Atlanta, GA - Results of a new randomized feasibility trial suggest that treating people with blood pressure (BP) in the prehypertensive range reduces incident hypertension [1].

The Trial of Preventing Hypertension (TROPHY) treated prehypertensive patients with the angiotensin receptor blocker (ARB) candesartan (Atacand, AstraZeneca) for two years, then with placebo for two years, and compared this group with one treated for the whole time with placebo.

Dr Stevo Julius with coauthor Dr Shawna D Nesbitt

"Over a period of four years, nearly two thirds of the placebo group developed stage 1 treatment-requiring hypertension, and that may be a very important message," said lead investigator Dr Stevo Julius (University of Michigan, Ann Arbor) of the findings. "Two years of treatment with candesartan in patients with prehypertension delayed the onset of stage 1 hypertension beyond two years (after the discontinuation of treatment), substantially suppressed the onset of stage 1 hypertension during the two years of treatment, and substantially prolonged the hypertension-free period. In addition, the treatment was well-tolerated."

Although the data are encouraging, the researchers are not recommending intermittent drug treatment for the estimated 25 million prehypertensive patients in the US on the basis of this one trial, given that the safety and cost effectiveness of this approach has not been proven. However, because of the high rate of transition from prehypertension to stage 1 hypertension seen in the trial, they are recommending more frequent follow-up of patients whose BP is in the range they studied. "We think the three-month interval (used in the trial) is a reasonable one," he told attendees here.

The study was published online March 14, 2006 in the New England Journal of Medicine, to coincide with its presentation at the American College of Cardiology 2006 Scientific Sessions.

Treating "high-normal" BPs

The term prehypertension has been a contentious one since it was introduced in the JNC 7 hypertension guidelines in 2003 [2]. Some felt the change in terminology from high-normal to prehypertension would give a "diagnosis" to essentially healthy people; the authors of the guidelines, however, meant for it to indicate that action should be taken before people became frankly hypertensive. They noted that lifestyle, not pharmacologic, intervention would be recommended for those who fell into the prehypertensive range.

"These approaches do work, but in a very select group of people, and they have not yet had an impact in any way on the frequency of [high] blood pressure or obesity," Julius said in an interview with heartwire. "We thought the time had come to look at whether pharmacological intervention in these people was feasible and whether we could postpone hypertension."

The trial was investigator-initiated and supported by AstraZeneca. Included were prehypertensive patients with repeated BP measures of 130 to 139 mm Hg systolic and 89 mm Hg diastolic or lower, or systolic pressure of 139 mm Hg or lower and diastolic pressure of 85 to 89 mm Hg—that is, those in the higher half of the prehypertensive range that begins at 120/80 mm Hg —the range that used to be called "high-normal." Julius pointed out their trial began before the change in the JNC 7 terminology,

A total of 809 patients from 71 centers were randomized to receive either 16 mg/day of candesartan or placebo for two years; all patients were then switched to placebo for an additional two years. The primary end point of the trial was the development of stage 1 hypertension, defined one of four ways:

• A BP of >140/>90 mm Hg at any three visits.
• A BP of >160/>100 mm Hg at any one visit.
• A BP of >140/>90 mm Hg at the final study visit.
• A decision by the study investigator, regardless of BP, that a given patient required treatment, particularly for target-organ damage.

Once a diagnosis of hypertension was made, patients were offered treatment free of charge with metoprolol and a diuretic; physicians could also prescribe other drugs at their discretion, with the exception of an ARB.

At the end of the study, data on 772 patients—391 in the candesartan group and 381 in the placebo group—were available for analysis. The average age of participants was younger than that of other hypertension trials, Julius noted, "but I wish they had been younger." BP was equal in both groups at the outset.

After two years, there was a highly significant difference in the development of hypertension between the two groups, Julius said, and the difference remained significant out to four years, even after treatment was stopped in the candesartan group.

TROPHY: Risk of hypertension after two years with candesartan vs placebo, followed by two years of placebo vs placebo

Time point	Candesartan, n	Placebo, n	Relative risk (95% CI)	p
At 2 years	53	154	66.3	<0.001
At 4 years	208	240	15.6	<0.007

"What is important to note are the safety data," he said, "and what you can clearly see is that serious adverse events in these two groups were the same, and low," occurring in 3.5% of candesartan patients and 5.6% of those on placebo—not a significant difference. "So there was no negative effect of treatment."

The effect was seen across all the subgroups they examined.

