[MUSIC PLAYS]

SPEAKER 1: What I would like to speak about today is two things. One is to review the results of the systolic blood pressure intervention trial, or SPRINT. And to give you a little bit of a behind-the-scenes perspective, not only on what were the results of this study, but how we achieved those results, at least at the University of Pittsburgh site-- talk about the main results of this study. And then the two questions that people often ask me, when we're talking about SPRINT, is what about the older adults in SPRINT? And what about quality of life?

The other thing that I would like to discuss is what are the possible implications for SPRINT results in primary care settings, which is where I spend my clinical time. I'm going to have some brief words about population health management, and then just maybe some questions to provoke some thoughts, as we head to lunch, about how we can expand current care models for hypertension. If we in fact think that the SPRINT results are ready for prime time.

So as we know, hypertension affects everyone, almost everyone as they get older. One billion adults worldwide, and you would think that with such a prevalent disease that's been around for such a long time that we would have it all figured out. We've got this all figured out. However, the target for systolic blood pressure lowering is uncertain. And I recently had a wonderful conversation with one of my mentors, Dr. Lewis Kohler, who's going to speak to you about statin management this afternoon.

And you feel like you're this young investigator, I was working on the SPRINT study, this is so exciting. And he was, like, oh Molly. When we did SHEP, you know, which is the study of systolic hypertension in elderly patients, you know, the study was done in the late '80s, everyone was all, oh, you're going to lower blood pressure. This is going to be terrible. What about the older elderly population? So he kind of helped me put this in a little bit of historical perspective that we've gone back and forth on blood pressure guidelines before.

Speaking of that, in 2014, late 2014, the most recent official hypertension guidelines, JNC 8, came out, and moved the systolic blood pressure goal from less than 140, which was what JNC 7 had said, to less than 150 for adults over 60. And this engendered a bunch of debate and dissent, including a dissenting article from some of the investigators who were on the JNC 8 committee. So into this landscape of hypertension, it's very prevalent. It's been around forever. We should know the best thing to do. Do we? Comes the SPRINT study.

The objective of the SPRINT study was to compare the benefit of a treatment of systolic blood pressure to a target of less than 120 millimeters of mercury, which I'll refer to as the intensive group or intensive goal, with a treatment target of less than 140 millimeters of mercury, which I'll refer to as the standard treatment or the standard goal. Neither one of these are what JNC 8 says, because SPRINT was obviously designed and rolling before those guidelines came out. But the standard treatment arc is more concordant with previous blood pressure guidelines.

So we did a randomized controlled, open label trial, that was conducted at 102 clinical sites, including our site here in Pittsburgh. They were organized into five clinical center networks in the United States.

I'm listing the SPRINT inclusion criteria for you here. SPRINT doesn't apply to everybody, you could argue, because it did not include all patients with hypertension. Specifically, we were looking for patients who were at least 50 years old, who had blood pressure on a certain number of medication that fell into the levels that I'll let you just peruse there. Just as an example, if you were on three or four medications, you couldn't have a blood pressure of higher than 160 or 150 millimeters of mercury, because it was believed that it would be too hard to get patients like that to the intensive treatment goal if they were randomized.

So patients who have resistant hypertension, which is thought to be a non-- a not insignificant number of our patients with hypertension, or patients who perhaps just could not take their medications correctly, were not selected to be in SPRINT. In addition to meeting the age and blood pressure level cutoff, patients also had to have a certain level of cardiovascular risk to be included in SPRINT, specifically they had to have had previous clinical or subclinical cardiovascular disease, but not stroke.

Chronic kidney disease, with an EGFR in the 20 to 59 range. They couldn't be on dialysis. A Framingham risk score for 10-year cardiovascular disease risk of greater than or equal to 15%, or be greater than or equal to 75 years old. And that last population was one that we specifically targeted for SPRINT. Our goal was to get at about a third of the participants in that older age category. There were multiple exclusion criteria. For the sake of brevity, and also to focus on sort of generalizability to your practice, the two most important ones for you to know about are no diabetic patients and no patients with prior stroke.

So what did we do? This is kind of like looking beneath the hood in SPRINT. How did we try to get patients at these blood pressure goals, and keep them there for the duration of a planned five-year study? We started them off very intensively. When patients were first randomized to the study, they came in for monthly visits with study staff for three months. And basically our goal was during that first three month period, to get them to the goal where they were supposed to be.

