Tuesday, February 28, 2017

How Low Should We Go with Blood Pressure in Older Adults? A Podcast with Mark Supiano

In this GeriPal Podcast we talk with Dr. Mark A. Supiano about a blood pressure management in older adults in the light of new evidence from the Systolic Blood Pressure Intervention Trial (SPRINT).

In particular, we talk about a recent paper he co-author with Jeff Williamson in the Journal of the American Geriatrics Society (JAGS) titled "Applying the Systolic Blood Pressure Intervention Trial Results to Older Adults."  We discuss issues around the generalizability of the SPRINT trial, whether it applies to a subset of frail older adults, and whether the way we measure blood pressure in a typical office visit influences how we should think about the question "how low should we go?"

GeriPal Podcasts can be found on:

Transcript of the podcast:

Eric: Welcome to the geripal podcast this is Eric Widera.

Sei: And Sei Lee.

Alex: And Alex Smith.

Eric: And before we get to our guest who's calling in all the way from Utah, maybe Sei you can give us a song that can give the audience a hint about what we're gonna be talking about today.

Sei: How about friends in low places.

Eric: Alright.

Sei and Alex sing “Friends in Low Places” by Garth Brooks:

All together now, Blame it all on my roots, I showed up in boots and ruined your black tie affair. I was the last one to know, was the last one to show, I was the last one you thought you'd see there. I saw the surprise, and the fear in his eyes, when I took a glass of champagne. Then I toasted you, said honey we may be through, but you'll never hear me complain. 'Cause I got friends in low places where the whiskey drowns and the beer chases my blues away. I'll be okay. I'll be okay.

Eric: So with that intro of a song, today we have Mark Supiano with us, he's a professor and chief of division of geriatrics University of Utah School of Medicine, and the director of the Salt Lake City grecc and Mark just got published, I think just a couple months ago, is that correct Mark? An article on plying the systolic blood pressure intervention trial or SPRINT trial results to older adults in a JAGS issue that just got published in January.

Mark: Yes, it appeared in the January issue, it was published online early I believe back in November, and I should also acknowledge the co-author with me on this article is Jeff Williamson from Wake Forest where he's the chief of the division of geriatrics there and has been integrally involved in the SPRINT trial.

Eric: I'm gonna just start us off, how should we be integrating the SPRINT trial into our clinical practices?

Mark: Sure, so that question Eric, is exactly what Jeff and I wanted to respond to after the main trial results were published in New England Journal of Medicine in late 2015 and then in May of 2016 the results from SPRINT specific to individuals age 75 years and older was published in JAMA. The questions that Jeff and I were getting from our colleagues in geriatric medicine and primary care providers was how we should be interpreting these results and what sense does it make for patients that we're seeing in our clinics?

So, there are several points to make I'll start with I think the first one really, which is the case with any randomized trial result that you are considering applying to the patient in front of you, is whether or not the patient would have met the eligibility criteria for that trial. So in other words does the trial finding generalize to the patient that I'm treating. So with that in mind there were several important exclusions to SPRINT I can name just several now, and then we can go into some of the details, but importantly SPRINT did not include individuals with a history of diabetes, it did not include individuals with prior history of stroke or had prevalent congestive heart failure, or dementia. Form the geriatrics perspective individuals for community dwellings, so ambulatory community dwelling adults, so not nursing home residents and without dementia at the time of entry into the trial.

So those were some of the important characteristics of the study population that need to be kept in mind when you're deliberating whether or not the SPRINT blood pressure target of 120 would be even something that you would consider applying to a particular patient.

Eric: And real quick overview for our audience who may have read the SPRINT trial over a year ago, what was the intervention again?

Mark: So, SPRINT is a bit unusual it was not a specific medication intervention, it was a single labeled design, the subjects knew, which arm they were in, two blood pressure targets. So a standard arm of a systolic blood pressure of below 140 and an intensive arm to a systolic blood pressure target of less than 120. To accomplish that blood pressure reduction on average required one additional medication, so at the time of entry in to the trial the average number of antihypertensive medications that the subjects were taking was two. And the average systolic blood pressure on entry was just over 140. To achieve the lower systolic target of 120 required on average an additional medication so that at the end of the study the individuals in the intensive arm were on three antihypertensive medications while the standard arm average remained at about two.

That said the separation in systolic blood pressure between the two groups occurred relatively quickly, so within the first six months there was more than a double digit difference in systolic blood pressure and that delta was maintains across the duration of the trial for the 75 and older group the delta ended up being about 11 millimeters of mercury between the two groups.

Eric: And if I remember correctly the mean blood pressure in the intervention group was actually slightly above 120, right?

Mark: That's correct it was, if I'm recalling correctly, it was 123 for the 75 and older population so, half the group had blood pressures above 123 and half were below that.

Eric: So what are we supposed to make of that, so if the goal was less than 120 but still the majority, or the mean was above 120 in the intervention group, how should that play into our clinical thinking?

Mark: Yeah, so that's a good question, there will be additional analysis done to determine the impact or the relationship between the achieved systolic blood pressure and outcomes, as opposed to assignment to treatment arm, right? So at the moment we have the results based on assignment to treatment arm, to address that question whether there was added benefit or potential harm in going based on the achieved systolic blood pressure those analysis have not yet been completed.

But let me take a tangent off on that question to make an additional very important point about measurement, because this is another question that does come up and something that Jeff and I highlighted in the Jags article, is that the protocol that was used in SPRINT to measure blood pressure at the time of entry and then at all of the follow up phys-eds is not the routine office blood pressure measurement that most patients are getting when they go in for a visit with their provider. So the blood pressure measurement details are that an automated blood pressure device was used, it was programed to require that the individual being sitting quietly for five minutes before any blood pressure measurement was taken, the appropriate details regarding cuff size and position of the arm and so forth were adhered to, and then after the five minute rest period three serial blood pressure measurements were taken most of the time without the observer in the room, and the average of those three serial blood pressure readings was then taken to be that clinic blood pressure for that day.

Eric: So that sounds very different, I just had a blood pressure check yesterday, what happened was, I walked in to the room, the nurse was talking to me the whole time, she slapped on a blood pressure cuff-

Mark: Probably over your jacket.

Eric: (laughs) It was actually not over my jacket, but over my shirt and over an undershirt. Throughout the whole intervention maybe it took like one minute to do the blood pressure check, again we were talking and chatting throughout the whole time. Different, it sounds like.

Mark: There's a difference, and the reason it's important is that there's pretty good evidence to suggest that this automated protocol result is anywhere between five to eight millimeters of mercury lower than the blood pressure that you just had done yesterday. Alright so if it was repeated using this protocol chances are your systolic pressure would be five to eight millimeters lower than what you obtained with the protocol that you described. So, the main message here is that if you intend to treat to the SPRINT target of 120, again this was the target as you pointed out earlier not everyone got to that level, but if you're attempting to bring someone's blood pressure down to that range, it really is important that you're utilizing the protocol that SPRINT used to make sure that you're not over-treating and getting someone down even lower.

Eric: And if we don't have a lot of control over the protocol in the short term, I mean this sounds like a great quality improvement project, should we be targeting slightly higher by five to eight millimeters mercury?

Mark: That would be the correct interpretation, Eric, although I want to take this opportunity to push back on your first part of the statement, I think we really should push back in fact to change our work flow and the procedures so that we are measuring blood pressure correctly and accurately. So the excuse that while the blood pressures of SPRINT aren't meaningful in the real world, my response to that is we need to make the real world a better place and utilize the appropriate protocol to measure blood pressure accurately so that we are in fact treating people to the right target. This also mitigates in part some of the white coat hypertension that we see more commonly in geriatric patients and just so I don't lose my train of thought if I can interject one more important measurement aspect because it will relate to the comments that I'm sure we'll get to in a minute about adverse events, but the other critical aspect of SPRINT and critical aspect for treating hypertension in older people, is to make sure that orthostatic hypotension is checked for and monitored.

So, an important exclusion to the SPRINT trial was if someone's standing blood pressure after one minute was below 110, it turns out that for the 75 and older population, ten percent of individuals who are screened for enrollment into SPRINT had blood pressures below 110 after one minute of standing. So this was not uncommon. So, the other take home is that before you entertain a lower target for the patients that you're seeing for their hypertension, it's obligatory that their orthostatic blood pressure be measured to make sure that you're not putting that patient at risk for this more intensive systolic blood pressure target down to 120.

Eric: When I think of orthostatic blood pressure I often think of not just systolic blood pressure of less than 110, but a drop in systolic blood pressure when standing, was that also part of it? Or was it just, if your blood pressure was less than 110?

Mark: So the exclusion was if the standing blood pressure was below 110, that said orthostatic blood pressure measurements were checked at every visit, which is another thing not commonly done in routine clinical practice, but is in fact a standard dating back to JNC 7 actually. So this is not new information, but again not commonly done. But the protocol for follow-up visits in SPRINT was that orthostatic blood pressures were assessed as well as carefully assessing for adverse events of hypotension, dizziness, and falls.

