|
|
||||||||||
|
J Am Coll Cardiol, 2004; 43:629-634, doi:10.1016/j.jacc.2003.08.051 © 2004 by the American College of Cardiology Foundation |



* Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
Vanderbilt University, Nashville, Tennessee, USA
University of California at San Diego, La Jolla, California, USA
Children's Hospital and Harvard Medical School, Boston, Massachusetts, USA
Manuscript received June 3, 2003; revised manuscript received August 1, 2003, accepted August 6, 2003.
* Reprint requests and correspondence: Dr. Scott Kinlay, Cardiovascular Division, Brigham and Women's Hospital, 75 Francis Street, Boston, Massachusetts 02115, USA.
skinlay{at}partners.org
| Abstract |
|---|
|
|
|---|
BACKGROUND: Endothelial function is a sensitive indicator of vascular health. Oxidant stress and oxidized low-density lipoprotein (LDL) impair endothelial function by reducing nitric oxide bioavailability in the artery wall.
METHODS: We randomly assigned 30 subjects with CAD to combined vitamin E (800 IU per day) and C (1,000 mg per day) or to placebos in a double-blind trial. Coronary artery endothelial function was measured as the change in coronary artery diameter to acetylcholine infusions (n = 18 patients), and brachial artery endothelial function was assessed by flow-mediated dilation (n = 25 patients) at baseline and six months. Plasma markers of oxidant stress (oxidized LDL and autoantibodies) were also measured.
RESULTS: Plasma alpha-tocopherol (p < 0.001) and ascorbic acid (p < 0.02) increased with active therapy. Compared to placebo, there was no improvement in coronary and brachial endothelial vasomotor function over six months. Although vitamins C and E tended to reduce F2-isoprostanes (p = 0.065), they failed to alter oxidized LDL or autoantibodies to oxidized LDL.
CONCLUSIONS: Long-term oral vitamins C and E do not improve key mechanisms in the biology of atherosclerosis or endothelial dysfunction, or reduce LDL oxidation in vivo.
| ||||||||
Short-term infusions of high concentrations of the antioxidant vitamin C can correct endothelial dysfunction in patients with a variety of cardiovascular risk factors (4,5). These studies provided evidence that oxidant stress impairs endothelial function and served as a "proof of principle" that antioxidants can improve endothelial function. Yet, in recent large-scale clinical trials, oral antioxidants failed to reduce cardiovascular events (69). Although the large trials have been criticized (10), the long-term effects of oral antioxidant vitamins on key biologic features of atherosclerosis are poorly defined in humans in vivo.
We conducted a randomized trial of the long-term effect of high-dose oral vitamins C and E on coronary endothelial function (primary end point) and the secondary end points of brachial artery endothelial function and oxidant stress in patients with stable coronary artery disease (CAD).
| Methods |
|---|
|
|
|---|
Treatment. The active treatment was natural vitamin E (RRR-alpha-tocopherol) (Bronson Labs, American Fork, Utah) 400 IU twice per day and vitamin C (ascorbic acid) (Bronson Labs) 500 mg twice per day. The staff of Investigational Pharmacy encapsulated the active and placebo vitamins and randomly allocated subjects to combined vitamins or placebos using a random number table. Treatment was stopped 12 h before final assessment of endothelial function to minimize acute effects of these drugs.
Coronary endothelium-dependent vasomotor function. Of the 25 patients who completed the study, 4 (16%) had mild single-vessel disease with a stenosis 30% to 50%, 17 (68%) had single-vessel disease with a stenosis >50%, 2 (8%) had two-vessel disease, and 2 (8%) had three-vessel disease (including a >50% stenosis in a side branch of a study vessel).
Coronary endothelium-dependent vasomotor function was assessed using a protocol established in our laboratory (12,13). The study vessel was selected solely on technical suitability. Control, acetylcholine (Miochol, OMJ Pharmaceuticals Inc., San German, Puerto Rico) at estimated final concentrations of 10-8M, 10-7M, 10-6M, and nitroglycerin at 25 µg/min were administered selectively into the study coronary artery at 0.8 ml/min for 3 min.
Coronary artery diameters were measured using quantitative coronary angiography (CMS-QCA, MEDIS, Leiden, Netherlands), blind to treatment and drug infusions. Vasomotor responses were calculated as percent change from baseline and the results from three artery segments were averaged (12).
