点击显示 收起
1 From the Division of Cardiovascular and Medical Sciences (DJS, GDOL, AR, PL, and PWM), the Nursing & Midwifery School (GM), and the Robertson Centre for Biostatistics (ADM), University of Glasgow, Glasgow, United Kingdom; the Departments of Haematology (RCT) and Pathological Biochemistry (DSJO), Glasgow Royal Infirmary, Glasgow, United Kingdom; the Department of Geriatric Medicine, Garnavel General Hospital, Glasgow, United Kingdom (EGS and JBM); and the Section of Gerontology and Geriatrics, Leiden University Medical Centre, Leiden, Netherlands (RGJW)
2 International Standard Randomized Controlled Trial Number (ISRCTN) 07337345. 3 Supported by a grant from the Healthcare Foundation (reference 112/57). 4 Reprints not available. Address reprint requests to DJ Stott, Academic Section of Geriatric Medicine, 3rd Floor Centre Block, Glasgow Royal Infirmary, Glasgow G4 0SF, United Kingdom. E-mail: d.j.stott{at}clinmed.gla.ac.uk.
ABSTRACT
Background: Homocysteine is an independent risk factor for vascular disease and is associated with dementia in older people. Potential mechanisms include altered endothelial and hemostatic function.
Objective: We aimed to determine the effects of folic acid plus vitamin B-12, riboflavin, and vitamin B-6 on homocysteine and cognitive function.
Design: This was a factorial 2 x 2 x 2, randomized, placebo-controlled, double-blind study with 3 active treatments: folic acid (2.5 mg) plus vitamin B-12 (500 µg), vitamin B-6 (25 mg), and riboflavin (25 mg). We studied 185 patients aged 65 y with ischemic vascular disease. Outcome measures included plasma homocysteine, fibrinogen, and von Willebrand factor at 3 mo and cognitive change (determined with the use of the Letter Digit Coding Test and on the basis of the Telephone Interview of Cognitive Status) after 1 y.
Results: The mean (±SD) baseline plasma homocysteine concentration was 16.5 ± 6.4 µmol/L. This value was 5.0 (95% CI: 3.8, 6.2) µmol/L lower in patients given folic acid plus vitamin B-12 than in patients not given folic acid plus vitamin B-12 but did not change significantly with vitamin B-6 or riboflavin treatment. Homocysteine lowering with folic acid plus vitamin B-12 had no significant effect, relative to the 2 other treatments, on fibrinogen, von Willebrand factor, or cognitive performance as measured by the Letter Digit Coding Test (mean change: –1; 95% CI: –2.3, 1.4) and the Telephone Interview of Cognitive Status (–0.7; 95% CI: –1.7, 0.4).
Conclusion: Oral folic acid plus vitamin B-12 decreased homocysteine concentrations in elderly patients with vascular disease but was not associated with statistically significant beneficial effects on cognitive function over the short or medium term.
Key Words: Elderly • homocysteine • folic acid • vitamin B-12 • riboflavin • vitamin B-6 • randomized controlled trial • cognitive function
INTRODUCTION
In the developed world, vascular disease is the predominant cause of disability and death and is a major contributor to cognitive decline in elderly people. Serum total homocysteine (tHcy) is an independent risk factor for vascular disease, including myocardial infarction and stroke (1). An elevated tHcy concentration is also associated with dementia (2). Several plausible biological mechanisms have been proposed for these associations, including an enhanced tendency for thrombosis mediated via increased endothelial disturbance (3), platelet activation, reduced cell expression of thrombomodulin, and inhibition of activated protein C (4).
tHcy is a sulfhydryl amino acid. Its precursor, methionine, is an essential amino acid derived from dietary protein. The enzymes responsible for metabolizing tHcy are cystathionine synthase, methionine synthase, and 5,10 methylenetetrahydrofolate reductase. The activity of these enzymes is dependent on 4 micronutrients: folic acid, vitamin B-12, riboflavin (vitamin B-2), and vitamin B-6 (pyridoxal 6-phosphate), deficiencies of which cause elevations in plasma tHcy. Folic acid is the most important of these vitamins in treatment regimens designed to reduce tHcy (1, 5); folic acid supplementation reduces tHcy across a wide range of erythrocyte folate concentrations. However, in older patients, vitamin B-12 deficiency is often an important contributor to elevated tHcy concentrations (6). The importance of riboflavin and vitamin B-6 is less clear; however, both act as cofactors in the enzymatic breakdown of tHcy (7), and supplementation might play a role in ensuring maximal reductions in tHcy.
