Vitamin B-12 supplementation improves arterial function in vegetarians with subnormal vitamin B-12 status

Abstract

Objective

Vegetarians are more vascular-healthy but those with subnormal vitamin B-12 status have impaired arterial endothelial function and increased intima-media thickness. We aimed to study the impact of vitamin B-12 supplementation on these markers, in the vegetarians.

Design

Double-blind, placebo controlled, randomised crossover study.

Setting

Community dwelling vegetarians.

Participants

Fifty healthy vegetarians (vegetarian diet for at least 6 years) were recruited. Intervention: Vitamin B-12 (500µg/day) or identical placebo were given for 12 weeks with 10 weeks of placebo-washout before crossover (n=43), and then open label vitamin B-12 for additional 24 weeks (n=41).

Measurement

How-mediated dilation of brachial artery (FMD) and intima-media thickness (IMT) of carotid artery were measured by ultrasound.

Results

The mean age of the subjects was 45+9 years and 22 (44%) were male. Thirty-five subjects (70%) had serum B-12 levels <150pmol/l. Vitamin B-12 supplementation significantly increased serum vitamin B-12 levels (p<0.0001) and lowered plasma homocysteine (p<0.05). After vitamin B-12 supplementation but not placebo, significant improvement of brachial FMD (6.3±1.8% to 6.9±1.9%; p<0.0001) and in carotid IMT (0.69±0.09mm to 0.67±0.09 mm, p<0.05) were found, with further improvement in FMD (to 7.4±1.7%; p<0.0001) and IMT (to 0.65±0.09mm; p<0.001) after 24 weeks open label vitamin B-12. There were no significant changes in blood pressures or lipid profiles. On multivariate analysis, changes in B-12 (β=0.25; p=0.02) but not homocysteine were related to changes in FMD, (R=0.32; F value=3.19; p=0.028).

Conclusions

Vitamin B-12 supplementation improved arterial function in vegetarians with subnormal vitamin B-12 levels, proposing a novel strategy for atherosclerosis prevention.

Introduction

It is commonly believed that vegetarian diets are cardioprotective, but epidemiologic studies of vegetarians have failed to show significant reductions in cardiovascular mortality, when compared with omnivores of similar lifestyles (1). Certain groups of vegetarians have a high prevalence of vitamin B-12 deficiency (70%), high salt intake and raised serum triglyceride concentrations (2, 3), which may predispose to atherosclerosis. We have recently reported that Chinese vegetarians have impaired arterial endothelial function and intima-media thickening (4), key early events in atherogenesis and cardiovascular diseases (5., 6., 7.), predictive of cardiovascular and stroke outcomes (6., 7., 8., 9.). We therefore aimed to study whether vitamin B-12 supplementation in vegetarian with subnormal vitamin B-12 status can improve these markers of early arterial disease, as a novel preventive strategy for atherosclerotic diseases.

Subjects and methods

To evaluate the impact of vitamin B-12 supplementation on atherogenic processes, 50 community dwelling vegetarians were studied. They had been vegetarians for six years or more (median 14 years), life-long non-smokers, had no known major vascular, renal or hepatic diseases and were not on regular medications or vitamins supplements. A locally validated food frequency questionnaire was administered to estimate dietary intake over the previous one week (10). After enrolment the subjects were randomized to receive oral vitamin B-12 (500mcg/day) or image-matched placebo for 12 weeks, in a double-blind cross-over fashion, with 10 weeks washout before cross-over, and subsequently open label vitamin B-12 for additional 24 weeks (Figure 1). Subjects were reviewed at the beginning and end of each 12-week period. On each occasion, subjects attended after a 14-h fast (except for the study medication) for a blood test, measurement of resting supine blood pressure and a vascular ultrasound study (as described below). Ultrasound studies were analysed by a single observer, blinded in every case to subject identity and to the stage of the experiment, as described in previous similar studies from our group (11, 12). Vitamin B-12 and folate levels in serum were measured by immunoassay (DPC dual count solid-phase noboil kits), with the upper limit measured being 1771pmol/liter and 85.8nmol/liter respectively by serial dilution, and total homocysteine levels were assessed on frozen stored plasma samples by enzymatic immunoassay (Abbott IMX Analyzer; Abbott Park, Illinois) (13). This study complies with the Declaration of Helsinki, in that our institutional appointed ethics committee has approved the research protocol and informed consent was obtained from the subjects.

