Quick Reference for Health Care Providers
The Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) is a long-term, multicenter trial conducted on a large group of participants ages 55 and older with Stage 1 or Stage 2 hypertension and at least one other cardiovascular disease (CVD) risk factor. Supported by the National Heart, Lung, and Blood Institute, ALLHAT includes the largest clinical study to date to compare the effectiveness of medications to treat hypertension. ALLHAT also included a cholesterol-lowering study in a subset of patients participating in the hypertension study. This study evaluated statin drug treatment versus "usual care." Both studies measured mortality as well as cardiovascular events, such as myocardial infarction (fatal and nonfatal), as primary or secondary outcomes.
The ALLHAT hypertension study results indicate that less costly, traditional diuretics are more effective than newer medicines at lowering high blood pressure and preventing some forms of heart disease. In the ALLHAT cholesterol study, both the statin and usual care groups had a substantial decrease in cholesterol levels. The 10 percent difference in cholesterol levels between the groups was insufficient to produce significant reductions in deaths and produced only a small, non-significant decrease in the rates of heart attacks and strokes in the statin group relative to usual care.
With more than 42,000 original enrollees, ALLHAT provides an opportunity to examine the effects of CVD treatment on patients that had been underrepresented in or excluded from previous clinical trials. Nearly one-half of participants in both studies were women; more than one-third were African American; and more than one-third had diagnosed diabetes. In addition, significant proportions of participants were over age 60 and were Hispanic.
ALLHAT was conducted primarily in community-based settings covering a vast geographic area; most of the 623 clinical sites were in private practices in the continental United States, Puerto Rico, the U.S. Virgin Islands, and eastern Canada. Other clinic settings included dozens of Department of Veterans Affairs hospitals, academic centers, military clinics, and community health centers.
The trial began in February 1994 and concluded in March 2002; patient follow-up averaged 4.9 years. The final results of both the ALLHAT hypertension and the lipid studies are published in the December 18, 2002, issue of The Journal of the American Medical Association (JAMA).
Hypertension Trial Results
The ALLHAT hypertension study was a randomized, double-blind trial that sought to compare traditional antihypertensive medicines with newer classes of antihypertensives. The study enrolled 42,418 individuals; average blood pressure at enrollment was 146/84 mm Hg or 156/89 mm Hg if untreated.
When ALLHAT was initiated, use of diuretics and beta-blockers was steadily declining as prescriptions for three newer classes of antihypertensive drugs--calcium channel blocker, angiotensin converting enzyme (ACE) inhibitor, and alpha-adrenergic blocker--grew, in spite of expert recommendations that the traditional medications be preferred because of their proven effectiveness and lower cost. Research on the newer medications largely compared the drugs' effectiveness to placebo; ALLHAT sought to identify which medication worked best.
For the study's traditional (standard control) arm, a diuretic (chlorthalidone) was selected over a beta-blocker because diuretics had been studied more extensively. In addition, although evidence from several randomized trials found comparable CVD outcomes when initial treatment with diuretic was compared to beta-blocker, diuretics were considered clinically advantageous for the types of patients to be selected for ALLHAT--older patients and those at higher risk of CVD complications.
ALLHAT demonstrated that chlorthalidone is superior at preventing CVD events compared to each of the treatment drugs studied: calcium channel blocker (amlodipine), ACE inhibitor (lisinopril), and alpha-adrenergic blocker (doxazosin). Each of the newer drugs had significantly higher rates of one or more forms of cardiovascular disease, and lisinopril and doxazosin had higher rates of combined CVD. None of the three test treatments differed significantly from chlorthalidone in rates of major coronary heart disease (CHD) events and all-cause mortality.
The doxazosin arm was stopped early (in March 2000) due to a 25 percent higher rate of combined CVD and a two-fold higher rate of heart failure compared to the diuretic arm. The results were published in the April 19, 2000, JAMA.
The remaining 33,357 patients stayed on study drugs through the end of the study, for an average of 4.9 years. The differences observed between those on diuretics and other study drugs were not conclusive at the time the doxazosin arm was stopped. These differences became apparent by the study's end.
All three classes of drugs reported on in the December 18 issue of JAMA--diuretics, calcium channel blockers, and ACE inhibitors--have been previously shown to lower blood pressure and reduce cardiovascular complications. In head-to-head comparisons, the diuretics were shown to be superior in treating high blood pressure and preventing cardiovascular events.
