| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
1 Department of Pharmacology, University of the Philippines College of Medicine, Manila
2 Philippine Heart Center, Quezon City
3 Department of Biochemistry and Molecular Biology, University of the Philippines College of Medicine, Manila
Jose B Nevado, Jr, MD, Tel: +240 301 482 9625, Fax: +(632) 526 4197 301 451 7090, Email: jolnev2000{at}yahoo.com, 547 Pedro Gil Street, Ermita, Manila, Philippines.
ABSTRACT
Homocysteinemia is a risk factor for cardiovascular diseases. Folic acid combined with vitamins B6 and B12 is effective in lowering homocysteine levels. This randomized placebo-controlled study was designed to determine the effect of a folic acid-based supplement on secondary prevention of clinical events in non-ST-segment elevation acute coronary syndromes. The study comprised 240 patients with either unstable angina or non-ST-elevation myocardial infarction in the previous 2 weeks who were randomized to a folate group (n =116) or a placebo group (n =124). The folate group received 1 mg folic acid, 400 µg vitamin B12, and 10 mg vitamin B6 daily. Clinical outcomes within 6 months were assessed. The composite endpoint of death, nonfatal acute coronary syndrome, and serious re-hospitalization was significantly higher in the folate group; serious re-hospitalization alone was significantly higher in this group. Advanced age and diabetes increased susceptibility to the composite outcome. Folic acid-based supplementation is not beneficial and may even be harmful in the secondary prevention of cardiovascular events in patients with unstable angina and non-ST-elevation myocardial infarction. Further studies on the safety of such supplements are suggested. Controlled Clinical Trials Registry no. ISRCTN30249553.
Key Words: Unstable angina • Angina • Unstable • Folic Acid • Homocysteine • Myocardial Infraction • Randomized Controlled Trial
INTRODUCTION
A raised plasma homocysteine level is a risk factor for cardiovascular diseases. Extensive systematic reviews summarize the numerous clinical studies and reveal a strong association between high homocysteine levels and coronary heart disease, stroke, and peripheral venous thrombosis.1,2 A causative effect of homocysteine was further supported by studies demonstrating that homocysteinemia can induce endothelial dysfunction in humans.3,4 Considering that homocysteine might be a modifiable risk factor for cardiovascular diseases, clinical trials were initiated to determine the therapeutic effect of lowering homocysteine levels. Almost all of these trials made use of folic acid, which has been shown to significantly decrease homocysteine levels. Folic acid serves as a cofactor in the main pathway for converting homocysteine to nontoxic methionine. In 1998, a meta-analysis demonstrated that folic acid supplementation of 0.5 to 5 mg per day reduced homocysteine levels by a quarter to a third. In the same study, vitamin B12 was estimated to contribute a 7% mean reduction in homocysteine levels at 0.5 mg per day.5 Daily intake of 2 mg folic acid also significantly reduced homocysteine levels by 1.8 µmol · L–1 in patients with previous myocardial infarction (MI) or unstable angina after 6 months of treatment.6 By lowering homocysteine levels with folic acid supplementation, usually in combination with vitamins B6 and B12, clinical studies have shown trends toward a parallel decrease in the occurrence of cardiovascular events such as stroke, MI, and peripheral vascular disease. In 2001, a research group in Switzerland observed that folic acid supplementation decreased by a third the rate of coronary artery restenosis in patients who had undergone coronary angioplasty.7 Since then, several clinical trials have been initiated to assess the cardiovascular effects of folic acid supplementation.8–12 Unfortunately, no significant reduction in mortality or nonfatal cardiac events was noted in any of these studies. A recent meta-analysis of randomized controlled trials did not find conclusive clinical benefit of using folic acid in cardiovascular diseases.13
Non-ST-elevation myocardial infarction (NSTEMI) and unstable angina belong to the spectrum of acute coronary syndromes (ACS) that imply partial occlusion of the coronary arteries, leading to ischemic events. These conditions are significant indicators of impending MI which can be fatal or may require intensive hospital care. Unstable angina and NSTEMI have high rates of recurrence in ACS and mortality within 6 months after an event.14 Blood homocysteine levels are higher in patients with unstable angina.15 In addition to being a risk factor, elevated homocysteine is also implicated in poorer outcomes and greater myocardial injury in unstable angina and NSTEMI.16 In epidemiologic studies, low homocysteine levels seemed to confer better long-term outcomes in patients with coronary heart disease.17 Thus, in view of current evidence, it was hypothesized that lowering homocysteine levels might reduce subsequent clinical events in patients with recent unstable angina or NSTEMI. As no randomized controlled trial on the use of folic acid-based supplementation in this subset was available, this study sought to determine whether such supplements could lower the incidence of mortality, nonfatal acute coronary events, serious re-hospitalization, or a composite of these endpoints in such patients, especially in the first 6 months after an event.
PATIENTS AND METHODS
Patients with unstable angina or NSTEMI in the previous 2 weeks were screened for inclusion in the study. The study recruited 243 patients from 5 medical centers in the Philippines: Philippine Heart Center (133), Philippine General Hospital (92), Quirino Memorial Medical Center (7), Ospital ng Maynila Medical Center (6), and East Avenue Medical Center (2). The Ethical Review Board of the Research Implementation and Development Office of the University of the Philippines College of Medicine approved the protocol.
All patients provided informed consent. However, the number of prospects assessed for eligibility was not routinely recorded, including their reasons for non-participation. Of the recruited patients, 118 were assigned to folic acid supplementation and 125 were assigned to the placebo group. Exclusion criteria included hemodynamic instability (cardiogenic shock, ongoing chest pain, unresolved and new-onset end-organ damage, and unstable congestive heart failure in the previous 2 weeks), significant liver disease (classic signs and symptoms, or liver enzyme levels 3 times the upper limit of normal, or prothrombin time >1.5 times normal), significant renal disease (creatinine >180 µmol · dL–1 or requiring dialysis), hemoglobin <10 g · dL–1, high-output failure, inability to provide adequate self-care, malignancy or any terminal illness, age <18 years, pregnancy, and residency outside Metro Manila or the adjacent provinces of Cavite and Rizal. In view of the finding that folic acid-based supplements increased the rate of in-stent restenosis, patients who had revascularization procedures were excluded.8 Those who joined the study and then had a revascularization procedure were advised to discontinue intake of supplements, but were included in the final analysis according to their original assignment. Baseline characteristics were collected by interview with a questionnaire.
The study is a double-blind placebo-controlled trial. Patients underwent randomization using predetermined randomization provided to the pharmacy by the study statistician. A member of the pharmacy staff, who had sole knowledge of patient assignment, randomly allocated and dispensed the treatment. Patients in the folate group were instructed to ingest a supplement tablet (1 mg folic acid, 400 µg vitamin B12, and 10 mg vitamin B6) daily for 6 months in addition to their prescribed treatment for unstable angina or NSTEMI. The control group received a placebo tablet daily for 6 months with their conventional treatment. Conventional treatment included anticoagulants, platelet inhibitors, beta blockers, and nitrovasodilators. Any suspected serious adverse effect of the supplement resulted in termination of the treatment.
Follow-up was conducted after 1, 3, and 6 months of treatment. The duration of treatment was based on available data that indicated >90% of adverse cardiac events after onset of NSTEMI or unstable angina occur within the first 6 months after an acute coronary event.14,18 The patients’ attending physicians also conducted follow-up examinations at their own discretion. Outcomes beyond 6 months after the start of the trial were not included in the analysis. The primary outcomes consisting of death, nonfatal ACS, and serious re-hospitalization were assessed using a hierarchal designation, that is, in decreasing priority. Accordingly, only 1 clinical endpoint was assigned to each patient. ACS was defined as acute MI, NSTEMI, or high- or medium-risk unstable angina. Serious re-hospitalization was defined as an emergency admission to hospital, which required airway or circulatory support for at least 1 day, as decided by the attending physician.
Homocysteine levels were determined using an Abbott IMx analyzer (Abbott Laboratories, Abbott Park, IL, USA). Dyslipidemia was defined as total cholesterol >190 mg · dL–1, low-density lipoprotein >160 mg · dL–1, or use of a cholesterol-lowering drug.
Analysis was intention-to-treat. Baseline characteristics and clinical endpoints were analyzed with appropriate tests, using STATA version 6.0 software (Stata Corp., College Station, TX, USA). Forest graphs were constructed using the software of Clark O, Djulbegovic B. Forest plots in excel software (data sheet), 2001, available at: www.evidencias.com. The sample size (243) was less than the projected 404 patients expected to be recruited to have a power of 0.80 to detect a 20% reduction in the composite endpoint alone, using the 31.4% incidence in the PRISM-PLUS study.14 Nevertheless, a significant divergence emerged early between the 2 groups to warrant termination of the study. Due to the inadequate sample size, adverse events are summarized and described without attempting statistical comparison.
RESULTS
The study recruited 243 subjects. However, 3 patients were excluded after recruitment (Figure 1
): 1 from each group was found to have persistently increased creatinine and required chronic renal support, despite a normal creatinine level in the emergency room; another in the folate group had persistent anemia due to thalassemia, after an initially normal hemoglobin. These patients were excluded during the first 3 days after inclusion as such conditions can theoretically augment the benefit of folic acid-based supplementation. No adverse outcome was assessed in these patients upon exclusion, and they did not receive their allocated treatment beyond 1 month, the duration for the minimum volume of supplement provided. Three patients were lost to follow-up, thus using intention-to-treat analysis, 2 in the folate group were assumed dead, while the data of 1 in the placebo group were considered missing. Compliance with the treatment was also determined (Table 1); there was no significant difference between groups. Those taking the folic acid-based supplement tended to have decreased compliance in the early part of the study, but the level of compliance became almost equal towards the end, with more patients in the placebo group sustaining their intake until 6 months. The main reasons for discontinuing treatment included the occurrence of a possible adverse event (18 patients), inability to present for follow-up due to financial constraints (48), failure to comply due to attending physician’s advice (2), transfer to areas remote from the follow-up centers (12), and unspecified reasons (2). Thus, data from 240 subjects were available for analysis. Table 2 summarizes the baseline characteristics of the participants according to treatment group. There was no significant difference in patient profiles between groups. The baseline homocysteine levels did not differ significantly despite an individual in the placebo group having a level of 159.5 µmol · L–1 (upper normal limit, 16 µmol · L–1); no significant difference was noted when this subject’s data were excluded.
|
|
|
). The composite endpoint of all 3 outcomes was significantly higher in the folate group than the placebo group. There were significantly more instances of re-hospitalization in the folate group. Most re-hospitalizations were cardiac related (3 hypertensive emergencies, 3 cases of acute pulmonary edema, and 1 of bradycardia), implying that cardiac events were increased by use of the supplements. These observations of possible harm persisted even if all patients lost to follow-up were considered as missing data. To demonstrate the time-course of the effect of folate supplementation on clinical outcomes, probability curves for death and the composite outcome were constructed (Figure 2). A trend suggestive of earlier mortality in the folate group was observed. However, after 6 months, the mortality rates did not differ significantly. In contrast, a marked divergence in the probability of event-free status was noted; this continued to diverge with time, indicating the persistence of harm in the folate group. Nonfatal outcomes (re-hospitalization and nonfatal MI) increased from the 5th week onwards in the folate group, implying that supplement intake could be harmful as early as 5 weeks after administration, and harm continued as long as the supplement was taken. Again, this finding persisted even if patients lost to follow-up were considered as missing data.
|
|
. The advanced aged subgroup (>45 years in men, >55 years in women) in the folate group was significantly more vulnerable to the composite outcome, mainly due to more serious non-ACS events. Diabetes mellitus conferred a poorer prognosis in terms of the composite outcome in the folate group, with a strong trend towards more nonfatal ACS. On the other hand, folate treatment showed a significant protective effect against nonfatal ACS in those with a previous stroke.
|
summarizes the non-serious adverse events reported during the study. Combining the serious non-ACS-related events and non-serious adverse events, more events were noted in the folate group. Dyspnea was the most common adverse effect, followed by dizziness and rashes. Dizziness was noted in 3 patients, all in the folate group. Raised blood pressure and rashes were not discriminated as side effects of the supplement, because they were also observed in the placebo group. Most of these events in the folate group occurred within 1 month after the onset of the treatment, suggesting a causative effect of the active ingredients in the treatment.
|
The study found that the composite endpoint of death, nonfatal ACS, and serious re-hospitalization was increased by the use of a folic acid-based supplement in patients with medium- or high-risk unstable angina or NSTEMI. Notably, a consistently harmful trend was observed for the other primary endpoints (death and nonfatal ACS), although the small sample size limited the detection of significance. While the incidence of all primary outcomes was higher in the folate group, the most significant increase was in the number of serious re-hospitalizations. Most of the events were cardiovascular in nature, implying that most re-hospitalizations were due to cardiac complications, and such additional events could be direct effects of the multivitamin intake. The presence of increased harm in the folate group was highly probable. Moreover, the fact that serious re-hospitalizations usually occurred earlier in the folate group than the placebo group also implies possible harm due to the use of the supplement.