Figure 1.

Design of cross-over double-blind randomised placebo- controlled trial of vitamin B-12 / placebo supplementation in vegetarians, with assessment at baseline (visit 1), week 12, weeks 22 and week 34 (V2-V4), followed by open-lable vitamin B-12 supplementation for 24 more weeks (V5)

Vascular studies

Endothelial function of the brachial artery was studied using high resolution ultrasound, as described previously (5, 11, 12). In brief, the diameter of the brachial artery was measured on B- mode ultrasound images, using a linear array transducer (L10- 5) with a median frequency of 7.5MHz and a standard Advanced Technology Laboratories 3000 system, and forearm tourniquet cuff placement to induce reactive hyperemia. Scans were acquired at rest, during reactive hyperemia (to induce flow-mediated endothelium-dependent dilation, FMD) and after 200μg sublingual glyceryltrinitrate (GTN, an endotheliumindependent dilator). FMD is predominantly due to endothelial nitric oxide release (14), and endothelial responses of the brachial artery measured by this method correlate significantly with both coronary endothelial function in the same subjects (15) and with the extent of coronary atherosclerosis (16). The accuracy, reproducibility and low intraobserver error for this measurement of arterial physiology have been demonstrated previously (17), as reported in our previous experiments (a mean relative difference of 5% in FMD over time) (18).

All carotid scans were performed by a single operator (PC), according to the standardized scanning protocol for the right and left carotid arteries described by Salonen and Salonen (19) and Blankenhorn et al (20), using images of the far wall of the distal 10 mm of the common carotid arteries. All scans were recorded on super-VHS videotape for subsequent off-line analysis for intima-media thickness (IMT), using a validated automatic edge-detecting and measurement software package, as described previously (21). The interobserver variability of mean IMT was 0.003±0.01mm (CV <1.0%).

Statistical Analysis

Descriptive data are expressed as mean ± SD. The primary study endpoints were the comparison between brachial FMD and carotid IMT on vitamin B-12 or on placebo, using the paired Student’s t-test. Other outcome variables were compared and adjusted for multiple comparisons (22).

Determinants of the vitamin B-12 related improvement in FMD and IMT (calculated for each subject) were explored by standard univariate and multivariate regression analyses, with age, changes in serum vitamin B-12 and plasma homocysteine levels as the independent variables. Statistical significance was inferred at a two-tailed p value <0.05.

Results

Mean age of the 50 subjects was 45±9 years and 22 (44%) were male, with the median duration of vegetarianism being 14 years (range 6-40 years). Two subjects were vegans taking no dairy products. Their baseline nutritional and clinical parameters were characterized by normal blood pressures and renal function, relatively low protein and cholesterol intake, lower range of blood cholesterol, glucose and vitamin B-12, but relatively high homocysteine, folate and triglyceride levels (Table 1). Twelve subjects (24%) had serum vitamin B-12 between 150-300 pmol/l and 35 subjects (70%) <150 pmol/l respectively. Forty-three subjects completed the 34 weeks randomized crossover trial, and 7 subjects (without any adverse reaction) withdrew consent before completion. Twenty one subjects took vitamin B-12 initially and placebo subsequently, and vise-versa in 22 subjects (Figure 1). Forty-one of these forty-three subjects agreed and completed the subsequent 24 week open label trial of vitamin B-12 supplement.

Table 1.