Aging is associated with elevated tHcy concentrations (7) and a reduced activity of cystathionine synthase (8), one of the key tHcy-metabolizing enzymes. Therefore, older patients may be at particular risk of tHcy-mediated disease. We aimed to determine whether vitamin supplementation with folic acid plus vitamin B-12, vitamin B-6, and riboflavin reduces plasma tHcy, alters hemostatic and endothelial function, and affects cognitive function in elderly patients with vascular disease.
SUBJECTS AND METHODS
Patients
A total of 185 patients was studied in this 2-center, hospital-based, randomized controlled trial (Figure 1). Inclusion criteria were age 65 y and ischemic vascular disease, defined as one or more of the following: history of angina pectoris, previous acute myocardial infarction, evidence of major ischemia or previous acute myocardial infarction on the basis of a 12-lead electrocardiogram, ischemic stroke, transient ischemic attack, intermittent claudication, or surgery for peripheral arterial disease. Exclusion criteria included an acute vascular event <1 wk previously; major surgery <1 mo previously; any other major acute illness <1 mo previously; severe renal impairment (serum creatinine > 400 µmol/L); severe hepatic impairment; malignancy within the previous year (excluding local skin cancer); severe congestive heart failure (New York Heart Association class IV); total anterior cerebral infarct with major residual disability; malabsorption; inability to give informed consent (eg, due to dementia or dysphasia); major cognitive impairment (Mini-Mental State Examination score <19); existing treatment with riboflavin, vitamin B-6, vitamin B-12, or folic acid preparations; hemoglobin concentration < 10 g/dL; and mean cell volume >100 fL plus either a low red blood cell folate concentration (<280 ng/mL) or a low serum vitamin B-12 concentration (<250 pg/mL). Baseline characteristics are presented in Table 1. Written informed consent was obtained from all eligible patients. The study was approved by the relevant local hospital ethical committees. The study was registered on the Cochrane Central Register of Controlled Trials (2001) and conducted and reported according to CONSORT (Consolidated Standards of Reporting Trials) guidelines (9).
FIGURE 1.. Flow diagram of patient recruitment and progress throughout the study. F, folic acid; B12, vitamin B-12; B6, vitamin B-6; B2, riboflavin.
View this table:
TABLE 1. Baseline characteristics of all study groups1
Treatment
The patients initially entered a single-blind, placebo, run-in (2 capsules/d as per the randomized phase of the trial) phase lasting between 2 and 4 wk. Patients who successfully completed the run-in phase were randomly allocated to receive folic acid (2.5 mg) plus vitamin B-12 (400 µg) or placebo, vitamin B-6 (25 mg) or placebo, or riboflavin (25 mg) or placebo for 12 wk in a factorial 2 x 2 x 2 design. The various combinations of vitamins and placebo were all packaged so that the daily dose was provided in a total of 2 capsules (1 red and 1 white), irrespective of patient group. The capsules were visually identical for the run-in phase and for all arms of the trial. Treatment allocation was concealed from the patients and the investigators (double-blind). Allocation was determined at a site remote from the clinical study (Robertson Centre for Biostatistics) in randomized permuted blocks of 8, stratified by hospital center.
Study measures
Patients were assessed at the beginning and end of the placebo run-in and at 3, 6, and 12 mo after randomization in the active phase of the trial. Baseline measures included height, weight, blood pressure (taken using a standard sphygmomanometer as the mean of 2 recordings after 5 min sitting, diastolic phase V), a 12-lead electrocardiogram, the Barthel index, and a short instrumental activities of daily living (IADL) scale (10).