Brachial artery endothelium-dependent vasomotor function. Brachial artery endothelial function was assessed within 12 h of the coronary study under the same conditions. The brachial artery was imaged in longitudinal section with a Toshiba SSH-140A with a 7.5-MHz linear array transducer (14). A blood pressure cuff was positioned distal to the transducer, inflated to 50 mm Hg above systolic pressure for 5 min, and then released. The artery was scanned 50 to 70 s after cuff deflation (flow-mediated dilation) and 3 min after sublingual nitroglycerin (0.4 mg) (14). Hyperemia was measured from the Doppler velocity time integral. The ultrasound images were digitized at end-diastole and brachial artery diameters were measured blind to treatment assignment using a semi-automated edge detection program (CVI Analysis, Information Integrity, Stow, Massachusetts). Flow-mediated dilation and nitroglycerin dilation were reported as percent change from baseline diameter.
Measurement of blood vitamin levels and oxidant stress. Blood specimens were taken at baseline and six months and plasma was stored at 70°C. Plasma alpha-tocopherol and ascorbic acid concentrations were measured using high-performance liquid chromatography. Plasma free F2-isoprostanes were measured by stable-isotope-dilution mass spectrometry (15) in a single-batch run. The ratio of oxidized phospholipid epitopes associated with apoB-100 (detected by monoclonal antibody EO6) to number of apoB-100 particles (detected by monoclonal antibody MB24) served as an index of oxidized plasma LDL (16). Autoantibodies to oxidized LDL (immunoglobulin G [IgG]) were determined using alkaline phosphatase-labeled goat anti-human IgG and a chemiluminescent technique using Lumi-Phos (Lumigen). Antibody binding was expressed as relative light units in 100 ms (16).
Statistical analysis. The primary and secondary end points were absolute change in endothelial function over six months. The change in coronary artery dilation to acetylcholine was skewed and reported as medians and interquartile (25% to 75%) range, and differences were tested by the rank-sum test. The change in brachial artery endothelial function was normally distributed and reported as means (and standard errors) and differences assessed by unpaired t tests. Similarly, the changes in oxidized LDL and its IgG autoantibody were normally distributed and differences assessed by unpaired t tests. All statistical tests were intention-to-treat analyses using STATA (Statacorp, College Station, Texas).
Power of the study. We used the variability of change in coronary acetylcholine response observed in the vitamin group (standard deviation = 13%) to estimate the power of the study to find important changes in coronary endothelial function. In other studies of coronary endothelial function, therapies that reduce coronary events generally improve the acetylcholine response by 15% to 25% (12,17,18). With a standard deviation of 13%, this study had a 74% power of finding a 16% difference and a 90% power of finding a 20% difference at the 0.05 level.
| Results |
|---|
|
|
|---|
|
Plasma vitamin concentrations. The baseline vitamin concentrations were similar in both groups of subjects. Over six months, the plasma alpha-tocopherol and ascorbate concentrations doubled in the subjects assigned to vitamins and were significantly greater than the change in the placebo group (Table 2).
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Oral antioxidant vitamins and endothelial function. The role of increased oxidant stress on endothelial dysfunction in patients with a variety of coronary risk factors is supported by the reversal of this dysfunction with intra-arterial delivery of supraphysiologic doses of the antioxidant vitamin C (4).
However, studies that investigated long-term administration of oral antioxidant vitamins that achieve high physiologic concentrations have yielded contradictory effects on peripheral artery endothelial function (4,5). Some studies show improvement in peripheral endothelial function with oral supplementation of vitamin E (19,20) or vitamin C (21,22), whereas others show a lack of benefit for vitamin E (23,24), vitamin C (25,26), or their combination (27). Our study was the first to examine the effect of oral antioxidant vitamins on coronary endothelial function.
The inconsistent results of previous studies may be related to several factors. To effectively scavenge superoxide vitamin C must be present at concentrations of 1 to 10 mmol/l or higher (28). These concentrations were achieved in short-term parenteral studies, but long-term oral administration typically raises vitamin C concentrations to only 0.1 mmol/l (28), similar to the present study. In the short term, vitamin C could also improve endothelial function by changes in intracellular redox state by sparing intracellular glutathione (29), by reduced degradation of tetrahydropterina co-factor required for nitric oxide synthesis (30), and by reduced protein kinase C activation (31). The negative results of the present study suggest that the short-term effects may not persist with long-term administration of vitamin C.
