Hypertension Highlights: Does It All Begin and End With ALLHAT?

Linda Brookes Good, MSc


March 18, 2010

In This Article

This Hypertension Highlights newsletter kicked off 7 years ago in December 2002 with the first results of what is still the largest hypertension trial ever conducted. I'm referring, of course, to the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack (ALLHAT) trial.

For this issue, we report on the 10-year follow-up analysis, which not surprisingly, reached the same conclusion it did 7 years ago: no medication is better for hypertension than thiazide-type diuretics (or at least chlorthalidone) -- although almost 50 years after their introduction, they appear to be losing ever-more appeal.

Even with several classes of antihypertensive drugs available whose therapeutic benefits have been proven over and over in large, well-designed clinical trials, high blood pressure control rates continue to be far from ideal. They hover around 30% in the United States and are much lower in most other countries. Future patients may therefore be best served by antihypertensive vaccines. Several of these vaccines, which are directed against angiotensin I or II, are in development.

Perhaps the best news at the end of 2009 is that drinking 2 glasses of champagne per day may actually reduce cardiovascular disease risk. Shall we raise our champagne glasses and drink to the Eighth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 8), due in 2010?

ALLHAT Investigators Stick to Same Conclusions in the Long Term

Long-term follow-up data from the ALLHAT trial support the main conclusion of the original trial, that thiazide-type diuretics should be preferred as first-step treatment in most older patients with hypertension, according to William Cushman, MD (Veterans Affairs Medical Center, Memphis, Tennessee). Presenting the data at the 2009 scientific sessions of the American Heart Association in November in Orlando, Florida, Dr. Cushman announced that 10-year follow-up confirmed that an older drug, the thiazide-type diuretic chlorthalidone, was as safe and effective as the calcium-channel blocker , amlodipine; an angiotensin-converting enzyme inhibitor, lisinopril; and the alpha-adrenergic blocker, doxazosin.[1] All comparator drugs were relatively new at the time ALLHAT, the largest hypertension trial to date, started in 1994.

The randomized, double-blind, multicenter ALLHAT trial compared the effect of treatment with amlodipine 2.5-10 mg/day, lisinopril 10-40 mg/day, or doxazosin 2-8 mg/day with a chlorthalidone 12.5-25 mg/day on a composite endpoint of fatal coronary heart disease (CHD) or nonfatal myocardial infarction.[2] The study involved 42,418 individuals age ≥ 55 years who had hypertension and ≥ 1 other risk factor for CHD. The doxazosin arm of the trial was stopped prematurely at 3 years when an increase in cardiovascular events was identified in those patients.[3] At 5-year follow-up, no difference existed between treatment with chlorthalidone and the other newer drugs in terms of the primary or other outcomes, including CHD, all-cause mortality, cardiovascular disease mortality, or end-stage renal disease.[4] Compared with chlorthalidone, amlodipine was associated with a 38% higher risk for congestive heart failure and lisinopril was associated with 10% more cardiovascular events, 19% more congestive heart failure, and 15% more stroke. The rate of stroke was 40% higher in blacks who received lisinopril vs chlorthalidone.

Ten-Year Follow-up

The 10-year follow-up analysis, which was carried out in US patients only, was based on in-trial data, as well as on several post-trial events identified from administrative databases, including the National Death Index, Social Security Administration, Centers for Medicare & Medicaid Services, and the United States Renal Data System. No information was available on post-trial blood pressure levels or additional drugs that ALLHAT participants took during the follow-up after the trial ended. The investigators made the assumption that patients were treated similarly, Dr. Cushman said. Mortality data were available on 41,719 patients. As in the main trial, no differences were seen in all-cause mortality and cardiovascular disease mortality (the primary outcome of this analysis) between amlodipine or lisinopril and chlorthalidone in these patients after they were removed from blinded therapy. Combined morbidity and mortality data were available on 27,246 patients. There were also no differences in CHD in these patients.

For patients hospitalized for heart failure or with fatal heart failure, the differences seen at earlier follow-up persisted, Dr. Cushman reported. The difference between lisinopril and chlorthalidone was no longer significant, but the significant difference between amlodipine and chlorthalidone was maintained, with a 12% higher rate of patients in the amlodipine group hospitalized for heart failure (P = .01 vs chlorthalidone). The higher rates of patients in the lisinopril group who were hospitalized for stroke or who had fatal stroke during the trial did not persist over long-term follow-up. Long-term (7-year) follow-up of patients in the doxazosin arm of the trial also showed no differences in cardiovascular disease, all-cause mortality, and number of patients hospitalized for heart failure or with fatal heart failure. There was a 7% higher risk for hospitalization for stroke or fatal stroke with doxazosin vs chlorthalidone, but this difference was not significant.

Analysis of post-trial differences in cardiovascular disease appeared to favor lisinopril vs chlorthalidone. However, because there was no difference in cardiovascular disease or cardiovascular disease mortality over the entire follow-up period, Dr. Cushman stressed that the (small) post-trial difference did not suggest that lisinopril should be preferred over chlorthalidone for initial treatment of hypertension. He believes that the 10-year results of ALLHAT support the recommendations of the JNC 7 report,[5] the current US hypertension guidelines, to initiate therapy with a thiazide-type diuretic, and "if you want to see the benefits that we have seen, you should keep the patients on the drugs."

The AHA-designated discussant, Daniel W. Jones, MD (University of Mississippi, Oxford) noted that after publication of the 5-year follow-up results of ALLHAT, hypertension guidelines still differed in their recommendations for initial therapy. In the United States, the JNC 7 guidelines followed the ALLHAT conclusions and recommended a thiazide-type diuretic as first-line treatment.[5] Dr. Jones said others might have reached an alternative interpretation of the ALLHAT results (ie, that several classes of drugs appear to provide cardiovascular protection). He noted that the European Society of Hypertension/European Society of Cardiology guidelines express no preference among 5 drug classes,[6] whereas in the United Kingdom, thiazides and beta-blockers are discouraged.[7]

"Blood pressure should be lowered with drugs demonstrated to reduce cardiovascular mortality using the least intrusive approach for an individual patient," Dr. Jones said. "Least intrusive might include the potential for adverse effects, cost, or inconvenience. For many clinicians, either approach will lead to use of thiazide-type diuretics as an initial choice for many patients." However, "we can be confident of cardiovascular protection with a number of drugs," he said.

Panel discussant Robert O. Bonow, MD (Northwestern University Feinberg School of Medicine, Chicago, Illinois) cautioned that although the adverse effects of doxazosin seen earlier in ALLHAT did not persist in long-term follow-up, this did not mean that doxazosin could be recommended as initial therapy in hypertension.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.