Baseline characteristics of 50 vegetarian subjects

Age (years)	45+9
Male (%)	44 (22/50)
Nutritional Intake Calorie (Kcal/day)	1871.2+587.3*
Protein (g/day)	82.9+36.7
Fat (g/day)	59.8+22.8
Systolic blood pressure (mmHg)	122+14
Diastolic blood pressure (mmHg)	80+1
Body mass index (kg/m2)	23.8+2.9
Waist-hip ratio	0.84+0.06
Vitamin B12 (pmol/l)	134+126
Folate (nmol/l)	39.4+10.4
Homocysteine (mmol/l)	16.7+11.0
Glucose (mmol/l)	4.8+0.6
Cholesterol (mg/day)	276.7+151.2
Total cholesterol (mmol/l)	4.6+0.9
Triglyceride (mmol/l)	1.6+0.9
HDL-cholesterol (mmol/l)	1.4+0.3
LDL-choleserol (mmol/l)	2.5+0.8

^*Expressed as the mean ± standard deviation; HDL: High density lipoprotein; LDL: Low density lipoprotien

The major outcomes are shown in Table 2 and Figure 2. After vitamin B-12 supplementation, there was a significant increase in vitamin B-12 and fall in homocysteine levels but no significant changes in blood pressure, serum glucose or LDL- cholesterol. Brachial FMD significantly increased (6.3±1.8% to 6.9±1.9%; p<0.0001) and carotid IMT decreased (0.69±0.09mm to 0.67±0.09mm; p<0.05) after vitamin B-12 but not after placebo treatment. There were no significant changes in GTN responses after either treatment period. On multivariate analysis, changes in vitamin B-12 (β=0.25; p=0.02) but not homocysteine level correlated with changes in FMD, after adjustment for age (model R value=0.32; F value=3.19; p=0.028).

Table 2.

outcomes of 12 week randomized placebo controlled crossover trial of oral vitamin B-12 supplement (500 mcg daily) in 43 vegetarians

	Baseline	After Placebo	After Vitamin B-12
Systolic BP (mmHg)	l22±l4	117±13	117±16
Diastolic BP (mmHg)	8l±l2	78±9	79±10
B-l2 (pmol/l)	l34.0±l25.6	185.7±145.4*	379.6±206.2#
Homocysteine (µmol/l)	l6.7±ll.0	13.1±5.0*	11.3±6.0**
Folate (nmol/l)	39.4±l0.4	39.2±14.2	40.0±12.3
LDL-C (mmol/l)	2.5±0.8	2.4±0.7	2.4±0.8
HDL-C (mmol/l)	l.4±0.3	1.4±0.4	1.4±0.3
Glucose (mmol/l)	5.l±0.9	5.1±1.0	5.1±0.6
Reactive Hyperemia (%)	883.9±97.l	900.7±99.9	920.7±75.5
GTN (%)	17.1±2.7	18.2±2.0	18.8±2.1

Compared with baseline

^*P<0.05;

^**p<0.01: #p<0.0001,paired t test. Reactive hyperemia = Velocity-time integral2 × Heart Rate2/Velocity-time integrall × Heart Ratel x100 BP: blood pressure; FMD: brachial flow mediated dilation; GTN: glyceryltrinitrate; HDL-C : high density lipoprotein bound cholesterol; IMT: intima media thickness; LDL-C - low density lipoprotein bound cholesterol

Figure 2.

Changes in brachial flow-mediated dilation (FMD) during vitamin B-12 / placebo supplementation (12 weeks, n=43) and open label vitamin B-12 supplementation (24 weeks) in 41 vegetarians

Further improvement in brachial FMD (7.4±1.7%, P <0.001) and carotid IMT (0.65±0.09mm, P<0.05) were documented after the additional 24 weeks open label vitamin B-12 treatment (Figure 2 & 3). No significant adverse effects including blood biochemical profiles and skin reaction were reported throughout the double-blind or open label treatment periods.

Figure 3.

Changes in carotid intima-media thickness (IMT) during vitamin B-l2 / placebo supplementation (l2 weeks, n=43) and open label vitamin B-l2 supplementation (24 weeks) in 4l vegetarians

Outcomes of 12 week randomized placebo controlled crossover trial of oral vitamin B-12 supplement (500 mcg daily) in 43 vegetarians

Discussion

We have found that vitamin B-l2 supplementation in vegetarians with subnormal vitamin B-l2 status significantly enhanced arterial endothelial function compared with placebo, and was associated with lower carotid IMT thickness. The effects appeared to be better correlated with the correction of vitamin B-l2 deficiency than with the lowering of homocysteine. We have incorporated l0 weeks washout before crossing over, to eliminate the possible carrying over effect of vitamin Bl2. However some vitamin B-l2 effects were still observed after cross-over to placebo, but this nevertheless would only under-estimate the benefit of vitamin Bl2 on endothelial function.

