VISP and INVEST

Summary. VISP1 enrolled 3680 adults with nondisabling cerebral infarction within 120 days at 56 hospitals, medical centers, and neurology private practices in the United States, Canada, and Scotland. Patients were randomly assigned to 25 mg pyridoxine (vitamin B6), 0.4 mg cobalamin (vitamin B12), and 2.5 mg folic acid (high-dose formulation) or 200 μg pyridoxine, 6 μg cobalamin, and 20 μg folic acid (low-dose formulation). All participants received referenced daily multivitamins and the best available medical and surgical management. The main outcome measures were recurrent cerebral infarction (primary outcome), coronary heart disease (CHD) events, and death (secondary outcomes). After a mean follow-up of 2 years, the unadjusted risk ratio for stroke, CHD event, or death was 1.0 (95% confidence interval, 0.8–1.1). The risk for ischemic stroke was 9.2% in the high-dose group and 8.8% in the low-dose group (risk ratio, 1.0; 95% confidence interval, 0.8–1.3). Moderate reduction of homocysteine levels had no effect on vascular outcomes, but there was a graded association between graded baseline homocysteine levels and cardiovascular risk. Comment. Homocystinuria, with high plasma homocysteine levels, has long been known to be associated with cerebrovascular disease. Many studies have demonstrated an association between elevated homocysteine and stroke. Elevations of homocysteine levels above 20 μmol/L have been shown to increase CHD eight-fold.2 The risk for homocysteine appears independent of other established risk factors. Folic acid, in combination with vitamin B6 and vitamin B12, has been used to decrease plasma homocysteine levels on the assumption that cardiovascular risk may be reversed, presumably by moderation of endothelial dysfunction. VISP assessed the effects of this intervention in patients with previous ischemic stroke; participants required homocysteine plasma levels at or above the 25th percentile of levels in the North American stroke population. There was a slight decrease in homocysteine level entry requirements throughout the study because the population homocysteine levels declined from 10.5 μmol/L initially (1997) to 9.5 μmol/L for men and 8.5 μmol/L for women (1999). Because of the randomization protocol, there were no differences in characteristics of the qualifying stroke in the high-dose and low-dose vitamin groups. The entry characteristics of the participants (mean between groups) were: 16% smokers, body mass index 28.4, total cholesterol level 203 mg/dL, low-density lipoprotein cholesterol level 122 mg/dL, high-density lipoprotein cholesterol level 45 mg/dL, triglycerides 174 mg/dL, and blood pressure 141/78 mm Hg. The only significant difference in these entry characteristics was a 2.8% higher smoking rate in the high-dose vitamin group. However, even with optimal medical management, the mean body mass indices and blood pressures did not decrease, and the mean low-density lipoprotein cholesterol remained above the optimal <100 mg/dL (mean 112 mg/dL). A total of 15% were smoking at 1 year (approximately 3.7% quit smoking during that 1 year and 3.4% started smoking). At the end of the study, mean total homocysteine levels were lower in the high-dose group (by 2.0 μmol/L), and folate and vitamin B12 levels were higher. There was little or no change in these levels in the low-dose group. The only significant differences in clinical outcome were based on initial baseline levels of total homocysteine. Graded associations of baseline homocysteine levels and probability of recurrent stroke were significant for the low-dose group (high initial levels vs. medium/low initial levels) and were seen but not significant for the high-dose group (Figure 1). Probability of stroke over time in the Vitamin Intervention for Stroke Prevention (VISP) study, by treatment group and total homocysteine level at baseline. Reprinted with permission from the American Medical Association from JAMA. 2004;291:565–575.1©2004, American Medical Association. All rights reserved. The homocysteine level reductions were considered modest in the high-dose group, possibly because of folate fortification in the US grain supply at the time of the study initiation, as pointed out by the authors. This could have accounted for the lowering of homocysteine levels in the general stroke population used as the standard for entry. The authors also point out that, based on previous analyses,3 a 2 μmol/L difference in homocysteine level should reduce stroke/coronary event risk by 13%. However, the differences in homocysteine levels reached in the two groups were too low to detect a significant effect by power calculations (31% power) vs. the planned 30% reduction in recurrent stroke expected (80% power). The study demonstrates the confounding effect on dietary studies of global changes in nutritional supplementation in the general population's diet. It also demonstrates that even with attempts at optimal risk-factor management, standard target guidelines for lipids, blood pressure, and body mass may not be reached in the period of 1 year, thus modulating the general risk reduction that might be achieved. The demonstration of baseline homocysteine levels with vascular outcomes as in previous studies does suggest the potential efficacy of lowering homocysteine levels in stroke patients, possibly with the use of higher doses of folate (5 mg) to achieve a greater homocysteine reduction. Dr. Liebson was a member of the VISP coronary end points committee but had no role in the design or conduct of the study itself. Summary. INVEST4 enrolled 22,576 patients with hypertension and coronary artery disease (CAD) to compare a calcium-antagonist strategy (CAS) (sustained-release verapamil) with a β-blocker strategy (atenolol). Addition