above 65 years), or follow-up time (0–5 years or >5 years).

Table 2. Relative Risk of ALS Death according to Use of Vitamin Supplements
SupplementsPersonsCasesRate × 105 Person-yearsRRaRRb (95% CI)
  • a

    Adjusted for age (5-year categories).

  • b

    Adjusted for age (1-year categories), smoking (never, past, current; cigarettes per day: 1–14, 15–24, 25 or more), and sex.

  • ALS = amyotrophic lateral sclerosis;

  • RR = relative risk; CI = confidence interval.

Vitamin E     
 No use735,3864146.41.0 (ref)1.0 (ref)
 Occasional58,218336.40.990.99 (0.69–1.41)
 15+ days/mo, <10 yr50,626173.80.590.59 (0.36–0.96)
 15+ days/mo, ≥10 yr20,28352.80.380.38 (0.16–0.92)
 p, trend    0.004
 Missing93,227567.01.011.00 (0.76–1.32)
Vitamin C     
 No use671,8643666.21.0 (ref)1.0 (ref)
 Occasional93,288587.01.171.16 (0.88–1.54)
 15+ days/mo, <10 yr59,839336.21.011.02 (0.72–1.46)
 15+ days/mo, ≥10 yr37,446185.40.810.81 (0.50–1.30)
 p, trend    0.75
 Missing95,303506.10.910.90 (0.67–1.22)
Multivitamin     
 No use596,7563346.41.0 (ref)1.0 (ref)
 Occasional108,740575.91.01.01 (0.7–1.34)
 15+ days/mo, <10 yr79,988496.91.161.19 (0.88–1.61)
 15+ days/mo, ≥10 yr71,126326.10.780.79 (0.55–1.13)
 p, trend    0.61
 Missing101,130536.10.910.91 (0.68–1.22)

Because of the strong correlation between use of vitamin E supplements and use of other vitamins, we also conducted analyses adjusting simultaneously for use of each vitamin supplement (including vitamin A). As shown in the Figure, in these analyses there was still a strong inverse association between vitamin E use and ALS mortality. In contrast, use of other vitamin supplements was not associated with ALS mortality (see Fig 1).

thumbnail image

Illustration 1. Fig. Relative risk (and 95% CI) of ALS death according to use of vitamin E, vitamin C, and multivitamin supplement, adjusted for age (1 year categories), smoking (never, ever, current: 1–4, 15–24, >=25 cigarettes/day), gender, and frequency and duration of use of vitamin E, vitamin C, multivitamin, and vitamin A supplements.

Download figure to PowerPoint

The numbers of individuals who reported regular use of vitamin E without other supplements were too low for meaningful analyses. We could, however, examine the association between regular use of multivitamin or vitamin C without regular use of vitamin E, because these patterns of use were relatively common. Compared with participants who used neither multivitamins nor vitamin E, participants who were regular users of multivitamins but did not use vitamin E were not at reduced risk of ALS (RR, 1.1; 95% CI, 0.85–1.43), whereas participants who were regular users of both multivitamins and vitamin E were at reduced risk (RR, 0.48; 95% CI, 0.27–0.88). Similarly, compared with participants who used neither vitamin C nor vitamin E, participants who were regular users of vitamin C but did not use vitamin E were not at reduced risk (RR, 1.24; 95% CI, 0.87–1.76), whereas those who were regular users of both vitamin C and vitamin E were at reduced risk (RR, 0.59; 95% CI, 0.36–0.97). Use of vitamin A supplements (mostly retinyl esters), which are not antioxidants, was not related to ALS mortality (RR for regular long-term users vs nonusers, 0.73; 95% CI, 0.23–2.26).

Discussion

  1. Top of page
  2. Abstract
  3. Subjects and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In this large prospective cohort, we found that individuals who regularly used vitamin E supplements for 10 or more years had less than half the risk of death from ALS than nonusers. In contrast, use of vitamin C or multivitamins was not associated with ALS risk.