A standard fasting venous blood sample was taken at both baseline visits and at 3 mo. Blood was anticoagulated with tripotassium citrate (0.109 mol/L; 9:1, by vol) and centrifuged at 2000 x g at 4 °C; citrated plasma aliquots were snap-frozen and stored at –50 °C until assayed for plasma total homocysteine (measured by HPLC), von Willebrand factor (measured by enzyme-linked immunosorbent assay; DAKO, High Wycombe, United Kingdom), and fibrinogen (Clauss assay, Coag-A-Mate X2; Organon Teknika, Cambridge, United Kingdom). Serum vitamin B-12 and red blood cell folate were measured with the chemiluminescence method (Centaur; Bayer, Newbury, United Kingdom). Plasma pyridoxal 5-phosphate was measured by using HPLC. Riboflavin status was assessed by measuring erythrocyte glutathione reductase activation coefficient (11).
General cognitive function was assessed by using the Telephone Interview for Cognitive status (TICSm; 12) at both baseline visits and at 6 and 12 mo; attention and speed of information processing were assessed by using the Letter Digit Coding Test (at both baseline visits and 12 mo) (13, 14). Baseline cognitive assessments were face-to-face, 6-mo reviews by telephone and 12-mo face-to-face reviews. The TICSm is composed of 21 items and has a maximum score of 39 (12); it was used in the Heart Protection Study (15). It correlates highly with the in-person Mini-Mental State Examination (MMSE) in healthy elderly subjects and in those with a diagnosis of Alzheimer disease (16). Serious adverse events, including incident vascular events, were recorded at each review.
Statistical analysis
Our target sample size was 200 individuals. The study was planned pragmatically, recognizing that it would not be powered to detect an effect on vascular endpoints but would provide valuable data in planning a large-scale intervention study. Data were analyzed by using the SAS version 8.02 software package (SAS Institute Inc, Cary, NC). All analyses were based on an intention-to-treat basis. Baseline summary statistics are presented as means ± SDs or as medians and interquartile ranges (IQRs) for continuous variables and as numbers and percentages for categorical variables. Baseline data are presented for all 8 groups of vitamin combinations; however, the analysis of each of the 3 active components of treatment was preplanned to be based on a factorial design, with a comparison of 1) all patients who received folic acid plus vitamin B-12 and those who did not receive folic acid plus vitamin B-12, 2) all patients who received riboflavin and those who did not receive riboflavin, and 3) all those who received vitamin B-6 and those who did not receive vitamin B-6.
The effects of the different vitamins on laboratory variables were analyzed by analysis of covariance (ANCOVA). The change from the mean of the 2 baseline measures was compared with the 3-mo measure, with adjustment for the baseline value of each variable. Statistical analysis of interactions of the different vitamin interventions on homocysteine concentrations was performed based on 3-factor ANCOVA (folic acid plus vitamin B-12, riboflavin, and vitamin B-6), seeking possible 3- and 2-factor interactions. An analysis of the changes in cognitive function was performed similarly to the laboratory analyses, but the change from baseline to 1 y of follow-up was examined. The effects of vitamins on incident vascular events were analyzed by calculating Mantel-Haenszel odds ratios and 95% CIs.