Their findings generate a variety of other questions, Julius concluded, including whether results would be better in younger subjects with prehypertension, whether a longer period of treatment or a large degree of BP-lowering would give better results, whether the results would be the same with other antihypertensive agents, and whether early treatment of prehypertension would ultimately decrease target-organ damage and improve cardiovascular outcomes.

"All of these are interesting questions, and important in a large number of patients; we see no research in that area in that group of patients," he said.

During the question period, though, Dr Jim Stein (University of Wisconsin, Madison) pointed out that although fewer patients transitioned to hypertension during the treatment phase (53 on treatment vs 154 on placebo), more in the group initially treated with candesartan developed hypertension than those on placebo during the second two-year period when they were switched to placebo (155 in the candesartan group vs 86 in the placebo group).

"What it really looks like to me is once people stop therapy, the blood pressure goes up inexorably, so I'm not quite sure you've really proved that we can delay hypertension; I think we've shown is that if we treat it, we lower the numbers," Stein said.

Julius responded, "We are not claiming that this is anything else but statistically significant, encouraging for us, and hope that it will be demonstrated again."

Changing the threshold?

Dr William Elliot (Source:Rush-Presbyterian Medical Center)

Dr William Elliot (Rush-Presbyterian Medical Center, Chicago, IL), also a TROPHY investigator, was the invited discussant for the presentation, although he said he'd promised to say "only nasty things."

He noted that he had the same concerns as Stein had expressed about the parallel curves and said this should be further analyzed.

"The big concern that I think most people have about TROPHY is that it may tend—although perhaps not intentionally—to give the idea that people should change medical practice because we are changing the threshold for blood-pressure treatment from 140/90 mm Hg, the traditional one, to one that includes people in the middle of the prehypertensive range," Elliott said.

The study was designed before the term prehypertension was coined, when it was "high-normal" BP, and "we abandoned the term because of the fact that, for most patients, when the doctor told them they had high-normal blood pressure, what the patients heard was, 'Hi! You have normal blood pressure,' " Elliott said wryly. "That was definitely the wrong take-home message."

TROPHY provides data that "suggest but do not prove that the institution of antihypertensive drug therapy has some benefit—at least on turning people into hypertensives who would otherwise be treated with the drug that we are supposed to be using, which of course is low-dose chlorthalidone."

The other concern is the potential cost, Elliott said. "I've been told the estimate of prehypertensive people in the US is 31% of the population, and that's about the same percentage who voted in the last general election. I quake in my boots to think that the American economy is strong enough to pay for that [medication]," he said.

Lifestyle intervention still the foundation

In an editorial accompanying publication of TROPHY [3], Dr Heribert Schunkert (University of Luebeck, Germany) writes that the study findings imply that the medical treatment of prehypertension does not just mask the subsequent development of overt hypertension but rather that important effects can be observed even after a long period of placebo "washout."

"Thus, these investigators successfully tested the hypothesis that prehypertensive levels of blood pressure and the intimately involved renin-angiotensin system, together or separately, are key players in a vicious cycle that ultimately leads to new-onset hypertension," Schunkert writes.

However, he points out that the findings should be interpreted with some caution: for example, because hypertension was diagnosed only by three readings together averaging above 140/90 mm Hg, those on candesartan would have lower overall averages because of the treatment period, thus perhaps "artificially" delaying the end point; in addition, the number of patients started on antihypertensive medications in year three and four were the same as in years one and two when one group was receiving candesartan. "All these points, taken together, suggest that, because of the study design, the difference between the two groups with respect to true prevention of hypertension may have been overestimated."

While drug therapy for prehypertension may be more convenient than lifestyle changes, he concludes, "too many questions remain open. Foremost we need to understand the price in terms of patient safety, and possibly, prohibitive financial costs of what would approach populationwide drug treatment."

Dr George Bakris (Rush University, Chicago, IL), a member of the American Heart Association Kidney Council, was extremely positive about the data, calling it a "dramatic result" and saying that Julius and colleagues should be congratulated for getting this trial done.

It has been established that cardiovascular risk begins at 115 mm Hg systolic, so that even though they are not technically hypertensive, people in the range of the BPs studied in this trial are still at increased risk, he told heartwire. If, by treating them, it's possible to prevent further acceleration of this process, it would, in the long run, make pharmacoeconomic sense by preventing hospitalizations associated with target-organ damage, he pointed out. "The bad news is that this trial was not powered to make broad generalizations," he said.

However, if the other features of metabolic syndrome, including obesity and lipid abnormalities, are added in, the risk is even higher, Bakris said. "In fact, I'm on record as saying that those with metabolic syndrome should be treated," he said.