Key findings from the ALLHAT hypertension study include:
- The chlorthalidone group had slightly better blood pressure control; systolic blood pressure was 1 mm lower than the amlodipine group and 2 mm lower than the lisinopril group.
- Participants taking amlodipine had a 38 percent increased risk of heart failure compared to those taking chlorthalidone. No significant end-point differences for stroke or combined CVD were found. Similarly, there were no significant differences for cancer incidence or mortality, or for hospitalization for gastrointestinal bleeding (assessed in a subset). Results were consistent across subgroups by age, gender, race, and diabetic status.
- The lisinopril group had significantly higher risk of stroke (15 percent), heart failure (19%), and angina (11%) compared to the chlorthalidone group. Increased rate of combined CVD was 10 percent. Results were consistent across subgroups by gender and diabetic status.
- The effectiveness of chlorthalidone compared to lisinopril was more remarkable in blacks than in non-blacks. Rates of stroke were 40% higher in the lisinopril group for blacks, with no difference in non-blacks; rates of combined CVD were 19 percent higher in blacks compared to 6 percent increased rate in non-blacks. In the lisinopril compared to the chlorthalidone group, the difference in systolic blood pressure was twice as great (4 mm Hg) among blacks compared to non-blacks.
- The antihypertensive medications were well tolerated. Among patients followed for five years, 80 percent of the chlorthalidone and of the amlodipine groups, and 73 percent of the lisinopril group, were still taking their assigned drug or another drug in the same class.
- ALLHAT confirmed results from previous studies regarding biochemical effects of chlorthalidone compared to amlodipine and lisinopril. The chlorthalidone group had more hypokalemia (lower potassium levels), slightly higher mean serum cholesterol and glucose levels, and slightly higher incidence of new cases of diabetes at four years follow-up. However, ALLHAT also demonstrated that these effects do not result in more adverse clinical outcomes when patients' serum potassium is monitored and they are provided potassium supplements when necessary. In addition, creatinine-based measures of renal function were more favorable in the amlodipine group overall and, in some cases, the lisinopril group; however, rates of end-stage renal disease were not significantly different compared to the chlorthalidone group.
The ALLHAT findings are consistent in large part with evidence from other clinical trials and should be widely applied in patient care. In summary, the results indicate that:
- Because of their superiority in preventing one or more major forms of CVD and their lower cost, thiazide-type diuretics should be the drugs of choice for initial treatment of hypertension in most patients requiring drug therapy.
- In patients who cannot tolerate a diuretic, therapy can be started with ACE inhibitors, calcium channel blockers, or beta-blockers. These medications have been shown to have CVD benefits compared with placebo. Alpha-blockers, however, should not be considered for initial therapy.
- Most hypertensive patients require more than one medication to adequately control blood pressure, and diuretics should be part of most multi-drug regimens. Behavioral approaches (eg, eating plan, physical activity, weight loss) should also be taken.
- Physicians and patients should consider changing antihypertensive therapy to a thiazide-type diuretic if a different type of medication is currently taken--even if the patient's blood pressure is well controlled (below 140/90 mm Hg) with the alternate medication. The patient is likely to benefit from the enhanced effectiveness of the diuretic as well as the reduced cost of the drug.
Lipid Study Results
The ALLHAT lipid lowering treatment study was conducted in a subset of ALLHAT participants; 10,355 individuals with mildly elevated LDL cholesterol levels were randomly assigned to receive "usual care" or drug therapy using a statin (pravastatin, 40 mg/day). Although the results of this nonblinded study are inconclusive, they are nevertheless consistent with earlier trials and current guidelines to reduce blood cholesterol levels and decrease the risk of cardiovascular events through lifestyle changes and drug treatment.
All of the participants in the lipid study received standard dietary and lifestyle advice to control their cholesterol levels. The use of cholesterol-lowering medication in these participants was not considered essential by then current guidelines. However, participants in the usual care group were prescribed cholesterol-lowering drugs (not provided by the study) when their physicians felt that pharmacological therapy was warranted by changes in the patients' clinical status such as a myocardial infarction or a marked increase in cholesterol level. Thirty percent of the usual care group received a cholesterol-lowering drug during the course of the study.
Cholesterol levels fell substantially in the usual care as well as the pravastatin treatment groups. Although the statin group had greater reductions in total cholesterol levels than the usual care group (17 percent reduction in the statin group versus 8 percent reduction in usual care), the difference between the two groups was relatively modest--largely due to increased use of statins in the usual care group.