Similar RCTs on the use of folic acid in coronary disease suggest potential harm. In the HOPE 2 trial, an increase in re-hospitalizations, other than those due to nonfatal MI and stroke, is be inferred.9 In fact, the data showed a significantly higher rate of re-hospitalization due to unstable angina in the multivitamin group. The occurrence of heart failure was also higher in the folate group, although no significant difference was detected. In the Norwegian Vitamin (NORVIT) trial, which was carried out in a country without mandatory folic acid supplementation in the diet, trends towards harm were also observed that could be due to folic acid-based supplements, in terms of mortality, MI, and composite cardiovascular outcomes.10 Our study is consistent with NORVIT and HOPE 2 as it noted that folic acid-based therapy is not beneficial in the secondary prevention of death and acute coronary events in patients with coronary heart disease. From an ethical point of view, this study underscores the notion that even for vitamins and other medications generally perceived to be safe, the need of safety studies is crucial before such interventions can be tested for their efficacy. To underscore our point, in a systematic review of relevant randomized placebo-controlled clinical trials of folate-based supplements in ischemic heart disease, we found no significant reductions in rates of mortality or cardiac events (Figure 4).7–11,19 The NORVIT study suggests that the harmful effects of these supplements could be due to vitamin B6.10 The individual and interactive effects of folic acid and vitamin B12 together with vitamin B6, should be assessed and clarified further, including their possible toxic metabolites and harmful interactions with other drugs. In this study, time-course analysis suggested that the harmful effects persisted during continued intake. Therefore, it is strongly recommended that studies on the safety of folic acid-based interventions should be undertaken prior to any further trials involving patients with coronary heart disease, including those involving other outcomes such as neuropsychiatric and cerebrovascular parameters.
|
The study also showed that clinical outcomes of folic acid-based supplementation were better in patients with previous stroke, especially in terms of nonfatal acute coronary events. Although this trial did not assess secondary stroke prevention, other clinical studies have shown a significant reduction in the occurrence of stroke.15,16 Combined, these findings suggest that the mechanism of cerebrovascular disease may not be similar to vasoocclusive disease in other parts of the body, such as the heart. Thus, folic acid supplementation might be selectively beneficial to cerebral pathologic lesions by mechanisms not necessarily pertinent to atherosclerosis, or even to the action of homocysteine. This issue could be addressed in subsequent studies.
It was also observed that there were more non-serious adverse events in the folate group, including skin irritation and rashes, as previously reported.7,19 Dyspnea was more common in the folate group. Dizziness was noted in the folate group but not in the placebo group; neither are commonly cited side effects of folic acid, vitamin B6 or B12. Hence, it is speculated that these could be idiopathic reactions in Filipinos, or more possibly, overlooked side effects. These side effects should be considered in subsequent human studies.
It should be noted that the study had significant limitations. The sample size was not large enough to detect significant differences in specific clinical outcomes. The compliance rate of approximately 60% probably attenuated the effect of folate treatment. On the other hand, such a compliance rate may actually reflect the real-life compliance rate that approximates genuine harm or benefit. The fact that this study was performed in a developing country such as the Philippines should also be considered in view of the lack of mandatory folic acid supplementation in food, and potential ethnical differences. Despite these limitations, it is not recommended to repeat the study in a larger sample, due to the safety issues. Alternatively, animal or cellular models might be more appropriate. Research on the mechanism of homocysteine-induced injury is clearly needed to direct alternative and more specific therapy for homocysteinemia.
ACKNOWLEDGMENTS
This study was made possible through the funds of the Department of Science and Technology-Grant-in-Aid of the Republic of the Philippines, under the supervision of the Philippine Council for Health Research and Development.
We would like to thank the following for their kind support: Senator Ramon Magsaysay Jr. for providing the research funds through his initiative; Dr. Felix Eduardo Punzalan for his advice, Ms. Marie Jeanne Berroya and Ms. Rosario Hapin for patiently supervising our progress; Pharmacare (Philippines) for supplying the supplements and placebo; Mark Anthony Javelosa, MSc, for the statistical consultation and the Department of Biochemistry and Molecular Biology of the UP College of Medicine and the Department of Medicine of the UP-Philippine General Hospital for administrative support.
Footnotes
4Drs. Imasa and Nevado contributed equally to the study
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:13-21
© 2009 by SAGE Publications
DOI: 10.1177/0218492309102494
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
p = 0.0606. 
p = 0.0275. 
p = 0.0694. ¶p = 0.0318. ACS = acute coronary syndrome, BMI = body mass index.