If we were successful in the first three months, then we saw them on a quarterly basis. At those quarterly or monthly visits, we titrated their blood pressure medications, as you can see. If they were in the intensive arm and not a goal we would increase their medications. And if they were in the standard arm, and their blood pressure was either lower than 130 at one visit, or less than 135 for two subsequent visits, we would decrease their medications. And as soon as they had a change in their blood pressure medication, they would again be coming back to see us on a monthly basis, until we were sure that they were at goal.

So I don't know about you guys, but I don't see my patients with blood pressure that often. And these were all in-person visits. This was coming in, seeing my study nurse, who basically, her full time job was to manage this group of about 150 patients, who were in the SPRINT study.

So this was in a very intensive protocol. And at least in my practice, this does not mirror what I do with my patients with high blood pressure, even in these higher risk groups. The primary outcome of this study was myocardial infarction, acute coronary syndrome stroke, heart failure or cardiovascular disease death. So this was a composite outcomes-- pardon me a composite outcome. The other outcomes that we were interested in were all-cause mortality. That's an important one, as well as some kidney and cognitive outcomes.

Who were the people who came in to SPRINT? We randomized close to 9,400 participants across the country. Again, about 150 here in Pittsburgh. They had an average age of almost 68 years. And we got pretty close to our target of having a third of the population-- that oldest group, we had 28% who were 75 years or older. Most of the non-white participants were black. And we had about 30% black patients in SPRINT. The average baseline systolic blood pressure was around 140. And all other indicators of cardiovascular disease risk, use of medications that would lower your cardiovascular disease risk and kidney function, were basically balanced between the treatment arms at baseline.

So all that work, that my fantastic nurse, Tina, did, did it work? This is not our data here in Pittsburgh. But this is basically showing that in the aggregate SPRINT population, that we had a really good separation of the blood pressure in the standard treatment arm, which is shown at the top in blue, and the intensive treatment arm, which is shown at the bottom in sort of a reddish-orange color. So basically, we got to where we wanted to go with that intensive protocol.

Did it make a difference in the outcomes that I specified to you? It did. So again, the standard treatment arm is in blue, and the intensive treatment arm is an orange. And what we saw is that separating pretty early, around the first year of the study, but then more significantly as time went on, there was a lower risk of having a cardiovascular event in the intensive arm, compared to the standard treatment arm. About a 25% lower risk, and that was statistically significant, significant to the point where we actually were supposed to continue for five years. And they stopped the trial early, because the data safety monitoring board said this is too big of a result for us to be able to continue this study, as it's currently randomized.

So we continue to observe people, after the data safety monitoring board told us that. But we no longer treated them according to protocol. Death from any cause was also significantly different between the two arms with a very similar hazard ratio. What was driving this? This is a good follow-up for Dr. Ramani's discussion. Basically what you'll see is that between the standard and secondary outcomes, there was no difference in heart attack, acute corners or stroke.

But there is a very significant difference in incidence of heart failure, and that was one of the main drivers of this significant composite outcome. For those of you who like to think about number needed to treat, which I think is a good way to think about this in a more global population health way, you would need to treat 61 patients on the intensive protocol on two to prevent one incidence of the primary outcome, which again was that composite cardiovascular outcome. You would need to treat 90 in the intensive protocol to prevent death from any cause, and 172% to prevent cardiovascular disease death. So those numbers needed to treat are on the line with other treatments that we recommend and consider standard good practice.

But what about the downside? So all this wonderful blood pressure control and decreased incidence of the primary outcome, there was a cost. There was no difference in serious adverse events between the two groups, but if you look at individual types of things that happened, hypotension was more common. Syncope was sort of borderline more common. And electrolyte abnormalities were all more common in the intensive treatment arm compared to the standard treatment arm. So that is something to consider, when you're sort of weighing, perhaps for an individual patient, you know, is the SPRINT protocol right for him or her. There was no difference in injurious falls between the two groups.

So there were a whole flurry of editorials that came out when this article was published in the New England Journal of Medicine in late 2015. In general, it was thought that it was a high quality study, and that the results were reliable. It was thought that this should require some urgent action on hypertension management, that these weren't just results that we should sit with. And specifically, there was also a thought that a move toward more aggressive goals will create a major challenge for clinicians. And that's what I want to spend sort of the second half of my comments today, talking about-- the sort of, what is the challenge? And how are ways that we could potentially meet it?

And they also had a side comment about hypertension prevention being weak in the US, but I'm going to kind of leave that to the side because this was not a primary prevention study. So one of the questions that I got after the SPRINT results came out, because a lot of people sent me e-mails and kind of asked for my thoughts, were, well, what about the older adults? And more recently, colleagues from the SPRINT study published a paper in JAMA that looked just very in a focused way on that population that was 75 or older. And basically what it showed is that the results were, if anything, even more dramatic in favor of intensive treatment in this arm. And it makes sense because these are the people who are more likely to have cardiovascular events, be hospitalized for heart failure.

So we saw similar results in this group with a number needed to treat of only 27. And I'll just remind you that it was 61 for the overall SPRINT population to prevent the primary outcome, and 41, compared to 90, in the overall SPRINT population to prevent death from any cause. They also did a really nice secondary analysis, where they said, well, let's not just look at older adults, but what about the older adults who we'd be really worried about putting on maybe a more intensive blood pressure medication, so people who are frail, people who have a low gait speed, it might be at higher risk for falls. And again, the benefits were consistent in those groups as well.

I've been most involved with the group who has been looking at the quality of life results for SPRINT, because one thing that you might argue would be it's great if you can get blood pressure, lower maybe, add some months or years to someone's life. But if they're miserable taking these medications or adhering to this more intensive protocol, sort of, what have we gained? I'm thinking about it from a patient-centered perspective.

And basically what these slides illustrate, on the left, we have the mental and physical components of the SF 36, which is a widely accepted quality of life measure, showing that the two groups-- and here, intensive treatment is in red, and standard treatment as in blue-- did not differ in either their mental or physical health at the beginning of the study. And nor did they differ at the end of the study or any time point in between. So basically, you know, it would be great to say, like, and it makes you feel great, too. But what we can say is that there is no detriment to quality of life in the folks who were in the more intensive treatment group.

On the right, that is actually a measure of depression, the PHQ 9. And again, a similar statement could be made. Depressive symptoms were similar in the two groups at baseline. And even though those lines look like they're diverging, there's no statistically significant difference at any time point.

The thing that I forgot to mention earlier, too, in terms of what it took to do this was that, on average, at the end of the study, the folks in the intensive arm were on one more medication than folks in the standard arm. So about two meds to get to goal in the standard arm and three in the intensive arm. I meant to mention that earlier, and I neglected to.

So my editorial comments, I was surprised. Actually, I had equipoise, which is what you're supposed to have when you're doing the clinical trial. If you're asking people to be in one of two groups, and feel good about yourself, you should think that either group could be successful. So I was surprised when we got the notification from the data safety monitoring board that the trial was being stopped early. But I didn't doubt the results.

I think we need some more information to really put this into context, and take this into practice about the impact on kidney disease. Cognitive function is one of the big ones that we're waiting for. And then also data will be coming out soon about the cost-effectiveness of SPRINT.

But what I really want to reflect on for the last few minutes that I have your attention before lunch is sort of how we might need to think creatively about responding to what are likely going to be new blood pressure guidelines that are coming out in response to SPRINT. As I indicated before, this style of blood pressure management is possible, but it's intensive. And I think it's impractical, in most clinical settings, using our current care models. So what would we do about that?

I'm going to skip that slide in the interest of time, because I'm sort of restating the same thing. So one thing that's become a very popular buzzword here at UPMC and elsewhere is population health management. And what does that mean when you think about hypertension? It means looking at not sort of an individual patient who's walking into your office. But sort of how would you take care of a group of patients, whether they be your individual patients, your practice as patients, or all of our patients that UPMC, with regard to their blood pressure.

So there is some data out there on the fact that population health management strategies can be successful in managing blood pressure. And the folks at Kaiser Permanente in northern California, which is an integrated health system, had a nice report in JAMA about three years ago showing that reports to primary care providers about blood pressure management, along with tools to help patients, can help to better control blood pressure in a patient population. And I'm going to repeat myself on purpose.

I think the important phrase in that sentence is "along with tools to help patients," because I think just sending more alerts and sending more reports to doctors, it's a nice FYI. And you open it, or at least I open those Manila envelopes when they come, and think, "Hmm. Good for me, Molly," or "Gosh, I need to do better." But I think the tools are really important.