RESULTS
Baseline characteristics for all 8 different vitamin combination groups are presented in Table 1. Mean (±SD) baseline red blood cell folate and vitamin B-12 concentrations were 312 ± 127 ng/mL and 362 ± 136 pg/mL, respectively, in the group treated with folic acid plus vitamin B-12 and 294 ± 120 ng/mL and 371 ± 123 pg/mL, respectively, in those not treated with folic acid plus vitamin B-12. At baseline the median MMSE score for all patients was 28 (IQR: 26–29). Vitamin supplementation increased vitamin concentrations and improved vitamin status as expected (Table 2). Mean (±SD) baseline fasting plasma homocysteine was 16.5 ± 6.4 µmol/L. This was reduced significantly after 3 mo of treatment in all treatment groups that received folic acid plus vitamin B-12. Reductions in tHcy were seen across the entire range of baseline red blood cell folate concentrations; however, they were greatest in subjects in the lowest baseline quintiles of concentrations of this vitamin (P = 0.019, 4 df; interaction test) (Figure 2). There were no statistically significant effects of vitamin B-6 or riboflavin supplementation on homocysteine concentrations (Table 2). No statistically significant 2- or 3-factor interactions between the different components of vitamin treatment on homocysteine concentrations were seen (3-factor ANCOVA).
View this table:
TABLE 2. Changes in vitamin and homocysteine concentrations from baseline to 3 mo in the different vitamin-treatment groups1
FIGURE 2.. Mean (±1 SE) change in total homocysteine concentrations with folic acid plus vitamin B-12 supplementation () compared with no folate and vitamin B-12 supplementation (•), by baseline concentration of red blood cell folate (quintile 1: 210 ng/mL; quintile 2: >210 to 252 ng/mL; quintile 3: >252 to 315 ng/mL; quintile 4: >315 to 387 ng/mL; quintile 5: >387 ng/mL). P values (analysis of covariance) for quintiles 1–5: <0.001, 0.004, 0.001, 0.004, and 0.039, respectively. P = 0.019 for the interaction between the baseline red blood cell folate concentration and the reduction in homocysteine concentrations with folic acid plus vitamin B-12.
No statistically significant 2- or 3-factor interactions between the different components of vitamin treatment on fibrinogen or von Willebrand factor were seen (3-factor ANCOVA) (Table 3). There was no significant difference in the 1-yr incidence of vascular events in subjects who received folic acid plus vitamin B-12 (17/92; 18%) compared with those who received no folic acid or vitamin B-12 (13/93; 14%); odds ratio 1.39 (95% CI: 0.63, 3.07).
View this table:
TABLE 3. Changes in hemostatic factors from baseline to 3 mo in the different vitamin-treatment groups1
There were no significant effects of any of the vitamins on change in cognitive function (as measured by TICSm and the Letter Digit Coding Test; Table 4). There was no evidence of beneficial effects on cognition from folic acid plus vitamin B-12, even in those in the lowest quintile of baseline red blood cell folate or of serum vitamin B-12 concentrations.
View this table:
TABLE 4. Effects of the different vitamin treatments on cognitive function1
The results of the preplanned marginal analyses are presented in Table 5. All patients who received folic acid plus vitamin B-12 were compared with all those who do not receive these vitamins, all who received riboflavin were compared with all those who did not receive riboflavin, and all those who received vitamin B-6 were compared with all those who did not receive vitamin B-6. Folic acid plus vitamin B-12 significantly decreased tHcy concentrations. However, there were no significant effects of any of the vitamins on fibrinogen or on the von Willebrand factor. No significant effects of any of the vitamins on cognition were seen; the between-group differences in change in cognition for all patients who received folic acid plus vitamin B-12 compared with all those who received no folic acid or vitamin B-12 were –0.7 (95% CI: –1.7, 0.4) for the TICSm score and –1.0 (–2.3, 0.4) for the Letter Digit Coding Test. Compared with the patients who did not receive riboflavin, the patients who received riboflavin had a change of –0.5 (95% CI: –1.5, 0.5) in the TICSm score and a change of 0.4 (–1.0, 1.7) in the Letter Digit Coding Test. Compared with those who received no vitamin B-6, the patients who received vitamin B-6 had a change of –0.1 (95% CI: –1.1, 0.9) in the TICSm score and a change of –0.6 (–2.0, 0.8) in the Letter Digit Coding Test.
