DLM: Vitamin E, C, B-carotene Supplements and All-Cause Mortality (2007-2010)

Citation:
 
Study Design:
Class:
- Click here for explanation of classification scheme.
Quality Rating:
Research Purpose:

Meta-analysis of the dose-response relationship between vitamin E supplementation and total mortality by using data from randomized controlled trials.

Inclusion Criteria:

Studies that met the following criteria:

  • Random allocation of participants
  • Use of vitamin E supplementation alone or combined with other vitamins or minerals
  • Presence of a control or placebo group
  • Study sample limited to men or non-pregnant women
  • Duration of vitamin E supplementation and follow-up longer than one year
  • Occurrence of at least 10 deaths in the trial.
Exclusion Criteria:

Any study that did not meet the above inclusion criteria.

Of the 2,170 studies identified, 2,134 were excluded for following reasons:

  • No original data
  • Not randomized and controlled
  • No control group without vitamin E
  • Vitamin E administration or follow-up after less than one year
  • Trials in children or pregnant or lactating women
  • Secondary publications of included trials.

Of the remaining 36 studies, 17 were excluded for the following reasons:

  • Fewer than 10 deaths in the trial: 12 studies
  • No mortality data available: Three studies
  • Mortality data not separated from a composite end point: Two studies.
Description of Study Protocol:
  • Recruitment: Searches via PubMed 1966 to 2004. Medline, 1,677 studies identified; Cochrane, 1,666 studies identified; manual search, 38 studies identified
  • Design: Meta-analysis of randomized controlled trials
  • Blinding used: Every study evaluated was controlled. 17 of 19 studies were placebo-controlled; 15 of 19 were double-blind and one was single-blind.

Intervention

Study subjects were given various doses of vitamin E, with or without other vitamins or minerals. Of the 19 studies analyzed:

  • Nine studies gave no other vitamin or mineral supplements
  • Two studies gave vitamin C as the only additional supplement
  • Four studies gave vitamin C and beta carotene as the only other supplements
  • Two studies gave vitamin C, beta carotene and selenium
  • One study gave vitamin C, beta carotene, selenium and zinc
  • One study gave multi-vitamin and mineral tablets (exact type not specified).

Statistical Analysis

All analyses were conducted according to the intent-to-treat principle.

  • Analyses were done using S-PLUS 2000 (Mathsoft, Inc., Seattle, WA) and HLM 5 (Scientific Software International, Lincolnwood, IL).
  • Two-way analyses were used for trials with a factorial design (all participants receiving vitamin E compared with those not receiving it). A two-level hierarchical logistic regression model was used to evaluate the association between vitamin E treatment and all-cause mortality. 
  • First within study level: Authors specified a logistic model for the probability of death as a function of random assignment to vitamin E treatment in each trial
  • In the second between-study level: First authors introduced an indicator variable for high-dose vitamin E trials (at least 400 IU per day) to evaluate differences in effect of high- and low-dose vitamin E supplementation. For dose-response analyses, authors replaced the previous indicator variable with a quadratic-linear spline model for log-dose of vitamin E (more flexible and less influenced by extreme values and more plausible than the usual linear or quadratic polynomial models).
  • Several sensitivity analyses were done to evaluate the robustness of the results. Sex distribution, mean age, use of other vitamins or minerals combined with vitamin E and mean duration of follow-up were each separately evaluated to see if any of those changed the results. The influence of each trial on the results was evaluated by removing each individual study from the analysis.
  • In all analyses, authors estimated and inferred vitamin E effects by using population-average models with robust standard errors. The logistic model results were transformed into risk differences and risk rations of the effect of vitamin E supplementation, compared with control for easier clinical interpretability. Confidence interval (CI) for risk differences and risk ratios were obtained by using the delta method.
Data Collection Summary:

Timing of Measurements

The selected trials were reported between 1993 and 2004.

Dependent Variables

Death in study participants over time of study.

Independent Variables

  • Vitamin E or placebo
  • Low-dose vitamin E, less than 400 IU
  • High-dose vitamin E, more than 400 IU per day.

 Control Variables

  • Age
  • Sex
  • Years of follow-up
  • Use of other supplements in addition to vitamin E.
Description of Actual Data Sample:

N

  • There were 19 randomized control trials with a total of 135,967 participants
  • Study participants ranged from 196 to 29,584 per trial
  • Nine trials tested vitamin E alone and 10 tested a combination of vitamin E and other vitamins or minerals
  • Vitamin E doses ranged from 16.5 IU to 2,000 IU per day (median, 400 IU per day).

Age

47 to 84 years.

Ethnicity

Studies included individuals from France, China, Finland, Italy, 19 different unspecified countries, the United States, Israel and Canada (specific ethnicity not given).

Other Relevant Demographics

  • Most studies targeted individuals at high risk for a chronic disease, most often coronary heart disease
  • Five studies did not focus on risk for chronic disease. These targeted "healthy adults," "general population," "healthy older adults" and "well-nourished older adults."