Sei: So first of all, I want to congratulate you on just an amazing study. And I have to say, it was great reading a study that was designed and really led by geriatricians, and I was thinking as I was reviewing it thinking about how it differed from the HYVET study where things that I was so interested in like frailty and cognition were measured at baseline so that I could really make a critical assessment of whether the patients in the trial were similar to the patients that I see. So, first I just wanted to say thank you for supplying us with the evidence to actually be able to say how well the trial data fits with the patients that we take care of.

Mark: Well, thanks for that comments Sei, and I want to give credit where credit is due and I want to thank Jeff Williamson and others on the SPRINT steering committee, but Jeff in particular really went to bat to incorporate the geriatric measures that you just sighted so, the cognitive battery, the gate speed, the frailty assessments that we've been able to report on, those were built into the design of the trial. And I think as geriatricians as you were just saying that we really should applaud finally, that we've been successful in getting these outcomes into a major trial like this because it is so important for the older patients. And let's also applaud the fact that, and it was actually thanks to the ARRA funding that there was an opportunity to augment the recruitment of individuals 75 and older into the trial, and so that population accounted for 28 percent of the entire cohort. So their 2800 individuals over the age of 75 who participated in the trial so we also need to thank the SPRINT steering committee and NHLBI and the other funders, including NIA, for that investment to allow that to happen. Thank the sites it's very difficult to recruit and retain older people into trials like this. And also, most importantly thank and acknowledge all of those participants who contributed to this information and really allowed us to make this kind of report.

Eric: So the other thing I want to thank, is going back to the Jags article that you recently published, you had a section in there called time to benefit. And thinking about medical decisions in the context of prognosis, how should we be thinking about the SPRINT results as far as time to benefit?

Mark: Well thanks for bringing that up Eric, it's something I've given a lot of thought to and again it is geriatricians and for the Geripal audience as you just alluded to, we need to carefully think about any treatment that we're gonna recommend to individuals, particularly frail, older individuals at advanced ages and I think with that said, the flip side of that is, we need to be careful not to under-treat older, frail people, who may derive from more intensive treatment of important cardiovascular risk factors such as high blood pressure.

With that in mind the information that's maybe somewhat lost in this that I will take this opportunity to emphasize is that similar to HYVET, you just brought up the HYVET study so, you may recall that HYVET, like SPRINT stopped early because the DSMB identified a significant mortality benefit in the group that in HYVET had been randomized to active treatment relative to placebo, and in SPRINT similarly the trial ended earlier than anticipated. The trial was designed to go a little over five years, it ended at 3.2 years of follow-up, when the DSMB noted the unequivocal benefit to the intensive treatment group in a mortality outcome, as well as the primary cardiac outcome. So with that in mind the other study that people may not be familiar with is that there was a follow on extension to HYVET that took people who had been in the placebo arm of HYVET and crossed them over to active treatment and the same benefit for the HYVET outcome of stoke, so the same benefit with about a 35 percent reduction in the hazard ratio for stroke occurred in that group that crossed over within the first 12 months of treatment. So, when you think about how high blood pressure develops over time and we think about these risks that accrue over a lifetime.

That said the benefits that we're talking about for reduction in cardiac morbidity and mortality and total mortality are occurring in the space of 12 to 24 months. So for someone with a life expectancy that exceeds one to two years, the potential for benefit is there. I think another point to move to then, and I'll be very interested in both of your input into this, but quality of life for this patient population supersedes, I almost said a different word but I've eradicated that word from my vocabulary given the current events in Washington, so the quality of life may supersede mortality for our patient population and these results are not yet published, they've been presented at a national meeting and the paper is under review but, the quality of life was another outcome that was assessed in SPRINT and the short take home was that there were no deleterious effects in quality of life in individuals randomized to the intensive arm.

And from our perspective for the geriatric population that also held true for the most frail subgroup of the 75 and older population. So these benefits occurred without a negative impact on quality of life, and finally speaking of frailty, we are often somewhat nihilistic about recommending treatment to an older frail individual, but it seems to me that the last thing an older frail individual needs is to develop heart failure or to have a stroke, because if they're already old and frail the morbidity and disability that is incumbent with the development of either of those outcomes could be devastating. So if we can successfully prevent additional disability that would be related to the development of congestive heart failure or a new stroke. It seems to me that that's an outcome that many of our patients would consider as important to them, and if so worth the additional costs and time and so forth that would be involved in one additional medication to get their blood pressure a little bit lower.

