Stopped Early for Benefit
Henry R. Black, MD: Hi. I'm Dr Henry Black. I'm adjunct professor of medicine at the Langone New York University School of Medicine, and I'm here today with my long-term friend and colleague, Dr Bill Cushman. Bill, thank you very much for doing this.
William C. Cushman, MD: Delighted to be here.
Dr Black: What I want to talk about is the SPRINT study,[1] which you've been a primary participant in. The top-line results were just released. Tell us a little bit about SPRINT: who was in it, what the hypothesis was, and how it compares to the ACCORD study, which you also participated in.
Dr Cushman: Sure. I'm Dr Bill Cushman. I'm from Memphis, Tennessee. And I'm chief of the preventive medicine section at the VA and professor of preventive medicine at the University of Tennessee.
I was a network principal investigator in SPRINT, which meant that I oversaw about a quarter of the sites. SPRINT was a study sponsored by the National Institutes of Health (NIH), primarily the National Heart, Lung, and Blood Institute (NHLBI). But other institutes—the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute on Aging, the National Institute of Neurological Disorders and Stroke—were also involved.
SPRINT was a study of 9361 participants who were randomized to either a lower, more intensive goal of less than 120 mm Hg systolic blood pressure (SBP) compared with a goal of less than 140 mm Hg systolic. That was considered standard when we designed the study, and all guidelines recommended at least getting below 140 mm Hg.
We recruited a participant pool of high-risk hypertensive patients with SBPs of ≥130 mm Hg. They could be on medications (the majority were), but they didn't have to be. Participants not only had to have elevated blood pressure, but they also had to be above age 50 and they had to have some other indices of risk: known cardiovascular disease, chronic kidney disease, or being above age 75, for example, or having a Framingham risk assessment for cardiovascular disease of ≥ 15% over 10 years.
They were randomly allocated to these two groups, with the intent of being followed for about 5 years. The primary outcome in SPRINT was a combined cardiovascular outcome that included myocardial infarction (MI), acute coronary syndrome other than MI, stroke, heart failure, or cardiovascular death.
Now, there are a lot of other outcomes in SPRINT, including whether this lower blood pressure goal would prevent dementia, changes on MRI, or chronic kidney disease. Those outcomes have not been stopped or announced yet, and we're still collecting data on that.
The cardiovascular outcomes were viewed as so positive in terms of the benefit that the Data and Safety Monitoring Board recommended to Gary Gibbons, the director of the NHLBI, that the cardiovascular part of the trial—and the intensive intervention in particular—should be stopped and that the investigators and the participants should be unblinded. And that was done.
Dr Black: Were the antihypertensive regimens prescribed, or was it whatever the docs wanted to do?
Dr Cushman: Good point. We actually recommended using the major classes that were proven to be of benefit in cardiovascular outcome trials in hypertension: either thiazide-type diuretics, ACE inhibitors, angiotensin receptor blockers, or calcium blockers. It was primarily those four classes, and they could be combined in whatever way the investigators wanted. We did put a lot of emphasis on using thiazide-type diuretics because of the ALLHAT[2] results.
But the way they could be combined was really up to the investigators. Now, if the participants had known coronary disease or some other indication for a beta-blocker, that could certainly be used. And then other drugs could be added. We had a very large formulary representative of all the major classes of drugs— not only those classes, but also beta-blockers, alpha blockers, aldosterone inhibitors (spironolactone or amiloride, for example).
We had a lot of drugs available. They were predominantly purchased for the study, by NIH. There were only two drugs that were donated by the pharmaceutical companies. The study was entirely funded by NIH.
SPRINT vs ACCORD
Dr Black: How is this different from ACCORD?[3]
Dr Cushman: In ACCORD, we had the same two SBP goals: less than 120 mm Hg compared with a SBP of less than 140 mm Hg. However, SPRINT is twice as large as ACCORD.
As you may remember, we did not show a significant benefit for the lower SBP goal for the overall cardiovascular outcome in the ACCORD trial. We did see a significant reduction in stroke of about 40%, but that was a secondary outcome. The primary outcomes in mortality were not reduced in ACCORD.
However, ACCORD was about half the size of SPRINT. And even though the ACCORD blood pressure study was done in patients with diabetes, on average, they were probably a little lower-risk than our SPRINT participants because of their somewhat younger age (average age, 62 years), the absence of real chronic kidney disease, and several other reasons.
Even though ACCORD didn't show a statistically significant benefit, it did show a 12% reduction in the cardiovascular outcome with a confidence interval that could have included up to a 27% benefit.
In contrast, SPRINT was twice as large, with a higher-risk population with an older average age. We excluded people with diabetes because that was being looked at in ACCORD. And we excluded people who'd had a prior stroke because that was being looked at in the SPS3[4] post-stroke study in terms of blood pressure goals.
Despite that, we had a very high-risk population. And what we found was about a third of a reduction in the primary cardiovascular events. That was significant.
We also saw, quite importantly, about a 25% reduction in all-cause mortality. That was surprising. The results are quite clear that there's dramatic benefit in terms of both cardiovascular events and total mortality.
Dr Black: You probably can't tell us this yet, but what was the blood pressure achieved in the less-than-140 group compared with the less-than-120 group?
Dr Cushman: We can't say that quite yet, but we can say that the intervention worked well, and our experience was fairly similar to that of ACCORD. In the ACCORD study, we ended up with an average SBP of 134 mm Hg in the standard group and about 119 mm Hg in the intensive group. And that was maintained throughout the study. Basically, we saw the same kinds of results and very nice separation in SBP through the SPRINT trial.
