Long-term ALLHAT Data Affirm Parity of Antihypertensive Meds

Passive follow-up for as much as 23 years after randomization in ALLHAT, a trial comparing various antihypertensive agents, reveals results similar to the study’s primary findings, according to a paper published online Monday in JAMA Network Open.

Researchers led by Jose-Miguel Yamal, PhD (UTHealth School of Public Health, Houston, TX), report that rates of cardiovascular mortality didn’t significantly differ among patients randomized to initial treatment with a thiazide-type diuretic, a calcium channel blocker, or an ACE inhibitor, with no differences for most secondary outcomes either.

Although use of an ACE inhibitor was associated with greater risks of stroke mortality and fatal/nonfatal hospitalized stroke compared with use of a diuretic, those relationships become nonsignificant after accounting for multiple comparisons.

Senior author Barry Davis, MD, PhD (UTHealth School of Public Health), who has helped lead ALLHAT since its inception, told TCTMD that because there was such a large investment of resources to conduct the trial, the investigators believed it would be worth spending more to perform extended, passive follow-up. He said these new findings are consistent with what was observed in the primary trial results published back in 2002, with a mean follow-up of 4.9 years, and in a prior analysis that extended follow-up out an additional 4 years.

He acknowledged that after the end of the initial trial period, patients were no longer under a controlled experiment and would have gone in different directions in terms of treatment and monitoring, which will influence the findings of an extended follow-up study like this.

But with so much work done in studying various classes of antihypertensive agents over the years, there is an overarching lesson, Davis said. “The most important thing is to control the blood pressure,” he said. “The medication that you use may be fine-tuned . . . but if the blood pressure is controlled, that’s what is most important.”

Prespecified Analysis of ALLHAT

The ALLHAT trial included patients 55 and older who were diagnosed with hypertension and had a documented history of CVD, atherosclerosis, and/or at least one other coronary heart disease risk factor. Investigators initially randomized patients to four initial treatment groups: thiazide-type diuretic (chlorthalidone), calcium channel blocker (amlodipine), ACE inhibitor (lisinopril), or alpha-blocker (doxazosin). The alpha-blocker arm was discontinued early on, however, due to a greater risk of CV events and was therefore excluded from extended follow-up.

The current prespecified analysis, with a mean follow-up of 13.7 years (maximum of 23.9 years), included 15,002 patients started on chlorthalidone, 8,898 on amlodipine, and 8,904 on lisinopril. Passive follow-up was performed by examining administrative data from sources that included the National Death Index, the Social Security Administration, and the Center for Medicare & Medicaid Services. Information on all-cause mortality was available for all patients (mean age 66.9 years; 46.9% women), with morbidity data available for a subset of 22,754 patients (mean age 68.7 years; 56.1% women).

The primary endpoint of this analysis was CV mortality, with rates per 100 participants of 23.7, 21.6, and 23.8 for the diuretic, calcium channel blocker, and ACE inhibitor groups, respectively. Compared with chlorthalidone, risks were no different for amlodipine (adjusted HR 0.97; 95% CI 0.89-1.05) or lisinopril (adjusted HR 1.06; 95% CI 0.97-1.15).

If the blood pressure is controlled, that’s what is most important. Barry Davis

Long-term risks of various secondary outcomes also didn’t differ across treatment groups, although lisinopril was associated with greater risks of stroke mortality (adjusted HR 1.19; 95% CI 1.03-1.37) and fatal/nonfatal hospitalized stroke (adjusted HR 1.11; 95% CI 1.03-1.20) compared with chlorthalidone. These differences were no longer significant after adjustment for the multiple comparisons performed.

“We, therefore, interpreted the magnitude of the HRs for consistency with in-trial analysis and previous posttrial analysis with particular caution,” the investigators write in their paper. “However, with 11 years of additional passive follow-up (2006-2017), the results for lisinopril versus chlorthalidone for stroke and stroke mortality are almost the same.”

Tailoring Therapy in the Modern Era

Commenting for TCTMD, Daniel Duprez, MD (University of Minnesota, Minneapolis), a member of the hypertension work group within the American College of Cardiology’s prevention of cardiovascular disease section, said that ALLHAT has been a landmark study in the realm antihypertensive medications.

But much has changed in the landscape of hypertension treatment since the trial started back in 1994, Duprez said, pointing to greater attention to target-organ damage associated with high BP and lower BP goals in current guidelines. Moreover, he said, ALLHAT included patients who were 55 and older, whereas now hypertension treatment would be initiated at younger ages.

Duprez was cautious, too, about interpreting the results of this follow-up study given the fact that much would have changed with the management of patients’ hypertension after the end of the initial trial period.

Perhaps most importantly, it is much more common in the current era to start patients on combination therapy—often in a single pill to reduce medication burden—rather than starting with one and then adding another if necessary, he indicated. “I would say now the tendency is to tailor your antihypertensive medications in function of the individual patient.”

The key message for clinicians is that the “absolute blood pressure decrease is very important,” Duprez said, also stressing the need to address other risk factors for coronary heart disease, like albuminuria, at the same time.

Davis reiterated how critical it is to lower BP using whichever medications are effective for a given patient. And although diuretics have “gotten a bad rap” due to an associated increase in blood sugar in some patients, clinicians shouldn’t dismiss them as an option, he said. “They are a very powerful medication and an excellent medication in the armamentarium of trying to reduce blood pressure, and they should always be considered.”