A limitation of this study is that use of vitamin E and other supplements was only assessed in 1982. As estimated from a subset of participants, however, a majority of regular vitamin E users in 1982 were still using these supplements 10 years later. Furthermore, because changes in vitamin E use during the follow-up would most likely attenuate the association between use of vitamin E and ALS, they cannot explain our results. Most users of vitamin E supplements also used vitamin C or multivitamins. Although neither of these supplements when used without vitamin E was associated with a reduced risk of ALS, it remains possible that a combination of vitamin E and C or vitamin E and multivitamin supplements is more effective than vitamin E alone. An additional limitation is that we had data only on ALS mortality, not incidence. Nevertheless, because vitamin E supplementation did not prolong survival of ALS patients in a large trial,21 the lower mortality among vitamin E users that we observed is most likely caused by a reduction in ALS incidence rather than an improved prognosis. The diagnoses of ALS were based on death certificates, which have been found to accurately identify only 70 to 90% of ALS or motor neuron disease cases.22–25 However, this would not materially bias the relative risk estimates unless the misclassification of the cause of death was strongly related to vitamin E use. Finally, as in all observational studies, we cannot exclude confounding by unmeasured factors.

Vitamin E (α-tocopherol) is a lipid soluble, highly efficient antioxidant, which is the primary cellular defense against membrane lipid peroxidation. The average daily dietary intake of vitamin E in the United States is between 8 and 11 α-tocopherol equivalents (12–16IU), which is close to the recommended dietary allowance.26 Most multivitamins contain 15 to 30mg of synthetic vitamin E (all-rac-α-tocopheryl acetate, equivalent to 7.5–15mg of RRR-α-tocopherol, or natural vitamin E), an amount that would have only a modest effect on tissue levels.27 In contrast, specific supplements typically contain 400mg of synthetic vitamin E, an amount that upon long-term daily supplementation increases plasma α-tocopherol concentrations threefold and doubles concentrations in the central nervous system.27 Our finding of a decreased risk of ALS among users of individual vitamin E supplements, but not among users of multivitamin supplements, thus is consistent with these findings. Vitamin C also has antioxidant properties. In a mouse model of familial ALS, vitamin C supplementation before disease onset did not delay the age at onset but significantly slowed the progression of paralysis.28 The lack of association between vitamin C supplements and ALS mortality in our study suggests that vitamin C alone may be ineffective in preventing ALS. Vitamin C is a water-soluble antioxidant with different properties than vitamin E and thus may not share the same biological effects in vivo.29

In a randomized trial, supplementation with 1,000IU/day of vitamin E for 1 year had a marginal beneficial effect on disease progression but did not prolong survival.21 The failure of vitamin E to prolong survival of ALS patients, however, does not exclude a beneficial effect of vitamin E supplementation, which in animal experiments was found to delay the disease onset but not its progression.10

Because SOD1 mutations account for only approximately 1 and 2% of all cases of ALS,7, 8 only a few of the cases included in our investigation are likely to have harbored a SOD1 mutation and most were sporadic (information on family history of ALS was not available). The role of oxidative stress in sporadic ALS is uncertain, but markers of increased oxidation have been reported in several studies.4, 6, 30–32 Furthermore, signs of increased compensatory response to oxidative stress3 have been reported in patients with sporadic ALS. These changes, however, could be a consequence rather than the cause of the neurodegenerative process.

Oxidative stress appears to act in concert with other mechanisms that have been implicated in the pathogenesis of motor neuron degeneration in ALS.33 Thus, oxidative reactions increase the susceptibility of motor neurons to excitotoxicity by disrupting glial glutamate transport,34–37 modulate immune activation,33 affect mitochondrial function, and promote protein aggregation.38 Vitamin E, by reducing oxidative stress, therefore could influence several downstream events that result in the death of motor neurons. Finally, in addition to being an antioxidant, vitamin E reduces platelet aggregation39 and smooth muscle cell proliferation.40 These mechanisms are likely to be beneficial in ischemic cardiovascular diseases and could contribute to a possible protective effect against ALS. The recent finding that a haplotype in the promoter region of the human vascular endothelial growth factor (VEGF) gene is associated with lower circulating levels of VEGF and with a highly significant increase in ALS risk41 provides a link between vascular perfusion and ALS.

The finding in this large prospective study that long-term users of vitamin E supplements had less than half the risk of dying from ALS than nonusers suggests that vitamin E supplementation may have a role in ALS prevention. Future studies will need to confirm this result. Because in our study most ALS cases were likely to be sporadic, whether or not a similar association exists for familial ALS remains to be determined.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Subjects and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported by a research grant from the NIH (Natural Institute of Neurological Diseases and Stroke, NS045893, A.A.).

References

  1. Top of page
  2. Abstract
  3. Subjects and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References