No Magic Standard Drugs Used
Dr Black: I want to emphasize that you got these results using standard drugs—nothing magical, nothing new, but with drugs that are readily available, and for the most part, relatively inexpensive for the practicing physician. You gave them guidance but no required protocol. The investigators were able to use their own judgment as to what to give individual patients.
Dr Cushman: Right. I do want to point out that our formulary did have selected drugs within the various classes. And it's important to point out that we actually used chlorthalidone as the thiazide-type diuretic the majority of the time that we used thiazides.
One reason we did that is because we didn't want there to be a differential use of chlorthalidone and hydrochlorothiazide between the two randomized groups. So whenever a thiazide-type diuretic was used in either group, overwhelmingly it was chlorthalidone.
Dr Black: We've been old fans of chlorthalidone since SHEP[5] days. This is a drug that I think has been used much more now that we've finally got some of those old SHEP results out. But it's still not as popular as using thiazides.
Did you look at resistant hypertensives here? Was there an upper limit of who you would enroll?
Dr Cushman: That's a very good point. We did exclude people who were on a lot of drugs, depending on what their upper-level SBP was. So, for example, if they were on one or no drugs, they could have a SBP as high as 180 mm Hg. But if they had more than one or two drugs on board, then there were progressively more restrictive upper ranges for the SBP.
Technically, we excluded the most resistant patients, because if we brought somebody in on four drugs with a blood pressure of 180 mm Hg, it would be very difficult to get them down below 120.
Dr Black: What about chronic kidney disease? Was there an upper limit of glomerular filtration rate (GFR) serum creatinine that was included here?
Dr Cushman: Yes. The goal was to include a fairly good-size population of patients with chronic kidney disease. We ended up with about 28% of the participants having chronic kidney disease, with estimated GFRs between 20 and 59 mL/min/1.73 m2. So with normal GFR, they could still get in the study if they were also at high risk.
But actually having chronic kidney disease, with a GFR of < 20 mL/min/1.73 m2, was a risk factor in and of itself that allowed them to get into the study if they were above age 50 and had the right blood pressure levels. [Editor's note: Speaker means to say eGFR 20-59 mL/min/1.73 m2; patients with eGFR < 20 mL/min/1.73 m2 were excluded.]
Dr Black: The Framingham Heart Study percentage was 1.5% per year—15% over 10 years. Is that right?
Dr Cushman: That's correct.
Dr Black: Anything else that you can tell us right now? Or do we have to wait for the details?
Dr Cushman: Obviously, people are interested in subgroups—for example, the chronic kidney disease population; patients over age 75, which was a prespecified subgroup; men vs women; blood pressure levels at baseline, etc.
These subgroups are still being analyzed and will be reported in the primary paper. But what we can say is that the results are consistent in the subgroups. In other words, there's no heterogeneity when you're comparing the different subgroups.
Implications for Guideline
Dr Black: Can you give us any hints about how long it's going to take before we actually see that?
Dr Cushman: Our intent is to publish the primary results paper before the end of 2015, so hopefully within a couple months.
Dr Black: That would be good. I think a lot of us have been waiting anxiously for these results—you probably more than anyone—to see how it really turns out.
Dr Cushman: Right.
Dr Black: I really appreciate you sharing this with us now. I look forward with tremendous interest to the final paper. The real question (which you probably can't answer either) is, do you think that the guideline committee is going to comment on this?
Dr Cushman: First,I want to point out that it's very important for people to realize that we measured blood pressure in the office in a very specific way. We used an automated machine (an Omron machine) that automatically waited 5 minutes and then took three blood pressures and averaged them. This blood pressure was measured while the staff were out of the room. So this is not typical of what's done in many office practices to date.
Dr Black: Right.
Dr Cushman: With the way it's done in office practices today, even if it's done with a good machine and by somebody who knows how to take blood pressures, it often is 5 or 10 mm Hg higher than that. It’s important for people to realize that when we talk about these blood pressure levels, we're talking about taking the blood pressure in a very specific way.
It's very consistent with the way we've done outcome studies in the past, even though in studies like SHEP the staff were in the room because they were taking blood pressures with a mercury manometer. But I think the results are very consistent with that.
In terms of guideline committees, I've been on various guideline committees, as you have. We're going to have to see how the medical community responds to this. It's not up to me, but I think there will be a fairly rapid response. There is a set of guidelines that I'm not involved with that's being developed by the American Heart Association and the American College of Cardiology on the treatment of hypertension.
Several of the key members of that committee have also been involved in SPRINT. I have a feeling that they'll wait to comment and include these results in whatever they say. Time will tell, but the results are so clear and so dramatic in terms of the benefit. Looking at adverse effects and the potential harmful effects, in general, the intervention was fairly well tolerated. And clearly the benefits outweigh the harms.
I think that there will be a guideline recommendation based on SPRINT. The key issue will be: Who does it apply to, and who does it not apply to?
Dr Black: You were always one of the leading advocates regarding not having evidence yet for the 140–mm Hg goal. I guess we do now.
Dr Cushman: That's right.
Dr Black: Bill, I want to thank you very much for all of your efforts and for participating in this conference. Thanks again.
Dr Cushman: Thank you very much, Henry.
Medscape Cardiology © 2015 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: SPRINT Hypertension Trial: Preliminary Results Discussed - Medscape - Sep 18, 2015.
Comments