Make New Friends, but Keep the Old

Abstract & Commentary

Synopsis: Thiazide diuretics should be the first-step antihypertensives because they are less expensive and more effective than calcium channel blockers or ace inhibitors.

Source: ALLHAT Study. JAMA. 2002;288:2981-2997.

This report is part of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) comparing the cardiovascular effects of various first-step drugs for the treatment of hypertension. Participants were 33,357 individuals whose mean age was 67 years (minimum age, 55). Also, 47% were women, 35% were black, 19% were Hispanic, and 36% were diabetic. They were recruited from 623 centers in 4 countries. Successfully recruited subjects were divided into 3 groups that were well matched for important variables, including cigarette smoking, body mass index, aspirin use, education, blood pressure, diabetes, and most other known cardiovascular risks. To be included, a subject had to have stage 1 or 2 hypertension and at least 1 additional risk factor for coronary heart disease events, including myocardial infarction (MI) or stroke (CVA) within the previous 6 months, left ventricular hypertrophy (LVH), type 2 diabetes, current cigarette smoking, high-density lipoprotein (HDL) cholesterol of less than 35 mg/dL, or documentation of other atherosclerotic cardiovascular disease (CVD). Those with heart failure were excluded. Participants were randomly assigned to 1 of 3 arms: chlorthalidone (representing the thiazides), amlodipine (representing the calcium channel blockers), or lisinopril (representing the ACE inhibitors). They discontinued whatever antihypertensive they were on when the study drug was started.

Patients were followed at 1, 3, 6, 9 and 12 months and every 4 months thereafter. Length of follow-up ranged from 3.6 years to 8.1 years (average, 4.9). At each visit, blood pressure (BP) was measured, and drugs were titrated to achieve a blood pressure of less than 140/90. Clinicians in the trial could add other antihypertensives at their discretion. Nonpharmacologic approaches were also recommended to patients.

The primary outcomes were fatal coronary heart disease or nonfatal MI. Secondary outcomes were all-cause mortality, stroke, combined coronary heart disease (including angina, revascularization, MI), and combined cardiovascular disease (including all coronary heart disease, stroke and heart failure).

In comparing outcomes between treatment groups, the following difference were noted:

  1. Amlodipine vs chlorthalidone: The Amlodipine group had a 38% higher risk of heart failure;
  2. Lisinopril vs chlorthalidone: The Lisinopril group had a 15% higher risk of stroke, a 19% higher risk of combined cardiovascular disease (including a 19% higher risk of heart failure), and a mean systolic BP at follow-up that was 2 mm Hg higher.

Angioedema occurred 4 times more commonly with lisinopril than with chlorthalidone, but cholesterol levels were higher and hypokalemia and new diabetes were more common with chlorthalidone than with either of the other drugs. All-cause mortality was the same between the groups.

Comment by Barbara A. Phillips, MD, MSPH

For those of us not classified by the pharmaceutical industry as "early adopters," these findings are vindicating. This report, which comes from the largest study of hypertension ever undertaken, clearly documents that old-fashioned diuretics are more effective and less costly than newer agents in lowering blood pressure and preventing cardiovascular complications. In their discussion, it is noted that the health care system would have saved $3.1 billion in estimated costs of antihypertensive drugs had the pattern of prescriptions for the treatment of hypertension remained at the 1982 level (when diuretic use comprised 56% of the antihypertensive market, compared with the current 27%). A companion article in the same issue of JAMA also notes that pravastatin did not reduce all-cause mortality or coronary heart disease significantly when compared with usual care in older patients.1

These and other findings have attracted the attention of the insurance industry and the media. My home-town newspaper2 carried an editorial, reprinted from the New York Times, which said, in part: "The study . . . carries a compelling warning about the deficiencies of drug testing. The newer drugs had never been forced to prove they were better than diuretics, and their manufacturers had no incentive to risk conducting such tests . . . The government should either force manufacturers to compare their new drugs with existing remedies when seeking marketing approval or conduct such studies itself."

Two messages got through to me: Most all antihypertensive regimens should include a diuretic, and newer isn’t necessarily better (though it’s almost always more expensive).


1. The ALLHAT Officers and Coordinators for the ALLHAT Collaborators Research Group. Major outcomes in moderately hypercholesterolemia, hypertensive patients randomized to pravastatin vs usual care. The antihypertensive and lipid-lowering treatment to prevent heart attack trial. (ALLHAT). JAMA. 2002;288: 2998-3007.

2. Test new drugs against the old. Lexington Herald Leader. December 20, 2002.

Dr. Phillips is Professor of Medicine at the University of Kentucky and Director of the Sleep Disorders Center, Samaritan Hospital in Lexington, KY.