Location

Authors were from:

  • Johns Hopkins Medial Institutions, Baltimore, MD, Division of Biostatistics
  • National Center for Epidemiology, Instituto de Salud Carlos III, Madrid, Spain
  • Cardiovascular Research Unit University of Edinburgh, Edinburgh, Scotland, United Kingdom.
Summary of Results:

Variables
(Dependent variable: One with each independent variable)

Treatment Group
(Measures and confidence intervals)

Control Group
(Measures and confidence intervals)

Statistical Significance of Group Difference
(Risk Ratio and 95% CI)

1: All-cause mortality vs. vitamin E or no vitamin E

Risk difference between groups: 10/10,000 people

95% CI: -18 to 38/10,000 people

1,022 per 10,000 people

RR 1.01 (0.98 to 1.04)

P>0.2

2: All-cause mortality vs. low-dose vitamin E (<400 IU per day)

Risk difference: -16 per 10,000 people

95% CI: -41 to 10 per 10,000 people

 

RR 0.98 (0.96 to 1.01)

P>0.2

3: All-cause mortality vs. high-dose vitamin E (>400 IU per day)

Risk difference: 39 per 10,000 people

95% CI: 3 to 74 per 10,000 people

 

RR 1.04

CI 1.01 to 1.07

P=0.035

Other Findings

  • All-cause mortality progressively increased as vitamin E dosage increased by more than 150 IU per day. For dosages less than 150 IU per day, all-cause mortality slightly but non-significantly decreased. Both linear and quadratic components of the dose-response model were statistically significant (P=0.027 and P=0.037).
  • In four-way analysis: Pooled risk difference between any amount of vitamin E and control was eight per 10,000 people (95% CI, -23 to 39 per 10,000 people; P>0.2). Pooled risk difference for low-dose vitamin E trials was -33 per 10,000 people (95% CI, -60 to -5 per 10,000 people; P=0.021). Pooled risk for high-dosage vitamin E trials was 34 per 10,000 people (95% CI, 5 to 64 per 10,000 people; P=0.022). Dose-response relationship for four-way analyses was very similar to that for two-way analyses; significant linear and quadratic components (P=0.027 and P=0.030, respectively).
  • There was no appreciable change after adjustment for differences in sex, age or mean follow-up across trials
  • When controlled for use of other vitamins or minerals, risk for all-cause mortality at low dosages of vitamin E decreased to -6 per 10,000 people (95% CI, -38 to 26 per 10,000 people) and risk for all-cause mortality with high dosages of vitamin E increased to 63 per 10,000 people (95% CI, 6 to 119 per 10,000 people)
  • Increased risk at high vitamin E dosages remained statistically significant after exclusion of each of the 11 high-dosage trials.
Author Conclusion:
  • High-dosage vitamin E supplementation (at least 400 IU per day) may increase all-cause mortality and should be avoided
  • High-dosage vitamin supplementation is often assumed to be at worst innocuous
  • In view of this study and previous studies showing an increased mortality associated with high-dosage beta carotene, use of any high-dosage vitamin supplementation should be discouraged until evidence of efficacy is documented from appropriately designed clinical trials.
Funding Source:
Industry:
R.A. Riemersma (Roche)
Pharmaceutical/Dietary Supplement Company:
University/Hospital: Instituto de Salud Carlos III (EPY 1261/02
Other:
Reviewer Comments:
  • This meta-analysis was clearly written, analysis of results was thorough and limitations of studies were well-described.
  • The fact that the Dietary Reference Intake for vitamin E for adults is actually very low and the tolerable upper limit is much higher than the dosages used in the high-dosage trials included in this meta-analysis makes this paper very important (Estimated Average Requirement,12.12 IU synthetic vitamin E or 18 IU natural vitamin E; Recommended Dietary Allowance, 15.15 IU synthetic vitamin E or 22.5 IU natural vitamin E; tolerable upper limit, 1,100 IU synthetic vitamin E or 1,500 IU natural vitamin E)
  • Based on the results of this study, the tolerable upper limit for vitamin E needs to be re-evaluated and perhaps reduced to 400 IU or less.
Quality Criteria Checklist: Primary Research
Relevance Questions
  1. Would implementing the studied intervention or procedure (if found successful) result in improved outcomes for the patients/clients/population group? (Not Applicable for some epidemiological studies) Yes
  2. Did the authors study an outcome (dependent variable) or topic that the patients/clients/population group would care about? Yes
  3. Is the focus of the intervention or procedure (independent variable) or topic of study a common issue of concern to dieteticspractice? Yes
  4. Is the intervention or procedure feasible? (NA for some epidemiological studies) Yes
 