View this table:
TABLE 5. Changes in homocysteine, fibrinogen, and the von Willebrand factor (baseline to 3 mo) and in cognitive function (1 y) in the different vitamin-treatment groups1
DISCUSSION
We found that folic acid plus vitamin B-12 had a large effect on tHcy in elderly patients with vascular disease; concentrations decreased by 33%. The effect was greatest in those with low baseline red blood cell folate and serum vitamin B-12 concentrations. However, folic acid plus vitamin B-12 supplementation reduced tHcy concentrations across the whole range of baseline folate concentrations, with no apparent ceiling effect, and in all except the top quintile of baseline vitamin B-12 concentrations.
We saw no significant effects of supplementation with vitamin B-6 or riboflavin, although there was a trend for vitamin B-6 to decrease tHcy. There were no statistically significant interactions between the different vitamins on changes in tHcy. Our results do not support previous suggestions that riboflavin interacts with folic acid to decrease tHcy by maximizing the catalytic activity of methylenetetrahydrofolate reductase (7). Riboflavin supplementation of older patients with biochemical deficiency of riboflavin does not affect tHcy (17). However, riboflavin status appears to be an important determinant of tHcy in homozygotes for the thermolabile (TT) variant of methylenetetrahydrofolate reductase, with high concentrations of tHcy in those with biochemical riboflavin deficiency (18). It is possible that riboflavin supplementation in this specific genetic subgroup might decrease tHcy.
The magnitude of reduction in tHcy that we observed with folic acid plus vitamin B-12 supplementation is larger than that seen in most other studies. The Vitamin Intervention for Stroke Prevention trial resulted in a reduction of 2 µmol tHcy/L with high-dose compared with low-dose folic acid plus vitamin B-6 and vitamin B-12 (7). High-dose folic acid decreased the tHcy concentration by 1.5 µmol/L in middle-aged patients with ischemic heart disease (19). The reductions seen after fortification of the diet with folic acid have ranged from 0.7 to 1.5 µmol/L (20, 21). The likely reasons for the large reduction in tHcy with folic acid plus vitamin B-12 supplementation seen in our study were the older age of our patients and the poor folate status at study entry. Both of these factors are associated with elevated tHcy concentrations.
We found no effect of folic acid plus vitamin B-12 supplementation on the von Willebrand factor, a marker of endothelial function. Previous studies of the effects of homocysteine-lowering vitamins (including folic acid) on flow-mediated (endothelium dependent) dilatation in the brachial artery have produced contradictory results (22, 23). Fibrinogen is the substrate for fibrin formation and a cofactor in platelet aggregation; therefore, increased concentrations may enhance thrombosis. We also found no effect of folic acid plus vitamin B-12 supplementation on fibrinogen. A randomized controlled trial of B-vitamin supplementation also found no significant effects of tHcy-lowering treatment on markers of clotting activation, although there was a trend for fibrin D-dimer to be reduced (24). Therefore, there is insufficient evidence to prove the hypothesis that an elevated tHcy concentration damages the endothelium and leads to a prothrombotic state. We did see a statistically significant reduction in fibrinogen in the group who received riboflavin alone, compared with placebo. However this was likely a chance finding. The marginal analysis, in which all patients who received riboflavin were compared with all who did not receive riboflavin, did not confirm any effect on fibrinogen.
Associations between tHcy and dementia (2, 25, 26)or cognitive impairment (27) have been reported. However, the relation of tHcy with cognitive decline is less certain (28). We found no significant effects on changes in cognition from tHcy lowering with folic acid plus vitamin B-12. Indeed, there was a trend for those receiving folic acid plus vitamin B-12 to do slightly worse than those not receiving this treatment, and the CIs for the effect were such that we are reasonably confident that these vitamins do not significantly improve cognition or protection against cognitive decline, at least in the short to medium term. Patients with dementia were excluded from our study; however, subjects with a mild degree of cognitive impairment (as indicated by the mean baseline MMSE score for the cohort) were included.