Vitamin E is a lipophilic antioxidant that does not scavenge superoxide (32). It acts by protecting LDL from oxidation (33), a property that would preserve nitric oxide (2). However, compared to other antioxidants, vitamin E exerts a weak antioxidant effect (3335).
Antioxidant vitamins and oxidant stress. Quantitative measurement of oxidant stress in vivo is difficult but can be based on measuring oxidatively modified molecules such as F2-isoprostanes (oxidized arachidonate containing phospholipids [15]) and circulating oxidized LDL (16). Oxidized LDL is highly immunogenic as well as atherogenic, and autoantibodies to oxidized LDL are another validated measure of oxidant stress (16). Although we found that the combination of vitamins C and E modestly reduced plasma F2-isoprostanes, it did not lower circulating oxidized LDL or autoantibodies to oxidized LDL, and this may explain the lack of benefit on endothelial function.
Power of the study. Because of ethical and technical constraints of invasive investigations, studies of coronary endothelial function typically investigate fewer subjects than other end point studies. Fortunately, testing of coronary endothelial function is highly reproducible, and changes can be demonstrated in relatively small sample sizes. The sample size of nine subjects in each group in our coronary study was within the range of other studies that found improvements in coronary endothelial function with cholesterol-lowering therapies (9 to 11 subjects per group) (17,36). If antioxidant vitamins had an effect comparable to cholesterol-lowering therapies, our sample size had a 74% to 90% power of finding a difference at the 0.05 level. Although we can be confident that the present study excluded a substantial and hence clinically relevant benefit of antioxidants on coronary endothelial function, a modest benefit could be excluded only by a much larger study.
Clinical implications. Antioxidant vitamins C and E are used widely in clinical practice. However, recent large-scale clinical trials of vitamin E (6,8,9) or vitamins E and C (7) have failed to reduce cardiovascular end points either in combination with other preventive therapies or alone. Our study provides a plausible explanation, as the high doses of vitamins used in this study were similar to or exceeded the doses of the large clinical trials, yet did not improve key features in the biology of atherosclerosis: endothelial function or lipid oxidation. The data in this study do not disprove the oxidant stress hypothesis. They do suggest that more potent and effective antioxidants are needed in this setting.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Riccioni, T. Bucciarelli, B. Mancini, F. Corradi, C. Di Ilio, P. A. Mattei, and N. D'Orazio Antioxidant Vitamin Supplementation in Cardiovascular Diseases Ann. Clin. Lab. Sci., January 1, 2007; 37(1): 89 - 95. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Castilla, R. Echarri, A. Davalos, F. Cerrato, H. Ortega, J. L. Teruel, M. F. Lucas, D. Gomez-Coronado, J. Ortuno, and M. A Lasuncion Concentrated red grape juice exerts antioxidant, hypolipidemic, and antiinflammatory effects in both hemodialysis patients and healthy subjects Am. J. Clinical Nutrition, July 1, 2006; 84(1): 252 - 262. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. H. Stone, D. M. Lloyd-Jones, S. Kinlay, B. Frei, W. Carlson, J. Rubenstein, T. C. Andrews, M. Johnstone, G. Sopko, H. Cole, et al. Effect of Intensive Lipid Lowering, With or Without Antioxidant Vitamins, Compared With Moderate Lipid Lowering on Myocardial Ischemia in Patients With Stable Coronary Artery Disease: The Vascular Basis for the Treatment of Myocardial Ischemia Study Circulation, April 12, 2005; 111(14): 1747 - 1755. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. A. Economides, L. Khaodhiar, A. Caselli, A. E. Caballero, H. Keenan, S.-E. Bursell, G. L. King, M. T. Johnstone, E. S. Horton, and A. Veves The Effect of Vitamin E on Endothelial Function of Micro- and Macrocirculation and Left Ventricular Function in Type 1 and Type 2 Diabetic Patients Diabetes, January 1, 2005; 54(1): 204 - 211. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |