Abstracts & Commentary
Synopsis: Folic acid and B vitamins may yet be proven to have a positive role through decreasing HC levels, but enthusiasm for such an approach is no longer appropriate.
Sources: Lange H, et.al. N Engl J Med. 2004;350:2673-2681. Herrmann HC. N Engl J Med. 2004:350:2708-2710.
Homocysteine (HC) has long been considered a putative risk factor for coronary disease (CAD). Many physicians prescribe routine folate supplementation, along with vitamin B6 and B12, to lower serum levels of HC, with the hopes of decreasing vascular risk. Considerable interest arose several years ago when Schnyder and colleagues reported on decreased coronary restenosis rates following angioplasty in individuals given folate therapy (N Engl J Med. 2001;345:1593-1600). Subsequently, other negative trials have appeared, and on balance, there is considerable confusion in the published literature as to whether HC is or is not a modifiable risk factor for CAD, particularly in the realm of percutaneous revascularization. This European study represents investigators from Bremen, Germany, Zwoll, and The Netherlands, who enrolled 636 patients in a randomized, placebo-controlled study assessing the efficacy of folic acid and B vitamins in individuals undergoing PCI with stenting. Standard restenosis indices were assessed using blinded, quantitative coronary-angiography. Reference diameter, minimal luminal diameter, and lesion length were calculated at base line, as well as at a 6 month angiogram. Individuals with clinical symptoms could undergo angiography prior to the scheduled follow-up procedure. Restenosis was defined as greater than 50% of luminal diameter; late luminal loss, acute gain, and a loss index were calculated. Therapy consisted of an initial intravenous bolus dose of 1 mg of folic acid, 5 mg of vitamin B6, and 1 mg of vitamin B12, followed by daily oral administration of 1.2 mg of folic acid, 48 mg of vitamin B6, and 60 µg vitamin B12, vs placebo.
Baseline characteristics were well matched; 20-25% were women; mean age was 61; 14% had diabetes; 30% smoked; and 60% had hypercholesterolemia. Forty percent of individuals had a previous MI or coronary bypass graft; 40% were on statins, 25% on ACE inhibitors, and 70% on beta-blockers; all received clopidogrel and aspirin. Routine follow-up angiography was scheduled at 6 months; target lesion revascularization (TLR) was recommended for restenosis of 75%-90%, 70 with symptoms or signs of ischemia, and in all patients when a stenosis exceeded 90%. Bare metal stents only were used. The primary angiographic end point was minimal lumen diameter within the target lesion at follow-up. The primary clinical endpoints were any event related to restenosis, including death, MI, and TLR. The results indicated that folate and vitamin B therapy did not reduce restenosis, compared to placebo; there even was a somewhat smaller 6-month-minimal lumen diameter in the treatment cohort. Women and diabetics, however, benefited from the experimental cocktail, as did individuals with high homocysteine levels (all NS). Clinically driven TLR was greater in the experimental group (7.6 vs 4.4%). The overall restenosis rate was 34.5% in the folate group, compared to 26.5 % in the placebo group P = 0.05.
Lange and colleagues conclude that the data do not provide any evidence that folic acid therapy, for primary or secondary prevention of CAD, is potentially harmful, since folate did not increase the incidence in death or infarction. However, they caution that this therapy should be considered a double-edged sword in patients with stents, with both antiproliferative and proproliferative actions potentially linked to the vitamin cocktail. The lower rates of restenosis in folate-treated women, diabetics, as well as those with markedly elevated HC, remain to be evaluated in subsequent trials. Statin therapy had no effect on restenosis rates.
Comment by Jonathan Abrams, MD
This study, in addition to another recent publication (Genser D. Am J Cardiol. 2002;89:495-499), does not support that HC is a modifiable risk factor favoring decreased restenosis. The study was carefully performed and the data appear to be highly reliable. There was a biphasic clinical response to folate, with the overall folate cohort having a smaller minimal lumen diameter than placebo at 6 months, but there was a suggestion of some benefit in women and diabetics (statistically insignificant). The treatment cocktail did lower HC levels from a mean of 12.2 µmol/L at baseline to 8.7 µmol/L at 1 month (P < 0.001), and 6 months (P < 0.001). Lange et al discuss the biochemical pathways by which HC can be metabolized. Research data that folate may promote intimal proliferation is provided, and the suggestion is made that PCI individuals without stenting may have a different mechanism of restenosis than those with stents, with the latter resulting in considerable smooth muscle cell matrix proliferation, and angioplasty alone, stimulating thrombus inflammation within the vascular cracks.
In the accompanying editorial by Herrmann, emphasis is placed on the use a cardioprotective cocktail in all patients who undergo PCI, as well as ". . . refocus on the underlying disease process in order to treat vulnerable plaques and atherosclerosis and prevent subsequent cardiac events and new and progressive atherosclerotic lesions." Thus, Herrmann recommends the use of aspirin, clopidogrel, statins, and ACE inhibitors, along with lifestyle recommendations including diabetes and hypertension control, smoking cessation, and exercise, for all patients who have receive a stent. I would emphasize that this recommendation is germane for all patients who have vascular disease, except for the, as yet, uncertain role of long-term clopidogrel. Herrmann’s emphasis is shared by me, stressing that atherosclerosis is a "chronic progressive disease," and that PCI is but a short-term, albeit dramatic, intervention which should not deflect physicians and patients from focusing on long term consequences of disease progression and its stabilization.
This study also re-emphasizes the need for well-designed clinical trials in order evaluate benefits and risks of therapies that appear to be beneficial on the surface, but may indeed not be useful or safe. The estrogen replacement story is well known, as are the negative results of antioxidant vitamins, a list to which we now must probably add folic acid replacement. In all of these areas, there were robust, although mixed, data supporting efficacy of each agent, yet large, well-designed trials resulted in red faces across academia and clinical practice, and retractions in multiple guidelines. Folic acid and B vitamins may yet be proven to have a positive role through decreasing HC levels, but enthusiasm for such an approach is no longer appropriate.
Dr. Abrams, Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque, is on the Editorial Board for Clinical Cardiology Alert.