So in the Kaiser program, the tools were the blood pressure Data Registry, which allowed them to generate the reports, shared best practices for physicians. They also gave decision support to PCPs in terms of helping them decide when to act on intensifying blood pressure management onto the patients they gave combination therapy, which it's been shown in other blood pressure studies to increase adherence, basically giving patients two medications, or in the same pill. And they also had follow-up with medical assistance. So there was support for the doctors and support for the patients. And what that resulted in-- and again, this was not a randomized controlled trial. So this is a pre-post study design is in the Kaiser group, which are shown here shown in the solid circles.

Over the course of this program, there was an improvement in controlled hypertension, from mid 40s up to 80% in their population, which was pretty impressive. Again, controlled here is not controlled a la SPRINT, less than 120 in the intensive arm, but this is less than 140 concurrent with the JNC 7 guidelines that were the current guidelines at the time that this study was done. Those improvements in the Kaiser population, that was having this population health program, were significantly better than those in the general population at the same time period.

So how do we think about population health management in real time? I think we have to think about, if we're thinking about our conventional or current models of care, better preparing patients and providers for the visit, which for me, oftentimes is wishing that I had more access to better blood pressure data, whether that be additional clinic readings, or home readings. And also, I'm thinking more about medication adherence review, either in preparation for the visit, as a part of the visit, or calling patients afterwards to make sure they really understand the medications that were perhaps adjusted during the visit.

I think there's also perhaps some potential to put some better clinical decision support into the electronic health record. And I don't want people to start throwing things at me, because I know we don't all need more alerts, and things firing in our faces as we're trying to take care of our patients. But I think a simple tool to identify patients who may benefit from more intensive blood pressure control protocols would be helpful, especially if they are-- kind of stealing a page from the folks at Kaiser-- tied to best practices and tools, not just an alert, but something that's going to help you take better care of the patient. And ultimately, the patient and the provider are going to have to make the final choices about what is the best goal and therapy.

So that all works thinking in the context of an office visit, but I would also argue that we probably have to think outside the office if we're really going to get where we want to go with blood pressure control, whether you're thinking about the current guidelines or the new guidelines that might be coming down the pike. I think remote collection of blood pressure is something that we've already started doing at UPMC. Again, Dr. Ramani mentioned remote collection of data for his patients with heart failure. It's something that we're looking at in the general hypertensive population as well.

But I think the thing that's still in my mind a very active thought is not "Can we collect the data?" We can, and we're doing it now. But what do we do with this blood pressure data once we get it? And just, you know, you can think for yourself right now, if you had a patient who was monitoring remotely at home, would you want to be alerted for high readings? Who should that alert go to, and how often? And if you find that a patient really isn't doing well, what would you do next? Would you want to have that patient have some remote advice and follow-up? Would you want an in-person follow-up, or some combination? And then who are the other people in our primary care team who can be assisting with management?

I think, in general, that new models of care should offer more support for patients and providers. For patients, I think things that would be helpful would be easier home measurement, easier medication regimens to promote adherence, and easier access to information and feedback on blood pressure.

A small thing that we did during SPRINT that actually made a big difference is patients always left their visit with their meds in a bag. And it sounds like such a small thing, but especially for some of our older patients, who it's like getting to the pharmacy, getting those meds, they were incredibly grateful, and these were standard generic medications. It's not going to break anyone's budget to-- that would not break anyone's budget to provide. But it's just something to think about when we think about the cost-effectiveness of what we're currently doing. For providers, I think, easier access to accurate blood pressure data would be helpful. Again-- again, exactly how I think is something we still need to figure out. And then more support for hypertension care for both our simple and more complex patients.

In summary, I think the SPRINT study will change, but not revolutionize, blood pressure treatment. Again, my mentor helped me keep some perspective on sort of the history of where we've been with blood pressure decisions. And thinking about "Is what's old new again?" in terms of sort of just stepping up the approaches, and in using things that we know work.

I think we will go back to the basics of blood pressure management, which again, to steal a line from Doctor Kohler is "Adherence is the ballgame," but we might be using some newer tools and out of office protocols to help our patients become more adherent to regimens. I think the increased emphasis on population health management and new models of care may be the real revolution. But we're going to have to work together to discover and disseminate best approaches.

[MUSIC PLAYS]