In accord with the small difference in cholesterol reductions, ALLHAT found that compared to the usual care group, the pravastatin group had only a small nonsignificant decrease (9 percent) in rates of heart attacks and strokes, and no reductions in deaths.
The growing use of statin therapy in the usual care group reflected general medical practice trends over the nearly 8-year course of the trial. During this time, evidence from five large statin trials demonstrating the effectiveness of cholesterol-lowering medications and continued development of new potent statin drugs brought about significant changes in the indications for and utilization of statins in the U.S. Similarly, the clinical guidelines for treating high blood cholesterol in adults released by the National Cholesterol Education Program (NCEP) in May 2001 lowered the threshold for patients for whom cholesterol-lowering medications were recommended. In fact, if these guidelines had been in effect at the start of the ALLHAT lipid study, nearly all participants would have met the criteria for drug therapy at enrollment.
The final results of the ALLHAT lipid study, therefore, demonstrate that:
- Both the pravastatin and the usual care groups attained substantial cholesterol reductions, resulting in a relatively modest cholesterol difference between them.
- Accordingly, the ALLHAT lipid-lowering treatment study found only a small (nonsignificant) decrease in CVD event rates for pravastatin compared to usual care, and no difference in mortality.
- The results do not alter current cholesterol treatment guidelines, which are based on a series of clinical trials with larger cholesterol reductions than that observed in ALLHAT.
The results of the ALLHAT lipid study are appropriately viewed in the context of prior cholesterol-lowering trials. The ALLHAT lipid study results for both heart attacks and mortality are within the statistical margin of error of the average results reported by earlier cholesterol-lowering trials attaining comparable differences in cholesterol levels between treatment and control groups.
Despite finding results consistent with those of other statin trials when viewed in context of their treatment differentials for cholesterol, the ALLHAT researchers have explored other possible explanations for the absence of a significant effect of pravastatin on clinical outcomes. First, they considered the differences between the population studied in ALLHAT and other large statin trials. ALLHAT participants as a group were older, included more women, and had greater racial and ethnic diversity. Subgroup analyses of the ALLHAT lipid study based on age, gender, and race failed to explain the absence of a bigger treatment difference. In particular, these analyses provided no basis for concluding that cholesterol-lowering treatment is less effective for women, the elderly, or African-Americans.
Furthermore, the PROSPER trial, a European study published in November 2002, specifically evaluated pravastatin for the prevention of myocardial infarction and stroke in patients ages 70 years to 82 years. PROSPER researchers found that pravastatin significantly reduced the incidence of fatal and nonfatal myocardial infarction (but not stroke). These results argue against attributing the absence of a significant treatment benefit in the ALLHAT lipid study to the inclusion of elderly patients.
Second, theALLHAT investigators considered the possibilty that the lack of a significant effect of pravastatin on clinical outcomes in ALLHAT was due to the fact that the trial was performed in hypertensive patients. A prior analysis of three earlier pravastatin trials suggested that hypertension might attenuate the beneficial impact of statin treatment on clinical outcomes. However, the recent results of the British Heart Protection Study (a trial of simvastatin in which heart attacks were reduced by 24% in patients with and without hypertension), and preliminary reports of highly positive results of the ASCOT trial (a trial of atorvastatin conducted in hypertensive patients), argue strongly against this hypothesis.
In summary, statin drugs have been shown by numerous trials to be safe and effective for the treatment of high blood cholesterol levels and to reduce the rate of heart attacks and other complications of cardiovascular disease. Although the differences in clinical outcomes in the ALLHAT lipid study observed in participants treated with pravastatin versus those receiving usual care are not statistically significant, they are consistent with the results of other large cholesterol-lowering trials when considered in context of the modest cholesterol difference between treatment groups. In the ALLHAT lipid study, statin treatment was effective in reducing cholesterol levels, including among participants in the usual care group who took statins. Furthermore, there were no reports of rhabdomyolysis or other life-threatening adverse effects in the ALLHAT lipid study.
The ALLHAT lipid study results do not alter current guidelines for cholesterol-lowering therapy in the prevention and treatment of cardiovascular disease. Cholesterol lowering by lifestyle changes and drug therapy is recommended to reduce CVD morbidity and mortality.
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES * National Institutes of Health * National Heart, Lung, and Blood Institute