| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
The Cooper Institute, Dallas Texas
Address reprint requests to: Conrad P. Earnest, PhD, The Cooper Institute, 12330 Preston Road, Dallas, TX 75230. E-mail: cearnest{at}cooperinst.org
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Methods: We examined 182 participants for selected plasma vitamin concentrations and clinically relevant variables including homocysteine, lipids and LDL-C oxidation indices at baseline and six months.
Results: We found no significant differences between groups for any parameter at baseline. Following six months of vitamin supplementation, we observed elevations in plasma concentrations of vitamin B6 (as pyridoxal 5'-phosphate; PLP), vitamin B12, folate, vitamin C, vitamin E and ß-carotene (p < 0.0001), all of which were significantly greater than respective placebo group changes (p < 0.0001). Homocysteine decreased in the treatment (8.38 ± 2.9 vs. 6.93 ± 2.5 µmol/L; p < 0.0001) and placebo group (8.17 ± 3.0 vs. 7.42 ± 2.2 µmol/L; p < 0.0001) from baseline to six months, respectively, with reductions in the treatment group being greater than placebo (p < 0.008). LDL-C oxidation indices were also improved as LDL-C oxidation rate was decreased (-0.39 µmol/min/g protein; p < 0.0003) and LDL-C lag time increased (11.3 min; p < 0.003) in supplemented participants. Further analysis also showed that LDL-C oxidation rate was lower (p < 0.0007) and LDL-C lag time longer (p < 0.0001) for the vitamin group than placebo treatment after six months.
Conclusion: We conclude that a multi-ingredient vitamin formula with antioxidant properties has measurable effects on homocysteine and LDL-C oxidation indices.
Key words: multivitamin, homocysteine, LDL-C oxidation
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
In a recent review, recommendations were made that the problem of suboptimal vitamin ingestion could be easily overcome through the purchase of "brand-name" multivitamins for an annual investment of about $10 to $30 [4]. Others also hold this opinion [15]. The authors further state that they are "aware of no evidence that multivitamins differ in bioavailability." While we are not in disagreement with this premise, the Dietary Supplement and Health Education Act does not require supplement companies to demonstrate absorption characteristics/bioavailability or product efficacy. Therefore, consumers must take purported health benefits somewhat on faith, rather than clinical evidence. Independent consumer advocacy laboratories also show that the reported vitamin concentration in some products varies widely and is often less than advertised label claims [16]. Thus, even if absorption is not the issue per se, nutrient bioavailability will certainly be affected by availability of ingredients claimed to be in the product. We propose that a better solution is to validate the efficacy of individual products directly.
We recently reported in an open-label, proof-of-concept pilot trial that participants supplementing with a complex multivitamin formula (e.g., 24-ingredient, Table 1) were capable of significantly elevating plasma concentrations of vitamins B6, B12, folic acid, vitamin C, vitamin E and beta-carotene. Clinical markers associated with cardiovascular disease (CVD) examined during this trial also showed significant reductions in homocysteine concentration, increases in LDL-C oxidation lag time and decreases in LDL-C oxidation rate [17]. The primary ingredients of this formula include vitamins A, D, C, E and K, as well as folic acid and vitamins B6 and B12. An elegant review of many of these ingredients has been detailed elsewhere [4]. Though the pitfalls of such a study design are obvious, the aim of our current investigation reported here is to examine this same formula in a more vigorous clinical follow-up trial. As the benefits of these vitamins are broad, we have chosen to focus this report on those indices specifically related to coronary heart disease. To accomplish this, we have examined the plasma concentrations of vitamins B6, B12, C, folic acid and E, as well as several clinical markers associated with CVD risk including homocysteine, LDL-C oxidation rate and LDL-C oxidation lag time.
|
|
METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Inclusion Criteria and Pretrial Screening
Inclusion criteria for this study necessitated that male and female participants have a homocysteine concentration not less than 8.0 and 7.0 µmol/L, respectively. All participants agreed to maintain their current diet and exercise habits during the trial and to have abstained from taking vitamin or mineral supplements for six weeks prior to screening. We excluded pregnant or lactating women from participation. Postmenopausal women both on and off hormone replacement therapy (HRT) were accepted into the trial; however, we asked those on HRT to remain on their current medication and dosage schedule and notify us if the regimen was changed. Participants currently on standard medical therapy (for conditions such as hypertension, hypercholesterolemia, diabetes, arthritis or other chronic diseases) were allowed to enter the study if they agreed to remain on their current therapy during the trial.
Participants were excluded from the trial if their BMI was <18.5 or >34.9, they had recently donated blood (<3 months) or failed to agree not to donate blood during the trial. Also excluded were participants with elevated blood pressure, total cholesterol, LDL-cholesterol or fasting plasma glucose requiring immediate drug therapy according to national guidelines (JNC VI, Adult Treatment Panel II, and American Diabetes Association). Participants were also excluded if they planned to move from the area during the trial, smoked more than two packs of cigarettes, consumed alcoholic beverages exceeding an average of three drinks per day or consumed coffee in excess of three cups per day.
We initially screened 1,275 potential participants by phone, with 447 qualifying to initiate the study. After completing an initial blood assessment, 259 participants remained eligible to participate, but 20 of these participants withdrew prior to randomization. Of the remaining 239 participants who formally entered the trial, 57 withdrew from the trial after initiating treatment. The time that elapsed between the screening and the first blood sample was less than two weeks. The initial screening for homocysteine was performed at the Cooper Clinic Clinical Laboratory because of its proximity. However, once successful screening had taken place, all blood assessments used as dependent variables were examined at an independent and blinded laboratory. Following a successful screening procedure, participants donated fasting blood samples to assess various indices (see below) before we gave them a 24-ingredient multivitamin/mineral formula (Cooper Complete, Dallas, TX USA) for 24 weeks (Table 1). A scientific advisory board assisted in the development of this formula based on an evidence-based, systematic review of the literature and their personal research performed in three nutrition research centers in the USA.
Treatment and Assessment Indices
Each participant consumed three tablets twice a day for six months and was verbally instructed to take them with a meal. During each visit, all participants returned their pill bottles, and a pill count was undertaken to ensure compliance. Blood assessments were obtained in a fasting condition (>12 hours) for a variety of parameters inclusive of a blood lipid profile, blood glucose concentration, muscle, kidney and hepatorenal indices (Chem-16), homocysteine, LDL-C oxidation rates (i.e., lag time and oxidation rate) and blood vitamin concentrations. Each blood collection consisted of obtaining approximately 50 mL of blood that was divided into one serum separator vacutainer (10 mL) and four K3 EDTA tubes (
40 mL). All samples were subsequently spun within three minutes of venous collection in a cold centrifuge at 1200 rpm for 12 minutes. Separated plasma and red blood cells were divided into four cryovials and placed in a -80°C freezer. Blood lipid profiles, fasting blood glucose concentration and muscle and hepatorenal indices were analyzed immediately at the Cooper Clinic (Dallas, TX USA) via a Dimension RXL analyzer (Oxford, Connecticut USA). The Cooper Clinic Lab is a College of American Pathologists certified lab. All frozen samples were shipped frozen on dry ice via overnight express to various labs for subsequent analyses (see below).
Homocysteine, Folate and Vitamins B12 and B6.
Blood samples were obtained to determine concentrations of homocysteine, folate, vitamin B12 and pyridoxal 5'-phosphate (the active circulating form of vitamin B6). Total homocysteine concentration in plasma was determined by high-performance liquid chromatography with fluorometric detection [18]. Plasma folate was measured by a microbial (Lactobacillus casei) assay in a 96-well plate [19,20]. Plasma pyridoxal 5'-phosphate was measured by the tyrosine decarboxylase apoenzyme method and plasma vitamin B12 was measured by radio-assay (Quantaphase II, Bio-Rad, Hercules, CA USA) [21]. Coefficients of variation for these assays were 8% for homocysteine, 13% for folate, 16% for pyridoxal 5'-phosphate and 7% for vitamin B12.
Vitamin E (
-tocopherol), Vitamin C (ascorbate), ß-Carotene and LDL-C Oxidation Indices.
Additional blood was also obtained for plasma -tocopherol, ascorbate and ß-carotene concentrations. Samples for plasma ascorbate were deproteinized with ice-cold 10% metaphosphoric acid and centrifuged. The supernatant was purged with nitrogen and stored below -20°C in foil-covered tubes. Plasma ascorbate concentrations were determined spectrophotometrically after derivatization with 2,4-dintrophenylhydrazine. The concentrations of -tocopherol and ß-carotene were measured in plasma and LDL-C following extraction by reverse-phase high-performance liquid chromatography [22]. The plasma concentrations of both -tocopherol and ß-carotene were standardized to total plasma lipids as described [23].
Two indices of oxidation were used in this study. First, the lipid peroxide content of oxidized LDL-C was measured by a modification of the thiobarbituric acid-reactive substances (TBARS) [24]. TBARS activity was expressed as malondialdehyde equivalents using freshly diluted 1,1,3,3-tetramethoxypropane as the standard. Second, the amount of conjugated dienes formed during LDL-C oxidation was determined by measuring the absorbance of LDL-C against a PBS blank at 234 nm following a 1:4 dilution of the samples in PBS [25]. Research has shown that dilution of an oxidized LDL-C sample to 1:2, 1:4 and 1:8 displays linearity and excellent recovery; the data are expressed as the increase in conjugated dienes over baseline (A234) [26]. The rate of LDL-C oxidation was determined from the propagation phase of the time-course curve using a spline function. We determined lag phase by drawing a tangent to the slope of the propagation phase and extrapolating it to the horizontal axis [26]. The lag time constitutes the interval from zero time to the intersection point.
Statistical Analysis
We calculated median and interquartile ranges for skewed variables and means and standard deviations for normally distributed variables by treatment group. Spearman correlations were computed to assess association between variables. The mean change in each variable was compared between treatment groups using ANCOVA with adjustment for age, body mass index, gender, HRT, statin use and baseline value. Results are reported as least squares adjusted means. Correlations were also performed to examine the relationship between changes in plasma vitamin levels and homocysteine, LDL-C lag time and lag rate. All statistical analyses were performed using SAS Version 8.1. Statistical significance refers to a value of two-tailed p < 0.05.
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Blood Chemistry, Plasma Vitamin and Homocysteine Concentration
We observed no significant changes for total cholesterol, LDL-C, triacylglycerols, kidney function, hepatorenal function or hemoglobin A1C following treatment (Table 2). However, a significant decrease in HDL-C was noted for the placebo group (-0.06 mmol/L; [95% CI; -0.11, -0.01]). Our analysis did show significant elevations in the plasma concentrations of vitamin B6 (as pyridoxal 5'-phosphate), vitamin B12, folate, vitamin C, vitamin E and beta-carotene (p < 0.0001; Table 3). Post-hoc analysis demonstrated that all measured vitamin concentrations were significantly elevated in the treatment group at six months (p < 0.0001). No significant changes in these values were observed for the placebo condition. Between group analyses also showed that the corresponding increase in plasma vitamin concentrations were statistically greater than those in the placebo treatment. Significant changes in homocysteine were observed as both groups demonstrated a reduction in homocysteine concentration (p < 0.0001). Further analysis revealed that the reduction observed in the treatment group was greater than that observed in the placebo group (p < 0.008).
|
|
LDL-C Oxidation Characteristics
Our analysis showed a significant treatment effect for each LDL-C oxidation index (Table 3; p < 0.0001). Specifically, LDL-C oxidation rate was decreased 0.39 µmol/min/g protein (p < 0.0003) and LDL-C lag time increased 11.3-minutes (p < 0.003) in the treatment group following six months of vitamin supplementation. These changes in LDL-C oxidation rate (p < 0.0007) and LDL-C lag time (p < 0.0001) were significantly different from the placebo treatment. Plasma changes in vitamin E were significantly correlated with changes in LDL-C lag rate (partial r, 0.037; p < 0.03), but not LDL-C lag time (partial r, 0.025; p < 0.07).
|
DISCUSSION |
|---|
Following six months of supplementation, we observed significant increases in plasma concentrations of vitamin B6, plasma folate and vitamin B12 in multivitamin treated participants, all of which are associated in some degree to circulating homocysteine concentrations [8–10,27]. The increase in plasma vitamin concentrations observed during our trial is matched by a 1.45 µmol/L decrease in homocysteine concentration for treated participants. Though the placebo group also showed a significant reduction in homocysteine, it was not of the same magnitude and was significantly less than the treatment group. In this regard, we cannot discount regression to the mean. However, we did attempt to control for this phenomenon by employing a screening of subjects prior to trial entry. In addition, the nature of the subject pool is such that each individual was already "skewed" compared to a broader population by having elevated homocysteine. It is interesting to note that no other parameter changed in similar fashion (i.e., LDL-oxidation indices). These data confirm our previous trial showing a 1.20-µmol/L reduction in homocysteine after three months of supplementation and are consistent with previous research showing an inverse relationship between folic acid, vitamins B6 and B12 and homocysteine [8–10,27]. This is clinically important, as several cohort studies have shown retrospectively and prospectively that low plasma folate and B vitamin concentrations are associated with CHD [28,29].
One of the most compelling reports examining folate and vitamin B status showed that grain products fortified with folic acid were associated with a substantial improvement in folate status and a decreased homocysteine concentration [11]. Perhaps the most intriguing feature of this report is the observation that homocysteine concentrations were an additional 10% lower among those who used vitamin supplements versus those who did not. The authors attributed the difference in homocysteine to higher vitamin B12 and B6 concentrations in participants using B vitamin supplements [11]. Furthermore, previous research in men and women receiving supplementation with increasing doses of folic acid has shown that the usual dependency of homocysteine on folate ingestion diminishes with supplementation and that vitamin B12 becomes the main determinant of plasma homocysteine concentration [30]. This finding suggests that fortification with folic acid and vitamin B12, rather than folic acid alone, may be more effective at lowering homocysteine concentrations. We believe that this observation is clinically relevant because it shows that homocysteine can be further improved via supplementation in a population already consuming food products fortified with folic acid [11]. For example, treated participants in our trial increased baseline plasma folate concentrations from 14.22 ± 6.83 nmol/L to 20.75 ± 8.26 nmol/L after six months of supplementation. Despite this increase in folate concentration, the strongest association between changes in homocysteine and the changes in individual plasma vitamin concentrations observed in our study was vitamin B12. It is conceivable that our population had sufficient dietary folate, yet was marginal in vitamin B12 status. Therefore, it seems prudent to recommend that adequate folic acid and vitamin B12 are included in a supplement routine if the goal is to reduce homocysteine concentration. Though we cannot conclude that our supplement will decrease the incidence of CHD, similar findings to ours support the idea that folic acid and vitamin B supplementation reduces risk factors associated with several chronic diseases including CHD [27], colon cancer in men and women [31,32] and breast cancer in women regularly consuming alcohol [33].
Our second clinical finding is the observation that our formulation improved two indices of LDL-C oxidation. Specifically, treated participants in our trial showed a 12% increase in LDL-C oxidation lag time and an 18% decrease in LDL-C oxidation lag rate. The results for LDL-C lag time in this trial are similar to those in our first uncontrolled trial. However, our overall data from our first trial for LDL-C lag rate was longer at three months and six months of supplementation [17]. Nonetheless, positive alterations in LDL-C oxidation have been associated with vitamin E, vitamin C and lycopene [34,35]. In our present study, we measured plasma concentrations of vitamins E and C, but not lycopene. However, lycopene has been shown to be easily absorbed and work synergistically with vitamin E relative to LDL-C oxidation [36]. Evidence has also shown that vitamin C attenuates LDL-C oxidation [37] and may complement the effects of vitamin E [38]. This latter point is still a matter for debate as Huang et al. recently demonstrated that supplementation with vitamin C and vitamin E alone reduced lipid peroxidation to a similar extent. However, supplementation with a combination of vitamins C and E conferred no benefit beyond that of either vitamin alone [39], and the weight of current evidence suggests that vitamin E is the more potent progenitor of LDL-C oxidation—an effect that may be dose dependent [40].