of the angiotensin-converting enzyme (ACE) inhibitor trandolapril and/or hydrochlorothiazide was allowed if target blood pressure was not reached. The study was randomized and single blinded with open-label methodology. After 24 months, the primary outcome of all-cause mortality, nonfatal myocardial infarction, or nonfatal stroke did not differ (9.9% in the CAS group vs. 10.2% in the non-CAS group [relative risk, 0.98; 95% confidence interval, 0.90–1.06]) (Figure 2). Two-year blood pressure control was similar in both groups. In the CAS group, more than 81% were still taking verapamil, more than 60% were also taking the ACE inhibitor, and more than 40% were taking the diuretic. In the non-CAS group, 77% remained on the β blocker, 60% were taking a diuretic, and more than 50% were taking an ACE inhibitor. Primary outcome of treatment strategy in the International Verapamil-Trandolapril Study (INVEST). Reprinted with permission from the American Medical Association from JAMA. 2003;290:2805–2816.4©2004 American Medical Association. All rights reserved. Comment. This study of the use of antihypertensive drug combinations is important because it is the first large, randomized prospective trial of antihypertensive treatment in patients exclusively with CAD and especially because of its comparison of drug combinations in accord with the latest Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) guidelines. The design of the study was complex because of the number of hypotheses tested in regard to the primary outcome. These included efficacy of a CAS vs. a non-CAS agent; the efficacy of addition of an ACE inhibitor in those with diabetes, renal failure, and heart failure; and the need to provide appropriate target blood pressure reduction dependent on JNC 7 guidelines, which recommend different target end points for those with diabetes and renal disease than other participants (<130/85 mm Hg vs. <140/90 mm Hg, respectively). At the conclusion of the study, most participants in the CAS group were on at least two agents, and most of those in the non-CAS group were on three agents. Nonstudy antihypertensive agents were observed in 43% of participants in both strategies. Two-year blood pressure control was similar in both groups, unlike the results of Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT),5 although approximately 30% of patients in both groups did not achieve blood pressures of <140/90 mm Hg, despite the availability of supplementary antihypertensive agents. Secondary outcomes were no different among groups (death, nonfatal myocardial infarction, nonfatal stroke, cardiovascular-related death, cardiovascular-related hospitalization). In a broad range of factors at baseline that might influence primary outcome in group comparisons, there were no significant differences, although there was borderline significance in the non-CAS vs. CAS comparison in those with congestive heart failure at baseline (relative risk for CAS, 1.21; 95% confidence interval, 0.99–1.47). The study conclusion stated that the CAS (verapamil—trandopril) strategy was as effective as the non-CAS (atenolol—hydrochlorothiazide) strategy. However, 44% of the CAS group was also receiving the diuretic, and 52% of the non-CAS group was receiving the ACE inhibitor. Thus, the study—although a comparison of a calcium channel blocker (CCB) vs. a β blocker—included a robust supplementation of an ACE inhibitor and diuretic in both groups. The use of aspirin (57%–58%) and lipid-lowering agents (41%) was the same in both groups. What is the background of this study? Previous outcome analyses from hypertension clinical trials and meta-analyses have demonstrated that diuretics compare equally and perhaps favorably with other drug classes. In CAD, ACE inhibitors appear to be superior to CCBs. The rate of cardiovascular events was 4% per year, roughly 70% of what would have been expected if antihypertensive drugs had not been used.6 What would have been the relative efficacy of antihypertensive agents if other risk-factor modifiers had been optimized (aspirin and lipid-lowering agents)? Since all participants had documented CAD, a case could have been made for initiating ACE inhibitors as a first-line drug in both groups. Then the point could be made that β blockers should also be used in both groups. In that case, CCBs and diuretics would be third-line agents. On the other hand, a diuretic would be a logical second-choice drug to act on blood volume if the initial agent was an ACE inhibitor. The Losartan Intervention for End point reduction in hypertension (LIFE) study of high-risk hypertensive patients agents demonstrated the efficacy of an angiotensin-receptor blocker with a supplementary diuretic compared with a β blocker and a supplementary diuretic.7 INVEST did demonstrate that a nondihydropyridine long-acting CCB was not harmful in CAD per se, contradictory to some previous clinical trial and meta-analysis results. Perhaps the most important lessons to be learned from this study, in regard to extrapolation to further clinical trial designs, are that comparisons of antihypertensive drug combinations do not appear to differ in primary outcome events in CAD patients; the efficacy of antihypertensive intervention must relate to the risk for outcome events, but other risk-factor interventions (aspirin, lipid lowering) must be taken into account; the initial use of an antihypertensive agent should conform to both previous evidence-based efficacy (e.g., ACE inhibitors in CAD) and pathophysiologic variants (increased renin-angiotensin system activity vs. blood volume); and, finally, the possibility of different responses to antihypertensive agents based on demographics must also be considered.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.