Validity Questions
1. Was the research question clearly stated? Yes
  1.1. Was (were) the specific intervention(s) or procedure(s) [independent variable(s)] identified? Yes
  1.2. Was (were) the outcome(s) [dependent variable(s)] clearly indicated? Yes
  1.3. Were the target population and setting specified? Yes
2. Was the selection of study subjects/patients free from bias? N/A
  2.1. Were inclusion/exclusion criteria specified (e.g., risk, point in disease progression, diagnostic or prognosis criteria), and with sufficient detail and without omitting criteria critical to the study? Yes
  2.2. Were criteria applied equally to all study groups? Yes
  2.3. Were health, demographics, and other characteristics of subjects described? Yes
  2.4. Were the subjects/patients a representative sample of the relevant population? Yes
3. Were study groups comparable? Yes
  3.1. Was the method of assigning subjects/patients to groups described and unbiased? (Method of randomization identified if RCT) Yes
  3.2. Were distribution of disease status, prognostic factors, and other factors (e.g., demographics) similar across study groups at baseline? ???
  3.3. Were concurrent controls or comparisons used? (Concurrent preferred over historical control or comparison groups.) Yes
  3.4. If cohort study or cross-sectional study, were groups comparable on important confounding factors and/or were preexisting differences accounted for by using appropriate adjustments in statistical analysis? Yes
  3.5. If case control study, were potential confounding factors comparable for cases and controls? (If case series or trial with subjects serving as own control, this criterion is not applicable.) ???
  3.6. If diagnostic test, was there an independent blind comparison with an appropriate reference standard (e.g., "gold standard")? Yes
4. Was method of handling withdrawals described? N/A
  4.1. Were follow-up methods described and the same for all groups? N/A
  4.2. Was the number, characteristics of withdrawals (i.e., dropouts, lost to follow up, attrition rate) and/or response rate (cross-sectional studies) described for each group? (Follow up goal for a strong study is 80%.) N/A
  4.3. Were all enrolled subjects/patients (in the original sample) accounted for? N/A
  4.4. Were reasons for withdrawals similar across groups? N/A
  4.5. If diagnostic test, was decision to perform reference test not dependent on results of test under study? N/A
5. Was blinding used to prevent introduction of bias? Yes
  5.1. In intervention study, were subjects, clinicians/practitioners, and investigators blinded to treatment group, as appropriate? Yes
  5.2. Were data collectors blinded for outcomes assessment? (If outcome is measured using an objective test, such as a lab value, this criterion is assumed to be met.) Yes
  5.3. In cohort study or cross-sectional study, were measurements of outcomes and risk factors blinded? Yes
  5.4. In case control study, was case definition explicit and case ascertainment not influenced by exposure status? N/A
  5.5. In diagnostic study, were test results blinded to patient history and other test results? N/A
6. Were intervention/therapeutic regimens/exposure factor or procedure and any comparison(s) described in detail? Were interveningfactors described? Yes
  6.1. In RCT or other intervention trial, were protocols described for all regimens studied? Yes
  6.2. In observational study, were interventions, study settings, and clinicians/provider described? N/A
  6.3. Was the intensity and duration of the intervention or exposure factor sufficient to produce a meaningful effect? N/A
  6.4. Was the amount of exposure and, if relevant, subject/patient compliance measured? Yes
  6.5. Were co-interventions (e.g., ancillary treatments, other therapies) described? Yes
  6.6. Were extra or unplanned treatments described? N/A
  6.7. Was the information for 6.4, 6.5, and 6.6 assessed the same way for all groups? Yes
  6.8. In diagnostic study, were details of test administration and replication sufficient? N/A
7. Were outcomes clearly defined and the measurements valid and reliable? Yes
  7.1. Were primary and secondary endpoints described and relevant to the question? Yes
  7.2. Were nutrition measures appropriate to question and outcomes of concern? Yes
  7.3. Was the period of follow-up long enough for important outcome(s) to occur? Yes
  7.4. Were the observations and measurements based on standard, valid, and reliable data collection instruments/tests/procedures? Yes
  7.5. Was the measurement of effect at an appropriate level of precision? Yes
  7.6. Were other factors accounted for (measured) that could affect outcomes? Yes
  7.7. Were the measurements conducted consistently across groups? Yes
8. Was the statistical analysis appropriate for the study design and type of outcome indicators? N/A
  8.1. Were statistical analyses adequately described and the results reported appropriately? Yes
  8.2. Were correct statistical tests used and assumptions of test not violated? Yes
  8.3. Were statistics reported with levels of significance and/or confidence intervals? Yes
  8.4. Was "intent to treat" analysis of outcomes done (and as appropriate, was there an analysis of outcomes for those maximally exposed or a dose-response analysis)? N/A
  8.5. Were adequate adjustments made for effects of confounding factors that might have affected the outcomes (e.g., multivariate analyses)? Yes
  8.6. Was clinical significance as well as statistical significance reported? Yes
  8.7. If negative findings, was a power calculation reported to address type 2 error? N/A
9. Are conclusions supported by results with biases and limitations taken into consideration? Yes
  9.1. Is there a discussion of findings? Yes
  9.2. Are biases and study limitations identified and discussed? Yes
10. Is bias due to study's funding or sponsorship unlikely? Yes
  10.1. Were sources of funding and investigators' affiliations described? Yes
  10.2. Was the study free from apparent conflict of interest? Yes