Several other randomized, double-blind, placebo-controlled trials have reported on the cognitive effects of folic acid with or without vitamin B-12 (29) and vitamin B-6 (30). Short-term (5 wk) treatment with folic acid, vitamin B-6, or vitamin B-12 showed no significant beneficial effects on cognition in 211 healthy women (31); this study included young and middle-aged subjects as well as 75 women aged >65 y. Similarly, no effects on cognition were seen in a 3-mo trial of vitamin B-6 supplementation in 76 healthy elderly men (30, 31). The VITAL (vitamins and acetyl-salicylic acid) trial examined the effects of 3 mo of treatment with folic acid plus vitamin B-12 in a factorial design (also including aspirin and antioxidant vitamins) in 149 subjects with dementia or mild cognitive impairment; no significant effects on global measures of cognition were seen (2). Two other studies of folic acid supplementation reported no beneficial effects on cognition in 11 (32) and 30 cognitively impaired patients (33); however, these studies were very underpowered because of the small sample sizes. Therefore, our study has extended knowledge on this issue by being the largest randomized controlled trial in older people to have examined the effects of folic acid and various B-vitamin supplements on cognition over the longest period of time.
No significant effect was seen on incident vascular events from tHcy lowering; however, this study was underpowered to detect any such effect. Two other randomized, double-blind, placebo-controlled trials have reported no significant effect of tHcy-lowering on the incidence of vascular events: the Vitamin Intervention for Stroke Prevention trial (34) and a study conducted in Cambridge in patients with ischemic heart disease (19). An open label study of folic acid in patients with coronary artery disease also found no effect (35). Other large randomized controlled trials (36) are due to report their findings soon, which should provide sufficient evidence to determine whether tHcy lowering prevents ischemic vascular disease.
Our study had several limitations. In particular, the duration was relatively short; therefore, longer-term cognitive benefits from tHcy lowering in older people cannot be excluded. Although our study, to date, is the largest randomized controlled trial to have assessed effects of these vitamins on cognition, it was not large enough to have the statistical power to exclude possible modest benefits. The need to co-prescribe folic acid with vitamin B-12 means that we cannot determine the relative effects of these vitamins on any of the outcomes; however, it was thought not to be ethical to give these vitamins separately because of the theoretical risk of harming those with covert vitamin B-12 deficiency by giving folic acid without vitamin B-12.
In conclusion we found that oral folic acid plus vitamin B-12 resulted in large reductions in plasma homocysteine in elderly patients with vascular disease. Neither riboflavin nor vitamin B-6 had any significant effects. The large reductions in tHcy achieved with folic acid plus vitamin B-12 were not accompanied by any demonstrable effects on endothelial or hemostatic function on the basis of circulating concentrations of fibrinogen and the von Willebrand factor. The lowering of tHcy concentrations with folic acid plus vitamin B-12 had no statistically significant effects on cognitive function over a 1-y period. Therefore, we found no evidence to support clinically important beneficial effects on cognitive function in older people with vascular disease from the short- to medium-term administration of these vitamins.
ACKNOWLEDGMENTS
We thank Melanie Shields and Jackie Scott (research nurses) for their hard work in patient recruitment and assessment and Mark Barber, Tricia Moylan, and Martin Whitehead for assistance with patient recruitment.
DJS contributed to the study design, supervision of clinical data collection, analysis of data, and writing of the manuscript. GM contributed to the patient recruitment and clinical data collection. GDOL, DSJO, and RCT contributed to the study design, laboratory analysis, and writing of the manuscript. AR contributed to the laboratory analysis and the writing of the manuscript. ADM contributed to the study design, randomization schedule, analysis of the data, and writing of the manuscript. PL contributed to the study design and writing of the manuscript. EGS and JBM helped supervise the clinical data collection and contributed to the writing of the manuscript. PWM provided consultation and contributed to the writing of the manuscript. RGJW provided consultation and contributed to the study design and writing of the manuscript. None of the authors had any conflicting or competing interests, and the funder of the study had no role in data collection, analysis, or interpretation of the data or in the writing of the report.
REFERENCES