Eric: So the one last question from my standpoint then, is when we're thinking about other risks from this, and especially since the trial ended early, sometimes risks of medications and treatment strategies build up over time, and the one that you talk about in the JAGS article that really popped out at me is kidney outcomes. How should we think about the SPRINT trial, because there were increased number of negative kidney outcomes, is that right?

Mark: Correct, so these are referred to as conditions of interest, so the event rates not only acute kidney injury and electrolyte abnormalities, primarily hyponatremia and low blood pressure were more common in the intensive group. That said, these have been found to be transitory, there was no increased rate of renal failure, in fact the number of renal outcomes was very low overall so there was no harm in terms of permanent renal injury that developed. So these transitory adverse events can be modified by dosage adjustments and other things that can be done quite easily, in turn if you develop a cardiac event, that's not as modifiable, right? If you have a heart attack or stroke, you can't go back and change that, so our take home was that these risks although present, are far out raised by the cardiovascular mortality benefits that were identified. The final point though with regard to your question, Eric, is that there are longer term studies that will be carried out in the SPRINT cohort to address your question about longer term kidney outcomes, so this is called the SPRINT ASK study, so aging seniors and kidney I think is the acronym. Geriatricians don't have quite as sexy acronyms as cardiologists.

Eric: Cardiologists are really good at it.

Mark: Yeah, we needed a cardiologist to help us with that, but SPRINT ASK will likely be funded this year, we don't have the definitive notice of award yet, but that's looking very encouraging and if funded that will allow additional follow-up for another year in the SPRINT participants, not only for their renal outcomes but the other area we've not yet touched on and I know we're running out of time but just very quickly, the other outcome of keen interest to all of us is the cognitive outcomes, and whether there is an impact on the development of incident dementia with the more intensive blood pressure reduction. So the SPRINT ASK extension will likewise collect neuropsychological outcomes for an additional year in participants who were in the SPRINT trial. So those outcomes on the cognitive outcomes are still forthcoming and that will obviously have a huge impact on how widely adopted the lower blood pressure target is going to reach.

Sei: How do you think these results should be interpreted for patients of the appropriate age, but patients for whom as a clinician we feel like this is a patient who is more frail than the average 80 year old, a patient for whom it seems they're more frail than the frailty of the average SPRINT participant?

Mark: Good question Sei, and we have to think back to the very first thing I said about generalizability is if in your clinical judgment the patient that you're seeing would not have been eligible to enter the SPRINT trial for the reasons that you've stated, we should be very cautious and thinking about generalizing and applying it to that individual. So one particular example would be, because this question has arisen relatively commonly, is what about residents of skilled nursing facilities or long term care residents. So that's a very different population where, as you're pointing out their life expectancy may be quite low and we need to think carefully about an intervention in that population and we don't want to over-generalize these findings to a more frail population such as you're describing.

Eric: So I want to thank you Mark for really the insightful comments and for writing the Jags paper, which we will put up on our website for our listeners to review.

Mark: Just to add one quick thing in response to Sei's last question, I thought you were also gonna ask what happened longitudinally with frailty, we do have those data and I actually just saw a sneak peek of those results yesterday, but we're preparing to present the longitudinal frailty outcomes at the international meeting of the Association of Gerontology and Geriatrics or the IAGG meeting in July in your home city of San Francisco-

Sei: Yes, we'll see you there.

Mark: So, come to our session, there's a SPRINT session symposium that's been accepted for presentation at IAGG so we'll have some additional data there, and again we look forward to fielding more questions and getting more input about this very important topic.

Eric: And we could have the in-person podcast!

Mark: Fantastic, on the scene right?

Eric: The sequel, we'll have you at that next one, we'll have you sing our closing verse with us.

Mark: Oh yeah, I've warned you not to invite me to sing.

Sei and Alex sing “Friends in Low Places” by Garth Brooks:

Blame it all on my roots, I showed up in boots, and ruined your black tie affair. I was the last one to know, I was the last one to show, I was the last one you thought you'd see there. I saw the surprise and the fear in his eyes, when I took his glass of champagne. And I toasted you said honey we may be through, but you'll never hear me complain. 'Cause I got friends in low places where the whiskey drowns and the beer chases my blues away. I'll be okay. I'll be okay. And I'm not big on social graces, think I'll slip on down to the oasis, oh I got friends in low places.

Transcript edited by Sean Lang-Brown


AKT said...

Another great podcast -- entertaining and insightful. Thanks guys!

Mary George said...
This comment has been removed by the author.