Vegetarian diets are generally believed beneficial to cardiovascular health. Vegetarians have been consistently shown to have lower serum cholesterol (23), and some short term trials have shown that vegetarian diets can lower blood pressure (24). Cohort studies however, have failed to show a reduction in cardiovascular mortality, especially among vegans who have more restrictive diets (l). We have previously reported that Chinese vegetarians have impaired endothelial function and increased carotid intima thickness, when compared with control subjects (4). These vegetarians showed a high prevalence (70%) of vitamin B-l2 deficiency, higher systolic blood pressures and serum triglycerides.

The majority (94%) of our trial subjects had subnormal vitamin B-l2 status. There have been controversies about the definition of vitamin B-l2 deficiency because of variability in clinical manifestations. It is generally agreed that metabolic deficiency, as evidenced by raised plasma methylmalonic acid or homocysteine, would constitute a compromized vitamin B- l2 status. Epidemiologic studies have shown that serum vitamin B-l2 level lower than 300pmol/l is associated with a significant risk of metabolic vitamin B-l2 deficiency (25).

Our subjects were carefully selected for freedom from chronic diseases. As folate intakes are high in the vegetarian diet, the high plasma homocysteine in our subjects were mostly attributable to vitamin B-12 deficiency. Raised homocysteine has been found to be an independent risk factor for endothelial dysfunction, cardiovascular diseases and stroke (11, 26). However large scale randomized placebo-controlled trials of homocysteine-lowering, by B vitamin supplementation in high risk subjects with normal vitamin B-12 and folate status, have reported conflicting results on benefit on cardiovascular and stroke outcomes (27., 28., 29., 30., 31., 32., 33.). Our studied population had a low vitamin B-12 levels, and we found an independent improvement in FMD correlated better with the increase in serum vitamin B-12, than with changes in plasma homocysteine. More recent studies have shown that other related metabolites in homocysteine metabolism e.g methyltetrafolate (34) and S-adenosylmethionine (SAM) (35) correlated better with FMD and carotid IMT than homocysteine. SAM is a metabolite of methionine, and is a major methyl donor potentially related to endothelial health (35). Vegetarians have low methionine intake and therefore are dependent on the re-methylation pathway of homocysteine (36), in which vitamin B12 and folate are co-enzymes. The daily methionine intake and blood SAM were not measured. However, in the context of vitamin B-12 deficiency in our vegetarian cohort, one might expect a decrease in the level of SAM and this should be readily reversible by vitamin B-12 supplementation, especially in the setting of normal folate levels.

Vitamin B-12 deficiency is prevalent among older people and vegetarians and is mostly asymptomatic. For vegetarians, apart from direct vitamin B-12 supplementation, vitamin B-12 deficiency is also preventable by higher and adequate intake of dairy products, eggs and fortified cereals.

There are several limitations in this study. Firstly, all our subjects were vegetarians who were folate replete with otherwise favorable cardiovascular risk profiles. The effects of vitamin B-12 supplementation in older patients with cardiovascular risk factors and possibly suboptimal folate status requires further study. Secondly the effect of vitamin B-12 supplementation over 24 weeks on carotid intima-media thickness was quite subtle and of uncertain biological significance. A longer term vitamin B12 supplementation is needed to confirm the benefit on carotid IMT. Moreover, whether these changes in surrogate markers of vascular health can be translated into prevention of clinical events in longer term needs to be tested further. Thirdly, the vegetarians in this study had relatively low vitamin B-12 status. It is uncertain that vitamin B-12 supplementation will have the same effects in vegetarians with better vitamin B-12 status.

We conclude that vitamin B-12 supplementation in asymptomatic vegetarians with compromized vitamin B-12 status may lead to a significant improvement in arterial endothelial function and carotid intima-media thickness, with potential benefit on cardiovascular health.

Disclosures

None