The effect of vitamin E (-tocopherol) on LDL-C oxidation kinetics has been thoroughly reviewed elsewhere [40]. In brief, vitamin E is postulated to prevent atherosclerotic disease via its antioxidant effects, attenuation of smooth muscle proliferation and/or a reduction in platelet adhesion, and of arterial plaque accumulation [41–43]. Despite these physiologic benefits, the clinical benefit of secondary prevention has not been shown, even when vitamin E is taken in high doses [44,45]. However, all of these studies have been relatively short in length and deal with individuals with known metabolic disease (i.e., high risk). There is also an issue for a potential dose dependency relationship, where one shorter trial has demonstrated a reduction in deaths from coronary artery disease and nonfatal myocardial infarctions [46].
An equally important issue is that mortality stemming from atherosclerotic disease often takes place over several decades and is affected by oxidative stress. Further, the occurrence of atherosclerotic plaque at the carotid bifurcation has been found to be inversely associated with vitamin E intake [42]. Thus, vitamin E may have more influence on the disease process and be appropriate for primary prevention with longer investigations and use, as high vitamin E intake has been shown to decrease LDL-C oxidation and inhibit lesion progression [40,43]. In keeping with this finding, we observed an increase in LDL-C oxidation lag time and a decrease in LDL-C oxidation rate following six months of supplementation using a formula containing 800 IU of vitamin E and 1,000 mg of vitamin C per day. These effects are similar to those noted by Jialal and Grundy, who showed that the combined supplementation of -tocopherol (400 IU/day) plus ascorbate (1,000 mg/day) and beta carotene (30 mg/day) resulted in the two-fold prolongation of the oxidative lag phase and a 40% decrease in the oxidation rate [25]. Interestingly, several studies have shown attenuation in atherosclerotic progression when vitamin E and C are taken together [47–49].
A final point to consider in our study is that some trials using ß-carotene show an increased risk of lung cancer among smokers [50]. However, this effect has not been observed in all trials. The obvious implication is that ß-carotene supplementation is contraindicated among smokers. It is also important to observe that the dose in our formula (3-mg) is much smaller than those observed in other trials (20–25 mg) noting this adverse effect [50]. Our vitamin A was composed of mixed ß-carotene and other carotenoids, primarily lutein and zeazanthin, ingredients which have shown in serum studies to have an inverse relationship to breast cancer [51,52].
|
CONCLUSION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
We also believe that it is important for the reader to distinguish between the potential for long-term prophylaxis via a nutrient support intervention designed to overcome suboptimal vitamin intakes and inadvertently assigning vitamin supplements the role of a pharmacological treatment. Our evidence suggests that it is biologically plausible to alter processes associated with disease progression and cause. Therefore, the context of each investigation should be kept in mind as "low risk" groups are seldom examined clinically (i.e., placebo controlled studies), but rather rely on epidemiological evidence. In this regard, only long-term follow up studies will be capable of addressing the issue of whether additional multivitamin supplementation will be beneficial in decreasing the incidence and mortality associated with certain chronic diseases.
|
ACKNOWLEDGMENTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received September 25, 2002. Accepted March 21, 2003.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Villa, C. Perri, R. Suriano, F. Cucinelli, S. Panunzi, M. Ranieri, C. Mele, and A. Lanzone L-Folic Acid Supplementation in Healthy Postmenopausal Women: Effect on Homocysteine and Glycolipid Metabolism J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4622 - 4629. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
